2012 AASHTO BRIDGE COMMITTEE AGENDA ITEM: 41 (REVISION 1)

SUBJECT: LRFD Bridge Design Specifications: Section 13, Articles 13.8.2 & 13.8.3

TECHNICAL COMMITTEE: T-7 Guardrail and Bridge Rail / T-5 Loads

REVISION ADDITION NEW DOCUMENT

DESIGN SPEC CONSTRUCTION SPEC MOVABLE SPEC

MANUAL FOR BRIDGE SEISMIC GUIDE SPEC COASTAL GUIDE SPEC
EVALUATION OTHER

DATE PREPARED: 12/8/11

DATE REVISED: 7/10/12

AGENDA ITEM:
Item #1

In Article 13.8.2, delete the 3rd paragraph as follows:

The design load for chain link or metal fabric fence shall be 0.015 ksf acting normal to the entire surface.

Item #2

Add new Article 13.8.3 the following at the end of Article 13.8.2:

13.8.3Design Wind Loads

The design wind load for chain link or metal fabric fence shall be taken as 0.015 ksf acting normal to the entire
surface. The load from the tributary area of chain link or metal fabric fence shall be applied to the longitudinal
members and posts of pedestrian railings, as well as connections between chain link or metal fabric and posts or
longitudinal members.

Item #3

In Article 13.8.2, change the title from Design Live Loads to Design Loads.

OTHER AFFECTED ARTICLES:

None

BACKGROUND:
This is to provide clarification on loads and load combinations for pedestrian railings with chain link fence.

ANTICIPATED EFFECT ON BRIDGES:

Clarification of loads will prevent unnecessary overdesign of posts on railings with chain link fence.
REFERENCES:
None

OTHER:
None
2012 AASHTO BRIDGE COMMITTEE AGENDA ITEM: 41 (REVISION 1)

SUBJECT: LRFD Bridge Design Specifications: Section 13, Articles 13.8.2 & 13.8.3

TECHNICAL COMMITTEE: T-7 Guardrail and Bridge Rail / T-5 Loads

REVISION ADDITION NEW DOCUMENT

DESIGN SPEC CONSTRUCTION SPEC MOVABLE SPEC

MANUAL FOR BRIDGE SEISMIC GUIDE SPEC COASTAL GUIDE SPEC
EVALUATION OTHER

DATE PREPARED: 12/8/11

DATE REVISED: 7/10/12

AGENDA ITEM:
Item #1

In Article 13.8.2, delete the 3rd paragraph as follows:

The design load for chain link or metal fabric fence shall be 0.015 ksf acting normal to the entire surface.

Item #2

Add new Article 13.8.3 the following at the end of Article 13.8.2:

13.8.3Design Wind Loads

The design wind load for chain link or metal fabric fence shall be taken as 0.015 ksf acting normal to the entire
surface. The load from the tributary area of chain link or metal fabric fence shall be applied to the longitudinal
members and posts of pedestrian railings, as well as connections between chain link or metal fabric and posts or
longitudinal members.

Item #3

In Article 13.8.2, change the title from Design Live Loads to Design Loads.

OTHER AFFECTED ARTICLES:

None

BACKGROUND:
This is to provide clarification on loads and load combinations for pedestrian railings with chain link fence.

ANTICIPATED EFFECT ON BRIDGES:

Clarification of loads will prevent unnecessary overdesign of posts on railings with chain link fence.
REFERENCES:
None

OTHER:
None
2012 AASHTO BRIDGE COMMITTEE AGENDA ITEM: 42 (REVISION 1)

SUBJECT: Guide Specification for the Design of Bonded FRP Systems for Repair and
Strengthening of Concrete Bridge Elements

TECHNICAL COMMITTEE: T-6 Fiber Reinforced Polymer Composites

REVISION ADDITION NEW DOCUMENT

DESIGN SPEC CONSTRUCTION SPEC MOVABLE SPEC

MANUAL FOR BRIDGE SEISMIC GUIDE SPEC COASTAL GUIDE SPEC
EVALUATION OTHER

DATE PREPARED: 3/7/12

DATE REVISED: 7/9/12

AGENDA ITEM:
Guide Specification for the Design of Bonded FRP Systems for Repair and Strengthening of Concrete Bridge
Elements See Attachment A

OTHER AFFECTED ARTICLES:

None

BACKGROUND:
This Guide Specification is based on the recommendations of NCHRP Reports 655 and 678 concerning the design
of FRP systems in the repair and strengthening of concrete bridges

ANTICIPATED EFFECT ON BRIDGES:

This Specification will provide guidance for the repair and strengthening of concrete bridges using FRP polymer
composite materials.

REFERENCES:
NCHRP 655 and NCHRP 678

OTHER:
None
ATTACHMENT A 2012 AGENDA ITEM 42 T-6 (REVISION 1)

GUIDE SPECIFICATION FOR THE DESIGN OF EXTERNALLY BONDED FRP SYSTEMS FOR
REPAIR AND STRENGTHENING OF CONCRETE BRIDGE ELEMENTS

This Guide Specification is based on NCHRP Report 655 - Recommended Guide Specification for the Design of
Externally Bonded FRP Systems for Repair and Strengthening of Concrete Bridge Elements (2010) and NCHRP
Report 678 - Design of FRP Systems for Strengthening Concrete Girders in Shear (2011). Additional background
information on the Specifications or specific design examples can be found in the above-mentioned NCHRP
documents.
SECTION 1: GENERAL REQUIREMENTS
1.1 SCOPE
These Guide Specifications are intended for the repair
and strengthening of reinforced and prestressed
highway bridge structures using externally bonded
fiber-reinforced polymer (FRP) composite systems.
These Guide Specifications supplement the AASHTO
LRFD Bridge Design Specifications, 6th Edition, 2012
(AASHTO LRFD). Except where specifically provided
below, all provisions of AASHTO LRFD shall apply.
These Guide Specifications state only the minimum
requirements necessary to provide for public safety and
are not intended to supplant proper training or the
exercise of judgment by the Engineer of Record. The
Owner or the Engineer of Record may require the
structural design or the quality of materials and
construction to exceed the minimum requirements.
These specifications employ the Load and Resistance
Factor Design (LRFD) methodology, in which the load
and resistance factors have been developed from
structural reliability theory based on current
probabilistic/statistical models of loads and structural
performance.
Seismic design shall be in accordance with either the
provisions in the appropriate sections of AASHTO
LRFD or the provisions in the AASHTO Guide
Specifications for LRFD Seismic Bridge Design.
Except where specifically provided below, all
provisions of the LRFD Bridge Construction
Specifications shall apply.
1.2 APPLICATION OF FRP
1.2.1 GENERAL
In general, procedures for the installation of FRP
systems are developed by the manufacturer and can
vary between different systems. Two optional types of
externally bonded FRP systems are available. The first
type consists of dry fiber-fabric sheets saturated with
an epoxy resin. This type of system is referred to as the
wet-layup technique. The second type of system
consists of pre-cured fiber/resin laminates
(manufacturer fabricated), which are bonded to the
concrete surface with an adhesive resin. The wet-layup
system is the more versatile of these systems because
the fabric/cloth are flexible and can conform to most
shapes in the field. Installation procedures may also
vary depending on the type and condition of the
C1.1
Article 1.1 discusses the scope of the guide
specifications, its applicability and limitations. This
article is analogous to the opening articles, Articles
X.1, of each of the sections of AASHTO LRFD.
The commentary is not intended to provide a complete
historical background concerning the development of
these or previous Specifications, nor is it intended to
provide a detailed summary of the studies and research
data reviewed in formulating the provisions of the
Specifications. However, references to North American
and international guidelines (ACI 440.2R-08, 2008;
ISIS Canada Design Manuals, 2001; fib technical
report bulletin 14, fib 2001; CNR-DT 200, 2006; JSCE,
2001; and AFGC, 2003; as well as relevant research
data dealing with externally bonded FRP reinforcement
for reinforced and prestressed concrete structures are
provided for those who wish to study the background
material in depth.
NCHRP Report 609 presents recommended
construction specifications concerning the use of
externally bonded FRP reinforcement for strengthening
concrete structures.
The commentary directs attention to other documents
that provide suggestions for carrying out the
requirements and intent of these Guide Specifications.
However, the commentary and references herein are
not part of these Guide Specifications.
C1.2.1
It is recommended that FRP applications be performed
by a contractor trained in accordance with the
installation procedures specified by the manufacturer.
Comprehensive guidelines in this regard are provided
in NCHRP Report 609, Recommended Construction
Specifications and Process Control Manual for Repair
and Retrofit of Concrete Structures Using Bonded FRP
Composites.
structure to be strengthened. The application of FRP
systems will not stop the ongoing corrosion of existing
steel reinforcement. The cause of corrosion to internal
steel reinforcement should be addressed and corrosion-
related deterioration should be repaired prior to
application of any FRP system.

1.2.2 Surface Preparation C1.2.2

The concrete surface should be prepared to a minimum The bond behavior of the FRP system is highly
concrete surface profile (CSP) 3 as defined by the dependent on a sound concrete substrate and on the
ICRI-surface-profile chips (ICRI 03732, NCHRP workmanship of the installation; either one or both can
Report 609). Localized out-of-plane variations, significantly influence the integrity of the FRP
including form lines, should not exceed 1/32 inch or strengthening system. Proper preparation and profiling
the tolerances recommended by the FRP system of the concrete substrate is necessary to achieve
manufacturer, whichever is smaller. Holes and voids optimum bond strength. Improper surface preparation
should be filled with a compatible epoxy paste can lead to premature debonding or delamination.
(thickened epoxy/putty). It is recommended that
surface preparation be accomplished using abrasive or
water-blasting techniques. All laitance, dust, dirt, oil,
curing compound, existing coatings, and any other
matter that could interfere with the bond between the
FRP system and concrete substrate should be removed.

When fibers are wrapped around corners, the corners

should be rounded to a minimum 1/2 inch radius to
prevent stress concentrations in the FRP system and
voids between the FRP system and the concrete. Rough
edges should also be smoothed by grinding or with
putty prior to FRP application.

1.2.3 Inspection, Evaluation, and Acceptance C1.2.3

FRP strengthening is usually performed on structurally
deficient or damaged RC beams. Before a
strengthening procedure is implemented, the extent of
deficiency and suitability of FRP strengthening should
be evaluated. Systems considered for FRP
strengthening should be inspected by a licensed
engineer or qualified inspector knowledgeable in FRP
systems and installation procedures. The following
should be recorded at the time of installation:
Date and time of installation
Ambient temperature, relative humidity, and
general weather observations and surface
temperature of concrete
Surface dryness, surface preparation methods and
resulting profile using the ICRC-surface-profile-
chips
Qualitative description of surface cleanliness
Type of auxiliary heat source, if applicable
Widths of cracks not injected with epoxy
Fiber fabric sheets or pre-cured laminate batch
number(s) and approximate locations in the
structure
When concrete and atmospheric temperatures exceed
90F, difficulties may be experienced in application of
the epoxy compound owing to acceleration of the
reaction and hardening rates. If ambient temperatures
above 90F are anticipated, work should be scheduled
when the temperature is lower, such as in the early
morning hours. If it is necessary to apply epoxy
compounds at temperatures exceeding 90 F, the work
should be supervised by a person experienced in
applying epoxy at high temperatures. Epoxy systems
formulated for elevated temperature are available (ACI
503R-93).
At temperatures below 40F, difficulties may occur due
to deceleration of the reaction rates. The presence of
frost or ice crystals may also be detrimental to the bond
between the FRP and the concrete.
Evaluate moisture content or outgassing of the concrete
by determining if moisture will collect at bond lines
between old concrete and epoxy adhesive before epoxy
has cured. This may be accomplished by taping a 4 x 4
ft (1 x 1 m) polyethylene sheet to the concrete surface.
 Batch numbers, mixture ratios, mixing times, and
qualitative descriptions of the appearance of all
mixed resins, including primers, putties,
saturants, adhesives, and coatings mixed for the
day
Observations of progress of cure of resins
Conformance with installation procedures
Location and size of any delaminations or air
voids
General progress of work
Level of curing of resin in accordance with
ASTM D2582.
Adhesion strength
If moisture collects on the underside of the
polyethylene sheet before epoxy would cure, then
allow concrete to dry sufficiently to prevent the
possibility of a moisture barrier between old concrete
and new epoxy (ACI 503R-93).
During installation, sample cups of mixed resin should
be prepared according to a predetermined sampling
plan and retained for testing to determine level of
curing in accordance with ASTM D2583. The relative
cure of the resin can also be evaluated on the project
site by physical observation of resin tackiness and
hardness of work surfaces or hardness of retained resin
samples.
For bond-critical applications, tension adhesion testing
of cored samples should be conducted using the
methods in ACI 503R or ASTM D 7234 or the method
described by ISIS (1998). The sampling frequency
should be specified. Tension adhesion strengths should
exceed 200 psi and exhibit failure of the concrete
substrate (ACI 440.2R-08).
The necessary evaluation criteria for repair of existing
concrete structures and post repair evaluation criteria
are well established in the following documents:
ACI 201.1R: Guide for Making a Condition
Survey of Concrete in Service
ACI 224.1R: Causes, Evaluation, and Repair of
Cracks in Concrete
ACI 364.1R-94: Guide for Evaluation of
Concrete Structures Prior to Rehabilitation
ACI 440.2R-08: Guide for the Design and
Construction of Externally Bonded FRP Systems
for Strengthening Concrete Structures
ACI 503R: Use of Epoxy Compounds with
Concrete
ACI 546R: Concrete Repair Guide
International Concrete Repair Institute (ICRI)
ICRI 03730: Guide for Surface Preparation for
the Repair of Deteriorated Concrete Resulting
from Reinforcing Steel Corrosion
International Concrete Repair Institute (ICRI)
ICRI 03733: Guide for Selecting and Specifying
Materials for Repairs of Concrete Surfaces
NCHRP Report 609: Recommended
Construction Specifications Process Control
Manual for Repair and Retrofit of Concrete
Structures Using Bonded FRP Composites
Relevant specifications and guidelines provided by
FRP manufacturers should also be carefully reviewed
prior to the design of any strengthening system.
1.3 DEFINITIONS

Definitions and terms related to the use of fiber-

reinforced polymeric (FRP) materials in bridge
engineering, rehabilitation and retrofit shall be as
stipulated in ASTM D3878. Terms related to adhesives
shall be as specified in ASTM D907.

Definitions and terms related to highway bridge design

shall be as stipulated in AASHTO LRFD.

1.3.1 Symbols and Notation

Variables used throughout the guide specifications as well as references to their usage are listed alphabetically below:
A f , A frp = effective area of FRP reinforcement for shear-friction (in2)
Ag = gross area of column section (in2)
Ah = area of one leg of the horizontal reinforcement (in2)
As = area of nonprestressed steel tension reinforcement (in2)
As' = area of steel compression reinforcement (in2)
Ast = total area of longitudinal steel reinforcement (in2)
Avf = area of steel reinforcement required to develop strength in shear friction (in2)
b = width of rectangular section (in)
b f , b frp = width of the FRP reinforcement (in.)
bv = effective shear web width (in)
bw = girder width (in)
C = clamping force across the crack face (kips)
c = depth of the concrete compression zone (in)
Dg = external diameter of circular column (in)
d f , d frp = effective FRP shear reinforcement depth (in)
ds = distance from extreme compression surface to the centroid of nonprestressed tension
reinforcement (in.)
dv = effective shear depth (in)
Ea = modulus of elasticity of adhesive (ksi)
Ec = modulus of elasticity of the concrete (ksi)
E f , E frp = tensile modulus of the FRP reinforcement in the direction of structural action
fc = stress in concrete at strain c (ksi)
f c
' = 28 - day compression strength of the concrete (ksi)
f '
cc
= compressive strength of confined concrete (ksi)
f frpu = characteristic value of the tensile strength of FRP reinforcement (ksi)
flfrp = ultimate confinement pressure due to FRP strengthening (ksi)
f peel = peel stress at the FRP reinforcement concrete interface (ksi)
fs = stress in the steel tension reinforcement at development of nominal flexural resistance (ksi)
f s
' = stress in the steel compression reinforcement at development of nominal flexural resistance (ksi)
fy = specified yield stress of steel reinforcement (ksi)
f yf = yield strength of steel reinforcement for shear-friction (ksi)
Ga = characteristic value of the shear modulus of adhesive (ksi)
h = depth of section (in); overall thickness or depth of a member (in.)
IT = moment of inertia of an equivalent FRP transformed section, neglecting any contribution of
concrete in tension (in4)
ka = a coefficient that defines the effectiveness of the specific anchorage system
ke = strength reduction factor applied for unexpected eccentricities
k2 = multiplier for locating resultant of the compression force in the concrete
Ld = development length (in)
lu = unsupported length of compression member (in)
Mr = factored resistance of a steel-reinforced concrete rectangular section strengthened with FRP
reinforcement externally bonded to the beam tension surface (kip-in)
Mu = factored moment at the reinforcement end-termination (kip-in)
Nb = FRP reinforcement tensile strength per unit width at a tensile strain of 0.005 (kips/in)
N e = effective tensile strength per unit width of the FRP reinforcement (kips/in)
frp

N frp,w = tensile strength of a closed (wrapped) jacket (kips/in)

Ns = FRP reinforcement strength per unit width at a tensile strain of 0.004 (kips/in)
N ut = the characteristic value of the tension strength per unit width of the FRP reinforcement (kips/in)
Pr = factored axial load resistance (kips)
r = girder corner radius (in)
sv = spacing of FRP reinforcement (in)
T f ,T frp = tension force in the FRP reinforcement (kips)
Tr = the factored torsion strength of a concrete member strengthened with an externally bonded FRP
system (kip-in)
ta = thickness of the adhesive layer (in)
t frp = thickness of the FRP reinforcement (in)
Vc = the nominal shear strength provided by the concrete (kips)
V f ,V frp = the nominal shear strength provided by the externally bonded FRP reinforcement (kips)
Vni = nominal shear-friction strength (kips)
Vp = component of the effective prestressing force in the direction of applied shear (kips)
Vr = factored shear strength of a concrete member strengthened with an externally bonded FRP system
(kips)
Vs = nominal shear strength provided by the transverse steel reinforcement (kips)
Vu = factored shear force at the reinforcement end-termination (kips)
w f , w frp = total width of FRP reinforcement (in)
y = distance from the extreme compression surface to the neutral axis of a transformed section,
neglecting any contribution of concrete in tension (in)
a = angle between FRP reinforcement principal direction and the longitudinal axis of the member;
angle between the shear-friction reinforcement and the shear plane ()
a1 = ratio of average stress in rectangular compression block to the specified concrete compressive
strength
ec = strain in concrete (in/in)
e frp = strain in FRP reinforcement (in/in)
e ut
frp
= characteristic value of the tensile failure strain of the FRP reinforcement (in/in)
eo = the concrete strain corresponding to the maximum stress of the concrete stress-strain curve (in/in)
m = coefficient of friction
 h = strain limitation coefficient that is less than unity
na = Poissons ratio of adhesive
ta = characteristic value of the limiting shear stress in the adhesive (ksi)
t int = interface shear transfer strength (ksi)
f frp = resistance factor for FRP component of resistance
1.4 DESIGN BASIS
1.4.1 Bridge elements strengthened with externally
bonded reinforcement shall be designed to achieve the
objectives of constructability, safety, and
serviceability, with regard to issues of inspectability,
economy and aesthetics, as stipulated in Article 2.5 of
AASHTO LRFD.
1.4.2 The provisions of these Guide Specifications are
limited to concretes with specified compressive
strength f c not exceeding 8 ksi.
1.4.3 The design value of the strength or the failure
strain of the FRP reinforcement used for strengthening
shall be computed in accordance with ASTM D7290
standard practice.
1.4.4 The provisions of these Guide Specifications shall
apply to bridge elements for which the factored
resistance satisfies the following requirement:
Rr hi ( DC + DW ) + (LL + IM ) (1.4.4-1)
where:
Rr = factored resistance computed in accordance with
Section 5 of AASHTO LRFD.
hi = load modifier specified in Article 1.3.2 of
AASHTO LRFD, for this guide spec, hi = 1.0
DC = force effects due to component and attachments
DW = force effects due to wearing surfaces and
utilities
LL = force effects due to live loads
IM = force effects due to dynamic load allowance
1.5 LIMIT STATES
Structural members shall satisfy Eq. 1.3.2.1-1 of
AASHTO LRFD, for each limit state, unless otherwise
C1.4
The resistance criteria in these Guide Specifications
were developed using modern principles of structural
reliability analysis, which are consistent with those on
which AASHTO LRFD are based. Structural reliability
analysis takes the uncertainties in concrete, steel and
FRP material strengths and stiffnesses into account
using rational statistical models of these key
engineering parameters. The criteria for checking
safety and serviceability of structural members and
components that have been strengthened with
externally bonded FRP reinforcement are based on a
target reliability index, , equal to 3.5 under inventory
loading, which was the target value assumed in the
development of AASHTO LRFD. The factored
resistance and factored loads used in these checks are
consistent with those found in customary engineering
practice to facilitate their use in practice and to
minimize the likelihood of misinterpretation.
These Guide Specifications are intended to apply only
to bridge structural members and components that have
minimum strength requirements established in Article
1.4.4 prior to strengthening by externally bonded FRP
reinforcement. If such a minimum cannot be shown by
analysis or test to exist, the behavior of the
strengthened member will depend almost entirely on
the performance of the FRP reinforcement and if the
field application of the FRP is deficient or if the bridge
is accidentally overloaded, damage or failure may
occur without warning. The limitation on strength prior
to strengthening is intended to minimize the likelihood
of occurrence of such damage or failure.
C1.5
These Guide Specifications apply to Strength limit
states I and II, Serviceability limit states I, III and IV,
specified
The load factor, g i , in Eq. 1.3.2.1-1 of AASHTO
LRFD shall be as defined in AASHTO LRFD Tables
3.4.1-1, 3.4.1-2 and 3.4.1-3. The resistance factors, ,
are defined in Chapters 3, 4 and 5 of these Guide
Specifications.
1.5.1 Service Limit States
Structural members shall satisfy AASHTO LRFD Eq.
1.3.2.1-1 for the applicable combinations of factored
force effects as specified at each of the following
service limit states:
Service I - Load combination relating to the normal
operational use of the bridge with a 55 mph wind and
all loads taken at their nominal values.
Service III - Load combination for longitudinal
analysis relating to tension in prestressed concrete
superstructures with the objective of crack control and
to principal tension in the webs of segmental concrete
girders
Service IV - Load combination relating only to tension
in prestressed concrete columns with the objective of
crack control.
1.5.2 Strength Limit States
Structural members of a bridge shall satisfy AASHTO
LRFD Eq. 1.3.2.1-1 for the applicable combinations of
factored extreme force effects, specified as follows:
Strength I - Basic load combination relating to the
normal random vehicular use of the bridge without
wind.
Strength II - Load combination relating to the use of
the bridge by Owner-specified special design vehicles,
evaluation permit vehicles, or both without wind.
Strength III - Load combination relating to the bridge
Extreme Event limit states I and II, and the Fatigue
limit state, as defined in Article 3.4 of AASHTO
LRFD.
C1.5.1
Service limit states customarily are defined by
restrictions on stress, deformation, and crack width
under regular service conditions. Compression in
prestressed concrete components and tension in
prestressed bent caps are investigated using the Service
I load combination. The Service III limit-state load
combination is used to investigate tensile stresses in
prestressed concrete components.
The AASHTO LRFD Service II limit state load
combination is not applicable to concrete bridge
structures reinforced with FRP systems as it is only
applied to steel structures.
The live load specified in AASHTO LRFD reflects
current exclusion weight limits mandated by various
jurisdictions. Vehicles permitted under these limits
were in service for many years prior to 1993. For
longitudinal loading, there is no nationwide evidence
that these vehicles have caused cracking in existing
prestressed concrete components. The 0.80 factor on
live load in the Service III combination reflects the fact
that the event is expected to occur about once a year for
bridges with two traffic lanes, less often for bridges
with more than two traffic lanes, and about once a day
for bridges with a single traffic lane. The Service I
limit state load combination should be used for
checking tension related to transverse analysis of
concrete segmental girders.
C1.5.2
Design for strength limit states ensures that local and
global strength and stability are provided to resist the
specified load combinations that a bridge is expected to
experience in its design life. The background for the
load combination requirements in AASHTO LRFD is
presented in Nowak (1993).
exposed to wind velocity exceeding 55 mph.

Strength IV - Load combination relating to very high

dead load to live load force effect ratios.

Strength V - Load combination relating to normal

random vehicular use of the bridge with wind of 55
mph velocity.

1.5.3 Extreme-event Limit States C.1.5.3

Structural members of a bridge shall satisfy AASHTO Consideration of extreme-event limit states is aimed at
LRFD Eq. 1.3.2.1-1 for the applicable combinations of ensuring that the bridge structure survives a major
factored extreme force effects as specified at each of earthquake or flood, collision from a vessel or heavy
the following: vehicle, or ice flow, or possible foundation scour.

Extreme-event I - Load combination including

earthquake.

Extreme-event II - Load combination relating to ice

load, collision by vessels and vehicles, and certain
hydraulic events with a reduced live load other than
that which is part of the vehicular collision load.

1.5.4 Fatigue Limit State C1.5.4

Structural members, connections and components of a
bridge shall satisfy AASHTO LRFD Eq. 1.3.2.1-1 for
the Fatigue I limit-state load combination, the load
combination related to infinite load-induced fatigue
life.
The fatigue limit states place restrictions on stress
range resulting from a single design truck occurring at
the number of expected stress range cycles. Concrete
bridge structures are designed to provide a theoretically
infinite fatigue design life.

The load factor for the Fatigue I load combination

applied to a single design truck having the axle spacing
specified in AASHTO LRFD Article 3.6.1.4.1 reflects
load levels found to be representative of the maximum
stress range of the truck population. The Fatigue II
limit-state load combination is not applicable to
concrete bridge structures reinforced with FRP systems
as it is not generally applicable to concrete components
and connections.

1.6 LOADS AND LOAD COMBINATIONS

1.6.1 Loads C1.6.1

The loads defined in AASHTO LRFD Article 3.3.2 and
characterized in AASHTO LRFD Article 3.6 through
3.15 shall be applied for designing reinforced concrete
and prestressed highway bridge members strengthened
with externally bonded FRP reinforcement
The loads required for the design and evaluation of
concrete bridge structures reinforced with FRP systems
are classified in AASHTO LRFD as permanent and
transient loads.

1.6.2 Load Combinations

The load combination requirements shall be

determined in accordance with Article 3.4 of AASHTO
LRFD
1.6.3 Bridge Load Ratings and Evaluation
Bridge evaluations shall be performed using the
evaluation criteria stipulated in the AASHTO Manual
for Bridge Evaluation, Second Edition, 2011 (MBE).
Eq. 6A.4.2.1-1 of the MBE shall be used in
determining the load rating of each component and
connection subjected to a single force effect (i.e., axial
force, flexure, or shear).
The load rating shall be carried out at each applicable
limit state with the lowest value determining the
controlling rating factor. Limit states and load factors
for load rating shall be selected from MBE Table
6A.4.2.2-1.Interaction of load effects (i.e., axial-
bending interaction or shear-bending interaction), shall
be considered, as provided in the MBE under the
sections on resistance of structures, in developing the
rating.
C1.6.3
Bridge load ratings are performed for specific
purposes, such as: NBI and BMS reporting, local
planning and programming, determining load posting
or bridge strengthening needs, and overload permit
review. Live load models, evaluation criteria, and
evaluation procedures are selected based upon the
intended use of the load rating results. Live-load
models used in evaluation are comprised of the design
live load, legal loads, and permit loads.
Strength is the primary limit state for load rating;
service and fatigue limit states are selectively applied
in accordance with the provisions of the MBE. Live-
load models for load rating include:
Design Load: HL-93 Design Load per AASHTO
LRFD
Legal Loads: AASHTO Legal loads, as specified in
MBE Article 6A.4.4.2.1.a, and (2) The Notional Rating
Load as specified in MBE Article 6A.4.4.2.1.b or State
legal loads.
Permit Load: Actual Permit Truck
Bridges that do not satisfy the HL-93 design load check
should be evaluated for legal loads in accordance with
the provisions of MBE Article 6A.4.5 to determine the
need for load posting or strengthening. Legal loads for
ratings given in MBE Article 6A.4.4.2.1.a that model
routine commercial traffic are the same family of three
AASHTO trucks (Type 3, Type 3S2, and Type 3-3)
used in current and previous AASHTO evaluation
Manuals. The single-unit legal load models given in
MBE Article 6A.4.4.2.1.b represent the increasing
presence of Formula B multi-axle specialized hauling
vehicles in the traffic stream in many States.
1.7 REFERENCES

ACI (2005). Building code requirements for structural concrete (ACI Standard 318-05). American Concrete Institute,
Farmington Hills, MI.

ACI 440.2-R02 (2002) Guide for the design and construction for externally bonded FRP systems for strengthening
concrete structures, Evaluation of Concrete Structures Prior to Rehabilitation," American Concrete Institute,
Farmington Hills, Michigan.

ACI 440.2-R08 (2008) Guide for the design and construction for externally bonded FRP systems for strengthening
concrete structures, Evaluation of Concrete Structures Prior to Rehabilitation," American Concrete Institute,
Farmington Hills, Michigan.

AFGC (2003). Rparation et renforcement des structures en bton au moyen des matriaux composites, Association
Franaise de Gnie Civil, Dcembre, 92225 Bagneux Cedex, France.

CNR-DT 200 (2006), Guide for the Design and Construction of Externally Bonded FRP Systems for Strengthening
Existing Structures, Italian Advisory Committee on Technical Recommendations for construction. Rome, Italy.

fib (2001) technical report bulletin 14, Externally bonded FRP reinforcement for RC structures, published in Europe
(fib, CEB-FIP, 2001).

JSCE (2001). , Recommendations for Upgrading of Concrete Structures with Use of Continuous Fiber Sheets, Japan
Society of Civil Engineers.

ISIS (2001). ISIS Canada Design Manuals, Strengthening Reinforced Concrete Structures with Externally-Bonded
Fiber-Reinforced Polymers, Winnipeg, Manitoba.

MBE (2008). The Manual for Bridge Evaluation, AASHTO, Washington, D.C.

NCHRP Report 609 (2008). Recommended construction specifications and process control manual for repair and
retrofit of concrete structures using bonded FRP composites, Transportation Research Board, Washington, D.C.

NCHRP Report 655 (2010) Recommended Guide Specification for the Design of Externally Bonded FRP Systems for
Repair and Strengthening of Concrete Bridge Elements, Transportation Research Board, Washington, D.C.

NCHRP Report 678 (2011) Design of FRP Systems for Strengthening Concrete Girders in Shear, Transportation
Research Board, Washington, D.C.

Nowak, A. S. (1993). Live load model for highway bridges, Journal of Structural Safety, Volume 13, No. 1 and 2,
December, pp.53-66.
SECTION 2: MATERIAL REQUIREMENTS

2.1 SCOPE

This section defines the requirements for externally

bonded FRP composite material systems intended for
use for repair and strengthening of concrete bridge
elements.

2.2 MATERIAL REQUIREMENTS

2.2.1 The contractor shall submit for approval evidence

of acceptable quality control procedures followed in the
manufacture of the composite reinforcement system.
The quality control procedure shall include, but not be
limited to, specifications for raw material procurement,
the quality standards for the final product, in-process
inspection and control procedures, test methods,
sampling plans, criteria for acceptance or rejection, and
record keeping standards.

2.2.2 The contractor shall furnish information

describing the fiber, matrix, and adhesive systems
intended for use as reinforcing materials that is
sufficient to define their engineering properties.
Descriptions of the fiber system shall include the fiber
type, percent of fiber orientation in each direction, and
fiber surface treatments. Where required by the
Engineer of Record, the matrix and the adhesive shall be
identified by their commercial names and the
commercial names of each of their components, along
with their weight fractions with respect to the resin
system.

2.2.3 The contractor shall submit for approval test

results that demonstrate that constituent materials and
the composite system are in conformance with the
physical and mechanical property values stipulated by
the Engineer of Record. These tests shall be conducted
by a testing laboratory approved by the Engineer of
Record. For each property value, the batches from
which test specimens were drawn shall be identified and
the number of tested specimens from each batch, the
mean value, the minimum value, the maximum value,
and the coefficient of variation shall be reported. The
minimum number of tested samples shall be 10.

2.2.4 When cured under conditions identical to those of

the intended use, the composite material system as well
as the adhesive system, if used, shall conform to the
following requirements:

2.2.4.1 The characteristic value of the glass transition C2.2.4.1 The glass transition temperature, Tg, is the
temperature of the composite system, determined in approximate temperature value or temperature range at
accordance with ASTM E1640, shall be at least 40oF which the matrix changes from a glassy to a rubbery
higher than the maximum design temperature, TMaxDesign ,
defined in Section 3.12.2.2 of AASHTO
AASHT LRFD.
2.2.4.2 The characteristic value of the tensile failure
strain in the direction corresponding to the highest
percentage of fibers shall not be less than 1%, when the
tension test is conducted according to ASTM 3039.
2.2.4.3 The average value and coefficient of variation of
the moisture equilibrium content determined in
accordance with ASTM D 5229/D 5229M shall not be
greater than 2% and 10%, respectively. A minimum
sample size of 10 shall be used in the calculation of
these maximum values.
2.2.4.4 After conditioning in the following
environments, the characteristic value
v of the glass
transition temperature determined ed in accordance with
ASTM E1640 and the characteristic value of the tensile
strain, determined in accordance with ASTM D3039, of
the composite in the direction of interest shall retain
85% of the values established
ablished in Art. 2.2.4.1 and 2.2.4.2,
respectively.
A. Water: Samples shall be immersed in distilled water
having a temperature of 100 3F (38
(3 2C) and tested
after 1000, 3000, and 10000 hours of exposure.
B. Alternating ultraviolet light and condensat
condensation
humidity: Samples shall be conditioned in an apparatus
state. Above Tg, the composite softens and loses its
mechanical properties, as illustrated in Figure C2-1.
C2 In
addition, it is to be noted that Tg decreases as the
moisture content in the composite increases.
Figure C2-1, Effect of temperature on the properties of
polymer composite materials (Zureick and Kahn, 2001)
C2.2.4.3 The diffusion of moisture into organic
polymers results in pronounced changes in mechanical,
chemical, and thermophysical properties of practically
all composite reinforcing systems. All organic matrix
systems and organic reinforcing
reinforc fibers absorb moisture
to a certain degree. While both glass and carbon fibers
are considered to be impervious to moisture absorption,
aramid fibers absorb more moisture than many matrix
systems. In all cases when moisture migrates through
the matrix system and ultimately reaches the fiber-
fiber
matrix interface, adhesion of the matrix system to the
fibers become weak and the structural integrity of the
composite system degrades.
C2.2.4.4. The physical and mechanical properties of
FRP materials and of the concrete structure reinforced
with an externally bondednded reinforced system are
sensitive to various environmental conditions that can
affect one or more of the followings:
Chemical and/or physical changes in the
polymeric matrix.
Loss of bond at the fiber/matrix interface and
at the FRP-concrete
concrete interface.
Strength and stiffness degradation of the
reinforcing fibers.
The durability requirements in Article 2.2.4.4 are based
under Cycle 1 -UV exposure condition according to on those developed for CALTRANS (Steckel et al.,
ASTM G154 Standard Practice. Samples shall be tested 1999a, 1999b; Hawkins et al., 1999) and for GDOT
within two hours after removal from the apparatus. (Zureick, 2002).

C. Alkali: The sample shall be immersed in a saturated
solution of calcium hydroxide (pH ~11) at ambient
temperature of 73 3oF (23 2oC) for 1000, 3000, and
10000 hours prior to testing. The pH level shall be
monitored and the solution shall be maintained as
needed.
Cycle No 1 UV exposure condition uses UVA-340
lamps that simulate direct solar radiation and have
negligible UV output below 300nm, considered to be
the cut-on wavelength for terrestrial sunlight

D. Freeze-thaw: Composite samples shall be exposed to

100 repeated cycles of freezing and thawing in an
apparatus meeting the requirements of ASTM C666.

2.2.5 Where impact tolerance is stipulated by the

Engineer, the stipulated impact tolerance shall be
determined by ASTM D7136.

2.2.6 Adhesive: when adhesive material is used to bond

the FRP reinforcement to the concrete surface, the
following requirements shall be met:

2.2.6.1 After conditioning in the environments

stipulated in Article 2.2.4.4 A-D, the characteristic value
of the glass transition temperature of the adhesive
material determined in accordance with ASTM E1640,
shall be at least 40oF higher than the maximum design
temperature, TMaxDesign , defined in Section 3.12.2.2 of
AASHTO LRFD.

2.2.6.2 After conditioning in the environments C.2.2.6.2 The bond strength limit of 0.065 fc' is based
stipulated in Article 2.2.4.4 A-D, the bond strength
(ksi), determined by tests specified by the Engineer of on tests conducted by Naaman (1999) on reinforced
concrete beams strengthened with externally bonded
Record, shall be at least 200 psi or 0.065 f c' which FRP reinforcement.
ever is greater, where fc' (ksi) is the specified
compression strength of the concrete.
2.3 REFERENCES

Naaman, A. E. (1999). Repair and strengthening of reinforced concrete beams using CFRP laminates, Volumes. 1- 7,
Michigan Department of Transportation, Report No. RC-1372.

Hawkins, G. F. Johnson, and Nokes, (1999). Qualifications for Seismic Retrofitting of Bridge Columns Using
Composites, Volume 3: Composite Properties Characterization, Prepared for the California Department of
Transportation, Contract No. SA0A011, The Aerospace Corporation, El Segundo California.

Steckel, G.L., Bauer, J.L., Hawkins, G.F. and Vanik, (1999). Qualifications for Seismic Retrofitting of Bridge
Columns Using Composites, Volume 2: Composite Properties Characterization, Prepared for the California
Department of Transportation, Contract No. SA0A011, The Aerospace Corporation, El Segundo California

Steckel, G. L., Hawkins, G. F. and Bauer, J. L. (1999). Qualifications for Seismic Retrofitting of Bridge Columns
Using Composites, Volume 1: Composite Properties Characterization, Prepared for the California Department of
Transportation, Contract No. SA0A011, The Aerospace Corporation, El Segundo California.

Zureick, A. and Kahn, L. (2001). Rehabilitation of Reinforced Concrete Structures Using Fiber-Reinforced Polymer
Composites, ASM Handbook, Volume 21, Composites, pp. 906-913.

Zureick, A. (2002). Proposed Specifications-Polymeric Composite Materials for Rehabilitating Concrete Structures,
Report Prepared for the Georgia Department of Transportation, Atlanta, Georgia.
SECTION 3: MEMBERS UNDER FLEXURE

3.1 GENERAL REQUIREMENTS

The factored resistance of structural members

subjected to flexure shall equal or exceed the required
strength calculated for the factored loads and forces in
combinations stipulated by these Specifications.

Except where specifically provided below, all

provisions of AASHTO LRFD, Article 5.7.3, shall
apply.

3.2 DESIGN ASSUMPTIONS C3.2

The calculation of the strength of reinforced concrete The strength in flexure of a reinforced concrete
members externally reinforced with FRP materials member that has been additionally reinforced by an
shall be based on the following assumptions: externally bonded FRP plate is derived from the classic
Bernoulli-Navier hypothesis that plane sections remain
The distribution of strains over the depth of plane and perpendicular to the neutral axis during
the member is linear and conditions of force flexure. The stresses on the section can be determined
equilibrium and strain compatibility are from the constitutive relations for the concrete,
satisfied. reinforcing steel and FRP reinforcement, and the
flexural strength at any section is determined from
Perfect bond exists between the reinforcing
requirements for axial force and moment equilibrium at
steel and the concrete, and the FRP
that section.
reinforcement and the concrete.
The contribution of tension stresses in the
concrete to flexural strength is neglected.
The stress-strain behavior for FRP
reinforcement is linear-elastic to the point of
failure.
The stress-strain behavior of steel
reinforcement is bilinear, with elastic behavior
up to yielding and perfectly plastic behavior
thereafter.
The maximum usable compression strain in
the concrete is equal to 0.003 (in/in).
The maximum usable strain at the
FRP/concrete interface is 0.005 (in/in).
When concrete compressive strain is 0.003 (in/in) When the compression strain in the extreme fiber of the
under conditions of force equilibrium, it is permitted to concrete is less than 0.003 (in/in) at force equilibrium,
model the distribution of concrete stress in the Whitney compression stress block may no longer
compression as having a uniform stress of 0.85 fc over describe the compression resultant in the concrete
accurately, and a more exact representation of the
a depth a = b1c , in which c = depth to the neutral axis
distribution of concrete stress in the compression zone
from the compression face of the beam and 1 = stress is required. Eq. (3.2-1), which provides this
block factor specified in Article 5.7.2.2 of AASHTO representation, was presented by Desayi and Krishnan,
LRFD. S. (1964) and by Todeschini et al. (1964).
When concrete compressive strain is less than 0.003
(in/in) under conditions of force equilibrium, the
concrete compression stress distribution shall be
modeled as parabolic according to the following
equation:

2 ( 0.9 fc) (e c e o )
fc = (3.2-1)
1+ (e c eo )
2

fc
where: e o = 1.71 (3.2-2)
Ec

fc = stress in concrete (ksi) at strain e c

e c = strain in concrete (in/in)

fc = the 28 - day compression strength of the concrete

(ksi)

e o = the concrete strain (in/in) corresponding to the

maximum stress of the concrete stress-strain curve

Ec = the modulus of elasticity (ksi) of the concrete

specified in Section 5.4.2.4 of AASHTO LRFD

3.3 FATIGUE LIMIT STATES C3.3

3.3.1 Subjected to the fatigue load combination By limiting the maximum strain in the concrete to that
specified in Article 3.4.1 of AASHTO LRFD, the specified in Eq. 3.3-1, the stress range in the concrete
maximum compression strain in the concrete, e c , the will be kept within 0.40 fc' . Limiting the strain of the
strain in the steel reinforcement, e s , and the strain in steel reinforcement under service load to 80% of the
the FRP reinforcement, e frp , shall meet the following yield strain of the steel is equivalent to the
recommendation of ACI Committee 440, where the
requirements stress in the reinforcing steel under service load is
limited to 80% of the yield stress of the steel; this
f c' recommendation is based on the analytical work of El-
e c 0.36 (3.3-1)
Ec Tawil et al. (2001).

Strain limits on the FRP reinforcement are placed to

e s 0.8e y (3.3-2)
avoid creep-rupture of the reinforcement. Polymer
composites reinforced with carbon fibers are less
e frp he ufrp 0.005 (3.3-3) susceptible to creep rupture than those reinforced with
glass or aramid fibers. The recommended strain
where reduction factors of 0.55, 0.3, and 0.2 are based on the
recommendation of the ACI 440 Committee (ACI
e ufrp = characteristic value of the tensile failure strain of 440.2R-08).
the FRP reinforcement when tested in accordance with
As the design is often governed by service limit state,
ASTM D3039 (in/in)
FRP strains at Service I load combination are
sufficiently low that creep rupture of the FRP is
h = strain limitation coefficient that is less than unity.
typically not of concern.
The Engineer of Record shall stipulate the value of h
based on experimental data for the materials specified,
and this value shall be provided in the contract
documents. In the absence of such data, a value of h =
0.55, 0.3, and 0.2 shall be used for carbon, aramid, and
glass fiber reinforcement, respectively.

3.4 STRENGTH LIMIT STATES

3.4.1 Factored Flexural Resistance

3.4.1.1 Rectangular Sections C3.4.1

The factored resistance, M r , of a steel-reinforced
concrete rectangular section strengthened with FRP
reinforcement externally bonded to the beam tension
surface shall be taken as
M r As f s d s k2c As' f s' k2c d s'
frpTfrp h k2c
The factored resistance is in line with the design
strength determination in accordance with Article
5.7.3.2 of AASHTO LRFD, and is written so that the
design strength for a reinforced concrete flexural
member is simply augmented by the contribution of the
externally bonded FRP reinforcement. This format
ensures that when the FRP reinforcement is slight, the
design strength approaches that of a flexural member
without FRP reinforcement.

(3.4.1.1-1) It should be noted that the tensile area of the wet-layup

system is typically based upon the net-area of the
in which: fibers; whereas, the properties for a pre-cured system
are based upon the gross-laminate area.
Tfrp = b frp N b (3.4.1.1-
2)

e e
2 c - arctan c
e o e o
k2 = 1- (3.4.1.1-3)
e
2

b2 c
eo

2
e
Ln 1+ c
eo

b2 = (3.4.1.1-4)
e
c
eo

where:

As = area of nonprestressed steel tension

reinforcement

As' = area of steel compression reinforcement (in2)

fs = stress in the steel tension reinforcement at

development of nominal flexural resistance (ksi)

f s' = stress in the steel compression reinforcement

at development of nominal flexural resistance (ksi)
c = depth of the concrete compression zone (in)

d s = distance from extreme compression surface to the

centroid of nonprestressed tension reinforcement (in.)

h = depth of section (in)

Tfrp = tension force in the FRP reinforcement (kips)

= Resistance Factor defined in Article 5.5.4.2 of

AASHTO LRFD

f frp = FRP resistance factor equal to 0.85

k2 = multiplier for locating resultant of the compression

force in the concrete

b frp = width of the FRP reinforcement (in.)

N b = FRP reinforcement tensile strength per unit width,

corresponding to 0.5% a strain value of 0.005 in the
FRP reinforcement when subjected to tension in
accordance with ASTM D3039.

3.4.1.2 Flanged Sections C3.4.1.2

For flanged sections subjected to flexure about one axis
where the neutral axis, based upon the stress
distribution specified in Article 3.2, lies within the
flange, the factored resistance, M r , shall be computed
in accordance with Article 3.4.1.1. When the neutral
axis falls outside the flange, the factored flexural
resistance shall be determined by an analysis based on
the assumptions specified in Article 3.2.
For most practical cases involving flanged sections
strengthened externally with bonded FRP
reinforcement applied to the tension surface, the depth
of the neutral axis falls within the flange. When the
neutral axis falls below the compression flange, the
compression force exerted in the concrete is the sum of
two components, one of which corresponds to the
flange and one corresponds to the portion of the web
under compression. Due to the nature of the assumed
nonlinear stress-strain relationship of Eq. 3.2-1, the
determination of the compressive force requires
integration of the stress-strain function expressed in
Eq. 3.2-1 over the area of the cross-section.

3.4.1.3 Other Cross-Sections

For cross-sections other than rectangular or flanged

sections, the factored flexural resistance, M r , shall be
determined by an analysis based on the assumptions
specified in Article 3.2.

3.4.1.4 Prestressed Sections

For rectangular and nonrectangular prestressed

concrete sections subjected to flexure about one axis,
the factored flexural resistance shall be determined by
an analysis based on the assumptions specified in
Article 3.2

3.4.2 Ductility requirements C3.4.2

The strain developed in the FRP reinforcement at the
ultimate limit state defined by Eq. (3.4.1.1-1) shall be
equal to or greater than 2.5 times the strain in the FRP
reinforcement at the point where the steel tension
reinforcement yields.
This provision ensures that the tension steel
reinforcement yields before the point of incipient
debonding of the externally bonded FRP
reinforcement, thereby enabling the development of a
ductile mode of flexural failure.

3.4.3 Detailing requirements C3.4.3

Flexural members shall be detailed to permit the
development of the factored resistance defined by Eq.
(3.4.1.1-1).
The externally bonded FRP reinforcement must be
installed and detailed in such a manner that the
assumptions in Section 3.2 are valid and the flexural
capacity defined in Eq. (3.4.1.1-1) can be fully
developed.

3.4.3.1 Development length C3.4.3.1

The tension development length, Ld, shall be taken as The minimum development length is required to allow
Tfrp the full tension strength of the FRP reinforcement to be
Ld (3.4.3.1-1) developed in the region of maximum moment. The
t int b frp
interface shear transfer strength limit of
where: t int = 0.065 f c is based on the recommendation of
Naaman and Lopez (1999) and Naaman et al. (1999)
TFRP = tensile force (kips) in the FRP reinforcement from tests conducted on uncracked and precracked
corresponding to an FRP strain of 0.005 and reinforced concrete beams externally bonded with FRP
reinforcement and subjected to bending with 300
t int = 0.065 f c is the interface shear transfer strength, freeze-thaw cycles. The limit also represents a lower
bound for experimental data conducted on short-term
(ksi). direct tension tests of FRP reinforcement bonded to
concrete surfaces (Haynes, 1997; Binzindavyi and
Neale, 1999).

3.4.3.2 Reinforcement End-Termination Peeling C3.4.3.2

The peel stress at the point of end-termination of The end termination of an externally bonded
externally bonded reinforcement shall meet the reinforcement system, when subjected to combined
following requirement: shear and flexure, may separate in the form of
debonding in three different modes: critical diagonal
f peel 0.065 fc' (3.4.3.2-1) crack debonding with (Yao and Teng 2007) or without
(Oehlers and Seracino, 2004) concrete cover
in which: separation; concrete cover separation (Teng et al.,
2002); and plate end interfacial debonding (Teng et al.,
2002).
3E t
1/4

f peel = t av a frp (3.4.3.2-2)

E frp ta Critical diagonal crack debonding may occur where the
FRP end is located in a zone of high shear force and
the amount of steel shear reinforcement is limited. In
G t h - y
( )
1/2
such a case a major diagonal shear crack forms and
t av = Vu + a
M u frp intersects the FRP, and then propagates towards the
frp frp a
E t t IT
end. This failure mode is suppressed if the shear
 (3.4.3.2-3)
and where:
h = overall thickness or depth of a member (in.)
y = distance from the extreme compression surface to
the neutral axis of a transformed section, neglecting
any contribution of concrete in tension (in).
IT = moment of inertia of an equivalent FRP
transformed section, neglecting any contribution of
concrete in tension (in4)
t a = thickness of the adhesive layer (in)
t frp = thickness of the FRP reinforcement (in)
Ea = Youngs modulus of adhesive (ksi)
Ga = characteristic value of the shear modulus of
adhesive when tested in accordance with ASTM D5656
(ksi).
n a = Poissons ratio of adhesive, taken as equal to 0.35
t a = characteristic value of the limiting shear stress in
the adhesive (ksi), determined according to ASTM D
5656. In the absence of experimental data, a value of
5 (ksi) can be used.
Vu = factored shear force at the reinforcement end-
termination (kips)
M u = factored moment at the reinforcement end-
termination (kip-in)
strength of the strengthened member remains higher
than the flexural strength.
In beams with heavy steel shear reinforcement,
multiple diagonal cracks of smaller widths instead of a
single major shear crack dominate the behavior, so
concrete cover separation may take over as the
controlling debonding failure mode. Failure of the
concrete cover is initiated by a crack near the FRP end
due to the stress concentration at that point. The crack
then propagates to and then along the level of steel
tension reinforcement. This mode of failure has been
demonstrated experimentally for beams with externally
bonded steel plates (Jones et al., 1988; Oehlers and
Moran, 1990) and FRP reinforcement (Malek et al.,
1998; Lopez and Naaman, 2003; Yao and Teng 2007).
Plate-end interfacial debonding is also initiated by high
interfacial shear and normal stresses near the end of the
FRP that exceed the strength of the weakest element,
generally the concrete. Debonding in this case
propagates from the end of the FRP towards the
middle, near the FRP-concrete interface. Note that this
failure mode is only likely to occur when the FRP is
significantly narrower than the beam section.
In summary, provided that shear failure is suppressed
(through shear strengthening, if needed), stress
concentrations near the FRP reinforcement end may
result in debonding through the concrete layer near the
level of the longitudinal steel (or, rarely, near the FRP-
concrete interface).
Although a wide range of predictive models that
include numerical, fracture mechanics, data-fitting, and
strength of material-based methods have been
developed to address this complex mode of failure
(Yao, 2004), the equations presented in 3.4.3.2 are
based on the approximate analysis of Roberts (1989),
due to its simplicity for design purposes.
At present, there is no standard test method for
determining the peel strength between an FRP
reinforcement system and a concrete surface. Until
such a test method is developed, the ASTM Standard
Test Method D 3167 is recommended for determining
the peel strength within the adhesive layer. ASTM
D3167 is used for determining the peel resistance of
adhesive bonds between one rigid adhered and one
flexible adhered. For cases in which the peeling occurs
within the concrete layer, it is recommended that the
peeling strength be limited to 0.065 f c' .
3.5 REFERENCES

ACI 440.2R-08 (2008). Guide for the Design and Construction of Externally Bonded FRP Systems for Strengthening
Concrete Structures, American Concrete Institute, Farmington Hills, Michigan.

Bizindavyi, L. and Neale K. W. (1999). Transfer Length and Bond Strengths for Composite Bonded to Concrete,
ASCE Journal of Composites for Construction, Vol. 3, No. 4, November, pp. 153-160.

Desayi, P. and Krishnan, S. (1964). Equation for the stress-strain curve of concrete. Journal the American Concrete
Institute 61(3): 345-350.

El-Tawil, S., Ogunc, C., Okeil, A. M., and Shahawy, M. (2001) Static and Fatigue Analyses of RC Beams
Strengthened with CFRP Laminates, Journal of Composites for Construction, ASCE, Vol. 5, No. 4, pp. 258-267.

Jones, R., Swamy, R. N., and Charif, A. (1988). Plate separation and anchorage of reinforced concrete beams
strengthened with epoxy-bonded steel plates, The Structural Engineer, Volume 66, No. 5, pp. 85-94.

Haynes, L. (1997). An investigation of bond between concrete and externally bonded carbon fiber reinforced plastic
plates, M.S. special report submitted to the faculty of the School of Civil Engineering in partial fulfillment of the
requirements of the degree of Master of Science in Civil Engineering, Georgia Institute of Technology, Atlanta,
Georgia.

Lopez, M. M., Naaman A.E. (2003). Concrete Cover Failure or Tooth type failure on RC beams strengthened with
FRP laminates, Proceedings of the Sixth International Symposium on Fiber Reinforced Polymer (FRP) for Reinforced
Concrete Structures (FRPRCS-6). pp. 317-326.

Malek, A.; Saadatmanesh, H.; and Ehsani, M. (1998). Prediction of Failure Load of R/C Beams Strengthened with
FRP Plate Due to Stress Concentrations at the Plate End, ACI Structural Journal, Volume 95, No. 1, pp. 142-152.

Naaman, A. E. and Lopez, M. (1999). Repair and strengthening of reinforced concrete beams using CFRP
Laminates, Behavior of beams subjected to freeze-thaw cycles, Michigan Department of Transportation, Report No.
RC-1372., April

Naaman, A. E. , Lopez, M., and Pinkerton, L. (1999). Repair and strengthening of reinforced concrete beams using
CFRP Laminates , Behavior of beams under Cyclic loading at low temperature, Michigan Department of
Transportation, Report No. RC-1372., April.

Oehlers, D. J. and Moran, J. P. (1990). "Premature failure of externally plated reinforced concrete beams." Journal of
the Structural Division of the American Society of Civil Engineers, Vol. 116, No. 4, pp. 978-995.

Oehlers D. J. and Seracino R. (2004). Design of FRP and Steel Plated RC Structures, Elsevier Science Publishers,
Etc., Oxford, UK.

Roberts, T. M. (1989). Approximate analysis of shear and normal stress concentrations in the adhesive layer of plated
RC beams, The Structural Engineer, Volume 67, No.12/20 June, pp. 229-233.

Teng, T. G., Chen, J. F., Smith, S. T., and Lam, L. (2002). FRP Strengthened RC Structures, John Wiley & Sons, Ltd,
Chichester, UK.

Todeschini, C.E., Bianchini, A.C. and Kesler, C.E. (1964). Behavior of concrete columns reinforced with high
strength steels. ACI Journal 61(6):701-716.

Yao, J. (2004). Debonding failures in RC beams and slabs strengthened with FRP plates, Ph.D. Thesis, The Hong
Kong Polytechnic University.
Yao J. and Teng, J. G. (2007). Plate end debonding in FRP-plated RC beams-I-Experiments, Engineering
Structures, 29, pp. 2457-2471.
SECTION 4: MEMBERS UNDER SHEAR AND TORSION
4.1 GENERAL REQUIREMENTS
The factored shear and torsion resistance of structural
members at all sections shall equal or exceed the
required strength in shear or in torsion calculated for
the factored loads and forces in combinations stipulated
in Article 3.4 of AASHTO LRFD.
Except where specifically provided below, all
provisions of AASHTO LRFD shall apply.
4.2 STRENGTHENING SCHEMES
FRP shear reinforcement should be attached to a beam,
as shown in Figure 4.2-1 with (a) side bonding, in
which the FRP is only bonded to the sides, (b) U-wrap,
in which FRP U-jackets are bonded to both the sides
and soffit, with or without anchors, and (c) complete
wrapping, in which the FRP is wrapped around the
entire cross section.
(a) (b) (c)
Figure 4.2-1 Strengthening Scheme: Cross-Sectional
View - (a) Side bonding, (b) U-wrap, and (c) Complete
wrap
For all wrapping schemes, the transverse FRP
reinforcement shall be applied symmetrically on both
sides of the element as discrete strips, or continuously
along the portion of the member length to be
strengthened. The fibers of the FRP may also be
oriented at various angles to meet a range of
strengthening requirements as shown in Figure 4.2-2.
C4.1
The shear strengthening guidelines provided in this
section are the recommendations from NCHRP Report
655 and NCHRP Report 678.
The provisions for strengthening reinforced concrete
structural members and components for shear and torsion
using externally bonded FRP reinforcement have been
developed with the assumption that all design
requirements for shear and torsion in Article 5.8 of
AASHTO LRFD shall apply, except as specifically
provided for in Section 4. Any duplication of provisions
in these two documents is intended solely to facilitate the
use and interpretation of provisions in Section 4.
C4.2
Typical FRP strengthening schemes for beams and
columns are summarized as follows. Beams are typically
limited to U-wrap and side bonding applications since the
integral slab makes it impractical to completely wrap
such members:
Side bonding (Fig. 4.2-1a) is the least effective FRP shear
reinforcement scheme due to premature debonding under
shear loading and should be avoided if possible. Side
bonding does not allow for the development of the shear-
resisting mechanism based on a parallel chord truss
model that was first proposed by Ritter (1899), due to the
lack of tensile diagonals.
U-jacketing (Fig. 4.2-1b) is the most common externally
bonded shear strengthening method for reinforced
concrete beams and girders. The key drawback of this
system is the possibility of premature debonding of the
FRP, which may reduce its effectiveness. Despite this
drawback, the system is quite popular in practice, due to
its simplicity.
U-jacketing combined with anchorage (Fig. C4.2-1) aims
to increase the effectiveness of FRP by anchoring the
fibers, preferably, in the compression zone. Properly
designed anchors may result in the fibers reaching their
tensile capacity, permitting the jacket to behave as if it
were completely wrapped.
 Center-to-Center Spacing of FRP Strip (sf)
Width of FRP Strips (wf)
(a)
Center-to-Center Spacing of FRP Strip (sf )
Width of FRP Strips (wf )
(b)
Figure 4.2-2
2 Strengthening Scheme: Side View - (a)
Fibers at 90 Direction, and (b) Fibers at Inclined
Direction
4.3 STRENGTH IN SHEAR
The factored shear strength, Vr, of a concrete member
strengthened with an externally bonded FRP system
shall equal or exceed the required shear strength, Vu,
determined from the effect of the factored loads.
Figure C4.2-1
1 Jacketing with anchors.
Different types of anchorage systems are available for
shear strengthening with FRP. Anchoring
A externally
bonded FRP shear reinforcement helps reduce the
potential for premature failure due to debonding.
Examples of mechanical anchorage systems consisting of
FRP composite plates and concrete anchor bolts, and FRP
spikes/ties are available in
n the literature [NCHRP Report
678]. However, it should be noted that additional bonded
horizontal FRP strips can not ensure FRP rupture failure.
Thus, it is recommended that equation 4.3.2-5 be used to
calculate the FRP contribution, realizing that such
approach
pproach will result in conservative estimates.
estimates
Complete wrapping (Fig. 4.2-1c)
4.2 of the cross section is
the most effective scheme and is commonly used in
strengthening columns where there is sufficient access for
such application, as it ensures maximum straining
stra of the
fibers
C4.3
Shear strengthening of reinforced concrete members
using FRP reinforcement may be provided by bonding the
external reinforcement (typically in the form of sheets)
with the principal fiber direction as parallel as practically
possible to that of maximum
ximum principal tensile stresses, so
that the effectiveness of FRP is maximized. For the most
common case of structural members subjected to lateral
loads, in which loads are perpendicular to the member
axis (e.g. beams under gravity loads or columns under
seismic forces), the maximum principal stress trajectories
in the shear-critical
critical zones form an angle with the member
axis which may be taken roughly equal to 45o. However,
it is normally more practical to attach the external FRP
reinforcement with the pri principal fiber direction
perpendicular to the member axis.
Experimental and analytical investigations of the
behavior of reinforced concrete members strengthened in

4.3.1 Factored Strength
The factored shear strength, Vr, shall be defined as:
( )
Vr = f Vc +Vs +Vp + f frpV frp (4.3.1-1)
where:
Vc = the nominal shear strength provided by the
concrete in accordance with Articles 5.8.3.3 of
AASHTO LRFD.
Vs = the nominal shear strength provided by the
transverse steel reinforcement in accordance
with Article 5.8.3.3 of AASHTO LRFD;
V p = component of the effective prestressing force in
the direction of applied shear as specified in
Article 5.8.3.3 of AASHTO LRFD;
V frp = the nominal shear strength provided by the
externally bonded FRP system in accordance
with Article 4.3;
= Resistance Factor defined in Article 5.5.4.2 of
AASHTO LRFD
shear have revealed the following failure modes:
1 - Steel yielding followed by FRP debonding
2 - Steel yielding followed by FRP fracture
3 - FRP debonding before steel yielding
4 - Diagonal concrete crushing
Depending on the amount of usable steel shear
reinforcement in the structural element, FRP debonding
can occur either before or after steel yielding. The third
failure mode is, in fact, highly unlikely to occur if proper
detailing is provided.
Diagonal concrete crushing in the direction perpendicular
to the tension field can be suppressed by limiting the total
amount of steel and FRP reinforcement. Note that fracture
of the FRP reinforcement is highly unlikely to occur
because the strain when FRP debonds is substantially
lower than that corresponding to the FRP fracture
strength.
C4.3.1
The shear provisions in Article 4.3 draw upon the
traditional ACI approach embodied by Chapter 11 of the
ACI Standard 318-05, supplemented by the report of ACI
Committee 440.2R-08 (ACI, 2008).
The contribution of the externally bonded FRP
reinforcement to shear strength is based on fiber
orientation and an assumed crack pattern following the
formulation of Khalifa, et al (1998). Its contribution to
member shear strength may be treated analogously to the
treatment of internal steel, assuming that the FRP plate
carries only normal stresses in the principal FRP material
direction and that at the ultimate limit state in shear
(concrete diagonal tension), the FRP develops an
effective strain in the principal material direction of
approximately 0.004. This limiting strain is conservative
with respect to what tests have indicated (Sato et al.,
1996; Araki et al., 1997; Triantafillou, 1998, Carolin, and
Taljsten,2005; Chajes et al., 1995; Deniaud and Cheng,
2001;). Such a limiting strain value was also proposed by
Priestley et al. (1996) to control circular bridge column
dilation and was adopted by ACI Committee 440 (2002).
f frp = FRP resistance factor equal to 0.85

4.3.2 Strength provided by externally bonded FRP C4.3.2

reinforcement

The contribution of FRP (Vfrp) can be computed using The contribution of the FRP to the shear strength of a
the 45 truss model as: member is based on an assumed crack pattern of 45o and
the fiber orientation (angle between the principal fiber
A frp f d f ( sin a f + cos a f ) orientation and longitudinal axis of member in Fig. 4.2-
V frp =
fe
2). Eq. (4.3.2-1) is analogous to the equation for shear
sf
appearing in Chapter 11 of ACI Standard 318-05. The
A frp E f e fe d f ( sin a f + cos a f ) effective strength per unit width in Eq. (4.3.2-1) may be
= (4.3.2-1) taken equal to the mean FRP stress along the shear crack.
sf
The value of this stress at each location along the shear
= r f E f e fe bv d f ( sin a f + cos a f ) crack depends mainly on the strengthening scheme
(complete wrapping, U-jacketing, anchored U-jacketing,
where: and side bonding) and on the bond stress slip relation at
the FRP-concrete interface (Triantafillou 1998).
Afrp = Area of FRP (in2) covering two sides of the
It should be noted that the tensile area of the wet-layup
beam and can be determined by 2n f t f w f ( n f system is typically based upon the net-area of the fibers;
is number of FRP plies) whereas, the properties for a pre-cured system are based
upon the gross-laminate area.
t f = FRP reinforcement thickness (in)

w f = width of the strip (in)

f fe = effective stress of FRP (ksi)

d f = effective depth of FRP measured from the top of

FRP reinforcement to the centroid of the
longitudinal reinforcement (in)

s f = center-to-center spacing of FRP (in)

f = angle of inclination of FRP with respect to the

longitudinal axis of the member as shown in
Figure 4.2-2 (deg.)

E f = modulus of elasticity of FRP (ksi)

fe = effective strain of FRP (in/in)

f = reinforcement ratio of FRP

bv = effective web width (in) taken as the minimum

web width within the effective depth ( d f )

The FRP shear reinforcement ratio, f , is determined

as:

For discrete strips

 2n f t f w f
f (4.3.2-2)
bv s f

For continuous sheets

2n f t f
f (4.3.2-3)
bv
The effective strain ( fe ) can be expressed as:

For Full Anchorage (Rupture Failures Expected):
Complete Wrap or U-Wrap with Anchors:
fe Rf fu
in which:
R f 0.088 4( f E f ).67 1.0
For Other Anchorage (Non-Rupture Failures more
likely): Side bonding or U-Wrap:
e fe = R f e fu 0.004 (4.3.2-5)
in which:
R f 0.066 3( f E f ) .67 1.0
The effective strain, fe , is largely dependent on the
failure modes as discussed in NCHRP Report 6788
Appendix ASections A3 and A4. Therefore, the
(4.3.2-4) experimental database contained in NCHRP Report 678
was grouped by the failure mode of the test specimens,
i.e., either as debonding or rupture of the FRP and then
regression analyses was performed to obtain Eq. 4.3.2-4
and 4.3.2-5.
The effective strain, fe , represents the average strain
experienced by the FRP at shear failure of the
strengthened member.
The upper bound for the quantity f Ef in Eq. 4.3.2-4 and
4.3.2-5 is 300 ksi (NCHRP Report 678). Substituting this
value in these two equations results in the lower bound
value of Rf shown in the two equations.

The reduction factors Rf were developed from tests in

which the loading was at a distance from the support
sufficient to assume plane sections before deformation
remain plane after deformation, i.e. shallow beam
behavior.

4.3.3 Limitations

4.3.3.1 Shear span-to-depth ratio

These provisions are only applicable to beams with a

shear span-to-depth ratio greater than 2.5.

4.3.3.2 Maximum FRP Shear Reinforcement C4.3.3.2

The amount of FRP should be determined so that the This provision is required to avoid web crushing failure
nominal shear strength should not exceed the nominal of FRP strengthened beams due to excessive transverse
shear strength calculated by shear reinforcement (both FRP and steel stirrups).

Vn 0.25 f cbv dv Vp
(AASHTO 5.8.3.3-2)

where:

Vn = Vc + Vs + Vfrp
4.3.3.3 Maximum Spacing of FRP Shear
Reinforcement

The clear spacing between externally bonded FRP

shear reinforcement shall not exceed the maximum
permitted spacing ( smax ) in accordance with AASHTO
LRFD, expressed as:

If vu 0.125 f c' then smax 0.8dv 24in.

(AASHTO 5.8.2.7-1)
If vu 0.125 f c' then smax 0.4dv 12in.
(AASHTO 5.8.2.7-2)
where:
vu = the shear stress calculated in accordance with
AASHTO LRFD Article 5.8.2.9 (ksi) and

d v = effective shear depth as defined in AASHTO

LRFD - Article 5.8.2.9 (in.)

4.4 STRENGTH IN TORSION C4.4

The factored torsion strength, Tr, of a concrete member
strengthened with an externally bonded FRP system
shall equal or exceed the required torsion strength, Tu,
determined from the effect of the factored loads.
Strengthening for increased torsional capacity may be
required in conventional beams and columns, as well as
in box girders and other structural members with hollow
sections. The principles applied to strengthening in shear
are also valid, for the most part, for the case of torsion.

The user of these Guide Specifications is cautioned that,

in contrast to the provisions in Articles 4.2 and 4.3,
supporting experimental data on the enhancement of the
capacity of a member to withstand torsion by externally
bonded FRP reinforcement does not exist. Accordingly,
in situations where this limit state is considered, the
Engineer of Record should consider the option of
confirmatory testing.

4.4.1 Factored strength in torsion

The factored torsion strength, Tr, shall be defined as:

Tr = fTn + f frpT frp (4.4.1-1)

in which:

Tn = nominal strength in torsion specified in Article

5.8.3.6 of AASHTO LRFD;

T frp = nominal shear strength provided by the externally

bonded FRP system in accordance with Article 4.4.2;

= Resistance Factor defined in Article 5.5.4.2 of

AASHTO LRFD

f frp = 0.65

4.4.2 Nominal strength in torsion

Externally bonded FRP reinforcement used to

strengthen members in torsion shall be completely
wrapped, as defined in Article 4.2. The nominal
strength in torsion shall be calculated as follows:

For intermittent FRP reinforcement,

N efrp d frpa t x1 y1
T frp = (4.4.2-1)
sf

For continuous FRP reinforcement

T frp = N efrp d frpa t x1 y1 (4.4.2-2)

in which

(
a t = 0.66 + 0.33 y1 x1 1.5 )
where:

x1 = lesser dimension of the member (in)

y1 = larger dimension of the member (in)

N efrp = effective tensile strength per unit width (kip/in)

of the FRP reinforcement determined as:

1
N efrp = N s + N frp ,w - N s (4.4.2-3)
2

where N frp,w is the tensile strength (kip/in) of a closed

(wrapped) jacket, defined as:

N frp,w = 0.5N ut N s

N ut = nominal tensile strength of the FRP

reinforcement (kip/in);

Ns = strength of FRP reinforcement (kip/in)

corresponding to a strain of 0.004 (in/in)
4.5 REFERENCES

ACI (2005). Building code requirements for structural concrete (ACI Standard 318-05). American Concrete Institute,
Farmington Hills, MI.

ACI Committee 440 (2002). Guide to the design and construction of externally bonded FRP systems for strengthening
concrete structures (ACI 440.2R-02). American Concrete Institute, Farmington Hills, MI

ACI Committee 440 (2008). Guide to the design and construction of externally bonded FRP systems for strengthening
concrete structures (ACI 440.2R-08). American Concrete Institute, Farmington Hills, MI

Araki, N., Matsuzaki, Y., Nakano, K., Kataoka, T., and Fukuyama, H. (1997). Shear capacity of retrofitted rc members
with continuous fiber sheets. Non-Metallic (FRP) Reinforcement for Concrete Structures, Japan Concrete Institute, 1,
515-522.

Brosens, K. and Van Gemert, D. (1999), Anchorage design for externally bonded carbon fiber reinforced polymer
laminates, Proceedings of Fourth International Symposium on FRP Reinforcement for Concrete Structures, Baltimore,
USA, 635-645.

Carolin, A. and Taljsten, B. (2005). Experimental study of strengthening for increased shear bearing capacity, ASCE
J. Comp. Constr., 9(6), 488-496.

Chajes, M. J., Januska, T. F., Mertz, D. R., Thomson, T. A., and Finch, W. W. (1995). Shear strengthening of
reinforced concrete beams using externally applied composite fabrics. ACI Struct. J., 92(3), May-June, 295-303

Deniaud, C. and Cheng, R. (2001). Shear behavior of reinforced concrete T-beams with externally bonded fiber-
reinforced polymer sheets, ACI Struct. J., 98(3), May-June, 386-394

Holzenkmpfer,P. (1994), Ingenieurmodelle des verbundes geklebter bewehrung fr betonbauteile. Dissertation, TU

Braunschweig (In German).

Khalifa, Ahmed, Gold, William J., Nanni, A., and Abdel Aziz, M.I. (1998). Contribution of Externally Bonded FRP to
Shear Capacity of FRP Members, ASCE Journal of Composites for Construction, Vol 2, No. 4, pp. 195-202.

Monti, G., Santinelli, F. and Liotta, M. A. (2004a), Shear strengthening of beams with composite materials,
Proceedings of the International Conference on FRP Composites in Civil Engineering CICE 2004, Ed. R. Seracino,
Adelaide, Australia, 569-577.

Monti, G., Santinelli, F., and Liotta, M.A. (2004b). Mechanics of shear FRP-strengthening of RC beams. ECCM 11,
Rhodes, Greece.

Priestly, M. J.N., Seible, F., and Calvi, M. (1996). Seismic design and retrofit of bridges, John Wiley and Sons, Inc., NY

Ritter, W. (1899). Die Bauweise Hennebique, Schweizerische, Bauzeitung, Vol. 33, No. 7 pp. 5961.

Sato, Y., Ueda, T., Kakuta, Y., and Tanaka, T. (1996). Shear reinforcing effect of carbon fiber sheet attached to side of
reinforced concrete beams. Advanced Composite Materials in Bridges and Structures, M. M. El-Badry, ed., 621-627.

Triantafillou, T. C. (1998), Shear strengthening of reinforced concrete beams using epoxy-bonded FRP composites,
ACI Structural Journal, 95(2), 107-115.

NCHRP Report 655 (2010) Recommended Guide Specification for the Design of Externally Bonded FRP Systems for
Repair and Strengthening of Concrete Bridge Elements, Transportation Research Board, Washington, D.C.

NCHRP Report (2011) Design of FRP Systems for Strengthening Concrete Girders in Shear, Transportation Research
Board, Washington, D.C
SECTION 5: MEMBERS UNDER COMBINED AXIAL FORCE AND FLEXURE

.
5.1 GENERAL REQUIREMENTS

The factored resistance of structural members

subjected to combined axial forces and flexure
shall equal or exceed the required strength at all
sections calculated for the factored loads and forces
in combinations stipulated by these Guide
Specifications.

Except where specifically provided below, all

provisions of Article 6.9 of AASHTO LRFD shall
apply.

5.2 METHODS FOR STRENGTHENING

WITH FRP REINFORCEMENT

5.2.1 Columns shall be strengthened with FRP

reinforcement using the complete wrapping
method specified in Article 4.2.

5.3 COLUMNS IN AXIAL COMPRESSION

5.3.1 General Requirements C5.3.1

The factored axial load resistance, Pr , for a
confined column shall be taken as follows:
For members with spiral reinforcement
( ) (
Pr = 0.85f 0.85 f cc' Ag - Ast + f y Ast
(5.3.1-1)
For members with tie reinforcement
( ) (
Pr = 0.80f 0.85 f cc' Ag - Ast + f y Ast
(5.3.1-2)
where:
The design procedure for columns strengthened
with FRP is the same as for reinforcement concrete
columns without strengthening. However, the
concrete compressive strength f c' is substituted by
the increased confined concrete compressive
)
strength f cc' as calculated according to Article
5.3.2.2.
The multipliers of 0.85 and 0.80 in equations 5.3.1-
1 and 5.3.1-2 reflect the effect of minimum
accidental eccentricities of axial force (0.05h and
0.10h, respectively, for columns with spiral or tied
)
reinforcement), which impart small end moments
to columns. Columns with eccentricities greater
than these values must be designed using the
provisions of Section 5.5 to take these extra
moments into account.

f = resistance factor specified in Article 5.5.4.2 of
AASHTO LRFD
Ag = gross area of section (in2)
Confined circular columns sustain ultimate axial
strains that are far greater than those of non-
confined columns. Any gain in strength due to
strain hardening of the steel reinforcement is not
accounted for in the above equation, thus providing
additional safety. This gain is a function of the
ultimate axial strains, unless buckling of the steel
Ast = total area of longitudinal reinforcement, (in2). reinforcement initiates failure of the column.

f y = specified yield strength of reinforcement (ksi)

f cc' = compressive strength of the confined concrete

(ksi) determined according to Article 5.3.2.2.

5.3.2 Short Columns in Compression C5.3.2

Columns in compression shall be fully wrapped

over the entire length. The provisions in Article 5.3.2 apply to short
columns in which second-order effects are
negligible and the limit state of instability can be
ignored.

5.3.2.1 Limitations C5.3.2.1

Provisions in this section shall apply to circular The limitations are similar to those in the Canadian
columns in which the slenderness parameter lu D guidelines for column strengthening (ISIS 2001).
The limitation on column slenderness in this
does not exceed 8 and to rectangular columns in
section ensures that the development of column
which the aspect ratio, h b does not exceed 1.5, strength not prevented by column instability.
the minimum radius of corners is one inch, and the
slenderness parameter, lu b , does not exceed 9,
where:

D = external diameter of the circular member (in)

b = smaller dimension of the rectangular member

(in)

h = larger dimension of the rectangular member

(in)

5.3.2.2 Confinement in Columns C5.3.2.2

The compressive strength of the confined concrete, The bonding of FRP sheets, where the fiber
f cc' , shall be determined from: orientation is perpendicular to the column axis to
limit the circumferential strains in the column,
constitutes confinement. Various confinement
2f
f cc' = f c' 1+ 'l models have been developed over the years and
fc (5.3.2.2-1) comparisons among the most common models
have been presented by Rocca et al. (2008). The
The confinement pressure due to FRP expression for the compressive strength of
strengthening, f l (ksi) for circular columns shall confined concrete adopted in these guides is similar
to that of ISIS Canada due to its simplicity. The
be determined as: stress-strain curve for concrete confined by FRP
reinforcement can be considered to be bilinear, but
2 N frp f c' 1 differs from the situation where the confinement is
f l frp c provided by spiral reinforcement or steel jacketing.
2 ke (5.3.2.2-2)
D The secondary stiffness depends on the hoop
stiffness of the confining reinforcement.
where:
The maximum value of the confinement pressure
ke is a strength reduction factor applied for
unexpected eccentricities. It shall be taken as
follows:
ke = 0.80 for tied columns, and
ke = 0.85 for spiral columns.
N frp = Strength per width of FRP reinforcement
(kip/in) corresponding to a strain of
0.004 (in/in).
c = 0.75, material resistance factor for concrete,
frp = 0.65,
The confinement pressure shall be greater or equal
to 600 psi.
For rectangular columns, the diameter D in Eq.
(5.3.2.2-2) shall be replaced with the smaller
dimension of the width and depth.
5.3.3 Slender columns
Columns not meeting the limitations on slenderness
in 5.3.2.1 shall be designated as slender and their
design shall be based on forces and moments
determined from rational analysis. Such an analysis
shall take into account the influence of forces,
deflections and foundation rotations, and duration
of loads on member stiffness and on the
development of moments, shears and axial forces.
5.4 COLUMNS IN COMBINED AXIAL
COMPRESSION AND BENDING
5.4.1 General requirements
Members subjected to moment in combination with
axial load shall be designed for the maximum
moment that can accompany the axial load. The
factored axial force at given eccentricity shall not
exceed Pr given in Section 5.3.1. The maximum
required moment, Mu, shall be magnified, as
appropriate, for slenderness effects.
specified in Eq. 5.3.2.2-2 was established to limit
the axial compression strains in over strengthened
columns. The minimum confinement pressure of
600 psi reflects the fact that the effectiveness of the
confinement pressure depends upon a certain level
of ductility. Relevant background related to the
maximum and minimum values of confinement
pressure in FRP reinforcement jackets in axially
loaded columns is given by Thriault and Neale
(2000).
When equation 5.3.2.2-2 is applied to rectangular
columns after replacing D with the smaller
dimension of the rectangular section, the factored
axial strength estimated from Eq. 5.3.1-1 or 5.3.1-2
errors on the conservative side. At present, this is
justified owing to the limited properly documented
available test data.
The gain in strength provided by the confinement
of rectangular sections is very little compared to
that attainable for circular sections. As a result,
neither minimum nor maximum limits are specified
for rectangular sections since the attainable
confinement pressure, which relies on ductility
development, is very limited for rectangular
columns.
C5.3.3
The provisions for short columns in Article 5.3.2
are adequate for the majority of rehabilitation
projects because second-order structural actions
leading to instability seldom would occur. There is
only limited test data to support the development of
column strength provisions in situations where this
is not the case. In such situations, the required
column strength should be determined by rational
analysis, supplemented by confirmatory testing,
where feasible.
C5.4.1
The design procedure for the members
strengthened with FRP is the same as for
reinforcement concrete members without
strengthening. However, the concrete compressive
strength f c' is substituted by the increased confined
concrete compressive strength f cc' as calculated
 according to articles 5.3.2.2.

5.4.2 Design Basis

Design of columns subject to combinations of axial

force and flexure shall be based on stress and strain
compatibility. The maximum usable strain in the
extreme concrete compression fiber shall be
assumed to equal 0.003 (in/in).

Externally bonded FRP reinforcement of columns

strengthened to withstand end moments due to
lateral load shall be reinforced over a distance from
the column ends equal to the maximum column
dimension or the distance over which the moment
exceeds 75% of the maximum required moment,
whichever distance is larger.

The tensile strength of the FRP reinforcement in

the longitudinal direction of the column shall be
determined by rational analysis. However, the
strength in the longitudinal direction shall not be
less than 50% of the strength in the perimeter
direction.

5.4.3. Limitations C5.4.3

The contribution of the FRP reinforcement to

column capacity shall not be considered at
eccentricity ratios greater than those corresponding
to balanced strain conditions, at which tension
reinforcement reaches the strain corresponding the
steel yield strength and concrete in compression
reaches an ultimate strain of 0.003 (in/in) at any
cross section.
Provisions in Article 5.4 are limited to members
subjected to combined axial loading and bending
where failures occur by concrete crushing in
compression rather than reinforcement yielding in
tension. If the eccentricity of axial force present in
the member is greater than 0.10h for the spirally
reinforced columns or 0.05h for tied columns,
strengthening requires externally bonded FRP
reinforcement to withstand force in the longitudinal
direction of the column in addition to its perimeter.

5.5 AXIAL TENSION

5.5.1 Limitations

Members that are axially loaded in tension shall be

reinforced symmetrically with respect to the
column cross-section principal axes.

5.5.2 General requirements C5.5.2

The factored axial load resistance, Pr, for an axially FRP systems can be used to provide additional
loaded member with externally bonded FRP tensile strength to concrete members. The tension
reinforcement shall be determined as: strength provided by the FRP is limited by the
design tensile strength of the FRP and the ability to
Pr As f y frp N frp w frp transfer stresses into the substrate through bond.
The effective strain in the FRP can be determined
(5.5.2-1) based on the criteria given for shear strengthening.

where: For members completely wrapped by the FRP

= Resistance Factor defined in Article 5.5.4.2 of
AASHTO LRFD
f frp = 0.5
N frp = tensile strength per unit width in the load
direction (kip/in) at a strain value of 0.005.
systems, loss of the aggregate interlock of concrete
occurs at fiber strain less than the ultimate fiber
strain. To preclude this mode of failure, the
maximum design strain should be limited to 0.4%:
e fe = 0.004 0.75e fu
where:

w frp = total width of FRP reinforcement along the e fe is the effective strain level in FRP
cross section (in). reinforcement attained at failure (in/in)

e fu is the design rupture strain of FRP

reinforcement (in/in)

5.6 REFERENCES

ISIS (2001). ISIS Canada Design Manuals, Strengthening Reinforced Concrete Structures with Externally-
Bonded Fiber-Reinforced Polymers, Winnipeg, Manitoba.

Mirmiran, A., Shahawy, M. (1997). Behavior of concrete columns confined by fiber composites. J. Struct.
Engrg. ASCE 123(5): 583-590.

Mirmiran, A., Shahawy, M., Samaan, M., El Echary, H., Mastrapa, J.C. and Pico, O. (1998) Effect of column
parameters on FRP-confined concrete. J. Composites for Construction, ASCE 2(4): 175-185.

Saaman, M., Mirmiran, A. and Shahawy, M. (1998). Model of concrete confined by fiber composites. J.
Struct. Engrg. ASCE 124(9):1025-1031.

Rocca, S., Galati, N. and Nanni, N. (2008) ASCE Journal of Composites for Construction, Vol. 12, No.
1,February, pp.80-92.

Thriault, M. and Neale, K.W. (2000). Design equations for axially loaded reinforced concrete columns
strengthened with fibre reinforced polymer wraps, Canadian Journal of Civil Engineering, 27(5): 1011.1020.

Val, D. (2003). Reliability of fiber-reinforced polymer-confined reinforced concrete columns. J. Struct. Engrg.
ASCE 129(8): 1122-1130.
2012 AASHTO BRIDGE COMMITTEE AGENDA ITEM: 43 (REVISION 1)

SUBJECT: AASHTO LRFD Guide Specifications for Design of Concrete-Filled FRP Tubes
for Flexural and Axial Members

TECHNICAL COMMITTEE: T-6 Fiber Reinforced Polymer Composites

REVISION ADDITION NEW DOCUMENT

DESIGN SPEC CONSTRUCTION SPEC MOVABLE SPEC

MANUAL FOR BRIDGE SEISMIC GUIDE SPEC COASTAL GUIDE SPEC
EVALUATION OTHER

DATE PREPARED: 2/14/12

DATE REVISED: 7/9/12

AGENDA ITEM:
AASHTO LRFD Guide Specifications for Design of Concrete-Filled FRP Tubes for Flexural and Axial Members
See Attachment B

OTHER AFFECTED ARTICLES:

None

BACKGROUND:
Guide Specifications were developed in conjunction with the AASHTO TIG initiative for concrete-filled FRP tube
type structures.

ANTICIPATED EFFECT ON BRIDGES:

Provides for an alternate type of construction for bridges and culverts.

REFERENCES:
None

OTHER:
None
ATTACHMEN B 2012 AGENDA ITEM 43 T-6

AASHTO LRFD GUIDE SPECIFICATIONS FOR

DESIGN OF CONCRETE-FILLED FRP TUBES FOR
FLEXURAL AND AXIAL MEMBERS
 TABLE OF CONTENTS

SECTION 1: INTRODUCTION

1.1 SCOPE .................................................................................................................................... 1

1.2 DEFINITIONS ........................................................................................................................ 2
1.3 LIMITATIONS........................................................................................................................ 2
1.4 DESIGN PHILOSOPHY ......................................................................................................... 2
1.5 REFERENCES ........................................................................................................................ 3
Section 2.......................................................................................................................................... 4
CONCRETE-FILLED FRP TUBES (CFFTs)................................................................................ 4
2.1 SCOPE .................................................................................................................................... 4
2.2 DEFINITIONS ........................................................................................................................ 4
2.3 NOTATION ............................................................................................................................ 4
2.4 LIMITATIONS........................................................................................................................ 7
2.5 VERIFICATION OF COMPOSITE ACTION..........................................................................8
2.6 MATERIAL PROPERTIES ..................................................................................................... 9
2.6.1 FRP Tube ......................................................................................................................... 9
2.6.1.1 General......................................................................................................................... 9
2.6.1.2 Tensile and Compressive Ultimate Strengths and Strains .............................................. 9
2.6.1.3 Modulus of Elasticity.................................................................................................. 10
2.7 LIMIT STATES..................................................................................................................... 11
2.7.1 Service Limit State ......................................................................................................... 11
2.7.2 Fatigue and Creep Rupture Limit State ........................................................................... 11
2.7.3 Strength Limit State........................................................................................................ 12
2.7.3.1 General....................................................................................................................... 12
2.7.3.2 Resistance Factors ...................................................................................................... 12
2.7.3.3 Stability...................................................................................................................... 15
2.7.4 Extreme Event Limit State.............................................................................................. 15
2.8 DESIGN CONSIDERATIONS .............................................................................................. 15
2.8.1 General........................................................................................................................... 15
2.8.2 Effect of Imposed Deformation....................................................................................... 15
2.9 DESIGN FOR FLEXURE WITH NO AXIAL COMPRESSION............................................ 15
2.9.1 General........................................................................................................................... 16
 2.9.2 Assumptions................................................................................................................... 16
2.9.2.1 General....................................................................................................................... 16
2.9.2.2 Service and Fatigue and Creep Rupture Limit States ................................................... 16
2.9.2.3 Strength and Extreme Event Limit States .................................................................... 17
2.9.3 Minimum Longitudinal Tension Reinforcement.............................................................. 18
2.9.4 Minimum Reinforcement in the Transverse or Hoop Direction ....................................... 18
2.9.5 Factored Flexural Resistance .......................................................................................... 18
2.9.5.1 General Method.......................................................................................................... 19
2.9.5.2 Simplified Method...................................................................................................... 19
2.9.6 Deformation ................................................................................................................... 19
2.9.7 Control of Cracking........................................................................................................ 20
2.9.8 Stress Limit for Concrete................................................................................................ 20
2.10 DESIGN FOR AXIAL COMPRESSION ............................................................................... 20
2.10.1 General........................................................................................................................... 20
2.10.2 Assumptions................................................................................................................... 20
2.10.3 Factored Axial Compressive Resistance.......................................................................... 21
2.10.4 Minimum Reinforcement in the Hoop Direction ............................................................. 23
2.10.5 Stress Limit for Concrete................................................................................................ 23
2.11 DESIGN FOR COMBINED FLEXURE AND AXIAL COMPRESSION............................... 24
2.11.1 General........................................................................................................................... 24
2.11.2 Assumptions................................................................................................................... 24
2.11.3 Factored Resistance........................................................................................................ 25
2.11.3.1 General Method ...................................................................................................... 26
2.11.3.2 Simplified Method .................................................................................................. 26
2.11.4 Evaluation of Slenderness Effects ................................................................................... 27
2.11.5 Minimum Reinforcement................................................................................................ 27
2.11.6 Deformation ................................................................................................................... 28
2.11.7 Control of Cracking........................................................................................................ 28
2.11.8 Stress Limit for Concrete................................................................................................ 28
2.12 DESIGN FOR SHEAR EFFECTS ......................................................................................... 28
2.12.1 General........................................................................................................................... 28
2.12.2 Nominal Shear Resistance .............................................................................................. 29
2.12.3 Minimum Shear Reinforcement ...................................................................................... 31
 2.12 CONNECTIONS ................................................................................................................... 32
2.13 REFERENCES ...................................................................................................................... 33
Section 3........................................................................................................................................ 35
3.1 SCOPE .................................................................................................................................. 35
3.2 DEFINITIONS ...................................................................................................................... 35
3.3 NON-STANDARD DOCUMENTS ....................................................................................... 37
3.4 MATERIAL AND MANUFACTURE ................................................................................... 37
3.4.1 Fibers ............................................................................................................................. 37
3.4.2 Matrix Resins ................................................................................................................. 37
3.4.3 Fillers and Additives....................................................................................................... 37
3.4.4 Manufacturing Process ................................................................................................... 38
3.5 PHYSICAL PROPERITES .................................................................................................... 38
3.5.1 Fiber Content ................................................................................................................. 38
3.5.2 Glass Transition Temperature ......................................................................................... 38
3.5.3 Longitudinal and Transverse Coefficients of Thermal Expansion (CTE) ......................... 39
3.6 MECHANICAL PROPERITES ............................................................................................. 39
3.6.1 Tensile Strength ............................................................................................................. 39
3.6.2 Tensile Modulus of Elasticity ......................................................................................... 40
3.6.3 Ultimate Tensile Strain ................................................................................................... 40
3.6.4 Compressive Strength..................................................................................................... 41
3.6.5 Ultimate Compressive Strain .......................................................................................... 41
3.7 DURABILITY PROPERTIES ............................................................................................... 41
3.7.1 Moisture Absorption....................................................................................................... 41
3.7.2 Resistance to Alkaline Environment ............................................................................... 42
3.8 SAMPLING........................................................................................................................... 42
3.8.1 Sampling Frequency and Number of Specimens ............................................................. 42
3.8.2 Rejection ........................................................................................................................ 42
3.9 CERTIFICATION ................................................................................................................. 43
3.9.1 Documents ..................................................................................................................... 43
3.9.2 QC/QA........................................................................................................................... 43
3.9.3 Product Certification ...................................................................................................... 44
3.9.4 Markings ........................................................................................................................ 45
3.10 REFERENCES ...................................................................................................................... 46
 Section 1

INTRODUCTION

1.1 SCOPE C1.1

These Specifications present provisions for the
analysis and design of concrete-filled fiber reinforced
polymer (FRP) tubes (CFFT) for use as structural
components in bridges. Design methodology presented in
this specification allows CFFTs to be designed as flexural
members, axial compression members, or members
subjected to combined flexural and axial compression, in
addition to shear. CFFT bridge components may include
beams, arches, columns and piles.
These Specifications are not intended to supplant
proper training or the exercise of judgment by the Design
Professional, and state only the minimum requirements
necessary to provide public safety. The Owner or the
Design Professional may require the sophistication of the
design and/or the quality of materials and construction to
be higher than the minimum requirements.
FRP materials have emerged as an alternative
material to steel reinforcement for concrete structures.
They offer advantages over steel reinforcement due to
their noncorrosive nature and nonconductive behavior.
FRP is a also a versatile material that can be produced
in many forms such as reinforcing bars, grids, rigid
plates, flexible sheets and several structural shapes,
including tubes. This specification is focused on one
application of FRP in the form of tubes used as
structural stay-in-place forms filled with concrete
[Fardis and Khalili (1981), Nanni and Bradford (1995),
Mirmiran and Shahawy (1996), Davol (1998),
Burgueo (1999), Fam (2000), Fam and Rizkalla
(2001), Fam and Rizkalla (2002)]. Due to differences
in the physical and mechanical behavior of FRP
materials as opposed to steel, particularly when used as
stay-in-place structural forms, unique guidance on the
engineering and construction of bridge components
using this technology is needed.

The Design Professional shall be familiar with the

provisions of the AASHTO LRFD Bridge Design
Specifications, 6th Ed. (AASHTO 2012) and the latest
interim revisions, as well as with the design of
conventional reinforced concrete structures and structures
exposed to earth loading.

The commentary directs attention to other documents

that provide suggestions for carrying out the requirements
and intent of these Specifications. However, those
documents and this commentary are not intended to be a
part of these Specifications.
1.2 DEFINITIONS

Composite Action - A condition in which two or more elements or components are made to act together by eliminating
relative movements at their interface.

Design Professional - The architect, engineer, architectural firm, or engineering firm responsible for the design of the
bridge and issuing Contract Documents or administering the Work under Contract Documents, or both.

Fiber - Any fine thread-like natural or synthetic object of mineral or organic origin. Note: This term is generally used
for materials whose length is at least 100 times its diameter.

Fiber, Aramid - Highly oriented organic fiber derived from polyamide incorporating an aromatic ring structure.

Fiber, Carbon - Fiber produced by heating organic precursor materials containing a substantial amount of carbon,
such as rayon, polyacrylonitrile (PAN), or pitch in an inert environment.

Fiber, Glass - Fiber drawn from an inorganic product of fusion that has cooled without crystallizing.

1.3 LIMITATIONS C1.3

Composite action between the concrete core and FRP
tube is required for a CFFT member to develop its
stiffness and strength as defined in these specifications.
CFFTs which do not have sufficient bond between the
concrete core and FRP tube to ensure composite action
are not addressed in these Specifications. Composite
action shall be verified in accordance with section 2.5 of
this specification.
Bond may be achieved by shear interlock
mechanism, such as a roughened inner surface of the
FRP tube and/or friction, which can be further
enhanced by the use of low-shrinkage or expansive
concrete.FRP materials demonstrate a linear-elastic
behavior up to failure and do not demonstrate yielding,
which is the basis for plastic hinge formation and
moment redistribution.

The assumed failure mechanism of CFFTs used as

flexural members shall not be based on the formation of
plastic hinges. CFFTs shall not be used as ductile
earthquake resisting elements.

1.4 DESIGN PHILOSOPHY C1.4

These Specifications are based on limit state design The limit states specified herein are intended to
principles where structural components shall be provide for a buildable, serviceable bridge, capable of
proportioned to satisfy the requirements at all appropriate safely carrying design loads for a specified lifetime.
service, fatigue and creep rupture, strength and extreme
event limit states. In many instances, serviceability or
fatigue and creep rupture limits may control the design.

Provisions related to limit states analyses, general

design and location features, loads and load factors, and
structural analysis and evaluation shall be in compliance
with the AASHTO LRFD.

1.5 REFERENCES

AASHTO. (2012). AASHTO LRFD Bridge Design Specifications, 6th Edition with Interims. Washington
DC: American Association of State Highway and Transportation Officials.

Burgueo, R. (1999). System Characterization and Design of Modular Fiber Reinforced Polymer (FRP) Short- and
Medium-Span Bridges. (Doctoral dissertation, University of California, San Diego, 1999). (UMI No.
9928617)

Davol, A. (1998). Structural Characterization of Concrete Filled Fiber Reinforced Shells. (Doctoral dissertation,
University of California, San Diego, 1998).

Fam, A. (2000). Concrete-Filled Fibre-Reinforced Polymer Tubes for Axial and Flexural Structural Members.
(Doctoral dissertation, Univeristy of Manitoba, 2000).

Fam, A. & Rizkalla, S. (2001). Confinement Model for Axially Loaded Concrete Confined by Circular Fiber-
Reinforced Polymer Tubes. ACI Structural Journal 98(4), 451-461.

Fam, A. & Rizkalla, S. (2002). Flexural Behavior of Concrete-Filled Fiber-Reinforced Polymer Circular Tubes.
Journal of Composites for Construction. 6(2), 123-132.

Fardis, M. N. and Khalili, H. Concrete Encased in Fibreglass-Reinforced Plastic, ACI Structural Journal, Title No.
78-38, Nov.-Dec. 1981, pp. 440-446.

Mirmiran, A. & Shahawy, M. (1996) A new concrete-filled hollow FRP composite column. Composites Part B:
Engineering. Special Issue on Infrastructure. Elsevier Science Ltd., 27B(3-4), 263-268.

Nanni, A., & Bradford, N. (1995) FRP Jacketed Concrete Under Uniaxial Compression. Construction and Building
Materials. 9(2). 115-124.
 Section 2

CONCRETE-FILLED FRP TUBES (CFFTs)

2.1 SCOPE

This section contains the provisions for the design of

concrete-filled FRP tubes (CFFTs) for use as components
of a bridge structure.

2.2 DEFINITIONS

Component - A constituent part of a structure.

Composite Action - A condition in which two or more elements or components are made to act together by eliminating
relative movements at their interface.

Creep - Time-dependent plastic strain in a material subjected to sustained stress.

Creep Rupture - The gradual, time-dependent reduction of tensile strength due to permanent loading that leads to a
premature failure of the section.

Dynamic Load Allowance - An increase in the applied static force effects to account for the dynamic interaction
between the bridge and moving vehicles.

Specified Strength of Concrete - The nominal compressive strength of concrete specified for the work and assumed for
design and analysis of new structures.

Stay-in-Place Form - A permanent form for the concrete that remains in place after construction is finished.

2.3 NOTATION

Ac = effective concrete area (in2)

CE = environmental reduction factor

dv = effective shear depth (in)

D = average diameter of circular FRP tube (in)

Di = inner diameter of circular FRP tube (in)

Do = outer diameter of circular FRP tube (in)

Ec = elastic modulus of concrete (ksi)

Efh = elastic modulus of the tube laminate in the hoop direction (ksi)

Efl = elastic modulus of the tube laminate in the longitudinal direction (ksi)

Es = elastic modulus of steel (ksi)

EI = flexural stiffness of concrete-filled FRP tube (kip-in2)

fc = unconfined concrete design compressive strength (ksi)

fcc = confined concrete design compressive strength (ksi)

ffcu = design compressive strength of FRP laminate in longitudinal direction considering reductions for

service environment (ksi)

f*fcu = compressive strength of FRP laminate in longitudinal direction for product certification as reported

by manufacturers (ksi)

ffe = tensile hoop stress in the tube laminate (ksi)

ffuh = design tensile strength of FRP laminate in hoop direction considering reductions for service

environment (ksi)

f*fuh = tensile strength of FRP laminate in hoop direction for product certification as reported by

manufacturers (ksi)

fful = design tensile strength of FRP laminate in longitudinal direction considering reductions for service

environment (ksi)

f*ful = tensile strength of FRP laminate in longitudinal direction for product certification as reported by

manufacturer (ksi)

fl = maximum confinement pressure (ksi)

fr = modulus of rupture (ksi)

Ifl = moment of inertia of the FRP tube (in4)

Is = moment of inertia of a steel tube (in4)

Mcr = cracking moment (kip-in)

Mn = nominal flexural resistance (kip-in)

Mnb = nominal flexural resistance corresponding to the balance point (kip-in)

Mr = factored flexural resistance (kip-in)

n = modular ratio = Efl/Ec

Pe = Euler buckling load (kip)

Pn = nominal axial compressive resistance (kip)

Pnb = nominal axial compressive resistance corresponding to the balance point (kip)

P'n = axial compressive resistance (kip)

Pr = factored axial compressive resistance (kip)

Sc = section modulus of the extreme fiber of the composite section where tensile stress is caused by

externally applied loads (in3)

t = effective structural thickness of FRP tube (in)

Vc = nominal shear resistance provided by the concrete (kip)

Vf = nominal shear resistance provided by the FRP shell (kip)

Vn = nominal shear resistance (kip)

Vr = factored shear resistance (kip)

Vs = nominal shear resistance provided by steel reinforcement (kip)

X = embedment length of a CFFT member into a reinforced concrete footing, pile cap or beam (in)

ccu = maximum confined longitudinal compressive strain (in/in)

fe = tensile strain in the hoop direction of the FRP tube (in/in)

fcu = design ultimate compressive strain of the FRP laminate in the longitudinal direction considering

reductions for service environment (in/in)

*fcu = ultimate compressive strain of FRP laminate in longitudinal direction for product certification as

reported by manufacturers (in/in)

fuh = design ultimate tensile strain of the FRP laminate in the hoop direction considering reductions for

service environment (in/in)

*fuh = ultimate tensile strain of FRP laminate in hoop direction for product certification as reported by

manufacturers (in/in)

ful = design ultimate strain of the FRP laminate in the longitudinal direction considering reductions for

service environment (in/in)

*ful = ultimate tensile strain of FRP laminate in longitudinal direction for product certification as reported

by manufacturers (in/in)

fv = hoop strain resulting from shear (in/in)

= reinforcement ratio

b = reinforcement ratio producing balanced strain conditions in bending without axial load

= resistance factor

f = FRP system reduction factor for flexural resistance

2.4 LIMITATIONS C2.4

Seismic design is not addressed in this document.
However, where applicable, it shall be considered as one
of the loading cases. CFFTs shall not be used as ductile
earthquake resisting elements.
Design provisions in the following sections should not
be applied to CFFTs with tubes having fibers in the
longitudinal direction only.
Tubes with fibers oriented only in the longitudinal
direction tend to fail prematurely under horizontal
shear by splitting in a plane parallel to the fibers (Fam
2000) before developing their full flexural strength.
Performance of that type must be verified by tests.
Strength enhancement due to confinement in CFFT
compression members with concrete strength fc
exceeding 10 ksi (70 MPa) has not been experimentally
verified (ACI 440.2R-08). Further experimental

These provisions shall not be applied to CFFT
members with unconfined concrete strength fc exceeding
10 ksi (70 MPa).
evidence show that confinement effectiveness using
FRP jackets is reduced as the concrete strength
increases (Mandal et al 2005).

2.5 VERIFICATION OF COMPOSITE ACTION

Testing to verify composite action shall consist of the

following:

3 full-scale static flexural tests to failure

3 full-scale flexural fatigue tests to 2,000,000

cycles using anticipated service loads and a load
ratio consistent with the expected demands on
the structure followed by static tests to failure
for each specimen C2.5

CFFT specimens must exhibit no slip between
shell and concrete before ultimate failure in
static or fatigue tests. CFFT specimens must
retain 90% of the mean strength and stiffness of
the flexural specimens not subjected to fatigue.
Composite action is required for all types of CFFTs
covered by this Guide Specification. CFFTs subjected
to axial compression or combined axial compression
and flexural loading benefit from greater levels of
core/shell friction due to the Poissons effect.

Testing to verify composite action is mandatory for tubes

subjected to only flexural and shear forces (i.e. beams).
Testing requirements for members subjected to axial
compression or combined axial compression and flexural
loading may be waived or reduced at the discretion of the
Engineer.

It is the responsibility of the engineer to ensure that the

manufacturer's testing documentation is in accordance
with the requirements of this specification.
 NUMBERING IN SECTIONS BELOW WAS
REVISED TO ACCOMMODATE NEW SECTION
2.5

2.6 MATERIAL PROPERTIES

2.6.1 FRP Tube

2.6.1.1 General

The FRP tube shall conform to the material

specifications reported in Section 3, and comply with the
applicable requirements of Articles 1.3 and 2.4.

2.6.1.2 Tensile and Compressive Ultimate C2.6.1.2

Strengths and Strains
The strength values as reported by manufacturers
The design ultimate tensile strength of the FRP tube provide a 99.87 percent probability that the indicated
material in the longitudinal direction, fful, and the values are exceeded by similar FRP coupons provided
transverse or hoop direction, ffuh, shall be taken as: that the number of tested specimens complies with the
provisions of Article 3.8.1. The manufacturer should
f ful CE f ful
* provide a description of the method used to obtain the
(2.6.1.2-1) reported strength.

f fuh CE f fuh
* The material properties provided by the
(2.6.1.2-2) manufacturers are considered as initial properties that
do not include the effects of long term exposure to the
The design ultimate compressive strength of the FRP environment. Because long term exposure to various
tube material in the longitudinal direction, ffcu shall be environments may reduce the tensile strength and creep
taken as: rupture and fatigue endurance of FRP tubes, the
material properties used in all design equations are
f fcu C E f fcu
* reduced based on type and level of environmental
(2.6.1.2-3) exposure.

The design ultimate tensile strain of the FRP tube

material in the longitudinal direction, ful , and the
transverse or hoop direction, fuh , shall be taken as:

ful C E *ful
(2.6.1.2-4)
 fuh C E *fuh
(2.6.1.2-5)

The design ultimate compressive strain of the FRP tube

material in the longitudinal direction, fcu shall be taken
as:

fcu C E *fcu
(2.6.1.2-6)

In Equations 2.6.1.2-1 through 2.6.1.2-4, CE shall be

taken as 0.85 for a carbon-based FRP, 0.65 for glass, and
0.75 for aramid. If the bridge is exposed to aggressive
environments, CE shall be taken as 0.5 for glass and 0.7
for aramid.

The quantities
*
f ful *
, f fuh *
, f fcu , *ful , *fuh , and *fcu
are the ultimate tensile strengths and strains in the
longitudinal and hoop directions and the ultimate
compressive strength and strain in the longitudinal
direction for product certification as reported by the
manufacturer and shall be defined as the average value
calculated for a frequency and number of specimens as
specified in Article 3.8.1 minus three times the standard
deviation. In hybrid tubes, where different types of fibers
are used in the longitudinal and transverse or hoop
directions, different CE shall be used in the two directions
according to the type of fibers.

2.6.1.3 Modulus of Elasticity

The FRP tube material shall be treated as linearly

elastic in tension and compression. The modulus of
elasticity in the longitudinal direction, Efl, and the
transverse or hoop direction, Efh, shall be calculated as:

E fl f ful ful (2.6.1.3-1)

E fh f fuh fuh (2.6.1.3-2)

2.7 LIMIT STATES

2.7.1 Service Limit State

Under services loads, CFFT members shall be

analyzed as fully elastic structures. Checks to be
performed at service limit state shall be related to
deformations, cracking and concrete stresses as specified
in Articles 2.9.6, 2.9.7 and 2.9.8 for CFFT flexural
members and Articles 2.11.6, 2.11.7 and 2.11.8 for CFFT
members under flexure and axial compression. The FRP
tube stresses at service are addressed in Article 2.7.2.

The loads to be considered in this analysis shall be as

defined in Section 3 of AASHTO LRFD.

2.7.2 Fatigue and Creep Rupture Limit State C2.7.2

The maximum longitudinal tensile stress in the FRP
tube under all sustained loads plus fatigue loading ffl,s
shall not exceed the following limits:
For carbon-based FRP: 0.55fful
(longitudinal)
For glass-based FRP: 0.20fful
To prevent creep-rupture of the FRP tube under
sustained stress or failure due to cyclic stresses and
fatigue, the longitudinal stress level in the tube under
these stress conditions should be limited. Because
these stress levels will be within the the elastic range of
the CFFT member, the stress can be computed through
an elastic analysis of the cracked transformed CFFT
section.

(longitudinal)

For aramid-based FRP: 0.30fful

(longitudinal)

The loads to be considered in this analysis consist of

all permanent loads, and the fatigue load as defined in
Article 3.6.1.4 of AASHTO LRFD. The load factors for
DC and DW, EV, and EH shall be 1.0, and the load factor
for the fatigue vehicle plus impact shall be 0.75.

Hoop direction stresses shall be checked for fatigue

and sustained loads when the FRP tube is relied upon for If the axial stress in CFFT members exceeds 0.65 fc,
confinement of the concrete core and if the concrete radial cracking occurs and the FRP tube is loaded in
compressive stress level under fatigue and sustained loads the circumferential direction. In this case, creep
is higher than 0.65 fc in accordance with Article 2.10.5 rupture and fatigue should also be checked.

2.7.3 Strength Limit State

2.7.3.1 General

The strength limit states to be considered shall be those

of strength and stability.

Factored resistance shall be the product of nominal

resistance as determined in accordance with the applicable
provisions of Articles 2.8, 2.9, 2.10, 2.11 and 2.12 unless
another limit state is specifically identified, and the
resistance factor as specified in Article 2.7.3.2.

The loads to be considered in this analysis shall be as

defined in Section 3 of AASHTO LRFD.

2.7.3.2 Resistance Factors C2.7.3.2

The ratio of the cross-sectional area of the FRP tube to
the cross-sectional area of the concrete core is defined as
the reinforcement ratio . In CFFT members subjected
to flexure without axial compression loads, the balanced
reinforcement ratio b is defined as the reinforcement
ratio which results in tensile rupture and compressive
crushing of the FRP tube under longitudinal stresses
occurring simultaneously, in accordance with the
assumptions in Article 2.9.2.2. The resistance factor for
flexure shall be computed as follows.
0.55 if b confinement effect is insignificant). On the other hand,
0.65 if 1.4 b on the compression side under a bi-axial state of stress
Tension failure of CFFT flexural and axial-flexural
members is typically by rupture of the tube on the
tension side under longitudinal tensile stresses (i.e. a
uniaxial state of stress). Compression failure of CFFT
flexural and axial-flexural members is not by concrete
crushing as in conventional reinforced concrete
members but is triggered primarily by failure of the
tube, which is immediately followed by crushing of
concrete as a secondary failure. For CFFT flexural
members it is predominantly governed by compression
failure of the tube under longitudinal compressive
stresses, where the hoop tensile stresses (i.e.
for CFFT axial and axial-flexural members,
compression failure is governed by failure of the tube
involving longitudinal compressive and hoop tensile
(i.e. confinement effect) stresses.

The resistance factors proposed here are essentially

 the same as those used in FRP-reinforced concrete
0.3 0.25 if b 1.4 b structures governed by FRP rupture for tension failure

b (2.7.3.2-1) and classic concrete crushing for compression failure
(AASHTO 2012). As stated above, compression failure
in which: of CFFT members is different. Until further reliability
analysis is carried out and new factors are established,
4t it is recommended to use the current factors.
b b
D (2.7.3.2-2)
In pure bending, without axial loads, there is a
unique tube thickness that provides the balanced
and the balanced tube thickness tb is:
condition when the extreme tension and compression
fibers of the FRP tube reach the respective ultimate
f cD
tb 2 Sin(2 ) longitudinal strengths, simultaneously. Equations
4 f ful f fcu 2.7.3.2-3 to 2.7.3.2-6 have been established based on
an equivalent concrete stress block with parameters
in radian and . These parameters have the conventionally
known definition and have been established for a large
(2.7.3.2-3) range of concrete compressive strengths of 3 to 15 ksi
and ultimate longitudinal compressive strain of the
f fcu FRP tube ranging from 2 to 10 times the concrete
a cos 1 2
f ful f fcu compressive strain that corresponds to the peak
(2.7.3.2-4) strength fc. It is to be noted that the concrete core in
CFFT under pure bending is assumed to be represented
0.917
fcu by an unconfined stress-strain curve in compression,
7.325 f c 1.086 with an extended strain softening that is limited by the
c (2.7.3.2-5) tube ultimate longitudinal compressive strain. The two
conditions used to establish and are: i) the area
under the non-linear stress-strain curve is equal to the
fcu
0.247
rectangular stress area, and ii) the location of the
0.633
f c 0.222
c
centroid of the area under the nonlinear curve is the
(2.7.3.2-6) same as that of the rectangular stress distribution.
While this technique conventionally applies to
in which: rectangular sections, is adopted here for the circular
section for simplicity. Both AASHTO LRFD (2012)
f c n and ACI 318 (2008a) also use the same and for any
c cross-section geometry.
Ec n 1 ; (2.7.3.2-7)

f c
n 0. 8
2.5 ; (2.7.3.2-8)

E c 1265 f c 1000
(2.7.3.2-9)

where:
fc = the concrete compressive strength (ksi)

fful = the longitudinal ultimate tensile strength of the FRP

tube (ksi)

ffcu = the longitudinal ultimate compressive strength of the

FRP tube (ksi)

The following limits shall apply:

f ful f fcu
(2.7.3.2-10)

2 fcu 10
c ; (2.7.3.2-11)

3ksi f c 15ksi
(2.7.3.2-12)

In CFFT members subjected to combined bending and

axial loads and governed by failure on the tension side by
rupture of the tube under longitudinal stresses, the
resistance factor shall be taken as 0.55.

In CFFT members subjected to combined bending and

axial loads and governed by failure on the compression
side and when is greater than 1.4 b , the resistance
factor shall be taken as 0.65.

In CFFT members subjected to pure axial loads, the

resistance factor shall be taken as 0.65. Members can
be considered to be subjected to pure axial load if the
eccentricity is less than 0.1D.

Resistance factor for shear shall be taken as 0.75.

2.7.3.3 Stability

The structure consisting of CFFT members as a whole

and its components shall be designed to resist sliding,
overturning, and uplift. Effect of eccentricity of loads
shall be considered in the analysis and design.

2.7.4 Extreme Event Limit State

The structure consisting of CFFT members as a whole

and its components shall be proportioned to resist collapse
due to extreme events, specified in Table 3.4.1-1 of
AASHTO LRFD, as may be appropriate to the site and
application.

2.8 DESIGN CONSIDERATIONS

2.8.1 General

Components and connections shall be designed to

resist load combinations as specified in Section 3 of
AASHTO LRFD, at all stages during the life of the
structure, including those during construction. Load
factors shall be as specified in Section 3 of AASHTO
LRFD, with additional permanent load factors as defined
in Article 2.7.2 for the fatigue and creep rupture limit
state.

As specified in Section 4 of AASHTO LRFD,

equilibrium and strain compatibility shall be maintained
in the analysis.

2.8.2 Effect of Imposed Deformation

The effect of imposed deformations due to shrinkage,

temperature change, creep, and vertical and/or horizontal
movements of supports shall be investigated.

2.9 DESIGN FOR FLEXURE WITH NO AXIAL

COMPRESSION
2.9.1 General C2.9.1

The CFFT shall be detailed, fabricated and constructed
such that full composite action is achieved between the
tube and concrete. The concrete shall be sufficiently
consolidated to achieve the levels of strength, stiffness
and ductility assumed in the design.
Consolidation of concrete within the FRP tube may
be achieved by internal or external vibration, or
through the use of self-consolidating concrete (SCC).
Selection of a consolidation method should consider
stability of the concrete and, if mechanical vibration is
performed, the ability of the FRP tube to withstand
vibration-induced deformations and stresses without
loss of structural integrity

2.9.2 Assumptions

2.9.2.1 General

The following assumptions may be used to determine

the flexural resistance of CFFTs at the limit states
indicated:

2.9.2.2 Service and Fatigue and Creep Rupture Limit

States

Composite action exists between the FRP tube

and the concrete core. Composite action shall
be verified in accordance with section 2.5.

The longitudinal strains in the concrete vary

linearly over the depth of the member and are
proportional to the distance from the neutral axis,
except in components or regions of components
for which conventional strength of materials is
inappropriate.

The tensile strength of the concrete is neglected.

The modular ratio, n = Efl/Ec is rounded to the

nearest integer; for permanent loads, a modular
ratio of 2n shall apply.
2.9.2.3 Strength and Extreme Event Limit States

Composite action exists between the FRP tube

and the concrete core. Composite action shall
be verified in accordance with section 2.5.

The longitudinal strains in the concrete vary

linearly over the depth of the member and are
proportional to the distance from the neutral axis,
except in components or regions of components
for which conventional strength of materials is
inappropriate.

The tensile strength of the concrete is neglected.

The FRP tube is sturdy enough to contribute

longitudinally in compression.

Concrete is partially confined when circular

tubes are used. The FRP tube controls failure,
whether in tension or compression. As such, the
concrete longitudinal compressive strain at
failure may exceed 0.003.

Tension failure is defined by tensile rupture of

the FRP tube under uniaxial longitudinal stresses,
while compression failure is defined by crushing
of the FRP tube under uniaxial longitudinal
stresses. Confinement in flexure is not significant
enough to weaken the tube under longitudinal
compression or longitudinal tension.

The concrete stress-strain relationship proposed

by Popovics (1973), with an extended strain
softening beyond the usual ultimate compressive
strain of 0.003, may be used in calculating the
ultimate moment capacity based on equilibrium
and strain compatibility in Article 2.9.5.1.

The stress-strain relationships of the FRP tube in

tension and compression shall be taken as linear
elastic as specified in Article 2.6.1.3.
2.9.3 Minimum Longitudinal Tension
Reinforcement

At any section of a flexural CFFT component, the FRP

reinforcement shall be adequate to develop a factored
flexural resistance Mr equal to the lesser of 1.2Mcr or 1.33
times the factored moment required by the applicable
strength load combinations specified in Article 5.7.3.3.2
of AASHTO LRFD, where:

Mcr Sc fr (2.9.3-1)

In which:

fr = 0.37 f c' for normal-weight concrete

where:

Sc = the section modulus of the extreme fiber of the

composite section where tensile stress is caused by
externally applied loads (in3).

2.9.4 Minimum Reinforcement in the Transverse or

Hoop Direction

At any cross-section, the FRP tube shall provide

sufficient confinement to the concrete core to justify the
use of the concrete stress-strain relationship employed for
the strength analysis. In addition, the minimum transverse
shear reinforcement requirement of Article 2.12.3 shall be
satisfied.

2.9.5 Factored Flexural Resistance

The factored flexural resistance of a section Mr shall be

taken as:

M r M n (2.9.5-1)
where:

Mn = the nominal moment capacity determined either by

the general method in Article 2.9.5.1 or the simplified
method in Article 2.9.5.2.

2.9.5.1 General Method

The nominal moment capacity Mn shall be calculated

based on a rigorous cross-sectional analysis that satisfies
equilibrium and strain compatibility, and utilizes
appropriate material constitutive relationships for the
concrete and FRP, and failure modes, satisfying the
provisions of Article 2.9.2.2.

2.9.5.2 Simplified Method

For tension failure-controlled circular CFFT members, C2.9.5.2

and in lieu of the rigorous equilibrium and strain
compatibility approach, Mn may be calculated as follows:
Equation 2.9.5-2 is an empirical formula developed
by best fitting of experimental data and finite element
0.815
4t f ful model results of a parametric study. Both the
M n 0.0045 D f c 100
3

Do f c
o (2.9.5-2) experimental and numerical data covered a wide range
of CFFT geometric and mechanical properties (Fam
and Son, 2008).
where:

t = the structural wall thickness of the tube (in)

Do = the outer diameter (in)

fful = the design tensile strength of FRP laminate in the

longitudinal direction (ksi)

fc = the concrete unconfined compressive strength (ksi)

2.9.6 Deformation

Control of deflections shall be considered in

accordance with AASHTO LRFD.
2.9.7 Control of Cracking C2.9.7

Crack width shall be controlled by limiting the tensile
stress level in the FRP tube in accordance with Article
2.7.2.
Crack width is indirectly limited by controlling the
stress level in the FRP tube under service load for
creep rupture and fatigue failure. Also, cracking in
CFFT members is invisible because of the tube and
does not pose aesthetic or durability concerns.

2.9.8 Stress Limit for Concrete C2.9.8

Compression stresses in the concrete shall be
investigated at the Service Limit State Load Combination
I specified in Table 3.4.1-1 of AASHTO LRFD. The
stress limitation of 0.45 fc shall apply to concrete in
compression.
This provision is to control the effects of concrete
creep, including excessive deformation with time. Note
that the unconfined concrete strength is used, given the
insignificant confinement effects in CFFTs under
flexure.

2.10 DESIGN FOR AXIAL COMPRESSION

2.10.1 General

Design of CFFT members with FRP tubes having

sufficient strength and stiffness in the hoop direction, in
accordance with Article 2.10.4, may include the effect of
confinement that leads to increasing the concrete core
compressive strength from fc to fcc.

2.10.2 Assumptions

The following assumptions may be used to determine

axial resistance of CFFTs:

Axial load is applied over the entire cross-section

of the concrete core and FRP tube

The concrete core and the FRP tube are fully

bonded. As such, the axial shortening is equal in
both the tube and concrete.
 The FRP tube is sturdy enough to contribute
longitudinally in compression.

The tube acts as a bi-axially loaded membrane

subjected to hoop tensile and longitudinal
compressive stresses. The longitudinal stress
weakens the tube in the hoop direction.

The unconfined concrete compressive strength

does not exceed 10 ksi.

Failure, and hence the ultimate axial load, occur

when the FRP tube fractures under the bi-axial
state of stress.

2.10.3 Factored Axial Compressive Resistance C2.10.3

The factored axial compressive resistance of a
nonprestressed section Pr shall be taken as:
Pr Pn
in which:
Pn is the nominal axial resistance determined as:
Pn 0.85Pn'
Several models that simulate the stress-strain
behavior of FRP-confined concrete compression
members are available in the literature, including fully
empirical models such as Samaan et al (1998) and Lam
(2.10.3-1) and Teng (2003) or semi empirical models with some
rational mechanics basis such as Spoelstra and Monti
(1999) and Fam and Rizkalla (2001).
The stress-strain model developed by Lam and Teng
(2003) has been adopted in the ACI440.2R-08 design
guide and is recommended to be used for CFFT
members in this document due to its simplicity
(2.10.3-2) combined with its reasonable accuracy.

Pn' is the axial compressive resistance and is determined The FRP tube shares some of the axial loading
for circular CFFT members from: capacity in the longitudinal direction, hence, the
second term in Eq. 2.10.3-3.

Di2
Pn' 0.85 f cc D t E fl ccu (2.10.3-3)
4

in which:

fcc is the concrete confined strength calculated as:

f cc f c 3.3 f f l (2.10.3-4)

fl is the maximum confinement pressure calculated as:

 f l min( f l1 , f l 2 ) (2.10.3-5)

2 E fh t fe
f l1 (2.10.3-6)
D

0.01
1.5
'c
fl2 (2.10.3-7)
12 fe
0.45

f ' c ' c

where:

Di = the inner diameter of the FRP tube (in)

D = the average diameter of the FRP tube (in)

Efl = the design longitudinal modulus of the tube (ksi)

f = a reduction factor equal to 0.95

fe = the effective hoop strain level in the tube at the

ultimate compressive capacity of the CFFT. fe shall be
taken as 0.55 fuh
fuh = the design ultimate tensile strain of the tube in the
hoop direction.
Efh = the design modulus of the tube in the hoop direction.
Where the confining pressure, fl is governed by the
limiting value fl2 obtained using (2.10.3-7), fcc shall be
calculated as:
The effective hoop strain level in the tube at
ultimate is reduced from the ultimate value by a factor
taken here as 0.55. This is attributed to several factors
that may include (a) the multiaxial state of stress in the
tube, in which the longitudinal compressive stresses
weaken the tube in the circumferential direction in
accordance to Tsai-Wu failure criteria (Daniel and
Ishai, 1994) and (b) stress concentration regions caused
by cracking of concrete as it dilates.

fcc fc 0.01 E2 (2.10.3-8)

in which:

f *cc f 'c
E2 (2.10.3-9)
ccu
 f *cc fc 3.3 f fl1
(2.10.3-10)
In the presence of shear, the tube resists additional
In presence of a shear force acting on the section, fe circumferential tensile stresses. As such, the full tensile
shall be replaced by fe fv as per section 2.12.2. capacity may not be available for confinement.

The maximum longitudinal compressive strain ccu

shall be determined as:

f l fe
0.45

ccu
c 1.5 12 0.01 (2.10.3-11)
f c c

in which:

f c n
c (2.10.3-12)
Ec n 1

f c
n 0. 8 (2.10.3-13)
2.5

E c 1265 f c 1000 (2.10.3-14)

2.10.4 Minimum Reinforcement in the Hoop C2.10.4

Direction
The FRP tube shall have sufficient thickness and
stiffness to produce a confinement pressure fl calculated
according to equation 2.10.3-5 of not less than 0.08 times
the unconfined concrete strength fc.
Based on the tests by Lam and Teng (2003), the
ratio of confinement pressure to unconfined concrete
strength should not be less than 0.08. This is the
minimum level of confinement required to assure a
nondescending second branch in the stress-strain curve.

2.10.5 Stress Limit for Concrete C2.10.5

At service limit states, the axial compressive stress in
concrete shall not exceed 0.65fc.
To ensure that radial cracking of concrete will not
occur under service loads, the transverse strain in
concrete shall remain below its cracking strain at
service load levels. This corresponds to limiting the
compressive stress in the concrete to 0.65fc (ACI
440.2R-08). By maintaining the specified stress in the
concrete at service, the stress in the FRP tube will be
relatively low. Service load tensile stress in the tube
should never exceed the creep-rupture stress limit.
2.11 DESIGN FOR COMBINED FLEXURE AND
AXIAL COMPRESSION

2.11.1 General C2.11.1

The CFFT shall be detailed, fabricated and constructed
such that full composite action is achieved between the
tube and concrete. The concrete shall be sufficiently
consolidated to achieve the levels of strength, stiffness
and ductility assumed in the design. Flexure may be
induced by an eccentric axial load and/or by transverse
loading.
Design of CFFT members with FRP tubes having
sufficient strength and stiffness in the hoop direction, in
accordance with Article 2.10.4, may include the effect of
confinement that may lead to increasing the concrete core
compressive strength from fc to fcc, depending on the
eccentricity e, where e is the ratio of the applied moment
Mu, including slenderness effect, to the compression load
Pu.
Fam et al (2003) have demonstrated experimentally
and analytically that the effect of concrete confinement
is reduced as the eccentricity of axial load (or the ratio
of moment to axial load) is increased. The extreme
limits are full confinement at concentric loading and
partial confinement at pure bending, where it is
hypothesized that the peak concrete strength remains at
the unconfined level of fc but the material experiences
an extended strain softening beyond the typical
crushing strain of 0.003. In this case failure in
compression is governed by crushing of the tube under
longitudinal compressive stresses.

2.11.2 Assumptions

The following assumptions may be used in determining

the resistance of CFFTs to combined axial and flexural
loading:

Composite action exists between the FRP tube

and the concrete core. Composite action shall
be verified in accordance with section 2.5.

The longitudinal strains in the concrete vary

linearly over the depth of the member and are
proportional to the distance from the neutral axis,
except in components or regions of components
for which conventional strength of materials is
inappropriate.

The tensile strength of the concrete is neglected.

The FRP tube has sufficient strength and

stiffness to contribute longitudinally in
 compression.

Depending on the eccentricity, concrete

experiences different levels of confinement when
circular tubes are used. The FRP tube controls
failure, whether in tension or compression. As
such, the concrete longitudinal compressive
strain at failure may exceed 0.003.

The tube acts as a bi-axially loaded membrane on

the compression side subjected to hoop tensile
and longitudinal compressive stresses. The
longitudinal stress weakens the tube in the hoop
direction.

Tension failure is defined by tensile rupture of

the FRP tube under uniaxial longitudinal stresses,
while compression failure occurs when the FRP
tube fractures under the bi-axial state of stress.

In the general method in Article 2.11.3.1, the

concrete stress-strain relationship based on the
variable confinement model of Fam et al (2003)
shall be used in calculating the ultimate moment
and axial load based on equilibrium and strain
compatibility. For each eccentricity, a different
stress-strain curve with a different ultimate
longitudinal compressive strain is used.

In the simplified method in Article 2.11.3.2, a

single concrete stress-strain relationship based on
full confinement under concentric compression
shall be used for all ranges of eccentricity
between concentric compression and the
balanced point. The model by Lam and Teng
(2003) in Article 2.10.3 may be used.

The stress-strain relationships of the FRP tube in

tension and compression shall be taken as linear
elastic as specified in Article 2.6.1.3.

2.11.3 Factored Resistance C2.11.3.

The designer shall use one of the two methods stated Either the rigorous method of Article 2.11.3.1 or the
below to establish factored resistance under combined more conservative simplified method of Article
2.11.3.2 may be used to determine the interaction
bending and axial compression. diagram for combined compression and bending.

2.11.3.1 General Method

The nominal moment and axial load capacity ( Pn,

Mn) shall be calculated based on a rigorous cross-
sectional analysis that satisfies equilibrium and strain
compatibility, and utilizes appropriate material
constitutive relationships for the concrete and FRP, and
failure modes, satisfying the provisions of Article 2.11.2.

2.11.3.2 Simplified Method C2.11.3.2

In lieu of the general method in Article 2.10.3.1, the This simplified method is approximate and is based
nominal moment and axial load ( Pn, Mn) interaction on an equivalent rectangular approximation of the
resistance envelope may conservatively be simplified as a confined concrete stress distribution.
bi-linear curve between the pure axial load point ( Pn, 0),
the balanced point( Pnb, Mnb) and the pure bending
point (0, Mn). Where the equations of this section
calculate a balanced axial load less than 0 (i.e. axial
tension) the interaction diagram shall be approximated
using a straight line between the pure axial load point and
pure bending point.

The pure axial compression resistance Pn shall be

calculated in accordance with Article 2.10.3. The pure
flexural resistance Mn shall be calculated in accordance
with Article 2.8.5. The balanced point resistance shall be
calculated as follows:

Dt D2
Pnb E fl ccu (2c D) i 1 f cc (2 a Sin 2 a )
2c 8
(2.11.3.2-1)

D3 t D3
M nb E fl ccu i 1 f cc Sin 3 a
8c 12

(2.11.3.2-2)

in which:
 t ccu
c ( Do ) (2.11.3.2-3)
2 ccu ful

21c
a ArcCos(1 ) (2.11.3.2-4)
Di

where:

1 0.83

1 0.88

ccu is calculated using equation 2.10.3-11

2.11.4 Evaluation of Slenderness Effects

Evaluation of member slenderness and moment

magnification shall follow the provisions of the AASHTO
LRFD as applicable to reinforced concrete members. The
value of EI for use in determining Pe shall be calculated
using Efl and Ifl in place of Es and Is, respectively when
applying provisions of the AASHTO LRFD, where Ifl is
the moment of inertia of the cross-section of the FRP
tube.

2.11.5 Minimum Reinforcement

At any cross-section of the CFFT member, the FRP

tube shall provide sufficient confinement to the concrete
core to justify the use of the concrete stress-strain
relationship employed for the strength analysis. The
provisions of article 2.10.4 shall apply. Also, the
minimum transverse shear reinforcement requirement of
Article 2.12.3 shall be satisfied.

At any cross-section of the CFFT member, the FRP

tube shall be adequate to develop a factored flexural
resistance Mn equal to the lesser of 1.2Mcr or 1.33 times
the factored moment required by the applicable strength
load combinations specified in the AASHTO LRFD. Mcr
shall be calculated taking into consideration the presence
of the factored axial load Pn.

2.11.6 Deformation

Control of deflections shall be considered in

accordance with AASHTO LRFD.

2.11.7 Control of Cracking C2.11.7

Crack width is indirectly controlled by limiting the Cracking in CFFT members is not visible and does
tensile stress level in the FRP tube in accordance with not cause aesthetic or durability concerns.
Article 2.7.2.

2.11.8 Stress Limit for Concrete

Compression stresses in the concrete shall be

investigated at the Service Limit State Load Combination
I specified in Table 3.4.1-1 of AASHTO LRFD. The
stress limitation of 0.45 fc shall apply to CFFT members
governed by tension failure and 0.65 fc shall apply to
CFFT members governed by compression failure in
accordance with Article 2.10.5.

2.12 DESIGN FOR SHEAR EFFECTS

2.12.1 General C2.12.1

The critical section for shear near a support in a CFFT In CFFT members, longitudinal and transverse
member shall be determined in accordance with the (shear) reinforcement are integral and coupled in the
provisions of AASHTO LRFD for reinforced concrete form of the same FRP tube. As such, combined shear
members. and flexure or shear and axial loading places the tube
in a state of bi-axial stresses.
In lieu of the methods specified herein, the resistance
of members in shear may be determined by satisfying the
conditions of equilibrium and compatibility of strains and
by using experimentally verified stress-strain relationships
for the transverse reinforcement provided by the FRP tube
and for diagonally cracked concrete.
2.12.2 Nominal Shear Resistance C2.12.2

The factored shear resistance of a section Vr shall be Compared with a steel-reinforced section, with
taken as: equal areas of longitudinal reinforcement, a cross
section reinforced with FRP flexural reinforcement
V r V n (2.12.2-1)
after cracking has a smaller depth to the neutral axis
because of the lower modulus of FRP reinforcement.
As a result, the shear resistance provided by both
The nominal shear resistance, Vn, shall be determined aggregate interlock and compressed concrete is
as: reduced. Furthermore, the contribution of the FRP
flexural reinforcement by dowel action is less than
Vn Vc V f Vs steel reinforcement because of their lower strength and
(2.12.2-2)
stiffness in the transverse direction (ACI 440.1R-06).
In case of CFFTs, the shear resistance of the FRP tube
When no longitudinal steel reinforcement is used in the is simplified to a circumferential tensile stress.
CFFT member, the concrete shear strength, Vc (kip), shall
be taken as:

Vc 0.158 f c' Ac
(2.12.2-3)

where:

Ac = effective concrete area, which is the area of the

compression zone of the cracked transformed
section under service axial load and moment
acting on the section of interest based on the
assumption of linearly elastic material response.
In lieu of the detailed sectional analysis, and
assuming no axial compression force is acting on
the section, Ac may be calculated based on a
neutral axis depth of 0.3Do (in2)

f c' = unconfined concrete compressive strength (ksi)

If internal longitudinal steel reinforcement is used, Vc,

shall be taken as defined in AASHTO LRFD.

The FRP tube shear strength, Vf, shall be taken as:

V f 2td v f fe
(2.12.2-4)

where:
t = the structural wall thickness of the FRP tube (in)

dv = the effective depth, taken as 0.8 Do (in)

The tensile hoop stress in the FRP tube, ffe, shall be

limited to:
The hoop tensile strain in the FRP tube under shear
f fe 0.004 E fh is limited to 0.004 to control shear crack width and
(2.12.2-5) maintain shear integrity in the concrete.

where: The effective Youngs modulus of the tube laminate

Efh = the elastic modulus of the tube laminate in the hoop
direction (ksi)
The combined effects of shear and confinement on the
hoop strain level in the FRP tube shall be considered in
checking the ultimate strain in the tube. This may be
achieved by deducting the hoop strain due to shear force
and limiting the total hoop strain to 0.004 to ensure shear
integrity of the confined concrete core, in both the general
and simplified methods. In the simplified method, this
may be achieved by substituting the effective hoop strain
fe in Eq. 2.10.3-6 and 2.10.3-7 by fe fv where the
hoop strain resulting from shear at the same section is
calculated as follows:
in the circumferential direction, Efh, is used along with
the total structural wall thickness t of the tube to
determine its shear contribution. Alternatively, in the
special case of [0/90] cross-ply laminates, the thickness
and modulus of the combined [90] hoop layers only
could be used.
The hoop tensile strain in the FRP tube due to shear
reduces the capacity of the tube for confinement under
axial compression loads. This must be accounted for.

Vu Vc
fv (2.12.2-6)
2td v E fh

subject to the limitation:

fe fv 0.004
(2.12.2-7)

where:

Vu = the factored shear force acting on the section of

interest (kip)

If axial compression load is applied with shear, the

0.004 strain in Eq. 2.12.2-5 shall be reduced by the
amount of hoop strain in the tube due to confinement
under the applied factored axial compression load

If internal transverse steel reinforcement is used, Vs,

shall be taken as defined in AASHTO LRFD.

2.12.3 Minimum Shear Reinforcement C2.12.3

Transverse shear reinforcement is required where:
Vu 0.5Vc
(2.12.3-1)
The minimum tube thickness tmin to satisfy minimum
shear reinforcement shall be taken as:
The minimum requirements for transverse
reinforcement area is to prevent or restrain shear failure
in members where the sudden formation of cracks can
lead to excessive distress (joint ACI-ASCE Committee
426 (1974)). In the case of CFFT, the minimum
reinforcement is reflected in terms of a minimum
thickness of the FRP tube

f c'
t min 0.0079 Do
f fuh
(2.12.3-2)

where:

ffuh = the design ultimate tensile strength of the tube in the

hoop direction (ksi), calculated from Equation 2.6.1.2-2

Do = the outer diameter of the FRP tube (in)

2.13 CONNECTIONS
C2.13

CFFT members subjected to flexure, shear and/or axial

CFFT members secured into concrete footings, pile
loads may be connected to other reinforced concrete (RC)
caps or cap beams of adequate size can develop their
structural components of adequate size such as footings,
full flexural capacity before a premature bond-slip
pile caps or beams by direct embedment of the CFFT into
failure occurs, provided that the embedment length is
the RC member without the use of additional
sufficient. Equations 2.12-1 and 2.12-2 are based on
reinforcement.
the work by Sadeghian and Fam (2010), including a
1.5 safety factor, where the resistance is provided by a
For CFFT members subjected to flexure without axial
combined bearing and friction mechanism.
loads, the design embedment length X, in inches, of the
CFFT member into the RC structural component shall be
The concrete block in which the CFFT member is
calculated as:
fixed into needs to be checked under the internal

3 M n bearing and friction forces induced by the CFFT

X 1.5 (2.13-1)
D f c member. A procedure has been recommended by
Sadeghian and Fam (2010).

where:

Mn = the nominal moment capacity of the CFFT member,

calculated in accordance with Article 2.8.5 (k-in)

fc = the concrete compressive strength of the RC

component (ksi)

D = the average diameter of the FRP tube (in)

For CFFT members subjected to combined flexure and

axial compression, the design embedment length X of the

CFFT member into the RC structural component shall be

the maximum of X calculated using equation 2.12-1, and:
 1

X 1.5
3 M nb 1 0.5 M nb 3
D f c Pnb D 8

(2.13-2)

where:

Pnb = the balanced axial load (kip), determined from

equation 2.10.3.2-1

Mnb = the balanced moment (k-in), determined from

equation 2.10.3.2-2

fc = concrete compressive strength of the concrete

component (ksi)

Depending on the size of the RC member

accommodating the CFFT embedment, confinement
reinforcement may be needed in the RC member to
control splitting cracks.

2.14 REFERENCES

ACI (2008a). "Building Code Requirements for Structural Concrete and Commentary." ACI Publication No. 318-08.
American Concrete Institute, Farmington Hills, MI.

ACI (2008b). "Guide for the Design and Construction of Externally Bonded FRP Systems for Strengthening Concrete
Structures. ACI Publication No. 440.2R-08. American Concrete Institute, Farmington Hills, MI.

Burgueo, R. (1999). "System Characterization and Design of Modular Fiber Reinforced Polymer (FRP) Short- and
Medium- Span Bridges." (Doctoral dissertation, University of California, San Diego, 1999). (UMI No.
9928617)

Burgueo, R. and Bhide, K. (2006). Shear Response of Concrete-Filled FPR Composite Cylindrical Shells. Journal
of Structural Engineering, ASCE, 132(6): 949-960.

Davol, A., Burgueo, R. and Seible, F. (2001). Flexural behavior of circular concrete filled FRP shells. Journal of
Structural Engineering, ASCE, 127(7): 810-817.
Fam, A. (2000). "Concrete-Filled Fibre-Reinforced Polymer Tubes for Axial and Flexural Structural Members."
(Doctoral dissertation, University of Manitoba, 2000).

Fam, A., Cole, B. and Mandal, S. (2007). Composite tubes as an alternative to steel spirals for concrete members in
bending and shear. Construction and Building Materials, 21(2007): 347-355.

Fam, Amir Z. and Rizkalla, Sami H., (2001) Confinement Model for Axially Loaded Concrete Confined by FRP
Tubes, ACI Structural Journal, 98(4):251-461.

Lam, L. and Teng, J. (2003) Design-Oriented Stress-Strain Model for FRP-Confined Concrete, Construction and
Building Materials, Vol 17, pp.471-489.

Fam, A., Flisak, B. and Rizkalla, S. (2003) Experimental and Analytical Investigations of Concrete-Filled Fiber-
Reinforced Polymer Tubes Subjected to Combined Bending and Axial Loads, ACI Structural Journal,
100(4):499-509.

Mirmiran, A. and Shahaway,M. (1997). Behavior of concrete columns confined by fiber composites. Journal of
Structural Engineering, ASCE, 123(5): 583-590.

Mandal, S., Hoskin, A. and Fam, A. (2005) Influence of Concrete Strength on Confinement Effectiveness of Fiber-
Reinforced Polymer Circular Jackets, ACI Structural Journal, 102(3):383-392.

Spoelstra, M. R. and Monti, G. (1999) FRP-Confined Concrete Model, ASCE Journal of Composites for
Construction, 3(3):143-150.
 Section 3

MATERIAL SPECIFICATIONS

3.1 SCOPE

This section covers provisions governing testing,

evaluation, and acceptance, and describes permitted
constituent materials for use in the manufacturing of
CFFTs.

3.2 DEFINITIONS

Binder- Chemical treatment applied to the random arrangement of fibers to give integrity to mats, roving, and fabric.
Specific binders are utilized to promote chemical compatibility with the various laminating resins used.

Catalyst- A substance that accelerates a chemical reaction and enables it to proceed under conditions more mild than
otherwise required, and which is not, itself, permanently changed by the reaction. See Initiator or Hardener.

Epoxy - A thermosetting polymer that is the reaction product of epoxy resin and an amino hardener.

Fiber - Any fine thread-like natural or synthetic object of mineral or organic origin. Note: This term is generally used
for materials whose length is at least 100 times its diameter.

Filament - See Fiber.

Filament Winding A process that involves winding a resin saturated strand of filament around a rotating mandrel.

Filler - A relatively inert substance added to a resin to alter its properties or to lower cost or density. Sometimes the
term is used specifically to mean particulate additives. Also called extenders.

Glass Transition Temperature - The midpoint of the temperature range over which an amorphous material (such as
glass or a high polymer) changes from (or to) a brittle, vitreous state to (or from) a plastic state.

Hardener - In a two-component adhesive or coating, the chemical component that causes the resin component to cure.

Initiator - A source of free radicals, which are groups of atoms that have at least one unpaired electron, used to start
the curing process for unsaturated polyester and vinyl ester resins. Peroxides are the most common source of free
radicals. See Catalyst.

Matrix - The material in which reinforcing fiber of a composite system is embedded.

Monomer - An organic molecule of relatively low molecular weight that creates a solid polymer by reacting with itself
or other compounds of low molecular weight, or both.

Nominal durability Property - A value no greater than the average of at least 25 specimens tested according to a
specified method. This definition is applicable to moisture absorption and resistance to alkaline environment.

Polyester - One of a large group of synthetic resins, mainly produced by reaction of dibasic acids with dihydroxy
alcohols; commonly prepared for applications by mixing with a vinyl-group monomer and free-radical catalysts at
ambient temperatures and used as binders for resin mortars and concretes, fiber laminate (mainly glass), adhesives,
and the like. Commonly referred to as "unsaturated polyester".

Polymer - A high molecular weight organic compound, natural or synthetic, containing repeating units.

Polymerization - The reaction in which two or more molecules of the same substance combine to form a compound
containing the same elements and in the same proportions but of higher molecular weight.

Process Modification - Any change in properties of constituents, proportions of constituents, or production

parameters.

Production Lot Any lot of CFFTs produced from start to end with the same constituent materials used in the same
proportions without changing any production parameter.

Pultrusion - A continuous process for manufacturing composites that have a uniform cross-sectional shape. The
process consists of pulling a fiber-reinforcing material through a resin impregnation bath, then through a shaping die
where the resin is subsequently cured.

Resin - Polymer material such as polyester, vinylester, or epoxy used as the matrix in a composite to bind the
reinforcement fibers together.

Resin Infusion - A process for manufacturing composites in which dry fibers are placed within a closed cavity and
subsequently wetted out with resin introduced via pressure. Also known as Vacuum Infusion or Resin Transfer
Molding.

Roving - A number of tows or ends of fibers collected into a parallel bundle.

Sizing - Surface treatment or coating applied to filaments to improve the filament-to-resin bond and to impart
processing and durability attributes.

Strength Property as Reported by Manufacturer - A value no greater than the average minus three times the standard
deviation of at least 25 specimens tested according to a specified method. This definition is applicable to tensile
strength, compressive strength, and shear strength of the composite.

Tensile Modulus of Elasticity - A value no greater than the average of at least 25 specimens tested according to a
specified method.

Thermoset - Resin that is formed by cross-linking polymer chains that cannot be melted and recycled because the
polymer chains form a three-dimensional network.
Tow - A bundle of continuous filaments

Vinyl Ester - A thermosetting resin containing both vinyl and ester components, and cured by additional
polymerization initiated by free-radical generation. Vinyl esters are used as binders for fiber laminates and adhesives.

Yarn - An assemblage of twisted filaments formed into a continuous length that is suitable for use in weaving textile
materials.

3.3 NON-STANDARD DOCUMENTS

Test methods from ACI 440.3R-04 are referenced in

this Section when an ASTM standard is not available.
Because these test methods are not written in mandatory-
language, purchaser and manufacturer shall agree on the
protocol to be used.

3.4 MATERIAL AND MANUFACTURE

3.4.1 Fibers

Fiber sizings and coupling agents shall be compatible

with the resin system used to impregnate them.

3.4.2 Matrix Resins

Commercial grades of vinyl ester and epoxy resin

systems are permitted provided the finished product meets
the physical and durability requirements of this Section.

Styrene is permitted to be added to the polymer resin

during processing. Added styrene shall be less than 10
percent by mass of the polymer resin. The amount of
styrene, as a mass percentage of the polymer resin, added
during processing shall be reported.

3.4.3 Fillers and Additives

Commercial grade inorganic fillers such as kaolin clay,

calcium carbonate, and alumina trihydrate shall not
exceed 20 percent by mass of the polymer resin
constituent.
 Commercial grade additives and process-aids, such as
release agents, low profile shrink additives, initiators,
promoters, hardeners, catalysts, pigments, fire-retardants,
and ultra-violet inhibitors are permitted and depend on the
processing method. Shrink additives, if used, shall be less
than 20 percent by mass of the polymer resin.

Commercial grade inorganic or organic non-woven

surfacing mats or veils are permitted.

3.4.4 Manufacturing Process

The manufacturer shall produce the FRP tube by any

composite manufacturing method that is capable of
producing a tube that meets the physical, mechanical, and
durability requirements of this Specification. Process
modifications are not permitted during the production lot.
The manufacturer shall document the process used and
report the date of production and production lot size.

3.5 PHYSICAL PROPERITES

3.5.1 Fiber Content

Fiber content shall be measured by ASTM D 3171 or

ASTM D 2584. Fiber content shall be high enough to
meet the mechanical property requirements of the FRP
tube laminate. The manufacturer shall report the fiber
content of the end product by volume or by mass in
accordance to the method used.

3.5.2 Glass Transition Temperature C3.5.2

The glass transition temperature of the resin shall not FRP tubes should not be used in environments with
be less than 212F. The glass transition temperature shall a service temperature higher than the glass transition
be measured on a coupon cut from an FRP tube laminate temperature of the resin used for their manufacturing.
using either the differential scanning calorimetry (DSC)
method in ASTM E 1356 or the dynamic mechanical
analysis (DMA) method in ASTM E 1640. When using
the DSC method, test results for both the first scan
(according to ASTM 1356) and the second scan shall be
reported.
3.5.3 Longitudinal and Transverse Coefficients of
Thermal Expansion (CTE)

The coefficient of Thermal Expansion (CTE) of the

tube may vary in the longitudinal and circumferential
directions of the tube depending on the laminate
architecture and type of fibers and resins.

3.6 MECHANICAL PROPERITES

3.6.1 Tensile Strength C3.6.1

The tensile strength shall be determined for two Acceptance limits for FRP tube mechanical
purposes: (1) product certifications from multiple properties shall be agreed upon by the owner, and
production lots; and (2) manufacturer's quality control material supplier, and shall be included in the contract
(QC) and purchaser's quality assurance (QA) for each documents.
production lot. The tensile strength shall be determined
for both the longitudinal and hoop directions of the tube
laminate.

The tensile strength as reported by the manufacturer

for product certification shall be measured according to
ASTM Test Method D 3039, or other tension test method
designed to determine tensile properties of composite
laminates at a frequency and number of specimens as
specified in Article 3.8.1. The manufacturing method of
the test specimens should be similar to the method used to
manufacture the CFFT. Approval of the engineer shall be
required for acceptance of test specimens produced by a
different manufacturing method. The manufacturer shall
report the individual test results.

The tensile strength for QC/QA purposes of a

production lot shall be measured according to ASTM Test
Method D 3039, or other tension test method designed to
determine tensile properties of composite laminates at a
frequency and number of specimens as specified in
Article 3.8.1. The manufacturer shall report the
individual test results. The strength of each specimen
shall be reported and shall not be less than the tensile
strength reported by the manufacturer. If the strength is
less than the tensile strength as reported by manufacturer,
the production lot shall be rejected.
3.6.2 Tensile Modulus of Elasticity

The tensile modulus of elasticity shall be determined

for the purpose of product certification. The modulus of
elasticity shall be determined for both the longitudinal and
hoop directions of the tube laminate. The tensile modulus
of elasticity may be calculated using equations 2.5.1.3-1
and 2.5.1.3-2. The tensile modulus of elasticity may also
be derived from specimens tested in accordance with
ASTM Test Method D 3039, or other tension test method
designed to determine tensile properties of composite
laminates at a frequency and number of specimens as
specified in Article 3.8.1. The manufacturing method of
the test specimens should be similar to the method used to
manufacture the CFFT. Approval of the engineer shall be
required for acceptance of test specimens produced by a
different manufacturing method. The manufacturer shall
report the individual test results.

3.6.3 Ultimate Tensile Strain

The ultimate tensile strain shall be determined for the

purpose of product certification. The ultimate tensile
strain shall be determined for both the longitudinal and
hoop directions of the tube laminate. The ultimate tensile
strain shall be derived from specimens tested in
accordance with ASTM Test Method D 3039, or other
tension test method designed to determine tensile
properties of composite laminates at a frequency and
number of specimens as specified in Article 3.8.1. The
manufacturing method of the test specimens should be
similar to the method used to manufacture the CFFT.
Approval of the engineer shall be required for acceptance
of test specimens produced by a different manufacturing
method. The manufacturer shall report the individual test
results.

For laminates that display a linear-elastic response, the

tensile strain may be calculated based on the tensile
strength and the tensile modulus of elasticity as reported
by the manufacturer.
3.6.4 Compressive Strength

The compressive strength shall be determined for the

purpose of product certification. The compressive
strength shall be determined for the longitudinal direction
of the tube laminate. The compressive strength shall be
derived from specimens tested in accordance with ASTM
Test Method D 6641, or other compression test method
designed to determine compressive properties of
composite laminates at a frequency and number of
specimens as specified in Article 3.8.1. The
manufacturing method of the test specimens should be
similar to the method used to manufacture the CFFT.
Approval of the engineer shall be required for acceptance
of test specimens produced by a different manufacturing
method. The manufacturer shall report the individual test
results.

3.6.5 Ultimate Compressive Strain

The ultimate compressive strain shall be determined

for the purpose of product certification. The ultimate
compressive strain shall be determined for the
longitudinal direction of the tube laminate. The ultimate
compressive strain shall be derived from specimens tested
in accordance with ASTM Test Method D 6641, or other
compression test method designed to determine
compressive properties of composite laminates at a
frequency and number of specimens as specified in
Article 3.8.1. The manufacturing method of the test
specimens should be similar to the method used to
manufacture the CFFT. Approval of the engineer shall be
required for acceptance of test specimens produced by a
different manufacturing method. The manufacturer shall
report the individual test results.

3.7 DURABILITY PROPERTIES

3.7.1 Moisture Absorption

Moisture absorption tests shall be determined for the

purpose of product certification in accordance with
ASTM D 570 using a water temperature of 122F at a
frequency and number of specimens as specified in
Article 3.8.1. The manufacturing method of the test
specimens should be similar to the method used to
manufacture the CFFT. Approval of the engineer shall be
required for acceptance of test specimens produced by a
different manufacturing method. The individual moisture
absorption test results shall be reported and their average
shall be less than 1.0 percent.

3.7.2 Resistance to Alkaline Environment

Resistance to alkaline environment tests shall be

determined for the purpose of product certification in
accordance with ACI 440.3R Test Method B.6 at a
frequency and number of specimens as specified in
Article 3.8.1. The manufacturing method of the test
specimens should be similar to the method used to
manufacture the CFFT. Approval of the engineer shall be
required for acceptance of test specimens produced by a
different manufacturing method. The manufacturer shall
report the individual test results and the test method.
Minimum strength retention values have not yet been
established and should be agreed upon between the
manufacturer and the purchaser.

3.8 SAMPLING

3.8.1 Sampling Frequency and Number of

Specimens

For the determination of each of the mechanical and

durability properties for QC/QA tests, at least five
samples of sufficient size to perform the required tests
shall be obtained from each production lot.

For the determination of each of the mechanical and

durability properties for product certification, at least 25
samples of sufficient size to perform the required tests
shall be obtained from five different production lots of the
same laminate architecture. Tests for the determination of
the mechanical and durability properties for product
certification shall be repeated at least every three years.

3.8.2 Rejection
 The owner has the option to reject material that fails to
conform to the requirements of this Specification as per
the contract documents.

3.9 CERTIFICATION

3.9.1 Documents

When specified in the purchase order or contract, the

purchaser shall be furnished with:

Documentation showing the constituents, their

quantities, and their properties, as provided by
the suppliers; and

Certification stating samples representing each

production lot have been tested and inspected as
indicated in this Specification and the
requirements have been met.

Certifications shall bear the signature of a responsible,

authorized representative of the FRP tube manufacturer.

3.9.2 QC/QA

When specified in the purchase order or contract, a

report of the results of QC/QA tests shall be furnished.
Test reports may be transmitted to the purchaser by
electronic services. The content of the electronically
transmitted document shall conform to any existing
agreement between the purchase and the seller. The test
report shall include the following for a given production
lot traceable to the identifying marking on the FRP tubes
furnished:

Test date and laboratory where testing was

performed;

Laminate architecture and thickness

Production lot number or identifying marking

 Test standard used or description of testing
method

Description of test specimens tested(i.e.,

specimen dimensions)

Results of individual test values for each

specimen

Description of testing apparatus (i.e., load frame

type and capacity, extensometer, data acquisition
software, and other pertinent details).

3.9.3 Product Certification

When specified in the purchase order or contract, a

report of the test results of product certification shall be
furnished. Test reports may be transmitted to the
purchaser by electronic services. The test report shall
include the following for the FRP tubes furnished:

Test date and laboratory where testing was

performed;

Resin matrix, fillers, and additives;

Fiber type(s) used

Description of test specimens tested (i.e.,

specimen dimensions)

Description of the testing method when the

choice of more than one test method or variation
has been specified in this document;

Laminate architecture and thickness;

Physical properties: fiber content with indication

of reference by mass or volume and glass
transition temperature. Mechanical properties as
reported by the manufacturer: tensile strength
and modulus of elasticity in both longitudinal
and hoop directions

Nominal durability properties: moisture

absorption, and resistance to alkaline
 environment.

Description of testing apparatus (i.e., load frame

type and capacity, extensometer, data acquisition
software, and other pertinent details).

3.9.4 Markings

Each tube shall be identified with a corresponding

production lot number. Production lot numbers shall be
printed on each tube by means of a durable tag or
marking. Each tube shall be labeled with the following
information:
A symbol to identify the manufacturer;

The nominal tube diameter and thickness;

A marking to indicate the laminate longitudinal
tensile strength as reported by the manufacturer;

A marking to indicate the longitudinal modulus

of elasticity as reported by the manufacturer.

Tube markings may take the form of bar coding,

physical marking on the product, color coding by any
method, or any means determined by the manufacturer
such that a sample of tube is identified with a particular
production lot.
3.10 REFERENCES

ACI Committee 440. 2004. Guide to Test Methods for Fiber-Reinforced Polymer (FRPs) for Reinforcing or
Strengthening Concrete Structures. ACI 440.3R-04. American Concrete Institute, Farmington Hills, MI.

ASTM D 570. Test Method for Water Absorption of Plastics. American Society for Testing and Materials, West
Conshohocken, PA.

ASTM D 2584. Standard Test Method for Ignition Loss of Cured Reinforced Resins. American Society for Testing
and Materials, West Conshohocken, PA.

ASTM D 3039. Standard Test Method for Tensile Properties of Polymer Matrix Composite Materials. American
Society for Testing and Materials, West Conshohocken, PA.

ASTM D 3171. Standard Test Methods for Constituent Content of Composite Materials. American Society for
Testing and Materials, West Conshohocken, PA.

ASTM E 1356. Standard Test Method for Assignment of the Glass Transition Temperatures by Differential Scanning
Calorimetry. American Society for Testing and Materials, West Conshohocken, PA.

ASTM E 1640. Standard Test Method for Assignment of the Glass Transition Temperature by Dynamic Mechanical
Analysis. American Society for Testing and Materials, West Conshohocken, PA.
2012 AASHTO BRIDGE COMMITTEE AGENDA ITEM: 44

SUBJECT: LRFD Bridge Design Specifications: Section 3, Article 3.4.1

TECHNICAL COMMITTEE: T-5 Loads / T-15 Substructures and Retaining Walls

REVISION ADDITION NEW DOCUMENT

DESIGN SPEC CONSTRUCTION SPEC MOVABLE SPEC

MANUAL FOR BRIDGE SEISMIC GUIDE SPEC COASTAL GUIDE SPEC
EVALUATION OTHER

DATE PREPARED: 1/6/12

DATE REVISED: 5/14/12

AGENDA ITEM:
Item #1

Revise the Strength IV bullet in the 2nd paragraph of Article 3.4.1:

Strength IV Load combination relating to very high dead load to live load force effect ratios in bridge
superstructures.

Item #2

Revise the last sentence of the 5th paragraph of Article C3.4.1:

This load combination is not applicable to can control during investigation of construction stages, substructures,
and bearing design. Other load combinations adequately address substructures and bearings.

OTHER AFFECTED ARTICLES:

None

BACKGROUND:
Eliminate discrepancies and misunderstanding in the application of the Strength IV load combination such as:
1. Strength IV used in a cantilever retaining wall design example in NHI Course No. 132068, May 2001
2. Strength IV used as a construction load combination when that topic is addressed in Article 3.4.2.
3. Strength IV used in bearing design.
It was only intended for bridge superstructures with high dead to live load ratios.

ANTICIPATED EFFECT ON BRIDGES:

Minimize overdesign and the number of load combinations required.

REFERENCES:
Load and Resistance Factor Design (LRFD) for Highway Bridge substructures: Reference Manual and
Participant Workbook, NHI Course No. 132068, May 2001.
OTHER:
None
2012 AASHTO BRIDGE COMMITTEE AGENDA ITEM: 45

SUBJECT: LRFD Bridge Design Specifications: Section 3, Article 3.6.1.1.1 (WAI 43)

TECHNICAL COMMITTEE: T-5 Loads

REVISION ADDITION NEW DOCUMENT

DESIGN SPEC CONSTRUCTION SPEC MOVABLE SPEC

MANUAL FOR BRIDGE SEISMIC GUIDE SPEC COASTAL GUIDE SPEC
EVALUATION OTHER

DATE PREPARED: 9/30/11

DATE REVISED: 5/3/12

AGENDA ITEM:
In Article 3.6.1.1.1, revise the 1st paragraph as follows:

Generally, Unless specified otherwise, the width of the design lanes should be taken as 12.0 ft. Tthe number of
design lanes should be determined by taking the integer part of the ratio w/12.0, where w is the clear roadway width
in ft feet between curbs and/or barriers. Possible future changes in the physical or functional clear roadway width of
the bridge should be considered.

OTHER AFFECTED ARTICLES:

None

BACKGROUND:
Previously, the standard design lane width of 12.0 ft was implied but not stated. This change clarifies the issue.

ANTICIPATED EFFECT ON BRIDGES:

None

REFERENCES:
None

OTHER:
None
2011 AASHTO BRIDGE COMMITTEE AGENDA ITEM: 46

SUBJECT: LRFD Bridge Design Specifications: Section 4, Various Articles (WAI 42)

TECHNICAL COMMITTEE: T-5 Loads

REVISION ADDITION NEW DOCUMENT

DESIGN SPEC CONSTRUCTION SPEC MOVABLE SPEC

MANUAL FOR BRIDGE SEISMIC GUIDE SPEC COASTAL GUIDE SPEC
EVALUATION OTHER

DATE PREPARED: 1/2/11

DATE REVISED: 5/3/12

AGENDA ITEM:

Item #1

In Article 4.3 Notation, modify g as follows:

g = live load distribution factor representing the number of design lanes; acceleration of gravity (ft/sec.2)
(4.6.2.2.1) (C4.7.4.3.2)

In Article 4.6.2.2.1, modify g as follows:

g = live load distribution factor representing the number of design lanes

Item #2

Revise the titles of the Tables in Articles 4.6.2.2.2 and 4.6.2.2.3 as follows:

Table 4.6.2.2.2a-1 Live Load Distribution Factor of Live Load per Lane for Moment and Shear in Interior Beams
with Wood Decks

Table 4.6.2.2.2b-1 Live Load Distribution Factor of Live Loads per Lane for Moment in Interior Beams

Table 4.6.2.2.2c-1 Live Load Distribution Factor of Live Load per Lane for Moment in Interior Beams with
Corrugated Steel Plank Decks

Table 4.6.2.2.2d-1 Live Load Distribution Factor of Live Loads per Lane for Moment in Exterior Longitudinal
Beams

Table 4.6.2.2.2f-1 Live Load Distribution Factor of Live Load per Lane for Transverse Beams for Moment and
Shear

Table 4.6.2.2.3a-1 Live Load Distribution Factor of Live Load per Lane for Shear in Interior Beams

Table 4.6.2.2.3b-1 Live Load Distribution Factor of Live Load per Lane for Shear in Exterior Beams
Item #3

Revise the 1st paragraph of Article 4.6.2.2.2a as follows:

The live load flexural moment and shear for interior beams with transverse wood decks may be determined
by applying the lane fraction live load distribution factor, g, specified in Table 4.6.2.2.2a-1 and Eq. 4.6.2.2.2a-1.

Revise the 1st paragraph of Article 4.6.2.2.2b as follows:

The live load flexural moment for interior beams with concrete decks may be determined by applying the lane
fraction live load distribution factor, g, specified in Table 4.6.2.2.2b-1.

Revise Article 4.6.2.2.2c as follows:

The live load flexural moment for interior beams with corrugated steel plank deck may be determined by
applying the lane fraction live load distribution factor, g, specified in Table 4.6.2.2.2c-1.

Revise the 1st paragraph of Article 4.6.2.2.2d as follows:

The live load flexural moment for exterior beams may be determined by applying the lane fraction live load
distribution factor, g, specified in Table 4.6.2.2.2d-1.

Revise the 1st sentence in the 2nd paragraph of Article 4.6.2.2.2f as follows:

The fractions live load distribution factor, g, provided in Table 4.6.2.2.2f-1 shall be used in conjunction with
the 32.0-kip axle design load alone.

Revise the 1st sentence in the 1st paragraph of Article 4.6.2.2.3a as follows:

The live load shear for interior beams may be determined by applying the lane fractions live load distribution
factor, g, specified in Table 4.6.2.2.3a-1.

Revise the 1st sentence in the 1st paragraph of Article 4.6.2.2.3b as follows:

The live load shear for exterior beams shall may be determined by applying the lane fractions live load
distribution factor, g, specified in Table 4.6.2.2.3b-1.

Revise the 1st and 2nd sentences in the 1st paragraph of Article 4.6.2.2.3c as follows:

Shear in the exterior beam at the obtuse corner of the bridge shall be adjusted when the line of support is
skewed. The value of the correction factor shall be obtained from Table 4.6.2.2.3c-1. It is applied to the lane
fraction live load distribution factor, g, specified in Table 4.6.2.2.3a-1 for interior beams and in Table 4.6.2.2.3b-1
for exterior beams.

Item #4

In the first column, revise the 3rd row of Table 4.6.2.2.2b-1; 3rd row of 4.6.2.2.2d-1; 2nd column heading of Table
4.6.2.2.2f-1; 1st and 2nd row of Table 4.6.2.2.2e-1; 3rd row of 4.6.2.2.3a-1; 3rd row of 4.6.2.2.3b-1; 1st row of Table
4.6.2.2.3c-1 as follows:
Table 4.6.2.2.2b-1
Concrete Deck, or Filled Grid,
Partially Filled Grid, or Unfilled
Grid Deck Composite with
Reinforced Concrete Slab on
Steel or Concrete Beams;
Concrete T-Beams, T- and
Double T-Sections

Table 4.6.2.2.2d-1
Concrete Deck, or Filled Grid,
Partially Filled Grid, or Unfilled
Grid Deck Composite with
Reinforced Concrete Slab on
Steel or Concrete Beams;
Concrete T-Beams, T- and
Double T-Sections

Table 4.6.2.2.2f-1
Type of Deck Live Load Distribution Factors for Range of Applicability
Fraction of Wheel Load to
Each Floorbeam

Table 4.6.2.2.2e-1 Reduction of Live Load Distribution Factors for Moment in Longitudinal Beams on
Skewed Supports
Concrete Deck, or Filled Grid,
Partially Filled Grid, or Unfilled
Grid Deck Composite with
Reinforced Concrete Slab on
Steel or Concrete Beams;
Concrete T-Beams, T- and
Double T-Sections
Concrete Deck on Concrete
Spread Box Beams, Cast-in-
Place Multicell Box, Concrete
Box Beams and Double T-
Sections used in Multibeam
Decks

Table 4.6.2.2.3a-1
Concrete Deck, or Filled Grid,
Partially Filled Grid, or Unfilled
Grid Deck Composite with
Reinforced Concrete Slab on
Steel or Concrete Beams;
Concrete T-Beams, T- and
Double T-Sections

Table 4.6.2.2.3b-1
Concrete Deck, or Filled Grid,
Partially Filled Grid, or Unfilled
Grid Deck Composite with
Reinforced Concrete Slab on
 Steel or Concrete Beams;
Concrete T-Beams, T- and
Double T-Sections

Table 4.6.2.2.3c-1 Correction Factors for Live Load Distribution Factors for Support Shear of the Obtuse
Corner
Concrete Deck, or Filled Grid,
Partially Filled Grid, or Unfilled
Grid Deck Composite with
Reinforced Concrete Slab on
Steel or Concrete Beams;
Concrete T-Beams, T- and
Double T-Sections

OTHER AFFECTED ARTICLES:

None

BACKGROUND:
Item #1: Consistent use of terminology between articles and tables.

Item #2: In LFD, for the approximate method, live load distribution factors were applied to a wheel line. In LRFD
for the approximate method live load distribution factors are applied to the entire lane. Both are then applied to the
supporting members, not the design lane. Since there is one truck per lane the words per lane were used in the
girder distribution table titles in Article 4.6.2.2 to distinguish from number of wheel lines. However, as the title
implies, if load distribution factors are applied per lane one might think the live load distribution factor, g, is
supposed to be multiplied by the number of design lanes instead of being applied to each girder or for whole width
design multiplied by the number of girders. This could lead to a misunderstanding of how much live load is
applied.
Table 4.6.2.2.2f-1 is a slightly different case because the distribution of live load is a fraction of axle load
The articles describe the distribution factors in the tables as fractions of lanes applied to beams. The titles and
articles do not say it means a complete design vehicle instead of a wheel line by using per Lane.
The table titles should be brief as well as grammatically and technically correct.

Item #3: Consistent use of terminology between articles and tables.

Item #4: Editorial: add or after Concrete Deck to separate it from the group of grid type decks. Add a comma in
the second row of Table 4.6.2.2.2e-1 for clarity. Change column heading in Table 4.6.2.2.2f -1 for consistent use
of terminology and delete Wheel since axle loads are being distributed, per the article.

ANTICIPATED EFFECT ON BRIDGES:

None

REFERENCES:
AASHTO LRFD BDS, 5th Edition, with 2011Interims.
AASHTO Standard Specification for Highway Bridges, 17th Ed, 2004.
OTHER:
None
2012 AASHTO BRIDGE COMMITTEE AGENDA ITEM: 47

SUBJECT: AASHTO Guide Specifications for LRFD Seismic Bridge Design, Section 8,
Articles 8.4.1, 8.4.2 & 8.5 (WAI-1C)

TECHNICAL COMMITTEE: T-3 Seismic

REVISION ADDITION NEW DOCUMENT

DESIGN SPEC CONSTRUCTION SPEC MOVABLE SPEC

MANUAL FOR BRIDGE SEISMIC GUIDE SPEC COASTAL GUIDE SPEC
EVALUATION OTHER

DATE PREPARED: 2/15/12

DATE REVISED: 5/22/12

AGENDA ITEM:
Item #1

Replace Article 8.4.1 with the following:

Reinforcing bars, deformed wire, cold-drawn wire, welded plain wire fabric and welded deformed wire
fabric shall conform to the material standards as specified in AASHTO LRFD Bridge Design Specifications.
For SDCs B and C, ASTM A 706 Grade 60 reinforcing steel shall be used in members where plastic
hinging is expected. ASTM A 615 Grade 60 reinforcing steel may be used in members where plastic hinging is
expected, with the Owners approval.
For SDC D, ASTM A 706 Grade 60 reinforcing steel shall be used in members where plastic hinging is
expected.
ASTM A 706 Grade 80 reinforcing steel, may be used in capacity protected members as specified in
Article 8.5, but shall not be used in members where plastic hinging is expected.

Item #2

Add the following paragraph to the end of Article C8.4.1:

Only ASTM A 706 Grade 60 or ASTM A 615 Grade 60 reinforcing steel should be used in members where
plastic hinging is expected.
The 80.0 ksi limit on reinforcing steel used in capacity protected members is established to limit the crack
widths that may occur with the use of such reinforcement. There is little data on the effectiveness of the use of
higher strength materials for capacity protected elements. The use of ASTM A 706 Grade 80 reinforcing is
permitted for use in capacity protected members due to its strength control and elongation characteristics. However,
the use of ASTM A 615 Grade 75 and ASTM A 615 Grade 80 are not currently permitted due to a lack of stress-
strain data, a lack of seismic test results, and their lower ultimate tensile strain values.
The requirement for plastic hinging and capacity protection of adjoining members do not apply to the
structures in SDC A. Therefore the above restrictions on the type and grade of reinforcing steel do not apply in SDC
A.
Item #3

In Table 8.4.2-1, add the words Grade 60 to the column labeled ASTM A 706.

Item #4

In Article 8.5, replace the definition of mo in the where list with the following:

mo = overstrength multiplier factor

= 1.2 for ASTM A 706 Grade 60 reinforcement
= 1.4 for ASTM A 615 Grade 60 reinforcement

Item #5

Add the following paragraph after the 3rd paragraph of Article 8.5:

When using ASTM A 706 Grade 80 reinforcing in capacity protected members, the expected nominal moment
capacity, Mne, shall be based on the expected concrete strength, fce, and an expected yield strength, fye, of 86.0 ksi
when the concrete strain reaches a magnitude of 0.003 or when the reinforcing steel strain reaches 0.06 for #11 bars
and larger or 0.09 for #10 bars or smaller.

OTHER AFFECTED ARTICLES:

None

BACKGROUND:
The AASHTO LRFD Bridge Design Specifications in 2012 T-10 Ballot Item is permitting the use of high strength
reinforcing bars with minimum yield strength of 100 ksi in non-seismic regions. The yield strength of ASTM
A706 Grade 80 reinforcing steel (80 ksi) is within the AASHTO specified 100 ksi and could be safely used for
structures in low seismic zone and with some limitation in moderate to high seismic zones.

ANTICIPATED EFFECT ON BRIDGES:

Reduction in cost and improved constructability.

ASTM A 706 reinforcing steel is available in two grades, 60 ksi and 80 ksi yield strengths. ASTM A 706 Grade 60
is used in members expected to develop plastic hinges due to its superior mechanical properties. ASTM A706
Grade 80 reinforcing steel with yield strength of 80 ksi is commercially available in the United States and the use of
this higher strength steel could provide benefits to concrete bridge construction by reducing member cross sections
and reinforcement quantities, leading to savings in material, shipping, and placement costs. Reducing reinforcement
quantities will also reduce congestion problems leading to better quality of construction. ASTM A 706 Grade 80
reinforcing steel has not been fully evaluated for use in members expected to form plastic hinges. Consequently, the
use of ASTM A 706 Grade 80 reinforcing is not recommended for members expected to form plastic hinges. Only a
limited record of actual vs. nominal bar yield strengths is available; thus the expected yield strength is specified as
86 ksi. The limited test data available show that the expected yield may be higher, but it is not prudent to use
higher values until a more complete record of bar strengths is available.
REFERENCES:
1. Shahrooz, B. M., R. A. Miller, K. A. Harries, and H. G. Russell. 2011. Design of Concrete Structures Using
High-Strength Steel Reinforcement, NCHRP Report 679. Transportation Research Board, National Research
Council, Washington, DC.
2. CalTrans Seismic Design Criteria Article 3.4 Requirement, for Capacity protected Components, Version 1.6,
2010.
3. ASTM A706/706M-09B ASTM A706 / A706M - 09b Standard Specification for Low-Alloy Steel Deformed
and Plain Bars for Concrete Reinforcement.

OTHER:
Lead State: AK, CA,WA
Industry:
FHWA:
ATTACHMENT A 2012 AGENDA ITEM 48 T-3

(PROVIDED ON CD)

APPENDIX B DESIGN EXAMPLES

2012 AASHTO BRIDGE COMMITTEE AGENDA ITEM: 48

SUBJECT: AASHTO Guide Specifications for Seismic Isolation Design: Appendix B (new)

TECHNICAL COMMITTEE: T-3 Seismic

REVISION ADDITION NEW DOCUMENT

DESIGN SPEC CONSTRUCTION SPEC MOVABLE SPEC

MANUAL FOR BRIDGE SEISMIC GUIDE SPEC COASTAL GUIDE SPEC
EVALUATION OTHER

DATE PREPARED: 2/13/12

DATE REVISED:

AGENDA ITEM:
See Attachment A (provided on CD) - Add new Appendix to Guide Specifications for Seismic Isolation Design:
Appendix B - Design Examples.

OTHER AFFECTED ARTICLES:

Table of Contents for Guide Specifications should be updated to reflect addition of new Appendix. If this new
appendix triggers a new edition of the Specifications (i.e. 4 th Edition), a Preface to this new edition may be
necessary.

BACKGROUND:
This revision adds a new appendix to the Guide Specifications for Seismic isolation Design, which contains the
fourteen design examples recently completed under NCHRP project 20-7/262. These examples illustrate the design
process for an isolated bridge and the design of related hardware in accordance with the Guide Specifications.

In particular, the fourteen examples demonstrate the application of isolation to a range of bridges for varying
seismic hazard, site classification, isolator type, and bridge type. In general, each example illustrates the suitability
of the bridge for isolation (or otherwise), and presents calculations for preliminary design using the Simplified
Method (Art 7.1, AASHTO, 2010), preliminary and final isolator design, and detailed analysis using the Multi-
Modal Spectral Analysis procedure (Art 7.3, AASHTO 2010). Detailed designs of the superstructure, substructure
(piers) and foundations are not included.

Text books on this topic are rare and these examples are believed to be the only set of their kind. They are therefore
considered to be a valuable resource for engineers wishing to design a seismically isolated bridge and publishing
them as an Appendix to the Guide Specifications will ensure both access and a wide distribution.

ANTICIPATED EFFECT ON BRIDGES:

An increase in the number of seismically isolated bridges in the US is expected as designers become familiar with
the benefits of the technique, the design process, and the sizing of isolation hardware.
REFERENCES:
AASHTO, 2010. Guide Specifications for Seismic Isolation Design, Third Edition, American Association State
Highway and Transportation Officials, Washington DC.

OTHER:
None
ATTACHMENT A 2012 AGENDA ITEM T-3

APPENDIX B DESIGN EXAMPLES

SUMMARY

The Examples in this Appendix illustrate the design process for a seismically isolated bridge in
accordance with the requirements of these Guide Specifications.

Fourteen examples are presented to demonstrate the application of isolation to a range of bridges for
varying seismic hazard, site classification, isolator type, and bridge type. In general, each example
illustrates the suitability of the bridge for isolation (or otherwise), and presents calculations for
preliminary design using the Simplified Method (Art 7.1), preliminary and final isolator design, and
detailed analysis using the Multi-Modal Spectral Analysis procedure (Art 7.3). Design of the
superstructure, substructure and foundation is not included.

Text books on this topic are rare and these examples are believed to be the only set of their kind. They are
included in this Appendix as a resource for engineers wishing to design a seismically isolated bridge.

ACKNOWLEDGEMENTS

These Design Examples were prepared under NCHRP Project 20-7/262 by the following authors:
Ian Buckle, Professor University of Nevada Reno
Moustafa Al-Ani, Visiting Researcher, University of Nevada Reno
Eric Monzon, Graduate Assistant, University of Nevada Reno

The authors are grateful for the oversight provided by the NCHRP Panel for this project.
Members of this Panel were:
Ralph E. Anderson, PE, SE, Engineer of Bridges and Structures, Illinois DOT
Barry Bowers, PE, Structural Design Support Engineer, South Carolina DOT
Derrell Manceaux, PE, Structural Design Engineer, FHWA Resource Center CO
Gregory Perfetti, PE, State Bridge Design Engineer, North Carolina DOT
Richard Pratt, PE, Chief Bridge Engineer, Alaska DOT
Hormoz Seradj, PE, Steel Bridge Standards Engineer, Oregon DOT
Kevin Thompson, PE, Deputy Division Chief, California DOT
Edward P Wasserman, PE, Civil Engineering Director - Structures Division, Tennessee DOT

The authors are also grateful for the assistance of the Project Working Group. Members of this Group
were:
Tim Huff, Tennessee DOT
Allaoua Kartoum, California DOT
Elmer Marx, Alaska DOT
Albert Nako, Oregon DOT
David Snoke, North Carolina DOT
Daniel Tobias, Illinois DOT
 CONTENTS (move to page xiv of the Guide Specifications)

INTRODUCTION 1
1.1 Background 1
1.2 Design Examples 1
1.3 Design Methodology 5
1.4 Presentation of Design Examples 5
1.5 Summary of Results 5
1.6 References 8

SECTION 1. PC GIRDER BRIDGE EXAMPLES 9

Example 1.0 (Benchmark #1) 9

A. Bridge and Site Data 10
B. Analyze Bridge in Longitudinal Direction 12
C. Analyze Bridge in Transverse Direction 23
D. Calculate Design Values 25
E. Design of Lead Rubber Isolators 26

Example 1.1 (Site Class D) 36

A. Bridge and Site Data 37
B. Analyze Bridge in Longitudinal Direction 39
C. Analyze Bridge in Transverse Direction 50
D. Calculate Design Values 52
E. Design of Lead Rubber Isolators 53

Example 1.2 (S1 = 0.6g) 63

A. Bridge and Site Data 64
B. Analyze Bridge in Longitudinal Direction 66
C. Analyze Bridge in Transverse Direction 77
D. Calculate Design Values 79
E. Design of Lead Rubber Isolators 80

Example 1.3 (Spherical Friction Isolators) 90

A. Bridge and Site Data 91
B. Analyze Bridge in Longitudinal Direction 93
C. Analyze Bridge in Transverse Direction 104
D. Calculate Design Values 106
E. Design of Spherical Friction Isolators 107

Example 1.4 (Eradiquake Isolators) 113

A. Bridge and Site Data 114
B. Analyze Bridge in Longitudinal Direction 116
C. Analyze Bridge in Transverse Direction 127
D. Calculate Design Values 129
E. Design of Eradiquake Isolators 130
Example 1.5 (Unequal Pier Heights) 140
A. Bridge and Site Data 141
B. Analyze Bridge in Longitudinal Direction 143
C. Analyze Bridge in Transverse Direction 154
D. Calculate Design Values 156
E. Design of Lead Rubber Isolators 157

Example 1.6 (Skew = 450) 167

A. Bridge and Site Data 168
B. Analyze Bridge in Longitudinal Direction 170
C. Analyze Bridge in Transverse Direction 182
D. Calculate Design Values 184
E. Design of Lead Rubber Isolators 185

SECTION 2. STEEL PLATE GIRDER BRIDGE EXAMPLES 195

Example 2.0 (Benchmark #2) 195

A. Bridge and Site Data 196
B. Analyze Bridge in Longitudinal Direction 198
C. Analyze Bridge in Transverse Direction 209
D. Calculate Design Values 210
E. Design of Lead Rubber Isolators 211

Example 2.1 (Site Class D) 221

A. Bridge and Site Data 222
B. Analyze Bridge in Longitudinal Direction 224
C. Analyze Bridge in Transverse Direction 237
D. Calculate Design Values 238
E. Design of Lead Rubber Isolators 239

Example 2.2 (S1 = 0.6g) 250

A. Bridge and Site Data 251
B. Analyze Bridge in Longitudinal Direction 253
C. Analyze Bridge in Transverse Direction 266
D. Calculate Design Values 268
E. Design of Lead Rubber Isolators 269

Example 2.3 (Spherical Friction Isolators) 279

A. Bridge and Site Data 280
B. Analyze Bridge in Longitudinal Direction 282
C. Analyze Bridge in Transverse Direction 293
D. Calculate Design Values 294
E. Design of Spherical Friction Isolators 295

Example 2.4 (Eradiquake Isolators) 301

A. Bridge and Site Data 302
B. Analyze Bridge in Longitudinal Direction 304
C. Analyze Bridge in Transverse Direction 315
D. Calculate Design Values 316
E. Design of Eradiquake Isolators 317
Example 2.5 (Unequal Pier Heights) 327
A. Bridge and Site Data 328
B. Analyze Bridge in Longitudinal Direction 330
C. Analyze Bridge in Transverse Direction 342
D. Calculate Design Values 344
E. Design of Lead Rubber Isolators 345

Example 2.6 (Skew = 450) 352

A. Bridge and Site Data 353
B. Analyze Bridge in Longitudinal Direction 355
C. Analyze Bridge in Transverse Direction 367
D. Calculate Design Values 369
E. Design of Lead Rubber Isolators 370

 INTRODUCTION

1.1 BACKGROUND

Today about 200 bridges have been designed and constructed in the U.S. using the AASHTO Guide
Specifications for Seismic Isolation Design but this figure is a fraction of the potential number of
applications and falls far short of the number of isolated bridges in other countries (Buckle et. al., 2006).

One of the major barriers to implementation is that isolation is a significant departure from conventional
seismic design and one that is not routinely taught in university degree courses. Furthermore, very few
text books on this topic have been published and those that are available focus on applications to
buildings rather than bridges. The absence of formal instruction and lack of reference material, means
that many designers are not familiar with the approach and uncomfortable using the technique, despite the
potential for significant benefits.

In an effort to address this need, an NCHRP 20-7 project was funded in 2010 to develop and publish a
number of examples to illustrate the design process and design of related hardware in accordance with the
Guide Specifications.

Fourteen examples were developed under this Project illustrating the application of isolation to a range of
bridges for varying seismic hazard, site classification, isolator type, and bridge type. In general, each
example illustrates the suitability of the bridge for isolation (or otherwise), and presents calculations for
preliminary design using the Simplified Method (Art 7.1), preliminary and final isolator design, and
detailed analysis using a Multi-Modal Spectral Analysis procedure (Art 7.3). Detailed design of the
superstructure, substructure (piers) and foundations is not included. These examples are presented in this
Appendix.

1.2 DESIGN EXAMPLES

The fourteen examples are summarized in Table 1. It will be seen they fall into two sets: one based on a
PC-girder bridge with short spans and multiple column piers (Benchmark Bridge #1) and the other on a
steel plate-girder bridge with long spans and single column piers (Benchmark Bridge #2). For each bridge
there are six variations as shown in Table 1. Both benchmark bridges have the following attributes:
Seismic Hazard: Spectral Acceleration at 1.0 sec (S1) = 0.2g
Site class: B (rock)
Pier heights: Uniform
Skew: None
Isolator: Lead-Rubber Bearings (LRB)

These five attributes are varied (one at a time) to give 12 additional examples as shown in Table 1.
Variations covered include S1 = 0.6g, Site Class D, unequal pier heights, 450 skew, Spherical Friction
Bearing (SFB) and Eradiquake (EQS) isolators.

Brief descriptions of the two benchmark bridges are given in the following sections.

 Table 1. Seismic Isolation Design Examples.

Site Column size

Example S1 Spans Girders Skew Isolator
class and heights

EXAMPLE SET 1: PC Girder Bridge, short spans, multi-column concrete piers

1.0
0.2g 3 6 PC girders
Benchmark B 2 x 3-col piers 00 LRB
Zone 2 25-50-25 ft (AASHTO Type II)
Bridge #1
3 6 PC girders
1.1 Zone 3 D 2 x 3-col piers 00 LRB
25-50-25 ft (AASHTO Type II)

0.6g 3 6 PC girders 0
1.2 B 2 x 3-col piers 0 LRB
Zone 4 25-50-25 ft (AASHTO Type II)

0.2g 3 6 PC girders 0
1.3 B 2 x 3-col piers 0 SFB
Zone 2 25-50-25 ft (AASHTO Type II)

0.2g 3 6 PC girders
1.4 B 2 x 3-col piers 00 EQS
Zone 2 25-50-25 ft (AASHTO Type II)

0.2g 3 6 PC girders 2 x 3-col piers 0

1.5 B 0 LRB
Zone 2 25-50-25 ft (AASHTO Type II) unequal height

0.2g 3 6 PC girders 0
1.6 B 2 x 3-col piers 45 LRB
Zone 2 25-50-25 ft (AASHTO Type II)

EXAMPLE SET 2: Steel Plate Girder Bridge, long spans, single-column concrete piers

2.0
0.2g 3 3 steel plate 2 x single-col 0
Benchmark B 0 LRB
Zone 2 105-152.5-105 ft girders with slab piers.
Bridge #2
3 3 steel plate 2 x single-col 0
2.1 Zone 3 D 0 LRB
105-152.5-105 ft girders with slab piers

0.6g 3 3 steel plate 2 x single-col

2.2 B piers 00 LRB
Zone 4 105-152.5-105 ft girders with slab

0.2g 3 3 steel plate 2 x single-col

2.3 B piers 00 SFB
Zone 2 105-152.5-105 ft girders with slab

0.2g 3 3 steel plate 2 x single-col 0

2.4 B 0 EQS
Zone 2 105-152.5-105 ft girders with slab piers
2 x single-col
0.2g 3 3 steel plate 0
2.5 B piers with 0 LRB
Zone 2 105-152.5-105 ft girders with slab
unequal height
0.2g 3 3 steel plate 2 x single-col 0
2.6 B 45 LRB
Zone 2 105-152.5-105 ft girders with slab piers

1.2.1 Benchmark Bridge No. 1

Benchmark Bridge No. 1 is a straight,

3-span, slab-and-girder structure with
three columns at each pier and seat-
type abutments. The spans are
continuous over the piers with span
lengths of 25 ft, 50 ft, and 25 ft for a
total length of 100 ft (Figure 1.1). The
superstructure comprises six
AASHTO Type II girders spaced at
7.17 ft with 3.1 ft overhangs for a
total width of 42.5 ft. The total weight
of the superstructure is 651 k.

The two piers each consist of three

circular columns spaced at 14 ft.,
longitudinal steel ratio of 1%, and a
transverse steel ratio of 1%. The
plastic shear capacity of each column
(in single curvature) is 25 k. The
height of the superstructure is
approximately 20 ft above ground.

The bridge is located on a rock site

where the PGA = 0.4g, SS = 0.75g and
S1 = 0.20g.

Figure 1.1. Plan, Side View, and Pier Elevation for

3-Span Benchmark Bridge No. 1.

1.2.2 Benchmark Bridge No. 2

Benchmark bridge No. 2 is a straight, 3-span, steel plate-girder structure with single column piers and
seat-type abutments. The spans are continuous over the piers with span lengths of 105 ft, 152.5 ft, and
105 ft for a total length of 362.5 ft (Figure 1.2). The girders are spaced 11.25 ft apart with 3.75 ft
overhangs for a total width of 30 ft. The built-up girders are composed of 1.625 in by 22.5 in top and
bottom flange plates and 0.9375 in. by 65 in. web plate. The reinforced concrete deck slab is 8.125 in
thick with 1.875 in. haunch. The support and intermediate cross-frames are of V-type configuration as
shown in Figure 1.3. Cross-frame spacing is about 15 ft throughout the bridge length. The total weight of
superstructure is 1,651 kips.

All the piers are single concrete columns with a longitudinal steel ratio of 1%, and transverse steel ratio of
1%. The plastic shear capacity (in single curvature) is 128k. The height of the superstructure is
approximately 24 ft above the ground.

The bridge is located on a rock site where the PGA = 0.4g, SS = 0.75g and S1 = 0.20g.

 1050 1526 1050

300

Figure 1.2. Plan of 3-Span Benchmark Bridge No. 2.

Deckslab

Plategirders

Crossframe

Isolators

240
Singlecolumnpierand
hammerheadcapbeam

Figure 1.3. Typical Section of Superstructure and Elevation at Pier of Benchmark

Bridge No. 2.

1.3 DESIGN METHODOLOGY

All of the isolation systems used in the above design examples have nonlinear properties in order to be
near-rigid for non-seismic loads but soften for earthquake loads. To avoid having to use nonlinear
methods of analysis, equivalent linear springs and viscous damping is assumed to represent the nonlinear
properties of isolators. But since these equivalent properties are dependent on displacement, an iterative
approach is required to obtain a solution. This approach is sometimes known as the Direct Displacement
Method.

Each of the 14 examples has been designed using the same assumptions and design methodology. This
methodology has five basic steps as below:

Step A. Determine bridge and site data including required performance criteria
Step B. Analyze bridge for earthquake in longitudinal direction using the Simplified Method to
obtain initial estimates for use in Multi-modal Spectral Analysis

Step C. Analyze bridge for earthquake in transverse direction using the Simplified Method to
obtain initial estimates for use in Multi-modal Spectral Analysis

Step D. Combine results from Steps B and C and obtain design values for displacements and
forces

Step E. Design isolators

Further detail for each step is given in Table 2.

1.4 PRESENTATION OF DESIGN EXAMPLES

The same 2-column format is used for each design example. A step-by-step design procedure, based on
the methodology in the previous section, is given in the left-hand column and the application of this
procedure to the example in hand is given in the right- hand column. The left hand column is therefore the
same for all examples. The right hand column changes from example to example. Each example is
presented as a stand-alone exercise to improve readability. However, in these circumstances, repetition of
some material is unavoidable.

References to provisions in the AASHTO Specifications are made throughout the examples using the
following notation:
GSID refers to Guide Specifications Seismic Isolation Design, AASHTO 2010
LRFD refers to LRFD Bridge Design Specifications, AASHTO 2008

1.5 SUMMARY OF RESULTS

Table 3 summarizes the results of the 14 designs. The basic dimensions of each isolator required to
achieve (or almost achieve) the desired performance are given in this Table. Column shear forces and
superstructure displacements are also given for each bridge.

 Table 2. Methodology and Steps in Design of Seismically Isolated Bridge.

METHODOLOGY
Assume equivalent linear springs and viscous damping can be used to represent the nonlinear hysteretic properties of
isolators, so that linear methods of analysis methods may be used to determine response. Since equivalent properties
are dependent on displacement, an iterative approach is required to obtain a solution. The methodology below uses
the Simplified Method to obtain initial estimates of displacement for use in an iterative solution using the Multi-Modal
Spectral Analysis Method.

STEP A. BRIDGE AND SITE DATA

A1. Obtain bridge properties: weight, geometry, substructure stiffnesses and capacities, isolators, soil conditions
A2. Determine seismic hazard at site (acceleration coefficients and soil factors); plot response spectrum
A3. Determine required performance of isolated bridge (e.g. elastic columns for design earthquake)

STEP B. ANALYZE BRIDGE IN LONGITUDINAL DIRECTION.

B1. Apply response spectrum in longitudinal direction of bridge and use Simplified Method to analyze a single
degree-of-freedom model of bridge to obtain first estimate of superstructure displacement and required properties
of each isolator necessary to obtain desired performance (i.e. find d, characteristic strength, Qdj, and post elastic
stiffness, Kdj for each isolator j)

B2. Apply response spectrum in longitudinal direction of bridge and use Multimodal Spectral Analysis Method to
analyze 3-dimensional, multi-degree-of-freedom model of bridge and obtain final estimates of superstructure
displacement and required properties of each isolator to obtain desired performance. [Use the results from the
Simplified Method to determine equivalent spring elements to represent the isolators in the 3-D model used in
this analysis.]
|
Obtain longitudinal and transverse displacements (uL, vL) for each isolator
Obtain longitudinal and transverse displacements for superstructure
Obtain biaxial column moments and shears at critical locations

STEP C. ANALYZE BRIDGE IN TRANSVERSE DIRECTION

Repeat B1 and B2 above for response spectrum applied in transverse direction.
Obtain longitudinal and transverse displacements (uT, vT) for each isolator
Obtain longitudinal and transverse displacements for superstructure
Obtain biaxial column moments and shears at critical locations

STEP D. COMBINE RESULTS AND OBTAIN DESIGN VALUES

Combine results from longitudinal and transverse analyses using the (100L+30T) & (30L+100T) rule to obtain
design values for isolator and superstructure displacement, moment and shear.
Check that required performance is satisfied.

STEP E. DESIGN ISOLATORS

Select isolator type (e.g. lead-rubber isolator, spherical friction isolator, Eradiquake isolator)
Design isolators to have the required characteristic strengths (Qdj) and post elastic stiffnesses (Kdj) calculated
above. If actual values of Qdj and Kdj differ significantly from above values, reanalyze bridge. Check design for
strain limit state, and vertical stability, Conduct upper and lower analyses using minimum and maximum
properties to account for isolator aging, temperature effects, scragging, contamination and wear. Revise design if
Table required
3 performance objective is not satisfied.

 Table 3. Summary of Seismic Isolator Designs.

Super-
Isolator size Isolator size Diam. Rubber
structure
including without lead Shear Column
Ex. ID resultant
mounting mounting core modulus shear (k)
displace-
plates (in) plates (in) (in) (psi)
ment (in)

EXAMPLE SET 1: PC GIRDER BRIDGE (Column yield shear force = 25.0 k)

Benchmark 17.00 x 17.00 13.00 dia. x

1.0 1 x 11.50 (H) 10.00(H)
1.61 60 18.03 1.72

Site 17.25 x 17.25 13.25 dia. x

1.1 Class D x 11.875(H) 10.375(H)
1.97 60 25.55* 3.96

20.25 x 20.25 16.25 dia. x

1.2 S1=0.6g
x 16.75(H) 15.25(H)
1.97 60 29.15* 7.32

16.25 x 16.25 12.25 dia. R (in) PTFE

1.3 SFB isolator
x 4.50(H) x 4.50(H)
18.03 1.72
39.0 15GF

32.0 x 18.0 18.0 x 18.0 Polyurethane springs

1.4 EQS isolator
x 4.00(H) x 4.00(H)
18.03 1.72
4 1.25 dia.

17.00 x 17.00 13.00 dia. 19.56 (P1)

1.5 H1=0.5H2
x 11.50(H) x 10.00(H)
1.61 60
2.56 (P2)
2.32

16.00 x 16.00 12.00 dia.

1.6 450 skew 1.63 60 28.32* 1.61
x 10.00(H) x 8.50(H)

EXAMPLE SET 2: STEEL PLATE GIRDER BRIDGE (Column yield shear force) = 128 k)

Benchmark 17.50 x 17.50 13.50 dia.

2.0 2 x 5.50(H) x 4.00(H)
3.49 100 71.74 1.82

21.25 x 21.25 17.25 dia.

2.1 Class D
x 8.125(H) x 6.625(H)
4.13 60 121.0 3.79

24.0 x 24.0 20.0 dia.

2.2 S1=0.6g
x 12.625(H) x 11.125(H)
4.68 60 175.0* 8.21

17.75 x 17.75 13.75 dia. R (in) PTFE

2.3 SFB isolator
x 9.00(H) x 7.00(H)
71.74 1.82
27.75 25GF

36.0 x 23.0 23.0 x 23.0 Polyurethane springs

2.4 EQS isolator
6.20(H) x 6.20(H)
71.74 1.82
4 2.75 dia.

17.50 x 17.50 13.50 dia. 87.56 (P1)

2.5 H1=0.5H2
x 5.875(H) x 4.375(H)
3.49 100
47.53 (P2)
2.05

17.50 x 17.50 13.50 dia.

2.6 450 skew 3.49 100 106.8 1.69
x 5.50(H) x 4.00(H)

Note: * exceeds column yield shear force.

For the PC Girder Bridge, the elastic performance criterion is satisfied in 4 of the 7 cases. But in three
cases (soft soils, higher hazard, and extreme skew), it is not possible for the LRB system to keep the
column shear forces below yield. However the excess is small (less than 16%) and essentially elastic
behavior is to be expected. It is noted that these three cases use LRB isolators which, for this bridge, are
governed by vertical stability requirements. It is possible that the SFB and EQS systems might be able to
achieve fully elastic behavior in these cases, since they are not as sensitive to stability requirements.

For the Steel Plate Girder Bridge, the elastic criterion is satisfied in 6 of the 7 cases. The exception is the
case where S1 =0.6g (Example 2.2)and it is clear that for this level of seismicity either some level of yield
must be accepted, or the column increased in size to increase its elastic strength. As noted in Example 2.2,
a pushover analysis of this column will quickly determine the ductility demand during this earthquake and
a judgment can them be made whether it is acceptably small. It is noted that the value of this demand will
be significantly less than if isolation had not been used in the design.

For the other six cases it is shown that isolation designs may be found (using LRB, SFB and EQS
systems), for softer soils (Site Class D) and asymmetric geometry (unequal column heights and high
skew), and still keep the columns elastic. This improved performance compared to the PC Girder Bridge
is due to the Steel Plate Girder Bridge being heavier with fewer isolators (12 vs 24), a fact that favors
most isolation systems.

It is interesting to note in Table 3 that, although both bridges are significantly different in weight and
length, they have similar displacements for the same hazard, soil conditions and geometry. This is
because, when these bridges are isolated, they have similar fundamental periods and therefore respond to
the same hazard in similar ways.

1.6 REFERENCES

AASHTO, 2008. LRFD Bridge Design Specifications, American Association State Highway and
Transportation Officials, Washington DC

AASHTO, 2010. Guide Specifications for Seismic Isolation Design, Third Edition, American Association
State Highway and Transportation Officials, Washington DC

Buckle, I.G., Constantinou, M., Dicleli, M., Ghasemi, H., 2006. Seismic Isolation of Highway Bridges,
Special Report MCEER-06-SP07, Multidisciplinary Center Earthquake Engineering Research, University
at Buffalo, NY

 SECTION 1
PC Girder Bridge, Short Spans, Multi-Column Piers

DESIGN EXAMPLE 1.0: Benchmark Bridge #1

Design Examples in Section 1

Site
ID Description S1 Pier height Skew Isolator type
Class
1.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
1.1 Change site class 0.2g D Same 0 Lead rubber bearing
Change spectral
1.2 0.6g B Same 0 Lead rubber bearing
acceleration, S1
Spherical friction
1.3 Change isolator to SFB 0.2g B Same 0
bearing
1.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
1.5 Change column height 0.2g B H1=0.5 H2 0 Lead rubber bearing
1.6 Change angle of skew 0.2g B Same 450 Lead rubber bearing

DESIGN PROCEDURE
STEP A: BRIDGE AND SITE DATA
A1. Bridge Properties
Determine properties of the bridge:
number of supports, m
number of girders per support, n
angle of skew
weight of superstructure including railings,
curbs, barriers and to the permanent loads,
WSS
weight of piers participating with
superstructure in dynamic response, WPP
weight of superstructure, Wj, at each
support
stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge. The calculation of these
quantities requires careful consideration of
several factors such as the use of cracked
sections when estimating column or wall
flexural stiffness, foundation flexibility,
and effective column height.
column shear strength (minimum value).
This will usually be derived from the
minimum value of the column flexural
yield strength, the column height, and
whether the column is acting in single or
double curvature in the direction under
consideration.
allowable movement at expansion joints
isolator type if known, otherwise to be
determined.
A2. Seismic Hazard
Determine seismic hazard at site:
acceleration coefficients
site class and site factors
seismic zone
Plot response spectrum.
Use Art. 3.1 GSID to obtain peak ground and spectral
acceleration coefficients. These coefficients are the
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (e.g. Figures
3.10.2.1-1 to 3.10.2.1-21 LRFD).
Use Art. 3.2 to obtain Site Class and corresponding Site
Factors (Fpga, Fa and Fv). These data are the same as for
DESIGN EXAMPLE 1.0 (Benchmark #1)
A1. Bridge Properties, Example 1.0
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 6
Number of columns per support = 3
Angle of skew = 00
Weight of superstructure including permanent
loads, WSS = 650.52 k
Weight of superstructure at each support:
o W1 = 44.95 k
o W2 = 280.31 k
o W3 = 280.31 k
o W4 = 44.95 k
Participating weight of piers, WPP = 107.16 k
Effective weight (for calculation of period),
Weff = Wss + WPP = 757.68 k
Stiffness of each pier in the longitudinal
direction:
o Ksub,pier1,long = 172.0 k/in
o Ksub,pier2,long = 172.0 k/in
Stiffness of each pier in the transverse direction:
o Ksub,pier1,trans = 687.0 k/in
o Ksub,pier2,trans = 687.0 k/in
Minimum column shear strength based on
flexural yield capacity of column = 25 k
Displacement capacity of expansion joints
(longitudinal) = 2.0 in for thermal and other
movements.
Lead rubber isolators
A2. Seismic Hazard, Example 1.0
Acceleration coefficients for bridge site are given in
design statement as follows:
PGA = 0.40
S1 = 0.20
SS = 0.75
Bridge is on a rock site with shear wave velocity in
upper 100 ft of soil = 3,000 ft/sec.
Table 3.10.3.1-1 LRFD gives Site Class as B.
Tables 3.10.3.2-1, -2 and -3 LRFD give following
Site Factors:
Fpga = 1.0
Fa = 1.0
Fv = 1.0
10

conventional bridges and Art 3.2 refers the designer to

the corresponding articles in the LRFD Specifications,
i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3,
LRFD.

Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give

modified spectral acceleration coefficients that include
site effects as follows: As = Fpga PGA = 1.0(0.40) = 0.40
As = Fpga PGA SDS = Fa SS = 1.0(0.75) = 0.75
SDS = Fa SS SD1 = Fv S1 = 1.0(0.20) = 0.20
SD1 = Fv S1

Seismic Zone is determined by value of SD1 in Since 0.15 < SD1 < 0.30, bridge is located in Seismic
accordance with provisions in Table5-1 GSID. Zone 2.

These coefficients are used to plot design response Design Response Spectrum is as below:
spectrum as shown in Fig. 4-1 GSID.
0.9

0.8

0.7

0.6

Acceleration
0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Period(s)

A3. Performance Requirements A3. Performance Requirements, Example 1.0

Determine required performance of isolated bridge In this example, assume the owner has specified full
during Design Earthquake (1000-yr return period). functionality following the earthquake and therefore
the columns must remain elastic (no yield).
Examples of performance that might be specified by
the Owner include: To remain elastic the maximum lateral load on the
Reduced displacement ductility demand in pier must be less than the load to yield any one
columns, so that bridge is open for emergency column (25 k).
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in The maximum shear in the column must therefore be
columns or yield elsewhere in bridge), so that less than 25 k in order to keep the column elastic and
bridge is fully functional and open to all vehicles meet the required performance criterion.
immediately following earthquake.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.

11

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for Isolator Force, F

non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most have
bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use Ku
equivalent linear springs and viscous damping to
represent the isolators, so that linear methods of analysis
disol
may be used to determine response. Since equivalent Ku
dy Isolator
Ku Displacement, d
properties such as Kisol are dependent on displacement (d),
and the displacements are not known at the beginning of
the analysis, an iterative approach is required. Note that in Kd
Art 7.1, GSID, keff is used for the effective stiffness of an
isolator unit and Keff is used for the effective stiffness of a disol = Isolator displacement
dy = Isolator yield displacement
combined isolator and substructure unit. To minimize Fisol = Isolator shear force
confusion, Kisol is used in this document in place of keff. Fy = Isolator yield force
Kd = Post-elastic stiffness of isolator
There is no change in the use of Keff and Keff,j, but Ksub is Kisol = Effective stiffness of isolator
used in place of ksub. Ku
Qd
= Loading and unloading stiffness (elastic stiffness)
= Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement (disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 1.0
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.20 2.0
GSID

(2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not
occur under non-seismic loads (e.g. wind) but

12

 low enough that yield will occur during an

earthquake. Experience has shown that
taking Qd to be 5% of the bridge weight is a
good starting point, i.e. 0.05 0.05 650.52 32.53
0.05 (B-2)
(3) Post-yield stiffness, Kd
Art 12.2 GSID requires that all isolators
exhibit a minimum lateral restoring force at
the design displacement, which translates to a
minimum post yield stiffness Kd,min given by
Eq. B-3.

Art. 650.52
0.025 0.05 0.05 16.26 /
12.2 , (B-3) 2.0
GSID
Experience has shown that a good starting
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 1.0
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength, ,

Qd, and stiffness, Kd, values in proportion to the dead o Qd, 1 = 2.25 k
load applied at that support: o Qd, 2 = 14.02 k
o Qd, 3 = 14.02 k
o Qd, 4 = 2.25 k

, (B-4)
and
and
,
, (B-5) o Kd,1 = 1.12 k/in
o Kd,2 = 7.01 k/in
o Kd,3 = 7.01 k/in
o Kd,4 = 1.12 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 1.0
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of , ,
the isolators at support j (Kisol,j) and the stiffness of , ,
the substructure Ksub,j. See figure below for
definitions (after Fig. 7.1-1 GSID). o 1 = 2.25x10-4
o 2 = 8.49x10-2
An expression for Keff,j, is given in Eq.7.1-6 GSID, but o 3 = 8.49x10-2
a more useful formula as follows (MCEER 2006): o 4 = 2.25x10-4
,
, (B-6)
1 ,
where ,
1
, ,
(B-7)
, , o Keff,1 = 2.25 k/in
and Ksub,j for the piers are given in Step A1. For the o Keff,2 = 13.47 k/in
abutments, take Ksub,j to be a large number, say 10,000 o Keff,3 = 13.47 k/in
k/in, unless actual stiffness values are available. Note o Keff,4 = 2.25 k/in
that if the default option is chosen, unrealistically high

13

 values for Ksub,j will give unconservative results for

column moments and shear forces.

F
Kd
Qd Kisol

dy disol
Superstructure
Isolator Effective Stiffness, Kisol

F
Isolator(s), Kisol
Ksub

Substructure, Ksub
dsub

Substructure Stiffness, Ksub

dsub disol F

d
Keff

d = disol + dsub

Combined Effective Stiffness, Keff

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 1.0
Calculate the total effective stiffness, Keff, of the
bridge:

Eq. , 31.43 /
7.1-6 , ( B-8)
GSID

B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 1.0
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1

(B-9) o disol,1 = 2.00 in

,
1 o disol,2 = 1.84 in
o disol,3 = 1.84 in
o disol,4 = 2.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the effective stiffness of the isolation system Example 1.0
at support j, Kisol,j, for all supports:
,
, ,
,
,
(B-10) o Kisol,1 = 2.25 k/in
, ,
, o Kisol,2 = 14.61 k/in
o Kisol,3 = 14.61 k/in
o Kisol,4 = 2.25 k/in

14

B1.7 Substructure Displacement at Each Support
B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j,
Example 1.0
for all supports:
, ,

, , (B-11)
o dsub,1 = 4.49x10-4 in
o d sub,2 = 1.57x10-1 in
o d sub,3 = 1.57x10-1 in
o d sub,4 = 4.49x10-4 in
B1.8 Lateral Load in Each Substructure Support B1.8 Lateral Load in Each Substructure, Example
Calculate the shear at support j, Fsub,j, for all 1.0
supports:
, , , (B-12) , , ,

where values for Ksub,j are given in Step A1. o F sub,1 = 4.49 k
o F sub,2 = 26.93 k
o F sub,3 = 26.93 k
o F sub,4 = 4.49 k

B1.9 Column Shear Force at Each Support B1.9 Column Shear Force at Each Support,
Calculate the shear in column k at support j, Fcol,j,k, Example 1.0
assuming equal distribution of shear for all columns at
support j:
,
, ,
#
,
, ,
# (B-13) o F col,2,1-3 8.98 k
o F col,3,1-3 8.98 k

Use these approximate column shear forces as a check These column shears are less than the yield capacity
on the validity of the chosen strength and stiffness of each column (25 k) as required in Step A3 and the
characteristics. chosen strength and stiffness values in Step B1.1 are
therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 1.0
damping ratio, , of the bridge:

Eq. 757.68
7.1-5 2 (B-14) 2 2
GSID 386.4 31.43

and = 1.57 sec

Eq. 2 , , , and taking dy,j = 0:

(B-15)
7.1-10 , , ,
GSID 2 , , 0
0.31
, , ,
where dy,j is the yield displacement of the isolator. For
friction-based isolators, dy,j = 0. For other types of
isolators dy,j is usually small compared to disol,j and has
negligible effect on , Hence it is suggested that for
the Simplified Method, set dy,j = 0 for all isolator

15

types. See Step B2.2 where the value of dy,j is revisited
for the Multimode Spectral Analysis Method.
B1.11 Damping Factor
Calculate the damping factor, BL, and the
displacement, d, of the bridge:
Eq. .
, 0.3
.
7.1-3 (B-16)
1.7, 0.3
GSID
Eq. 9.79 (B-17)
7.1-4
GSID
B1.12 Convergence Check
Compare the new displacement with the initial value
assumed in Step B1.1. If there is close agreement, go
to the next step; otherwise repeat the process from
Step B1.3 with the new value for displacement as the
assumed displacement.
This iterative process is amenable to solution using a
spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
B1.11 Damping Factor, Example 1.0
Since 0.31 0.3
1.70
and
9.79 9.79 0.2 1.57
1.81
1.70
B1.12 Convergence Check, Example 1.0
Since the calculated value for displacement, d
(=1.81 in) is not close to that assumed at the
beginning of the cycle (Step B1.1, d = 2.0), use the
value of 1.81 in as the new assumed displacement and
repeat from Step B1.3.
After one iteration, convergence is reached at a
superstructure displacement of 1.76 in, with an
effective period of 1.52 seconds, and a damping factor
of 1.70 (33% damping ratio). The displacement in the
isolators at Pier 1 is 1.61 in and the effective stiffness
of the same isolators is 15.69 k/in.
See spreadsheet in Table B1.12-1 for results of final
iteration.
Ignoring the weight of the cap beams, the sum of the
column shears at a pier must equal the total isolator
shear force. Hence, approximate column shear (per
column) = 15.69(1.61)/3 = 8.42 k which is less than
the maximum allowable (25k) if elastic behavior is to
be achieved (as required in Step A3).
Also the superstructure displacement = 1.76 in, which
is less than the available clearance of 2.0 in.
Therefore the above solution is acceptable and go to
Step B2.
Note that available clearance (2.0 in) is greater than
minimum required which is given by:
8 8 0.20 1.52
1.43
1.7
16

 Table B1.12-1 Simplified Method Solution for Design Example 1.0 Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
650.52 107.16 757.68 0.2 6

StepB1.1 d 1.72 Assumeddisplacement

Qd 32.53 Characteristicstrength
Kd 16.26 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 44.95 2.25 1.12 10000 2.40E04 2.40 1.76 2.40 4.23E04 4.23 3.958 7.444
Pier1 280.31 14.02 7.01 172.0 9.12E02 14.38 1.61 15.69 1.47E01 25.33 22.623 44.611
Pier2 280.31 14.02 7.01 172.0 9.12E02 14.38 1.61 15.69 1.47E01 25.33 22.623 44.611
Abut2 44.95 2.25 1.12 10000 2.40E04 2.40 1.76 2.40 4.23E04 4.23 3.958 7.444
Total 650.52 32.526 16.263 K eff,j 33.557 59.107 53.161 104.109
Step B1.4

StepB1.10 T eff 1.52 Effectiveperiod

0.33 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.75

B L 1.70 DampingFactor
d 1.75 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 0.375 0.187 0.400 1.66 0.413
Pier1 2.336 1.168 2.615 1.47 2.757
Pier2 2.336 1.168 2.615 1.47 2.757
Abut2 0.375 0.187 0.400 1.66 0.413

17

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and
then applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 1.0

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 (see
elastic stiffness, Kd,i, of each isolator i as follows: Table B1.12-1) by the number of isolators at each
, support (n = 6), the following values for Qd /isolator
, (B-19)
and Kd /isolator are obtained:
and
, o Qd, 1 = 2.25/6 = 0.37 k
, ( B-20)
o Qd, 2 = 14.02/6= 2.34 k
where values for Qd,j and Kd,j are obtained from the o Qd, 3 = 14.02/6= 2.34 k
final cycle of iteration in the Simplified Method (Step o Qd, 4 = 2.25/6 = 0.37 k
B1. 12 and
o Kd,1 = 1.12/6 = 0.19 k/in
o Kd,2 = 7.01/6 = 1.17 k/in
o Kd,3 = 7.01/6 = 1.17 k/in
o Kd,4 = 1.12/6 = 0.19 k/in

B2.2 Initial Stiffness and Yield Displacement B2.2 Initial Stiffness and Yield Displacement,
Calculate the initial stiffness, Ku,i, and the yield Example 1.0
displacement, dy,i, for each isolator i as follows: Since the isolator type has been specified in Step A1
(1) For friction-based isolators: Ku,i = and dy,i = 0. to be an elastomeric bearing (i.e. not a friction-based
(2) For other types of isolators, and in the absence of bearing), calculate Ku,i and dy,i for an isolator on Pier 1
isolator-specific information, take as follows:
, 10 , 10 1.17 11.7 /
, 10 , ( B-21)
and then and
, 2.34
,
( B-22) , 0.22
, , , 11.7 1.17
, ,

As expected, the yield displacement is small

compared to the expected isolator displacement (~2
in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.

B2.3 Isolator Effective Stiffness, Kisol,i B2.3 Isolator Effective Stiffness, Kisol,i, Example 1.0
Calculate the isolator stiffness, Kisol,i, of each isolator Dividing the results for Kisol (Step B1.12) among the 6
i: isolators at each support, the following values for Kisol
/isolator are obtained:
, o Kisol,1 = 2.40/6 = 0.40 k/in
, (B -23)
o Kisol,2 = 15.69/6 = 2.62 k/in
o Kisol,3 = 15.69/6 = 2.62 k/in
o Kisol,4 = 2.40/6 = 0.40 k/in

18

B2.4 Three-Dimensional Bridge Model B2.4 Three-Dimensional Bridge Model, Example

Using computer-based structural analysis software, 1.0
create a 3-dimensional model of the bridge with the Although the bridge in this Design Example is regular
isolators represented by spring elements. The and is without skew or curvature, a 3-dimensional
stiffness of each isolator element in the horizontal finite element model was developed for this Step, as
axes (Kx and Ky in global coordinates, K2 and K3 in shown below.
typical local coordinates) is the Kisol value calculated
in the previous step. For bridges with regular
geometry and minimal skew or curvature, the
superstructure may be represented by a single stick
provided the load path to each individual isolator at
each support is explicitly modeled, usually by a rigid
cap beam and a set of rigid links. If the geometry is
irregular, or if the bridge is skewed or curved, a finite
element model is recommended to accurately capture
the load carried by each individual isolator. If the
piers have an unusual weight distribution, such as a .
pier with a hammerhead cap beam, a more rigorous
model is justified.

B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in
Step A2 is for a 5% damped system. It is modified in
this step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the
effective period of the bridge, Teff, by the damping
factor, BL.
B.2.5 Composite Design Response Spectrum,
Example 1.0
From the final results of Simplified Method (Step
B1.12), BL = 1.70 and Teff = 1.52 sec. Hence the
transition in the composite spectrum from 5% to 33%
damping occurs at 0.8 Teff = 0.8 (1.46) = 1.22 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.22 sec by 1.70.
0.9

0.8

0.7
Acceleration(g)

0.6

0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Period(s)

B2.6 Multimodal Analysis of Finite Element
Model
Input the composite response spectrum as a user-
specified spectrum in the software, and define a load
case in which the spectrum is applied in the
longitudinal direction. Analyze the bridge for this
load case.
B2.6 Multimodal Analysis of Finite Element
Model, Example 1.0
Results of modal analysis of the example bridge are
summarized in Table B2.6-1 Here the modal periods
and mass participation factors of the first 12 modes
are given. The first two modes are the principal
longitudinal and transverse modes with periods of
1.41 and 1.35 sec respectively. The period of the
longitudinal mode (1.41 sec) is close to that
calculated in the Simplified Method.

19

 Table B2.6-1 Modal Properties of Bridge

Example 1.0 First Iteration
Mode Period ModalParticipatingMassRatios
No. Sec UX UY UZ RX RY RZ
1 1.410 0.761 0.000 0.000 0.000 0.003 0.000
2 1.346 0.000 0.738 0.031 0.059 0.000 0.534
3 1.325 0.000 0.000 0.000 0.000 0.000 0.217
4 0.187 0.000 0.000 0.000 0.000 0.000 0.000
5 0.186 0.125 0.000 0.000 0.000 0.000 0.000
6 0.121 0.000 0.000 0.000 0.000 0.000 0.000
7 0.121 0.000 0.000 0.000 0.000 0.000 0.006
8 0.104 0.000 0.034 0.183 0.107 0.000 0.064
9 0.101 0.000 0.000 0.000 0.000 0.226 0.000
10 0.095 0.000 0.121 0.081 0.184 0.000 0.041
11 0.094 0.000 0.000 0.000 0.000 0.000 0.000
12 0.074 0.000 0.002 0.000 0.003 0.000 0.000

Computed values for the isolator displacements due to

a longitudinal earthquake are as follows (numbers in
parentheses are those used to calculate the initial
properties to start iteration from the Simplified
Method):

o disol,1 = 1.65 (1.76) in

o disol,2 = 1.47 (1.61) in
o disol,3 = 1.47 (1.61) in
o disol,4 = 1.65 (1.76) in

B2.7 Convergence Check B2.7 Convergence Check, Example 1.0

Compare the resulting displacements at the The new superstructure displacement is 1.65 in, more
superstructure level (d) to the assumed than a 5% difference from the displacement assumed
displacements. These displacements can be obtained at the start of the Multimode Spectral Analysis.
by examining the joints at the top of the isolator
spring elements. If in close agreement, go to Step
B2.9. Otherwise go to Step B2.8.

B2.8 Update Kisol,i, Keff,j, and BL B2.8 Update Kisol,i, Keff,j, and BL, Example 1.0
Use the calculated displacements in each isolator Updated values for Kisol,i are given below (previous
element to obtain new values of Kisol,i for each values are in parentheses):
isolator as follows:
, o Kisol,1 = 0.41 (0.40) k/in
, , (B-24) o Kisol,2 = 2.76 (2.62) k/in
,
o Kisol,3 = 2.76 (2.62) k/in
Recalculate Keff,j : o Kisol,4 = 0.41 (0.40) k/in

Eq. , ,
Updated values for Keff,j, , BL and Teff are given below
7.1-6 ,
(B-25) (previous values are in parentheses):
, ,
GSID
o Keff,1 = 2.48 (2.40) k/in
Recalculate system damping ratio, : o Keff,2 = 15.48 (14.38) k/in
o Keff,3 = 15.48 (14.38) k/in
Eq. 2 , , ,
o Keff,4 = 2.48 (2.40) k/in
7.1-10 , , ,
(B-26)
GSID o = 27% (33%)
o BL = 1.66 (1.70)
Recalculate system damping factor, BL: o Teff = 1.41 (1.52) sec
The updated composite response spectrum is shown
. below:
Eq. 0.3
. ( B-27)
7.1-3 1.7 0.3

20

 GSID 0.9

0.8

Obtain the effective period of the bridge from the 0.7

Acceleration(g)
multi-modal analysis and with the revised damping 0.6

factor (Eq. B-27), construct a new composite 0.5

response spectrum. Go to Step B2.6. 0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Period(s)

B2.6 Multimodal Analysis Second Iteration,

Example 1.0
Reanalysis gives the following values for the isolator
displacements (numbers in parentheses are from the
previous cycle):

o disol,1 = 1.66 (1.65) in

B2.7 Convergence Check
Compare the resulting displacements at the
superstructure level (d) to the assumed
displacements. These displacements can be obtained
by examining the joints at the top of the isolator
spring elements. If in close agreement, go to Step
B2.9. Otherwise go to Step B2.8.
o disol,2 = 1.47 (1.47) in
o disol,3 = 1.47 (1.47) in
o disol,4 = 1.66 (1.65) in
B2.7 Convergence Check, Example 1.0
The new superstructure displacement is 1.66 in, less
than a 1% difference from the displacement assumed
at the start of the second cycle of Multimode Spectral
Analysis.

B2.9 Superstructure and Isolator Displacements
Once convergence has been reached, obtain
o superstructure displacements in the
longitudinal (xL) and transverse (yL) directions
of the bridge, and
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions of the bridge, for
each isolator, for this load case (i.e.
longitudinal loading). These displacements
may be found by subtracting the nodal
displacements at each end of each isolator
spring element.
B2.9 Superstructure and Isolator Displacements,
Example 1.0
From the above analysis:
o superstructure displacements in the
longitudinal (xL) and transverse (yL) directions
are:
xL= 1.66 in
yL= 0.0 in
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions are:
o Abutments: uL = 1.66 in, vL = 0.00 in
o Piers: uL = 1.47 in, vL = 0.00 in

All isolators at same support have the same

displacements.

B2.10 Pier Bending Moments and Shear Forces B2.10 Pier Bending Moments and Shear Forces,
Obtain the pier bending moments and shear forces in Example 1.0
the longitudinal (MPLL, VPLL) and transverse (MPTL, Maximum bending moments and shear forces in the
VPTL) directions at the critical locations for the columns in the longitudinal (MPLL,VPLL) and
longitudinally-applied seismic loading. transverse (MPTL,VPTL) directions are:
Exterior Columns:
o MPLL= 0 kft
o MPTL= 240 kft

21

 o VPLL= 15.81 k
o VPTL= 0 k
Interior Columns:
o MPLL= 0 kft
o MPTL= 235 kft
o VPLL= 15.24 k
o VPTL= 0 k

Both piers have the same distribution of bending

moments and shear forces among the columns.

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 1.0
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1

Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces
ator
ure to long. to long. due to
EQ EQ long. EQ
1 0.69 0.00 0.36
2 0.69 0.00 0.39

Abut 3 0.69 0.00 0.39

ment 4 0.69 0.00 0.39
5 0.69 0.00 0.39
6 0.69 0.00 0.36
1 4.04 0.00 0.13
2 4.05 0.00 0.19
3 4.05 0.00 0.22
Pier
4 4.05 0.00 0.22
5 4.05 0.00 0.19
6 4.04 0.00 0.13

22

 STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake C1. Analysis for Transverse Earthquake,
Repeat the above process, starting at Step B1, for Example 1.0
earthquake loading in the transverse direction of the Key results from repeating Steps B1 and B2
bridge. Support flexibility in the transverse direction (Simplified and Mulitmode Spectral Methods) are:
is to be included, and a composite response spectrum o Teff = 1.43 sec
is to be applied in the transverse direction. Obtain o Superstructure displacements in the longitudinal
isolator displacements in the longitudinal (uT) and (xT) and transverse (yT) directions are as follows:
transverse (vT) directions of the bridge, and the xT = 0 and yT = 1.53 in
biaxial bending moments and shear forces at critical o Isolator displacements in the longitudinal (uT)
locations in the columns due to the transversely- and transverse (vT) directions are as follows:
applied seismic loading. Abutments uT = 0.00 in, vT = 1.53 in
Piers uT = 0.00 in, vT = 1.49 in
o Maximum bending moments and shear forces in
the columns in the longitudinal (MPLL,VPLL) and
transverse (MPTL,VPTL) directions are:
Exterior Columns:
o MPLT= 108 kft
o MPTT= 1 kft
o VPLT= 0.06 k
o VPTT= 14.87 k
Interior Columns:
o MPLT= 120 kft
o MPTT= 0 kft
o VPLT= 0 k
o VPTT= 17.29 k

Both piers have the same distribution of

bending moments and shear forces among the
columns.

o Isolator shear and axial forces are in Table C1-1.

23

 Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces due
ator
ure to transv. to transv. to transv.
EQ EQ EQ
1 0.00 0.65 3.50
2 0.00 0.65 1.93

Abut 3 0.00 0.65 0.68

ment 4 0.00 0.65 0.68
5 0.00 0.65 1.93
6 0.00 0.65 3.50
1 0.02 3.98 12.56
2 0.01 4.00 1.14
3 0.00 4.01 2.29
Pier
4 0.00 4.01 2.29
5 0.01 4.00 1.14
6 0.02 3.98 12.56

24

STEP D. CALCULATE DESIGN VALUES

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements at Pier 1,

Following the provisions in Art. 2.1 GSID, and Example 1.0
illustrated in Fig. 2.1-1 GSID, calculate the total To illustrate the process, design displacements for the
design displacement, dt, for each isolator by outside isolator on Pier 1 are calculated below.
combining the displacements from the longitudinal
(uL and vL) and transverse (uT and vT) cases as Load Case 1:
follows: u1 = uL + 0.3uT = 1.0(1.47) + 0.3(0) = 1.47 in
u1 = uL + 0.3uT (D-1) v1 = vL + 0.3vT = 1.0(0) + 0.3(1.49) = 0.45 in
v1 = vL + 0.3vT (D-2) R1 = = 1.47 0.45 = 1.54 in
R1 = (D-3)
Load Case 2:
u2 = 0.3uL + uT (D-4) u2 = 0.3uL + uT = 0.3(1.47) + 1.0(0) = 0.44 in
v2 = 0.3vL + vT (D-5) v2 = 0.3vL + vT = 0.3(0) + 1.0(1.49) = 1.49in
R2 = (D-6) R2 = = 0.44 1.49 = 1.55in

dt = max(R1, R2) (D-7)

D2. Design Moments and Shears
Calculate design values for column bending moments
and shear forces for all piers using the same
combination rules as for displacements.
Governing Case:
Total design displacement, dt = max(R1, R2)
= 1.55in
D2. Design Moments and Shears in Pier 1,
Example 1.0
Design moments and shear forces are calculated for
Pier 1, Column 1, below to illustrate the process.

Alternatively this step may be deferred because the Load Case 1:

above results may not be final. Upper and lower VPL1= VPLL + 0.3VPLT = 1.0(15.81) + 0.3(0.06)
bound analyses are required after the isolators have = 15.83 k
been designed as described in Art 7. GSID. These VPT1= VPTL + 0.3VPTT = 1.0(0.02) + 0.3(14.87)
analyses are required to determine the effect of = 4.48 k
possible variations in isolator properties due age, R1 = = 15.83 4.48 = 16.45 k
temperature and scragging in elastomeric systems.
Accordingly the results for column shear in Steps Load Case 2:
B2.10 and C are likely to increase once these VPL2= 0.3VPLL + VPLT = 0.3(15.81) + 1.0(0.06)
analyses are complete. = 4.80 k
VPT2= 0.3VPTL + VPTT = 0.3(0.02) + 1.0(14.87)
= 14.88 k
R2 = = 4.80 14.88 = 15.63 k

Governing Case:
Design column shear = max (R1, R2)
= 16.45 k

25

STEP E. DESIGN OF LEAD-RUBBER (ELASTOMERIC) ISOLATORS

A lead-rubber isolator is an elastomeric
bearing with a lead core inserted on its vertical
centreline. When the bearing and lead core are
deformed in shear, the elastic stiffness of the
lead provides the initial stiffness (Ku).With
increasing lateral load the lead yields almost
perfectly plastically, and the post-yield
stiffness Kd is given by the rubber alone. More
details are given in MCEER 2006.

While both circular and rectangular bearings

are commercially available, circular bearings
are more commonly used. Consequently the
procedure given below focuses on circular
bearings. The same steps can be followed for
rectangular bearings, but some modifications
will be necessary. When sizing the physical dimensions of the bearing, plan dimensions (B, dL) should be rounded
up to the next 1/4 increment, while the total thickness of elastomer, Tr, is specified in multiples of the layer
thickness. Typical layer thicknesses for bearings with lead cores are 1/4 and 3/8. High quality natural rubber
should be specified for the elastomer. It should have a shear modulus in the range 60-120 psi and an ultimate
elongation-at-break in excess of 5.5. Details can be found in rubber handbooks or in MCEER 2006.

The following design procedure assumes the isolators are bolted to the masonry and sole plates. Isolators that use
shear-only connections (and not bolts) require additional design checks for stability which are not included below.
See MCEER 2006.

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

E1. Required Properties E1. Required Properties, Example 1.0

Obtain from previous work the properties required of The design of one of the exterior isolators on a pier is
the isolation system to achieve the specified given below to illustrate the design process for lead-
performance criteria (Step A1). rubber isolators.
required characteristic strength, Qd / isolator
required post-elastic stiffness, Kd / isolator From previous work
total design displacement, dt, for each isolator Qd / isolator = 2.34 k
maximum applied dead and live load (PDL, Kd / isolator = 1.17 k/in
PLL) and seismic load (PSL) which includes Total design displacement, dt = 1.55 in
seismic live load (if any) and overturning PDL = 45.52 k
forces due to seismic loads, at each isolator, PLL = 15.50 k
and PSL = 12.56 k
maximum wind load, PWL PWL = 1.76 k < Qd OK

E2. Isolator Sizing
E2.1 Lead Core Diameter E2.1 Lead Core Diameter, Example 1.0
Determine the required diameter of the lead plug, dL,
using:
2.34
(E-1) 1.61
0.9 0.9 0.9
See Step E2.5 for limitations on dL

26

E2.2 Plan Area and Isolator Diameter E2.2 Plan Area and Isolator Diameter, Example
Although no limits are placed on compressive stress 1.0
in the GSID, (maximum strain criteria are used
instead, see Step E3) it is useful to begin the sizing
process by assuming an allowable stress of, say, 1.0
ksi.

Then the bonded area of the isolator is given by:

(E-2)
1.0 45.52 15.50
61.02
1.0 1.0
and the corresponding bonded diameter (taking into
account the hole required to accommodate the lead
core) is given by:
4
(E-3)
4 4 61.02
1.61

Round the bonded diameter, B, to nearest quarter = 8.96 in

inch, and recalculate actual bonded area using
Round B up to 9.0 in and the actual bonded area is:
(E-4)
4
9.0 1.61 61.57
4
Note that the overall diameter is equal to the bonded
diameter plus the thickness of the side cover layers
(usually 1/2 inch each side). In this case the overall
diameter, Bo is given by:

1.0 (E-5)
Bo = 9.0 + 2(0.5) = 10.0 in

E2.3 Elastomer Thickness and Number of Layers E2.3 Elastomer Thickness and Number of Layers,
Since the shear stiffness of the elastomeric bearing is Example 1.0
given by:
(E-6)
where G = shear modulus of the rubber, and
Tr = the total thickness of elastomer, Select G, shear modulus of rubber, = 100 psi (0.1 ksi)
it follows Eq. E-5 may be used to obtain Tr given a
required value for Kd Then
0.1 61.57
(E-7) 5.27
1.17
A typical range for shear modulus, G, is 60-120 psi.
Higher and lower values are available and are used in
special applications.

If the layer thickness is tr, the number of layers, n, is

given by:
5.27
(E-8) 21.09
0.25
rounded up to the nearest integer.
Round to nearest integer, i.e. n = 22

27

Note that because of rounding the plan dimensions

and the number of layers, the actual stiffness, Kd, will
not be exactly as required. Reanalysis may be
necessary if the differences are large.

E2.4 Overall Height E2.4 Overall Height, Example 1.0

The overall height of the isolator, H, is given by:

1 2 ( E-9) 22 0.25 21 0.125 2 1.5 11.125

where ts = thickness of an internal shim (usually

about 1/8 in), and
tc = combined thickness of end cover plate
(0.5 in) and outer plate (1.0 in)

E2.5 Size Checks E2.5 Size Checks, Example 1.0

Experience has shown that for optimum performance Since B=9.0 check
of the lead core it must not be too small or too large.
The recommended range for the diameter is as 9.0 9.0
follows: 3 6
(E-10)
3 6 i.e., 3.0 1.5

Since dL = 1.6, lead core size is acceptable.

Art. 12.2 GSID requires that the isolation system 0.025 0.025 45.52
provides a lateral restoring force at dt greater than the , 0.66 /
1.73
restoring force at 0.5dt by not less than W/80. This
equates to a minimum Kd of 0.025W/d. As
0.1 64.15
0.025 1.17 / ,
5.5
,

E3. Strain Limit Check E3. Strain Limit Check, Example 1.0
Art. 14.2 and 14.3 GSID requires that the total
applied shear strain from all sources in a single layer
of elastomer should not exceed 5.5, i.e.,
Since
, + 0.5 5.5 (E-11)
45.52
0.739
where , , , are defined below. 61.57
(a) is the maximum shear strain in the layer due to
compression and is given by: G = 0.1 ksi
(E-12) and
where Dc is shape coefficient for compression in 61.57
8.71
circular bearings = 1.0, , G is shear 9.0 0.25
modulus, and S is the layer shape factor given by: then
(E-13) 1.0 0.739
0.849
0.1 8.71
(b) , is the shear strain due to earthquake loads
and is given by:

28

 1.58
, (E-14) , 0.282
4.5

(c) is the shear strain due to rotation and is given

by:
0.375 9.0 0.01
(E-15) 0.221
0.25 5.5
where Dr is shape coefficient for rotation in circular
bearings = 0.375, and is design rotation due to DL, Substitution in Eq E-11 gives
LL and construction effects Actual value for may
not be known at this time and value of 0.01 is , 0.5 0.85 0.28 0.5 0.22 1.24
suggested as an interim measure, including 5.5
uncertainties (see LRFD Art. 14.4.2.1)

E4. Vertical Load Stability Check E4. Vertical Load Stability Check, Example 1.0
Art 12.3 GSID requires the vertical load capacity of
all isolators be at least 3 times the applied vertical
loads (DL and LL) in the laterally undeformed state.

Further, the isolation system shall be stable under

1.2(DL+SL) at a horizontal displacement equal to
either
2 x total design displacement, dt, if in Seismic Zone 1
or 2, or
1.5 x total design displacement, dt, if in Seismic Zone
3 or 4.

E4.1 Vertical Load Stability in Undeformed State E4.1 Vertical Load Stability in Undeformed State,
The critical load capacity of an elastomeric isolator at Example 1.0
zero shear displacement is given by

4
1 1 3 3 0.1 0.3
2 (E-16)
0.3 1 0.67 8.71 15.55
where
9.0
322.1
Ts = total shim thickness 64

15.48 322.1
910.89 /
1 0.67 5.5
E = elastic modulus of elastomer = 3G
0.1 61.57
64 1.12 /
4.5
It is noted that typical elastomeric isolators have high
shape factors, S, in which case:
4
1 (E-17)

and Eq. E-16 reduces to:

(E-18) 1.12 910.89 100.33

Check that:

29

 100.33
3 (E-19) 1.64 3
45.52 15.5

E4.2 Vertical Load Stability in Deformed State E4.2 Vertical Load Stability in Deformed State,
The critical load capacity of an elastomeric isolator at Example 1.0
shear displacement may be approximated by: Since bridge is in Zone 2, 2 2 1.55 3.10

(E-20) 3.10
2 2.44
where 9.0
Ar = overlap area between top and bottom plates
of isolator at displacement (Fig. 2.2-1
GSID) 2.44 2.44
= 0.570
4
2

Agross = 4 0.570 100.33 57.16

It follows that:
(E-21)

Check that:
57.16
1 (E-22) 0.85 1
1.2 1.2 1.2 45.52 12.56

E5. Design Review E5. Design Review, Example 1.0

The basic dimensions of the isolator designed above are
as follows:

10.00 in (od) x 11.125 in (high) x 1.61 in dia. lead core

and the volume, excluding steel end and cover plates, is

638 in3.

This design satisfies the shear strain limit criteria, but

not the vertical load stability ratio in the undeformed
and deformed states.

A redesign is therefore required and the easiest way to

increase the Pcr is to increase the shape factor, S, since
the bending stiffness of an isolator is a function of the
shape factor squared. See equations in Step E4.1. To
increase S, increase the bonded area Ab while keeping tr
constant (Eq. E-13). But to keep Kd constant while
increasing Ab and Tr is constant, decrease the shear
modulus, G (Eq. E-6).

This redesign is outlined below. After repeating the

calculation for diameter of lead core, the process begins
by reducing the shear modulus to 60 psi (0.06 ksi) and

30

 increasing the bonded diameter to 12 in.

E2.1
2.34
1.61
0.9 0.9

E2.2
5.5 1.17
107.25
0.06

4 4 107.25
1.61 11.80

Round B to 12 in and the actual bonded area becomes:

12 1.61 111.06
4

Bo = 12 + 2(0.5) = 13 in

E2.3
0.06 111.06
5.71
1.17

5.71
22.82
0.25

Round to nearest integer, i.e. n = 23.

E2.4
23 0.25 22 0.125 2 1.5 11.5
E2.5
Since B=12 check

12 12
3 6

i.e., 4 2

Since dL = 1.61, the size of lead core is too small, and

there are 2 options: (1) Accept the undersize and check
for adequate performance during the Quality Control
Tests required by GSID Art. 15.2.2; or (2) Only have
lead cores in every second isolator, in which case the
core diameter, in those isolators with cores, will be 2 x
1.61 = 2.27 in (which satisfies above criterion).

0.06 111.06
1.16 / ,
5.75

E3.
45.52
0.41
111.06

31

 111.06
11.78
12 0.25

1.0 0.41
0.580
0.06 11.27

1.55
, 0.270
5.75

0.375 12 0.01
0.376
0.25 5.75

, 0.5 0.580 0.270 0.5 0.376

1.04 5.5
E4.1
3 3 0.06 0.18

0.18 1 0.67 11.78 16.93

12
1017.88
64
16.93 1017.88
2996.42 /
5.75

0.06 111.06
1.159 /
5.75

1.159 2996.42 185.13

185.13
3.03 3
45.52 15.50

E4.2
3.10
2 2.62
12

2.62 2.62
0.674

0.674 185.13 124.84

124.84
1.86 1
1.2 1.2 45.52 12.56

E5.
The basic dimensions of the redesigned isolator are as
follows:

32

 13.0 in (od) x 11.5 in (high) x 1.61 in dia. lead core and

its volume (excluding steel end and cover plates) is 1128
in3.

This design meets all the design criteria but is about 75%
larger by volume than the previous design. This increase
in size is driven by the need to satisfy the vertical load
stability ratio of 3.0 in the undeformed state.

E6. Minimum and Maximum Performance Check E6. Minimum and Maximum Performance Check,
Art. 8 GSID requires the performance of any Example 1.0
isolation system be checked using minimum and Minimum Property Modification factors are:
maximum values for the effective stiffness of the min,Kd = 1.0
system. These values are calculated from minimum min,Qd = 1.0
and maximum values of Kd and Qd, which are found
using system property modification factors, as which means there is no need to reanalyze the bridge
indicated in Table E6-1. with a set of minimum values.
Table E6-1. Minimum and maximum values
for Kd and Qd. Maximum Property Modification factors are:
max,a,Kd = 1.1
Eq. max,a,Qd = 1.1
8.1.2-1 Kd,max = Kd max,Kd (E-23)
GSID max,t,Kd = 1.1
Eq. max,t,Qd = 1.4
8.1.2-2 Kd,min = Kd min,Kd (E-24)
GSID max,scrag,Kd = 1.0
Eq. max,scrag,Qd = 1.0
8.1.2-3 Qd,max = Qd max,Qd (E-25) Applying a system adjustment factor of 0.66 for an
GSID other bridge, the maximum property modification
Eq. factors become:
8.1.2-4 Qd,min = Qd min,Qd (E-26)
GSID max,a,Kd = 1.0 + 0.1(0.66) = 1.066
max,a,Qd = 1.0 + 0.1(0.66) = 1.066
Determination of the system property modification
factors should include consideration of the effects of max,t,Kd = 1.0 + 0.1(0.66) = 1.066
temperature, aging, scragging, velocity, travel (wear) max,t,Qd = 1.0 + 0.4(0.66) = 1.264
and contamination as shown in Table E6-2. In lieu of
tests, numerical values for these factors can be max,scrag,Kd = 1.0
obtained from Appendix A, GSID. max,scrag,Qd = 1.0

Table E6-2. Minimum and maximum values for Therefore the maximum overall modification factors
system property modification factors.
max,Kd = 1.066(1.066)1.0 = 1.14
Eq. min,Kd = (min,t,Kd) (min,a,Kd) max,Qd = 1.066(1.264)1.0 = 1.35
8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-27)
GSID (min,scrag,Kd) Since the possible variation in upper bound properties
Eq. max,Kd = (max,t,Kd) (max,a,Kd) exceeds 15% a reanalysis of the bridge is required to
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-28) determine performance with these properties.
GSID (max,scrag,Kd)
Eq. min,Qd = (min,t,Qd) (min,a,Qd) The upper-bound properties are:
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-29) Qd,max = 1.35 (2.34) = 3.16 k
GSID (min,scrag,Qd) and
Kd,ma x=1.14(1.16) = 1.32 k/in
Eq. max,Qd = (max,t,Qd) (max,a,Qd)
(E-30)
8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd)

33

 GSID (max,scrag,Qd)

Adjustment factors are applied to individual -factors

(except v) to account for the likelihood of
occurrence of all of the maxima (or all of the
minima) at the same time. These factors are applied
to all -factors that deviate from unity but only to the
portion of the -factor that is greater than, or less
than, unity. Art. 8.2.2 GSID gives these factors as
follows:
1.00 for critical bridges
0.75 for essential bridges
0.66 for all other bridges

As required in Art. 7 GSID and shown in Fig. C7-1

GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Qd,max. As indicated in Fig C7-1 GSID, maximum
displacements will probably be given by the first case
(Kd,min and Qd,min) and maximum forces by the second
case (Kd,max and Qd,max).

E7. Design and Performance Summary E7. Design and Performance Summary, Example 1.0
E7.1 Isolator dimensions E7.1 Isolator dimensions, Example 1.0
Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E7.1-1.
Overall diameter (includes cover layer)
Overall height Table E7.1-1 Isolator Dimensions
Diameter lead core
Isolator Overall size Overall size Diam.
Bonded diameter Location including without lead core
Number of rubber layers mounting mounting (in)
Thickness of rubber layers plates (in) plates (in)
Total rubber thickness Under
Thickness of steel shims edge 17.0 x 17.0 13.0 dia.
Shear modulus of elastomer 1.61
girder x 11.5 (H) x 10.0 (H)
on Pier 1
Check all dimensions with manufacturer.
Isolator No. of Rubber Total Steel
Location rubber layers rubber shim
layers thick- thick- thick-
ness ness ness
(in) (in) (in)
Under
edge
23 0.25 5.75 0.125
girder
on Pier 1

Shear modulus of elastomer = 60 psi

E7.2 Bridge Performance E7.2 Bridge Performance, Example 1.0

Summarize bridge performance Bridge performance is summarized in Table E7.2-1
Maximum superstructure displacement where it is seen that the maximum column shear is
(longitudinal) 18.03 k. This less than the column plastic shear (25 k)
Maximum superstructure displacement and therefore the required performance criterion is

34

 (transverse) satisfied (fully elastic behavior). Furthermore the

Maximum superstructure displacement
(resultant)
Maximum column shear (resultant)
Maximum column moment (about transverse
axis)
Maximum column moment (about
longitudinal axis)
Maximum column torque
Check required performance as determined in Step
A3, is satisfied.
maximum longitudinal displacement is 1.66 in which is
less than the 2.0 in available at the abutment expansion
joints and is therefore acceptable.
Table E7.2-1 Summary of Bridge Performance
Maximum superstructure
1.66 in
displacement (longitudinal)
Maximum superstructure
1.54 in
displacement (transverse)
Maximum superstructure
1.72 in
displacement (resultant)
Maximum column shear (resultant) 18.03 k
Maximum column moment about
242 kft
transverse axis
Maximum column moment about
121 kft
longitudinal axis
Maximum column torque 1.82 kft

35

 SECTION 1
PC Girder Bridge, Short Spans, Multi-Column Piers

DESIGN EXAMPLE 1.1: Site Class D

Design Examples in Section 1

Site
ID Description S1 Pier height Skew Isolator type
Class
1.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
1.1 Change site class 0.2g D Same 0 Lead rubber bearing
Change spectral
1.2 0.6g B Same 0 Lead rubber bearing
acceleration, S1
Spherical friction
1.3 Change isolator to SFB 0.2g B Same 0
bearing
1.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
1.5 Change column height 0.2g B H1=0.5 H2 0 Lead rubber bearing
1.6 Change angle of skew 0.2g B Same 450 Lead rubber bearing

36

DESIGN PROCEDURE
STEP A: BRIDGE AND SITE DATA
A1. Bridge Properties
Determine properties of the bridge:
number of supports, m
number of girders per support, n
angle of skew
weight of superstructure including railings,
curbs, barriers and to the permanent loads,
WSS
weight of piers participating with
superstructure in dynamic response, WPP
weight of superstructure, Wj, at each
support
stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge. The calculation of these
quantities requires careful consideration of
several factors such as the use of cracked
sections when estimating column or wall
flexural stiffness, foundation flexibility,
and effective column height.
column shear strength (minimum value).
This will usually be derived from the
minimum value of the column flexural
yield strength, the column height, and
whether the column is acting in single or
double curvature in the direction under
consideration.
allowable movement at expansion joints
isolator type if known, otherwise to be
determined
A2. Seismic Hazard
Determine seismic hazard at site:
acceleration coefficients
site class and site factors
seismic zone
Plot response spectrum.
Use Art. 3.1 GSID to obtain peak ground and spectral
acceleration coefficients. These coefficients are the
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (e.g. Figures
3.10.2.1-1 to 3.10.2.1-21 LRFD).
Use Art. 3.2 to obtain Site Class and corresponding Site
Factors (Fpga, Fa and Fv). These data are the same as for
DESIGN EXAMPLE 1.1 (Site Class D)
A1. Bridge Properties, Example 1.1
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 6
Number of columns per support = 3
Angle of skew = 00
Weight of superstructure including permanent
loads, WSS = 650.52 k
Weight of superstructure at each support:
o W1 = 44.95 k
o W2 = 280.31 k
o W3 = 280.31 k
o W4 = 44.95 k
Participating weight of piers, WPP = 107.16 k
Effective weight (for calculation of period),
Weff = Wss + WPP = 757.68 k
Stiffness of each pier in the longitudinal
direction:
o Ksub,pier1,long = 172.0 k/in
o Ksub,pier2,long = 172.0 k/in
Stiffness of each pier in the transverse direction:
o Ksub,pier1,trans = 687.0 k/in
o Ksub,pier2,trans = 687.0 k/in
Minimum column shear strength based on
flexural yield capacity of column = 25 k
Displacement capacity of expansion joints
(longitudinal) = 2.0 in for thermal and other
movements.
Lead rubber isolators
A2. Seismic Hazard, Example 1.1
Acceleration coefficients for bridge site are given in
design statement as follows:
PGA = 0.40
S1 = 0.20
SS = 0.75
Bridge is on a stiff soil site with shear wave velocity
in upper 100 ft of soil = 1,000 ft/sec.
Table 3.10.3.1-1 LRFD gives Site Class as D.
Tables 3.10.3.2-1, -2 and -3 LRFD give following
Site Factors:
Fpga = 1.1
Fa = 1.2
Fv = 2.0
37

conventional bridges and Art 3.2 refers the designer to

the corresponding articles in the LRFD Specifications,
i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3,
LRFD.

Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give

modified spectral acceleration coefficients that include As = Fpga PGA = 1.1(0.40) = 0.44
site effects as follows: SDS = Fa SS = 1.2(0.75) = 0.9
As = Fpga PGA SD1 = Fv S1 = 2.0(0.20) = 0.40
SDS = Fa SS
SD1 = Fv S1

Seismic Zone is determined by value of SD1 in Since 0.30 < SD1 < 0.50, bridge is located in Seismic
accordance with provisions in Table5-1 GSID. Zone 3.

These coefficients are used to plot design response Design Response Spectrum is as below:
spectrum as shown in Fig. 4-1 GSID.
1.0
0.9
0.8

Acceleration(g)
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Period(s)

A3. Performance Requirements A3. Performance Requirements, Example 1.1

Determine required performance of isolated bridge In this example, assume the owner has specified full
during Design Earthquake (1000-yr return period). functionality following the earthquake and therefore
the columns must remain elastic (no yield).
Examples of performance that might be specified by
the Owner include: To remain elastic the maximum lateral load on the
Reduced displacement ductility demand in pier must be less than the load to yield any one
columns, so that bridge is open for emergency column (25 k).
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in The maximum shear in the column must therefore be
columns or yield elsewhere in bridge), so that less than 25 k in order to keep the column elastic and
bridge is fully functional and open to all vehicles meet the required performance criterion.
immediately following earthquake.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.

38

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for Isolator Force, F

non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most have
bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use Ku
equivalent linear springs and viscous damping to
represent the isolators, so that linear methods of analysis
disol
may be used to determine response. Since equivalent Ku
dy Isolator
Ku Displacement, d
properties such as Kisol are dependent on displacement (d),
and the displacements are not known at the beginning of
the analysis, an iterative approach is required. Note that in Kd
Art 7.1, GSID, keff is used for the effective stiffness of an
isolator unit and Keff is used for the effective stiffness of a disol = Isolator displacement
dy = Isolator yield displacement
combined isolator and substructure unit. To minimize Fisol = Isolator shear force
confusion, Kisol is used in this document in place of keff. Fy = Isolator yield force
Kd = Post-elastic stiffness of isolator
There is no change in the use of Keff and Keff,j, but Ksub is Kisol = Effective stiffness of isolator
used in place of ksub. Ku
Qd
= Loading and unloading stiffness (elastic stiffness)
= Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement (disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 1.1
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.40 4.0
GSID

(2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not
occur under non-seismic loads (e.g. wind) but

39

 low enough that yield will occur during an Due to larger estimated displacements (Eq B-1) than
earthquake. Experience has shown that for the benchmark bridge, Qd is increased to 7.5% of
taking Qd to be 5% of the bridge weight is a the bridge weight to introduce additional damping
good starting point, i.e. and reduce these displacements as much as possible,
0.05 (B-2) i.e.,
(3) Post-yield stiffness, Kd
Art 12.2 GSID requires that all isolators 0.075 0.075 650.52 48.79
exhibit a minimum lateral restoring force at
the design displacement, which translates to a Also, in view of these larger displacements, the post
minimum post yield stiffness Kd,min given by yield stiffness is increased to 0.1W/d, to give
Eq. B-3. essentially the same value for Kd found to be
satisfactory in Example 1.0.
Art.
0.025
12.2 , (B-3) 650.52
GSID 0.1 0.1 16.26 /
4.0
Experience has shown that a good starting
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 1.1
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength, ,

Qd, and stiffness, Kd, values in proportion to the dead o Qd, 1 = 3.37 k
load applied at that support: o Qd, 2 = 21.02 k
o Qd, 3 = 21.02 k
o Qd, 4 = 3.37 k

, (B-4)
and
and
,
, (B-5) o Kd,1 = 1.12 k/in
o Kd,2 = 7.01 k/in
o Kd,3 = 7.01 k/in
o Kd,4 = 1.12 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 1.1
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of , ,
the isolators at support j (Kisol,j) and the stiffness of , ,
the substructure Ksub,j. See figure below for
definitions (after Fig. 7.1-1 GSID). o 1 = 1.97x10-4
o 2 = 7.35x10-2
An expression for Keff,j, is given in Eq.7.1-6 GSID, but o 3 = 7.35x10-2
a more useful formula as follows (MCEER 2006): o 4 = 1.97x10-4
,
, (B-6)
1 ,
where ,
1
, ,
(B-7)
, , o Keff,1 = 1.97 k/in
and Ksub,j for the piers are given in Step A1. For the o Keff,2 = 11.78 k/in
abutments, take Ksub,j to be a large number, say 10,000 o Keff,3 = 11.78 k/in
k/in, unless actual stiffness values are available. Note o Keff,4 = 1.97 k/in
that if the default option is chosen, unrealistically high

40

 values for Ksub,j will give unconservative results for

column moments and shear forces.

F
Kd
Qd Kisol

dy disol
Superstructure
Isolator Effective Stiffness, Kisol

F
Isolator(s), Kisol
Ksub

Substructure, Ksub
dsub

Substructure Stiffness, Ksub

dsub disol F

d
Keff

d = disol + dsub

Combined Effective Stiffness, Keff

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 1.1
Calculate the total effective stiffness, Keff, of the
bridge:

Eq. , 27.50 /
7.1-6 , ( B-8)
GSID

B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 1.1
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1

(B-9) o disol,1 = 4.00 in

,
1 o disol,2 = 3.73 in
o disol,3 = 3.73 in
o disol,4 = 4.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the effective stiffness of the isolation system Example 1.1
at support j, Kisol,j, for all supports:
,
, ,
,
,
(B-10) o Kisol,1 = 1.97 k/in
, ,
, o Kisol,2 = 12.65 k/in
o Kisol,3 = 12.65 k/in
o Kisol,4 = 1.97 k/in

41

B1.7 Substructure Displacement at Each Support
B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j,
Example 1.1
for all supports:
, ,

, , (B-11)
o dsub,1 = 7.86x10-4 in
o d sub,2 = 2.74x10-1 in
o d sub,3 = 2.74x10-1 in
o d sub,4 = 7.86x10-4 in
B1.8 Lateral Load in Each Substructure Support B1.8 Lateral Load in Each Substructure, Example
Calculate the shear at support j, Fsub,j, for all 1.1
supports:
, , , (B-12) , , ,

where values for Ksub,j are given in Step A1. o F sub,1 = 7.86 k
o F sub,2 = 47.13 k
o F sub,3 = 47.13 k
o F sub,4 = 7.86 k

B1.9 Column Shear Force at Each Support B1.9 Column Shear Force at Each Support,
Calculate the shear in column k at support j, Fcol,j,k, Example 1.1
assuming equal distribution of shear for all columns at
support j:
,
, ,
#
,
, ,
# (B-13) o F col,2,1-3 15.71 k
o F col,3,1-3 15.71 k

Use these approximate column shear forces as a check These column shears are less than the yield capacity
on the validity of the chosen strength and stiffness of each column (25 k) as required in Step A3 and the
characteristics. chosen strength and stiffness values in Step B1.1 are
therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 1.1
damping ratio, , of the bridge:

Eq. 757.68
7.1-5 2 (B-14) 2 2
GSID 386.4 27.50

and = 1.68 sec

Eq. 2 , , , and taking dy,j = 0:

(B-15)
7.1-10 , , ,
GSID 2 , , 0
0.27
, , ,
where dy,j is the yield displacement of the isolator. For
friction-based isolators, dy,j = 0. For other types of
isolators dy,j is usually small compared to disol,j and has
negligible effect on , Hence it is suggested that for
the Simplified Method, set dy,j = 0 for all isolator

42

types. See Step B2.2 where the value of dy,j is revisited
for the Multimode Spectral Analysis Method.
B1.11 Damping Factor
Calculate the damping factor, BL, and the
displacement, d, of the bridge:
Eq. .
, 0.3
.
7.1-3 (B-16)
1.7, 0.3
GSID
Eq. 9.79 (B-17)
7.1-4
GSID
B1.12 Convergence Check
Compare the new displacement with the initial value
assumed in Step B1.1. If there is close agreement, go
to the next step; otherwise repeat the process from
Step B1.3 with the new value for displacement as the
assumed displacement.
This iterative process is amenable to solution using a
spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
B1.11 Damping Factor, Example 1.1
Since 0.27 0.3
1.65
and
9.79 9.79 0.4 1.68
3.98
1.65
B1.12 Convergence Check, Example 1.1
Since the calculated value for displacement, d
(=3.98 in) is very close to that assumed at the
beginning of the cycle (Step B1.1, d = 4.0), use the
value of 3.98 in as the displacement for the start of the
Multimode Spectral Analysis.
The final values to be used for the Multimode Spectral
Analysis are; an effective period of 1.68 seconds, a
damping factor of 1.65 (27% damping ratio). The
displacement in the isolators at Pier 1 is 3.71 in and
the effective stiffness of the same isolators is
12.68 k/in.
See spreadsheet in Table B1.12-1 for results of final
iteration.
Ignoring the weight of the cap beams, the sum of the
column shears at a pier must equal the total isolator
shear force. Hence, approximate column shear (per
column) = 12.68(3.71)/3 = 15.68 k which is less than
the maximum allowable (25k) if elastic behavior is to
be achieved (as required in Step A3).
Also the superstructure displacement = 3.98 in, which
is more than the available clearance of 2.0 in. It is
therefore apparent that the clearance should be
increased to 4.0 in, in which case the above solution is
acceptable and go to Step B2.
Note that if the available clearance is increased to 4.0
in, it will satisfy the minimum requirement given by:
8 8 0.40 1.68
3.26
1.65
43

 Table B1.12-1 Simplified Method Solution for Design Example 1.1 Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
650.52 107.16 757.68 0.4 6

StepB1.1 d 3.98 Assumeddisplacement

Qd 48.79 Characteristicstrength
Kd 16.26 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 44.95 3.37 1.12 10000 1.97E04 1.97 3.98 1.97 7.84E04 7.84 13.417 31.220
Pier1 280.31 21.02 7.01 172.0 7.37E02 11.81 3.71 12.68 2.73E01 47.00 77.946 187.100
Pier2 280.31 21.02 7.01 172.0 7.37E02 11.81 3.71 12.68 2.73E01 47.00 77.946 187.100
Abut2 44.95 3.37 1.12 10000 1.97E04 1.97 3.98 1.97 7.84E04 7.84 13.417 31.220
Total 650.52 48.789 16.260 K eff,j 27.554 109.686 182.726 436.639
Step B1.4

StepB1.10 T eff 1.68 Effectiveperiod

0.27 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.65

B L 1.65 DampingFactor
d 3.98 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 0.562 0.187 0.328 3.81 0.335
Pier1 3.504 1.168 2.113 3.41 2.195
Pier2 3.504 1.168 2.113 3.41 2.195
Abut2 0.562 0.187 0.328 3.81 0.335

44

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and
then applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 1.1

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 (see
elastic stiffness, Kd,i, of each isolator i as follows: Table B1.12-1) by the number of isolators at each
, support (n = 6), the following values for Qd /isolator
, (B-19)
and Kd /isolator are obtained:
and
, o Qd, 1 = 3.37/6 = 0.56 k
, ( B-20)
o Qd, 2 = 21.02/6= 3.50 k
where values for Qd,j and Kd,j are obtained from the o Qd, 3 = 21.02/6= 3.50 k
final cycle of iteration in the Simplified Method (Step o Qd, 4 = 3.37/6 = 0.56 k
B1. 12 and
o Kd,1 = 1.12/6 = 0.19 k/in
o Kd,2 = 7.01/6 = 1.17 k/in
o Kd,3 = 7.01/6 = 1.17 k/in
o Kd,4 = 1.12/6 = 0.19 k/in

B2.2 Initial Stiffness and Yield Displacement B2.2 Initial Stiffness and Yield Displacement,
Calculate the initial stiffness, Ku,i, and the yield Example 1.1
displacement, dy,i, for each isolator i as follows: Since the isolator type has been specified in Step A1
(1) For friction-based isolators: Ku,i = and dy,i = 0. to be an elastomeric bearing (i.e. not a friction-based
(2) For other types of isolators, and in the absence of bearing), calculate Ku,i and dy,i for an isolator on Pier 1
isolator-specific information, take as follows:
, 10 , 10 1.17 11.7 /
, 10 , ( B-21)
and then and
, 3.50
,
( B-22) , 0.33
,
, , 11.7 1.17
, ,

As expected, the yield displacement is small

compared to the expected isolator displacement (~4
in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.

B2.3 Isolator Effective Stiffness, Kisol,i B2.3 Isolator Effective Stiffness, Kisol,i, Example 1.1
Calculate the isolator stiffness, Kisol,i, of each isolator Dividing the results for Kisol (Step B1.12) among the 6
i: isolators at each support, the following values for Kisol
/isolator are obtained:
, o Kisol,1 = 1.97/6 = 0.33 k/in
, (B -23)
o Kisol,2 = 12.68/6 = 2.11 k/in
o Kisol,3 = 12.68/6 = 2.11 k/in

45

 o Kisol,4 = 1.97/6 = 0.33 k/in

B2.4 Three-Dimensional Bridge Model B2.4 Three-Dimensional Bridge Model, Example

Using computer-based structural analysis software, 1.1
create a 3-dimensional model of the bridge with the Although the bridge in this Design Example is regular
isolators represented by spring elements. The and is without skew or curvature, a 3-dimensional
stiffness of each isolator element in the horizontal finite element model was developed for this Step, as
axes (Kx and Ky in global coordinates, K2 and K3 in shown below.
typical local coordinates) is the Kisol value calculated
in the previous step. For bridges with regular
geometry and minimal skew or curvature, the
superstructure may be represented by a single stick
provided the load path to each individual isolator at
each support is explicitly modeled, usually by a rigid
cap beam and a set of rigid links. If the geometry is
irregular, or if the bridge is skewed or curved, a finite
element model is recommended to accurately capture
the load carried by each individual isolator. If the
piers have an unusual weight distribution, such as a .
pier with a hammerhead cap beam, a more rigorous
model is justified.

B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in
Step A2 is for a 5% damped system. It is modified in
this step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the
effective period of the bridge, Teff, by the damping
factor, BL.
B.2.5 Composite Design Response Spectrum,
Example 1.1
From the final results of Simplified Method (Step
B1.12), BL = 1.65 and Teff = 1.68 sec. Hence the
transition in the composite spectrum from 5% to 27%
damping occurs at 0.8 Teff = 0.8 (1.68) = 1.34 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.34 sec by 1.65.
1.0
0.9
0.8
Acceleration(g)

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Period(s)

B2.6 Multimodal Analysis of Finite Element
Model
Input the composite response spectrum as a user-
specified spectrum in the software, and define a load
case in which the spectrum is applied in the
longitudinal direction. Analyze the bridge for this
load case.
B2.6 Multimodal Analysis of Finite Element
Model, Example 1.1
Results of modal analysis of the example bridge are
summarized in Table B2.6-1 Here the modal periods
and mass participation factors of the first 12 modes
are given. The first two modes are the principal
longitudinal and transverse modes with periods of
1.55 and 1.49 sec respectively.

46

 Table B2.6-1 Modal Properties of Bridge

Example 1.1 First Iteration
Mode Period ModalParticipatingMassRatios
No. Sec UX UY UZ RX RY RZ
1 1.550 0.756 0.000 0.000 0.000 0.002 0.000
2 1.492 0.000 0.737 0.000 0.031 0.000 0.529
3 1.467 0.000 0.000 0.000 0.000 0.000 0.210
4 0.189 0.000 0.000 0.000 0.000 0.000 0.000
5 0.189 0.130 0.000 0.000 0.000 0.000 0.000
6 0.121 0.000 0.000 0.000 0.000 0.000 0.000
7 0.121 0.000 0.000 0.000 0.000 0.000 0.007
8 0.104 0.000 0.036 0.000 0.180 0.000 0.026
9 0.101 0.000 0.000 0.237 0.000 0.184 0.000
10 0.095 0.000 0.120 0.000 0.083 0.000 0.086
11 0.094 0.000 0.000 0.014 0.000 0.011 0.000
12 0.074 0.000 0.002 0.000 0.000 0.000 0.001

Computed values for the isolator displacements due to

a longitudinal earthquake are as follows (numbers in
parentheses are those used to calculate the initial
properties to start iteration from the Simplified
Method):

o disol,1 = 3.73 (3.98) in

o disol,2 = 3.36 (3.71) in
o disol,3 = 3.36 (3.71) in
o disol,4 = 3.36 (3.98) in

B2.7 Convergence Check B2.7 Convergence Check, Example 1.1

Compare the resulting displacements at the
superstructure level (d) to the assumed
displacements. These displacements can be obtained
by examining the joints at the top of the isolator
spring elements. If in close agreement, go to Step
B2.9. Otherwise go to Step B2.8.
The new superstructure displacement is 3.73 in, more
than a 5% difference from the displacement assumed
at the start of the Multimode Spectral Analysis
(3.98 in).

B2.8 Update Kisol,i, Keff,j, and BL B2.8 Update Kisol,i, Keff,j, and BL, Example 1.1
Use the calculated displacements in each isolator Updated values for Kisol,i are given below (previous
element to obtain new values of Kisol,i for each values are in parentheses):
isolator as follows:
, o Kisol,1 = 0.34 (0.33) k/in
, , (B-24) o Kisol,2 = 2.21 (2.11) k/in
,
o Kisol,3 = 2.21 (2.11) k/in
Recalculate Keff,j : o Kisol,4 = 0.34 (0.33) k/in

Eq. , ,
Updated values for Keff,j, , BL and Teff are given below
7.1-6 ,
(B-25) (previous values are in parentheses):
, ,
GSID
o Keff,1 = 2.03 (1.97) k/in
Recalculate system damping ratio, : o Keff,2 = 12.64 (12.68) k/in
o Keff,3 = 12.64 (12.68) k/in
Eq. 2 , , ,
o Keff,4 = 2.03 (1.97) k/in
7.1-10 , , ,
(B-26)
GSID o = 23% (27%)
o BL = 1.59 (1.65)
Recalculate system damping factor, BL: o Teff = 1.55 (1.68) sec

. The updated composite response spectrum is shown

Eq. 0.3
. ( B-27) below:
7.1-3 1.7 0.3

47

 GSID 1.0
0.9
0.8
Obtain the effective period of the bridge from the

Acceleration(g)
0.7
multi-modal analysis and with the revised damping 0.6
factor (Eq. B-27), construct a new composite 0.5
response spectrum. Go to Step B2.6. 0.4
0.3
0.2
0.1
0.0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Period(s)

B2.6 Multimodal Analysis Second Iteration,

Example 1.1
Reanalysis gives the following values for the isolator
displacements (numbers in parentheses are from the
previous cycle):

o disol,1 = 3.81 (3.73) in

o disol,2 = 3.41 (3.36) in
o disol,3 = 3.41 (3.36) in
o disol,4 = 3.81 (3.73) in
o
B2.7 Convergence Check B2.7 Convergence Check, Example 1.1
Compare the resulting displacements at the The new superstructure displacement is 3.81 in, a 2%
superstructure level (d) to the assumed difference from the displacement assumed at the start
displacements. These displacements can be obtained of the second cycle of Multimode Spectral Analysis
by examining the joints at the top of the isolator (3.73 in).
spring elements. If in close agreement, go to Step
B2.9. Otherwise go to Step B2.8.

B2.9 Superstructure and Isolator Displacements
Once convergence has been reached, obtain
o superstructure displacements in the
longitudinal (xL) and transverse (yL) directions
of the bridge, and
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions of the bridge, for
each isolator, for this load case (i.e.
longitudinal loading). These displacements
may be found by subtracting the nodal
displacements at each end of each isolator
spring element.
B2.9 Superstructure and Isolator Displacements,
Example 1.1
From the above analysis:
o superstructure displacements in the
longitudinal (xL) and transverse (yL) directions
are:
xL= 3.81 in
yL= 0.0 in
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions are:
o Abutments: uL = 3.81 in, vL = 0.00 in
o Piers: uL = 3.41 in, vL = 0.00 in

All isolators at same support have the same

displacements.
B2.10 Pier Bending Moments and Shear Forces B2.10 Pier Bending Moments and Shear Forces,
Obtain the pier bending moments and shear forces in Example 1.1
the longitudinal (MPLL, VPLL) and transverse (MPTL, Maximum bending moments and shear forces in the
VPTL) directions at the critical locations for the columns in the longitudinal (MPLL,VPLL) and
longitudinally-applied seismic loading. transverse (MPTL,VPTL) directions are:
Exterior Columns:
o MPLL= 0 kft
o MPTL= 360 kft
o VPLL= 22.91 k

48

 o VPTL= 0.02 k
Interior Columns:
o MPLL= 0 kft
o MPTL= 354 kft
o VPLL= 22.08 k
o VPTL= 0.00 k

Both piers have the same distribution of bending

moments and shear forces among the columns.

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 1.1
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1

Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces
ator
ure to long. to long. due to
EQ EQ long. EQ
1 1.29 0.00 0.60
2 1.29 0.00 0.61

Abut 3 1.29 0.00 0.61

ment 4 1.29 0.00 0.61
5 1.29 0.00 0.61
6 1.29 0.00 0.60
1 7.53 0.00 0.20
2 7.54 0.00 0.24
3 7.55 0.00 0.25
Pier
4 7.55 0.00 0.25
5 7.54 0.00 0.24
6 7.53 0.00 0.20

49

 STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake C1. Analysis for Transverse Earthquake,
Repeat the above process, starting at Step B1, for Example 1.1
earthquake loading in the transverse direction of the Key results from repeating Steps B1 and B2
bridge. Support flexibility in the transverse direction (Simplified and Mulitmode Spectral Methods) are:
is to be included, and a composite response spectrum o Teff = 1.56 sec
is to be applied in the transverse direction. Obtain o Superstructure displacements in the longitudinal
isolator displacements in the longitudinal (uT) and (xT) and transverse (yT) directions are as follows:
transverse (vT) directions of the bridge, and the xT = 0 and yT = 3.53 in
biaxial bending moments and shear forces at critical o Isolator displacements in the longitudinal (uT)
locations in the columns due to the transversely- and transverse (vT) directions are as follows:
applied seismic loading. Abutments uT = 0.00 in, vT = 3.53 in
Piers uT = 0.00 in, vT = 3.45 in
o Maximum bending moments and shear forces in
the columns in the longitudinal (MPLL,VPLL) and
transverse (MPTL,VPTL) directions are:
Exterior Columns:
o MPLT= 153 kft
o MPTT= 0 kft
o VPLT= 0.06 k
o VPTT= 20.829 k
Interior Columns:
o MPLT= 172 kft
o MPTT= 0 kft
o VPLT= 0.00 k
o VPTT= 24.68 k

Both piers have the same distribution of

bending moments and shear forces among the
columns.

o Isolator shear and axial forces are in Table C1-1.

50

 Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces due
ator
ure to transv. to transv. to transv.
EQ EQ EQ
1 0.00 1.22 4.37
2 0.00 1.22 2.41

Abut 3 0.00 1.22 0.83

ment 4 0.00 1.22 0.83
5 0.00 1.22 2.41
6 0.00 1.22 4.37
1 0.02 7.50 15.56
2 0.01 7.52 1.79
3 0.00 7.53 2.72
Pier
4 0.00 7.53 2.72
5 0.01 7.52 1.79
6 0.02 7.50 15.56

51

STEP D. CALCULATE DESIGN VALUES

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements at Pier 1,

Following the provisions in Art. 2.1 GSID, and Example 1.1
illustrated in Fig. 2.1-1 GSID, calculate the total To illustrate the process, design displacements for the
design displacement, dt, for each isolator by outside isolator on Pier 1 are calculated below.
combining the displacements from the longitudinal
(uL and vL) and transverse (uT and vT) cases as Load Case 1:
follows: u1 = uL + 0.3uT = 1.0(3.41) + 0.3(0.01) = 3.41 in
u1 = uL + 0.3uT (D-1) v1 = vL + 0.3vT = 1.0(0) + 0.3(3.44) = 1.03 in
v1 = vL + 0.3vT (D-2) R1 = = 3.41 1.03 = 3.56 in
R1 = (D-3)
Load Case 2:
u2 = 0.3uL + uT (D-4) u2 = 0.3uL + uT = 0.3(3.41) + 1.0(0.01) = 1.03 in
v2 = 0.3vL + vT (D-5) v2 = 0.3vL + vT = 0.3(0) + 1.0(3.44) = 3.44 in
R2 = (D-6) R2 = = 1.03 3.44 = 3.59 in

dt = max(R1, R2) (D-7)

D2. Design Moments and Shears
Calculate design values for column bending moments
and shear forces for all piers using the same
combination rules as for displacements.
Governing Case:
Total design displacement, dt = max(R1, R2)
= 3.59 in
D2. Design Moments and Shears in Pier 1,
Example 1.1
Design moments and shear forces are calculated for
Pier 1, Column 1, below to illustrate the process.

Alternatively this step may be deferred because the Load Case 1:

above results may not be final. Upper and lower VPL1= VPLL + 0.3VPLT = 1.0(22.91) + 0.3(0.06)
bound analyses are required after the isolators have = 22.93 k
been designed as described in Art 7. GSID. These VPT1= VPTL + 0.3VPTT = 1.0(0.02) + 0.3(20.83)
analyses are required to determine the effect of = 6.27 k
possible variations in isolator properties due age, R1 = = 22.93 6.27 = 23.77 k
temperature and scragging in elastomeric systems.
Accordingly the results for column shear in Steps Load Case 2:
B2.10 and C are likely to increase once these VPL2= 0.3VPLL + VPLT = 0.3(22.91) + 1.0(0.06)
analyses are complete. = 6.93 k
VPT2= 0.3VPTL + VPTT = 0.3(0.02) + 1.0(20.83)
= 20.84 k
R2 = = 6.93 20.84 = 21.96 k

Governing Case:
Design column shear = max (R1, R2)
= 23.77 k

52

STEP E. DESIGN OF LEAD-RUBBER (ELASTOMERIC) ISOLATORS

A lead-rubber isolator is an elastomeric
bearing with a lead core inserted on its vertical
centreline. When the bearing and lead core are
deformed in shear, the elastic stiffness of the
lead provides the initial stiffness (Ku).With
increasing lateral load the lead yields almost
perfectly plastically, and the post-yield
stiffness Kd is given by the rubber alone. More
details are given in MCEER 2006.

While both circular and rectangular bearings

are commercially available, circular bearings
are more commonly used. Consequently the
procedure given below focuses on circular
bearings. The same steps can be followed for
rectangular bearings, but some modifications
will be necessary. When sizing the physical dimensions of the bearing, plan dimensions (B, dL) should be rounded
up to the next 1/4 increment, while the total thickness of elastomer, Tr, is specified in multiples of the layer
thickness. Typical layer thicknesses for bearings with lead cores are 1/4 and 3/8. High quality natural rubber
should be specified for the elastomer. It should have a shear modulus in the range 60-120 psi and an ultimate
elongation-at-break in excess of 5.5. Details can be found in rubber handbooks or in MCEER 2006.

The following design procedure assumes the isolators are bolted to the masonry and sole plates. Isolators that use
shear-only connections (and not bolts) require additional design checks for stability which are not included below.
See MCEER 2006.

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

E1. Required Properties E1. Required Properties, Example 1.1

Obtain from previous work the properties required of The design of one of the exterior isolators on a pier is
the isolation system to achieve the specified given below to illustrate the design process for lead-
performance criteria (Step A1). rubber isolators.
required characteristic strength, Qd / isolator
required post-elastic stiffness, Kd / isolator From previous work
total design displacement, dt, for each isolator Qd / isolator = 3.50 k
maximum applied dead and live load (PDL, Kd / isolator = 1.17 k/in
PLL) and seismic load (PSL) which includes Total design displacement, dt = 3.59 in
seismic live load (if any) and overturning PDL = 45.52 k
forces due to seismic loads, at each isolator, PLL = 15.50 k
and PSL = 15.56 k
maximum wind load, PWL PWL = 1.76 k < Qd OK

E2. Isolator Sizing
E2.1 Lead Core Diameter E2.1 Lead Core Diameter, Example 1.1
Determine the required diameter of the lead plug, dL,
using:
3.50
(E-1) 1.97
0.9 0.9 0.9
See Step E2.5 for limitations on dL

53

E2.2 Plan Area and Isolator Diameter E2.2 Plan Area and Isolator Diameter, Example
Although no limits are placed on compressive stress 1.1
in the GSID, (maximum strain criteria are used
instead, see Step E3) it is useful to begin the sizing
process by assuming an allowable stress of, say, 1.0
ksi.

Then the bonded area of the isolator is given by:

(E-2)
1.0 45.52 15.50
61.02
1.0 1.0
and the corresponding bonded diameter (taking into
account the hole required to accommodate the lead
core) is given by:
4
(E-3)
4 4 61.02
1.97

Round the bonded diameter, B, to nearest quarter = 9.03 in

inch, and recalculate actual bonded area using
Round B up to 9.25 in and the actual bonded area is:
(E-4)
4
9.25 1.97 64.15
4
Note that the overall diameter is equal to the bonded
diameter plus the thickness of the side cover layers
(usually 1/2 inch each side). In this case the overall
diameter, Bo is given by:

1.0 (E-5)
Bo = 9.25 + 2(0.5) = 10.25 in

E2.3 Elastomer Thickness and Number of Layers E2.3 Elastomer Thickness and Number of Layers,
Since the shear stiffness of the elastomeric bearing is Example 1.1
given by:
(E-6)
where G = shear modulus of the rubber, and
Tr = the total thickness of elastomer, Select G, shear modulus of rubber, = 100 psi (0.1 ksi)
it follows Eq. E-5 may be used to obtain Tr given a
required value for Kd Then
0.1 64.15
(E-7) 5.49
1.17
A typical range for shear modulus, G, is 60-120 psi.
Higher and lower values are available and are used in
special applications.

If the layer thickness is tr, the number of layers, n, is

given by:
5.49
(E-8) 21.97
0.25
rounded up to the nearest integer.
Round to nearest integer, i.e. n = 22

54

Note that because of rounding the plan dimensions

and the number of layers, the actual stiffness, Kd, will
not be exactly as required. Reanalysis may be
necessary if the differences are large.

E2.4 Overall Height E2.4 Overall Height, Example 1.1

The overall height of the isolator, H, is given by:

1 2 ( E-9) 22 0.25 21 0.125 2 1.5 11.125

where ts = thickness of an internal shim (usually

about 1/8 in), and
tc = combined thickness of end cover plate
(0.5 in) and outer plate (1.0 in)

E2.5 Size Checks E2.5 Size Checks, Example 1.1

Experience has shown that for optimum performance Since B=9.0 check
of the lead core it must not be too small or too large.
The recommended range for the diameter is as 9.25 9.25
follows: 3 6
(E-10)
3 6 i.e., 3.08 1.54

Since dL = 1.97, lead core size is acceptable.

Art. 12.2 GSID requires that the isolation system 0.025 0.025 45.52
provides a lateral restoring force at dt greater than the , 0.29 /
3.96
restoring force at 0.5dt by not less than W/80. This
equates to a minimum Kd of 0.025W/d. As
0.1 64.15
0.025 1.17 / ,
5.5
,

E3. Strain Limit Check E3. Strain Limit Check, Example 1.1
Art. 14.2 and 14.3 GSID requires that the total
applied shear strain from all sources in a single layer
of elastomer should not exceed 5.5, i.e.,
Since
, + 0.5 5.5 (E-11)
45.52
0.710
where , , , are defined below. 64.15
(a) is the maximum shear strain in the layer due to
compression and is given by: G = 0.1 ksi
(E-12) and
where Dc is shape coefficient for compression in 64.15
8.83
circular bearings = 1.0, , G is shear 9.25 0.25
modulus, and S is the layer shape factor given by: then
(E-13) 1.0 0.710
0.804
0.1 8.83
(b) , is the shear strain due to earthquake loads
and is given by:

55

 , (E-14) 3.59
, 0.653
5.5
(c) is the shear strain due to rotation and is given
by:
(E-15) 0.375 9.25 0.01
0.233
where Dr is shape coefficient for rotation in circular 0.25 5.5
bearings = 0.375, and is design rotation due to DL,
LL and construction effects Actual value for may Substitution in Eq E-11 gives
not be known at this time and value of 0.01 is
suggested as an interim measure, including , 0.5 0.804 0.653 0.5 0.233
uncertainties (see LRFD Art. 14.4.2.1) 1.57
5.5

E4. Vertical Load Stability Check E4. Vertical Load Stability Check, Example 1.1
Art 12.3 GSID requires the vertical load capacity of
all isolators be at least 3 times the applied vertical
loads (DL and LL) in the laterally undeformed state.

Further, the isolation system shall be stable under

1.2(DL+SL) at a horizontal displacement equal to
either
2 x total design displacement, dt, if in Seismic Zone 1
or 2, or
1.5 x total design displacement, dt, if in Seismic Zone
3 or 4.

E4.1 Vertical Load Stability in Undeformed State E4.1 Vertical Load Stability in Undeformed State,
The critical load capacity of an elastomeric isolator at Example 1.1
zero shear displacement is given by

4
1 1 3 3 0.1 0.3
2 (E-16)
0.3 1 0.67 8.83 15.97
where
9.25
359.37
Ts = total shim thickness 64
15.97 359.37
1043.68 /
1 0.67 5.5
E = elastic modulus of elastomer = 3G
0.1 64.15
64 1.17 /
5.5
It is noted that typical elastomeric isolators have high
shape factors, S, in which case:
4
1 (E-17)

and Eq. E-16 reduces to:

(E-18) 1.17 1043.68 109.61

56

Check that:
109.61
3 (E-19) 1.80 3
45.52 15.5

E4.2 Vertical Load Stability in Deformed State E4.2 Vertical Load Stability in Deformed State,
The critical load capacity of an elastomeric isolator at Example 1.1
shear displacement may be approximated by: Since bridge is in Zone 3, 1.5 1.5 3.59 5.38

(E-20) 5.38
2 1.90
where 9.25
Ar = overlap area between top and bottom plates
of isolator at displacement (Fig. 2.2-1
GSID)
1.90 1.90
= 0.303
4
2

Agross = 4 0.303 109.61 33.24

It follows that:
(E-21)

Check that:
33.24
1 (E-22) 0.47 1
1.2 1.2 1.2 45.52 15.56

E5. Design Review E5. Design Review, Example 1.1

The basic dimensions of the isolator designed above are
as follows:

10.25 in (od) x 11.125 in (high) x 1.97 in dia. lead core

and the volume, excluding steel end and cover plates, is

670 in3.

This design satisfies the shear strain limit criteria, but

not the vertical load stability ratio in the undeformed
and deformed states.

A redesign is therefore required and the easiest way to

increase the Pcr is to increase the shape factor, S, since
the bending stiffness of an isolator is a function of the
shape factor squared. See equations in Step E4.1. To
increase S, increase the bonded area Ab while keeping tr
constant (Eq. E-13). But to keep Kd constant while
increasing Ab and Tr is constant, decrease the shear
modulus, G (Eq. E-6).

This redesign is outlined below. After repeating the

calculation for diameter of lead core, the process begins

57

 by reducing the shear modulus to 60 psi (0.06 ksi) and

increasing the bonded diameter to 12.25 in.

E2.1
3.50
1.97
0.9 0.9

E2.2

12.25 1.97 114.81

4

Bo = 12.25 + 2(0.5) = 13.25 in

E2.3
0.06 114.81
5.90
1.17

5.90
23.60
0.25

Round to nearest integer, i.e. n = 24.

E2.4
24 0.25 23 0.125 2 1.5 11.875
E2.5
Since B=12 check

12.25 12.25
3 6

i.e., 4.08 2.04

Since dL = 1.97, the size of lead core is too small, and

there are 2 options: (1) Accept the undersize and check
for adequate performance during the Quality Control
Tests required by GSID Art. 15.2.2; or (2) Only have
lead cores in every second isolator, in which case the
core diameter, in those isolators with cores, will be 2 x
1.97 = 2.79 in (which satisfies above criterion).

0.06 114.81
1.15 / ,
6.0

E3.
45.52
0.396
114.81

114.81
11.93
12.25 0.25

1.0 0.396
0.554
0.06 11.93

58

 3.59
, 0.598
6.00

0.375 12.25 0.01

0.375
0.25 6

, 0.5 0.554 0.598 0.5 0.375

1.34 5.5
E4.1
3 3 0.06 0.18

0.18 1 0.67 11.93 17.35

12.25
1105.39
64
17.35 1105.39
3197.07 /
6

0.06 114.81
1.148 /
6

1.148 3197.07 190.33

190.33
3.12 3
45.52 15.50

E4.2
5.38
2 2.23
12.25

2.23 2.23
0.459

0.459 190.33 87.37

87.37
1.24 1
1.2 1.2 45.52 15.56

E5.
The basic dimensions of the redesigned isolator are as
follows:

13.25 in (od) x 11.875 in (high) x 1.97 in dia. lead core

and its volume (excluding steel end and cover plates) is
1224 in3.

This design meets all the design criteria but is about 80%
larger by volume than the previous design. This increase
in size is driven by the need to satisfy the vertical load
stability ratio of 3.0 in the undeformed state.

59

E6. Minimum and Maximum Performance Check E6. Minimum and Maximum Performance Check,
Art. 8 GSID requires the performance of any Example 1.1
isolation system be checked using minimum and Minimum Property Modification factors are:
maximum values for the effective stiffness of the min,Kd = 1.0
system. These values are calculated from minimum min,Qd = 1.0
and maximum values of Kd and Qd, which are found
using system property modification factors, as which means there is no need to reanalyze the bridge
indicated in Table E6-1. with a set of minimum values.
Table E6-1. Minimum and maximum values
for Kd and Qd. Maximum Property Modification factors are:
max,a,Kd = 1.1
Eq. max,a,Qd = 1.1
8.1.2-1 Kd,max = Kd max,Kd (E-23)
GSID max,t,Kd = 1.1
Eq. max,t,Qd = 1.4
8.1.2-2 Kd,min = Kd min,Kd (E-24)
GSID max,scrag,Kd = 1.0
Eq. max,scrag,Qd = 1.0
8.1.2-3 Qd,max = Qd max,Qd (E-25) Applying a system adjustment factor of 0.66 for an
GSID other bridge, the maximum property modification
Eq. factors become:
8.1.2-4 Qd,min = Qd min,Qd (E-26)
GSID max,a,Kd = 1.0 + 0.1(0.66) = 1.066
max,a,Qd = 1.0 + 0.1(0.66) = 1.066
Determination of the system property modification
factors should include consideration of the effects of max,t,Kd = 1.0 + 0.1(0.66) = 1.066
temperature, aging, scragging, velocity, travel (wear) max,t,Qd = 1.0 + 0.4(0.66) = 1.264
and contamination as shown in Table E6-2. In lieu of
tests, numerical values for these factors can be max,scrag,Kd = 1.0
obtained from Appendix A, GSID. max,scrag,Qd = 1.0

Table E6-2. Minimum and maximum values for Therefore the maximum overall modification factors
system property modification factors.
max,Kd = 1.066(1.066)1.0 = 1.14
Eq. min,Kd = (min,t,Kd) (min,a,Kd) max,Qd = 1.066(1.264)1.0 = 1.35
8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-27)
GSID (min,scrag,Kd) Since the possible variation in upper bound properties
Eq. max,Kd = (max,t,Kd) (max,a,Kd) exceeds 15% a reanalysis of the bridge is required to
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-28) determine performance with these properties.
GSID (max,scrag,Kd)
Eq. min,Qd = (min,t,Qd) (min,a,Qd) The upper-bound properties are:
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-29) Qd,max = 1.35 (3.50) = 4.73 k
GSID (min,scrag,Qd) and
Kd,ma x=1.14(1.15) = 1.31 k/in
Eq. max,Qd = (max,t,Qd) (max,a,Qd)
8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd) (E-30)
GSID (max,scrag,Qd)

Adjustment factors are applied to individual -factors

(except v) to account for the likelihood of
occurrence of all of the maxima (or all of the
minima) at the same time. These factors are applied
to all -factors that deviate from unity but only to the
portion of the -factor that is greater than, or less
than, unity. Art. 8.2.2 GSID gives these factors as
follows:

60

 1.00 for critical bridges

0.75 for essential bridges
0.66 for all other bridges

As required in Art. 7 GSID and shown in Fig. C7-1

GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Qd,max. As indicated in Fig C7-1 GSID, maximum
displacements will probably be given by the first case
(Kd,min and Qd,min) and maximum forces by the second
case (Kd,max and Qd,max).

E7. Design and Performance Summary E7. Design and Performance Summary, Example 1.1
E7.1 Isolator dimensions E7.1 Isolator dimensions, Example 1.1
Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E7.1-1.
Overall diameter (includes cover layer)
Overall height Table E7.1-1 Isolator Dimensions
Diameter lead core
Isolator Overall size Overall size Diam.
Bonded diameter Location including without lead
Number of rubber layers mounting mounting core
Thickness of rubber layers plates (in) plates (in) (in)
Total rubber thickness Under
Thickness of steel shims edge 17.25 x 17.25 13.25 dia. x
Shear modulus of elastomer 1.97
girder x 11.875 (H) 10.375(H)
on Pier 1
Check all dimensions with manufacturer.

Isolator No. of Rubber Total Steel

Location rubber layers rubber shim
layers thick- thick- thick-
ness ness ness
(in) (in) (in)
Under
edge
24 0.25 6 0.125
girder
on Pier 1

Shear modulus of elastomer = 60 psi

E7.2 Bridge Performance E7.2 Bridge Performance, Example 1.1

Summarize bridge performance
Maximum superstructure displacement
(longitudinal)
Maximum superstructure displacement
(transverse)
Maximum superstructure displacement
(resultant)
Maximum column shear (resultant)
Maximum column moment (about transverse
axis)
Maximum column moment (about
longitudinal axis)
Maximum column torque
Bridge performance is summarized in Table E7.2-1
where it is seen that the maximum column shear is 25.55
k. This is slightly more than the column plastic shear
strength (25 k), but is sufficiently close as to allow the
column to remain essentially elastic. Furthermore the
maximum longitudinal displacement is 3.81 in which is
less than the 4.0 in available at the abutment expansion
joints and is therefore acceptable.
Table E7.2-1 Summary of Bridge Performance

61

Check required performance as determined in Step Maximum superstructure

3.81 in
A3, is satisfied. displacement (longitudinal)
Maximum superstructure
3.53 in
displacement (transverse)
Maximum superstructure
3.96 in
displacement (resultant)
Maximum column shear (resultant) 25.55 k
Maximum column moment about
362 kft
transverse axis
Maximum column moment about
172 kft
longitudinal axis
Maximum column torque 2.83 kft

62

 SECTION 1
PC Girder Bridge, Short Spans, Multi-Column Piers

DESIGN EXAMPLE 1.2: S1 = 0.6g

Design Examples in Section 1

Site
ID Description S1 Pier height Skew Isolator type
Class
1.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
1.1 Change site class 0.2g D Same 0 Lead rubber bearing
Change spectral
1.2 0.6g B Same 0 Lead rubber bearing
acceleration, S1
Spherical friction
1.3 Change isolator to SFB 0.2g B Same 0
bearing
1.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
1.5 Change column height 0.2g B H1=0.5 H2 0 Lead rubber bearing
1.6 Change angle of skew 0.2g B Same 450 Lead rubber bearing

63

DESIGN PROCEDURE
STEP A: BRIDGE AND SITE DATA
A1. Bridge Properties
Determine properties of the bridge:
number of supports, m
number of girders per support, n
angle of skew
weight of superstructure including railings,
curbs, barriers and to the permanent loads,
WSS
weight of piers participating with
superstructure in dynamic response, WPP
weight of superstructure, Wj, at each
support
stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge. The calculation of these
quantities requires careful consideration of
several factors such as the use of cracked
sections when estimating column or wall
flexural stiffness, foundation flexibility,
and effective column height.
column shear strength (minimum value).
This will usually be derived from the
minimum value of the column flexural
yield strength, the column height, and
whether the column is acting in single or
double curvature in the direction under
consideration.
allowable movement at expansion joints
isolator type if known, otherwise to be
determined
A2. Seismic Hazard
Determine seismic hazard at site:
acceleration coefficients
site class and site factors
seismic zone
Plot response spectrum.
Use Art. 3.1 GSID to obtain peak ground and spectral
acceleration coefficients. These coefficients are the
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (e.g. Figures
3.10.2.1-1 to 3.10.2.1-21 LRFD).
Use Art. 3.2 to obtain Site Class and corresponding Site
Factors (Fpga, Fa and Fv). These data are the same as for
DESIGN EXAMPLE 1.2 (S1 = 0.6g)
A1. Bridge Properties, Example 1.2
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 6
Number of columns per support = 3
Angle of skew = 00
Weight of superstructure including permanent
loads, WSS = 650.52 k
Weight of superstructure at each support:
o W1 = 44.95 k
o W2 = 280.31 k
o W3 = 280.31 k
o W4 = 44.95 k
Participating weight of piers, WPP = 107.16 k
Effective weight (for calculation of period),
Weff = Wss + WPP = 757.68 k
Stiffness of each pier in the longitudinal
direction:
o Ksub,pier1,long = 172.0 k/in
o Ksub,pier2,long = 172.0 k/in
Stiffness of each pier in the transverse direction:
o Ksub,pier1,trans = 687.0 k/in
o Ksub,pier2,trans = 687.0 k/in
Minimum column shear strength based on
flexural yield capacity of column = 25 k
Displacement capacity of expansion joints
(longitudinal) = 2.0 in for thermal and other
movements.
Lead rubber isolators
A2. Seismic Hazard, Example 1.2
Acceleration coefficients for bridge site are given in
design statement as follows:
PGA = 0.40
S1 = 0.60
SS = 0.75
Bridge is on a rock site with shear wave velocity in
upper 100 ft of soil = 3,000 ft/sec.
Table 3.10.3.1-1 LRFD gives Site Class as B.
Tables 3.10.3.2-1, -2 and -3 LRFD give following
Site Factors:
Fpga = 1.0
Fa = 1.0
Fv = 1.0
64

conventional bridges and Art 3.2 refers the designer to

the corresponding articles in the LRFD Specifications,
i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3,
LRFD.

Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give

modified spectral acceleration coefficients that include As = Fpga PGA = 1.0(0.40) = 0.40
site effects as follows: SDS = Fa SS = 1.0(0.75) = 0.75
As = Fpga PGA SD1 = Fv S1 = 1.0(0.60) = 0.60
SDS = Fa SS
SD1 = Fv S1

Seismic Zone is determined by value of SD1 in Since 0.50 < SD1, bridge is located in Seismic Zone 4.
accordance with provisions in Table5-1 GSID.

Design Response Spectrum is as below:

These coefficients are used to plot design response
spectrum as shown in Fig. 4-1 GSID. 0.9

0.8

0.7

Acceleration(g)
0.6

0.5

0.4

0.3

0.2

0.1

0.0
0 0.5 1 1.5 2 2.5 3
Period(s)

A3. Performance Requirements A3. Performance Requirements, Example 1.2

Determine required performance of isolated bridge In this example, assume the owner has specified full
during Design Earthquake (1000-yr return period). functionality following the earthquake and therefore
the columns must remain elastic (no yield).
Examples of performance that might be specified by
the Owner include: To remain elastic the maximum lateral load on the
Reduced displacement ductility demand in pier must be less than the load to yield any one
columns, so that bridge is open for emergency column (25 k).
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in The maximum shear in the column must therefore be
columns or yield elsewhere in bridge), so that less than 25 k in order to keep the column elastic and
bridge is fully functional and open to all vehicles meet the required performance criterion.
immediately following earthquake.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.

65

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for Isolator Force, F

non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most have
bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use Ku
equivalent linear springs and viscous damping to
represent the isolators, so that linear methods of analysis
disol
may be used to determine response. Since equivalent Ku
dy Isolator
Ku Displacement, d
properties such as Kisol are dependent on displacement (d),
and the displacements are not known at the beginning of
the analysis, an iterative approach is required. Note that in Kd
Art 7.1, GSID, keff is used for the effective stiffness of an
isolator unit and Keff is used for the effective stiffness of a disol = Isolator displacement
dy = Isolator yield displacement
combined isolator and substructure unit. To minimize Fisol = Isolator shear force
confusion, Kisol is used in this document in place of keff. Fy = Isolator yield force
Kd = Post-elastic stiffness of isolator
There is no change in the use of Keff and Keff,j, but Ksub is Kisol = Effective stiffness of isolator
used in place of ksub. Ku
Qd
= Loading and unloading stiffness (elastic stiffness)
= Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement (disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 1.2
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.60 6.0
GSID

(2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not
occur under non-seismic loads (e.g. wind)

66

 but low enough that yield will occur during
an earthquake. Experience has shown that
taking Qd to be 5% of the bridge weight is a
good starting point, i.e.
0.05 (B-2)
(3) Post-yield stiffness, Kd
Art 12.2 GSID requires that all isolators
exhibit a minimum lateral restoring force at
the design displacement, which translates to a
minimum post yield stiffness Kd,min given by
Eq. B-3.
Due to larger estimated displacements (Eq B-1) than
for the benchmark bridge, Qd is increased to 7.5% of
the bridge weight to introduce additional damping
and reduce these displacements as much as possible,
i.e., 0.075 0.075 650.52 49.79
Also, in view of these larger displacements, the post
yield stiffness is increased to 0.1W/d, to give
essentially the same value for Kd found to be
satisfactory in Example 1.0.

Art. 650.52
0.025 0.1 0.1 16.26 /
12.2 , (B-3) 4.0
GSID
Experience has shown that a good starting
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 1.2
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength, ,

Qd, and stiffness, Kd, values in proportion to the dead o Qd, 1 = 3.37 k
load applied at that support: o Qd, 2 = 21.02 k
o Qd, 3 = 21.02 k
o Qd, 4 = 3.37 k

, (B-4)
and
and
,
, (B-5) o Kd,1 = 1.12 k/in
o Kd,2 = 7.01 k/in
o Kd,3 = 7.01 k/in
o Kd,4 = 1.12 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 1.2
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of , ,
the isolators at support j (Kisol,j) and the stiffness of , ,
the substructure Ksub,j. See figure below for
definitions (after Fig. 7.1-1 GSID). o 1 = 1.69x10-4
o 2 = 6.24x10-2
An expression for Keff,j, is given in Eq.7.1-6 GSID, but o 3 = 6.24x10-2
a more useful formula as follows (MCEER 2006): o 4 = 1.69x10-4
,
, (B-6)
1 ,
where ,
1
, ,
(B-7)
, , o Keff,1 = 1.69 k/in
and Ksub,j for the piers are given in Step A1. For the o Keff,2 = 10.10 k/in
abutments, take Ksub,j to be a large number, say 10,000 o Keff,3 = 10.10 k/in
k/in, unless actual stiffness values are available. Note o Keff,4 = 1.69 k/in
that if the default option is chosen, unrealistically high

67

 values for Ksub,j will give unconservative results for

column moments and shear forces.

F
Kd
Qd Kisol

dy disol
Superstructure
Isolator Effective Stiffness, Kisol

F
Isolator(s), Kisol
Ksub

Substructure, Ksub
dsub

Substructure Stiffness, Ksub

dsub disol F

d
Keff

d = disol + dsub

Combined Effective Stiffness, Keff

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 1.2
Calculate the total effective stiffness, Keff, of the
bridge:

Eq. , 23.57 /
7.1-6 , ( B-8)
GSID

B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 1.2
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1

(B-9) o disol,1 = 6.00 in

,
1 o disol,2 = 5.65 in
o disol,3 = 5.65 in
o disol,4 = 6.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the effective stiffness of the isolation system Example 1.2
at support j, Kisol,j, for all supports:
,
, ,
,
,
(B-10) o Kisol,1 = 1.69 k/in
, ,
, o Kisol,2 = 10.73 k/in
o Kisol,3 = 10.73 k/in
o Kisol,4 = 1.69 k/in

68

B1.7 Substructure Displacement at Each Support
B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j,
Example 1.2
for all supports:
, ,

, , (B-11)
o dsub,1 = 1.01x10-3 in
o d sub,2 = 3.52x10-1 in
o d sub,3 = 3.52x10-1 in
o d sub,4 = 1.01x10-3 in
B1.8 Lateral Load in Each Substructure Support B1.8 Lateral Load in Each Substructure, Example
Calculate the shear at support j, Fsub,j, for all 1.2
supports:
, , , (B-12) , , ,

where values for Ksub,j are given in Step A1. o F sub,1 = 10.11 k
o F sub,2 = 60.59 k
o F sub,3 = 60.59 k
o F sub,4 = 10.11 k

B1.9 Column Shear Force at Each Support B1.9 Column Shear Force at Each Support,
Calculate the shear in column k at support j, Fcol,j,k, Example 1.2
assuming equal distribution of shear for all columns at
support j:
,
, ,
#
,
, ,
# (B-13) o F col,2,1-3 20.20 k
o F col,3,1-3 20.20 k

Use these approximate column shear forces as a check These column shears are less than the yield capacity
on the validity of the chosen strength and stiffness of each column (25 k) as required in Step A3 and the
characteristics. chosen strength and stiffness values in Step B1.1 are
therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 1.2
damping ratio, , of the bridge:

Eq. 757.68
7.1-5 2 (B-14) 2 2
GSID 386.4 23.57

and = 1.81 sec

Eq. 2 , , , and taking dy,j = 0:

(B-15)
7.1-10 , , ,
GSID 2 , , 0
0.21
, , ,
where dy,j is the yield displacement of the isolator. For
friction-based isolators, dy,j = 0. For other types of
isolators dy,j is usually small compared to disol,j and has
negligible effect on , Hence it is suggested that for
the Simplified Method, set dy,j = 0 for all isolator

69

types. See Step B2.2 where the value of dy,j is revisited
for the Multimode Spectral Analysis Method.
B1.11 Damping Factor
Calculate the damping factor, BL, and the
displacement, d, of the bridge:
Eq. .
, 0.3
.
7.1-3 (B-16)
1.7, 0.3
GSID
Eq. 9.79 (B-17)
7.1-4
GSID
B1.12 Convergence Check
Compare the new displacement with the initial value
assumed in Step B1.1. If there is close agreement, go
to the next step; otherwise repeat the process from
Step B1.3 with the new value for displacement as the
assumed displacement.
This iterative process is amenable to solution using a
spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
B1.11 Damping Factor, Example 1.2
Since 0.21 0.3
1.53
and
9.79 9.79 0.6 1.81
6.94
1.53
B1.12 Convergence Check, Example 1.2
Since the calculated value for displacement, d
(=6.94 in) is not close to that assumed at the
beginning of the cycle (Step B1.1, d = 6.0), use the
value of 6.94 in as the new assumed displacement and
repeat from Step B1.3.
After three iterations, convergence is reached at a
superstructure displacement of 7.44 in, with an
effective period of 1.87 seconds, and a damping factor
of 1.47 (18% damping ratio). The displacement in the
isolators at Pier 1 is 7.03 in and the effective stiffness
of the same isolators is 10.00 k/in.
See spreadsheet in Table B1.12-1 for results of final
iteration.
Ignoring the weight of the cap beams, the sum of the
column shears at a pier must equal the total isolator
shear force. Hence, approximate column shear (per
column) = 10(7.03)/3 = 23.43 k which is less than the
maximum allowable (25k) if elastic behavior is to be
achieved (as required in Step A3).
The superstructure displacement = 7.44 in, which is
much greater than the available clearance of 2.0 in.
There are two options; (1) increase the available
clearance to allow for this displacement, or (2) accept
that abutment pounding is likely to occur. However,
the minimum required clearance is given by:
8 8 0.60 1.87
6.11
1.47
Therefore, the available clearance needs to be
increased to above this minimum value, say 8.0 in.
70

 Table B1.12-1 Simplified Method Solution for Design Example 1.2 Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
650.52 107.16 757.68 0.6 6

StepB1.1 d 7.44 Assumeddisplacement

Qd 48.79 Characteristicstrength
Kd 16.26 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 44.95 3.37 1.12 10000 1.58E04 1.58 7.44 1.58 1.17E03 11.73 25.085 87.305
Pier1 280.31 21.02 7.01 172.0 5.81E02 9.45 7.03 10.00 4.09E01 70.30 147.869 523.218
Pier2 280.31 21.02 7.01 172.0 5.81E02 9.45 7.03 10.00 4.09E01 70.30 147.869 523.218
Abut2 44.95 3.37 1.12 10000 1.58E04 1.58 7.44 1.58 1.17E03 11.73 25.085 87.305
Total 650.52 48.789 16.260 K eff,j 22.046 164.070 345.907 1,221.047
Step B1.4

StepB1.10 T eff 1.87 Effectiveperiod

0.18 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.47

B L 1.47 DampingFactor
d 7.49 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 0.562 0.187 0.263 7.05 0.267
Pier1 3.504 1.168 1.666 6.45 1.711
Pier2 3.504 1.168 1.666 6.45 1.711
Abut2 0.562 0.187 0.263 7.05 0.267

71

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and
then applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 1.2

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 (see
elastic stiffness, Kd,i, of each isolator i as follows: Table B1.12-1) by the number of isolators at each
, support (n = 6), the following values for Qd /isolator
, (B-19)
and Kd /isolator are obtained:
and
, o Qd, 1 = 3.37/6 = 0.56 k
, ( B-20)
o Qd, 2 = 21.02/6= 3.50 k
where values for Qd,j and Kd,j are obtained from the o Qd, 3 = 21.02/6= 3.50 k
final cycle of iteration in the Simplified Method (Step o Qd, 4 = 3.37/6 = 0.56 k
B1. 12 and
o Kd,1 = 1.12/6 = 0.19 k/in
o Kd,2 = 7.01/6 = 1.17 k/in
o Kd,3 = 7.01/6 = 1.17 k/in
o Kd,4 = 1.12/6 = 0.19 k/in

B2.2 Initial Stiffness and Yield Displacement B2.2 Initial Stiffness and Yield Displacement,
Calculate the initial stiffness, Ku,i, and the yield Example 1.2
displacement, dy,i, for each isolator i as follows: Since the isolator type has been specified in Step A1
(1) For friction-based isolators: Ku,i = and dy,i = 0. to be an elastomeric bearing (i.e. not a friction-based
(2) For other types of isolators, and in the absence of bearing), calculate Ku,i and dy,i for an isolator on Pier 1
isolator-specific information, take as follows:
, 10 , 10 1.17 11.7 /
, 10 , ( B-21)
and then and
, 3.50
,
( B-22) , 0.33
,
, , 11.7 1.17
, ,

As expected, the yield displacement is small

compared to the expected isolator displacement (~2
in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.

B2.3 Isolator Effective Stiffness, Kisol,i B2.3 Isolator Effective Stiffness, Kisol,i, Example 1.2
Calculate the isolator stiffness, Kisol,i, of each isolator Dividing the results for Kisol (Step B1.12) among the 6
i: isolators at each support, the following values for Kisol
/isolator are obtained:
, o Kisol,1 = 1.58/6 = 0.26 k/in
, (B -23)
o Kisol,2 = 10.00/6 = 1.67 k/in
o Kisol,3 = 10.00/6 = 1.67 k/in

72

 o Kisol,4 = 1.58/6 = 0.26 k/in

B2.4 Three-Dimensional Bridge Model B2.4 Three-Dimensional Bridge Model, Example

Using computer-based structural analysis software, 1.2
create a 3-dimensional model of the bridge with the Although the bridge in this Design Example is regular
isolators represented by spring elements. The and is without skew or curvature, a 3-dimensional
stiffness of each isolator element in the horizontal finite element model was developed for this Step, as
axes (Kx and Ky in global coordinates, K2 and K3 in shown below.
typical local coordinates) is the Kisol value calculated
in the previous step. For bridges with regular
geometry and minimal skew or curvature, the
superstructure may be represented by a single stick
provided the load path to each individual isolator at
each support is explicitly modeled, usually by a rigid
cap beam and a set of rigid links. If the geometry is
irregular, or if the bridge is skewed or curved, a finite
element model is recommended to accurately capture
the load carried by each individual isolator. If the
piers have an unusual weight distribution, such as a .
pier with a hammerhead cap beam, a more rigorous
model is justified.

B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in
Step A2 is for a 5% damped system. It is modified in
this step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the
effective period of the bridge, Teff, by the damping
factor, BL.
B.2.5 Composite Design Response Spectrum,
Example 1.2
From the final results of Simplified Method (Step
B1.12), BL = 1.47 and Teff = 1.87 sec. Hence the
transition in the composite spectrum from 5% to 18%
damping occurs at 0.8 Teff = 0.8 (1.87) = 1.50 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.50 sec by 1.47.
0.9

0.8

0.7
Acceleration(g)

0.6

0.5

0.4

0.3

0.2

0.1

0.0
0 0.5 1 1.5 2 2.5 3
Period(s)

B2.6 Multimodal Analysis of Finite Element
Model
Input the composite response spectrum as a user-
specified spectrum in the software, and define a load
case in which the spectrum is applied in the
longitudinal direction. Analyze the bridge for this
load case.
B2.6 Multimodal Analysis of Finite Element
Model, Example 1.2
Results of modal analysis of the example bridge are
summarized in Table B2.6-1 Here the modal periods
and mass participation factors of the first 12 modes
are given. The first two modes are the principal
longitudinal and transverse modes with periods of
1.73 and 1.68 sec respectively.

73

 Table B2.6-1 Modal Properties of Bridge

Example 1.2 First Iteration
Mode Period ModalParticipatingMassRatios
No. Sec UX UY UZ RX RY RZ
1 1.727 0.751 0.000 0.000 0.000 0.002 0.000
2 1.675 0.000 0.736 0.000 0.031 0.000 0.528
3 1.644 0.000 0.000 0.000 0.000 0.000 0.210
4 0.191 0.000 0.000 0.000 0.000 0.000 0.000
5 0.191 0.135 0.000 0.000 0.000 0.000 0.000
6 0.122 0.000 0.000 0.000 0.000 0.000 0.000
7 0.122 0.000 0.000 0.000 0.000 0.000 0.007
8 0.104 0.000 0.038 0.000 0.178 0.000 0.027
9 0.101 0.000 0.000 0.237 0.000 0.184 0.000
10 0.095 0.000 0.120 0.000 0.086 0.000 0.086
11 0.094 0.000 0.000 0.014 0.000 0.011 0.000
12 0.074 0.000 0.002 0.000 0.000 0.000 0.001

Computed values for the isolator displacements due to

a longitudinal earthquake are as follows (numbers in
parentheses are those used to calculate the initial
properties to start iteration from the Simplified
Method):

o disol,1 = 6.99 (7.44) in

o disol,2 = 6.41 (7.03) in
o disol,3 = 6.41 (7.03) in
o disol,4 = 6.99 (7.44) in

B2.7 Convergence Check B2.7 Convergence Check, Example 1.2

Compare the resulting displacements at the
superstructure level (d) to the assumed
displacements. These displacements can be obtained
by examining the joints at the top of the isolator
spring elements. If in close agreement, go to Step
B2.9. Otherwise go to Step B2.8.
The new superstructure displacement is 6.99 in, more
than a 5% difference from the displacement assumed
at the start of the Multimode Spectral Analysis
(7.44 in).

B2.8 Update Kisol,i, Keff,j, and BL B2.8 Update Kisol,i, Keff,j, and BL, Example 1.2
Use the calculated displacements in each isolator Updated values for Kisol,i are given below (previous
element to obtain new values of Kisol,i for each values are in parentheses):
isolator as follows:
, o Kisol,1 = 0.27 (0.26) k/in
, , (B-24) o Kisol,2 = 1.72 (1.67) k/in
,
o Kisol,3 = 1.72 (1.67) k/in
Recalculate Keff,j : o Kisol,4 = 0.27 (0.26) k/in

Eq. , ,
Updated values for Keff,j, , BL and Teff are given below
7.1-6 ,
(B-25) (previous values are in parentheses):
, ,
GSID
o Keff,1 = 1.61 (1.58) k/in
Recalculate system damping ratio, : o Keff,2 = 10.01 (9.45) k/in
o Keff,3 = 10.01 (9.45) k/in
Eq. 2 , , ,
o Keff,4 = 1.61 (1.58) k/in
7.1-10 , , ,
(B-26)
GSID o = 17% (18%)
o BL = 1.44 (1.47)
Recalculate system damping factor, BL: o Teff = 1.73 (1.87) sec

. The updated composite response spectrum is shown

Eq. 0.3
. ( B-27) below:
7.1-3 1.7 0.3

74

 GSID 0.9

0.8

Obtain the effective period of the bridge from the 0.7

Acceleration(g)
multi-modal analysis and with the revised damping 0.6

factor (Eq. B-27), construct a new composite 0.5

response spectrum. Go to Step B2.6. 0.4

0.3

0.2

0.1

0.0
0 0.5 1 1.5 2 2.5 3
Period(s)

B2.6 Multimodal Analysis Second Iteration,

Example 1.2
Reanalysis gives the following values for the isolator
displacements (numbers in parentheses are from the
previous cycle):

o disol,1 = 7.05 (6.99) in

B2.7 Convergence Check
Compare the resulting displacements at the
superstructure level (d) to the assumed
displacements. These displacements can be obtained
by examining the joints at the top of the isolator
spring elements. If in close agreement, go to Step
B2.9. Otherwise go to Step B2.8.
o disol,2 = 6.45 (6.41) in
o disol,3 = 6.45 (6.41) in
o disol,4 = 7.05 (6.99) in
B2.7 Convergence Check, Example 1.2
The new superstructure displacement is 7.05 in, less
than a 1% difference from the displacement assumed
at the start of the second cycle of Multimode Spectral
Analysis.

B2.9 Superstructure and Isolator Displacements
Once convergence has been reached, obtain
o superstructure displacements in the
longitudinal (xL) and transverse (yL) directions
of the bridge, and
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions of the bridge, for
each isolator, for this load case (i.e.
longitudinal loading). These displacements
may be found by subtracting the nodal
displacements at each end of each isolator
spring element.
B2.9 Superstructure and Isolator Displacements,
Example 1.2
From the above analysis:
o superstructure displacements in the
longitudinal (xL) and transverse (yL) directions
are:
xL= 7.05 in
yL= 0.0 in
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions are:
o Abutments: uL = 7.05 in, vL = 0.00 in
o Piers: uL = 6.45 in, vL = 0.00 in

All isolators at same support have the same

displacements.

B2.10 Pier Bending Moments and Shear Forces B2.10 Pier Bending Moments and Shear Forces,
Obtain the pier bending moments and shear forces in Example 1.2
the longitudinal (MPLL, VPLL) and transverse (MPTL, Maximum bending moments and shear forces in the
VPTL) directions at the critical locations for the columns in the longitudinal (MPLL,VPLL) and
longitudinally-applied seismic loading. transverse (MPTL,VPTL) directions are:
Exterior Columns:
o MPLL= 0 kft
o MPTL= 443 kft

75

 o VPLL= 27.03 k
o VPTL= 0.02 k
Interior Columns:
o MPLL= 0 kft
o MPTL= 436 kft
o VPLL= 26.05 k
o VPTL= 0.00 k

Both piers have the same distribution of bending

moments and shear forces among the columns.

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 1.2
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1

Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces
ator
ure to long. to long. due to
EQ EQ long. EQ
1 1.89 0.00 0.85
2 1.89 0.00 0.84

Abut 3 1.89 0.00 0.84

ment 4 1.89 0.00 0.84
5 1.89 0.00 0.84
6 1.89 0.00 0.85
1 11.05 0.00 0.27
2 11.06 0.00 0.30
3 11.06 0.00 0.28
Pier
4 11.06 0.00 0.28
5 11.06 0.00 0.30
6 11.05 0.00 0.27

76

 STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake C1. Analysis for Transverse Earthquake,
Repeat the above process, starting at Step B1, for Example 1.2
earthquake loading in the transverse direction of the Key results from repeating Steps B1 and B2
bridge. Support flexibility in the transverse direction (Simplified and Mulitmode Spectral Methods) are:
is to be included, and a composite response spectrum o Teff = 1.66 sec
is to be applied in the transverse direction. Obtain o Superstructure displacements in the longitudinal
isolator displacements in the longitudinal (uT) and (xT) and transverse (yT) directions are as follows:
transverse (vT) directions of the bridge, and the xT = 0 and yT = 6.54 in
biaxial bending moments and shear forces at critical o Isolator displacements in the longitudinal (uT)
locations in the columns due to the transversely- and transverse (vT) directions are as follows:
applied seismic loading. Abutments uT = 0.00 in, vT = 6.54 in
Piers uT = 0.00 in, vT = 6.42 in
o Maximum bending moments and shear forces in
the columns in the longitudinal (MPLL,VPLL) and
transverse (MPTL,VPTL) directions are:
Exterior Columns:
o MPLT= 173 kft
o MPTT= 0 kft
o VPLT= 0.03 k
o VPTT= 22.99 k
Interior Columns:
o MPLT= 198 kft
o MPTT= 0 kft
o VPLT= 0.00 k
o VPTT= 28.09 k

Both piers have the same distribution of

bending moments and shear forces among the
columns.

o Isolator shear and axial forces are in Table C1-1.

77

 Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces due
ator
ure to transv. to transv. to transv.
EQ EQ EQ
1 0.00 1.79 3.48
2 0.00 1.79 1.92
Abut 3 0.00 1.79 0.61
ment 4 0.00 1.79 0.61
5 0.00 1.79 1.92
6 0.00 1.79 3.48
1 0.01 11.00 11.52
2 0.01 11.03 2.16
3 0.00 11.04 2.31
Pier
4 0.00 11.04 2.31
5 0.01 11.03 2.16
6 0.01 11.00 11.52

78

STEP D. CALCULATE DESIGN VALUES

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements at Pier 1,

Following the provisions in Art. 2.1 GSID, and Example 1.2
illustrated in Fig. 2.1-1 GSID, calculate the total To illustrate the process, design displacements for the
design displacement, dt, for each isolator by outside isolator on Pier 1 are calculated below.
combining the displacements from the longitudinal
(uL and vL) and transverse (uT and vT) cases as Load Case 1:
follows: u1 = uL + 0.3uT = 1.0(6.45) + 0.3(0) = 6.45 in
u1 = uL + 0.3uT (D-1) v1 = vL + 0.3vT = 1.0(0) + 0.3(6.42) = 1.93 in
v1 = vL + 0.3vT (D-2) R1 = = 6.45 1.93 = 6.73 in
R1 = (D-3)
Load Case 2:
u2 = 0.3uL + uT (D-4) u2 = 0.3uL + uT = 0.3(6.45) + 1.0(0) = 1.94 in
v2 = 0.3vL + vT (D-5) v2 = 0.3vL + vT = 0.3(0) + 1.0(6.42) = 6.42 in
R2 = (D-6) R2 = = 1.94 6.42 = 6.71 in

dt = max(R1, R2) (D-7) Governing Case:

D2. Design Moments and Shears
Calculate design values for column bending moments
and shear forces for all piers using the same
combination rules as for displacements.
Total design displacement, dt = max(R1, R2)
= 6.71in
D2. Design Moments and Shears in Pier 1,
Example 1.2
Design moments and shear forces are calculated for
Pier 1, Column 1, below to illustrate the process.

Alternatively this step may be deferred because the Load Case 1:

above results may not be final. Upper and lower VPL1= VPLL + 0.3VPLT = 1.0(27.03) + 0.3(0.03)
bound analyses are required after the isolators have = 27.04 k
been designed as described in Art 7. GSID. These VPT1= VPTL + 0.3VPTT = 1.0(0.02) + 0.3(22.99)
analyses are required to determine the effect of = 6.92 k
possible variations in isolator properties due age, R1 = = 27.04 6.92 = 27.91 k
temperature and scragging in elastomeric systems.
Accordingly the results for column shear in Steps Load Case 2:
B2.10 and C are likely to increase once these VPL2= 0.3VPLL + VPLT = 0.3(27.03) + 1.0(0.03)
analyses are complete. = 8.14 k
VPT2= 0.3VPTL + VPTT = 0.3(0.02) + 1.0(22.99)
= 23.00 k
R2 = = 8.14 23.00 = 24.40 k

Governing Case:
Design column shear = max (R1, R2)
= 27.91 k

79

STEP E. DESIGN OF LEAD-RUBBER (ELASTOMERIC) ISOLATORS

A lead-rubber isolator is an elastomeric
bearing with a lead core inserted on its vertical
centreline. When the bearing and lead core are
deformed in shear, the elastic stiffness of the
lead provides the initial stiffness (Ku).With
increasing lateral load the lead yields almost
perfectly plastically, and the post-yield
stiffness Kd is given by the rubber alone. More
details are given in MCEER 2006.

While both circular and rectangular bearings

are commercially available, circular bearings
are more commonly used. Consequently the
procedure given below focuses on circular
bearings. The same steps can be followed for
rectangular bearings, but some modifications
will be necessary. When sizing the physical dimensions of the bearing, plan dimensions (B, dL) should be rounded
up to the next 1/4 increment, while the total thickness of elastomer, Tr, is specified in multiples of the layer
thickness. Typical layer thicknesses for bearings with lead cores are 1/4 and 3/8. High quality natural rubber
should be specified for the elastomer. It should have a shear modulus in the range 60-120 psi and an ultimate
elongation-at-break in excess of 5.5. Details can be found in rubber handbooks or in MCEER 2006.

The following design procedure assumes the isolators are bolted to the masonry and sole plates. Isolators that use
shear-only connections (and not bolts) require additional design checks for stability which are not included below.
See MCEER 2006.

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

E1. Required Properties E1. Required Properties, Example 1.2

Obtain from previous work the properties required of The design of one of the exterior isolators on a pier is
the isolation system to achieve the specified given below to illustrate the design process for lead-
performance criteria (Step A1). rubber isolators.
required characteristic strength, Qd / isolator
required post-elastic stiffness, Kd / isolator From previous work
total design displacement, dt, for each isolator Qd / isolator = 3.50 k
maximum applied dead and live load (PDL, Kd / isolator = 1.17 k/in
PLL) and seismic load (PSL) which includes Total design displacement, dt = 6.71 in
seismic live load (if any) and overturning PDL = 45.52 k
forces due to seismic loads, at each isolator, PLL = 15.50 k
and PSL = 11.52 k
maximum wind load, PWL PWL = 1.76 k < Qd OK

E2. Isolator Sizing
E2.1 Lead Core Diameter E2.1 Lead Core Diameter, Example 1.2
Determine the required diameter of the lead plug, dL,
using:
3.50
(E-1) 1.97
0.9 0.9 0.9
See Step E2.5 for limitations on dL

80

E2.2 Plan Area and Isolator Diameter E2.2 Plan Area and Isolator Diameter, Example
Although no limits are placed on compressive stress 1.2
in the GSID, (maximum strain criteria are used
instead, see Step E3) it is useful to begin the sizing
process by assuming an allowable stress of, say, 1.0
ksi.

Then the bonded area of the isolator is given by:

(E-2)
1.0 45.52 15.50
61.02
1.0 1.0
and the corresponding bonded diameter (taking into
account the hole required to accommodate the lead
core) is given by:
4
(E-3)
4 4 61.02
1.61

Round the bonded diameter, B, to nearest quarter = 9.03 in

inch, and recalculate actual bonded area using
Round B up to 9.25 in and the actual bonded area is:
(E-4)
4
9.25 1.97 64.15
4
Note that the overall diameter is equal to the bonded
diameter plus the thickness of the side cover layers
(usually 1/2 inch each side). In this case the overall
diameter, Bo is given by:

1.0 (E-5)
Bo = 9.25 + 2(0.5) = 10.25 in

E2.3 Elastomer Thickness and Number of Layers E2.3 Elastomer Thickness and Number of Layers,
Since the shear stiffness of the elastomeric bearing is Example 1.2
given by:
(E-6)
where G = shear modulus of the rubber, and
Tr = the total thickness of elastomer, Select G, shear modulus of rubber, = 100 psi (0.1 ksi)
it follows Eq. E-5 may be used to obtain Tr given a
required value for Kd Then
0.1 64.15
(E-7) 5.49
1.17
A typical range for shear modulus, G, is 60-120 psi.
Higher and lower values are available and are used in
special applications.

If the layer thickness is tr, the number of layers, n, is

given by:
5.49
(E-8) 21.97
0.25
rounded up to the nearest integer.
Note that because of rounding the plan dimensions Round to nearest integer, i.e. n = 22

81

and the number of layers, the actual stiffness, Kd, will

not be exactly as required. Reanalysis may be
necessary if the differences are large.

E2.4 Overall Height E2.4 Overall Height, Example 1.2

The overall height of the isolator, H, is given by:

1 2 ( E-9) 22 0.25 21 0.125 2 1.5 11.125

where ts = thickness of an internal shim (usually

about 1/8 in), and
tc = combined thickness of end cover plate
(0.5 in) and outer plate (1.0 in)

E2.5 Size Checks E2.5 Size Checks, Example 1.2

Experience has shown that for optimum performance Since B=9.25 check
of the lead core it must not be too small or too large.
The recommended range for the diameter is as 9.25 9.25
follows: 3 6
(E-10)
3 6 i.e., 3.08 1.54

Since dL = 1.97, lead core size is acceptable.

Art. 12.2 GSID requires that the isolation system 0.025 0.025 45.52
provides a lateral restoring force at dt greater than the , 0.16 /
7.32
restoring force at 0.5dt by not less than W/80. This
equates to a minimum Kd of 0.025W/d. As
0.1 64.15
0.025 1.17 / ,
5.5
,

E3. Strain Limit Check E3. Strain Limit Check, Example 1.2
Art. 14.2 and 14.3 GSID requires that the total
applied shear strain from all sources in a single layer
of elastomer should not exceed 5.5, i.e.,
Since
, + 0.5 5.5 (E-11)
45.52
0.710
where , , , are defined below. 64.15
(a) is the maximum shear strain in the layer due to
compression and is given by: G = 0.1 ksi
(E-12) and
where Dc is shape coefficient for compression in 64.15
8.83
circular bearings = 1.0, , G is shear 9.25 0.25
modulus, and S is the layer shape factor given by: then
(E-13) 1.0 0.710
0.804
0.1 8.83
(b) , is the shear strain due to earthquake loads
and is given by:

82

 6.71
, (E-14) , 1.222
5.5

(c) is the shear strain due to rotation and is given

by:
0.375 9.25 0.01
(E-15) 0.233
0.25 5.5
where Dr is shape coefficient for rotation in circular
bearings = 0.375, and is design rotation due to DL, Substitution in Eq E-11 gives
LL and construction effects Actual value for may
not be known at this time and value of 0.01 is , 0.5 0.804 1.222 0.5 0.233
suggested as an interim measure, including 2.14
uncertainties (see LRFD Art. 14.4.2.1) 5.5

E4. Vertical Load Stability Check E4. Vertical Load Stability Check, Example 1.2
Art 12.3 GSID requires the vertical load capacity of
all isolators be at least 3 times the applied vertical
loads (DL and LL) in the laterally undeformed state.

Further, the isolation system shall be stable under

1.2(DL+SL) at a horizontal displacement equal to
either
2 x total design displacement, dt, if in Seismic Zone 1
or 2, or
1.5 x total design displacement, dt, if in Seismic Zone
3 or 4.

E4.1 Vertical Load Stability in Undeformed State E4.1 Vertical Load Stability in Undeformed State,
The critical load capacity of an elastomeric isolator at Example 1.2
zero shear displacement is given by

4
1 1 3 3 0.1 0.3
2 (E-16)
0.3 1 0.67 8.83 15.97
where
9.25
359.37
Ts = total shim thickness 64

15.97 359.37
1043.68 /
1 0.67 5.5
E = elastic modulus of elastomer = 3G
0.1 64.15
64 1.17 /
5.5
It is noted that typical elastomeric isolators have high
shape factors, S, in which case:
4
1 (E-17)

and Eq. E-16 reduces to:

(E-18) 1.17 1043.68 109.61

Check that:

83


3 (E-19) 109.61

1.80 3
45.52 15.5

E4.2 Vertical Load Stability in Deformed State E4.2 Vertical Load Stability in Deformed State,
The critical load capacity of an elastomeric isolator at Example 1.2
shear displacement may be approximated by: Since bridge is in Zone 4, 1.5 1.5 6.71
10.07 in

(E-20) Since
where
Ar = overlap area between top and bottom plates
0
of isolator at displacement (Fig. 2.2-1
GSID)
Therefore, clearly the design displacement is too large
= 4 for this initial design. A redesign should be
2 undertaken with increased isolator dimensions. As a
general rule, the minimum isolator diameter, B,
should be of the order of 1.5 to ensure a sufficient
Agross = 4 overlap area.

It follows that:
(E-21)

Check that:

1 (E-22)
1.2

E5. Design Review E5. Design Review, Example 1.2

The redesign is outlined below. After repeating the
calculation for diameter of lead core, the process
begins by reducing the shear modulus to 60 psi (0.06
ksi), increasing the bonded diameter to 1.5 =
15.11 in, rounded up to 15.25 in.

E2.1
3.50
1.97
0.9 0.9

E2.2

15.25 1.97 179.61

4

Bo = 15.25 + 2(0.5) = 16.25 in

E2.3
0.06 179.61
9.23
1.17

84

 9.23
36.91
0.25
Round to nearest integer, i.e. n = 37
E2.4
37 0.25 36 0.125 2 1.5 16.75
E2.5
Since B = 15.25 in, check

15.25 15.25
3 6

i.e., 5.08 2.54

Since dL = 1.97, the size of lead core is too small, and

there are 2 options: (1) accept the undersize and check
for adequate performance during the Quality Control
Tests required by GSID Art. 15.2.2; or (2) only have
lead cores in every second isolator, in which case the
core diameter, in isolators with cores, will be 2 x
1.97 = 2.79 in (which satisfies above criterion).

0.06 179.61
1.17 / ,
9.25

E3.
45.52
0.25
179.61

179.61
15.00
15.25 0.25

1.0 0.25
0.282
0.06 15.00

6.71
, 0.727
9.25

0.375 15.25 0.01

0.377
0.25 9.25

, 0.5 0.282 0.727 0.5 0.377

1.20 5.5
E4.1
3 3 0.06 0.18

0.18 1 0.67 15.00 27.30

15.25
2654.91
64
27.30 2654.91
7835.21 /
9.25

85

 0.06 179.61
1.165 /
9.25

1.165 7835.21 300.15

300.15
4.92 3
45.52 15.50

E4.2
10.07
2 1.70
15.25

1.70 1.70
0.224

0.224 300.15 67.33

67.33
1.02 1
1.2 1.2 45.52 11.52

E5.
The basic dimensions of the redesigned isolator are as
follows:

16.25 in (od) x 16.75 in (high) x 1.97 in dia. lead core

and its volume (excluding steel end and cover plates)
is 2852 in3.

E6. Minimum and Maximum Performance Check E6. Minimum and Maximum Performance Check,
Art. 8 GSID requires the performance of any Example 1.2
isolation system be checked using minimum and Minimum Property Modification factors are:
maximum values for the effective stiffness of the min,Kd = 1.0
system. These values are calculated from minimum min,Qd = 1.0
and maximum values of Kd and Qd, which are found
using system property modification factors, as which means there is no need to reanalyze the bridge
indicated in Table E6-1. with a set of minimum values.
Table E6-1. Minimum and maximum values
for Kd and Qd. Maximum Property Modification factors are:
max,a,Kd = 1.1
Eq. max,a,Qd = 1.1
8.1.2-1 Kd,max = Kd max,Kd (E-23)
GSID max,t,Kd = 1.1
Eq. max,t,Qd = 1.4
8.1.2-2 Kd,min = Kd min,Kd (E-24)
GSID max,scrag,Kd = 1.0
Eq. max,scrag,Qd = 1.0
8.1.2-3 Qd,max = Qd max,Qd (E-25) Applying a system adjustment factor of 0.66 for an
GSID other bridge, the maximum property modification
Eq. factors become:
8.1.2-4 Qd,min = Qd min,Qd (E-26)
GSID max,a,Kd = 1.0 + 0.1(0.66) = 1.066

86

Determination of the system property modification max,a,Qd = 1.0 + 0.1(0.66) = 1.066

factors should include consideration of the effects of
temperature, aging, scragging, velocity, travel (wear)
and contamination as shown in Table E6-2. In lieu of
tests, numerical values for these factors can be
obtained from Appendix A, GSID.
Table E6-2. Minimum and maximum values for
system property modification factors.
max,t,Kd = 1.0 + 0.1(0.66) = 1.066
max,t,Qd = 1.0 + 0.4(0.66) = 1.264
max,scrag,Kd = 1.0
max,scrag,Qd = 1.0
Therefore the maximum overall modification factors

Eq. min,Kd = (min,t,Kd) (min,a,Kd) max,Kd = 1.066(1.066)1.0 = 1.14

8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-27) max,Qd = 1.066(1.264)1.0 = 1.35
GSID (min,scrag,Kd)
Eq. max,Kd = (max,t,Kd) (max,a,Kd) Since the possible variation in upper bound properties
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-28) exceeds 15% a reanalysis of the bridge is required to
GSID (max,scrag,Kd) determine performance with these properties.
Eq. min,Qd = (min,t,Qd) (min,a,Qd)
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-29) The upper-bound properties are:
GSID (min,scrag,Qd) Qd,max = 1.35 (3.50) = 4.73 k
Eq. max,Qd = (max,t,Qd) (max,a,Qd) and
8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd) (E-30) Kd,ma x=1.14(1.17) = 1.33 k/in
GSID (max,scrag,Qd)

Adjustment factors are applied to individual -factors

(except v) to account for the likelihood of
occurrence of all of the maxima (or all of the
minima) at the same time. These factors are applied
to all -factors that deviate from unity but only to the
portion of the -factor that is greater than, or less
than, unity. Art. 8.2.2 GSID gives these factors as
follows:
1.00 for critical bridges
0.75 for essential bridges
0.66 for all other bridges

As required in Art. 7 GSID and shown in Fig. C7-1

GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Qd,max. As indicated in Fig C7-1 GSID, maximum
displacements will probably be given by the first case
(Kd,min and Qd,min) and maximum forces by the second
case (Kd,max and Qd,max).

87

E7. Design and Performance Summary E7. Design and Performance Summary, Example 1.2
E7.1 Isolator dimensions E7.1 Isolator dimensions, Example 1.2
Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E7.1-1.
Overall diameter (includes cover layer)
Overall height Table E7.1-1 Isolator Dimensions
Diameter lead core
Isolator Overall size Overall size Diam.
Bonded diameter Location including without lead
Number of rubber layers mounting mounting core
Thickness of rubber layers plates (in) plates (in) (in)
Total rubber thickness Under
Thickness of steel shims edge 20.25 x 20.25 16.25 dia.
Shear modulus of elastomer 1.97
girder x 16.75(H) x 15.25(H)
on Pier 1
Check all dimensions with manufacturer.

Isolator No. of Rubber Total Steel

Location rubber layers rubber shim
layers thick- thick- thick-
ness ness ness
(in) (in) (in)
Under
edge
37 0.25 9.25 0.125
girder
on Pier 1

Shear modulus of elastomer = 60 psi

E7.2 Bridge Performance E7.2 Bridge Performance, Example 1.2

Summarize bridge performance
Maximum superstructure displacement
(longitudinal)
Maximum superstructure displacement
(transverse)
Maximum superstructure displacement
(resultant)
Maximum column shear (resultant)
Maximum column moment (about transverse
axis)
Maximum column moment (about
longitudinal axis)
Maximum column torque
Check required performance as determined in Step
A3, is satisfied.
Bridge performance is summarized in Table E7.2-1
where it is seen that the maximum column shear is 29.15
k. This is more than the column plastic shear (25 k) and
therefore the required performance criterion is not
satisfied (fully elastic behavior). Clearly the seismic
demand (S1 = 0.6g) is too high and the column too small
for isolation to give fully elastic response. The next step
might be to conduct a pushover analysis of the pier to
determine if the ductility demand at 7.32 in is
acceptable. If not, and this is an existing bridge, jacket
the column (as well as isolate the bridge). If a new
design, increase the size of the column and thereby
increase its strength. It is noted that the maximum
longitudinal displacement (7.05 in) exceeds the available
clearance, and as noted in Section B1.12, this gap needs
to be increased to say 8.0 in to meet the requirements for
isolation. An alternative is to accept pounding at the
abutments. The consequential damage is not likely to be
life-threatening and easily repaired.

88

 Table E7.2-1 Summary of Bridge Performance

Maximum superstructure
7.05 in
displacement (longitudinal)
Maximum superstructure
6.54 in
displacement (transverse)
Maximum superstructure
7.32 in
displacement (resultant)
Maximum column shear (resultant) 29.15 k
Maximum column moment about
444 kft
transverse axis
Maximum column moment about
199 kft
longitudinal axis
Maximum column torque 3.21 kft

89

 SECTION 1
PC Girder Bridge, Short Spans, Multi-Column Piers

DESIGN EXAMPLE 1.3: Spherical Friction Isolators

Design Examples in Section 1

Site
ID Description S1 Pier height Skew Isolator type
Class
1.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
1.1 Change site class 0.2g D Same 0 Lead rubber bearing
Change spectral
1.2 0.6g B Same 0 Lead rubber bearing
acceleration, S1
Spherical friction
1.3 Change isolator to SFB 0.2g B Same 0
bearing
1.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
1.5 Change column height 0.2g B H1=0.5 H2 0 Lead rubber bearing
1.6 Change angle of skew 0.2g B Same 450 Lead rubber bearing

90


DESIGN PROCEDURE
STEP A: BRIDGE AND SITE DATA
A1. Bridge Properties
Determine properties of the bridge:
number of supports, m
number of girders per support, n
angle of skew
weight of superstructure including railings,
curbs, barriers and to the permanent loads,
WSS
weight of piers participating with
superstructure in dynamic response, WPP
weight of superstructure, Wj, at each
support
piers heights (clear dimensions)
stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge
column flexural yield strength (minimum
value)
column shear strength (minimum value)
allowable movement at expansion joints
isolator type if known, otherwise to be
determined
A2. Seismic Hazard
Determine seismic hazard at site:
acceleration coefficients
site class and site factors
seismic zone
Plot response spectrum.
Use Art. 3.1 GSID to obtain peak ground and spectral
acceleration coefficients. These coefficients are the
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (e.g. Figures
3.10.2.1-1 to 3.10.2.1-21 LRFD).
Use Art. 3.2 to obtain Site Class and corresponding Site
Factors (Fpga, Fa and Fv). These data are the same as for
DESIGN EXAMPLE 1.3 (Spherical Friction
Isolator)
A1. Bridge Properties, Example 1.3
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 6
Number of columns per support = 3
Angle of skew = 00
Weight of superstructure including permanent
loads, WSS = 650.52 k
Weight of superstructure at each support:
o W1 = 44.95 k
o W2 = 280.31 k
o W3 = 280.31 k
o W4 = 44.95 k
Participating weight of piers, WPP = 107.16 k
Effective weight (for calculation of period),
Weff = Wss + WPP = 757.68 k
Stiffness of each pier in the longitudinal
direction:
o Ksub,pier1,long = 172.0 k/in
o Ksub,pier2,long = 172.0 k/in
Stiffness of each pier in the transverse direction:
o Ksub,pier1,trans = 687.0 k/in
o Ksub,pier2,trans = 687.0 k/in
Minimum flexural yield strength of one column =
425 kft (plastic moment capacity)
Displacement capacity of expansion joints
(longitudinal) = 2.0 in for thermal and other
movements.
Spherical friction bearing isolators
A2. Seismic Hazard, Example 1.3
Acceleration coefficients for bridge site are given in
design statement as follows:
PGA = 0.40
S1 = 0.20
SS = 0.75
Bridge is on a rock site with shear wave velocity in
upper 100 ft of soil = 3,000 ft/sec.
Table 3.10.3.1-1 LRFD gives Site Class as B.
Tables 3.10.3.2-1, -2 and -3 LRFD give following
Site Factors:
Fpga = 1.0
Fa = 1.0
Fv = 1.0
91

conventional bridges and Art 3.2 refers the designer to

the corresponding articles in the LRFD Specifications,
i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3,
LRFD.

Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give

modified spectral acceleration coefficients that include
site effects as follows: As = Fpga PGA = 1.0(0.40) = 0.40
As = Fpga PGA SDS = Fa SS = 1.0(0.75) = 0.75
SDS = Fa SS SD1 = Fv S1 = 1.0(0.20) = 0.20
SD1 = Fv S1

Seismic Zone is determined by value of SD1 in Since 0.15 < SD1 < 0.30, bridge is located in Seismic
accordance with provisions in Table5-1 GSID. Zone 2.

These coefficients are used to plot design response Design Response Spectrum is as below:
spectrum as shown in Fig. 4-1 GSID.
0.9

0.8

0.7

0.6

Acceleration
0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Period(s)

A3. Performance Requirements A3. Performance Requirements, Example 1.3

Determine required performance of isolated bridge
during Design Earthquake (1000-yr return period).
Examples of performance that might be specified by
the Owner include:
Reduced displacement ductility demand in
columns, so that bridge is open for emergency
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in
columns or yield elsewhere in bridge), so that
bridge is fully functional and open to all vehicles
immediately following earthquake.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.
In this example, assume the owner has specified full
functionality following the earthquake and therefore
the columns must remain elastic (no yield).
To remain elastic the maximum lateral load on the
pier must be less than the load to yield the column.
This load is taken as the plastic moment capacity
(strength) of the column (425 kft, see above) divided
by the overall column height (17 ft). This calculation
assumes the column is acting as a simple cantilever
in single curvature in the longitudinal direction.
Hence load to yield column = 425 /17 = 25.0 k
The maximum shear in the column must therefore be
less than 25 k in order to keep the column elastic and
meet the required performance criterion.

92

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for Isolator Force, F

non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most have
bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use Ku
equivalent linear springs and viscous damping to
represent the isolators, so that linear methods of analysis
disol
may be used to determine response. Since equivalent Ku
dy Isolator
Ku Displacement, d
properties such as Kisol are dependent on displacement (d),
and the displacements are not known at the beginning of
the analysis, an iterative approach is required. Note that in Kd
Art 7.1, GSID, keff is used for the effective stiffness of an
isolator unit and Keff is used for the effective stiffness of a disol = Isolator displacement
dy = Isolator yield displacement
combined isolator and substructure unit. To minimize Fisol = Isolator shear force
confusion, Kisol is used in this document in place of keff. Fy = Isolator yield force
Kd = Post-elastic stiffness of isolator
There is no change in the use of Keff and Keff,j, but Ksub is Kisol = Effective stiffness of isolator
used in place of ksub. Ku
Qd
= Loading and unloading stiffness (elastic stiffness)
= Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement (disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 1.3
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.20 2.0
GSID

(2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not
occur under non-seismic loads (e.g. wind) but

93

 low enough that yield will occur during an

earthquake. Experience has shown that
taking Qd to be 5% of the bridge weight is a
good starting point, i.e. 0.05 0.05 650.52 32.53
0.05 (B-2)
(3) Post-yield stiffness, Kd
Art 12.2 GSID requires that all isolators
exhibit a minimum lateral restoring force at
the design displacement, which translates to a
minimum post yield stiffness Kd,min given by
Eq. B-3.

Art. 650.52
0.025 0.05 0.05 16.26 /
12.2 , (B-3) 2.0
GSID
Experience has shown that a good starting
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 1.3
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength, ,

Qd, and stiffness, Kd, values in proportion to the dead o Qd, 1 = 2.25 k
load applied at that support: o Qd, 2 = 14.02 k
o Qd, 3 = 14.02 k
o Qd, 4 = 2.25 k

, (B-4)
and
and
,
, (B-5) o Kd,1 = 1.12 k/in
o Kd,2 = 7.01 k/in
o Kd,3 = 7.01 k/in
o Kd,4 = 1.12 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 1.3
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of , ,
the isolators at support j (Kisol,j) and the stiffness of , ,
the substructure Ksub,j. See figure below for
definitions (after Fig. 7.1-1 GSID). o 1 = 2.25x10-4
o 2 = 8.49x10-2
An expression for Keff,j, is given in Eq.7.1-6 GSID, but o 3 = 8.49x10-2
a more useful formula as follows (MCEER 2006): o 4 = 2.25x10-4
,
, (B-6)
1 ,
where ,
1
, ,
(B-7)
, , o Keff,1 = 2.25 k/in
and Ksub,j for the piers are given in Step A1. For the o Keff,2 = 13.47 k/in
abutments, take Ksub,j to be a large number, say 10,000 o Keff,3 = 13.47 k/in
k/in, unless actual stiffness values are available. Note o Keff,4 = 2.25 k/in
that if the default option is chosen, unrealistically high

94

 values for Ksub,j will give unconservative results for

column moments and shear forces.

F
Kd
Qd Kisol

dy disol
Superstructure
Isolator Effective Stiffness, Kisol

F
Isolator(s), Kisol
Ksub

Substructure, Ksub
dsub

Substructure Stiffness, Ksub

dsub disol F

d
Keff

d = disol + dsub

Combined Effective Stiffness, Keff

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 1.3
Calculate the total effective stiffness, Keff, of the
bridge:

Eq. , 31.43 /
7.1-6 , ( B-8)
GSID

B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 1.3
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1

(B-9) o disol,1 = 2.00 in

,
1 o disol,2 = 1.84 in
o disol,3 = 1.84 in
o disol,4 = 2.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the effective stiffness of the isolation system Example 1.3
at support j, Kisol,j, for all supports:
,
, ,
,
,
(B-10) o Kisol,1 = 2.25 k/in
, ,
, o Kisol,2 = 14.61 k/in
o Kisol,3 = 14.61 k/in
o Kisol,4 = 2.25 k/in

95

B1.7 Substructure Displacement at Each Support
B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j,
Example 1.3
for all supports:
, ,

, , (B-11)
o dsub,1 = 4.49x10-4 in
o d sub,2 = 1.57x10-1 in
o d sub,3 = 1.57x10-1 in
o d sub,4 = 4.49x10-4 in
B1.8 Lateral Load in Each Substructure Support B1.8 Lateral Load in Each Substructure, Example
Calculate the shear at support j, Fsub,j, for all 1.3
supports:
, , , (B-12) , , ,

where values for Ksub,j are given in Step A1. o F sub,1 = 4.49 k
o F sub,2 = 26.93 k
o F sub,3 = 26.93 k
o F sub,4 = 4.49 k

B1.9 Column Shear Force at Each Support B1.9 Column Shear Force at Each Support,
Calculate the shear in column k at support j, Fcol,j,k, Example 1.3
assuming equal distribution of shear for all columns at
support j:
,
, ,
#
,
, ,
# (B-13) o F col,2,1-3 8.98 k
o F col,3,1-3 8.98 k

Use these approximate column shear forces as a check These column shears are less than the yield capacity
on the validity of the chosen strength and stiffness of each column (25 k) as required in Step A3 and the
characteristics. chosen strength and stiffness values in Step B1.1 are
therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 1.3
damping ratio, , of the bridge:

Eq. 757.68
7.1-5 2 (B-14) 2 2
GSID 386.4 31.43

and = 1.57 sec

Eq. 2 , , , and taking dy,j = 0:

(B-15)
7.1-10 , , ,
GSID 2 , , 0
0.31
, , ,
where dy,j is the yield displacement of the isolator. For
friction-based isolators, dy,j = 0. For other types of
isolators dy,j is usually small compared to disol,j and has
negligible effect on , Hence it is suggested that for
the Simplified Method, set dy,j = 0 for all isolator

96

types. See Step B2.2 where the value of dy,j is revisited
for the Multimode Spectral Analysis Method.
B1.11 Damping Factor
Calculate the damping factor, BL, and the
displacement, d, of the bridge:
Eq. .
, 0.3
.
7.1-3 (B-16)
1.7, 0.3
GSID
Eq. 9.79 (B-17)
7.1-4
GSID
B1.12 Convergence Check
Compare the new displacement with the initial value
assumed in Step B1.1. If there is close agreement, go
to the next step; otherwise repeat the process from
Step B1.3 with the new value for displacement as the
assumed displacement.
This iterative process is amenable to solution using a
spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
B1.11 Damping Factor, Example 1.3
Since 0.31 0.3
1.70
and
9.79 9.79 0.2 1.57
1.81
1.70
B1.12 Convergence Check, Example 1.3
Since the calculated value for displacement, d
(=1.81 in) is not close to that assumed at the
beginning of the cycle (Step B1.1, d = 2.0), use the
value of 1.81 in as the new assumed displacement and
repeat from Step B1.3.
After one iteration, convergence is reached at a
superstructure displacement of 1.76 in, with an
effective period of 1.52 seconds, and a damping factor
of 1.70 (33% damping ratio). The displacement in the
isolators at Pier 1 is 1.61 in and the effective stiffness
of the same isolators is 15.69 k/in.
See spreadsheet in Table B1.12-1 for results of final
iteration.
Ignoring the weight of the cap beams, the sum of the
column shears at a pier must equal the total isolator
shear force. Hence, approximate column shear (per
column) = 15.69(1.61)/3 = 8.42 k which is less than
the maximum allowable (25k) if elastic behavior is to
be achieved (as required in Step A3).
Also the superstructure displacement = 1.76 in, which
is less than the available clearance of 2.0 in.
Therefore the above solution is acceptable and go to
Step B2.
Note that available clearance (2.0 in) is greater than
minimum required which is given by:
8 8 0.20 1.52
1.43
1.7
97

 Table B1.12-1 Simplified Method Solution for Design Example 1.3 Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
650.52 107.16 757.68 0.2 6

StepB1.1 d 1.72 Assumeddisplacement

Qd 32.53 Characteristicstrength
Kd 16.26 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 44.95 2.25 1.12 10000 2.40E04 2.40 1.76 2.40 4.23E04 4.23 3.958 7.444
Pier1 280.31 14.02 7.01 172.0 9.12E02 14.38 1.61 15.69 1.47E01 25.33 22.623 44.611
Pier2 280.31 14.02 7.01 172.0 9.12E02 14.38 1.61 15.69 1.47E01 25.33 22.623 44.611
Abut2 44.95 2.25 1.12 10000 2.40E04 2.40 1.76 2.40 4.23E04 4.23 3.958 7.444
Total 650.52 32.526 16.263 K eff,j 33.557 59.107 53.161 104.109
Step B1.4

StepB1.10 T eff 1.52 Effectiveperiod

0.33 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.75

B L 1.70 DampingFactor
d 1.75 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 0.375 0.187 0.400 1.66 0.413
Pier1 2.336 1.168 2.615 1.47 2.757
Pier2 2.336 1.168 2.615 1.47 2.757
Abut2 0.375 0.187 0.400 1.66 0.413

98

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and
then applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 1.3

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 (see
elastic stiffness, Kd,i, of each isolator i as follows: Table B1.12-1) by the number of isolators at each
, support (n = 6), the following values for Qd /isolator
, (B-19)
and Kd /isolator are obtained:
and
, o Qd, 1 = 2.25/6 = 0.37 k
, ( B-20)
o Qd, 2 = 14.02/6= 2.34 k
where values for Qd,j and Kd,j are obtained from the o Qd, 3 = 14.02/6= 2.34 k
final cycle of iteration in the Simplified Method (Step o Qd, 4 = 2.25/6 = 0.37 k
B1. 12 and
o Kd,1 = 1.12/6 = 0.19 k/in
o Kd,2 = 7.01/6 = 1.17 k/in
o Kd,3 = 7.01/6 = 1.17 k/in
o Kd,4 = 1.12/6 = 0.19 k/in

B2.2 Initial Stiffness and Yield Displacement B2.2 Initial Stiffness and Yield Displacement,
Calculate the initial stiffness, Ku,i, and the yield Example 1.3
displacement, dy,i, for each isolator i as follows: Since the isolator type has been specified in Step A1
(1) For friction-based isolators: Ku,i = and dy,i = 0. to be an elastomeric bearing (i.e. not a friction-based
(2) For other types of isolators, and in the absence of bearing), calculate Ku,i and dy,i for an isolator on Pier 1
isolator-specific information, take as follows:
, 10 , 10 1.17 11.7 /
, 10 , ( B-21)
and then and
, 2.34
,
( B-22) , 0.22
, , , 11.7 1.17
, ,

As expected, the yield displacement is small

compared to the expected isolator displacement (~2
in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.

B2.3 Isolator Effective Stiffness, Kisol,i B2.3 Isolator Effective Stiffness, Kisol,i, Example 1.3
Calculate the isolator stiffness, Kisol,i, of each isolator Dividing the results for Kisol (Step B1.12) among the 6
i: isolators at each support, the following values for Kisol
/isolator are obtained:
, o Kisol,1 = 2.40/6 = 0.40 k/in
, (B -23)
o Kisol,2 = 15.69/6 = 2.62 k/in
o Kisol,3 = 15.69/6 = 2.62 k/in

99

 o Kisol,4 = 2.40/6 = 0.40 k/in

B2.4 Three-Dimensional Bridge Model B2.4 Three-Dimensional Bridge Model, Example

Using computer-based structural analysis software, 1.3
create a 3-dimensional model of the bridge with the Although the bridge in this Design Example is regular
isolators represented by spring elements. The and is without skew or curvature, a 3-dimensional
stiffness of each isolator element in the horizontal finite element model was developed for this Step, as
axes (Kx and Ky in global coordinates, K2 and K3 in shown below.
typical local coordinates) is the Kisol value calculated
in the previous step. For bridges with regular
geometry and minimal skew or curvature, the
superstructure may be represented by a single stick
provided the load path to each individual isolator at
each support is explicitly modeled, usually by a rigid
cap beam and a set of rigid links. If the geometry is
irregular, or if the bridge is skewed or curved, a finite
element model is recommended to accurately capture
the load carried by each individual isolator. If the
piers have an unusual weight distribution, such as a .
pier with a hammerhead cap beam, a more rigorous
model is justified.

B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in
Step A2 is for a 5% damped system. It is modified in
this step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the
effective period of the bridge, Teff, by the damping
factor, BL.
B.2.5 Composite Design Response Spectrum,
Example 1.3
From the final results of Simplified Method (Step
B1.12), BL = 1.70 and Teff = 1.52 sec. Hence the
transition in the composite spectrum from 5% to 33%
damping occurs at 0.8 Teff = 0.8 (1.46) = 1.22 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.22 sec by 1.70.
0.9

0.8

0.7
Acceleration(g)

0.6

0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Period(s)

B2.6 Multimodal Analysis of Finite Element
Model
Input the composite response spectrum as a user-
specified spectrum in the software, and define a load
case in which the spectrum is applied in the
longitudinal direction. Analyze the bridge for this
load case.
B2.6 Multimodal Analysis of Finite Element
Model, Example 1.3
Results of modal analysis of the example bridge are
summarized in Table B2.6-1 Here the modal periods
and mass participation factors of the first 12 modes
are given. The first two modes are the principal
longitudinal and transverse modes with periods of
1.41 and 1.35 sec respectively. The period of the
longitudinal mode (1.41 sec) is close to that
calculated in the Simplified Method.

100

 Table B2.6-1 Modal Properties of Bridge

Example 1.3 First Iteration
Mode Period ModalParticipatingMassRatios
No. Sec UX UY UZ RX RY RZ
1 1.410 0.761 0.000 0.000 0.000 0.003 0.000
2 1.346 0.000 0.738 0.031 0.059 0.000 0.534
3 1.325 0.000 0.000 0.000 0.000 0.000 0.217
4 0.187 0.000 0.000 0.000 0.000 0.000 0.000
5 0.186 0.125 0.000 0.000 0.000 0.000 0.000
6 0.121 0.000 0.000 0.000 0.000 0.000 0.000
7 0.121 0.000 0.000 0.000 0.000 0.000 0.006
8 0.104 0.000 0.034 0.183 0.107 0.000 0.064
9 0.101 0.000 0.000 0.000 0.000 0.226 0.000
10 0.095 0.000 0.121 0.081 0.184 0.000 0.041
11 0.094 0.000 0.000 0.000 0.000 0.000 0.000
12 0.074 0.000 0.002 0.000 0.003 0.000 0.000

Computed values for the isolator displacements due to

a longitudinal earthquake are as follows (numbers in
parentheses are those used to calculate the initial
properties to start iteration from the Simplified
Method):

o disol,1 = 1.65 (1.76) in

o disol,2 = 1.47 (1.61) in
o disol,3 = 1.47 (1.61) in
o disol,4 = 1.65 (1.76) in

B2.7 Convergence Check B2.7 Convergence Check, Example 1.3

Compare the resulting displacements at the The new superstructure displacement is 1.65 in, more
superstructure level (d) to the assumed than a 5% difference from the displacement assumed
displacements. These displacements can be obtained at the start of the Multimode Spectral Analysis.
by examining the joints at the top of the isolator
spring elements. If in close agreement, go to Step
B2.9. Otherwise go to Step B2.8.

B2.8 Update Kisol,i, Keff,j, and BL B2.8 Update Kisol,i, Keff,j, and BL, Example 1.3
Use the calculated displacements in each isolator Updated values for Kisol,i are given below (previous
element to obtain new values of Kisol,i for each values are in parentheses):
isolator as follows:
, o Kisol,1 = 0.41 (0.40) k/in
, , (B-24) o Kisol,2 = 2.76 (2.62) k/in
,
o Kisol,3 = 2.76 (2.62) k/in
Recalculate Keff,j : o Kisol,4 = 0.41 (0.40) k/in

Eq. , ,
Updated values for Keff,j, , BL and Teff are given below
7.1-6 ,
(B-25) (previous values are in parentheses):
, ,
GSID
o Keff,1 = 2.48 (2.40) k/in
Recalculate system damping ratio, : o Keff,2 = 15.48 (14.38) k/in
o Keff,3 = 15.48 (14.38) k/in
Eq. 2 , , ,
o Keff,4 = 2.48 (2.40) k/in
7.1-10 , , ,
(B-26)
GSID o = 27% (33%)
o BL = 1.66 (1.70)
Recalculate system damping factor, BL: o Teff = 1.41 (1.52) sec
The updated composite response spectrum is shown
. below:
Eq. 0.3
. ( B-27)
7.1-3 1.7 0.3

101

 GSID 0.9

0.8

Obtain the effective period of the bridge from the 0.7

Acceleration(g)
multi-modal analysis and with the revised damping 0.6

factor (Eq. B-27), construct a new composite 0.5

response spectrum. Go to Step B2.6. 0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Period(s)

B2.6 Multimodal Analysis Second Iteration,

Example 1.3
Reanalysis gives the following values for the isolator
displacements (numbers in parentheses are from the
previous cycle):

o disol,1 = 1.66 (1.65) in

B2.7 Convergence Check
Compare the resulting displacements at the
superstructure level (d) to the assumed
displacements. These displacements can be obtained
by examining the joints at the top of the isolator
spring elements. If in close agreement, go to Step
B2.9. Otherwise go to Step B2.8.
o disol,2 = 1.47 (1.47) in
o disol,3 = 1.47 (1.47) in
o disol,4 = 1.66 (1.65) in
B2.7 Convergence Check, Example 1.3
The new superstructure displacement is 1.66 in, less
than a 1% difference from the displacement assumed
at the start of the second cycle of Multimode Spectral
Analysis.

B2.9 Superstructure and Isolator Displacements
Once convergence has been reached, obtain
o superstructure displacements in the
longitudinal (xL) and transverse (yL) directions
of the bridge, and
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions of the bridge, for
each isolator, for this load case (i.e.
longitudinal loading). These displacements
may be found by subtracting the nodal
displacements at each end of each isolator
spring element.
B2.9 Superstructure and Isolator Displacements,
Example 1.3
From the above analysis:
o superstructure displacements in the
longitudinal (xL) and transverse (yL) directions
are:
xL= 1.66 in
yL= 0.0 in
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions are:
o Abutments: uL = 1.66 in, vL = 0.00 in
o Piers: uL = 1.47 in, vL = 0.00 in

All isolators at same support have the same

displacements.

B2.10 Pier Bending Moments and Shear Forces B2.10 Pier Bending Moments and Shear Forces,
Obtain the pier bending moments and shear forces in Example 1.3
the longitudinal (MPLL, VPLL) and transverse (MPTL, Maximum bending moments and shear forces in the
VPTL) directions at the critical locations for the columns in the longitudinal (MPLL,VPLL) and
longitudinally-applied seismic loading. transverse (MPTL,VPTL) directions are:
Exterior Columns:
o MPLL= 0 kft
o MPTL= 240 kft

102

 o VPLL= 15.81 k
o VPTL= 0 k
Interior Columns:
o MPLL= 0 kft
o MPTL= 235 kft
o VPLL= 15.24 k
o VPTL= 0 k

Both piers have the same distribution of bending

moments and shear forces among the columns.

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 1.3
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1.

Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces
ator
ure to long. to long. due to
EQ EQ long. EQ
1 0.69 0.00 0.36
2 0.69 0.00 0.39

Abut 3 0.69 0.00 0.39

ment 4 0.69 0.00 0.39
5 0.69 0.00 0.39
6 0.69 0.00 0.36
1 4.04 0.00 0.13
2 4.05 0.00 0.19
3 4.05 0.00 0.22
Pier
4 4.05 0.00 0.22
5 4.05 0.00 0.19
6 4.04 0.00 0.13

103

 STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake C1. Analysis for Transverse Earthquake,
Repeat the above process, starting at Step B1, for Example 1.3
earthquake loading in the transverse direction of the Key results from repeating Steps B1 and B2
bridge. Support flexibility in the transverse direction (Simplified and Mulitmode Spectral Methods) are:
is to be included, and a composite response spectrum o Teff = 1.43 sec
is to be applied in the transverse direction. Obtain o Superstructure displacements in the longitudinal
isolator displacements in the longitudinal (uT) and (xT) and transverse (yT) directions are as follows:
transverse (vT) directions of the bridge, and the xT = 0 and yT = 1.53 in
biaxial bending moments and shear forces at critical o Isolator displacements in the longitudinal (uT)
locations in the columns due to the transversely- and transverse (vT) directions are as follows:
applied seismic loading. Abutments uT = 0.00 in, vT = 1.53 in
Piers uT = 0.00 in, vT = 1.49 in
o Maximum bending moments and shear forces in
the columns in the longitudinal (MPLL,VPLL) and
transverse (MPTL,VPTL) directions are:
Exterior Columns:
o MPLT= 108 kft
o MPTT= 1 kft
o VPLT= 0.06 k
o VPTT= 14.87 k
Interior Columns:
o MPLT= 120 kft
o MPTT= 0 kft
o VPLT= 0 k
o VPTT= 17.29 k

Both piers have the same distribution of

bending moments and shear forces among the
columns.

o Isolator shear and axial forces are in Table C1-1.

104

 Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces due
ator
ure to transv. to transv. to transv.
EQ EQ EQ
1 0.00 0.65 3.50
2 0.00 0.65 1.93

Abut 3 0.00 0.65 0.68

ment 4 0.00 0.65 0.68
5 0.00 0.65 1.93
6 0.00 0.65 3.50
1 0.02 3.98 12.56
2 0.01 4.00 1.14
3 0.00 4.01 2.29
Pier
4 0.00 4.01 2.29
5 0.01 4.00 1.14
6 0.02 3.98 12.56

105

STEP D. CALCULATE DESIGN VALUES

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements at Pier 1,

Following the provisions in Art. 2.1 GSID, and Example 1.3
illustrated in Fig. 2.1-1 GSID, calculate the total To illustrate the process, design displacements for the
design displacement, dt, for each isolator by outside isolator on Pier 1 are calculated below.
combining the displacements from the longitudinal
(uL and vL) and transverse (uT and vT) cases as Load Case 1:
follows: u1 = uL + 0.3uT = 1.0(1.47) + 0.3(0) = 1.47 in
u1 = uL + 0.3uT (D-1) v1 = vL + 0.3vT = 1.0(0) + 0.3(1.49) = 0.45 in
v1 = vL + 0.3vT (D-2) R1 = = 1.47 0.45 = 1.54 in
R1 = (D-3)
Load Case 2:
u2 = 0.3uL + uT (D-4) u2 = 0.3uL + uT = 0.3(1.47) + 1.0(0) = 0.44 in
v2 = 0.3vL + vT (D-5) v2 = 0.3vL + vT = 0.3(0) + 1.0(1.49) = 1.49in
R2 = (D-6) R2 = = 0.44 1.49 = 1.55in

dt = max(R1, R2) (D-7)

D2. Design Moments and Shears
Calculate design values for column bending moments
and shear forces for all piers using the same
combination rules as for displacements.
Governing Case:
Total design displacement, dt = max(R1, R2)
= 1.55in
D2. Design Moments and Shears in Pier 1,
Example 1.3
Design moments and shear forces are calculated for
Pier 1, Column 1, below to illustrate the process.

Alternatively this step may be deferred because the Load Case 1:

above results may not be final. Upper and lower VPL1= VPLL + 0.3VPLT = 1.0(15.81) + 0.3(0.06)
bound analyses are required after the isolators have = 15.83 k
been designed as described in Art 7. GSID. These VPT1= VPTL + 0.3VPTT = 1.0(0.02) + 0.3(14.87)
analyses are required to determine the effect of = 4.48 k
possible variations in isolator properties due age, R1 = = 15.83 4.48 = 16.45 k
temperature and scragging in elastomeric systems.
Accordingly the results for column shear in Steps Load Case 2:
B2.10 and C are likely to increase once these VPL2= 0.3VPLL + VPLT = 0.3(15.81) + 1.0(0.06)
analyses are complete. = 4.80 k
VPT2= 0.3VPTL + VPTT = 0.3(0.02) + 1.0(14.87)
= 14.88 k
R2 = = 4.80 14.88 = 15.63 k

Governing Case:
Design column shear = max (R1, R2)
= 16.45 k

106

STEP E. DESIGN OF SPHERICAL FRICTION ISOLATORS

The Spherical Friction Bearing (SFB) isolator has
an articulated slider to permit rotation, and a POLISHED STAINLESS STEEL SURFACE

spherical sliding interface. It has lateral stiffness

due to the curvature of this interface. These
isolators are capable of carrying very large axial
loads and can be designed to have long periods of SEAL
vibration (5 seconds or longer).

The main components of an SFB isolator are a STAINLESS STEEL R

stainless steel concave spherical plate, an ARTICULATED SLIDER COMPOSITE LINER MATERIAL

articulated slider and a housing plate as illustrated (ROTATIONAL PART)

in figure above. In this figure, the concave spherical plate is facing down. The bearings may also be installed with
this surface facing up as in the figure below. The side of the articulated slider in contact with the concave
spherical surface is coated with a low-friction composite material, usually PTFE. The other side of the slider is
also spherical but lined with stainless steel and sits in a spherical cavity coated with PTFE.

Spherical friction bearings are described by the same equation

of motion as conventional pendulums. As a consequence their
period of vibration is directly proportional to the radius of
curvature of the concave surface. See figure at right. Long
period shifts are therefore possible with surfaces that have large
radii of curvature. Friction between the articulated slider and the
R
concave surface dissipates energy and the weight of the bridge
acts as a restoring force, due to the curvature of the sliding W

surface. Restoringforce

The required values for Qd and Kd determine the coefficient of Friction
friction at the sliding interface and the radius of curvature.
D

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

E1. Required Properties E1. Required Properties, Example 1.3

Obtain from previous work the properties required of The design of one of the pier isolators is given below
the isolation system to achieve the specified to illustrate the design process for spherical friction
performance criteria (Step A1). isolators.
required characteristic strength, Qd / isolator
required post-elastic stiffness, Kd / isolator From previous work
the total design displacement, dt, for each Qd / isolator = 2.34 k
isolator Kd / isolator = 1.17 k/in
maximum applied dead load, PDL Total design displacement, dt = 1.55 in
maximum live load, PLL and PDL = 45.52 k
maximum wind load, PWL PLL = 15.50 k
PWL = 1.76 k < Qd OK

E2. Isolator Dimensions
E2.1 Radius of Curvature E2.1 Radius of Curvature, Example 1.3
Determine the required radius of curvature, R, using:

45.52
(E-1) 38.91 39.0
1.17

107

E2.2 Coefficient of Friction E2.2 Coefficient of Friction, Example 1.3

Determine the required coefficient of friction, ,
using:
2.34
0.0514 5.14%
(E-2) 45.52

E2.3 Material Selection E2.3 Material Selection, Example 1.3

Based on the required coefficient of friction select an Select 15GF Teflon and size disc to achieve required
appropriate PTFE compound and contact pressure, c, contact pressure of 6,500 psi (Step E2.4).
from Table E2.3-1.

Table E2.3-1 Material Properties

PTFE
Compound Contact
(Filled and Pressure, c (%)
Unfilled (psi)
Teflon)
1,000 11.93
Unfilled 2,000 8.70
(UF) 3,000 7.03
6,500 5.72
Glass-filled 1,000 14.61
15% by 2,000 10.08
weight 3,000 8.49
(15GF) 6,500 5.27
Glass-filled 1,000 13.20
25% by 2,000 11.20
weight 3,000 9.60
(25GF) 6,500 5.89

E2.4 Disk Diameter E2.4 Disk Diameter, Example 1.3

Determine the required contact area, Ac, and disk
diameter, d, using:

45.52
(E-3) 7.00
6.5
and
4 4 7.00
(E-4) 2.99 3.00

E2.5 Isolator Diameter E2.5 Isolator Diameter, Example 1.3

Determine the required diameter of the concave As the bridge is in Seismic Zone 2,
surface, Lchord, and overall isolator width, B, using:
= 2(dt) = 2(1.55) = 3.10 in
2 (E-5)
2 2 3.10 1.50 9.20
and
2 (E-6)
Select s = 1.5 in:

108

where: 9.20 2 1.5 12.20 12.25

= 2 x total design displacement, dt, if in Seismic
Zone 1 or 2, or
1.5 x total design displacement, dt, if in Seismic
Zone 3 or 4.
s = width of shoulder of concave plate.

E2.6 Isolator Height E2.6 Isolator Height, Example 1.3

E2.6.1 Rise E2.6.1 Rise, Example 1.3

Determine the rise of the concave surface, h, using:

9.20
(E-7) 0.27
8 8 38.91

E2.6.2 Throat Thickness E2.6.2 Throat Thickness, Example 1.3

Determine the required throat thickness, t, based on Assume safe bearing stress below isolator:
the minimum required bearing area, Ab, such that the
maximum allowable bearing stress, bearing, is not bearing = 2.0 ksi.
exceeded on either the sole plate above or the
masonry plate below, depending on whether the
isolator is installed with concave surface facing up or
down.
45.52 15.50
(E-8) 30.51
2.0

4 4 30.51
6.23
(E-9)

0.5 (E-10) 0.5 6.23 3.0 1.62 1.75

This assumes a 45 distribution of compressive stress

through the throat to the support plates.

E2.6.3 Total Height E2.6.3 Total Height, Example 1.3

Determine the thickness of concave plate, T1, using:

(E-11) 0.27 1.75 2.02 2.25

Thickness of slider plate (T2) will vary with detail for

socket that holds articulated slider and rotation
requirement. Check with manufacturer for value. For
estimating purposes take T2 = T1. 2.25

Then total height of isolator:

(E-12) 2.25 2.25 4.50

E3. Design Summary E3. Design Summary, Example 1.3

Overall diameter = 12.25 in
Overall height = 4.50 in (est.)
Radius concave surface = 39.0 in

109

 PTFE is 15% GF; contact pressure = 6,500 psi

Diameter PTFE sliding disc = 3.00 in
E4. Minimum and Maximum Performance Check E4. Minimum and Maximum Performance Check,
Art. 8 GSID requires the performance of any Example 1.3
isolation system be checked using minimum and For a spherical friction isolator, property modification
maximum values for the effective stiffness of the factors are applied to Qd only.
system. These values are calculated from minimum
and maximum values of Kd and Qd, which are found Minimum Property Modification factors are:
using system property modification factors, as min = 1.0
indicated in Table E4-1.
which means there is no need to reanalyze the bridge
Table E4-1. Minimum and maximum values with a set of minimum values.
for Kd and Qd.
Maximum Property Modification factors are (GSID
Eq. Appendix A):
8.1.2-1 Kd,max = Kd max,Kd (E-13)
GSID max,a = 1.1
Eq. max,c = 1.0
8.1.2-2 Kd,min = Kd min,Kd (E-14) max,tr = 1.2
GSID max,t = 1.2
Eq.
8.1.2-3 Qd,max = Qd max,Qd (E-15)
GSID
Eq.
8.1.2-4 Qd,min = Qd min,Qd (E-16)
GSID

Determination of the system property modification

factors should include consideration of the effects of
temperature, aging, scragging, velocity, travel (wear)
and contamination as shown in Table E4-2. In lieu of
tests, numerical values for these factors can be
obtained from Appendix A, GSID.

Table E4-2. Minimum and maximum values for

system property modification factors.

Eq. min,Kd = (min,t,Kd) (min,a,Kd)

8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-17)
GSID (min,scrag,Kd)
Eq. max,Kd = (max,t,Kd) (max,a,Kd)
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-18)
GSID (max,scrag,Kd)
Eq. min,Qd = (min,t,Qd) (min,a,Qd)
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-19)
GSID (min,scrag,Qd)
Eq. max,Qd = (max,t,Qd) (max,a,Qd)
8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd) (E-20)
GSID (max,scrag,Qd)

Adjustment factors are applied to individual -factors Applying a system adjustment factor of 0.66 for an
(except v) to account for the likelihood of other bridge, the maximum property modification
occurrence of all of the maxima (or all of the factors become:
minima) at the same time. These factors are applied max,a = 1.0 + 0.1(0.66) = 1.066
to all -factors that deviate from unity but only to the max,c = 1.0

110

portion of the -factor that is greater than, or less
than, unity. Art. 8.2.2 GSID gives these factors as
follows:
1.00 for critical bridges
0.75 for essential bridges
0.66 for all other bridges
As required in Art. 7 GSID and shown in Fig. C7-1
GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Qd,max. As indicated in Fig C7-1 GSID, maximum
displacements will probably be given by the first case
(Kd,min and Qd,min) and maximum forces by the second
case (Kd,max and Qd,max).
max,t = 1.0 + 0.2(0.66) = 1.132
max,a = 1.0 + 0.2(0.66) = 1.132
Therefore the maximum overall modification factors
max = 1.066(1.0)(1.132)(1.132) = 1.37
Since the possible variation in upper bound properties
exceeds 15% a reanalysis of the bridge is required to
determine performance with these properties.
The upper-bound properties are:
Qd,max = 1.37 (2.34) = 3.21 k
Kdmax = Kd = 1.17 k/in

E5. Design and Performance Summary, Example

E5. Design and Performance Summary
1.3

E5.1 Isolator dimensions E5.1 Isolator dimensions, Example 1.3

Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E5.1-1.
Overall diameter of isolator
Overall height Table E5.1-1 Isolator Dimensions
Radius of curvature of concave plate
Isolator Overall size Overall size Radius
Diameter of PTFE disc Location including without (in)
PTFE Compound mounting mounting
PTFE contact pressure plates (in) plates (in)
Under
edge 16.25x16.25 12.25 dia. x
39.0
girder x4.5(H) (est) 4.50(H) (est)
on Pier 1

PTFE is 15% Glass-filled; 6,500 psi contact pressure;

3.00 in diameter.

E5.2 Bridge Performance E5.2 Bridge Performance, Example 1.3

Summarize bridge performance
Maximum superstructure displacement
(longitudinal)
Maximum superstructure displacement
(transverse)
Maximum superstructure displacement
(resultant)
Maximum column shear (resultant)
Maximum column moment (about transverse
axis)
Maximum column moment (about
longitudinal axis)
Maximum column torque
Bridge performance is summarized in Table E5.2-1
where it is seen that the maximum column shear is
18.03 k. This less than the column plastic shear (25 k)
and therefore the required performance criterion is
satisfied (fully elastic behavior). Furthermore the
maximum longitudinal displacement is 1.66 in which
is less than the 2.0 in available at the abutment
expansion joints and is therefore acceptable.

Check required performance as determined in Step

A3, is satisfied.

111

 Table E5.2-1 Summary of Bridge Performance

Maximum superstructure
1.66 in
displacement (longitudinal)
Maximum superstructure
1.54 in
displacement (transverse)
Maximum superstructure
1.72 in
displacement (resultant)
Maximum column shear (resultant) 18.03 k
Maximum column moment about
242 kft
transverse axis
Maximum column moment about
121 kft
longitudinal axis
Maximum column torque 1.82 kft

112

 SECTION 1
PC Girder Bridge, Short Spans, Multi-Column Piers

DESIGN EXAMPLE 1.4: Eradiquake Isolators

Design Examples in Section 1

Site
ID Description S1 Pier height Skew Isolator type
Class
1.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
1.1 Change site class 0.2g D Same 0 Lead rubber bearing
Change spectral
1.2 0.6g B Same 0 Lead rubber bearing
acceleration, S1
Spherical friction
1.3 Change isolator to SFB 0.2g B Same 0
bearing
1.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
1.5 Change column height 0.2g B H1=0.5 H2 0 Lead rubber bearing
1.6 Change angle of skew 0.2g B Same 450 Lead rubber bearing

113

DESIGN PROCEDURE
STEP A: BRIDGE AND SITE DATA
A1. Bridge Properties
Determine properties of the bridge:
number of supports, m
number of girders per support, n
angle of skew
weight of superstructure including railings,
curbs, barriers and to the permanent loads,
WSS
weight of piers participating with
superstructure in dynamic response, WPP
weight of superstructure, Wj, at each
support
piers heights (clear dimensions)
stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge
column flexural yield strength (minimum
value)
column shear strength (minimum value)
allowable movement at expansion joints
isolator type if known, otherwise to be
determined
A2. Seismic Hazard
Determine seismic hazard at site:
acceleration coefficients
site class and site factors
seismic zone
Plot response spectrum.
Use Art. 3.1 GSID to obtain peak ground and spectral
acceleration coefficients. These coefficients are the
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (e.g. Figures
3.10.2.1-1 to 3.10.2.1-21 LRFD).
Use Art. 3.2 to obtain Site Class and corresponding Site
Factors (Fpga, Fa and Fv). These data are the same as for
DESIGN EXAMPLE 1.4 (EQS Isolators)
A1. Bridge Properties, Example 1.4
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 6
Number of columns per support = 3
Angle of skew = 00
Weight of superstructure including permanent
loads, WSS = 650.52 k
Weight of superstructure at each support:
o W1 = 44.95 k
o W2 = 280.31 k
o W3 = 280.31 k
o W4 = 44.95 k
Participating weight of piers, WPP = 107.16 k
Effective weight (for calculation of period),
Weff = Wss + WPP = 757.68 k
Stiffness of each pier in the longitudinal
direction:
o Ksub,pier1,long = 172.0 k/in
o Ksub,pier2,long = 172.0 k/in
Stiffness of each pier in the transverse direction:
o Ksub,pier1,trans = 687.0 k/in
o Ksub,pier2,trans = 687.0 k/in
Minimum flexural yield strength of one column =
425 kft (plastic moment capacity)
Displacement capacity of expansion joints
(longitudinal) = 2.0 in for thermal and other
movements.
Eradiquake isolators
A2. Seismic Hazard, Example 1.4
Acceleration coefficients for bridge site are given in
design statement as follows:
PGA = 0.40
S1 = 0.20
SS = 0.75
Bridge is on a rock site with shear wave velocity in
upper 100 ft of soil = 3,000 ft/sec.
Table 3.10.3.1-1 LRFD gives Site Class as B.
Tables 3.10.3.2-1, -2 and -3 LRFD give following
Site Factors:
Fpga = 1.0
Fa = 1.0
Fv = 1.0
114

conventional bridges and Art 3.2 refers the designer to

the corresponding articles in the LRFD Specifications,
i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3,
LRFD.

Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give

modified spectral acceleration coefficients that include
site effects as follows: As = Fpga PGA = 1.0(0.40) = 0.40
As = Fpga PGA SDS = Fa SS = 1.0(0.75) = 0.75
SDS = Fa SS SD1 = Fv S1 = 1.0(0.20) = 0.20
SD1 = Fv S1

Seismic Zone is determined by value of SD1 in Since 0.15 < SD1 < 0.30, bridge is located in Seismic
accordance with provisions in Table5-1 GSID. Zone 2.

These coefficients are used to plot design response Design Response Spectrum is as below:
spectrum as shown in Fig. 4-1 GSID.
0.9

0.8

0.7

0.6

Acceleration
0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Period(s)

A3. Performance Requirements A3. Performance Requirements, Example 1.4

Determine required performance of isolated bridge
during Design Earthquake (1000-yr return period).
Examples of performance that might be specified by
the Owner include:
Reduced displacement ductility demand in
columns, so that bridge is open for emergency
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in
columns or yield elsewhere in bridge), so that
bridge is fully functional and open to all vehicles
immediately following earthquake.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.
In this example, assume the owner has specified full
functionality following the earthquake and therefore
the columns must remain elastic (no yield).
To remain elastic the maximum lateral load on the
pier must be less than the load to yield the column.
This load is taken as the plastic moment capacity
(strength) of the column (425 kft, see above) divided
by the overall column height (17 ft). This calculation
assumes the column is acting as a simple cantilever
in single curvature in the longitudinal direction.
Hence load to yield column = 425 /17 = 25.0 k
The maximum shear in the column must therefore be
less than 25 k in order to keep the column elastic and
meet the required performance criterion.

115

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for Isolator Force, F

non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most have
bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use Ku
equivalent linear springs and viscous damping to
represent the isolators, so that linear methods of analysis
disol
may be used to determine response. Since equivalent Ku
dy Isolator
Ku Displacement, d
properties such as Kisol are dependent on displacement (d),
and the displacements are not known at the beginning of
the analysis, an iterative approach is required. Note that in Kd
Art 7.1, GSID, keff is used for the effective stiffness of an
isolator unit and Keff is used for the effective stiffness of a disol = Isolator displacement
dy = Isolator yield displacement
combined isolator and substructure unit. To minimize Fisol = Isolator shear force
confusion, Kisol is used in this document in place of keff. Fy = Isolator yield force
Kd = Post-elastic stiffness of isolator
There is no change in the use of Keff and Keff,j, but Ksub is Kisol = Effective stiffness of isolator
used in place of ksub. Ku
Qd
= Loading and unloading stiffness (elastic stiffness)
= Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement (disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 1.4
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.20 2.0
GSID

(2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not
occur under non-seismic loads (e.g. wind) but

116

 low enough that yield will occur during an

earthquake. Experience has shown that
taking Qd to be 5% of the bridge weight is a
good starting point, i.e.
0.05 (B-2)
(3) Post-yield stiffness, Kd 0.05 0.05 650.52 32.53
Art 12.2 GSID requires that all isolators
exhibit a minimum lateral restoring force at
the design displacement, which translates to a
minimum post yield stiffness Kd,min given by
Eq. B-3.

Art. 650.52
0.025 0.05 0.05 16.26 /
12.2 , (B-3) 2.0
GSID
Experience has shown that a good starting
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 1.4
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength, ,

Qd, and stiffness, Kd, values in proportion to the dead o Qd, 1 = 2.25 k
load applied at that support: o Qd, 2 = 14.02 k
o Qd, 3 = 14.02 k
o Qd, 4 = 2.25 k

, (B-4)
and
and
,
, (B-5) o Kd,1 = 1.12 k/in
o Kd,2 = 7.01 k/in
o Kd,3 = 7.01 k/in
o Kd,4 = 1.12 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 1.4
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of , ,
the isolators at support j (Kisol,j) and the stiffness of , ,
the substructure Ksub,j. See figure below for
definitions (after Fig. 7.1-1 GSID). o 1 = 2.25x10-4
o 2 = 8.49x10-2
An expression for Keff,j, is given in Eq.7.1-6 GSID, but o 3 = 8.49x10-2
a more useful formula as follows (MCEER 2006): o 4 = 2.25x10-4
,
, (B-6)
1 ,
where ,
1
, ,
(B-7)
, , o Keff,1 = 2.25 k/in
and Ksub,j for the piers are given in Step A1. For the o Keff,2 = 13.47 k/in
abutments, take Ksub,j to be a large number, say 10,000 o Keff,3 = 13.47 k/in
k/in, unless actual stiffness values are available. Note o Keff,4 = 2.25 k/in
that if the default option is chosen, unrealistically high

117

 values for Ksub,j will give unconservative results for

column moments and shear forces.

F
Kd
Qd Kisol

dy disol
Superstructure
Isolator Effective Stiffness, Kisol

F
Isolator(s), Kisol
Ksub

Substructure, Ksub
dsub

Substructure Stiffness, Ksub

dsub disol F

d
Keff

d = disol + dsub

Combined Effective Stiffness, Keff

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 1.4
Calculate the total effective stiffness, Keff, of the
bridge:

Eq. , 31.43 /
7.1-6 , ( B-8)
GSID

B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 1.4
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1

(B-9) o disol,1 = 2.00 in

,
1 o disol,2 = 1.84 in
o disol,3 = 1.84 in
o disol,4 = 2.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the effective stiffness of the isolation system Example 1.4
at support j, Kisol,j, for all supports:
,
, ,
,
,
(B-10) o Kisol,1 = 2.25 k/in
, ,
, o Kisol,2 = 14.61 k/in
o Kisol,3 = 14.61 k/in
o Kisol,4 = 2.25 k/in

118

B1.7 Substructure Displacement at Each Support
B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j,
Example 1.4
for all supports:
, ,

, , (B-11)
o dsub,1 = 4.49x10-4 in
o d sub,2 = 1.57x10-1 in
o d sub,3 = 1.57x10-1 in
o d sub,4 = 4.49x10-4 in
B1.8 Lateral Load in Each Substructure Support B1.8 Lateral Load in Each Substructure, Example
Calculate the shear at support j, Fsub,j, for all 1.4
supports:
, , , (B-12) , , ,

where values for Ksub,j are given in Step A1. o F sub,1 = 4.49 k
o F sub,2 = 26.93 k
o F sub,3 = 26.93 k
o F sub,4 = 4.49 k

B1.9 Column Shear Force at Each Support B1.9 Column Shear Force at Each Support,
Calculate the shear in column k at support j, Fcol,j,k, Example 1.4
assuming equal distribution of shear for all columns at
support j:
,
, ,
#
,
, ,
# (B-13) o F col,2,1-3 8.98 k
o F col,3,1-3 8.98 k

Use these approximate column shear forces as a check These column shears are less than the yield capacity
on the validity of the chosen strength and stiffness of each column (25 k) as required in Step A3 and the
characteristics. chosen strength and stiffness values in Step B1.1 are
therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 1.4
damping ratio, , of the bridge:

Eq. 757.68
7.1-5 2 (B-14) 2 2
GSID 386.4 31.43

and = 1.57 sec

Eq. 2 , , , and taking dy,j = 0:

(B-15)
7.1-10 , , ,
GSID 2 , , 0
0.31
, , ,
where dy,j is the yield displacement of the isolator. For
friction-based isolators, dy,j = 0. For other types of
isolators dy,j is usually small compared to disol,j and has
negligible effect on , Hence it is suggested that for
the Simplified Method, set dy,j = 0 for all isolator

119

types. See Step B2.2 where the value of dy,j is revisited
for the Multimode Spectral Analysis Method.
B1.11 Damping Factor
Calculate the damping factor, BL, and the
displacement, d, of the bridge:
Eq. .
, 0.3
.
7.1-3 (B-16)
1.7, 0.3
GSID
Eq. 9.79 (B-17)
7.1-4
GSID
B1.12 Convergence Check
Compare the new displacement with the initial value
assumed in Step B1.1. If there is close agreement, go
to the next step; otherwise repeat the process from
Step B1.3 with the new value for displacement as the
assumed displacement.
This iterative process is amenable to solution using a
spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
B1.11 Damping Factor, Example 1.4
Since 0.31 0.3
1.70
and
9.79 9.79 0.2 1.57
1.81
1.70
B1.12 Convergence Check, Example 1.4
Since the calculated value for displacement, d
(=1.81 in) is not close to that assumed at the
beginning of the cycle (Step B1.1, d = 2.0), use the
value of 1.81 in as the new assumed displacement and
repeat from Step B1.3.
After one iteration, convergence is reached at a
superstructure displacement of 1.76 in, with an
effective period of 1.52 seconds, and a damping factor
of 1.70 (33% damping ratio). The displacement in the
isolators at Pier 1 is 1.61 in and the effective stiffness
of the same isolators is 15.69 k/in.
See spreadsheet in Table B1.12-1 for results of final
iteration.
Ignoring the weight of the cap beams, the sum of the
column shears at a pier must equal the total isolator
shear force. Hence, approximate column shear (per
column) = 15.69(1.61)/3 = 8.42 k which is less than
the maximum allowable (25k) if elastic behavior is to
be achieved (as required in Step A3).
Also the superstructure displacement = 1.76 in, which
is less than the available clearance of 2.0 in.
Therefore the above solution is acceptable and go to
Step B2.
Note that available clearance (2.0 in) is greater than
minimum required which is given by:
8 8 0.20 1.52
1.43
1.7
120

 Table B1.12-1 Simplified Method Solution for Design Example 1.4 Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
650.52 107.16 757.68 0.2 6

StepB1.1 d 1.72 Assumeddisplacement

Qd 32.53 Characteristicstrength
Kd 16.26 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 44.95 2.25 1.12 10000 2.40E04 2.40 1.76 2.40 4.23E04 4.23 3.958 7.444
Pier1 280.31 14.02 7.01 172.0 9.12E02 14.38 1.61 15.69 1.47E01 25.33 22.623 44.611
Pier2 280.31 14.02 7.01 172.0 9.12E02 14.38 1.61 15.69 1.47E01 25.33 22.623 44.611
Abut2 44.95 2.25 1.12 10000 2.40E04 2.40 1.76 2.40 4.23E04 4.23 3.958 7.444
Total 650.52 32.526 16.263 K eff,j 33.557 59.107 53.161 104.109
Step B1.4

StepB1.10 T eff 1.52 Effectiveperiod

0.33 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.75

B L 1.70 DampingFactor
d 1.75 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 0.375 0.187 0.400 1.66 0.413
Pier1 2.336 1.168 2.615 1.47 2.757
Pier2 2.336 1.168 2.615 1.47 2.757
Abut2 0.375 0.187 0.400 1.66 0.413

121

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and
then applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 1.4

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 (see
elastic stiffness, Kd,i, of each isolator i as follows: Table B1.12-1) by the number of isolators at each
, support (n = 6), the following values for Qd /isolator
, (B-19)
and Kd /isolator are obtained:
and
, o Qd, 1 = 2.25/6 = 0.37 k
, ( B-20)
o Qd, 2 = 14.02/6= 2.34 k
where values for Qd,j and Kd,j are obtained from the o Qd, 3 = 14.02/6= 2.34 k
final cycle of iteration in the Simplified Method (Step o Qd, 4 = 2.25/6 = 0.37 k
B1. 12 and
o Kd,1 = 1.12/6 = 0.19 k/in
o Kd,2 = 7.01/6 = 1.17 k/in
o Kd,3 = 7.01/6 = 1.17 k/in
o Kd,4 = 1.12/6 = 0.19 k/in

B2.2 Initial Stiffness and Yield Displacement B2.2 Initial Stiffness and Yield Displacement,
Calculate the initial stiffness, Ku,i, and the yield Example 1.4
displacement, dy,i, for each isolator i as follows: Since the isolator type has been specified in Step A1
(1) For friction-based isolators: Ku,i = and dy,i = 0. to be an elastomeric bearing (i.e. not a friction-based
(2) For other types of isolators, and in the absence of bearing), calculate Ku,i and dy,i for an isolator on Pier 1
isolator-specific information, take as follows:
, 10 , 10 1.17 11.7 /
, 10 , ( B-21)
and then and
, 2.34
,
( B-22) , 0.22
, , , 11.7 1.17
, ,

As expected, the yield displacement is small

compared to the expected isolator displacement (~2
in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.

B2.3 Isolator Effective Stiffness, Kisol,i B2.3 Isolator Effective Stiffness, Kisol,i, Example 1.4
Calculate the isolator stiffness, Kisol,i, of each isolator Dividing the results for Kisol (Step B1.12) among the 6
i: isolators at each support, the following values for Kisol
/isolator are obtained:
, o Kisol,1 = 2.40/6 = 0.40 k/in
, (B -23)
o Kisol,2 = 15.69/6 = 2.62 k/in
o Kisol,3 = 15.69/6 = 2.62 k/in
o Kisol,4 = 2.40/6 = 0.40 k/in

122

B2.4 Three-Dimensional Bridge Model B2.4 Three-Dimensional Bridge Model, Example

Using computer-based structural analysis software, 1.4
create a 3-dimensional model of the bridge with the Although the bridge in this Design Example is regular
isolators represented by spring elements. The and is without skew or curvature, a 3-dimensional
stiffness of each isolator element in the horizontal finite element model was developed for this Step, as
axes (Kx and Ky in global coordinates, K2 and K3 in shown below.
typical local coordinates) is the Kisol value calculated
in the previous step. For bridges with regular
geometry and minimal skew or curvature, the
superstructure may be represented by a single stick
provided the load path to each individual isolator at
each support is explicitly modeled, usually by a rigid
cap beam and a set of rigid links. If the geometry is
irregular, or if the bridge is skewed or curved, a finite
element model is recommended to accurately capture
the load carried by each individual isolator. If the
piers have an unusual weight distribution, such as a .
pier with a hammerhead cap beam, a more rigorous
model is justified.

B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in
Step A2 is for a 5% damped system. It is modified in
this step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the
effective period of the bridge, Teff, by the damping
factor, BL.
B.2.5 Composite Design Response Spectrum,
Example 1.4
From the final results of Simplified Method (Step
B1.12), BL = 1.70 and Teff = 1.52 sec. Hence the
transition in the composite spectrum from 5% to 33%
damping occurs at 0.8 Teff = 0.8 (1.46) = 1.22 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.22 sec by 1.70.
0.9

0.8

0.7
Acceleration(g)

0.6

0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Period(s)

B2.6 Multimodal Analysis of Finite Element
Model
Input the composite response spectrum as a user-
specified spectrum in the software, and define a load
case in which the spectrum is applied in the
longitudinal direction. Analyze the bridge for this
load case.
B2.6 Multimodal Analysis of Finite Element
Model, Example 1.4
Results of modal analysis of the example bridge are
summarized in Table B2.6-1 Here the modal periods
and mass participation factors of the first 12 modes
are given. The first two modes are the principal
longitudinal and transverse modes with periods of
1.41 and 1.35 sec respectively. The period of the
longitudinal mode (1.41 sec) is close to that
calculated in the Simplified Method.

123

 Table B2.6-1 Modal Properties of Bridge

Example 1.4 First Iteration
Mode Period ModalParticipatingMassRatios
No. Sec UX UY UZ RX RY RZ
1 1.410 0.761 0.000 0.000 0.000 0.003 0.000
2 1.346 0.000 0.738 0.031 0.059 0.000 0.534
3 1.325 0.000 0.000 0.000 0.000 0.000 0.217
4 0.187 0.000 0.000 0.000 0.000 0.000 0.000
5 0.186 0.125 0.000 0.000 0.000 0.000 0.000
6 0.121 0.000 0.000 0.000 0.000 0.000 0.000
7 0.121 0.000 0.000 0.000 0.000 0.000 0.006
8 0.104 0.000 0.034 0.183 0.107 0.000 0.064
9 0.101 0.000 0.000 0.000 0.000 0.226 0.000
10 0.095 0.000 0.121 0.081 0.184 0.000 0.041
11 0.094 0.000 0.000 0.000 0.000 0.000 0.000
12 0.074 0.000 0.002 0.000 0.003 0.000 0.000

Computed values for the isolator displacements due to

a longitudinal earthquake are as follows (numbers in
parentheses are those used to calculate the initial
properties to start iteration from the Simplified
Method):

o disol,1 = 1.65 (1.76) in

o disol,2 = 1.47 (1.61) in
o disol,3 = 1.47 (1.61) in
o disol,4 = 1.65 (1.76) in

B2.7 Convergence Check B2.7 Convergence Check, Example 1.4

Compare the resulting displacements at the The new superstructure displacement is 1.65 in, more
superstructure level (d) to the assumed than a 5% difference from the displacement assumed
displacements. These displacements can be obtained at the start of the Multimode Spectral Analysis.
by examining the joints at the top of the isolator
spring elements. If in close agreement, go to Step
B2.9. Otherwise go to Step B2.8.

B2.8 Update Kisol,i, Keff,j, and BL B2.8 Update Kisol,i, Keff,j, and BL, Example 1.4
Use the calculated displacements in each isolator Updated values for Kisol,i are given below (previous
element to obtain new values of Kisol,i for each values are in parentheses):
isolator as follows:
, o Kisol,1 = 0.41 (0.40) k/in
, , (B-24) o Kisol,2 = 2.76 (2.62) k/in
,
o Kisol,3 = 2.76 (2.62) k/in
Recalculate Keff,j : o Kisol,4 = 0.41 (0.40) k/in

Eq. , ,
Updated values for Keff,j, , BL and Teff are given below
7.1-6 ,
(B-25) (previous values are in parentheses):
, ,
GSID
o Keff,1 = 2.48 (2.40) k/in
Recalculate system damping ratio, : o Keff,2 = 15.48 (14.38) k/in
o Keff,3 = 15.48 (14.38) k/in
Eq. 2 , , ,
o Keff,4 = 2.48 (2.40) k/in
7.1-10 , , ,
(B-26)
GSID o = 27% (33%)
o BL = 1.66 (1.70)
Recalculate system damping factor, BL: o Teff = 1.41 (1.52) sec
The updated composite response spectrum is shown
. below:
Eq. 0.3
. ( B-27)
7.1-3 1.7 0.3

124

 GSID 0.9

0.8

Obtain the effective period of the bridge from the 0.7

Acceleration(g)
multi-modal analysis and with the revised damping 0.6

factor (Eq. B-27), construct a new composite 0.5

response spectrum. Go to Step B2.6. 0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Period(s)

B2.6 Multimodal Analysis Second Iteration,

Example 1.4
Reanalysis gives the following values for the isolator
displacements (numbers in parentheses are from the
previous cycle):

o disol,1 = 1.66 (1.65) in

o disol,2 = 1.47 (1.47) in
o disol,3 = 1.47 (1.47) in
o disol,4 = 1.66 (1.65) in

B2.7 Convergence Check B2.7 Convergence Check, Example 1.4

Compare the resulting displacements at the
superstructure level (d) to the assumed
displacements. These displacements can be obtained
by examining the joints at the top of the isolator
spring elements. If in close agreement, go to Step
B2.9. Otherwise go to Step B2.8.
The new superstructure displacement is 1.66 in, less
than a 1% difference from the displacement assumed
at the start of the second cycle of Multimode Spectral
Analysis.

B2.9 Superstructure and Isolator Displacements
Once convergence has been reached, obtain
o superstructure displacements in the
longitudinal (xL) and transverse (yL) directions
of the bridge, and
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions of the bridge, for
each isolator, for this load case (i.e.
longitudinal loading). These displacements
may be found by subtracting the nodal
displacements at each end of each isolator
spring element.
B2.9 Superstructure and Isolator Displacements,
Example 1.4
From the above analysis:
o superstructure displacements in the
longitudinal (xL) and transverse (yL) directions
are:
xL= 1.66 in
yL= 0.0 in
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions are:
o Abutments: uL = 1.66 in, vL = 0.00 in
o Piers: uL = 1.47 in, vL = 0.00 in

All isolators at same support have the same

displacements.

B2.10 Pier Bending Moments and Shear Forces
Obtain the pier bending moments and shear forces in
the longitudinal (MPLL, VPLL) and transverse (MPTL,
VPTL) directions at the critical locations for the
longitudinally-applied seismic loading.
B2.10 Pier Bending Moments and Shear Forces,
Example 1.4
Maximum bending moments and shear forces in the
columns in the longitudinal (MPLL,VPLL) and
transverse (MPTL,VPTL) directions are:
Exterior Columns:
o MPLL= 0 kft

125

 o MPTL= 240 kft

o VPLL= 15.81 k
o VPTL= 0 k
Interior Columns:
o MPLL= 0 kft
o MPTL= 235 kft
o VPLL= 15.24 k
o VPTL= 0 k

Both piers have the same distribution of bending

moments and shear forces among the columns.

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 1.4
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1

Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces
ator
ure to long. to long. due to
EQ EQ long. EQ
1 0.69 0.00 0.36
2 0.69 0.00 0.39

Abut 3 0.69 0.00 0.39

ment 4 0.69 0.00 0.39
5 0.69 0.00 0.39
6 0.69 0.00 0.36
1 4.04 0.00 0.13
2 4.05 0.00 0.19
3 4.05 0.00 0.22
Pier
4 4.05 0.00 0.22
5 4.05 0.00 0.19
6 4.04 0.00 0.13

126

 STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake C1. Analysis for Transverse Earthquake,
Repeat the above process, starting at Step B1, for Example 1.4
earthquake loading in the transverse direction of the Key results from repeating Steps B1 and B2
bridge. Support flexibility in the transverse direction (Simplified and Mulitmode Spectral Methods) are:
is to be included, and a composite response spectrum o Teff = 1.43 sec
is to be applied in the transverse direction. Obtain o Superstructure displacements in the longitudinal
isolator displacements in the longitudinal (uT) and (xT) and transverse (yT) directions are as follows:
transverse (vT) directions of the bridge, and the xT = 0 and yT = 1.53 in
biaxial bending moments and shear forces at critical o Isolator displacements in the longitudinal (uT)
locations in the columns due to the transversely- and transverse (vT) directions are as follows:
applied seismic loading. Abutments uT = 0.00 in, vT = 1.53 in
Piers uT = 0.00 in, vT = 1.49 in
o Maximum bending moments and shear forces in
the columns in the longitudinal (MPLL,VPLL) and
transverse (MPTL,VPTL) directions are:
Exterior Columns:
o MPLT= 108 kft
o MPTT= 1 kft
o VPLT= 0.06 k
o VPTT= 14.87 k
Interior Columns:
o MPLT= 120 kft
o MPTT= 0 kft
o VPLT= 0 k
o VPTT= 17.29 k

Both piers have the same distribution of

bending moments and shear forces among the
columns.

o Isolator shear and axial forces are in Table C1-1.

127

 Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces due
ator
ure to transv. to transv. to transv.
EQ EQ EQ
1 0.00 0.65 3.50
2 0.00 0.65 1.93

Abut 3 0.00 0.65 0.68

ment 4 0.00 0.65 0.68
5 0.00 0.65 1.93
6 0.00 0.65 3.50
1 0.02 3.98 12.56
2 0.01 4.00 1.14
3 0.00 4.01 2.29
Pier
4 0.00 4.01 2.29
5 0.01 4.00 1.14
6 0.02 3.98 12.56

128

STEP D. CALCULATE DESIGN VALUES

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements at Pier 1,

Following the provisions in Art. 2.1 GSID, and Example 1.4
illustrated in Fig. 2.1-1 GSID, calculate the total To illustrate the process, design displacements for the
design displacement, dt, for each isolator by outside isolator on Pier 1 are calculated below.
combining the displacements from the longitudinal
(uL and vL) and transverse (uT and vT) cases as Load Case 1:
follows: u1 = uL + 0.3uT = 1.0(1.47) + 0.3(0) = 1.47 in
u1 = uL + 0.3uT (D-1) v1 = vL + 0.3vT = 1.0(0) + 0.3(1.49) = 0.45 in
v1 = vL + 0.3vT (D-2) R1 = = 1.47 0.45 = 1.54 in
R1 = (D-3)
Load Case 2:
u2 = 0.3uL + uT (D-4) u2 = 0.3uL + uT = 0.3(1.47) + 1.0(0) = 0.44 in
v2 = 0.3vL + vT (D-5) v2 = 0.3vL + vT = 0.3(0) + 1.0(1.49) = 1.49in
R2 = (D-6) R2 = = 0.44 1.49 = 1.55in

dt = max(R1, R2) (D-7)

D2. Design Moments and Shears
Calculate design values for column bending moments
and shear forces for all piers using the same
combination rules as for displacements.
Governing Case:
Total design displacement, dt = max(R1, R2)
= 1.55in
D2. Design Moments and Shears in Pier 1,
Example 1.4
Design moments and shear forces are calculated for
Pier 1, Column 1, below to illustrate the process.

Alternatively this step may be deferred because the Load Case 1:

above results may not be final. Upper and lower VPL1= VPLL + 0.3VPLT = 1.0(15.81) + 0.3(0.06)
bound analyses are required after the isolators have = 15.83 k
been designed as described in Art 7. GSID. These VPT1= VPTL + 0.3VPTT = 1.0(0.02) + 0.3(14.87)
analyses are required to determine the effect of = 4.48 k
possible variations in isolator properties due age, R1 = = 15.83 4.48 = 16.45 k
temperature and scragging in elastomeric systems.
Accordingly the results for column shear in Steps Load Case 2:
B2.10 and C are likely to increase once these VPL2= 0.3VPLL + VPLT = 0.3(15.81) + 1.0(0.06)
analyses are complete. = 4.80 k
VPT2= 0.3VPTL + VPTT = 0.3(0.02) + 1.0(14.87)
= 14.88 k
R2 = = 4.80 14.88 = 15.63 k

Governing Case:
Design column shear = max (R1, R2)
= 16.45 k

129

STEP E. DESIGN OF ERADIQUAKE ISOLATORS

An EradiQuake Isolator (EQS) is a sliding
isolation bearing composed of a multi-
directional sliding disc bearing and lateral
springs. Each spring assembly consists of a
cylindrical polyurethane spring and a spring
piston. The piston keeps the spring straight as
the isolator moves in different directions. The
disc bearing and springs are housed in a
mirror-finished stainless steel lined box.

The required values for Qd and Kd determine

the coefficient of friction at the sliding
interface and the properties of the springs.

The sliding interface is typically comprised of

stainless steel and PTFE. Energy is dissipated
during sliding while the springs provide a
restoring force. PTFE is an attractive material
in that at sliding slow speeds it has a low
coefficient of friction which is ideal for
accommodating thermal effects, and at higher
speeds the friction becomes greater and acts as
an effective energy dissipator during seismic events. The polyurethane springs are designed such that they are
never in tension. Their basic design and composition is derived from the die-spring industry.

Design and materials conform to the LRFD Specifications. Steel components are designed in accordance with
Section 6, while the disc bearing is designed and constructed per Section 14.

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

Notation
A Area Kd Stiffness when sliding (Total spring rate)
A1 Area based on dead load k1 Stiffness (spring rate) for one spring
A2 Area based on total load L Length
AS Spring area LGB Guide bar length
BBB Bearing block plan dimension LS Spring length
BBox Guide box plan dimension (out to out LSI Installed spring length
dimension of guide bars) LL Live Load
BBP Base plate length L1 Spring length based on max long term
BPTFE PTFE dimension displacement
BSP Slide plate (guide box top) length L2 Spring length based on max short term
DD Disc outer diameter displacement
DPTFE PTFE diameter M Moment
DS Spring outer diameter MN Factored moment
dL Service (long term) displacement PDL Dead load
dT Total seismic displacement PLL Live load
E Elastic modulus PSL Seismic live load
F Spring force PWL Wind load
FY Yield stress Qd Characteristic strength
H Isolator height SG Gross shape factor of disc
IDS Spring inner diameter TBB Bearing block thickness

130

TBP Base plate thickness Bearing rotation

TD Disc thickness Inner to outer diameter ratio
TGB Guide bar thickness W Wind displacement
TSP Slide plate thickness C Maximum average compression strain
W Isolator weight Coefficient of friction
W Plan width of isolator

WBP Base plate length
WL Wind load
Z Plastic modulus

E1. Required Properties E1. Required Properties, Example 1.4

Obtain from previous work the properties required of The design of one of the exterior isolators on a pier is
the isolation system to achieve the specified given below to illustrate the design process for an
performance criteria (Step A1). EQS isolator.
the required characteristic strength, Qd, per
isolator From previous work:
the required post-elastic stiffness, Kd, per Qd/isolator = 2.34 k
isolator Kd/isolator = 1.17 k/in
the total design displacement, dt, for each Total design displacement, dt = 1.55 in
isolator PDL = 45.52 k
the maximum applied dead and live load, PDL, PLL = 15.50 k
PLL, and seismic load, PSL,which includes PSL = 12.56 k
seismic live load (if any) and overturning
forces due to seismic loads, at each isolator Calculated for this design:
the wind load per isolator, PWL, and PWL = 1.76 k
the thermal displacement of the superstructure dL = +/- 0.25 in
at each isolator, dL.

E2. Isolator Sizing

E2.1 Size the Disc E2.1 Size Disc, Example 1.4
Estimate the disc outer diameter based on an average
compressive stress of 4.5 ksi using the gross plan
area.
4
4.5

(E-1)

0.53
0.5362 4.17 4.50
Estimate disc thickness based on a gross shape factor
of 2.4:
4.5
(E-2) 0.47 0.50
4 4 2.4

Check rotational capacity and adjust disc thickness if

required. Use standard disc design rotation, , of

0.02 radians, and a maximum compression strain, c,
of 0.10 for this calculation.

0.02 4.50
(E-3)
0.45
2 2 2 0.10

131


E2.2 Size the Springs E2.2 Size the Springs, Example 1.4

E2.2.1 Calculate Installed Spring Length E2.2 Calculate Installed Spring Length, Example
Assume 60% max compressive strain on the MER 1.4
spring for short term loading, 40% max compressive
strain for long term loading. Add 20% of long term
loading strain for elastomer compression set.
Then
2.5 (E-4) 2.5 0.25 0.63

And using a load combination of two times seismic

total design displacement (for Seismic Zone 2) plus
50% service (thermal):

2 0.5 2 1.55 0.5 0.25

(E-5) 5.38
0.60 0.60

Then required spring length is given by:

max , (E-6) 5.38

E2.2.2 Check Wind Displacements E2.2.2 Check Wind Displacements, Example 1.4
Calculate displacement due to wind as follows:

0.25 1.8 0.25 2.3

(E-7) 1.02
1.2
If the displacement due to wind is too large, add
spring precompression equal to the wind In this example, the wind displacement is acceptable
displacement to the spring length in the transverse and no adjustment of spring length is required.
direction. Precompression doubles the stiffness over
the precompressed displacement. If the wind
displacements are still too large, consider increasing
the spring stiffness in the transverse direction, or
using sacrificial shear keys.

E2.2.3 Calculate Spring Diameter E2.2.3 Calculate Spring Diameter, Example 1.4
Assume only one spring per side is used to meet
spring rate requirements, i.e. let k1=Kd, and take the
elastic modulus for polyurethane spring to be 6.0 ksi.
Since Since Kd = 1.17 k/in

6.0 1.2 5.38

(E-8) 1.08
6.0
it follows that
(E-9) Take initial value for = 0.20 and then
6.0
and
4 1.08
4 1.19 1.25
(E-10) 1 0.2
1

132

E2.2.4 Adjust Spring Length Using Nominal E2.2.4 Adjust Spring Length Using Nominal
Diameters Diameters, Example 1.4
For manufacturing purposes it is advantageous to use Use 1-1/4 in for the spring OD, and 7/16 in for the
standard diameters and adjust the spring length spring ID then
according to the actual value of to fine tune the
stiffness (spring rate).

0.44
(E-11) 0.35
1.25

1 (E-12) 1.25 1 0.35 1.08

4 4

(E-13) 6.0 1.08

5.38
1.20

Check that Ls is greater than either L1 and L2

max ,

Note that LS is the installed spring length. The actual

size of the springs may be slightly different than the
installed size. Springs are pre-compressed to provide
additional wind resistance if needed and account for
compression set in the elastomer.

E2.3 Size the PTFE Pad E2.3 Size the PTFE Pad, Example 1.4

E2.3.1 Calculate Coefficient of Friction E2.3.1 Calculate Required Coefficient of Friction,

Calculate the required coefficient of friction from: Example 1.4

(E-14)
Select PTFE and polished stainless steel as the
sliding surfaces. Low coefficients of friction are
possible with these materials at high contact stresses.
In general the friction coefficient decreases with
increasing pressure.
2.3
0.05
46
A value of 0.05 is lower than the dry PTFE sliding
material can achieve at design pressures. Two
alternatives are available: (1) design with a higher Qd
and then reanalyzing bridge response, and (2) use
EQS bearings with lubricated surfaces at some
isolator locations to reduce the global coefficient of
friction. However, because displacements are small,
two pieces of PTFE can be used, one dimpled and
lubricated ( 0.02 ), the other dry ( 0.07 ).
The dry PTFE area will need to comprise 60% of the
total area to achieve an overall coefficient of 0.05,
assuming the same contact stress across both pieces of
PTFE.

E2.3.2 Calculate Required Area of PTFE E2.3.2 Calculate Required Area of PTFE, Example
Calculate required area of PTFE using allowable 1.4
contact stresses in GSID Table 16.4.1-1. For service
loads (i.e. dead load) allowable average stress is 3.5
ksi, and then:

46
(E-15) 13.1
3.5 3.5
Check area required under dead plus live load using

133

an allowable average stress of 4.5 ksi (as permitted in

LRFD Sec 14.)
46 16
(E-16) 13.8
4.5 4.5
Then required area is
max , (E-17) max , 13.8

Since the structure design rotation of 0.01 radians is

only one-half of the disc design rotation, the limits on
the PTFE edge contact stresses (GSID Table 16.4.1-
1) do not govern.

E2.3.3 Calculate Size of PTFE Pad E2.3.3 Calculate Size of PTFE Pad, Example 1.4
For a circular PTFE pad, the diameter is given by: In this example, two rectangular pieces of PTFE with
different friction coefficients, are being used to
achieve the particularly low coefficient of friction that
4
(E-18a) is required overall. These pieces are separated by a
distance of 2dt to prevent the dry side becoming
lubricated during seismic excitation. The dimensions
For a square PTFE pad, the side dimension is given are such that the two pieces form a square of side
by: BPTFE which is given by:

(E-18b)

1.55 1.55 13.8 5.57 5.63

E2.4 Size the Bearing Block E2.4 Size the Bearing Block, Example 1.4

E2.4.1 Calculate Bearing Plan Dimension E2.4.1 Calculate Bearing Plan Dimension,
Two criteria must be checked to determine the Example 1.4
bearing block plan dimension. The disc must fit
under the block with some clearance, and the PTFE
must fit on top of the block with at least 1/8 in edge
clearance.
1.15 (E-19) 1.15 4.50 5.18

2 0.125 (E-20) 5.63 2 0.125 5.88

max , (E-21) max , 5.88 6.00

E2.4.2 Calculate Bearing Block Thickness E2.4.2 Calculate Bearing Block Thickness,
The thickness of bearing block must be sufficient to Example 1.4
ensure that the springs can be attached on each side
of the block, allowing for a 30% increase in diameter 1.3 1.25 1.63 1.75
upon spring compression.

1.3 (E-22) Size is ok. No need to resize spring diameters.

Note that if TBB is too large, reduce the diameter of

the springs and increase their number.

134

E2.5 Size the Box E2.5 Size the Box, Example 1.4

E2.5 .1 Calculate Guide Bar Thickness E2.5.1 Calculate Guide Bar Thickness, Example
(a) Guide Bar Force 1.4
Guide bars resist the spring forces. They are
modeled as cantilever beams, with the fixed end of
the cantilever located where the guide bar meets the
slide plate. Assume the resisting length of guide bar
to be three times the diameter of the spring. The
moment arm is one-half of the bearing block
thickness, plus 0.20 in. Forces corresponding to two
times the seismic displacement, imposed during
prototype testing, are used to design the guide bar.

2 0.5 (E-23) 1.2 2 1.55 0.5 0.25 3.87

(b) Guide Bar Moment

0.5 0.20 (E-24) 0.5 1.75 0.20 3.87 4.16

Since the effective length of guide bar resisting this

moment is assumed to be 3Ds, the bending moment
per inch of guide bar is:
4.16
(E-25) 1.11 /
3 3 1.25
(c) Guide Bar Thickness
Using a load factor of 1.75, a resistance factor of
1.00, and assuming 50 ksi steel:

1.75 1.00 (E-26)

then
1.75 1.75 1.11
(E-27) 0.039 /
50
But since
(E-28)
4
then
4 (E-29) 4 0.039 0.39 0.50

E2.5.2 Calculate Guide Bar Length E2.5.2 Calculate Guide Bar Length, Example 1.4

2 (E-30) 0.50 6.00 2 5.38 17.26 17.25

E2.5.3 Calculate Guide Bar Width E2.5.3 Calculate Guide Bar Width, Example 1.4

1.75 0.5 0.50 2.0

0.5 (E-31)
E2.5.4 Calculate Size of Box and Slide Plate E2.5.4 Calculate Size of Box and Slide Plate,
Calculate side dimension of box Example 1.4
(E-32)
17.25 0.50 17.75
Choose plan dimension of slide plate equal to, or
slightly larger than, the box, i.e.

135

 Take
(E-33)
18.00

E2.5.5 Calculate Box Top (Slide Plate) Thickness E2.5.5 Calculate Box Top (Slide Plate) Thickness,
Make the slide plate (guide box top) the same Example 1.4
thickness as the guide bars, with a minimum value of
in. 0.75

E2.6 Size the Lower Plate E2.6 Size the Lower Plate, Example 1.4
(a) Thickness
Use inch minimum thickness unless otherwise
required by State DOT specifications. 0.75
(b) Width
Since GSID provisions for prototype testing require
the isolator to be displaced to twice the design
displacement (for Seismic Zone 2), the base plate
must be wide enough to allow such movement
without interference from the anchor bolts.
4 8 (E-34) 17.75 4 1.55 8 31.95 32.0

(c) Length
Take
(E-35) 18.00

(d) Anchor Bolts

Design anchor bolts per LRFD specifications,
increase WBP if necessary.

E3. Design Summary E3 Design Summary, Example 1.4

Overall dimensions of isolator are: Width = WBP = 32.0 in
Width = WBP Length = BSP = 18.00 in
Length = BSP Height is given by:
Height is given by:
0.20 (E-36) 0.75 0.50 1.75 0.75 0.20 3.95

136

E4. Minimum and Maximum Performance Check
Art. 8 GSID requires the performance of any
isolation system be checked using minimum and
maximum values for the effective stiffness of the
system. These values are calculated from minimum
and maximum values of Kd and Qd, which are found
using system property modification factors, as
indicated in Table E4-1.
Table E4-1. Minimum and maximum values
for Kd and Qd.
E4. Minimum and Maximum Performance Check,
Example 1.4
For Eradiquake isolators, Modification Factors are
applied to both Qd and Kd, because both frictional and
elastomeric (urethane) elements are used in these
isolators.
Minimum Property Modification factors are:
min,Kd = 1.0
min,Qd = 1.0

Eq. which means there is no need to reanalyze the bridge

8.1.2-1 Kd,max = Kd max,Kd (E-37) with a set of minimum values.
GSID
Eq. Maximum Property Modification factors are (GSID
8.1.2-2 Kd,min = Kd min,Kd (E-38) Appendix A):
GSID max,a,Kd = 1.0
Eq. max,a,Qd = 1.2
8.1.2-3 Qd,max = Qd max,Qd (E-39)
GSID max,t,Kd = 1.3
Eq. max,t,Qd = 1.5
8.1.2-4 Qd,min = Qd min,Qd (E-40)
GSID max,tr,Kd = 1.0
max,tr,Qd = 1.0
Determination of the system property modification
max,c,Kd = 1.0
factors should include consideration of the effects of
max,c,Qd = 1.1
temperature, aging, scragging, velocity, travel (wear)
and contamination as shown in Table E4-2. In lieu of
tests, numerical values for these factors can be
obtained from Appendix A, GSID.

Table E4-2. Minimum and maximum values for

system property modification factors.

Eq. min,Kd = (min,t,Kd) (min,a,Kd)

8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-41)
GSID (min,scrag,Kd)
Eq. max,Kd = (max,t,Kd) (max,a,Kd)
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-42)
GSID (max,scrag,Kd)
Eq. min,Qd = (min,t,Qd) (min,a,Qd)
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-43)
GSID (min,scrag,Qd)
Eq. max,Qd = (max,t,Qd) (max,a,Qd)
8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd) (E-44)
GSID (max,scrag,Qd)

Adjustment factors are applied to individual -factors
(except v) to account for the likelihood of
occurrence of all of the maxima (or all of the
minima) at the same time. These factors are applied
to all -factors that deviate from unity but only to the
portion of the -factor that is greater than, or less
than, unity. Art. 8.2.2 GSID gives these factors as
follows:
Applying a system adjustment factor of 0.66 for an
other bridge, the maximum property modification
factors become:
max,a,Kd = 1.0 + 0.0(0.66) = 1.00
max,a,Qd = 1.0 + 0.2(0.66) = 1.13

137

 1.00 for critical bridges max,t,Kd = 1.0 + 0.3(0.66) = 1.20

0.75 for essential bridges
0.66 for all other bridges
As required in Art. 7 GSID and shown in Fig. C7-1
GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Qd,max. As indicated in Fig C7-1 GSID, maximum
displacements will probably be given by the first case
(Kd,min and Qd,min) and maximum forces by the second
case (Kd,max and Qd,max).
max,t,Qd = 1.0 + 0.5(0.66) = 1.33
max,tr,Kd = 1.0 + 0.0(0.66) = 1.00
max,tr,Qd = 1.0 + 0.0(0.66) =1.00
max,c,Kd = 1.0 + 0.0(0.66) = 1.00
max,c,Qd = 1.0 + 0.0(0.66) =1.00
Therefore the maximum overall modification factors
max,Kd = (1.00)(1.20)(1.00)(1.00) = 1.20
max,Qd = (1.13)(1.33)(1.00)(1.00) = 1.50

Since the possible variation in upper bound properties

exceeds 15% a reanalysis of the bridge is required to
determine performance with these properties.

The upper-bound properties are:

Qd,max = 1.50(2.34) = 3.51 k
and
Kd,ma x=1.20(1.17) = 1.40 k/in
E5. Design and Performance Summary, Example
E5. Design and Performance Summary
1.4
E5.1 Isolator dimensions E5.1 Isolator dimensions, Example 1.4
Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E5.1-1.
Overall size of lower plate
Overall size of box (top plate) Table E5.1-1 Isolator Dimensions
Overall height
Isolator Overall size Overall size Diam.
Size of disc Location including without Disc
Size of PTFE pad mounting mounting (in)
Number of polyurethane springs plate (in) plate (in)
Diameter of polyurethane springs Under
edge 32.0 x 18.0 17.75x17.75
Check all dimensions with manufacturer. 4.50
girder x 4.0(H) x 4.0(H)
on Pier 1
Isolator Size PTFE No. Diam.
Location pad poly- poly-
(in) urethane urethane
springs springs
(in)
Under
edge
5.63 x 5.63 4 1.25
girder
on Pier 1

E5.2 Bridge Performance E5.2 Bridge Performance, Example 1.4

Summarize bridge performance
Maximum superstructure displacement
(longitudinal)
Maximum superstructure displacement
(transverse)
Maximum superstructure displacement
(resultant)
Maximum column shear (resultant)
Bridge performance is summarized in Table E5.2-1
where it is seen that the maximum column shear is
18.03 k. This less than the column plastic shear (25 k)
and therefore the required performance criterion is
satisfied (fully elastic behavior). Furthermore the
maximum longitudinal displacement is 1.66 in which
is less than the 2.0 in available at the abutment
expansion joints and is therefore acceptable.

138

 Maximum column moment (about transverse Table E5.2-1 Summary of Bridge Performance
axis)
Maximum column moment (about Maximum superstructure
1.66 in
longitudinal axis) displacement (longitudinal)
Maximum column torque Maximum superstructure
1.54 in
displacement (transverse)
Check required performance as determined in Step Maximum superstructure
1.72 in
A3, is satisfied. displacement (resultant)
Maximum column shear (resultant) 18.03 k
Maximum column moment about
242 kft
transverse axis
Maximum column moment about
121 kft
longitudinal axis
Maximum column torque 1.82 kft

139

 SECTION 1
PC Girder Bridge, Short Spans, Multi-Column Piers

DESIGN EXAMPLE 1.5: H1=0.5 H2

Design Examples in Section 1

Site
ID Description S1 Pier height Skew Isolator type
Class
1.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
1.1 Change site class 0.2g D Same 0 Lead rubber bearing
Change spectral
1.2 0.6g B Same 0 Lead rubber bearing
acceleration, S1
Spherical friction
1.3 Change isolator to SFB 0.2g B Same 0
bearing
1.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
1.5 Change column height 0.2g B H1=0.5 H2 0 Lead rubber bearing
1.6 Change angle of skew 0.2g B Same 450 Lead rubber bearing

140


DESIGN PROCEDURE
STEP A: BRIDGE AND SITE DATA
A1. Bridge Properties
Determine properties of the bridge:
number of supports, m
number of girders per support, n
angle of skew
weight of superstructure including railings,
curbs, barriers and to the permanent loads,
WSS
weight of piers participating with
superstructure in dynamic response, WPP
weight of superstructure, Wj, at each
support
stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge. The calculation of these
quantities requires careful consideration of
several factors such as the use of cracked
sections when estimating column or wall
flexural stiffness, foundation flexibility,
and effective column heights.
column shear strength (minimum value).
This will usually be derived from the
minimum value of the column flexural
yield strength, the column height, and
whether the column is acting in single or
double curvature in the direction under
consideration.
allowable movement at expansion joints
isolator type if known, otherwise to be
determined
A2. Seismic Hazard
Determine seismic hazard at site:
acceleration coefficients
site class and site factors
seismic zone
Plot response spectrum.
Use Art. 3.1 GSID to obtain peak ground and spectral
acceleration coefficients. These coefficients are the
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (e.g. Figures
3.10.2.1-1 to 3.10.2.1-21 LRFD).
Use Art. 3.2 to obtain Site Class and corresponding Site
DESIGN EXAMPLE 1.5 (Unequal Pier
Heights: H1=0.5H2)
A1. Bridge Properties, Example 1.5
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 6
Number of columns per support = 3
Angle of skew = 00
Weight of superstructure including permanent
loads, WSS = 650.52 k
Weight of superstructure at each support:
o W1 = 44.95 k
o W2 = 280.31 k
o W3 = 280.31 k
o W4 = 44.95 k
Participating weight of piers, WPP = 107.16 k
Effective weight (for calculation of period),
Weff = Wss + WPP = 757.68 k
Stiffness of each pier in the longitudinal
direction:
o Ksub,pier1,long = 172.0 k/in
o Ksub,pier2,long = 21.5 k/in
Stiffness of each pier in the transverse direction:
o Ksub,pier1,trans = 687.0 k/in
o Ksub,pier2,trans = 86.0 k/in
Minimum column shear strength of shorter
column based on flexural yield capacity of
column = 25 k
Displacement capacity of expansion joints
(longitudinal) = 2.0 in for thermal and other
movements.
Lead rubber isolators
A2. Seismic Hazard, Example 1.5
Acceleration coefficients for bridge site are given in
design statement as follows:
PGA = 0.40
S1 = 0.20
SS = 0.75
Bridge is on a rock site with shear wave velocity in
upper 100 ft of soil = 3,000 ft/sec.
Table 3.10.3.1-1 LRFD gives Site Class as B.
Tables 3.10.3.2-1, -2 and -3 LRFD give following
Site Factors:
Fpga = 1.0
Fa = 1.0
141

Factors (Fpga, Fa and Fv). These data are the same as for Fv = 1.0
conventional bridges and Art 3.2 refers the designer to
the corresponding articles in the LRFD Specifications,
i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3,
LRFD.

Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give

modified spectral acceleration coefficients that include
site effects as follows: As = Fpga PGA = 1.0(0.40) = 0.40
As = Fpga PGA SDS = Fa SS = 1.0(0.75) = 0.75
SDS = Fa SS SD1 = Fv S1 = 1.0(0.20) = 0.20
SD1 = Fv S1

Seismic Zone is determined by value of SD1 in Since 0.15 < SD1 < 0.30, bridge is located in Seismic
accordance with provisions in Table5-1 GSID. Zone 2.

These coefficients are used to plot design response Design Response Spectrum is as below:
spectrum as shown in Fig. 4-1 GSID.
0.9

0.8

0.7

0.6

Acceleration
0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Period(s)

A3. Performance Requirements A3. Performance Requirements, Example 1.5

Determine required performance of isolated bridge In this example, assume the owner has specified full
during Design Earthquake (1000-yr return period). functionality following the earthquake and therefore
the columns must remain elastic (no yield).
Examples of performance that might be specified by
the Owner include: To remain elastic the maximum lateral load on the
Reduced displacement ductility demand in pier must be less than the load to yield any one
columns, so that bridge is open for emergency column (25 k).
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in The maximum shear in any of the shorter columns
columns or yield elsewhere in bridge), so that must therefore be less than 25 k in order to keep
bridge is fully functional and open to all vehicles these columns elastic and meet the required
immediately following earthquake. performance criterion.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.

142

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for Isolator Force, F

non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most have
bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use Ku
equivalent linear springs and viscous damping to
represent the isolators, so that linear methods of analysis
disol
may be used to determine response. Since equivalent Ku
dy Isolator
Ku Displacement, d
properties such as Kisol are dependent on displacement (d),
and the displacements are not known at the beginning of
the analysis, an iterative approach is required. Note that in Kd
Art 7.1, GSID, keff is used for the effective stiffness of an
isolator unit and Keff is used for the effective stiffness of a disol = Isolator displacement
dy = Isolator yield displacement
combined isolator and substructure unit. To minimize Fisol = Isolator shear force
confusion, Kisol is used in this document in place of keff. Fy = Isolator yield force
Kd = Post-elastic stiffness of isolator
There is no change in the use of Keff and Keff,j, but Ksub is Kisol = Effective stiffness of isolator
used in place of ksub. Ku
Qd
= Loading and unloading stiffness (elastic stiffness)
= Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement (disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 1.5
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.20 2.0
GSID

(2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not
occur under non-seismic loads (e.g. wind)

143

 but low enough that yield will occur during

an earthquake. Experience has shown that
taking Qd to be 5% of the bridge weight is a
good starting point, i.e. 0.05 0.05 650.52 32.53
0.05 (B-2)
(3) Post-yield stiffness, Kd
Art 12.2 GSID requires that all isolators
exhibit a minimum lateral restoring force at
the design displacement, which translates to a
minimum post yield stiffness Kd,min given by
Eq. B-3.

Art. 650.52
0.025 0.05 0.05 16.26 /
12.2 , (B-3) 2.0
GSID
Experience has shown that a good starting
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 1.5
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength, ,

Qd, and stiffness, Kd, values in proportion to the dead o Qd, 1 = 2.25 k
load applied at that support: o Qd, 2 = 14.02 k
o Qd, 3 = 14.02 k
o Qd, 4 = 2.25 k

, (B-4)
and
and
,
, (B-5) o Kd,1 = 1.12 k/in
o Kd,2 = 7.01 k/in
o Kd,3 = 7.01 k/in
o Kd,4 = 1.12 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 1.5
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of , ,
the isolators at support j (Kisol,j) and the stiffness of , ,
the substructure Ksub,j. See figure below for
definitions (after Fig. 7.1-1 GSID). o 1 = 2.25x10-4
o 2 = 8.49x10-2
An expression for Keff,j, is given in Eq.7.1-6 GSID, but o 3 = 9.67x10-1
a more useful formula as follows (MCEER 2006): o 4 = 2.25x10-4
,
, (B-6)
1 ,
where ,
1
, ,
(B-7)
, , o Keff,1 = 2.25 k/in
and Ksub,j for the piers are given in Step A1. For the o Keff,2 = 13.47 k/in
abutments, take Ksub,j to be a large number, say 10,000 o Keff,3 = 10.57 k/in
k/in, unless actual stiffness values are available. Note o Keff,4 = 2.25 k/in
that if the default option is chosen, unrealistically high

144

 values for Ksub,j will give unconservative results for

column moments and shear forces.

F
Kd
Qd Kisol

dy disol
Superstructure
Isolator Effective Stiffness, Kisol

F
Isolator(s), Kisol
Ksub

Substructure, Ksub
dsub

Substructure Stiffness, Ksub

dsub disol F

d
Keff

d = disol + dsub

Combined Effective Stiffness, Keff

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 1.5
Calculate the total effective stiffness, Keff, of the
bridge:

Eq. , 28.53 /
7.1-6 , ( B-8)
GSID

B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 1.5
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1

(B-9) o disol,1 = 2.00 in

,
1 o disol,2 = 1.84 in
o disol,3 = 1.02 in
o disol,4 = 2.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the effective stiffness of the isolation system Example 1.5
at support j, Kisol,j, for all supports:
,
, ,
,
,
(B-10) o Kisol,1 = 2.25 k/in
, ,
, o Kisol,2 = 14.61 k/in
o Kisol,3 = 20.79 k/in
o Kisol,4 = 2.25 k/in

145

B1.7 Substructure Displacement at Each Support
B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j,
Example 1.5
for all supports:
, ,

, , (B-11)
o dsub,1 = 4.49x10-4 in
o d sub,2 = 1.57x10-1 in
o d sub,3 = 9.83x10-1 in
o d sub,4 = 4.49x10-4 in
B1.8 Lateral Load in Each Substructure Support B1.8 Lateral Load in Each Substructure, Example
Calculate the shear at support j, Fsub,j, for all 1.5
supports:
, , , (B-12) , , ,

where values for Ksub,j are given in Step A1. o F sub,1 = 4.49 k
o F sub,2 = 26.93 k
o F sub,3 = 21.14 k
o F sub,4 = 4.49 k

B1.9 Column Shear Force at Each Support B1.9 Column Shear Force at Each Support,
Calculate the shear in column k at support j, Fcol,j,k, Example 1.5
assuming equal distribution of shear for all columns at
support j:
,
, ,
#
,
, ,
# (B-13) o F col,2,1-3 8.98 k
o F col,3,1-3 7.05 k

Use these approximate column shear forces as a check These column shears are less than the yield capacity
on the validity of the chosen strength and stiffness of each column (25 k) as required in Step A3 and the
characteristics. chosen strength and stiffness values in Step B1.1 are
therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 1.5
damping ratio, , of the bridge:

Eq. 757.68
7.1-5 2 (B-14) 2 2
GSID 386.4 28.53

and = 1.65 sec

Eq. 2 , , , and taking dy,j = 0:

(B-15)
7.1-10 , , ,
GSID 2 , , 0
0.27
, , ,
where dy,j is the yield displacement of the isolator. For
friction-based isolators, dy,j = 0. For other types of
isolators dy,j is usually small compared to disol,j and has
negligible effect on , Hence it is suggested that for
the Simplified Method, set dy,j = 0 for all isolator

146

types. See Step B2.2 where the value of dy,j is revisited
for the Multimode Spectral Analysis Method.
B1.11 Damping Factor
Calculate the damping factor, BL, and the
displacement, d, of the bridge:
Eq. .
, 0.3
.
7.1-3 (B-16)
1.7, 0.3
GSID
Eq. 9.79 (B-17)
7.1-4
GSID
B1.12 Convergence Check
Compare the new displacement with the initial value
assumed in Step B1.1. If there is close agreement, go
to the next step; otherwise repeat the process from
Step B1.3 with the new value for displacement as the
assumed displacement.
This iterative process is amenable to solution using a
spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
B1.11 Damping Factor, Example 1.5
Since 0.27 0.3
1.67
and
9.79 9.79 0.2 1.65
1.94
1.67
B1.12 Convergence Check, Example 1.5
Since the calculated value for displacement, d
(=1.94 in) is close to that assumed at the beginning of
the cycle (Step B1.1, d = 2.0), use the value of 1.94 in
as the assumed displacement.
See spreadsheet in Table B1.12-1 for results of
simplified method.
Ignoring the weight of the cap beams, the sum of the
column shears at a pier must equal the total isolator
shear force. Hence, approximate column shear (per
column) = 14.87(1.78)/3 = 8.82 k which is less than
the maximum allowable (25k) if elastic behavior is to
be achieved (as required in Step A3).
Also the superstructure displacement = 1.94 in, which
is less than the available clearance of 2.0 in.
Therefore the above solution is acceptable and go to
Step B2.
Note that available clearance (2.0 in) is greater than
minimum required which is given by:
8 8 0.20 1.63
1.95
1.67
147

 Table B1.12-1 Simplified Method Solution for Design Example 1.5 Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
650.52 107.16 757.68 0.2 6

StepB1.1 d 1.94 Assumeddisplacement

Qd 32.53 Characteristicstrength
Kd 16.26 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 44.95 2.25 1.12 10000 2.28E04 2.28 1.94 2.28 4.42E04 4.42 4.351 8.565
Pier1 280.31 14.02 7.01 172.0 8.64E02 13.69 1.78 14.87 1.54E01 26.51 24.983 51.330
Pier2 280.31 14.02 7.01 21.5 9.99E01 10.74 0.97 21.47 9.68E01 20.80 13.581 40.291
Abut2 44.95 2.25 1.12 10000 2.28E04 2.28 1.94 2.28 4.42E04 4.42 4.351 8.565
Total 650.52 32.526 16.260 K eff,j 28.996 56.155 47.267 108.752
Step B1.4

StepB1.10 T eff 1.63 Effectiveperiod

0.28 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.67

B L 1.67 DampingFactor
d 1.92 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 0.375 0.187 0.381 1.88 0.386
Pier1 2.336 1.168 2.478 1.68 2.558
Pier2 2.336 1.168 3.579 0.94 3.653
Abut2 0.375 0.187 0.381 1.88 0.386

148

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and
then applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 1.5

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 (see
elastic stiffness, Kd,i, of each isolator i as follows: Table B1.12-1) by the number of isolators at each
, support (n = 6), the following values for Qd /isolator
, (B-19)
and Kd /isolator are obtained:
and
, o Qd, 1 = 2.25/6 = 0.37 k
, ( B-20)
o Qd, 2 = 14.02/6= 2.34 k
where values for Qd,j and Kd,j are obtained from the o Qd, 3 = 14.02/6= 2.34 k
final cycle of iteration in the Simplified Method (Step o Qd, 4 = 2.25/6 = 0.37 k
B1. 12 and
o Kd,1 = 1.12/6 = 0.19 k/in
o Kd,2 = 7.01/6 = 1.17 k/in
o Kd,3 = 7.01/6 = 1.17 k/in
o Kd,4 = 1.12/6 = 0.19 k/in

B2.2 Initial Stiffness and Yield Displacement B2.2 Initial Stiffness and Yield Displacement,
Calculate the initial stiffness, Ku,i, and the yield Example 1.5
displacement, dy,i, for each isolator i as follows: Since the isolator type has been specified in Step A1
(1) For friction-based isolators: Ku,i = and dy,i = 0. to be an elastomeric bearing (i.e. not a friction-based
(2) For other types of isolators, and in the absence of bearing), calculate Ku,i and dy,i for an isolator on Pier 1
isolator-specific information, take as follows:
, 10 , 10 1.17 11.7 /
, 10 , ( B-21)
and then and
, 2.34
,
( B-22) , 0.22
, , , 11.7 1.17
, ,

As expected, the yield displacement is small

compared to the expected isolator displacement (~2
in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.

B2.3 Isolator Effective Stiffness, Kisol,i B2.3 Isolator Effective Stiffness, Kisol,i, Example 1.5
Calculate the isolator stiffness, Kisol,i, of each isolator Dividing the results for Kisol (Step B1.12) among the 6
i: isolators at each support, the following values for Kisol
/isolator are obtained:
, o Kisol,1 = 2.28/6 = 0.38 k/in
, (B -23)
o Kisol,2 = 14.87/6 = 2.48 k/in
o Kisol,3 = 21.47/6 = 3.58 k/in

149

 o Kisol,4 = 2.28/6 = 0.38 k/in

B2.4 Three-Dimensional Bridge Model B2.4 Three-Dimensional Bridge Model, Example

Using computer-based structural analysis software, 1.5
create a 3-dimensional model of the bridge with the Although the bridge in this Design Example is
isolators represented by spring elements. The without skew or curvature, a 3-dimensional finite
stiffness of each isolator element in the horizontal element model was developed for this Step, as shown
axes (Kx and Ky in global coordinates, K2 and K3 in below.
typical local coordinates) is the Kisol value calculated
in the previous step. For bridges with regular
geometry and minimal skew or curvature, the
superstructure may be represented by a single stick
provided the load path to each individual isolator at
each support is explicitly modeled, usually by a rigid
cap beam and a set of rigid links. If the geometry is
irregular, or if the bridge is skewed or curved, a finite
element model is recommended to accurately capture
the load carried by each individual isolator. If the .
piers have an unusual weight distribution, such as a
pier with a hammerhead cap beam, a more rigorous
model is justified.

B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in
Step A2 is for a 5% damped system. It is modified in
this step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the
effective period of the bridge, Teff, by the damping
factor, BL.
B.2.5 Composite Design Response Spectrum,
Example 1.5
From the final results of Simplified Method (Step
B1.12), BL = 1.67 and Teff = 1.63 sec. Hence the
transition in the composite spectrum from 5% to 28%
damping occurs at 0.8 Teff = 0.8 (1.63) = 1.30 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.30 sec by 1.67.
0.9

0.8

0.7
Acceleration(g)

0.6

0.5

0.4

0.3

0.2

0.1

0.0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Period(s)

B2.6 Multimodal Analysis of Finite Element
Model
Input the composite response spectrum as a user-
specified spectrum in the software, and define a load
case in which the spectrum is applied in the
longitudinal direction. Analyze the bridge for this
load case.
B2.6 Multimodal Analysis of Finite Element
Model, Example 1.5
Results of modal analysis of the example bridge are
summarized in Table B2.6-1 Here the modal periods
and mass participation factors of the first 12 modes
are given. The first two modes are the principal
longitudinal and transverse modes with periods of
1.54 and 1.37 sec respectively. While no significant
coupling is observed in the first two modes, the third
mode is a strongly coupled transverse and rotational
mode, as might be expected due to the lack of

150

 symmetry in the transverse direction.

Table B2.6-1 Modal Properties of Bridge

Example 1.5 First Iteration
Mode Period ModalParticipatingMassRatios
No. Sec UX UY UZ RX RY RZ
1 1.543 0.792 0.000 0.000 0.000 0.001 0.000
2 1.365 0.000 0.408 0.000 0.015 0.000 0.057
3 1.286 0.000 0.340 0.000 0.010 0.000 0.698
4 0.370 0.025 0.000 0.000 0.000 0.001 0.000
5 0.252 0.000 0.068 0.000 0.015 0.000 0.107
6 0.243 0.000 0.000 0.000 0.000 0.000 0.003
7 0.187 0.062 0.000 0.000 0.000 0.000 0.000
8 0.120 0.000 0.000 0.000 0.000 0.000 0.003
9 0.104 0.000 0.007 0.000 0.226 0.000 0.001
10 0.102 0.000 0.000 0.246 0.000 0.197 0.000
11 0.096 0.000 0.069 0.000 0.041 0.000 0.013
12 0.095 0.000 0.000 0.021 0.000 0.017 0.000

Computed values for the isolator displacements due to

a longitudinal earthquake are as follows (numbers in
parentheses are those used to calculate the initial
properties to start iteration from the
SimplifiedMethod):

o disol,1 = 1.88 (1.94) in

o disol,2 = 1.68 (1.78) in
o disol,3 = 0.94 (0.97) in
o disol,4 = 1.88 (1.94) in

B2.7 Convergence Check B2.7 Convergence Check, Example 1.5

Compare the resulting displacements at the The new superstructure displacement is 1.88 in, a 3%
superstructure level (d) to the assumed difference from the displacement assumed at the start
displacements. These displacements can be obtained of the Multimode Spectral Analysis.
by examining the joints at the top of the isolator
spring elements. If in close agreement, go to Step
B2.9. Otherwise go to Step B2.8.

B2.8 Update Kisol,i, Keff,j, and BL B2.8 Update Kisol,i, Keff,j, and BL, Example 1.5
Use the calculated displacements in each isolator As convergence was reached at the first iteration, it is
element to obtain new values of Kisol,i for each unnecessary to calculate updated properties.
isolator as follows:
,
, , (B-24)
,

Recalculate Keff,j :

Eq. , ,
7.1-6 , (B-25)
, ,
GSID

Recalculate system damping ratio, :

Eq. 2 , , ,
7.1-10 , , ,
(B-26)
GSID

Recalculate system damping factor, BL:

151

 Eq. .
0.3
.
7.1-3 ( B-27)
1.7 0.3
GSID

Obtain the effective period of the bridge from the

multi-modal analysis and with the revised damping
factor (Eq. B-27), construct a new composite
response spectrum. Go to Step B2.6.
B2.9 Superstructure and Isolator Displacements
Once convergence has been reached, obtain
o superstructure displacements in the
longitudinal (xL) and transverse (yL) directions
of the bridge, and
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions of the bridge, for
each isolator, for this load case (i.e.
longitudinal loading). These displacements
may be found by subtracting the nodal
displacements at each end of each isolator
spring element.
B2.9 Superstructure and Isolator Displacements,
Example 1.5
From the above analysis:
o superstructure displacements in the
longitudinal (xL) and transverse (yL) directions
are:
xL= 1.88 in
yL= 0.0 in
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions are:
o Abutments: uL = 1.88 in, vL = 0.00 in
o Pier 1: uL = 1.68 in, vL = 0.00 in
o Pier 2: uL = 0.94 in, vL = 0.00 in

All isolators at same support have the same

displacements.

B2.10 Pier Bending Moments and Shear Forces
Obtain the pier bending moments and shear forces in
the longitudinal (MPLL, VPLL) and transverse (MPTL,
VPTL) directions at the critical locations for the
longitudinally-applied seismic loading.
B2.10 Pier Bending Moments and Shear Forces,
Example 1.5
Maximum bending moments and shear forces in the
columns in the longitudinal (MPLL,VPLL) and
transverse (MPTL,VPTL) directions are:

Pier 1, Exterior Columns:

o MPLL= 1 kft
o MPTL= 246 kft
o VPLL= 16.21 k
o VPTL= 0.28 k
Pier 1, Interior Column:
o MPLL= 1 kft
o MPTL= 241 kft
o VPLL= 15.63 k
o VPTL= 0.27 k

Pier 2, Exterior Columns:

o MPLL= 1 kft
o MPTL= 233 kft
o VPLL= 8.77 k
o VPTL= 0.13 k
Pier 2, Interior Column:
o MPLL= 0 kft
o MPTL= 232 kft
o VPLL= 8.57 k
o VPTL= 0.12 k

152

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 1.5
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1

Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces
ator
ure to long. to long. due to
EQ EQ long. EQ
1 0.72 0.00 0.39
2 0.72 0.00 0.45
North 3 0.72 0.00 0.43
Abut
ment 4 0.72 0.00 0.43
5 0.72 0.00 0.45
6 0.72 0.00 0.39
1 4.15 0.00 0.20
2 4.16 0.00 0.17
3 4.16 0.00 0.15
Pier 1
4 4.16 0.00 0.15
5 4.16 0.00 0.17
6 4.15 0.00 0.20
1 3.34 0.00 0.18
2 3.35 0.00 0.22
3 3.35 0.00 0.25
Pier 2
4 3.35 0.00 0.25
5 3.35 0.00 0.22
6 3.34 0.00 0.18
1 0.72 0.00 0.33
2 0.72 0.00 0.40
South 3 0.72 0.00 0.40
Abut
ment 4 0.72 0.00 0.40
5 0.72 0.00 0.40
6 0.72 0.00 0.33

153

 STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake
Repeat the above process, starting at Step B1, for
earthquake loading in the transverse direction of the
bridge. Support flexibility in the transverse direction
is to be included, and a composite response spectrum
is to be applied in the transverse direction. Obtain
isolator displacements in the longitudinal (uT) and
transverse (vT) directions of the bridge, and the
biaxial bending moments and shear forces at critical
locations in the columns due to the transversely-
applied seismic loading.
C1. Analysis for Transverse Earthquake,
Example 1.5
Key results from repeating Steps B1 and B2
(Simplified and Mulitmode Spectral Methods) are:
o Teff = 1.43 sec
o Maximum superstructure displacements in the
longitudinal (xT) and transverse (yT) directions are
as follows:
North Abutment xT = 0.63 and yT = 1.93 in
Pier 1 xT = 0.63 and yT = 1.31 in
Pier 2 xT = 0.63 and yT = 1.54 in
South Abutment xT = 0.63 and yT = 2.24 in

o Maximum isolator displacements in the

longitudinal (uT) and transverse (vT) directions are
as follows:
North Abutment uT = 0.63 and vT = 1.93 in
Pier 1 uT = 0.60 and vT = 1.27 in
Pier 2 uT = 0.53 and vT = 1.39 in
South Abutment uT = 0.63 and vT = 2.24 in

o Maximum bending moments and shear forces in

the columns in the longitudinal (MPLL,VPLL) and
transverse (MPTL,VPTL) directions are:
Pier 1, Exterior Columns:
o MPLT= 116 kft
o MPTT= 25 kft
o VPLT= 2.69 k
o VPTT= 16.13 k
Pier 1, Interior Column:
o MPLT= 129 kft
o MPTT= 3 kft
o VPLT= 0.74 k
o VPTT= 18.71 k

Pier 2, Exterior Columns:

o MPLT= 241 kft
o MPTT= 27 kft
o VPLT= 1.65 k
o VPTT= 17.58 k
Pier 2, Interior Column:
o MPLT= 252 kft
o MPTT= 2 kft
o VPLT= 0.38 k
o VPTT= 18.75 k

o Isolator shear and axial forces are in Table C1-1.

154

 Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces due
ator
ure to transv. to transv. to transv.
EQ EQ EQ
1 0.24 0.74 2.50
2 0.15 0.75 1.07
North 3 0.05 0.75 0.37
Abut
ment 4 0.05 0.75 0.37
5 0.15 0.75 1.07
6 0.24 0.74 2.50
1 1.80 3.79 6.35
2 1.08 3.81 0.17
3 0.36 3.81 2.15
Pier 1
4 0.36 3.81 2.15
5 1.08 3.81 0.17
6 1.80 3.79 6.35
1 1.43 3.69 7.29
2 0.85 3.71 4.95
3 0.28 3.71 2.37
Pier 2
4 0.28 3.71 2.37
5 0.85 3.71 4.95
6 1.43 3.69 7.29
1 0.22 0.76 1.84
2 0.13 0.76 1.18
South 3 0.04 0.76 0.55
Abut
ment 4 0.04 0.76 0.55
5 0.13 0.76 1.18
6 0.22 0.76 1.84

155

STEP D. CALCULATE DESIGN VALUES

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements at Pier 1,

Following the provisions in Art. 2.1 GSID, and Example 1.5
illustrated in Fig. 2.1-1 GSID, calculate the total To illustrate the process, design displacements for the
design displacement, dt, for each isolator by outside isolator on Pier 1 are calculated below.
combining the displacements from the longitudinal
(uL and vL) and transverse (uT and vT) cases as Load Case 1:
follows: u1 = uL + 0.3uT = 1.0(1.68) + 0.3(0.60) = 1.86 in
u1 = uL + 0.3uT (D-1) v1 = vL + 0.3vT = 1.0(0) + 0.3(1.26) = 0.38 in
v1 = vL + 0.3vT (D-2) R1 = = 1.86 0.38 = 1.89 in
R1 = (D-3)
Load Case 2:
u2 = 0.3uL + uT (D-4) u2 = 0.3uL + uT = 0.3(1.68) + 1.0(0.60) = 1.10 in
v2 = 0.3vL + vT (D-5) v2 = 0.3vL + vT = 0.3(0) + 1.0(1.26) = 1.26 in
R2 = (D-6) R2 = = 1.10 1.26 = 1.68 in

dt = max(R1, R2) (D-7) Governing Case:

D2. Design Moments and Shears
Calculate design values for column bending moments
and shear forces for all piers using the same
combination rules as for displacements.
Total design displacement, dt = max(R1, R2)
= 1.89in
D2. Design Moments and Shears in Pier 1,
Example 1.5
Design moments and shear forces are calculated for
Pier 1, Column 1, below to illustrate the process.

Alternatively this step may be deferred because the Load Case 1:

above results may not be final. Upper and lower VPL1= VPLL + 0.3VPLT = 1.0(16.21) + 0.3(2.69)
bound analyses are required after the isolators have = 17.01 k
been designed as described in Art 7. GSID. These VPT1= VPTL + 0.3VPTT = 1.0(0.28) + 0.3(16.13)
analyses are required to determine the effect of = 5.12 k
possible variations in isolator properties due age, R1 = = 17.01 5.12 = 17.77 k
temperature and scragging in elastomeric systems.
Accordingly the results for column shear in Steps Load Case 2:
B2.10 and C are likely to increase once these VPL2= 0.3VPLL + VPLT = 0.3(16.21) + 1.0(2.69)
analyses are complete. = 7.55 k
VPT2= 0.3VPTL + VPTT = 0.3(0.28) + 1.0(16.13)
= 16.22 k
R2 = = 7.55 16.22 = 17.89 k

Governing Case:
Design column shear = max (R1, R2)
= 17.89 k

156

STEP E. DESIGN OF LEAD-RUBBER (ELASTOMERIC) ISOLATORS

A lead-rubber isolator is an elastomeric
bearing with a lead core inserted on its vertical
centreline. When the bearing and lead core are
deformed in shear, the elastic stiffness of the
lead provides the initial stiffness (Ku).With
increasing lateral load the lead yields almost
perfectly plastically, and the post-yield
stiffness Kd is given by the rubber alone. More
details are given in MCEER 2006.

While both circular and rectangular bearings

are commercially available, circular bearings
are more commonly used. Consequently the
procedure given below focuses on circular
bearings. The same steps can be followed for
rectangular bearings, but some modifications
will be necessary. When sizing the physical dimensions of the bearing, plan dimensions (B, dL) should be rounded
up to the next 1/4 increment, while the total thickness of elastomer, Tr, is specified in multiples of the layer
thickness. Typical layer thicknesses for bearings with lead cores are 1/4 and 3/8. High quality natural rubber
should be specified for the elastomer. It should have a shear modulus in the range 60-120 psi and an ultimate
elongation-at-break in excess of 5.5. Details can be found in rubber handbooks or in MCEER 2006.

The following design procedure assumes the isolators are bolted to the masonry and sole plates. Isolators that use
shear-only connections (and not bolts) require additional design checks for stability which are not included below.
See MCEER 2006.

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

E1. Required Properties E1. Required Properties, Example 1.5

Obtain from previous work the properties required of The design of one of the exterior isolators on a pier is
the isolation system to achieve the specified given below to illustrate the design process for lead-
performance criteria (Step A1). rubber isolators.
required characteristic strength, Qd / isolator
required post-elastic stiffness, Kd / isolator From previous work
total design displacement, dt, for each isolator Qd/isolator = 2.34 k
maximum applied dead and live load (PDL, Kd/isolator = 1.17 k/in
PLL) and seismic load (PSL) which includes Total design displacement, dt = 1.89 in
seismic live load (if any) and overturning PDL = 45.52 k
forces due to seismic loads, at each isolator, PLL = 15.50 k
and PSL = 6.35 k
maximum wind load, PWL PWL =1.76 k < Qd OK

E2. Isolator Sizing
E2.1 Lead Core Diameter E2.1 Lead Core Diameter, Example 1.5
Determine the required diameter of the lead plug, dL,
using:
2.34
(E-1) 1.61
0.9 0.9 0.9
See Step E2.5 for limitations on dL

157

E2.2 Plan Area and Isolator Diameter E2.2 Plan Area and Isolator Diameter, Example
Although no limits are placed on compressive stress 1.5
in the GSID, (maximum strain criteria are used
instead, see Step E3) it is useful to begin the sizing
process by assuming an allowable stress of, say, 1.0
ksi.

Then the bonded area of the isolator is given by:

(E-2)
1.0 45.52 15.50
61.02
1.0 1.0
and the corresponding bonded diameter (taking into
account the hole required to accommodate the lead
core) is given by:
4
(E-3)
4 4 61.02
1.61

Round the bonded diameter, B, to nearest quarter = 8.96 in

inch, and recalculate actual bonded area using
Round B up to 9.0 in and the actual bonded area is:
(E-4)
4
9.0 1.61 61.57
4
Note that the overall diameter is equal to the bonded
diameter plus the thickness of the side cover layers
(usually 1/2 inch each side). In this case the overall
diameter, Bo is given by:

1.0 (E-5)
Bo = 9.0 + 2(0.5) = 10.0 in

E2.3 Elastomer Thickness and Number of Layers E2.3 Elastomer Thickness and Number of Layers,
Since the shear stiffness of the elastomeric bearing is Example 1.5
given by:
(E-6)
where G = shear modulus of the rubber, and
Tr = the total thickness of elastomer, Select G, shear modulus of rubber, = 100 psi (0.1 ksi)
it follows Eq. E-5 may be used to obtain Tr given a
required value for Kd Then
0.1 61.57
(E-7) 5.27
1.17
A typical range for shear modulus, G, is 60-120 psi.
Higher and lower values are available and are used in
special applications.

If the layer thickness is tr, the number of layers, n, is

given by:
5.27
(E-8) 21.09
0.25
rounded up to the nearest integer.
Round to nearest integer, i.e. n = 22

158

Note that because of rounding the plan dimensions

and the number of layers, the actual stiffness, Kd, will
not be exactly as required. Reanalysis may be
necessary if the differences are large.

E2.4 Overall Height E2.4 Overall Height, Example 1.5

The overall height of the isolator, H, is given by:

1 2 ( E-9) 22 0.25 21 0.125 2 1.5 11.125

where ts = thickness of an internal shim (usually

about 1/8 in), and
tc = combined thickness of end cover plate
(0.5 in) and outer plate (1.0 in)

E2.5 Size Checks E2.5 Size Checks, Example 1.5

Experience has shown that for optimum performance Since B=9.0 check
of the lead core it must not be too small or too large.
The recommended range for the diameter is as 9.0 9.0
follows: 3 6
(E-10)
3 6 i.e., 3.0 1.5

Since dL = 1.61, lead core size is acceptable.

Art. 12.2 GSID requires that the isolation system 0.025 0.025 45.52
provides a lateral restoring force at dt greater than the , 0.40 /
2.27
restoring force at 0.5dt by not less than W/80. This
equates to a minimum Kd of 0.025W/d. As
0.1 61.56
0.025 1.12 / ,
5.5
,

E3. Strain Limit Check E3. Strain Limit Check, Example 1.5
Art. 14.2 and 14.3 GSID requires that the total
applied shear strain from all sources in a single layer
of elastomer should not exceed 5.5, i.e.,
Since
, + 0.5 5.5 (E-11)
45.52
0.739
where , , , are defined below. 61.57
(a) is the maximum shear strain in the layer due to
compression and is given by: G = 0.1 ksi
(E-12) and
where Dc is shape coefficient for compression in 61.57
8.71
circular bearings = 1.0, , G is shear 9.0 0.25
modulus, and S is the layer shape factor given by: then
(E-13) 1.0 0.739
0.849
0.1 8.71
(b) , is the shear strain due to earthquake loads
and is given by:

159

 1.89
, (E-14) , 0.344
5.5

(c) is the shear strain due to rotation and is given

by:
0.375 9.0 0.01
(E-15) 0.221
0.25 5.5
where Dr is shape coefficient for rotation in circular
bearings = 0.375, and is design rotation due to DL, Substitution in Eq E-11 gives
LL and construction effects Actual value for may
not be known at this time and value of 0.01 is , 0.5 0.849 0.344 0.5 0.221
suggested as an interim measure, including 1.30 5.5
uncertainties (see LRFD Art. 14.4.2.1)

E4. Vertical Load Stability Check E4. Vertical Load Stability Check, Example 1.5
Art 12.3 GSID requires the vertical load capacity of
all isolators be at least 3 times the applied vertical
loads (DL and LL) in the laterally undeformed state.

Further, the isolation system shall be stable under

1.2(DL+SL) at a horizontal displacement equal to
either
2 x total design displacement, dt, if in Seismic Zone 1
or 2, or
1.5 x total design displacement, dt, if in Seismic Zone
3 or 4.

E4.1 Vertical Load Stability in Undeformed State E4.1 Vertical Load Stability in Undeformed State,
The critical load capacity of an elastomeric isolator at Example 1.5
zero shear displacement is given by

4
1 1 3 3 0.1 0.3
2 (E-16)
0.3 1 0.67 8.71 15.55
where
9.0
322.1
Ts = total shim thickness 64

15.55 322.1
910.15 /
1 0.67 5.5
E = elastic modulus of elastomer = 3G
0.1 61.57
64 1.12 /
5.5
It is noted that typical elastomeric isolators have high
shape factors, S, in which case:
4
1 (E-17)

and Eq. E-16 reduces to:

(E-18) 1.12 910.15 100.27

Check that:

160


3 (E-19) 100.27

1.64 3
45.52 15.5

E4.2 Vertical Load Stability in Deformed State E4.2 Vertical Load Stability in Deformed State,
The critical load capacity of an elastomeric isolator at Example 1.5
shear displacement may be approximated by: Since bridge is in Zone 2, 2 2 1.89
3.79
(E-20)
where 3.79
2 2.27
Ar = overlap area between top and bottom plates 9.0
of isolator at displacement (Fig. 2.2-1
GSID)
= 4 2.27 2.27
0.480
2

Agross = 4
0.480 100.27 48.15
It follows that:
(E-21)

Check that:

1 (E-22)
1.2 48.15
0.79
1.2 1.2 45.52 6.35
1

E5. Design Review E5. Design Review, Example 1.5

The basic dimensions of the isolator designed above
are as follows:

10.00 in (od) x 11.125 in (high) x 1.61 in dia. lead

core and the volume, excluding steel end and cover
plates, is 638 in3.

This design satisfies the shear strain limit criteria, but

not the vertical load stability ratio in the undeformed
and deformed states.

A redesign is therefore required and the easiest way to

increase the Pcr is to increase the shape factor, S, since
the bending stiffness of an isolator is a function of the
shape factor squared. See equations in Step E4.1. To
increase S, increase the bonded area Ab while keeping
tr constant (Eq. E-13). But to keep Kd constant while
increasing Ab and Tr is constant, decrease the shear
modulus, G (Eq. E-6).

This redesign is outlined below. After repeating the

calculation for diameter of lead core, the process
begins by reducing the shear modulus to 60 psi (0.06

161

 ksi) and increasing the bonded diameter to 12 in.

E2.1
2.34
1.61
0.9 0.9

E2.2
5.5 1.17
107.25
0.06

4 4 107.25
1.61

11.80

Round B to 12 in and the actual bonded area

becomes:
12 1.61 111.06
4

Bo = 12 + 2(0.5) = 13 in

E2.3
0.06 111.06
5.71
1.17

5.71
22.82
0.25

Round to nearest integer, i.e. n = 23.

E2.4
23 0.25 22 0.125 2 1.5 11.5
E2.5
Since B=12 check

12 12
3 6

i.e., 4 2

Since dL = 1.61, the size of lead core is too small, and

there are 2 options: (1) Accept the undersize and
check for adequate performance during the Quality
Control Tests required by GSID Art. 15.2.2; or (2)
Only have lead cores in every second isolator, in
which case the core diameter, in those isolators with
cores, will be 2 x 1.61 = 2.27 in (which satisfies
above criterion).

0.06 111.06
1.16 / ,
5.75

E3.

162

 45.52
0.41
111.06

111.06
11.78
12 0.25

1.0 0.41
0.580
0.06 11.78

1.89
, 0.329
5.75

0.375 12 0.01
0.376
0.25 5.75

, 0.5 0.580 0.329 0.5 0.376

1.10 5.5
E4.1
3 3 0.06 0.18

0.18 1 0.67 11.78 16.93

12
1017.88
64
16.93 1017.88
2996.42 /
5.75

0.06 111.06
1.159 /
5.75

1.159 2996.42 185.13

185.13
3.03 3
45.52 15.50

E4.2
3.79
2 2.50
12

2.50 2.50
0.605

0.605 185.13 111.96

111.96
1.84 1
1.2 1.2 45.52 6.35

E5.

163

 The basic dimensions of the redesigned isolator are as

follows:

13.0 in (od) x 11.5 in (high) x 1.61 in dia. lead core

and its volume (excluding steel end and cover plates)
is 1128 in3.

This design meets all the design criteria but is about

75% larger by volume than the previous design. This
increase in size is driven by the need to satisfy the
vertical load stability ratio of 3.0 in the undeformed
state.
E6. Minimum and Maximum Performance Check E6. Minimum and Maximum Performance Check,
Art. 8 GSID requires the performance of any Example 1.5
isolation system be checked using minimum and Minimum Property Modification factors are:
maximum values for the effective stiffness of the min,Kd = 1.0
system. These values are calculated from minimum min,Qd = 1.0
and maximum values of Kd and Qd, which are found
using system property modification factors, as which means there is no need to reanalyze the bridge
indicated in Table E6-1. with a set of minimum values.
Table E6-1. Minimum and maximum values
for Kd and Qd. Maximum Property Modification factors are:
max,a,Kd = 1.1
Eq. max,a,Qd = 1.1
8.1.2-1 Kd,max = Kd max,Kd (E-23)
GSID max,t,Kd = 1.1
Eq. max,t,Qd = 1.4
8.1.2-2 Kd,min = Kd min,Kd (E-24)
GSID max,scrag,Kd = 1.0
max,scrag,Qd = 1.0
Eq.
8.1.2-3 Qd,max = Qd max,Qd (E-25)
GSID
Eq.
8.1.2-4 Qd,min = Qd min,Qd (E-26)
GSID

Determination of the system property modification

factors should include consideration of the effects of
temperature, aging, scragging, velocity, travel (wear)
and contamination as shown in Table E6-2. In lieu of
tests, numerical values for these factors can be
obtained from Appendix A, GSID.

Table E6-2. Minimum and maximum values for

system property modification factors.

Eq. min,Kd = (min,t,Kd) (min,a,Kd)

8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-27)
GSID (min,scrag,Kd)
Eq. max,Kd = (max,t,Kd) (max,a,Kd)
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-28)
GSID (max,scrag,Kd)
Eq. min,Qd = (min,t,Qd) (min,a,Qd)
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-29)
GSID (min,scrag,Qd)

164

 Eq. max,Qd = (max,t,Qd) (max,a,Qd)

8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd) (E-30)
GSID (max,scrag,Qd)

Adjustment factors are applied to individual -factors
(except v) to account for the likelihood of
occurrence of all of the maxima (or all of the
minima) at the same time. These factors are applied
to all -factors that deviate from unity but only to the
portion of the -factor that is greater than, or less
than, unity. Art. 8.2.2 GSID gives these factors as
follows:
1.00 for critical bridges
0.75 for essential bridges
0.66 for all other bridges
Applying a system adjustment factor of 0.66 for an
other bridge, the maximum property modification
factors become:
max,a,Kd = 1.0 + 0.1(0.66) = 1.066
max,a,Qd = 1.0 + 0.1(0.66) = 1.066
max,t,Kd = 1.0 + 0.1(0.66) = 1.066
max,t,Qd = 1.0 + 0.4(0.66) = 1.264
max,scrag,Kd = 1.0
max,scrag,Qd = 1.0

As required in Art. 7 GSID and shown in Fig. C7-1
GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Qd,max. As indicated in Fig C7-1 GSID, maximum
displacements will probably be given by the first case
(Kd,min and Qd,min) and maximum forces by the second
case (Kd,max and Qd,max).
Therefore the maximum overall modification factors
max,Kd = 1.066(1.066)1.0 = 1.14
max,Qd = 1.066(1.264)1.0 = 1.35
Since the possible variation in upper bound properties
exceeds 15% a reanalysis of the bridge is required to
determine performance with these properties.

The upper-bound properties are:

Qd,max = 1.35 (2.34) = 3.16 k
and
Kd,ma x=1.14(1.16) = 1.32 k/in

E7. Design and Performance Summary, Example

E7. Design and Performance Summary
1.5
E7.1 Isolator dimensions E7.1 Isolator dimensions, Example 1.5
Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E7.1-1.
Overall diameter (includes cover layer)
Overall height Table E7.1-1 Isolator Dimensions
Diameter lead core
Isolator Overall size Overall size Diam.
Bonded diameter Location including without lead
Number of rubber layers mounting mounting core
Thickness of rubber layers plates (in) plates (in) (in)
Total rubber thickness Under
Thickness of steel shims edge 17.0 x 17.0 13.0 dia.
Shear modulus of elastomer 1.61
girder x 11.50(H) x 10.0(H)
on Pier 1
Check all dimensions with manufacturer.

Isolator No. of Rubber Total Steel

Location rubber layers rubber shim
layers thick- thick- thick-
ness ness ness
(in) (in) (in)
Under
edge
23 0.25 5.75 0.125
girder
on Pier 1

165


E7.2 Bridge Performance
Summarize bridge performance
Maximum superstructure displacement
(longitudinal)
Maximum superstructure displacement
(transverse)
Maximum superstructure displacement
(resultant)
Maximum column shear (resultant)
Maximum column moment (about transverse
axis)
Maximum column moment (about
longitudinal axis)
Maximum column torque
Check required performance as determined in Step
A3, is satisfied.
Shear modulus of elastomer = 60 psi
E7.2 Bridge Performance, Example 1.5
Bridge performance is summarized in Table E7.2-1
where it is seen that the maximum column shear is
19.56 k. This less than the column plastic shear
strength (25 k) and therefore the required
performance criterion is satisfied (fully elastic
behavior). The maximum longitudinal displacement is
2.07 in, which is slightly more than the 2.0 in
available at the abutment expansion joints, and is
barely acceptable (light pounding may occur but not
likely to cause damage to the back wall).
Table E7.2-1 Summary of Bridge Performance
Maximum superstructure
2.07 in
displacement (longitudinal)
Maximum superstructure
2.24 in
displacement (transverse)
Maximum superstructure
2.32 in
displacement (resultant)
Maximum column shear (resultant) 19.56 k
Maximum column moment about
254 kft
transverse axis
Maximum column moment about
256 kft
longitudinal axis
Maximum column torque 6.84 kft
166

 SECTION 1
PC Girder Bridge, Short Spans, Multi-Column Piers

DESIGN EXAMPLE 1.6: Skew = 450

Design Examples in Section 1

Site
ID Description S1 Pier height Skew Isolator type
Class
1.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
1.1 Change site class 0.2g D Same 0 Lead rubber bearing
Change spectral
1.2 0.6g B Same 0 Lead rubber bearing
acceleration, S1
Spherical friction
1.3 Change isolator to SFB 0.2g B Same 0
bearing
1.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
1.5 Change column height 0.2g B H1=0.5 H2 0 Lead rubber bearing
1.6 Change angle of skew 0.2g B Same 450 Lead rubber bearing

167

DESIGN PROCEDURE
STEP A: BRIDGE AND SITE DATA
A1. Bridge Properties
Determine properties of the bridge:
number of supports, m
number of girders per support, n
angle of skew
weight of superstructure including railings,
curbs, barriers and to the permanent loads,
WSS
weight of piers participating with
superstructure in dynamic response, WPP
weight of superstructure, Wj, at each
support
stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge. The calculation of these
quantities requires careful consideration of
several factors such as the use of cracked
sections when estimating column or wall
flexural stiffness, foundation flexibility,
and effective column height.
column shear strength (minimum value).
This will usually be derived from the
minimum value of the column flexural
yield strength, the column height, and
whether the column is acting in single or
double curvature in the direction under
consideration.
allowable movement at expansion joints
isolator type if known, otherwise to be
determined
A2. Seismic Hazard
Determine seismic hazard at site:
acceleration coefficients
site class and site factors
seismic zone
Plot response spectrum.
Use Art. 3.1 GSID to obtain peak ground and spectral
acceleration coefficients. These coefficients are the
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (e.g. Figures
3.10.2.1-1 to 3.10.2.1-21 LRFD).
Use Art. 3.2 to obtain Site Class and corresponding Site
Factors (Fpga, Fa and Fv). These data are the same as for
DESIGN EXAMPLE 1.6 (Skew = 450)
A1. Bridge Properties, Example 1.6
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 6
Number of columns per support = 3
Angle of skew = 450
Weight of superstructure including permanent
loads, WSS = 678.62 k
Weight of superstructure at each support:
o W1 = 51.98 k
o W2 = 287.33 k
o W3 = 287.33 k
o W4 = 51.98 k
Participating weight of piers, WPP = 151.52 k
Effective weight (for calculation of period),
Weff = Wss + WPP = 830.14 k
Stiffness of each pier in the longitudinal
direction:
o Ksub,pier1,long = 307 k/in
o Ksub,pier2,long = 307 k/in
Stiffness of each pier in the transverse direction:
o Ksub,pier1,trans = 307 k/in
o Ksub,pier2,trans = 307 k/in
Minimum column shear strength based on
flexural yield capacity of column = 25 k
Displacement capacity of expansion joints
(longitudinal) = 2.0 in for thermal and other
movements.
Lead rubber isolators
A2. Seismic Hazard, Example 1.6
Acceleration coefficients for bridge site are given in
design statement as follows:
PGA = 0.40
S1 = 0.20
SS = 0.75
Bridge is on a rock site with shear wave velocity in
upper 100 ft of soil = 3,000 ft/sec.
Table 3.10.3.1-1 LRFD gives Site Class as B.
Tables 3.10.3.2-1, -2 and -3 LRFD give following
Site Factors:
Fpga = 1.0
Fa = 1.0
Fv = 1.0
168

conventional bridges and Art 3.2 refers the designer to

the corresponding articles in the LRFD Specifications,
i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3,
LRFD.

Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give

modified spectral acceleration coefficients that include
site effects as follows: As = Fpga PGA = 1.0(0.40) = 0.40
As = Fpga PGA SDS = Fa SS = 1.0(0.75) = 0.75
SDS = Fa SS SD1 = Fv S1 = 1.0(0.20) = 0.20
SD1 = Fv S1

Seismic Zone is determined by value of SD1 in Since 0.15 < SD1 < 0.30, bridge is located in Seismic
accordance with provisions in Table5-1 GSID. Zone 2.

These coefficients are used to plot design response Design Response Spectrum is as below:
spectrum as shown in Fig. 4-1 GSID.
0.9

0.8

0.7

0.6

Acceleration
0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Period(s)

A3. Performance Requirements A3. Performance Requirements, Example 1.6

Determine required performance of isolated bridge In this example, assume the owner has specified full
during Design Earthquake (1000-yr return period). functionality following the earthquake and therefore
the columns must remain elastic (no yield).
Examples of performance that might be specified by
the Owner include: To remain elastic the maximum lateral load on the
Reduced displacement ductility demand in pier must be less than the load to yield any one
columns, so that bridge is open for emergency column (25 k).
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in The maximum shear in the column must therefore be
columns or yield elsewhere in bridge), so that less than 25 k in order to keep the column elastic and
bridge is fully functional and open to all vehicles meet the required performance criterion.
immediately following earthquake.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.

169

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for Isolator Force, F

non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most have
bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use Ku
equivalent linear springs and viscous damping to
represent the isolators, so that linear methods of analysis
disol
may be used to determine response. Since equivalent Ku
dy Isolator
Ku Displacement, d
properties such as Kisol are dependent on displacement (d),
and the displacements are not known at the beginning of
the analysis, an iterative approach is required. Note that in Kd
Art 7.1, GSID, keff is used for the effective stiffness of an
isolator unit and Keff is used for the effective stiffness of a disol = Isolator displacement
dy = Isolator yield displacement
combined isolator and substructure unit. To minimize Fisol = Isolator shear force
confusion, Kisol is used in this document in place of keff. Fy = Isolator yield force
Kd = Post-elastic stiffness of isolator
There is no change in the use of Keff and Keff,j, but Ksub is Kisol = Effective stiffness of isolator
used in place of ksub. Ku
Qd
= Loading and unloading stiffness (elastic stiffness)
= Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement (disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 1.6
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.20 2.0
GSID

(2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not
occur under non-seismic loads (e.g. wind)

170

 but low enough that yield will occur during

an earthquake. Experience has shown that
taking Qd to be 5% of the bridge weight is a
good starting point, i.e.
0.05 (B-2) 0.05 0.05 678.62 33.93
(3) Post-yield stiffness, Kd
Art 12.2 GSID requires that all isolators
exhibit a minimum lateral restoring force at
the design displacement, which translates to a
minimum post yield stiffness Kd,min given by
Eq. B-3.

Art.
0.025
12.2 , (B-3) 678.62
GSID 0.05 0.05 16.97 /
2.0
Experience has shown that a good starting
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 1.6
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength, ,

Qd, and stiffness, Kd, values in proportion to the dead o Qd, 1 = 2.60 k
load applied at that support: o Qd, 2 = 14.37 k
o Qd, 3 = 14.37 k
o Qd, 4 = 2.60 k

, (B-4)
and
and
,
, (B-5) o Kd,1 = 1.30 k/in
o Kd,2 = 7.18 k/in
o Kd,3 = 7.18 k/in
o Kd,4 = 1.30 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 1.6
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of , ,
the isolators at support j (Kisol,j) and the stiffness of , ,
the substructure Ksub,j. See figure below for
definitions (after Fig. 7.1-1 GSID). o 1 = 2.60x10-4
o 2 = 4.79x10-2
An expression for Keff,j, is given in Eq.7.1-6 GSID, but o 3 = 4.79x10-2
a more useful formula as follows (MCEER 2006): o 4 = 2.60x10-4
,
, (B-6)
1 ,
where ,
1
, ,
(B-7)
, , o Keff,1 = 2.60 k/in
and Ksub,j for the piers are given in Step A1. For the o Keff,2 = 14.04 k/in
abutments, take Ksub,j to be a large number, say 10,000 o Keff,3 = 14.04 k/in
k/in, unless actual stiffness values are available. Note o Keff,4 = 2.60 k/in
that if the default option is chosen, unrealistically high

171

 values for Ksub,j will give unconservative results for

column moments and shear forces.

F
Kd
Qd Kisol

dy disol
Superstructure
Isolator Effective Stiffness, Kisol

F
Isolator(s), Kisol
Ksub

Substructure, Ksub
dsub

Substructure Stiffness, Ksub

dsub disol F

d
Keff

d = disol + dsub

Combined Effective Stiffness, Keff

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 1.6
Calculate the total effective stiffness, Keff, of the
bridge:

Eq. , 33.27 /
7.1-6 , ( B-8)
GSID

B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 1.6
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1

(B-9) o disol,1 = 2.00 in

,
1 o disol,2 = 1.91 in
o disol,3 = 1.91 in
o disol,4 = 2.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the effective stiffness of the isolation system Example 1.6
at support j, Kisol,j, for all supports:
,
, ,
,
,
(B-10) o Kisol,1 = 2.60 k/in
, ,
, o Kisol,2 = 14.71 k/in
o Kisol,3 = 14.71 k/in
o Kisol,4 = 2.60 k/in

172

B1.7 Substructure Displacement at Each Support
B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j,
Example 1.6
for all supports:
, ,

, , (B-11)
o dsub,1 = 5.20x10-4 in
o d sub,2 = 9.15x10-2 in
o d sub,3 = 9.15x10-2 in
o d sub,4 = 5.20x10-4 in
B1.8 Lateral Load in Each Substructure Support B1.8 Lateral Load in Each Substructure, Example
Calculate the shear at support j, Fsub,j, for all 1.6
supports:
, , , (B-12) , , ,

where values for Ksub,j are given in Step A1. o F sub,1 = 5.20 k
o F sub,2 = 28.08 k
o F sub,3 = 28.08 k
o F sub,4 = 5.20 k

B1.9 Column Shear Force at Each Support B1.9 Column Shear Force at Each Support,
Calculate the shear in column k at support j, Fcol,j,k, Example 1.6
assuming equal distribution of shear for all columns at
support j:
,
, ,
#
,
, ,
# (B-13) o F col,2,1-3 9.36 k
o F col,3,1-3 9.36 k

Use these approximate column shear forces as a check These column shears are less than the yield capacity
on the validity of the chosen strength and stiffness of each column (25 k) as required in Step A3 and the
characteristics. chosen strength and stiffness values in Step B1.1 are
therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 1.6
damping ratio, , of the bridge:

Eq. 830.14
7.1-5 2 (B-14) 2 2
GSID 386.4 33.27

and = 1.60 sec

Eq. 2 , , , and taking dy,j = 0:

(B-15)
7.1-10 , , ,
GSID 2 , , 0
0.31
, , ,
where dy,j is the yield displacement of the isolator. For
friction-based isolators, dy,j = 0. For other types of
isolators dy,j is usually small compared to disol,j and has
negligible effect on , Hence it is suggested that for
the Simplified Method, set dy,j = 0 for all isolator

173

types. See Step B2.2 where the value of dy,j is revisited
for the Multimode Spectral Analysis Method.
B1.11 Damping Factor
Calculate the damping factor, BL, and the
displacement, d, of the bridge:
Eq. .
, 0.3
.
7.1-3 (B-16)
1.7, 0.3
GSID
Eq. 9.79 (B-17)
7.1-4
GSID
B1.12 Convergence Check
Compare the new displacement with the initial value
assumed in Step B1.1. If there is close agreement, go
to the next step; otherwise repeat the process from
Step B1.3 with the new value for displacement as the
assumed displacement.
This iterative process is amenable to solution using a
spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
B1.11 Damping Factor, Example 1.6
Since 0.31 0.3
1.70
and
9.79 9.79 0.2 1.60
1.84
1.70
B1.12 Convergence Check, Example 1.6
Since the calculated value for displacement, d
(=1.84 in) is not close to that assumed at the
beginning of the cycle (Step B1.1, d = 2.0), use the
value of 1.84 in as the new assumed displacement and
repeat from Step B1.3.
After one iteration, convergence is reached at a
superstructure displacement of 1.80 in, with an
effective period of 1.55 seconds, and a damping factor
of 1.70 (33% damping ratio). The displacement in the
isolators at Pier 1 is 1.71 in and the effective stiffness
of the same isolators is 15.57 k/in.
See spreadsheet in Table B1.12-1 for results of final
iteration.
Ignoring the weight of the cap beams, the sum of the
column shears at a pier must equal the total isolator
shear force. Hence, approximate column shear (per
column) = 15.57(1.71)/3 = 8.87 k which is less than
the maximum allowable (25k) if elastic behavior is to
be achieved (as required in Step A3).
Also the superstructure displacement = 1.80 in, which
is less than the available clearance of 2.0 in.
Therefore the above solution is acceptable, and go to
Step B2.
Note that available clearance (2.0 in) is greater than
minimum required which is given by:
8 8 0.20 1.55
1.46
1.7
174

 Table B1.12-1 Simplified Method Solution for Design Example 1.6 Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
678.62 151.52 830.137 0.2 6

StepB1.1 d 1.80 Assumeddisplacement

Qd 32.53 Characteristicstrength
Kd 16.26 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 51.98 2.60 1.30 10000 2.74E04 2.74 1.80 2.74 4.94E04 4.94 4.677 8.887
Pier1 287.33 14.37 7.18 307.0 5.07E02 14.82 1.71 15.57 8.69E02 26.67 24.609 48.004
Pier2 287.33 14.37 7.18 307.0 5.07E02 14.82 1.71 15.57 8.69E02 26.67 24.609 48.004
Abut2 51.98 2.60 1.30 10000 2.74E04 2.74 1.80 2.74 4.94E04 4.94 4.677 8.887
Total 678.62 33.931 16.965 K eff,j 35.123 63.217 58.572 113.783
Step B1.4

StepB1.10 T eff 1.55 Effectiveperiod

0.33 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.76

B L 1.70 DampingFactor
d 1.79 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 0.433 0.217 0.457 1.65 0.479
Pier1 2.394 1.197 2.595 1.54 2.752
Pier2 2.394 1.197 2.595 1.54 2.752
Abut2 0.433 0.217 0.457 1.65 0.479

175

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and
then applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 1.6

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 (see
elastic stiffness, Kd,i, of each isolator i as follows: Table B1.12-1) by the number of isolators at each
, support (n = 6), the following values for Qd /isolator
, (B-19)
and Kd /isolator are obtained:
and
, o Qd, 1 = 2.60/6 = 0.43 k
, ( B-20)
o Qd, 2 = 14.37/6= 2.39 k
where values for Qd,j and Kd,j are obtained from the o Qd, 3 = 14.37/6= 2.39 k
final cycle of iteration in the Simplified Method (Step o Qd, 4 = 2.60/6 = 0.43 k
B1. 12 and
o Kd,1 = 1.30/6 = 0.22 k/in
o Kd,2 = 7.18/6 = 1.20 k/in
o Kd,3 = 7.18/6 = 1.20 k/in
o Kd,4 = 1.30/6 = 0.22 k/in

B2.2 Initial Stiffness and Yield Displacement B2.2 Initial Stiffness and Yield Displacement,
Calculate the initial stiffness, Ku,i, and the yield Example 1.6
displacement, dy,i, for each isolator i as follows: Since the isolator type has been specified in Step A1
(1) For friction-based isolators: Ku,i = and dy,i = 0. to be an elastomeric bearing (i.e. not a friction-based
(2) For other types of isolators, and in the absence of bearing), calculate Ku,i and dy,i for an isolator on Pier 1
isolator-specific information, take as follows:
, 10 , 10 1.20 12.0 /
, 10 , ( B-21)
and then and
, 2.39
,
( B-22) , 0.22
, , , 12.0 1.20
, ,

As expected, the yield displacement is small

compared to the expected isolator displacement (~2
in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.

B2.3 Isolator Effective Stiffness, Kisol,i
Calculate the isolator stiffness, Kisol,i, of each isolator
i:
,
, (B -23)
B2.3 Isolator Effective Stiffness, Kisol,i, Example 1.6
Dividing the results for Kisol (Step B1.12) among the 6
isolators at each support, the following values for Kisol
/isolator are obtained:
o Kisol,1 = 2.74/6 = 0.46 k/in
o Kisol,2 = 15.57/6 = 2.60 k/in

176

 o Kisol,3 = 15.57/6 = 2.60 k/in

o Kisol,4 = 2.74/6 = 0.46 k/in

B2.4 Three-Dimensional Bridge Model B2.4 Three-Dimensional Bridge Model, Example

Using computer-based structural analysis software, 1.6
create a 3-dimensional model of the bridge with the A 3-dimensional finite element model was developed
isolators represented by spring elements. The for this Step, as shown below.
stiffness of each isolator element in the horizontal
axes (Kx and Ky in global coordinates, K2 and K3 in
typical local coordinates) is the Kisol value calculated
in the previous step. For bridges with regular
geometry and minimal skew or curvature, the
superstructure may be represented by a single stick
provided the load path to each individual isolator at
each support is explicitly modeled, usually by a rigid
cap beam and a set of rigid links. If the geometry is
irregular, or if the bridge is skewed or curved, a finite
element model is recommended to accurately capture
the load carried by each individual isolator. If the
piers have an unusual weight distribution, such as a
pier with a hammerhead cap beam, a more rigorous .
model is justified.

B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in
Step A2 is for a 5% damped system. It is modified in
this step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the
effective period of the bridge, Teff, by the damping
factor, BL.
B.2.5 Composite Design Response Spectrum,
Example 1.6
From the final results of Simplified Method (Step
B1.12), BL = 1.70 and Teff = 1.55 sec. Hence the
transition in the composite spectrum from 5% to 33%
damping occurs at 0.8 Teff = 0.8 (1.55) = 1.24 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.24 sec by 1.70.
0.9

0.8

0.7
Acceleration(g)

0.6

0.5

0.4

0.3

0.2

0.1

0.0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Period(s)

B2.6 Multimodal Analysis of Finite Element
Model
Input the composite response spectrum as a user-
specified spectrum in the software, and define a load
case in which the spectrum is applied in the
longitudinal direction. Analyze the bridge for this
load case.
B2.6 Multimodal Analysis of Finite Element
Model, Example 1.6
Results of modal analysis of the example bridge are
summarized in Table B2.6-1, in which the X-direction
is along the bridge (longitudinal), and the Y-direction
is across the bridge (transverse). Here the modal
periods and mass participation factors of the first 12

177

 modes are given. The first three modes are the

principal isolation modes with periods of 1.47, 1.40
and 1.39 sec respectively. Mode shapes
corresponding to these three modes are plotted in
Figure B2.6-1. As can be seen, the first and third
modes are coupled translational modes whereas the
second mode is a pure torsional mode (rotation about
the Z-axis). Figure B2.6-1 also shows that the first
and second modes have approximately equal
displacement in the longitudinal and transverse
directions, and this is confirmed by the relative sizes
of the mass participation factors in Table B2.6-1. The
strong nature of coupling in these modes would
explain the presence of significant discrepancies
between the Simplified Method and the Multimodal
Analysis.

Table B2.6-1 Modal Properties of Bridge

Example 1.6 First Iteration
Mode Period ModalParticipatingMassRatios
No. Sec UX UY UZ RX RY RZ
1 1.466 0.352 0.399 0.000 0.016 0.001 0.274
2 1.401 0.000 0.000 0.000 0.000 0.000 0.236
3 1.392 0.384 0.337 0.000 0.016 0.001 0.232
4 0.227 0.000 0.000 0.000 0.000 0.000 0.012
5 0.227 0.071 0.071 0.000 0.006 0.000 0.048
6 0.201 0.000 0.000 0.000 0.000 0.000 0.000
7 0.201 0.000 0.000 0.000 0.000 0.000 0.012
8 0.149 0.006 0.008 0.000 0.537 0.024 0.006
9 0.140 0.000 0.000 0.313 0.000 0.232 0.000
10 0.114 0.000 0.000 0.000 0.000 0.000 0.014
11 0.108 0.086 0.085 0.000 0.063 0.002 0.058
12 0.103 0.003 0.003 0.000 0.000 0.000 0.002

Mode1

Combined longitudinal & transverse translational mode

178

 Mode2

In-plane rotational mode

Mode3

Combined longitudinal & transverse translational mode

Figure B2.6-1 First Three Mode Shapes for Isolated Bridge with 45 Skew (Example 1.6)

Computed values for the isolator displacements due to

a longitudinal earthquake are as follows (numbers in
parentheses are those used to calculate the initial
properties to start iteration from the Simplified
Method):

o disol,1 = 1.67 (1.80) in

o disol,2 = 1.56 (1.71) in
o disol,3 = 1.56 (1.71) in
o disol,4 = 1.67 (1.80) in

B2.7 Convergence Check B2.7 Convergence Check, Example 1.6

Compare the resulting displacements at the
superstructure level (d) to the assumed
displacements. These displacements can be obtained
by examining the joints at the top of the isolator
spring elements. If in close agreement, go to Step
B2.9. Otherwise go to Step B2.8.
The new superstructure displacement is 1.67 in, a 7%
difference from the displacement assumed at the start
of the Multimode Spectral Analysis.
Go to Step B2.8 and update properties for a second
cycle of iteration.

B2.8 Update Kisol,i, Keff,j, and BL B2.8 Update Kisol,i, Keff,j, and BL, Example 1.6
Use the calculated displacements in each isolator Updated values for Kisol,I (per isolator) are given
element to obtain new values of Kisol,i for each below (previous values are in parentheses):
isolator as follows:
, o Kisol,1 = 0.48 (0.46) k/in
, , (B-24) o Kisol,1 = 2.73 (2.60) k/in
,
o Kisol,1 = 2.73 (2.60) k/in
Recalculate Keff,j : o Kisol,1 = 0.48 (0.46) k/in

Eq. , ,
Updated values for Keff,j(per support), , BL and Teff are
7.1-6 ,
(B-25) given below (previous values are in parentheses):
, ,
GSID

179

 o Keff,1 = 2.86 (2.74) k/in

Recalculate system damping ratio, : o Keff,2 = 15.80 (14.82) k/in
o Keff,3 = 15.80 (14.82) k/in
Eq. 2 , , , o Keff,4 = 2.86 (2.74) k/in
7.1-10 , , ,
(B-26)
GSID o = 28% (33%)
o BL = 1.68 (1.70)
Recalculate system damping factor, BL: o Teff = 1.47 (1.55) sec
The updated composite response spectrum is shown
Eq. .
0.3 below:
.
7.1-3 ( B-27) 0.9
1.7 0.3
GSID 0.8

0.7

Acceleration(g)
Obtain the effective period of the bridge from the 0.6

multi-modal analysis and with the revised damping 0.5

0.4
factor (Eq. B-27), construct a new composite
0.3
response spectrum. Go to Step B2.6.
0.2

0.1

0.0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Period(s)

B2.9 Superstructure and Isolator Displacements B2.9 Superstructure and Isolator Displacements,
Once convergence has been reached, obtain Example 1.6
o superstructure displacements in the From the above analysis:
longitudinal (xL) and transverse (yL) directions o superstructure displacements in the
of the bridge, and longitudinal (xL) and transverse (yL) directions
o isolator displacements in the longitudinal (uL) are:
and transverse (vL) directions of the bridge, for xL= 1.17 in
each isolator, for this load case (i.e. yL= 1.17 in
longitudinal loading). These displacements
may be found by subtracting the nodal o isolator displacements in the longitudinal (uL)
displacements at each end of each isolator and transverse (vL) directions are:
spring element. o Abutments: uL = 1.17 in, vL = 1.17 in
o Pier2: uL = 1.09 in, vL = 1.09 in

All isolators at same support have the same

displacements.

B2.10 Pier Bending Moments and Shear Forces
Obtain the pier bending moments and shear forces in
the longitudinal (MPLL, VPLL) and transverse (MPTL,
VPTL) directions at the critical locations for the
longitudinally-applied seismic loading.
B2.10 Pier Bending Moments and Shear Forces,
Example 1.6
Maximum bending moments and shear forces in the
columns in the longitudinal (MPLL,VPLL) and
transverse (MPTL,VPTL) directions are:

Exterior Columns:
o MPLL= 170 kft
o MPTL= 170 kft
o VPLL= 15.34 k
o VPTL= 15.28 k
Interior Columns:
o MPLL= 147 kft
o MPTL= 147 kft
o VPLL= 13.40 k
o VPTL= 13.34 k
Both piers have the same distribution of bending

180

 moments and shear forces among the columns.

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 1.6
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1

Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces
ator
ure to long. to long. due to
EQ EQ long. EQ
1 0.56 0.55 1.10
2 0.56 0.56 1.07
North 3 0.56 0.56 0.79
Abut
ment 4 0.56 0.56 0.42
5 0.56 0.56 0.79
6 0.56 0.56 1.18
1 2.98 2.95 5.69
2 2.98 2.96 5.84
3 2.99 2.97 0.16
Pier 1
4 2.99 2.97 0.42
5 2.98 2.96 4.91
6 2.97 2.95 6.43
1 2.97 2.95 6.43
2 2.98 2.96 4.91
3 2.99 2.98 0.42
Pier 2
4 2.99 2.98 0.16
5 2.98 2.96 5.84
6 2.97 2.95 5.69
1 0.56 0.56 1.18
2 0.56 0.56 0.79
South 3 0.56 0.56 0.41
Abut
ment 4 0.56 0.56 0.79
5 0.56 0.56 1.06
6 0.56 0.56 1.10

181

 STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake
Repeat the above process, starting at Step B1, for
earthquake loading in the transverse direction of the
bridge. Support flexibility in the transverse direction
is to be included, and a composite response spectrum
is to be applied in the transverse direction. Obtain
isolator displacements in the longitudinal (uT) and
transverse (vT) directions of the bridge, and the
biaxial bending moments and shear forces at critical
locations in the columns due to the transversely-
applied seismic loading.
C1. Analysis for Transverse Earthquake,
Example 1.6
Key results from repeating Steps B1 and B2
(Simplified and Mulitmode Spectral Methods) are:
o Teff = 1.40 sec
o Maximum superstructure displacements in the
longitudinal (xT) and transverse (yT) directions are
as follows:
North Abutment xT = 1.17 and yT = 1.18 in
Pier 1 xT = 1.18 and yT = 1.19 in
Pier 1 xT = 1.18 and yT = 1.19 in
North Abutment xT = 1.18 and yT = 1.19 in

o Maximum isolator displacements in the

longitudinal (uT) and transverse (vT) directions are
as follows:

North Abutment uT = 1.18 and vT = 1.20 in

Pier 1 uT = 1.10 and vT = 1.11 in
Pier 1 uT = 1.10 and vT = 1.10 in
North Abutment uT = 1.19 and vT = 1.19 in

o Maximum bending moments and shear forces in

the columns in the longitudinal (MPLL,VPLL) and
transverse (MPTL,VPTL) directions are:
Exterior Columns:
o MPLT= 171 kft
o MPTT= 171 kft
o VPLT= 15.30 k
o VPTT= 15.31 k
Interior Columns:
o MPLT= 147 kft
o MPTT= 147 kft
o VPLT= 13.23 k
o VPTT= 13.23 k

o Isolator shear and axial forces are in Table C1-1.

182

 Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces due
ator
ure to transv. to transv. to transv.
EQ EQ EQ
1 0.56 0.57 1.19
2 0.56 0.57 1.26
North 3 0.56 0.57 0.98
Abut
ment 4 0.56 0.57 0.49
5 0.56 0.57 0.89
6 0.56 0.57 1.13
1 2.98 3.00 5.88
2 3.00 3.02 6.66
3 3.01 3.03 0.30
Pier 1
4 3.01 3.03 0.55
5 3.01 3.01 5.49
6 3.00 3.00 6.69
1 2.97 2.98 6.69
2 2.99 3.00 5.49
3 3.00 3.01 0.55
Pier 2
4 3.01 3.01 0.30
5 3.00 2.99 6.66
6 2.99 2.97 5.88
1 0.56 0.56 1.13
2 0.56 0.56 0.88
South 3 0.56 0.56 0.48
Abut
ment 4 0.56 0.56 0.97
5 0.56 0.56 1.26
6 0.56 0.56 1.19

183

STEP D. CALCULATE DESIGN VALUES

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements at Pier 1,

Following the provisions in Art. 2.1 GSID, and Example 1.6
illustrated in Fig. 2.1-1 GSID, calculate the total To illustrate the process, design displacements for the
design displacement, dt, for each isolator by outside isolator on Pier 1 are calculated below.
combining the displacements from the longitudinal
(uL and vL) and transverse (uT and vT) cases as Load Case 1:
follows: u1 = uL + 0.3uT = 1.0(1.09) + 0.3(1.09) = 1.42 in
u1 = uL + 0.3uT (D-1) v1 = vL + 0.3vT = 1.0(1.08) + 0.3(1.10) = 1.41 in
v1 = vL + 0.3vT (D-2) R1 = = 1.42 1.41 = 2.00 in
R1 = (D-3)
Load Case 2:
u2 = 0.3uL + uT (D-4) u2 = 0.3uL + uT = 0.3(1.09) + 1.0(1.09) = 1.42 in
v2 = 0.3vL + vT (D-5) v2 = 0.3vL + vT = 0.3(1.08) + 1.0(1.10) = 1.42 in
R2 = (D-6) R2 = = 1.42 1.42 = 2.01 in

dt = max(R1, R2) (D-7) Governing Case:

D2. Design Moments and Shears
Calculate design values for column bending moments
and shear forces for all piers using the same
combination rules as for displacements.
Total design displacement, dt = max(R1, R2)
= 2.01in
D2. Design Moments and Shears in Pier 1,
Example 1.6
Design moments and shear forces are calculated for
Pier 1, Column 1, below to illustrate the process.

Alternatively this step may be deferred because the Load Case 1:

above results may not be final. Upper and lower VPL1= VPLL + 0.3VPLT = 1.0(15.29) + 0.3(15.18)
bound analyses are required after the isolators have = 19.85 k
been designed as described in Art 7. GSID. These VPT1= VPTL + 0.3VPTT = 1.0(15.22) + 0.3(15.16)
analyses are required to determine the effect of = 19.77 k
possible variations in isolator properties due age, R1 = = 19.85 19.77 = 28.01 k
temperature and scragging in elastomeric systems.
Accordingly the results for column shear in Steps Load Case 2:
B2.10 and C are likely to increase once these VPL2= 0.3VPLL + VPLT = 0.3(15.29) + 1.0(15.18)
analyses are complete. = 19.77 k
VPT2= 0.3VPTL + VPTT = 0.3(15.22) + 1.0(15.16)
= 19.73 k
R2 = = 19.77 19.73 = 27.93 k

Governing Case:
Design column shear = max (R1, R2)
= 28.01 k

184

STEP E. DESIGN OF LEAD-RUBBER (ELASTOMERIC) ISOLATORS

A lead-rubber isolator is an elastomeric
bearing with a lead core inserted on its vertical
centreline. When the bearing and lead core are
deformed in shear, the elastic stiffness of the
lead provides the initial stiffness (Ku).With
increasing lateral load the lead yields almost
perfectly plastically, and the post-yield
stiffness Kd is given by the rubber alone. More
details are given in MCEER 2006.

While both circular and rectangular bearings

are commercially available, circular bearings
are more commonly used. Consequently the
procedure given below focuses on circular
bearings. The same steps can be followed for
rectangular bearings, but some modifications
will be necessary. When sizing the physical dimensions of the bearing, plan dimensions (B, dL) should be rounded
up to the next 1/4 increment, while the total thickness of elastomer, Tr, is specified in multiples of the layer
thickness. Typical layer thicknesses for bearings with lead cores are 1/4 and 3/8. High quality natural rubber
should be specified for the elastomer. It should have a shear modulus in the range 60-120 psi and an ultimate
elongation-at-break in excess of 5.5. Details can be found in rubber handbooks or in MCEER 2006.

The following design procedure assumes the isolators are bolted to the masonry and sole plates. Isolators that use
shear-only connections (and not bolts) require additional design checks for stability which are not included below.
See MCEER 2006.

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

E1. Required Properties E1. Required Properties, Example 1.6

Obtain from previous work the properties required of The design of one of the exterior isolators on a pier is
the isolation system to achieve the specified given below to illustrate the design process for lead-
performance criteria (Step A1). rubber isolators.
required characteristic strength, Qd / isolator
required post-elastic stiffness, Kd / isolator From previous work
total design displacement, dt, for each isolator Qd/isolator = 2.39 k
maximum applied dead and live load (PDL, Kd/isolator = 1.20 k/in
PLL) and seismic load (PSL) which includes Total design displacement, dt = 2.01 in
seismic live load (if any) and overturning PDL = 38.42k
forces due to seismic loads, at each isolator, PLL = 12.37 k
and PSL = 6.69 k
maximum wind load, PWL PWL = 1.76 k < Qd OK

E2. Isolator Sizing
E2.1 Lead Core Diameter E2.1 Lead Core Diameter, Example 1.6
Determine the required diameter of the lead plug, dL,
using:
2.39
(E-1) 1.63
0.9 0.9 0.9
See Step E2.5 for limitations on dL

185

E2.2 Plan Area and Isolator Diameter E2.2 Plan Area and Isolator Diameter, Example
Although no limits are placed on compressive stress 1.6
in the GSID, (maximum strain criteria are used
instead, see Step E3) it is useful to begin the sizing
process by assuming an allowable stress of, say, 1.0
ksi.

Then the bonded area of the isolator is given by:

(E-2)
1.0 45.52 15.50
50.79
1.0 1.0
and the corresponding bonded diameter (taking into
account the hole required to accommodate the lead
core) is given by:
4
(E-3)
4 4 50.79
1.63

Round the bonded diameter, B, to nearest quarter = 8.21 in

inch, and recalculate actual bonded area using
Round B up to 8.25 in and the actual bonded area is:
(E-4)
4
8.25 1.63 51.37
4
Note that the overall diameter is equal to the bonded
diameter plus the thickness of the side cover layers
(usually 1/2 inch each side). In this case the overall
diameter, Bo is given by:

1.0 (E-5)
Bo = 8.25 + 2(0.5) = 9.25 in

E2.3 Elastomer Thickness and Number of Layers E2.3 Elastomer Thickness and Number of Layers,
Since the shear stiffness of the elastomeric bearing is Example 1.6
given by:
(E-6)
where G = shear modulus of the rubber, and
Tr = the total thickness of elastomer, Select G, shear modulus of rubber, = 100 psi (0.1 ksi)
it follows Eq. E-5 may be used to obtain Tr given a
required value for Kd Then
0.1 51.37
(E-7) 4.29
1.20
A typical range for shear modulus, G, is 60-120 psi.
Higher and lower values are available and are used in
special applications.

If the layer thickness is tr, the number of layers, n, is

given by:
4.29
(E-8) 17.16
0.25
rounded up to the nearest integer.
Round to nearest integer, i.e. n = 18
Note that because of rounding the plan dimensions

186

and the number of layers, the actual stiffness, Kd, will

not be exactly as required. Reanalysis may be
necessary if the differences are large.

E2.4 Overall Height E2.4 Overall Height, Example 1.6

The overall height of the isolator, H, is given by:

1 2 ( E-9) 18 0.25 17 0.125 2 1.5 9.625

where ts = thickness of an internal shim (usually

about 1/8 in), and
tc = combined thickness of end cover plate
(0.5 in) and outer plate (1.0 in)

E2.5 Size Checks E2.5 Size Checks, Example 1.6

Experience has shown that for optimum performance Since B=8.25 check
of the lead core it must not be too small or too large.
The recommended range for the diameter is as 8.25 8.25
follows: 3 6
(E-10)
3 6 i.e., 2.75 1.38

Since dL = 1.63, lead core size is acceptable.

Art. 12.2 GSID requires that the isolation system 0.025 0.025 38.42
provides a lateral restoring force at dt greater than the , 0.44 /
2.17
restoring force at 0.5dt by not less than W/80. This
equates to a minimum Kd of 0.025W/d. As
0.1 51.37
0.025 1.14 / ,
4.5
,

E3. Strain Limit Check E3. Strain Limit Check, Example 1.6
Art. 14.2 and 14.3 GSID requires that the total
applied shear strain from all sources in a single layer
of elastomer should not exceed 5.5, i.e.,
Since
, + 0.5 5.5 (E-11)
38.42
0.75
where , , , are defined below. 51.37
(a) is the maximum shear strain in the layer due to
compression and is given by: G = 0.1 ksi

(E-12) and
where Dc is shape coefficient for compression in 51.37
7.93
circular bearings = 1.0, , G is shear 8.25 0.25

modulus, and S is the layer shape factor given by: then

(E-13) 1.0 0.75
0.943
0.1 7.93
(b) , is the shear strain due to earthquake loads
and is given by:

187

 2.01
, (E-14) , 0.446
4.5

(c) is the shear strain due to rotation and is given

by:
0.375 8.25 0.01
(E-15) 0.227
0.25 4.5
where Dr is shape coefficient for rotation in circular
bearings = 0.375, and is design rotation due to DL, Substitution in Eq E-11 gives
LL and construction effects Actual value for may
not be known at this time and value of 0.01 is , 0.5 0.943 0.446 0.5 0.227
suggested as an interim measure, including 1.50 5.5
uncertainties (see LRFD Art. 14.4.2.1)

E4. Vertical Load Stability Check E4. Vertical Load Stability Check, Example 1.6
Art 12.3 GSID requires the vertical load capacity of
all isolators be at least 3 times the applied vertical
loads (DL and LL) in the laterally undeformed state.

Further, the isolation system shall be stable under

1.2(DL+SL) at a horizontal displacement equal to
either
2 x total design displacement, dt, if in Seismic Zone 1
or 2, or
1.5 x total design displacement, dt, if in Seismic Zone
3 or 4.

E4.1 Vertical Load Stability in Undeformed State E4.1 Vertical Load Stability in Undeformed State,
The critical load capacity of an elastomeric isolator at Example 1.6
zero shear displacement is given by

4
1 1 3 3 0.1 0.3
2 (E-16)
0.3 1 0.67 7.93 12.93
where
8.25
227.40
Ts = total shim thickness 64

12.93 227.40
653.56 /
1 0.67 4.5
E = elastic modulus of elastomer = 3G
0.1 51.37
64 1.14 /
4.5
It is noted that typical elastomeric isolators have high
shape factors, S, in which case:
4
1 (E-17)

and Eq. E-16 reduces to:

(E-18) 1.14 653.56 85.81

Check that:

188


3 (E-19) 85.81

1.69 3
38.42 12.37

E4.2 Vertical Load Stability in Deformed State E4.2 Vertical Load Stability in Deformed State,
The critical load capacity of an elastomeric isolator at Example 1.6
shear displacement may be approximated by: Since bridge is in Zone 2, 2 2 2.01 4.02

(E-20) 4.02
2 2.13
where 8.25
Ar = overlap area between top and bottom plates
of isolator at displacement (Fig. 2.2-1
GSID)
2.13 2.13
= 0.406
4
2

Agross = 4
0.406 85.81 34.83
It follows that:
(E-21)

Check that:
34.83
1 (E-22) 0.66 1
1.2 1.2 1.2 38.42 6.69

E5. Design Review E5. Design Review, Example 1.6

The basic dimensions of the isolator designed above are
as follows:

9.25 in (od) x 9.625 in (high) x 1.63 in dia. lead core and

the volume, excluding steel end and cover plates, is 445
in3.

This design satisfies the shear strain limit criteria, but

not the vertical load stability ratio in the undeformed and
deformed states.

A redesign is therefore required and the easiest way to

increase the Pcr is to increase the shape factor, S, since
the bending stiffness of an isolator is a function of the
shape factor squared. See equations in Step E4.1. To
increase S, increase the bonded area Ab while keeping tr
constant (Eq. E-13). But to keep Kd constant while
increasing Ab and Tr is constant, decrease the shear
modulus, G (Eq. E-6).

This redesign is outlined below. After repeating the

calculation for diameter of lead core, the process begins
by reducing the shear modulus to 60 psi (0.06 ksi) and
increasing the bonded diameter to 11 in.

189

 E2.1
2.39
1.63
0.9 0.9

E2.2
4.5 1.20
90
0.06

4 4 90
1.63 10.83

Round B to 11 in and the actual bonded area becomes:

11 1.61 92.95
4

Bo = 11 + 2(0.5) = 12 in

E2.3
0.06 92.95
4.66
1.20

4.66
18.63
0.25

Round to nearest integer, i.e. n = 19.

E2.4
19 0.25 18 0.125 2 1.5 10
E2.5
Since B=11 check

11 11
3 6

i.e., 3.67 1.83

Since dL = 1.63, the size of lead core is too small, and

there are 2 options: (1) Accept the undersize and check
for adequate performance during the Quality Control
Tests required by GSID Art. 15.2.2; or (2) Only have
lead cores in every second isolator, in which case the
core diameter, in those isolators with cores, will be 2 x
1.63 = 2.31 in (which satisfies above criterion).

0.06 92.95
1.17 / ,
4.75

E3.
38.42
0.41
92.95

92.95
10.76
11 0.25

190

 1.0 0.41
0.640
0.06 10.76

2.01
, 0.422
4.75

0.375 11 0.01
0.382
0.25 4.75

, 0.5 0.640 0.422 0.5 0.382

1.25 5.5
E4.1
3 3 0.06 0.18

0.18 1 0.67 10.76 14.14

11
718.69
64
14.14 718.69
2139.24 /
4.75

0.06 92.95
1.174 /
4.75

1.174 2139.24 157.44

157.44
3.10 3
38.42 12.37

E4.2
4.02
2 2.39
11

2.39 2.39
0.546

0.546 157.44 85.96

85.96
1.63 1
1.2 1.2 38.42 6.69

E5.
The basic dimensions of the redesigned isolator are as
follows:

12 in (od) x 10 in (high) x 1.63 in dia. lead core and its

volume (excluding steel end and cover plates) is 792 in3.

191

 This design meets all the design criteria but is about 75%
larger by volume than the previous design. This increase
in size is dictated by the need to satisfy the vertical load
stability ratio of 3.0 in the undeformed state.

E6. Minimum and Maximum Performance Check E6. Minimum and Maximum Performance Check,
Art. 8 GSID requires the performance of any Example 1.6
isolation system be checked using minimum and Minimum Property Modification factors are:
maximum values for the effective stiffness of the min,Kd = 1.0
system. These values are calculated from minimum min,Qd = 1.0
and maximum values of Kd and Qd, which are found
using system property modification factors, as which means there is no need to reanalyze the bridge
indicated in Table E6-1. with a set of minimum values.
Table E6-1. Minimum and maximum values
for Kd and Qd. Maximum Property Modification factors are:
max,a,Kd = 1.1
Eq. max,a,Qd = 1.1
8.1.2-1 Kd,max = Kd max,Kd (E-23)
GSID max,t,Kd = 1.1
Eq. max,t,Qd = 1.4
8.1.2-2 Kd,min = Kd min,Kd (E-24)
GSID max,scrag,Kd = 1.0
Eq. max,scrag,Qd = 1.0
8.1.2-3 Qd,max = Qd max,Qd (E-25)
GSID
Eq.
8.1.2-4 Qd,min = Qd min,Qd (E-26)
GSID

Determination of the system property modification

factors should include consideration of the effects of
temperature, aging, scragging, velocity, travel (wear)
and contamination as shown in Table E6-2. In lieu of
tests, numerical values for these factors can be
obtained from Appendix A, GSID.

Table E6-2. Minimum and maximum values for

system property modification factors.

Eq. min,Kd = (min,t,Kd) (min,a,Kd)

8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-27)
GSID (min,scrag,Kd)
Eq. max,Kd = (max,t,Kd) (max,a,Kd)
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-28)
GSID (max,scrag,Kd)
Eq. min,Qd = (min,t,Qd) (min,a,Qd)
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-29)
GSID (min,scrag,Qd)
Eq. max,Qd = (max,t,Qd) (max,a,Qd)
8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd) (E-30)
GSID (max,scrag,Qd)

Adjustment factors are applied to individual -factors Applying a system adjustment factor of 0.66 for an
(except v) to account for the likelihood of other bridge, the maximum property modification
occurrence of all of the maxima (or all of the factors become:

192

minima) at the same time. These factors are applied
to all -factors that deviate from unity but only to the
portion of the -factor that is greater than, or less
than, unity. Art. 8.2.2 GSID gives these factors as
follows:
1.00 for critical bridges
0.75 for essential bridges
0.66 for all other bridges
max,a,Kd = 1.0 + 0.1(0.66) = 1.066
max,a,Qd = 1.0 + 0.1(0.66) = 1.066
max,t,Kd = 1.0 + 0.1(0.66) = 1.066
max,t,Qd = 1.0 + 0.4(0.66) = 1.264
max,scrag,Kd = 1.0
max,scrag,Qd = 1.0

As required in Art. 7 GSID and shown in Fig. C7-1
GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Qd,max. As indicated in Fig C7-1 GSID, maximum
displacements will probably be given by the first case
(Kd,min and Qd,min) and maximum forces by the second
case (Kd,max and Qd,max).
Therefore the maximum overall modification factors
max,Kd = 1.066(1.066)1.0 = 1.14
max,Qd = 1.066(1.264)1.0 = 1.35
Since the possible variation in upper bound properties
exceeds 15% a reanalysis of the bridge is required to
determine performance with these properties.

The upper-bound properties are:

Qd,max = 1.35 (2.39) = 3.23 k
and
Kd,ma x=1.14(1.17) = 1.34 k/in

E7. Design and Performance Summary E7. Design and Performance Summary, Example 1.6
E7.1 Isolator dimensions E7.1 Isolator dimensions, Example 1.6
Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E7.1-1.
Overall diameter (includes cover layer)
Overall height Table E7.1-1 Isolator Dimensions
Diameter lead core
Isolator Overall size Overall size Diam.
Bonded diameter Location including without lead
Number of rubber layers mounting mounting core
Thickness of rubber layers plates (in) plates (in) (in)
Total rubber thickness Under
Thickness of steel shims edge 16.0 x 16.0 12.0 dia.
Shear modulus of elastomer 1.63
girder x 10.0 (H) x 8.5(H)
on Pier 1

Isolator No. of Rubber Total Steel

Location rubber layers rubber shim
layers thick- thick- thick-
ness ness ness
(in) (in) (in)
Under
edge
19 0.25 4.75 0.125
girder
on Pier 1

Shear modulus of elastomer = 60 psi

E7.2 Bridge Performance E7.2 Bridge Performance, Example 1.6

Summarize bridge performance Bridge performance is summarized in Table E7.2-1
Maximum superstructure displacement where it is seen that the maximum column shear is
(longitudinal) 28.32 k. This is more than the column plastic shear

193

 Maximum superstructure displacement
(transverse)
Maximum superstructure displacement
(resultant)
Maximum column shear (resultant)
Maximum column moment (about transverse
axis)
Maximum column moment (about
longitudinal axis)
Maximum column torque
strength (25 k) and therefore the required performance
criterion is not satisfied (fully elastic behavior). Clearly
the additional column shear forces due to the skew are
too high to be reduced to below yield and give a fully
elastic response. However the displacement is only 1.61
in and the displacement ductility demand on the columns
is likely to be less than 2, thus indicating essentially
elastic behavior. If this is not acceptable, options
include: (1) jacketing the column if an existing bridge, or
(2) increasing the size of the column if a new bridge.

Check required performance as determined in Step It is noted that the maximum longitudinal displacement
A3, is satisfied. is 1.54 in which is less than the 2.0 in available at the
abutment expansion joints and therefore acceptable.

Table E7.2-1 Summary of Bridge Performance

Maximum superstructure
1.54 in
displacement (longitudinal)
Maximum superstructure
1.54 in
displacement (transverse)
Maximum superstructure
1.61 in
displacement (resultant)
Maximum column shear (resultant) 28.32 k
Maximum column moment about
227 kft
transverse axis
Maximum column moment about
227 kft
longitudinal axis
Maximum column torque 23.25 kft

194

SECTION 2
Steel Plate Girder Bridge, Long Spans, Single-Column Pier

DESIGN EXAMPLE 2.0: Benchmark Bridge #2

Design Examples in Section 2

Site
ID Description S1 Column height Skew Isolator type
Class
2.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
2.1 Change site class 0.2g D Same 0 Lead rubber bearing
2.2 Change spectral acceleration, S1 0.6g B Same 0 Lead rubber bearing
Spherical friction
2.3 Change isolator to SFB 0.2g B Same 0
bearing
2.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
2.5 Change column height 0.2g B H1=0.5H2 0 Lead rubber bearing
0
2.6 Change angle of skew 0.2g B Same 45 Lead rubber bearing

195

DESIGN PROCEDURE DESIGN EXAMPLE 2.0 (Benchmark #2)

STEP A: BRIDGE AND SITE DATA

A1. Bridge Properties
Determine properties of the bridge:
number of supports, m
number of girders per support, n
angle of skew
weight of superstructure including railings,
curbs, barriers and other permanent loads,
WSS
weight of piers participating with
superstructure in dynamic response, WPP
weight of superstructure, Wj, at each
support
stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge. The calculation of these
quantities requires careful consideration of
several factors such as the use of cracked
sections when estimating column or wall
flexural stiffness, foundation flexibility,
and effective column height.
column shear strength (minimum value).
This will usually be derived from the
minimum value of the column flexural
yield strength, the column height, and
whether the column is acting in single or
double curvature in the direction under
consideration.
allowable movement at expansion joints
isolator type if known, otherwise to be
selected
A1. Bridge Properties, Example 2.0
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 3
Angle of skew = 00
Number of columns per support = 1
Weight of superstructure including permanent
loads, WSS = 1651.32 k
Weight of superstructure at each support:
o W1 = 168.48 k
o W2 = 657.18 k
o W3 = 657.18 k
o W4 = 168.48 k
Participating weight of piers, WPP = 256.26 k
Effective weight (for calculation of period),
Weff = Wss + WPP = 1907.58 k
Pier heights are both 19ft (clear)
Stiffness of each pier in the both directions
(assume fixed at footing and single curvature
behavior) :
o Ksub,pier1 = 288.87 k/in
o Ksub,pier2 = 288.87 k/in
Minimum column shear strength based on
flexural yield capacity of column = 128 k
Displacement capacity of expansion joints
(longitudinal) = 2.5 in for thermal and other
movements
Lead-rubber isolators

A2. Seismic Hazard A2. Seismic Hazard, Example 2.0

Determine seismic hazard at site: Acceleration coefficients for bridge site are given in
acceleration coefficients design statement as follows:
site class and site factors PGA = 0.40
seismic zone S1 = 0.20
Plot response spectrum. SS = 0.75

Use Art. 3.1 GSID to obtain peak ground and spectral Bridge is on a rock site with shear wave velocity in
acceleration coefficients. These coefficients are the upper 100 ft of soil = 3,000 ft/sec.
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD Table 3.10.3.1-1 LRFD gives Site Class as B.
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (e.g. Figures Tables 3.10.3.2-1, -2 and -3 LRFD give following
3.10.2.1-1 to 3.10.2.1-21 LRFD). Site Factors:
Fpga = 1.0
Use Art. 3.2 to obtain Site Class and corresponding Site Fa = 1.0
Factors (Fpga, Fa and Fv). These data are the same as for Fv = 1.0
conventional bridges and Art 3.2 refers the designer to

196

the corresponding articles in the LRFD Specifications,

i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3,
LRFD.

Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give

modified spectral acceleration coefficients that include
site effects as follows: As = Fpga PGA = 1.0(0.40) = 0.40
As = Fpga PGA SDS = Fa SS = 1.0(0.75) = 0.75
SDS = Fa SS SD1 = Fv S1 = 1.0(0.20) = 0.20
SD1 = Fv S1

Seismic Zone is determined by value of SD1 in Since 0.15 < SD1 < 0.30, bridge is located in Seismic
accordance with provisions in Table 5-1 GSID. Zone 2.

These coefficients are used to plot design response Design Response Spectrum is as below:
spectrum as shown in Fig. 4-1 GSID.
0.9

0.8

0.7

0.6

Acceleration
0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Period(s)

A3. Performance Requirements A3. Performance Requirements, Example 2.0

Determine required performance of isolated bridge In this example, assume the owner has specified full
during Design Earthquake (1000-yr return period). functionality following the earthquake and therefore
the columns must remain elastic (no yield).
Examples of performance that might be specified by
the Owner include: To remain elastic the maximum lateral load on the
Reduced displacement ductility demand in pier must be less than the load to yield any one
columns, so that bridge is open for emergency column (128 k).
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in The maximum shear in the column must therefore be
columns or yield elsewhere in bridge), so that less than 128 k in order to keep the column elastic
bridge is fully functional and open to all vehicles and meet the required performance criterion.
immediately following earthquake.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.

197

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for

Isolator Force, F
non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most
have bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use Ku
equivalent linear springs and viscous damping to
represent the isolators, so that linear methods of analysis
may be used to determine response. Since equivalent dy disol
Isolator
Ku Ku
properties such as Kisol are dependent on displacement Displacement, d

(d), and the displacements are not known at the beginning

of the analysis, an iterative approach is required. Note Kd
that in Art 7.1, GSID, keff is used for the effective stiffness
of an isolator unit and Keff is used for the effective disol = Isolator displacement
stiffness of a combined isolator and substructure unit. To dy = Isolator yield displacement
Fisol = Isolator shear force
minimize confusion, Kisol is used in this document in Fy = Isolator yield force
place of keff. There is no change in the use of Keff and Keff,j, Kd
Kisol
= Post-elastic stiffness of isolator
= Effective stiffness of isolator
but Ksub is used in place of ksub. Ku = Loading and unloading stiffness (elastic stiffness)
Qd = Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement (disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 2.0
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.20 2.0
GSID

(2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not

198

occur under non-seismic loads (e.g. wind) but

low enough that yield will occur during an
earthquake. Experience has shown that
taking Qd to be 5% of the bridge weight is a
good starting point, i.e.
0.05 (B-2)

(3) Post-yield stiffness, Kd 0.05 0.05 1651.32 82.56

Art 12.2 GSID requires that all isolators
exhibit a minimum lateral restoring force at
the design displacement, which translates to a
minimum post yield stiffness Kd,min given by
Eq. B-3.

Art.
0.025
12.2 , (B-3)
GSID 1651.32
Experience has shown that a good starting 0.05 0.05 41.28 /
2.0
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 2.0
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength, ,

Qd, and stiffness, Kd, values in proportion to the dead o Qd, 1 = 8.42 k
load applied at that support: o Qd, 2 = 32.86 k
o Qd, 3 = 32.86 k
o Qd, 4 = 8.42 k

, (B-4)
and
and
,
, (B-5) o Kd,1 = 4.21 k/in
o Kd,2 = 16.43 k/in
o Kd,3 = 16.43 k/in
o Kd,4 = 4.21 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 2.0
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of , ,
the isolators at support j (Kisol,j) and the stiffness of , ,
the substructure Ksub,j. See figure below for
definitions (after Fig. 7.1-1 GSID). o 1 = 8.43x10-4
o 2 = 1.21x10-1
An expression for Keff,j, is given in Eq.7.1-6 GSID, but o 3 = 1.21x10-1
a more useful formula is as follows (MCEER 2006): o 4 = 8.43x10-4
,
, (B-6)
1 ,
where ,
1
, ,
(B-7) o Keff,1 = 8.42 k/in
, , o Keff,2 = 31.09 k/in
o Keff,3 = 31.09 k/in
and Ksub,j for the piers are given in Step A1. For the o Keff,4 = 8.42 k/in

199

abutments, take Ksub,j to be a large number, say 10,000

k/in, unless actual stiffness values are available. Note
that if the default option is chosen, unrealistically high
values for Ksub,j will give unconservative results for
column moments and shear forces.
F
Kd
Qd Kisol

dy disol
Superstructure
Isolator Effective Stiffness, Kisol

F
Isolator(s), Kisol
Ksub

Substructure, Ksub
dsub

Substructure Stiffness, Ksub

dsub disol F

d
Keff

d = disol + dsub

Combined Effective Stiffness, Keff

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 2.0
Calculate the total effective stiffness, Keff, of the
bridge:

Eq. , 79.02 /
7.1-6 , ( B-8)
GSID

B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 2.0
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1
, (B-9)
1 o disol,1 = 2.00 in
o disol,2 = 1.79 in
o disol,3 = 1.79 in
o disol,4 = 2.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the stiffness of the isolation system at Example 2.0
support j, Kisol,j, for all supports:
,
, ,
,
,
(B-10) o Kisol,1 = 8.43 k/in
, ,
, o Kisol,2 = 34.84 k/in
o Kisol,3 = 34.84 k/in
o Kisol,4 = 8.43 k/in

200



B1.7 Substructure Displacement at Each Support B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j, Example 2.0
for all supports:
, ,

, , (B-11) o dsub,1 = 0.002 in

o d sub,2 = 0.215 in
o d sub,3 = 0.215 in
o d sub,4 = 0.002 in
B1.8 Lateral Load in Each Substructure B1.8 Lateral Load in Each Substructure, Example
Calculate the lateral load in substructure j, Fsub,j, for 2.0
all supports:
, , , (B-12) , , ,

where values for Ksub,j are given in Step A1. o F sub,1 = 16.84 k
o F sub,2 = 62.18 k
o F sub,3 = 62.18 k
o F sub,4 = 16.84 k

B1.9 Column Shear Force at Each Support B1.9 Column Shear Force at Each Support,
Calculate the shear force in column k at support j, Example 2.0
Fcol,j,k, assuming equal distribution of shear for all
columns at support j:
,
, ,
#
,
, ,
# (B-13) o F col,2,1 = 62.18 k
o F col,3,1 = 62.18 k

Use these approximate column shear forces as a check These column shears are less than the plastic shear
on the validity of the chosen strength and stiffness capacity of each column (128k) as required in Step
characteristics. A3 and the chosen strength and stiffness values in
Step B1.1 are therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 2.0
damping ratio, , of the bridge:

Eq. 1907.58
7.1-5 2 (B-14) 2 2
GSID 386.4 79.02

and = 1.57 sec

Eq. 2 , , , and taking dy,j = 0:

7.1-10 (B-15)
, , ,
GSID 2 , , 0
0.30
, , ,
where dy,j is the yield displacement of the isolator. For
friction-based isolators, dy,j = 0. For other types of
isolators dy,j is usually small compared to disol,j and has
negligible effect on , Hence it is suggested that for
the Simplified Method, set dy,j = 0 for all isolator
types. See Step B2.2 where the value of dy,j is revisited

201

for the Multimode Spectral Analysis Method.

B1.11 Damping Factor B1.11 Damping Factor, Example 2.0

Calculate the damping factor, BL, and the Since 0.30 0.3
displacement, d, of the bridge:

Eq. .
, 0.3 1.70
.
7.1-3 (B-16)
1.7, 0.3
GSID and

9.79 9.79 0.2 1.57

Eq. 9.79 (B-17) 1.81
1.70
7.1-4
GSID
B1.12 Convergence Check B1.12 Convergence Check, Example 2.0
Compare the new displacement with the initial value Since the calculated value for displacement, d (=1.81)
assumed in Step B1.1. If there is close agreement, go is not close to that assumed at the beginning of the
to the next step; otherwise repeat the process from cycle (Step B1.1, d = 2.0), use the value of 1.81 as the
Step B1.3 with the new value for displacement as the new assumed displacement and repeat from Step
assumed displacement. B1.3.

This iterative process is amenable to solution using a
spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
After three iterations, convergence is reached at a
superstructure displacement of 1.65 in, with an
effective period of 1.43 seconds, and a damping factor
of 1.7 (30% damping ratio). The displacement in the
isolators at Pier 1 is 1.44 in and the effective stiffness
of the same isolators is 42.78 k/in.
See spreadsheet in Table B1.12-1 for results of final
iteration.
Ignoring the weight of the hammerhead, the column
shear force must equal the isolator shear force for
equilibrium. Hence column shear = 42.78 (1.44) =
61.60 k which is less than the maximum allowable
(128 k) if elastic behavior is to be achieved (as
required in Step A3).
Also the superstructure displacement = 1.65 in, which
is less than the available clearance of 2.5 in.

Therefore the above solution is acceptable and go to

Step B2.

Note that available clearance (2.5 in) is greater than

minimum required which is given by:

8 8 0.20 1.43
1.35
1.7

202

Table B1.12-1 Simplified Method Solution for Design Example 2.0 Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
1651.32 256.26 1907.58 0.2 3

StepB1.1 d 1.65 Assumeddisplacement

Qd 82.57 Characteristicstrength
Kd 50.04 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 168.48 8.424 5.105 10,000.00 0.001022 10.206 1.648 10.216 0.002 16.839 13.885 27.785
Pier1 657.18 32.859 19.915 288.87 0.148088 37.260 1.437 42.778 0.213 61.480 47.224 101.441
Pier2 657.18 32.859 19.915 288.87 0.148088 37.260 1.437 42.778 0.213 61.480 47.224 101.441
Abut2 168.48 8.424 5.105 10,000.00 0.001022 10.206 1.648 10.216 0.002 16.839 13.885 27.785
Total 1651.32 82.566 50.040 K eff,j 94.932 156.638 122.219 258.453
Step B1.4

StepB1.10 T eff 1.43 Effectiveperiod

0.30 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.71

B L 1.70 DampingFactor
d 1.65 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 2.808 1.702 3.405 1.69 3.363
Pier1 10.953 6.638 14.259 1.20 15.766
Pier2 10.953 6.638 14.259 1.20 15.766
Abut2 2.808 1.702 3.405 1.69 3.363

203

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and then
applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 2.0

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 (see
elastic stiffness, Kd,i, of each isolator i as follows: Table B1.12-1) by the number of isolators at each
, support (n = 3), the following values for Qd /isolator
, (B-19)
and Kd /isolator are obtained:
and o Qd, 1 = 8.42/3 = 2.81 k
, o Qd, 2 = 32.86/3=10.95 k
, ( B-20)
o Qd, 3 = 32.86/3 = 10.95 k
where values for Qd,j and Kd,j are obtained from the o Qd, 4 = 8.42/3 = 2.81 k
final cycle of iteration in the Simplified Method (Step
B1. 12 and
o Kd,1 = 5.10/3 = 1.70 k/in
o Kd,2 = 19.92/3 = 6.64 k/in
o Kd,3 = 19.92/3 = 6.64 k/in
o Kd,4 = 5.10/3 = 1.70 k/in

Note that the Kd values per support used above are

from the final iteration given in Table B1.12-1. These
are not the same as the initial values in Step B1.2,
because they have been adjusted from cycle to cycle,
such that the total Kd summed over all the isolators
satisfies the minimum lateral restoring force
requirement for the bridge, i.e. Kdtotal = 0.05 W/d. See
Step B1.1. Since d varies from cycle to cycle, Kd,j
varies from cycle to cycle.

B2.2 Initial Stiffness and Yield Displacement B2.2 Initial Stiffness and Yield Displacement,
Calculate the initial stiffness, Ku,i, and the yield Example 2.0
displacement, dy,i, for each isolator i as follows: Since the isolator type has been specified in Step A1
(1) For friction-based isolators Ku,i = and dy,i = 0. to be an elastomeric bearing (i.e. not a friction-based
(2) For other types of isolators, and in the absence of bearing), calculate Ku,i and dy,i for an isolator on Pier 1
isolator-specific information, take as follows:
, 10 , 10 6.64 66.4 /
, 10 , ( B-21)
and then and
, 10.95
,
( B-22) , 0.18
,
, , 66.4 6.64
, ,

As expected, the yield displacement is small

compared to the expected isolator displacement (~2

204

in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.

B2.3 Isolator Effective Stiffness, Kisol,i B2.3 Isolator Effective Stiffness, Kisol,i, Example 2.0
Calculate the isolator stiffness, Kisol,i, of each isolator Dividing the results for Kisol (Step B1.12) among the 3
i: isolators at each support, the following values for Kisol
/isolator are obtained:
, o Kisol,1 = 10.22/3 = 3.41 k/in
, (B -23)
o Kisol,2 = 42.78/3 = 14.26 k/in
o Kisol,3 = 42.78/3 = 14.26 k/in
o Kisol,4 = 10.22/3 = 3.41 k/in

B2.4 Three-Dimensional Bridge Model B2.4 Three-Dimensional Bridge Model, Example

Using computer-based structural analysis software, 2.0
create a 3-dimensional model of the bridge with the Although the bridge in this Design Example is regular
isolators represented by spring elements. The stiffness and is without skew or curvature, a 3-dimensional
of each isolator element in the horizontal axes (Kx and finite element model was developed for this Step, as
Ky in global coordinates, K2 and K3 in typical local shown below.
coordinates) is the Kisol value calculated in the
previous step. For bridges with regular geometry and
minimal skew or curvature, the superstructure may be .
represented by a single stick provided the load path
to each individual isolator at each support is explicitly
modeled, usually by a rigid cap beam and a set of rigid
links. If the geometry is irregular, or if the bridge is
skewed or curved, a finite element model is
recommended to accurately capture the load carried
by each individual isolator. If the piers have an
unusual weight distribution, such as a pier with a
hammerhead cap beam, a more rigorous model is
recommended.

B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in Step
A2 is for a 5% damped system. It is modified in this
step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the effective
period of the bridge, Teff, by the damping factor, BL.
B.2.5 Composite Design Response Spectrum,
Example 2.0
From the final results of Simplified Method (Step
B1.12), BL = 1.70 and Teff = 1.43 sec. Hence the
transition in the composite spectrum from 5% to 30%
damping occurs at 0.8 Teff = 0.8 (1.43) = 1.14 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.14 sec by 1.70.
0.8

0.7

0.6

0.5
Csm (g)

0.4

0.3

0.2

0.1

0
0 0.5 1 1.5 2 2.5 3 3.5 4
T (sec)

205

B2.6 Multimodal Analysis of Finite Element Model B2.6 Multimodal Analysis of Finite Element
Input the composite response spectrum as a user- Model, Example 2.0
specified spectrum in the software, and define a load Results of modal analysis of the example bridge are
case in which the spectrum is applied in the summarized in Table B2.6-1 Here the modal periods
longitudinal direction. Analyze the bridge for this and mass participation factors of the first 12 modes
load case. are given. The first three modes are the principal
transverse, longitudinal, and torsion modes with
periods of 1.60, 1.46 and 1.39 sec respectively. The
period of the longitudinal mode (1.46 sec) is very
close to that calculated in the Simplified Method. The
mass participation factors indicate there is no
coupling between these three modes (probably due to
the symmetric nature of the bridge) and the high
values for the first and second modes (92% and 94%
respectively) indicate the bridge is responding
essentially in a single mode of vibration in each
Mode Period Mass Participation Ratios
No Sec UX UY UZ RX RY RZ
1 1.604 0.000 0.919 0.000 0.952 0.000 0.697
2 1.463 0.941 0.000 0.000 0.000 0.020 0.000
3 1.394 0.000 0.000 0.000 0.000 0.000 0.231
4 0.479 0.000 0.003 0.000 0.013 0.000 0.002
5 0.372 0.000 0.000 0.076 0.000 0.057 0.000
6 0.346 0.000 0.000 0.000 0.000 0.000 0.000
7 0.345 0.000 0.001 0.000 0.010 0.000 0.000
8 0.279 0.000 0.003 0.000 0.013 0.000 0.002
9 0.268 0.000 0.000 0.000 0.000 0.000 0.000
10 0.267 0.058 0.000 0.000 0.000 0.000 0.000
11 0.208 0.000 0.000 0.000 0.000 0.129 0.000
12 0.188 0.000 0.000 0.000 0.000 0.000 0.001

direction. Similar results to that obtained by the

Simplified Method are therefore expected.

Table B2.6-1 Modal Properties of Bridge

Example 2.0 First Iteration
Computed values for the isolator displacements due to
a longitudinal earthquake are as follows (numbers in
parentheses are those used to calculate the initial
properties to start iteration from the Simplified
Method):

o disol,1 = 1.69 (1.65) in

o disol,2 = 1.20 (1.44) in
o disol,3 = 1.20 (1.44) in
o disol,4 = 1.69 (1.65) in

B2.7 Convergence Check B2.7 Convergence Check, Example 2.0

Compare the resulting displacements at the
superstructure level (d) to the assumed displacements.
These displacements can be obtained by examining
the joints at the top of the isolator spring elements. If
in close agreement, go to Step B2.9. Otherwise go to
Step B2.8.
The results for isolator displacements are close but
not close enough (15% difference at the piers)
Go to Step B2.8 and update properties for a second
cycle of iteration.

206

B2.8 Update Kisol,i, Keff,j, and BL B2.8 Update Kisol,i, Keff,j, and BL, Example 2.0
Use the calculated displacements in each isolator Updated values for Kisol,i are given below (previous
element to obtain new values of Kisol,i for each isolator values are in parentheses):
as follows:
, o Kisol,1 = 3.36 (3.41) k/in
, , (B-24) o Kisol,2 = 15.77 (14.26) k/in
,
o Kisol,3 = 15.77 (14.26) k/in
Recalculate Keff,j : o Kisol,4 = 3.36 (3.41) k/in

Eq. , ,
7.1-6 ,
(B-25) Since the isolator displacements are relatively close to
, ,
GSID previous results no significant change in the damping
ratio is expected. Hence Keff,j and are not
Recalculate system damping ratio, : recalculated and BL is taken at 1.70.

Eq. 2 , , ,
7.1-10 , , ,
(B-26)
GSID

Recalculate system damping factor, BL:

Eq. .
0.3
.
7.1-3 ( B-27)
1.7 0.3 Since the change in effective period is very small
GSID
(1.43 to 1.46 sec) and no change has been made to BL,
Obtain the effective period of the bridge from the there is no need to construct a new composite
multi-modal analysis and with the revised damping response spectrum in this case. Go back to Step B2.6
factor (Eq. B-27), construct a new composite response (see immediately below).
spectrum. Go to Step B2.6.
B2.6 Multimodal Analysis Second Iteration,
Example 2.0
Reanalysis gives the following values for the isolator
displacements (numbers in parentheses are those
from the previous cycle):

o disol,1 = 1.66 (1.69) in

o disol,2 = 1.15 (1.20) in
o disol,3 = 1.15 (1.20) in
o disol,4 = 1.66 (1.69) in

Go to Step B2.7

B2.7 Convergence Check B2.7 Convergence Check, Example 2.0

Compare results and determine if convergence has Satisfactory agreement has been reached on this
been reached. If so go to Step B2.9. Otherwise Go to second cycle. Go to Step B2.9
Step B2.8.

B2.9 Superstructure and Isolator Displacements B2.9 Superstructure and Isolator Displacements,
Once convergence has been reached, obtain Example 2.0
o superstructure displacements in the longitudinal From the above analysis:
(xL) and transverse (yL) directions of the bridge, o superstructure displacements in the
and longitudinal (xL) and transverse (yL) directions
o isolator displacements in the longitudinal (uL) are:
and transverse (vL) directions of the bridge, for xL= 1.69 in

207

each isolator, for this load case (i.e. yL= 0.0 in

longitudinal loading). These displacements
may be found by subtracting the nodal
displacements at each end of each isolator
spring element.
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions are:
o Abutments: uL = 1.66 in, vL = 0.00 in
o Piers: uL = 1.15 in, vL = 0.00 in

All isolators at same support have the same

displacements.

B2.10 Pier Bending Moments and Shear Forces B2.10 Pier Bending Moments and Shear Forces,
Obtain the pier bending moments and shear forces in Example 2.0
the longitudinal (MPLL, VPLL) and transverse (MPTL, Bending moments in single column pier in the
VPTL) directions at the critical locations for the longitudinal (MPLL) and transverse (MPTL) directions
longitudinally-applied seismic loading. are:
MPLL= 0
MPTL= 1602 kft

Shear forces in single column pier the longitudinal

(VPLL) and transverse (VPTL) directions are
VPLL=67.16 k
VPTL=0

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 2.0
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1

Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Sub- Long. Transv. Axial
Isol-
struc shear due shear due forces due
ator
ture to long. to long. to long.
EQ EQ EQ
1 5.63 0 1.29
Abut
2 5.63 0 1.30
ment
3 5.63 0 1.29
1 18.19 0 0.77
Pier 2 18.25 0 1.11
3 18.19 0 0.77

The difference between the longitudinal shear force in

the column (VPLL = 67.16k) and the sum of the
isolator shear forces at the same Pier (54.63 k) is
about 12.5 k. This is due to the inertia force
developed in the hammerhead cap beam which
weighs about 128 k and can generate significant
additional demand on the column (about a 23%
increase in this case).

208

STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake
Repeat the above process, starting at Step B1, for
earthquake loading in the transverse direction of the
bridge. Support flexibility in the transverse direction
is to be included, and a composite response spectrum
is to be applied in the transverse direction. Obtain
isolator displacements in the longitudinal (uT) and
transverse (vT) directions of the bridge, and the biaxial
bending moments and shear forces at critical locations
in the columns due to the transversely-applied seismic
loading.
C1. Analysis for Transverse Earthquake, Example
2.0
Key results from repeating Steps B1 and B2
(Simplified and Mulitmode Spectral Methods) are:
o Teff = 1.52 sec
o Superstructure displacements in the longitudinal
(xT) and transverse (yT) directions are as follows:
xT = 0 and yT = 1.75 in
o Isolator displacements in the longitudinal (uT)
and transverse (vT) directions as follows:
Abutments uT = 0.00 in, vT = 1.75 in
Piers uT = 0.00 in, vT = 0.71 in
o Pier bending moments in the longitudinal (MPLT)
and transverse (MPTT) directions are as follows:
MPLT = 1548.33 kft and MPTT = 0
o Pier shear forces in the longitudinal (VPLT) and
transverse (VPTT) directions are as follows:
VPLT = 0 and VPTT = 60.75 k
o Isolator shear and axial forces are in Table C1-1.

Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Long. Transv. Axial
Sub-
Isol- shear d shear due forces due
struct
ator ue to to transv. to transv.
ure
transv. EQ EQ
EQ
1 0.0 5.82 13.51
Abut
2 0.0 5.83 0
ment
3 0.0 5.82 13.51
1 0.0 15.40 26.40
Pier 2 0.0 15.57 0
3 0.0 15.40 26.40

The difference between the transverse shear force in

the column (VPLL = 60.75k) and the sum of the
isolator shear forces at the same Pier (46.37 k) is
about 14.4 k. This is due to the inertia force
developed in the hammerhead cap beam which
weighs about 128 k and can generate significant
additional demand on the column (about 31% ).

209

STEP D. CALCULATE DESIGN VALUES

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements at Pier 1,

Following the provisions in Art. 2.1 GSID, and Example 2.0
illustrated in Fig. 2.1-1 GSID, calculate the total To illustrate the process, design displacements for the
design displacement, dt, for each isolator by outside isolator on Pier 1 are calculated below.
combining the displacements from the longitudinal (uL
and vL) and transverse (uT and vT) cases as follows: Load Case 1:
u1 = uL + 0.3uT = 1.0(1.15) + 0.3(0) = 1.15 in
u1 = uL + 0.3uT (D-1) v1 = vL + 0.3vT = 1.0(0) + 0.3(0.71) = 0.21 in
v1 = vL + 0.3vT (D-2) R1 = = 1.15 0.21 = 1.17 in
R1 = (D-3)
Load Case 2:
u2 = 0.3uL + uT (D-4) u2 = 0.3uL + uT = 0.3(1.15) + 1.0(0) = 0.35 in
v2 = 0.3vL + vT (D-5) v2 = 0.3vL + vT = 0.3(0) + 1.0(0.71) = 0.71in
R2 = (D-6) R2 = = 0.35 0.71 = 0.79 in

dt = max(R1, R2) (D-7)

Governing Case:
Total design displacement, dt = max(R1, R2)
= 1.17 in

D2. Design Moments and Shears D2. Design Moments and Shears in Pier 1,
Calculate design values for column bending moments Example 2.0
and shear forces for all piers using the same Design moments and shear forces are calculated for
combination rules as for displacements. Pier 1 below, to illustrate the process.

Alternatively this step may be deferred because the Load Case 1:

above results may not be final. Upper and lower VPL1= VPLL + 0.3VPLT = 1.0(67.16) + 0.3(0) = 67.16 k
bound analyses are required after the isolators have VPT1= VPTL + 0.3VPTT = 1.0(0) + 0.3(60.75) = 18.23 k
been designed as described in Art 7. GSID. These R1 = = 67.16 18.23 = 69.59 k
analyses are required to determine the effect of
possible variations in isolator properties due age, Load Case 2:
temperature and scragging in elastomeric systems. VPL2= 0.3VPLL + VPLT = 0.3(67.16) + 1.0(0) = 20.15 k
Accordingly the results for column shear in Steps VPT2= 0.3VPTL + VPTT = 0.3(0) + 1.0(60.75) = 60.75 k
B2.10 and C are likely to increase once these analyses
R2 = = 20.15 60.75 = 64.00 k
are complete.
Governing Case:
Design column shear = max (R1, R2)
= 69.59 k

210

STEP E. DESIGN OF LEAD-RUBBER (ELASTOMERIC) ISOLATORS

A lead-rubber isolator is an elastomeric

bearing with a lead core inserted on its vertical
centreline. When the bearing and lead core are
deformed in shear, the elastic stiffness of the
lead provides the initial stiffness (Ku).With
increasing lateral load the lead yields almost
perfectly plastically, and the post-yield
stiffness Kd is given by the rubber alone. More
details are given in MCEER 2006.

While both circular and rectangular bearings

are commercially available, circular bearings
are more commonly used. Consequently the
procedure given below focuses on circular
bearings. The same steps can be followed for
rectangular bearings, but some modifications will be necessary. When sizing the physical dimensions of the
bearing, plan dimensions (B, dL) should be rounded up to the next 1/4 increment, while the total thickness of
elastomer, Tr, is specified in multiples of the layer thickness. Typical layer thicknesses for bearings with lead
cores are 1/4 and 3/8. High quality natural rubber should be specified for the elastomer. It should have a shear
modulus in the range 60-120 psi and an ultimate elongation-at-break in excess of 5.5. Details can be found in
rubber handbooks or in MCEER 2006.

The following design procedure assumes the isolators are bolted to the masonry and sole plates. Isolators that use
shear-only connections (and not bolts) require additional design checks for stability which are not included below.
See MCEER 2006.

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

E1. Required Properties E1. Required Properties, Example 2.0

Obtain from previous work the properties required of The design of one of the exterior isolators on a pier is
the isolation system to achieve the specified given below to illustrate the design process for lead-
performance criteria (Step A1). rubber isolators.
required characteristic strength, Qd / isolator
required post-elastic stiffness, Kd / isolator From previous work:
total design displacement, dt, for each isolator Qd / isolator = 10.95 k
maximum applied dead and live load (PDL, PLL) Kd / isolator = 6.76 k/in
and seismic load (PSL) which includes seismic Total design displacement, dt = 1.17 in
live load (if any) and overturning forces due to PDL = 187 k
seismic loads, at each isolator, and PLL = 123 k and PSL = 26.4 k (Table C1-1)
maximum wind load, PWL. PWL = 8.21 k < Qd OK

Note that the Kd value per isolator used above is from

the final iteration of the analysis. It is not the same as
the initial value in Step B2.1 (6.64 k/in) , because it
has been adjusted from cycle to cycle, such that the
total Kd summed over all the isolators satisfies the
minimum lateral restoring force requirement for the
bridge, i.e. Kdtotal = 0.05 W/d. See Step B1.1. Since d
varies from cycle to cycle, Kd,j varies from cycle to
cycle.

211

E2. Isolator Sizing

E2.1 Lead Core Diameter E2.1 Lead Core Diameter, Example 2.0
Determine the required diameter of the lead plug, dL,
using:
10.95
(E-1) 3.49
0.9 0.9 0.9
See Step E2.5 for limitations on dL

E2.2 Plan Area and Isolator Diameter E2.2 Plan Area and Isolator Diameter, Example
Although no limits are placed on compressive stress in 2.0
the GSID, (maximum strain criteria are used instead,
see Step E3) it is useful to begin the sizing process by
assuming an allowable stress of, say, 1.6 ksi.

Then the bonded area of the isolator is given by:

(E-2)
1.6

and the corresponding bonded diameter (taking into 187 123

account the hole required to accommodate the lead 193.75
1.6 1.6
core) is given by:
4
(E-3)

4 4 193.75
Round the bonded diameter, B, to nearest quarter inch, 3.49
and recalculate actual bonded area using
= 16.09 in
(E-4)
4 Round B up to 16.25 in and the actual bonded area is:

Note that the overall diameter is equal to the bonded

diameter plus the thickness of the side cover layers 16.25 3.49 197.84
4
(usually 1/2 inch each side). In this case the overall
diameter, Bo is given by:

1.0 (E-5) Bo = 16.25 + 2(0.5) = 17.25 in

E2.3 Elastomer Thickness and Number of Layers E2.3 Elastomer Thickness and Number of Layers,
Since the shear stiffness of the elastomeric bearing is Example 2.0
given by:
(E-6)
where G = shear modulus of the rubber, and
Tr = the total thickness of elastomer, Select G, shear modulus of rubber, = 100 psi (0.1 ksi)
it follows Eq. E-6 may be used to obtain Tr given a
required value for Kd Then
0.1 197.84
(E-7) 2.93
6.76
A typical range for shear modulus, G, is 60-120 psi.
Higher and lower values are available and are used in
special applications.

212

If the layer thickness is tr, the number of layers, n, is

given by:
2.93
(E-8) 11.72
0.25
rounded up to the nearest integer.
Round up to nearest integer, i.e. n = 12
Note that because of rounding the plan dimensions
and the number of layers, the actual stiffness, Kd, will
not be exactly as required. Reanalysis may be
necessary if the differences are large.

E2.4 Overall Height E2.4 Overall Height, Example 2.0

The overall height of the isolator, H, is given by:

1 2 ( E-9) 12 0.25 11 0.125 2 1.5 7.375

where ts = thickness of an internal shim (usually

about 1/8 in), and
tc = combined thickness of end cover plate (0.5
in) and outer plate (1.0 in)

E2.5 Lead Core Size Check E2.5 Lead Core Size Check, Example 2.0
Experience has shown that for optimum performance Since B=16.25 check
of the lead core it must not be too small or too large.
The recommended range for the diameter is as 16.25 16.25
follows: 3 6
(E-10)
3 6 i.e., 5.41 2.71

Since dL = 3.49, lead core size is acceptable.

E3. Strain Limit Check E3. Strain Limit Check, Example 2.0
Art. 14.2 and 14.3 GSID requires that the total applied Since
shear strain from all sources in a single layer of 187.0
elastomer should not exceed 5.5, i.e., 0.945
197.84

, + 0.5 5.5 (E-11) G = 0.1 ksi

where , , , are defined below. and

(a) is the maximum shear strain in the layer due to 197.84
15.50
compression and is given by: 16.25 0.25
(E-12) then
where Dc is shape coefficient for compression in 1.0 0.945
0.61
circular bearings = 1.0, , G is shear 0.1 15.50
modulus, and S is the layer shape factor given by:
(E-13)

(b) , is the shear strain due to earthquake loads and

is given by:
1.17
, (E-14) , 0.39
3.0

213

(c) is the shear strain due to rotation and is given

by:
0.375 16.25 0.01
(E-15) 1.32
0.25 3.0
where Dr is shape coefficient for rotation in circular
bearings = 0.375, and is design rotation due to DL, Substitution in Eq E-11 gives
LL and construction effects. Actual value for may
not be known at this time and a value of 0.01 is , 0.5 0.61 0.39 0.5 1.32
suggested as an interim measure, including 1.66
uncertainties (see LRFD Art. 14.4.2.1). 5.5

E4. Vertical Load Stability Check E4. Vertical Load Stability Check, Example 2.0
Art 12.3 GSID requires the vertical load capacity of
all isolators be at least 3 times the applied vertical
loads (DL and LL) in the laterally undeformed state.

Further, the isolation system shall be stable under

1.2(DL+SL) at a horizontal displacement equal to
either
2 x total design displacement, dt, if in Seismic Zone 1
or 2, or
1.5 x total design displacement, dt, if in Seismic Zone
3 or 4.

E4.1 Vertical Load Stability in Undeformed State E4.1 Vertical Load Stability in Undeformed State,
The critical load capacity of an elastomeric isolator at Example 2.0
zero shear displacement is given by

4
1 1 3 3 0.1 0.3
2 (E-16)
0.3 1 0.67 15.50 48.38
where
16.25
3,422.8
Ts = total shim thickness 64
48.38 3,422.8
55,201 /
1 0.67 3.0
E = elastic modulus of elastomer = 3G
0.1 197.84
64 6.59 /
3.0
It is noted that typical elastomeric isolators have high
shape factors, S, in which case:
4
1 (E-17)

and Eq. E-16 reduces to:

(E-18) 6.59 55,201 1895.5

Check that:

214

1895.5
3 (E-19) 6.11 3
187 123

E4.2 Vertical Load Stability in Deformed State E4.2 Vertical Load Stability in Deformed State,
The critical load capacity of an elastomeric isolator at Example 2.0
shear displacement may be approximated by: Since bridge is in Zone 2, 2 2 1.17 2.34

(E-20) 2.34
2 2.85
where 16.25
Ar = overlap area between top and bottom plates
of isolator at displacement (Fig. 2.2-1
GSID) 2.85 2.85
= 0.817
4
2

Agross = 4 0.817 1895.5 1548.6

It follows that:
(E-21)

Check that: 1548.6

6.17 1

1 (E-22) 1.2 1.2 187 26.4
1.2
E5. Design Review E5. Design Review, Example 2.0
The basic dimensions of the isolator designed above
are as follows:

17.25 in (od) x 7.375in (high) x 3.49 in dia. lead core

and the volume, excluding steel end and cover plates,

= 1,022 in3

Although this design satisfies all the required criteria,

the vertical load stability ratios (Eq. E-19 and E-22)
are much higher than required (6.11 vs 3.0) and total
rubber shear strain (1.66) is much less than the
maximum allowable (5.5), as shown in Step E3. In
other words, the isolator is not working very hard and
a redesign appears to be indicated to obtain a smaller
isolator with more optimal properties (as well as less
cost).

This redesign is outlined below. It begins by

increasing the allowable compressive stress from 1.6
to 3.2 ksi to obtain initial sizes. Remember that no

215

limits are placed on compressive stress in GSID, only

a limit on strain.
E2.1
10.95
3.49
0.9 0.9
E2.2
187 123
96.87
3.2 3.2

4 4 96.87
3.49 11.64

Round B up to 12.5 in and the actual bonded area

becomes:
12.5 3.49 113.16
4

Bo = 12.5 + 2(0.5) = 13.5 in

E2.3
0.1 113.16
1.67
6.76

1.67
6.7
0.25

Round up to nearest integer, i.e. n = 7.

E2.4
7 0.25 6 0.125 2 1.5 5.5
E2.5
Since B=12.5 check

12.5 12.5
3 6

i.e., 4.17 2.08

Since dL = 3.49, size of lead core is acceptable.

E3.
187.0
1.652
113.16

113.16
11.53
12.5 0.25

1.0 1.652
1.43
0.1 11.53

1.17
, 0.67
1.75

216

0.375 12.5 0.01

1.34
0.25 1.75

, 0.5 1.43 0.67 0.5 1.34

2.77 5.5
E4.1
3 3 0.1 0.3

0.3 1 0.67 11.53 26.89

12.5
1,198.4
64
26.89 1198.4
18,411.9 /
1.75

0.1 113.16
6.47 /
1.75

6.47 18411.9 1084.0

1084.0
3.50 3
187 123

E4.2
2.34
2 2.765
12.5

2.76 2.76
0.763

0.763 1084.0 827.15

827.15
3.30 1
1.2 1.2 187 26.4

E5.
The basic dimensions of the redesigned isolator are as
follows:

13.5 in (od) x 5.5 in (high) x 3.49 in dia. lead core

and the volume, excluding steel end and cover plates,

= 358 in3

This design reduces the excessive vertical stability

ratio of the previous design (it is now 3.50 vs 3.0

217

required) and the total layer shear strain is increased

(2.77 vs 5.5 max allowable). Furthermore, the isolator
volume is decreased from 1,022 in3 to 358 in3. This
design is clearly more efficient than the previous one.

E6. Minimum and Maximum Performance Check E6. Minimum and Maximum Performance Check,
Art. 8 GSID requires the performance of any isolation Example 2.0
system be checked using minimum and maximum Minimum Property Modification factors are:
values for the effective stiffness of the system. These min,Kd = 1.0
values are calculated from minimum and maximum min,Qd = 1.0
values of Kd and Qd, which are found using system
property modification factors, as indicated in Table which means there is no need to reanalyze the bridge
E6-1. with a set of minimum values.

Determination of the system property modification Maximum Property Modification factors are:
factors should include consideration of the effects of max,a,Kd = 1.1
temperature, aging, scragging, velocity, travel (wear) max,a,Qd = 1.1
and contamination as shown in Table E6-2. In lieu of
tests, numerical values for these factors can be max,t,Kd = 1.1
obtained from Appendix A, GSID. max,t,Qd = 1.4

max,scrag,Kd = 1.0
Table E6-1. Minimum and maximum values max,scrag,Qd = 1.0
for Kd and Qd. Applying a system adjustment factor of 0.66 for an
other bridge, the maximum property modification
Eq. factors become:
8.1.2-1 Kd,max = Kd max,Kd (E-23)
GSID max,a,Kd = 1.0 + 0.1(0.66) = 1.066
Eq. max,a,Qd = 1.0 + 0.1(0.66) = 1.066
8.1.2-2 Kd,min = Kd min,Kd (E-24)
GSID max,t,Kd = 1.0 + 0.1(0.66) = 1.066
Eq. max,t,Qd = 1.0 + 0.4(0.66) = 1.264
8.1.2-3 Qd,max = Qd max,Qd (E-25)
GSID max,scrag,Kd = 1.0
Eq. max,scrag,Qd = 1.0
8.1.2-4 Qd,min = Qd min,Qd (E-26)
GSID Therefore the maximum overall modification factors

max,Kd = 1.066(1.066)1.0 = 1.14

Table E6-2. Minimum and maximum values for max,Qd = 1.066(1.264)1.0 = 1.35
system property modification factors.
Since the possible variation in upper bound properties
Eq. min,Kd = (min,t,Kd) (min,a,Kd) exceeds 15% a reanalysis of the bridge is required to
8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-27) determine performance with these properties.
GSID (min,scrag,Kd)
Eq. max,Kd = (max,t,Kd) (max,a,Kd) The upper-bound properties are:
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-28) Qd,max = 1.35 (10.95) = 14.78 k
GSID (max,scrag,Kd) and
Eq. min,Qd = (min,t,Qd) (min,a,Qd) Kd,ma x=1.14(6.76) = 7.71 k/in
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-29)
GSID (min,scrag,Qd)
Eq. max,Qd = (max,t,Qd) (max,a,Qd)
8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd) (E-30)
GSID (max,scrag,Qd)

218

Adjustment factors are applied to individual -factors

(except v) to account for the likelihood of occurrence
of all of the maxima (or all of the minima) at the same
time. These factors are applied to all -factors that
deviate from unity but only to the portion of the -
factor that is greater than, or less than, unity. Art.
8.2.2 GSID gives these factors as follows:
1.00 for critical bridges
0.75 for essential bridges
0.66 for all other bridges

As required in Art. 7 GSID and shown in Fig. C7-1

GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Qd,max. As indicated in Fig C7-1 GSID, maximum
displacements will probably be given by the first case
(Kd,min and Qd,min) and maximum forces by the second
case (Kd,max and Qd,max).

E7. Design and Performance Summary, Example

E7. Design and Performance Summary
2.0
E7.1 Isolator dimensions E7.1 Isolator dimensions, Example 2.0
Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E7.1-1.
Overall diameter (includes cover layer)
Overall height Table E7.1-1 Isolator Dimensions
Diameter lead core Isolator Overall size Overall size Diam.
Bonded diameter Location including without lead
Number of rubber layers mounting mounting core
Thickness of rubber layers plates (in) plates (in) (in)
Total rubber thickness Under
Thickness of steel shims edge 17.5 x 17.5 13.5 dia.
Shear modulus of elastomer 3.49
girder x 5.5(H) x 4.0(H)
on Pier 1
Check all dimensions with manufacturer.

Isolator No. of Rubber Total Steel

Location rubber layers rubber shim
layers thick- thick- thick-
ness ness ness
(in) (in) (in)
Under
edge
7 0.25 1.75 0.125
girder
on Pier 1

Shear modulus of elastomer = 100 psi

E7.2 Bridge Performance E7.2 Bridge Performance, Example 2.0

Summarize bridge performance
Maximum superstructure displacement
(longitudinal)
Maximum superstructure displacement
(transverse)
Maximum superstructure displacement
Bridge performance is summarized in Table E7.2-1
where it is seen that the maximum column shear is
71.74k. This less than the column plastic shear (128k)
and therefore the required performance criterion is
satisfied (fully elastic behavior). Furthermore the
maximum longitudinal displacement is 1.69 in which

219

(resultant) is less than the 2.5in available at the abutment

Maximum column shear (resultant) expansion joints and is therefore acceptable.
Maximum column moment (about transverse
axis) Table E7.2-1 Summary of Bridge Performance
Maximum column moment (about longitudinal
axis) Maximum superstructure
1.69 in
Maximum column torque displacement (longitudinal)
Maximum superstructure
1.75 in
Check required performance as determined in Step displacement (transverse)
A3, is satisfied. Maximum superstructure
1.82 in
displacement (resultant)
Maximum column shear
71.74 k
(resultant)
Maximum column moment
1,657 kft
about transverse axis
Maximum column moment
1,676 kft
about longitudinal axis
Maximum column torque 21.44 kft

220

 SECTION 2
Steel Plate Girder Bridge, Long Spans, Single-Column Pier

DESIGN EXAMPLE 2.1: Site Class D

Design Examples in Section 2

Site
ID Description S1 Column height Skew Isolator type
Class
2.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
2.1 Change site class 0.2g D Same 0 Lead rubber bearing
2.2 Change spectral acceleration, S1 0.6g B Same 0 Lead rubber bearing
Spherical friction
2.3 Change isolator to SFB 0.2g B Same 0
bearing
2.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
2.5 Change column height 0.2g B H1=0.5H2 0 Lead rubber bearing
0
2.6 Change angle of skew 0.2g B Same 45 Lead rubber bearing

221

DESIGN PROCEDURE
STEP A: BRIDGE AND SITE DATA
A1. Bridge Properties
Determine properties of the bridge:
number of supports, m
number of girders per support, n
angle of skew
weight of superstructure including railings,
curbs, barriers and other permanent loads,
WSS
weight of piers participating with
superstructure in dynamic response, WPP
weight of superstructure, Wj, at each
support
stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge. The calculation of these
quantities requires careful consideration of
several factors such as the use of cracked
sections when estimating column or wall
flexural stiffness, foundation flexibility,
and effective column height.
column shear strength (minimum value).
This will usually be derived from the
minimum value of the column flexural
yield strength, the column height, and
whether the column is acting in single or
double curvature in the direction under
consideration.
allowable movement at expansion joints
isolator type if known, otherwise to be
selected
A2. Seismic Hazard
Determine seismic hazard at site:
acceleration coefficients
site class and site factors
seismic zone
Plot response spectrum.
Use Art. 3.1 GSID to obtain peak ground and spectral
acceleration coefficients. These coefficients are the
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (e.g. Figures
3.10.2.1-1 to 3.10.2.1-21 LRFD).
Use Art. 3.2 to obtain Site Class and corresponding Site
Factors (Fpga, Fa and Fv). These data are the same as for
conventional bridges and Art 3.2 refers the designer to
the corresponding articles in the LRFD Specifications,
DESIGN EXAMPLE 2.1 (Site Class D)
A1. Bridge Properties, Example 2.1
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 3
Angle of skew = 00
Number of columns per support = 1
Weight of superstructure including permanent
loads,, WSS = 1651.32 k
Weight of superstructure at each support:
o W1 = 168.48 k
o W2 = 657.18 k
o W3 = 657.18 k
o W4 = 168.48 k
Participating weight of piers, WPP = 256.26 k
Effective weight (for calculation of period),
Weff = WSS + WPP = 1907.58 k
Stiffness of each pier in the both directions:
o Ksub,pier1 = 288.87 k/in
o Ksub,pier2 = 288.87 k/in
Minimum column shear strength based on
flexural yield capacity of column = 128 k
Displacement capacity of expansion joints
(longitudinal) = 2.5 in (required to accommodate
thermal expansion and other movements)
Lead-rubber isolators
A2. Seismic Hazard, Example 2.1
Acceleration coefficients for bridge site are given in
design statement as follows:
PGA = 0.40
S1 = 0.20
SS = 0.75
Bridge is on a stiff soil site with a shear wave velocity
in upper 100 ft less than 1,200 ft/sec.
Table 3.10.3.1-1 LRFD gives Site Class as D.
Tables 3.10.3.2-1, -2 and -3 LRFD give following
Site Factors:
Fpga = 1.1
Fa = 1.2
Fv = 2.0
222

i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3,

LRFD.

Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give

modified spectral acceleration coefficients that include
site effects as follows: As = Fpga PGA = 1.1(0.40) = 0.44
As = Fpga PGA SDS = Fa SS = 1.2(0.75) = 0.90
SDS = Fa SS SD1 = Fv S1 = 2.0(0.20) = 0.40
SD1 = Fv S1

Seismic Zone is determined by value of SD1 in Since 0.30 < SD1 < 0.50, bridge is located in Seismic
accordance with provisions in Table 5-1 GSID. Zone 3.

These coefficients are used to plot design response

spectrum as shown in Fig. 4-1 GSID. Design Response Spectrum is as below :

A3. Performance Requirements A3. Performance Requirements, Example 2.1

Determine required performance of isolated bridge As in previous examples, the owner has specified full
during Design Earthquake (1000-yr return period). functionality following the earthquake and therefore
the columns must remain elastic (no yield).
Examples of performance that might be specified by
the Owner include: To remain elastic the maximum lateral load on the
Reduced displacement ductility demand in pier must be less than the load to yield any one
columns, so that bridge is open for emergency column (128 k).
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in The maximum shear in the column must therefore be
columns or yield elsewhere in bridge), so that less than 128 k in order to keep the column elastic
bridge is fully functional and open to all vehicles and meet the required performance criterion.
immediately following earthquake.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.

223

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for Isolator Force, F

non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most have
bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use equivalent Ku
linear springs and viscous damping to represent the
isolators, so that linear methods of analysis may be used to
determine response. Since equivalent properties such as Ku
dy disol
Isolator
Ku Displacement, d
Kisol are dependent on displacement (d), and the
displacements are not known at the beginning of the
analysis, an iterative approach is required. Note that in Kd
Art 7.1, GSID, keff is used for the effective stiffness of an
isolator unit and Keff is used for the effective stiffness of a disol = Isolator displacement
dy = Isolator yield displacement
combined isolator and substructure unit. To minimize Fisol = Isolator shear force
confusion, Kisol is used in this document in place of keff. Fy = Isolator yield force
Kd = Post-elastic stiffness of isolator
There is no change in the use of Keff and Keff,j, but Ksub is Kisol = Effective stiffness of isolator
Ku = Loading and unloading stiffness (elastic stiffness)
used in place of ksub. Qd = Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement ( disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 2.1
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.40 4.0
GSID

(2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not

224

 occur under non-seismic loads (e.g. wind) but

low enough that yield will occur during an
earthquake. Experience has shown that
taking Qd to be 5% of the bridge weight is a
good starting point, i.e.
0.05 (B-2) 0.05 0.05 1651.32 82.56

(3) Post-yield stiffness, Kd

Art 12.2 GSID requires that all isolators
exhibit a minimum lateral restoring force at
the design displacement, which translates to a
minimum post yield stiffness Kd,min given by
Eq. B-3.

Art. 1651.32
0.025 0.05 0.05 20.64 /
12.2 , (B-3) 4.0
GSID
Experience has shown that a good starting
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 2.1
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength,
Qd, and stiffness, Kd, values in proportion to the dead ,

load applied at that support: o Qd, 1 = 8.42 k

o Qd, 2 = 32.86 k
o Qd, 3 = 32.86 k
o Qd, 4 = 8.42 k
, (B-4)
and and

, (B-5) ,
o Kd,1 = 2.11 k/in
o Kd,2 = 8.25 k/in
o Kd,3 = 8.25 k/in
o Kd,4 = 2.11 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 2.1
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of
the isolators at support j (Kisol,j) and the stiffness of
the substructure Ksub,j. See figure below (after Fig.
7.1-1 GSID).

An expression for Keff,j, is given in Eq.7.1-6 GSID, but

a more useful formula as follows (MCEER 2006):

,
, (B-6)
1

225

where
, , , ,
(B-7)
, , , ,
and Ksub,j for the piers are given in Step A1. For the
abutments, take Ksub,j to be a large number, say o 1 = 4.21x10-4
10,000 k/in (unless actual stiffness values are o 2 = 5.85x10-2
available). Note that if the default option is chosen, o 3 = 5.85x10-2
unrealistically high values for Ksub,j will give o 4 = 4.21x10-4
unconservative results for column moments and shear
forces.
F
,
Kd ,
Qd Kisol
1

dy disol
Superstructure
Isolator Effective Stiffness, Kisol o Keff,1 = 4.21 k/in
o Keff,2 = 15.98 k/in
F
o Keff,3 = 15.98 k/in
Isolator(s), Kisol
Ksub o Keff,4 = 4.21 k/in
Substructure, Ksub
dsub

Substructure Stiffness, Ksub

dsub disol F

d
Keff

d = disol + dsub

Combined Effective Stiffness, Keff

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 2.1
Calculate the total effective stiffness, Keff, of the
bridge:
, 40.37 /
Eq.
7.1-6 , ( B-8)
GSID
B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 2.1
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1

(B-9) o disol,1 = 4.00 in

,
1 o disol,2 = 3.78 in
o disol,3 = 3.78 in
o disol,4 = 4.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the stiffness of the isolation system at Example 2.1
support j, Kisol,j, for all supports: ,
, ,
,
o Kisol,1 = 4.21 k/in
,
(B-10) o Kisol,2 = 16.91 k/in
, ,
, o Kisol,3 = 16.91 k/in
o Kisol,4 = 4.21 k/in

226

B1.7 Substructure Displacement at Each Support B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j, Example 2.1
for all supports:

, , (B-11) , ,

o dsub,1 = 0.002 in
o d sub,2 = 0.221 in
o d sub,3 = 0.221 in
o d sub,4 = 0.002 in

B1.8 Lateral Load in Each Substructure B1.8 Lateral Load in Each Substructure, Example
Calculate the shear at support j, Fsub,j, for all 2.1
supports:
, , , (B-12) , , ,

where values for Ksub,j are given in Step A1. o F sub,1 = 16.84 k
o F sub,2 = 63.91 k
o F sub,3 = 63.91 k
o F sub,4 = 16.84 k

B1.9 Column Shear Force at Each Support B1.9 Column Shear Force at Each Support,
Calculate the shear in column k at support j, Fcol,j,k, Example 2.1
assuming equal distribution of shear for all columns at
support j:
,
, ,
#
,
, ,
# (B-13) o F col,2,1 = 63.91 k
o F col,3,1 = 63.91 k

Use these approximate column shear forces as a check These column shears are less than the plastic shear
on the validity of the chosen strength and stiffness capacity of each column (128k) as required in Step
characteristics. A3 and the chosen strength and stiffness values in
Step B1.1 are therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 2.1
damping ratio, , of the bridge:

Eq. 1907.58
7.1-5 2 (B-14) 2 2
GSID 386.4 40.37

and = 2.20 sec

Eq. 2 , , , and taking dy,j = 0:

7.1-10 (B-15)
, , ,
GSID 2 , , 0
0.31
, , ,
where dy,j is the yield displacement of the isolator. For
friction-based isolators, dy,j = 0. For other types of

227

isolators dy,j is usually small compared to disol,j and has
negligible effect on , Hence it is suggested that for
the Simplified Method, set dy,j = 0 for all isolator
types. See Step B2.2 where the value of dy,j is revisited
for the Multimode Spectral Analysis Method.
B1.11 Damping Factor
Calculate the damping factor, BL, and the
displacement, d, of the bridge:
Eq. .
, 0.3
.
7.1-3 (B-16)
1.7, 0.3
GSID
Eq. 9.79 (B-17)
7.1-4
GSID
B1.12 Convergence Check
Compare the new displacement with the initial value
assumed in Step B1.1. If there is close agreement, go
to the next step; otherwise repeat the process from
Step B1.3 with the new value for displacement as the
assumed displacement.
This iterative process is amenable to solution using a
spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
B1.11 Damping Factor, Example 2.1
Since 0.31 0.3
1.70
and
9.79 9.79 0.4 2.20
5.06
1.70
B1.12 Convergence Check, Example 2.1
Since the calculated value for displacement, d (= 5.06)
is not close to that assumed at the beginning of the
cycle (Step B1.1, d = 4.0), use a value of say 5.0 in as
the new assumed displacement and repeat from Step
B1.3.
After three iterations, convergence is reached at a
superstructure displacement of 6.38 in, with an
effective period of 2.76 seconds, and a damping factor
of 1.7 (31% damping ratio). The displacement in the
isolators at Pier 1 is 6.16 in and the effective stiffness
of the same isolators is 10.49 k/in.
See spreadsheet in Table B1.12-1 for results of final
iteration.
Since the column shear force must equal the isolator
shear force for equilibrium, the column shear = 10.49
(6.16) = 64.62 k which is less than the maximum
allowable (128 k) if elastic behavior is to be achieved
(as required in Step A3).
But the superstructure displacement = 6.38 in, which
exceeds the available clearance of 2.5 in.
There are three choices here:
1. Increase the clearance at the abutment to say
7 in to avoid impact. (Note that the minimum
required is
8 8 0.40 2.76
5.20 .
1.7
2. Allow impact to happen which will damage
abutment back wall and require repair. This
option would violate the elastic performance
228

 requirement in Step A3.

3. Redesign the isolators. Since the column

shear is only one-half of the yield capacity
(64.62 vs 128 k) there is room to increase the
characteristic strength Qd and post yield
stiffness Kd to increase this shear and reduce
the displacements.

One of the many possible solutions here is to

increase Qd to 0.07W and Kd to 0.07 W/d. In
this case, it will be found that the
superstructure displacement reduces to 4.50
in, the effective period is 1.97 seconds, and
the damping factor remains at 1.7 (31%
damping ratio).

See spreadsheet in Table B1.12-2 for results

of the final iteration for this solution.

The displacement in the isolators at Pier 1 is

4.19 in and the effective stiffness of the same
isolators is 21.2 k/in. The column shear force
is therefore = 21.2 (4.19) = 88.83 k which is
much closer to the capacity of 128 k, but
remains elastic as required. Although this is a
much more efficient design, the
superstructure displacement (4.5 in) still
exceeds the capacity (2.5 in) and the
recommended option is to increase the
clearance at the abutments to, say, 5.0 in (the
minimum required using 8 SD1Teff/BL is 3.71
in). This option is revisited in Step E7.2.

Option 3 is recommended and the following

properties are assumed for the isolation system going
forward to the next step (Step B2):

0.07 0.07 1651.32 115.59

and
0.07 0.07 1651.32
25.69 /
4.5

229

 Table B1.12-1 Simplified Method Solution for Design Example 2.1 - First Solution, Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
1651.32 256.26 1907.58 0.4 3

StepB1.1 d 6.38 Assumeddisplacement

Qd 82.57 Characteristicstrength
Kd 12.94 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 168.48 8.424 1.320 10,000.00 0.000264 2.640 6.378 2.641 0.002 16.846 53.731 107.476
Pier1 657.18 32.859 5.150 288.87 0.036306 10.120 6.156 10.488 0.224 64.567 202.296 411.936
Pier2 657.18 32.859 5.150 288.87 0.036306 10.120 6.156 10.488 0.224 64.567 202.296 411.936
Abut2 168.48 8.424 1.320 10,000.00 0.000264 2.640 6.378 2.641 0.002 16.846 53.731 107.476
Total 1651.32 82.566 12.941 K eff,j 25.521 162.825 512.054 1,038.825
Step B1.4

StepB1.10 T eff 2.76 Effectiveperiod

0.31 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.74

B L 1.70 DampingFactor
d 6.37 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1
Pier1
Pier2
Abut2

230

 Table B1.12-2 Simplified Method Solution for Design Example 2.1 - Second Solution, Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
1651.32 256.26 1907.58 0.4 3

StepB1.1 d 4.50 Assumeddisplacement

Qd 115.59 Characteristicstrength
Kd 25.69 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 168.48 11.794 2.621 10,000.00 0.000524 5.240 4.498 5.243 0.002 23.581 53.043 106.115
Pier1 657.18 46.003 10.223 288.87 0.073375 19.747 4.192 21.196 0.308 88.861 192.861 399.872
Pier2 657.18 46.003 10.223 288.87 0.073375 19.747 4.192 21.196 0.308 88.861 192.861 399.872
Abut2 168.48 11.794 2.621 10,000.00 0.000524 5.240 4.498 5.243 0.002 23.581 53.043 106.115
Total 1651.32 115.592 25.687 K eff,j 49.974 224.883 491.808 1,011.974
Step B1.4

StepB1.10 T eff 1.97 Effectiveperiod

0.31 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.73

B L 1.70 DampingFactor
d 4.55 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 3.931 0.874 1.748 4.37 1.773
Pier1 15.334 3.408 7.065 3.78 7.464
Pier2 15.334 3.408 7.065 3.78 7.464
Abut2 3.931 0.874 1.748 4.37 1.773

231

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and then
applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 2.1

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 (see
elastic stiffness, Kd,i, of each isolator i as follows: Table B1.12-2) by the number of isolators at each
, support (n = 3), the following values for Qd /isolator
, (B-19)
and Kd /isolator are obtained:
And o Qd, 1 = 11.79/3 = 3.93 k
, o Qd, 2 = 46.00/3= 15.33 k
, ( B-20)
o Qd, 3 = 46.00/3 = 15.33 k
where values for Qd,j and Kd,j are obtained from the o Qd, 4 = 11.79/3 = 3.93 k
final cycle of iteration in the Simplified Method (Step
B1. 12 and
o Kd,1 = 2.62/3 = 0.87 k/in
o Kd,2 = 10.22/3 = 3.41 k/in
o Kd,3 = 10.22/3 = 3.41 k/in
o Kd,4 = 2.62/3 = 0.87 k/in

Note that the Kd values per support used above are

from the final iteration in Table B1.12-2 and are not
the same as the initial values in Step B1.2.This is
principally because Qd and Kd were changed in Step
B1.12 to reduce the superstructure displacements
from 6.38 in to 4.50in. Even if these parameters had
not been changed, the above Kd values would not be
the same as the values in Step B1.2, because they are
adjusted from cycle to cycle in the iteration process,
such that the total Kd summed over all the isolators
satisfies the minimum lateral restoring force
requirement for the bridge, i.e. Kdtotal = 0.05 W/d. See
Step B1.1. Since d varies from cycle to cycle, Kd,j
varies from cycle to cycle.

B2.2 Initial Stiffness and Yield Displacement
Calculate the initial stiffness, Ku,i, and the yield
displacement, dy,i, for each isolator i as follows:
(1) For friction-based isolators Ku,i = and dy,i = 0.
(2) For other types of isolators, and in the absence of
isolator-specific information, take
B2.2 Initial Stiffness and Yield Displacement,
Example 2.1
Since the isolator type has been specified in Step A1
to be an elastomeric bearing (i.e. not a friction-based
bearing), calculate Ku,i and dy,i for an isolator on Pier
1as follows:

, 10 , ( B-21) , 10 , 10 3.41 34.1 /

and then

232


( B-22)
B2.3 Isolator Effective Stiffness, Kisol,i
Calculate the isolator stiffness, Kisol,i, of each isolator
i:
(B -23)
B2.4 Three Dimensional Bridge Model
Using computer-based structural analysis software,
create a 3-dimensional model of the bridge with the
isolators represented by spring elements. The stiffness
of each isolator element in the horizontal axes (Kx and
Ky in global coordinates, K2 and K3 in typical local
coordinates) is the Kisol value calculated in the
previous step. For bridges with regular geometry and
minimal skew or curvature, the superstructure may be
represented by a single stick provided the load path
to each individual isolator at each support is explicitly
modeled, usually by a rigid cap beam and a set of rigid
links. If the geometry is irregular, or if the bridge is
skewed or curved, a finite element model is
recommended to accurately capture the load carried
by each individual isolator.
B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in Step
A2 is for a 5% damped system. It is modified in this
step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the effective
period of the bridge, Teff, by the damping factor, BL.
and
As expected, the yield displacement is small
compared to the expected isolator displacement (~ 5
in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.
B2.3 Isolator Effective Stiffness, Kisol,i, Example 2.1
Dividing the results for Kisol (Step B1.12) among the 3
isolators at each support, the following values for Kisol
/isolator are obtained:
o Kisol,1 = 5.24/3 = 1.75 k/in
o Kisol,2 = 21.20/3 = 7.07 k/in
o Kisol,3 = 21.20/3 = 7.07 k/in
o Kisol,4 = 5.24/3 = 1.75 k/in
B2.4 Three Dimensional Bridge Model, Example
2.1
Although the bridge in this Design Example is regular
and is without skew or curvature, a 3-dimensional
finite element model was developed for this Step, as
shown below.
.
B.2.5 Composite Design Response Spectrum,
Example 2.1
From the final results of Simplified Method (Step
B1.12), BL = 1.70 and Teff = 1.97 sec. Hence the
transition in the composite spectrum from 5% to 30%
damping occurs at 0.8 Teff = 0.8 (1.97) = 1.58 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.58 sec by 1.70.
233

B2.6 Multimodal Analysis of Finite Element Model
Input the composite response spectrum as a user-
specified spectrum in the software, and define a load
case in which the spectrum is applied in the
longitudinal direction. Analyze the bridge for this
load case.
B2.6 Multimodal Analysis of Finite Element
Model, Example 2.1
Results of modal analysis of the example bridge are
summarized in Table B2.6-1 Here the modal periods
and mass participation factors of the first 12 modes
are given. The first three modes are the principal
transverse, longitudinal, and torsion modes with
periods of 2.0, 1.91 and 1.87 sec respectively. The
period of the longitudinal mode (1.91 sec) is the
almost the same as calculated in the Simplified
Method (1.97 sec). The mass participation factors
indicate there is no coupling between these three
modes (probably due to the symmetric nature of the
bridge) and the high values for the first and second
modes (91% for each mode) indicate the bridge is
responding essentially in a single mode of vibration in
each direction. Similar results to those obtained by the
Simplified Method are therefore to be expected.

Table B2.6-1 Modal Properties of Bridge

Example 2.1 First Iteration

Mode Period UX UY UZ RX RY RZ
No Sec Unitless Unitless Unitless Unitless Unitless Unitless
1 1.998 0.000 0.909 0.000 0.911 0.000 0.689
2 1.905 0.910 0.000 0.000 0.000 0.020 0.000
3 1.874 0.000 0.000 0.000 0.000 0.000 0.229
4 0.488 0.000 0.001 0.000 0.041 0.000 0.001
5 0.372 0.000 0.000 0.074 0.000 0.055 0.000
6 0.354 0.000 0.000 0.000 0.000 0.000 0.002
7 0.346 0.000 0.003 0.000 0.017 0.000 0.002
8 0.283 0.000 0.006 0.000 0.018 0.000 0.004
9 0.251 0.000 0.000 0.000 0.000 0.000 0.000
10 0.251 0.089 0.000 0.000 0.000 0.001 0.000
11 0.207 0.000 0.000 0.000 0.000 0.128 0.000
12 0.188 0.000 0.000 0.000 0.000 0.000 0.001

Computed values for the isolator displacements due to

a longitudinal earthquake are as follows (numbers in
parentheses are those used to calculate the initial
properties to start iteration from the Simplified
Method):

o disol,1 = 4.37 (4.50) in

o disol,2 = 3.78 (4.19) in
o disol,3 = 3.78 (4.19) in
o disol,4 = 4.37 (4.50) in

B2.7 Convergence Check B2.7 Convergence Check, Example 2.1

Compare the resulting displacements at the
superstructure level (d) to the assumed displacements.
These displacements can be obtained by examining
the joints at the top of the isolator spring elements. If
in close agreement, go to Step B2.9. Otherwise go to
Step B2.8.
The results for the isolator displacements are close but
not close enough (10% difference at piers).
Go to Step B2.8 and update properties for a second
cycle of iteration.

234

B2.8 Update Kisol,i , Keff,j, and BL B2.8 Update Kisol,i , Keff,j, and BL, Example 2.1
Use the calculated displacements in each isolator Updated values for Kisol,i are given below (previous
element to obtain new values of Kisol,i for each isolator values are in parentheses):
as follows:
, o Kisol,1 = 1.77 (1.75) k/in
, , (B-24) o Kisol,2 = 7.46 (7.07) k/in
,
o Kisol,3 = 7.46 (7.07) k/in
Recalculate Keff,j : o Kisol,4 = 1.77 (1.75) k/in

Eq. , ,
7.1-6 ,
(B-25) Since the isolator displacements are relatively close to
, ,
GSID previous results no significant change in the damping
ratio is expected. Hence Keff,j and are not
Recalculate system damping ratio, : recalculated and BL is taken at 1.70.

Eq. 2 , , ,
7.1-10 , , ,
(B-26)
GSID

Recalculate system damping factor, BL:

Eq. .
0.3
.
7.1-3 ( B-27)
1.7 0.3
GSID
Since the change in effective period is very small
Obtain the effective period of the bridge from the (1.91 to 1.97 sec) and no change has been made to BL,
multi-modal analysis and with the revised damping there is no need to construct a new composite
factor (Eq. B-27), construct a new composite response response spectrum in this case. Go back to Step B2.6
spectrum. Go to Step B2.6. (see immediately below).

B2.6 Multimodal Analysis Second Iteration,

Example 2.1
Reanalysis gives the following values for the isolator
displacements (numbers in parentheses are those
from the previous cycle):

o disol,1 = 4.29 (4.37) in

o disol,2 = 3.68 (3.78) in
o disol,3 = 3.68 (3.78) in
o disol,4 = 4.29 (4.29) in

Go to Step B2.7

B2.7 Convergence Check B2.7 Convergence Check, Example 2.1

Compare results and determine if convergence has Satisfactory agreement has been reached on this
been reached. If so go to Step B2.9. Otherwise Go to second cycle (better than 2% at the abutments and 3%
Step B2.8. at the piers). Go to Step B2.9

B2.9 Superstructure and Isolator Displacements B2.9 Superstructure and Isolator Displacements,
Once convergence has been reached, obtain Example 2.1
o superstructure displacements in the longitudinal From the above analysis:
(xL) and transverse (yL) directions of the bridge, o superstructure displacements in the
and longitudinal (xL) and transverse (yL) directions
o isolator displacements in the longitudinal (uL) are:
and transverse (vL) directions of the bridge, for xL= 4.32 in
each isolator, for this load case (i.e. yL= 0.0 in

235

 longitudinal loading). These displacements o isolator displacements in the longitudinal (uL)

may be found by subtracting the nodal and transverse (vL) directions are:
displacements at each end of each isolator o Abutments: uL = 4.29 in, vL = 0.00 in
spring element. o Piers: uL = 3.68 in, vL = 0.00 in

All isolators at same support have the same

displacements.

B2.10 Pier Bending Moments and Shear Forces B2.10 Pier Bending Moments and Shear Forces,
Obtain the pier bending moments and shear forces in Example 2.1
the longitudinal (MPLL, VPLL) and transverse (MPTL, Bending moments in single column pier in the
VPTL) directions at the critical locations for the longitudinal (MPLL) and transverse (MPTL) directions
longitudinally-applied seismic loading. are:
MPLL= 0
MPTL= 2,661 kft

Shear forces in single column pier the longitudinal

(VPLL) and transverse (VPTL) directions are:

VPLL= 111.54 k
VPTL= 0

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 2.1
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1

Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces due
ator
ure to long. to long. to long.
EQ EQ EQ
1 7.60 0 1.56
Abut
2 13.98 0 1.54
ment
3 13.98 0 1.56
1 27.47 0 0.64
Pier 2 27.51 0 0.63
3 27.47 0 0.64

The difference between the longitudinal shear force in

the column (VPLL = 111.54 k) and the sum of the
isolator shear forces at the same Pier (82.45 k) is
about 29.1 k. This is due to the inertia force
developed in the hammerhead cap beam which
weighs about 128 k and can generate significant
additional demand on the column (about a 34%
increase in this case).

236

STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake
Repeat the above process, starting at Step B1, for
earthquake loading in the transverse direction of the
bridge. Support flexibility in the transverse direction
is to be included, and a composite response spectrum
is to be applied in the transverse direction. Obtain
isolator displacements in the longitudinal (uT) and
transverse (vT) directions of the bridge, and the biaxial
bending moments and shear forces at critical locations
in the columns due to the transversely-applied seismic
loading.
C1. Analysis for Transverse Earthquake, Example
2.1
Key results from repeating Steps B1 and B2
(Simplified and Mulitmode Spectral Methods) are:
o Teff = 1.92 sec
o Superstructure displacements in the longitudinal
(xT) and transverse (yT) directions are as follows:
xT = 0 and yT = 4.45 in
o Isolator displacements in the longitudinal (uT)
and transverse (vT) directions are as follows:
Abutments uT = 0.00 in, vT = 4.48in
Piers uT = 0.00 in, vT = 3.20 in
o Pier bending moments in the longitudinal (MPLT)
and transverse (MPTT) directions are as follows:
MPLT = 2,506 kft and MPTT = 0
o Pier shear forces in the longitudinal (VPLT) and
transverse (VPTT) directions as follows:
VPLT = 0 and VPTT = 92.06 k
o Isolator shear and axial forces are summarized in
Table C1-1.

Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Sub- Long. Transv. Axial
Isol-
struct shear due shear due forces due
ator
ure to transv. to transv. to transv.
EQ EQ EQ
1 0.0 7.79 18.04
Abut
2 0.0 7.80 0
ment
3 0.0 7.79 18.04
1 0.0 25.87 33.64
Pier 2 0.0 25.98 0
3 0.0 25.87 33.64

The difference between the longitudinal shear force in

the column (VPTT = 92.06 k) and the sum of the
isolator shear forces at the same Pier (77.72 k) is
about 14.3 k. This is due to the inertia force
developed in the hammerhead cap beam which
weighs about 128 k and can generate significant
additional demand on the column (about 18%).

237

STEP D. CALCULATE DESIGN VALUES

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements at Pier 1,

Following the provisions in Art. 2.1 GSID, and Example 2.1
illustrated in Fig. 2.1-1 GSID, calculate the total To illustrate the process, design displacements for the
design displacement, dt, for each isolator, by outside isolator on Pier 1 are calculated below.
combining the displacements from the longitudinal (uL
and vL) and transverse (uT and vT) cases as follows: Load Case 1:
u1 = uL + 0.3uT (D-1) u1 = uL + 0.3uT = 1.0(3.67) + 0.3(0) = 3.67 in
v1 = vL + 0.3vT (D-2) v1 = vL + 0.3vT = 1.0(0) + 0.3(3.20) = 0.96 in
R1 = (D-3) R1 = = 3.67 0.96 = 3.79 in

u2 = 0.3uL + uT (D-4) Load Case 2:

v2 = 0.3vL + vT (D-5) u2 = 0.3uL + uT = 0.3(3.67) + 1.0(0) = 1.10 in
R2 = (D-6) v2 = 0.3vL + vT = 0.3(0) + 1.0(3.20) = 3.20 in
R2 = = 1.10 3.20 = 3.38 in
dt = max(R1, R2) (D-7)
Governing Case:
Total design displacement, dt = max(R1, R2)
= 3.79in

D2. Design Moments and Shears D2. Design Moments and Shears at Pier 1,
Calculate design values for column bending moments Example 2.1
and shear forces for all piers using the same Design moments and shear forces are calculated for
combination rules as for displacements. Pier 1 below, to illustrate the process.

Alternatively this step may be deferred because the Load Case 1:

above results may not be final. Upper and lower VPL1= VPLL + 0.3VPLT = 1.0(111.54) + 0.3(0) = 111.5 k
bound analyses are required after the isolators have VPT1= VPTL + 0.3VPTT = 1.0(0) + 0.3(92.06) = 27.62 k
been designed as described in Art 7. GSID. These R1 = = 111.5 27.62 = 114.9 k
analyses are required to determine the effect of
possible variations in isolator properties due age, Load Case 2:
temperature and scragging in elastomeric systems. VPL2= 0.3VPLL + VPLT = 0.3(111.54) + 1.0(0) = 33.46 k
Accordingly the results for column shear in Steps VPT2= 0.3VPTL + VPTT = 0.3(0) + 1.0(92.06) = 92.06k
B2.10 and C are likely to increase once these analyses
R2 = = 33.46 92.06 = 97.95k
are complete.
Governing Case:
Design column shear = max (R1, R2)
= 114.9 k

238

STEP E. DESIGN OF LEAD-RUBBER (ELASTOMERIC) ISOLATORS

A lead-rubber isolator is an elastomeric

bearing with a lead core inserted on its vertical
centreline. When the bearing and lead core are
deformed in shear, the elastic stiffness of the
lead provides the initial stiffness (Ku).With
increasing lateral load the lead yields almost
perfectly plastically, and the post-yield
stiffness Kd is given by the rubber alone. More
details are given in MCEER 2006.

While both circular and rectangular bearings

are commercially available, circular bearings
are more commonly used. Consequently the
procedure given below focuses on circular
bearings. The same steps can be followed for
rectangular bearings, but some modifications will be necessary. When sizing the physical dimensions of the
bearing, plan dimensions (B, dL) should be rounded up to the next -inch increment, while the total thickness of
elastomer, Tr, is specified in multiples of the layer thickness. Typical layer thicknesses for bearings with lead
cores are inch and 3/8 in. High quality natural rubber should be specified for the elastomer. It should have a shear
modulus in the range 60-120 psi and an ultimate elongation-at-break in excess of 5.5. Details can be found in
rubber handbooks or in MCEER 2006.

The following design procedure assumes the isolators are bolted to the masonry and sole plates. Isolators that use
shear-only connections (and not bolts) require additional design checks for stability which are not included below.
See MCEER 2006.

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

E1. Required Properties E1. Required Properties, Example 2.1

Obtain from previous work the properties required of The design of one of the exterior isolators on a pier is
the isolation system to achieve the specified given below to illustrate the design process for a lead-
performance criteria (Step A1). rubber isolator.
required characteristic strength, Qd / isolator
required post-elastic stiffness, Kd / isolator From previous work
total design displacement, dt, for each isolator Qd / isolator = 15.33 k
maximum applied dead and live load (PDL, PLL) Kd / isolator = 3.41 k/in
and seismic load (PSL) which includes seismic Total design displacement, dt = 3.79 in
live load (if any) and overturning forces due to PDL = 187 k
seismic loads, at each isolator, and PLL = 123 k and PSL = 33.64 k (Table C1-1)
maximum wind load, PWL PWL = 8.21k < Qd OK

E2. Isolator Sizing

E2.1 Lead Core Diameter E2.1 Lead Core Diameter, Example 2.1
Determine the required diameter of the lead plug, dL,
using:
15.33
(E-1) 4.13
0.9 0.9 0.9

See Step E2.5 for limitations on dL

239

E2.2 Plan Area and Isolator Diameter E2.2 Plan Area and Isolator Diameter, Example
Although no limits are placed on compressive stress in 2.1
the GSID, (maximum strain criteria are used instead,
see Step E3) it is useful to begin the sizing process by
assuming an allowable stress of, say, 1.6 ksi.

Then the bonded area of the isolator is given by:

(E-2)
1.6

and the corresponding bonded diameter (taking into 187 123

account the hole required to accommodate the lead 193.75
1.6 1.6
core) is given by:
4
(E-3)

4 4 193.75
Round the bonded diameter, B, to nearest quarter inch, 4.13
and recalculate actual bonded area using
= 16.24 in
(E-4)
4 Round B up to 16.25 in and the actual bonded area is:

Note that the overall diameter is equal to the bonded

diameter plus the thickness of the side cover layers 16.25 4.13 194.02
4
(usually 1/2 inch each side). In this case the overall
diameter, Bo is given by:

1.0 (E-5)

Bo = 16.25 + 2(0.5) = 17.25 in

E2.3 Elastomer Thickness and Number of Layers E2.3 Elastomer Thickness and Number of Layers,
Since the shear stiffness of an elastomeric bearing is Example 2.1
given by:
(E-6)
where G = shear modulus of the rubber, and
Tr = the total thickness of elastomer, Select G, shear modulus of rubber = 100 psi (0.1ksi),
it follows Eq. E-6 may be used to obtain Tr given a then
required value for Kd
0.1 194.02
(E-7) 5.69
3.41

A typical range for shear modulus, G, is 60-120 psi.

Higher and lower values are available and are used in
special applications.

If the layer thickness is tr, the number of layers, n, is

given by:
(E-8) 5.69
22.8
0.25

240

rounded up to the nearest integer.

Round up to nearest integer, i.e. n = 23
Note that because of rounding the plan dimensions
and the number of layers, the actual stiffness, Kd, will
not be exactly as required. Reanalysis may be
necessary if the differences are large.

E2.4 Overall Height E2.4 Overall Height, Example 2.1

The overall height of the isolator, H, is given by:

1 2 ( E-9) 23 0.25 22 0.125 2 1.5 11.50

where ts = thickness of an internal shim (usually

about 1/8 in), and
tc = combined thickness of end cover plate (0.5
in) and outer plate (1.0 in)

E2.5 Lead Core Size Check E2.5 Lead Core Size Check, Example 2.1
Experience has shown that for optimum performance Since B=16.25 check
of the lead core it must not be too small or too large.
The recommended range for the diameter is as
follows:
16.25 16.25
(E-10)
3 6 3 6

i.e., 5.42 2.71

Since dL = 4.13, lead core size is acceptable.

E3. Strain Limit Check E3. Strain Limit Check, Example 2.1
Art. 14.2 and 14.3 GSID requires that the total applied
shear strain from all sources in a single layer of
elastomer should not exceed 5.5, i.e.,
Since
, + 0.5 5.5 (E-11) 187.0
0.964
194.02
where , , , are defined below.
(a) is the maximum shear strain in the layer due to
compression and is given by: G = 0.1 ksi

(E-12)
where Dc is shape coefficient for compression in and
194.02
circular bearings = 1.0, , G is shear 15.20
16.25 0.25
modulus, and S is the layer shape factor given by:
(E-13)
then
1.0 0.964
(b) , is the shear strain due to earthquake loads and 0.63
0.1 15.2
is given by:

, (E-14)

(c) is the shear strain due to rotation and is given

241

by:
3.79
(E-15) , 0.66
5.75
where Dr is shape coefficient for rotation in circular
bearings = 0.375, and is design rotation due to DL,
LL and construction effects . Actual value for may
not be known at this time and a value of 0.01 is 0.375 16.25 0.01
0.69
suggested as an interim measure, including 0.25 5.75
uncertainties (see LRFD Art. 14.4.2.1).
Substitution in Eq E-11 gives

, 0.5 0.63 0.66 0.5 0.69

1.64
5.5

E4. Vertical Load Stability Check E4. Vertical Load Stability Check, Example 2.1
Art 12.3 GSID requires the vertical load capacity of
all isolators be at least 3 times the applied vertical
loads (DL and LL) in the laterally undeformed state.

Further, the isolation system shall be stable under

1.2(DL+SL) at a horizontal displacement equal to
either
2 x total design displacement, dt, if in Seismic Zone 1
or 2, or
1.5 x total design displacement, dt, if in Seismic Zone
3 or 4.

E4.1 Vertical Load Stability in Undeformed State E4.1 Vertical Load Stability in Undeformed State,
The critical load capacity of an elastomeric isolator at Example 2.1
zero shear displacement is given by

4
1 1 3 3 0.1 0.3
2 (E-16)
0.3 1 0.67 15.20 46.54
where
16.25
3,422.8
Ts = total shim thickness 64

46.54 3,422.8
27,705 /
1 0.67 5.75
E = elastic modulus of elastomer = 3G
0.1 194.02
64 3.37 /
5.75
It is noted that typical elastomeric isolators have high
shape factors, S, in which case:

4
1 (E-17)

242

and Eq. E-16 reduces to:

(E-18) 3.37 27,705 960.54

Check that:
960.54
3 (E-19) 3.10 3
187 123

E4.2 Vertical Load Stability in Deformed State E4.2 Vertical Load Stability in Deformed State,
The critical load capacity of an elastomeric isolator at Example 2.1
shear displacement may be approximated by: Since bridge is in Zone 3,

1.5 1.5 3.79 5.69

(E-20)
where
5.69
Ar = overlap area between top and bottom plates 2 2.43
of isolator at displacement (Fig. 2.2-1 16.25
GSID)
= 4
2 2.43 2.43
0.564

Agross = 4

It follows that: 0.564 960.54 541.57

(E-21)

Check that: 541.57

2.10 1

1 (E-22) 1.2 1.2 187 33.64
1.2

E5. Design Review E5. Design Review, Example 2.1

The basic dimensions of the isolator designed above
are as follows:

17.25 in (od) x 11.50 in (high) x 4.13 in dia. lead core

and the volume, excluding the steel end and cover

plates = 2,337 in3

Although this design satisfies all the required criteria,

the total rubber shear strain (1.64) is much less than
the maximum allowable (5.5), as shown in Step E3. In
other words, the isolator is not working very hard and
a redesign appears to be worth exploring to see if a
more optimal design can be found. Since the plan
dimension appears to be about right to satisfy both
vertical stability requirements (Step E4.1 and E4.2)
the best way to optimize the design is to reduce its

243

 height. But reducing the height will increase the

stiffness, Kd, unless the shear modulus of the
elastomer is likewise reduced. In the redesign below,
the plan dimensions remain the same but the shear
modulus is reduced from 100 to 60 psi.

E2.1
15.33
4.13
0.9 0.9
E2.2

187 123
193.75
1.6 1.6

4 4 193.75
4.13

= 16.24 in

Round B up to 16.25 in and the actual bonded area is:

16.25 4.13 194.02

4

Bo = 16.25 + 2(0.5) = 17.25 in

E2.3
0.06 194.02
3.41
3.41

3.41
13.7
0.25
Round n up to 14

E2.4
14 0.25 13 0.125 2 1.5 8.125

E2.5
Since B = 16.25 check

16.25 16.25
3 6

5.42 2.71

Since dL = 4.13 in, size of lead core is acceptable.

E3.
187.0
0.964
194.02

194.02
15.20
16.25 0.25

244

 1.0 0.964
1.06
0.06 15.20

3.79
, 1.08
3.5

0.375 16.25 0.01

1.13
0.25 3.5

, 0.5 1.06 1.08 0.5 1.13

2.71 5.5

E4.1
3 3 0.06 0.18

0.18 1 0.67 15.20 27.93

16.25
3,422.8
64
27.93 3,422.8
27,309 /
3.5

0.06 194.02
3.33 /
3.5

3.33 27,309 946.82

946.82
3.05 3
187 123

E4.2
5.68
2 2.43
16.25

2.43 2.43
0.564

0.564 946.82 533.84

533.84
2.07 1
1.2 1.2 187 33.64

E5.
The basic dimensions of the redesigned isolator are as

245

 follows:

17.25 in (od) x 8.125 in (high) x 4.13 in dia. lead core

and the volume, excluding steel end and cover plates,

= 1,548 in3

This is a more optimal design. It is smaller than the

previous design (1,548 in3 vs 2,337 in3) while still
satisfying all the design criteria. Being smaller it
works harder to satisfy these requirements, as can be
seen by the increase in the total shear strain from 1.64
to 2.71 (but still less than the maximum allowable of
5.5).

E6. Minimum and Maximum Performance Check
Art. 8 GSID requires the performance of any isolation
system be checked using minimum and maximum
values for the effective stiffness of the system. These
values are calculated from minimum and maximum
values of Kd and Qd, which are found using system
property modification factors, as indicated in Table
E6-1.
E6. Minimum and Maximum Performance Check,
Example 2.1
Minimum Property Modification factors are
min,Kd = 1.0
min,Qd = 1.0
which means there is no need to reanalyze the bridge
with a set of minimum values.

Table E6-1. Minimum and maximum values Maximum Property Modification factors are
for Kd and Qd.
max,a,Kd = 1.1
Eq. max,a,Qd = 1.1
8.1.2-1 Kd,max = Kd max,Kd (E-23)
GSID max,t,Kd = 1.1
max,t,Qd = 1.4
Eq.
8.1.2-2 Kd,min = Kd min,Kd (E-24)
max,scrag,Kd = 1.0
GSID
max,scrag,Qd = 1.0
Eq.
8.1.2-3 Qd,max = Qd max,Qd (E-25)
Applying a system adjustment factor of 0.66 for an
GSID
other bridge, the maximum property modification
Eq.
factors become:
8.1.2-4 Qd,min = Qd min,Qd (E-26)
GSID
max,a,Kd = 1.0 + 0.1(0.66) = 1.066
max,a,Qd = 1.0 + 0.1(0.66) = 1.066
Determination of the system property modification
factors should include consideration of the effects of max,t,Kd = 1.0 + 0.1(0.66) = 1.066
temperature, aging, scragging, velocity, travel (wear) max,t,Qd = 1.0 + 0.4(0.66) = 1.264
and contamination as shown in Table E6-2. In lieu of
tests, numerical values for these factors can be max,scrag,Kd = 1.0
obtained from Appendix A, GSID. max,scrag,Qd = 1.0
Adjustment factors are applied to individual -factors Therefore the maximum overall modification factors
(except v) to account for the likelihood of occurrence
of all of the maxima (or all of the minima) at the same max,Kd = 1.066(1.066)1.0 = 1.14
time. These factors are applied to all -factors that max,Qd = 1.066(1.264)1.0 = 1.35
deviate from unity but only to the portion of the -
factor that is greater than, or less than, unity. Since the possible variation in upper bound properties
exceeds 15% a reanalysis of the bridge is required to
determine performance with these properties.

246

Art. 8.2.2 GSID gives these factors as follows: The upper-bound properties are:
1.00 for critical bridges Qd,max = 1.35 (15.33) = 20.70 k
0.75 for essential bridges and
0.66 for all other bridges Kd,ma x=1.14(3.41) = 3.89 k/in

Table E6-2. Minimum and maximum values for

system property modification factors.

Eq. min,Kd = (min,t,Kd) (min,a,Kd)

8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-27)
GSID (min,scrag,Kd)
Eq. max,Kd = (max,t,Kd) (max,a,Kd)
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-28)
GSID (max,scrag,Kd)
Eq. min,Qd = (min,t,Qd) (min,a,Qd)
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-29)
GSID (min,scrag,Qd)
Eq. max,Qd = (max,t,Qd) (max,a,Qd)
8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd) (E-30)
GSID (max,scrag,Qd)

Determination of the system property modification

factors should include consideration of the effects of
temperature, aging, scragging, velocity, travel (wear)
and contamination as shown in Table E6-2. In lieu of
tests, numerical values for these factors can be
obtained from Appendix A, GSID.

Adjustment factors are applied to individual -factors

(except v) to account for the likelihood of occurrence
of all of the maxima (or all of the minima) at the same
time. These factors are applied to all -factors that
deviate from unity but only to the portion of the -
factor that is greater than, or less than, unity. Art.
8.2.2 GSID gives these factors as follows:
1.00 for critical bridges
0.75 for essential bridges
0.66 for all other bridges

As required in Art. 7 GSID and shown in Fig. C7-1

GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Qd,max. As indicated in Fig C7-1 GSID, maximum
displacements will probably be given by the first case
(Kd,min and Qd,min) and maximum forces by the second
case (Kd,max and Qd,max).

247

 E7. Design and Performance Summary, Example

E7. Design and Performance Summary
2.1
E7.1 Isolator dimensions E7.1 Isolator dimensions, Example 2.1
Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E7.1-1.
Overall diameter (includes cover layer)
Overall height Table E7.1-1 Isolator Dimensions
Diameter lead core Isolator Overall size Overall size Diam.
Bonded diameter Location including without lead
Number of rubber layers mounting mounting core
Thickness of rubber layers plates (in) plates (in) (in)
Total rubber thickness Under
Thickness of steel shims edge 21.25 x 21.25 17.25 dia.
4.13
Shear modulus of elastomer girder x 8.125(H) x 6.625(H)
on Pier 1
Check all dimensions with manufacturer.

Isolator No. of Rubber Total Steel

Location rubber layers rubber shim
layers thick- thick- thick-
ness ness ness
(in) (in) (in)
Under
edge
14 0.25 3.50 0.125
girder
on Pier 1

Shear modulus of elastomer = 60 psi

E7.2 Bridge Performance E7.2 Bridge Performance, Example 2.1

Summarize bridge performance
Maximum superstructure displacement
(longitudinal)
Maximum superstructure displacement
(transverse)
Maximum superstructure displacement
(resultant)
Maximum column shear (resultant)
Maximum column moment (about transverse
axis)
Maximum column moment (about longitudinal
axis)
Maximum column torque
Check required performance as determined in Step
A3, is satisfied.
Bridge performance is summarized in Table E7.2-1
where it is seen that the maximum column shear is
121 k. This less than the column plastic shear (128k)
and therefore the required performance criterion is
satisfied (fully elastic behavior).
However the maximum longitudinal displacement is
3.79 in which is more than the 2.5 in available at the
abutment expansion joints. As discussed in Step
B1.12, the best option here is to increase the clearance
at the abutment to allow for this movement. Anything
less will lead to impact at the abutments and whereas
such damage will not be life threatening, it will not
satisfy the performance requirement in Step A3 (fully
elastic behavior).

248

 . Table E7.2-1 Summary of Bridge Performance

Maximum superstructure
3.67 in
displacement (longitudinal)
Maximum superstructure
3.20 in
displacement (transverse)
Maximum superstructure
3.79 in
displacement (resultant)
Maximum column shear
121 k
(resultant)
Maximum column moment
2809 kft
about transverse axis
Maximum column moment
2873 kft
about longitudinal axis
Maximum column torque 29 kft

249

 SECTION 2
Steel Plate Girder Bridge, Long Spans, Single-Column Pier

DESIGN EXAMPLE 2.2: S1 = 0.6g

Design Examples in Section 2

Site
ID Description S1 Column height Skew Isolator type
Class
2.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
2.1 Change site class 0.2g D Same 0 Lead rubber bearing
2.2 Change spectral acceleration, S1 0.6g B Same 0 Lead rubber bearing
Spherical friction
2.3 Change isolator to SFB 0.2g B Same 0
bearing
2.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
2.5 Change column height 0.2g B H1=0.5H2 0 Lead rubber bearing
0
2.6 Change angle of skew 0.2g B Same 45 Lead rubber bearing

250

DESIGN PROCEDURE
STEP A: BRIDGE AND SITE DATA
A1. Bridge Properties
Determine properties of the bridge:
number of supports, m
number of girders per support, n
angle of skew
weight of superstructure including railings,
curbs, barriers and other permanent loads,
WSS
weight of piers participating with
superstructure in dynamic response, WPP
weight of superstructure, Wj, at each
support
stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge. The calculation of these
quantities requires careful consideration of
several factors such as the use of cracked
sections when estimating column or wall
flexural stiffness, foundation flexibility,
and effective column height.
column shear strength (minimum value).
This will usually be derived from the
minimum value of the column flexural
yield strength, the column height, and
whether the column is acting in single or
double curvature in the direction under
consideration.
allowable movement at expansion joints
isolator type if known, otherwise to be
selected
A2. Seismic Hazard
Determine seismic hazard at site:
acceleration coefficients
site class and site factors
seismic zone
Plot response spectrum.
Use Art. 3.1 GSID to obtain peak ground and spectral
acceleration coefficients. These coefficients are the
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (Figures 3.10.2.1-
1 3.10.2.1-21LRFD).
Use Art. 3.2 to obtain Site Class and corresponding Site
Factors (Fpga, Fa and Fv). These data are the same as for
conventional bridges and Art 3.2 refers the designer to
the corresponding articles in the LRFD Specifications,
DESIGN EXAMPLE 2.2 (S1 = 0.6g)
A1. Bridge Properties, Example 2.2
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 3
Angle of skew = 00
Number of columns per support = 1
Weight of superstructure including permanent
loads, WSS = 1651.32 k
Weight of superstructure at each support:
o W1 = 168.48 k
o W2 = 657.18 k
o W3 = 657.18 k
o W4 = 168.48 k
Participating weight of piers, WPS = 256.26 k
Effective weight (for calculation of period),
Weff = Wss + Wps = 1907.58 k
Stiffness of each pier in the both directions:
o Ksub,pier1 = 288.87 k/in
o Ksub,pier2 = 288.87 k/in
Minimum column shear strength based on
flexural yield capacity of column = 128 k
Displacement capacity of expansion joints
(longitudinal) = 2.5 in for thermal and other
movements
Lead-rubber isolators
A2. Seismic Hazard, Example 2.2
Acceleration coefficients for bridge site are given in
design statement as follows:
PGA = 0.58
S1 = 0.60
SS = 1.38
Bridge is on a rock site with shear wave velocity in
upper 100 ft of soil = 3,000 ft/sec.
Table 3.10.3.1-1 LRFD gives Site Class as B.
Tables 3.10.3.2-1, -2 and -3 LRFD give following
Site Factors:
Fpga = 1.0
Fa = 1.0
Fv = 1.0
251

i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3,

LRFD.

Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give

modified spectral acceleration coefficients that include
site effects as follows:
As = Fpga PGA As = Fpga PGA = 1.0(0.58) = 0.58
SDS = Fa SS SDS = Fa SS = 1.0(1.38) = 1.38
SD1 = Fv S1 SD1 = Fv S1 = 1.0(0.60) = 0.60

Seismic Zone is determined by value of SD1 in Since 0.60 < SD1, bridge is located in Seismic Zone 4.
accordance with provisions in Table 5-1 GSID.

These coefficients are used to plot design response Design Response Spectrum is as below:
spectrum as shown in Fig. 4-1 GSID.

A3. Performance Requirements A3. Performance Requirements, Example 2.2

Determine required performance of isolated bridge As in previous examples, the owner has specified full
during Design Earthquake (1000-yr return period). functionality following the earthquake and therefore
the columns must remain elastic (no yield).
Examples of performance that might be specified by
the Owner include: To remain elastic the maximum lateral load on the
Reduced displacement ductility demand in pier must be less than the load to yield any one
columns, so that bridge is open for emergency column (128 k).
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in The maximum shear in the column must therefore be
columns or yield elsewhere in bridge), so that less than 128 k in order to keep the column elastic
bridge is fully functional and open to all vehicles and meet the required performance criterion.
immediately following earthquake.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.

252

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for

Isolator Force, F
non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most
have bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use Ku
equivalent linear springs and viscous damping to
represent the isolators, so that linear methods of analysis
may be used to determine response. Since equivalent Ku
dy disol
Isolator
Ku Displacement, d
properties such as Kisol are dependent on displacement
(d), and the displacements are not known at the
beginning of the analysis, an iterative approach is Kd
required. Note that in Art 7.1, GSID, keff is used for the
effective stiffness of an isolator unit and Keff is used for disol = Isolator displacement
the effective stiffness of a combined isolator and dy
Fisol
= Isolator yield displacement
= Isolator shear force
substructure unit. To minimize confusion, Kisol is used in Fy = Isolator yield force
Kd = Post-elastic stiffness of isolator
this document in place of keff. There is no change in the Kisol = Effective stiffness of isolator
use of Keff and Keff,j, but Ksub is used in place of ksub. Ku = Loading and unloading stiffness (elastic stiffness)
Qd = Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement (disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 2.2
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.60 6.0
GSID

(2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not

253

 occur under non-seismic loads (e.g. wind)

but low enough that yield does occur under
earthquake. Experience has shown that
taking Qd to be 5% of the bridge weight is a
good starting point, i.e.
0.05 (B-2) 0.05 0.05 1651.32 82.56
(3) Post-yield stiffness, Kd
Art 12.2 GSID requires that all isolators
exhibit a minimum lateral restoring force at
the design displacement, which translates to a
minimum post yield stiffness Kd,min given by
Eq. B-3.

Art.
0.025
12.2 , (B-3) 1651.32
GSID 0.05 0.05 13.76 /
6.0
Experience has shown that a good starting
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 2.2
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength,
Qd, and stiffness, Kd, values in proportion to the dead ,

load applied at that support: o Qd, 1 = 8.42 k

o Qd, 2 = 32.86 k
o Qd, 3 = 32.86 k
o Qd, 4 = 8.42 k
, (B-4)
and and

, (B-5) ,
o Kd,1 = 1.40 k/in
o Kd,2 = 5.48 k/in
o Kd,3 = 5.48 k/in
o Kd,4 = 1.40 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 2.2
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of
the isolators at support j (Kisol,j) and the stiffness of
the substructure Ksub,,j. See figure below (after Fig.
7.1-1 GSID). , ,

, ,
An expression for Keff,j, is given in Eq.7.1-6 GSID, but
a more useful formula as follows (MCEER 2006): o 1 = 2.81x10-6
,
(B-6) o 2 = 3.86x10-2
,
1 o 3 = 3.86x10-2
where o 4 = 2.81x10-6
, ,
(B-7)
, ,
and Ksub,j for the piers are given in Step A1. For ,
,
abutments, take Ksub,j to be a large number, say 1

254

 F o Keff,1 = 2.81k/in
Kd
o Keff,2 = 10.75 k/in
Qd Kisol
o Keff,3 = 10.75 k/in
o Keff,4 = 2.81 k/in
dy disol
Superstructure
Isolator Effective Stiffness, Kisol

F
Isolator(s), Kisol
Ksub

Substructure, Ksub
dsub

Substructure Stiffness, Ksub

dsub disol F

d
Keff

d = disol + dsub

Combined Effective Stiffness, Keff

10,000 k/in, unless actual stiffness values are

available. Note that if the default option is chosen,
unrealistically high values for Ksub,j will give
unconservative results for column moments and shear
forces.

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 2.2
Calculate the total effective stiffness, Keff, of the
bridge:
,
Eq.
7.1-6 , ( B-8) = 27.11 k/in
GSID

B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 2.2
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1

(B-9) o disol,1 = 6.00 in

,
1 o disol,2 = 5.78 in
o disol,3 = 5.78 in
o disol,4 = 6.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the stiffness of the isolation system at Example 2.2
support j, Kisol,j, for all supports:
,
, ,
,
,
(B-10) o Kisol,1 = 2.81 k/in
, ,
, o Kisol,2 = 10.75 k/in
o Kisol,3 = 10.75 k/in
o Kisol,4 = 2.81 k/in

255

B1.7 Substructure Displacement at Each Support B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j, Example 2.2
for all supports:

, , (B-11) , ,

o dsub,1 = 0 in
o d sub,2 = 0.22 in
o d sub,3 = 0.22 in
o d sub,4 = 0 in

B1.8 Substructure Shear at Each Support B1.8 Lateral Load in Each Substructure, Example
Calculate the shear at support j, Fsub,j, for all 2.2
supports:
, , , (B-12) , , ,

where values for Ksub,j are given in Step A1. o F sub,1 = 16.85 k
o F sub,2 = 64.49 k
o F sub,3 = 64.49 k
o F sub,4 = 16.85 k

B1.9 Column Shear at Each Support B1.9 Column Shear Force at Each Support,
Calculate the shear in column k at support j, Fcol,j,k, Example 2.2
assuming equal distribution of shear for all columns at
support j:
,
, ,
#
,
, ,
# (B-13) o F col,2,1 = 64.49 k
o F col,3,1 = 64.49 k

Use these approximate column shears as a check on These column shears are less than the plastic shear
the validity of the chosen strength and stiffness capacity of each column (128 k) as required in Step
characteristics. A3 and the chosen strength and stiffness values in
Step B1.1 are therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 2.2
damping ratio, , of the bridge:

Eq. 1907.58
7.1-5 2 (B-14) 2 2
GSID 386.4 27.11

and = 2.68 sec

Eq. 2 , , , and assuming dy,j = 0:

7.1-10 (B-15)
, , ,
GSID 2 , , ,
0.31
, , ,
where dy is the yield displacement of the isolator and
assumed to be negligible compared to d, i.e., take dy =

256

0 for the Simplified Method.
B1.11 Damping Factor
Calculate the damping factor, BL, and the
displacement, d, of the bridge:
Eq. .
, 0.3
.
7.1-3 (B-16)
1.7, 0.3
GSID
Eq. 9.79 (B-17)
7.1-4
GSID
B1.12 Convergence Check
Compare the new displacement with the initial value
assumed in Step B1.1. If there is close agreement, go
to the next step; otherwise repeat the process from
Step B1.3 with the new value for displacement as the
assumed displacement.
This iterative process is amenable to solution using a
spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
B1.11 Damping Factor, Example 2.2
Since 0.31 0.3
1.70
and
9.79 9.79 0.6 2.68
9.34
1.70
B1.12 Convergence Check, Example 2.2
Since the calculated value for displacement, d (= 9.34)
is not close to that assumed at the beginning of the
cycle (Step B1.1, d = 6.0), use the value of 9.34 as the
new assumed displacement and repeat from Step
B1.3.
After several iterations, convergence is reached at a
superstructure displacement of 14.19 in, with an
effective period of 4.11 seconds, and a damping factor
of 1.7 (32% damping ratio). The displacement in the
isolators at Pier 1 is 13.97 in and the effective
stiffness of the same isolators is 4.59 k/in.
See spreadsheet in Figure B1.12-1 for results of final
iteration.
Since the column shear must equal the isolator shear
for equilibrium, the column shear = 4.59 (13.97) =
64.12 k which is less than the maximum allowable
(128 k) if elastic behavior is to be achieved (as
required in Step A3).
But the superstructure displacement = 14.19 in, which
far exceeds the available clearance of 2.5 in.
There are three choices here:
1. Increase the clearance at the abutment to say
15 in to avoid impact. (Note that the
minimum required is
8 8 0.60 4.11
11.60 .
1.7
This could be expensive.
2. Allow impact to happen which will damage
abutment back wall and require repair. This
option would violate the elastic performance
requirement in Step A3.
3. Redesign the isolators. Since the column
257

 shear force is only one-half of the capacity

(64.12 vs 128 k) there is room to increase the
characteristic strength Qd and post yield
stiffness Kd to increase this shear and reduce
the displacements.

One of the many possible solutions here is to

increase Qd to 0.09W and Kd to 0.09W/d. In
this case, it will be found that the
superstructure displacement reduces to 8.00
in, the effective period is 2.31 seconds, and
the damping factor remains at 1.7 (31%
damping ratio).

See spreadsheet in Figure B1.12-2 for results

of final iteration.

The displacement in the isolators at Pier 1 is

7.60 in and the effective stiffness of the same
isolators is 14.42 k/in. The column shear is
therefore = 14.42 (7.60) = 109.6 k which is
still below the capacity of 128 k, so that the
column remains elastic as required. Although
this is a much more efficient design, the
superstructure displacement (8.0 in) still
exceeds the capacity (2.5 in) and the
recommended option is to increase the
clearance at the abutments to, say, 9.0 in (the
minimum required using 8 SD1Teff/BL is 6.6
in). This option is revisited in Step E7.2.

Option 3 is recommended and the following

properties are assumed for the isolation system going
forward to the next step (Step B2):

0.09 0.09 1651.32 148.62

and
0.09 0.09 1651.32
18.57 /
8.00

258

 Table B1.12-1 Simplified Method Solution for Design Example 2.2 - Final Iteration, First Solution

Qd = 0.05W

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
1651.32 256.26 1907.58 0.6 3

StepB1.1 d 14.20 Assumeddisplacement

Qd 82.57 Characteristicstrength
Kd 5.81 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 168.48 8.424 0.593 10,000.00 0.000119 1.186 14.198 1.187 0.002 16.847 119.607 239.227
Pier1 657.18 32.859 2.314 288.87 0.016151 4.591 13.974 4.665 0.226 65.196 459.182 925.780
Pier2 657.18 32.859 2.314 288.87 0.016151 4.591 13.974 4.665 0.226 65.196 459.182 925.780
Abut2 168.48 8.424 0.593 10,000.00 0.000119 1.186 14.198 1.187 0.002 16.847 119.607 239.227
Total 1651.32 82.566 5.815 K eff,j 11.555 164.085 1,157.577 2,330.014
Step B1.4

StepB1.10 T eff 4.11 Effectiveperiod

0.32 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.74

B L 1.70 DampingFactor
d 14.19 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1
Pier1
Pier2
Abut2

259

 Table B1.12-2 Simplified Method Solution for Design Example 2.2 - Final Iteration, Second Solution

Qd = 0.09W

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PS W eff S D1 n
1651.32 256.26 1907.58 0.6 3

StepB1.1 d 8.00 Assumeddisplacement

Qd 148.62 Characteristicstrength
Kd 18.58 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 168.48 15.163 1.895 10,000.00 0.000379 3.790 7.997 3.792 0.003 30.321 121.260 242.565
Pier1 657.18 59.146 7.393 288.87 0.052532 14.418 7.601 15.175 0.399 115.340 449.554 922.723
Pier2 657.18 59.146 7.393 288.87 0.052532 14.418 7.601 15.175 0.399 115.340 449.554 922.723
Abut2 168.48 15.163 1.895 10,000.00 0.000379 3.790 7.997 3.792 0.003 30.321 121.260 242.565
Total 1651.32 148.619 18.577 K eff,j 36.415 291.322 1,141.626 2,330.577
Step B1.4

StepB1.10 T eff 2.31 Effectiveperiod

0.31 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.73

B L 1.70 DampingFactor
d 7.99 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 5.054 0.632 1.264 7.70 1.288
Pier1 19.715 2.464 5.058 7.04 5.265
Pier2 19.715 2.464 5.058 7.04 5.265
Abut2 5.054 0.632 1.264 7.70 1.288

260

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and then
applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 2.2

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 by
elastic stiffness, Kd,i, of each isolator i as follows: the number of isolators at each support (n = 3), the
, following values for Qd /isolator and Kd /isolator are
, (B-19)
obtained:
and o Qd, 1 = 15.16/3 = 5.05 k
, o Qd, 2 = 59.15/3= 19.72 k
, ( B-20)
o Qd, 3 = 59.15/3 = 19.72 k
where values for Qd,j and Kd,j are obtained from the o Qd, 4 = 15.16/3 = 5.05 k
final cycle of iteration in the Simplified Method (Step
B1. 12 and
o Kd,1 =1.89/3 = 0.63 k/in
o Kd,2 = 7.39/3 = 2.43 k/in
o Kd,3 = 7.39/3 = 2.43 k/in
o Kd,4 = 1.89/3 = 0.63 k/in

Note that the Kd values per support used above are

from the final iteration given in Table B1.12-1. These
are not the same as the initial values in Step B1.2,
because they have been adjusted from cycle to cycle,
such that the total Kd summed over all the isolators
satisfies the minimum lateral restoring force
requirement for the bridge, i.e. Kdtotal = 0.05 W/d. See
Step B1.1. Since d varies from cycle to cycle, Kd,j
varies from cycle to cycle.

B2.2 Initial Stiffness and Yield Displacement B2.2 Initial Stiffness and Yield Displacement,
Calculate the initial stiffness, Ku,i, and the yield Example 2.2
displacement, dy,i, for each isolator i as follows: Since the isolator type has been specified in Step A1
to be an elastomeric bearing (i.e. not a friction-based
In the absence of isolator-specific information take bearing), calculate Ku,i and dy,i for an isolator on Pier 1
as follows:
, 10 , ( B-21) , 10 , 10 2.43 24.3 /
and then
,
and
, ( B-22) , 19.72
, , , 0.90
, , 24.3 2.43

As expected, the yield displacement is small

compared to the expected isolator displacement (~8

261

 in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.

B2.3 Isolator Effective Stiffness, kisol,i B2.3 Isolator Effective Stiffness, kisol,i, Example 2.2
Calculate the isolator stiffness, kisol,i, of each isolator Dividing the results for Kisol (Step B1.12) among the 3
i: isolators at each support, the following values for Kisol
/isolator are obtained:
, o Kisol,1 = 3.79/3 = 1.26 k/in
, (B -23)
o Kisol,2 = 15.18/3 = 5.06 k/in
o Kisol,3 = 15.18/3 = 5.06 k/in
o Kisol,4 = 3.79/3 = 1.26 k/in

B2.4 Finite Element Model B2.4 Finite Element Model, Example 2.2
Using computer-based structural analysis software,
create a 3-dimensional model of the bridge with the
isolators represented by spring elements. The stiffness .
of each isolator element in the horizontal axes (Kx and
Ky in global coordinates, K2 and K3 in typical local
coordinates) is the Kisol value calculated in the
previous step. For bridges with regular geometry and
minimal skew or curvature, the superstructure may be
represented by a single stick provided the load path
to each individual isolator at each support is explicitly
modeled, usually by a rigid cap beam and a set of rigid
links. If the geometry is irregular, or if the bridge is
skewed or curved, a finite element model is
recommended to accurately capture the load carried
by each individual isolator. If the piers have an
unusual weight distribution, such as a pier with a
hammerhead cap beam, a more rigorous model is
recommended.

B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in Step
A2 is for a 5% damped system. It is modified in this
step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the effective
period of the bridge, Teff, by the damping factor, BL.
B.2.5 Composite Design Response Spectrum, Ex
2.2
From the final results of Simplified Method (Step
B1.12), BL = 1.70 and Teff = 2.31 sec. Hence the
transition in the composite spectrum from 5% to 30%
damping occurs at 0.8 Teff = 0.8 (2.31) = 1.85 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.85
sec by 1.70.
1.6

1.4

1.2

1
Csm (g)

0.8

0.6

0.4

0.2

0
0 0.5 1 1.5 2 2.5 3 3.5 4

T (sec)

262

B2.6 Multimodal Analysis of Finite Element Model
Input the composite response spectrum as a user-
specified spectrum in the software, and define a load
case in which the spectrum is applied in the
longitudinal direction. Analyze the bridge for this
load case.
B2.6 Multimodal Analysis of Finite Element
Model, Example 2.2
Results of modal analysis of the example bridge are
summarized in Table B2.6-1 Here the modal periods
and mass participation factors of the first 12 modes
are given. The first three modes are the principal
transverse, longitudinal, and torsion modes with
periods of 2.304, 2.224 and 2.192 sec respectively.
The period of the longitudinal mode (2.30 sec) is
close to the period calculated in the Simplified
Method (2.31 sec). The mass participation factors
indicate there is no coupling between these three
modes (probably due to the symmetric nature of the
bridge) and the high values for the first and second
modes (90% for each mode) indicate the bridge is
behaving essentially in a single mode of vibration in
each direction. Similar results to those obtained by the
Simplified Method are therefore to be expected.

Table B2.6-1 Modal Properties of Bridge

Example 2.2 First Iteration
Mode Period MassParticipatingRatios
No Sec UX UY UZ RX RY RZ
1 2.304 0.000 0.904 0.000 0.898 0.000 0.686
2 2.224 0.903 0.000 0.000 0.000 0.020 0.000
3 2.192 0.000 0.000 0.000 0.000 0.000 0.228
4 0.499 0.000 0.004 0.000 0.052 0.000 0.003
5 0.372 0.000 0.000 0.074 0.000 0.055 0.000
6 0.364 0.000 0.000 0.000 0.000 0.000 0.003
7 0.347 0.000 0.004 0.000 0.019 0.000 0.003
8 0.285 0.000 0.007 0.000 0.018 0.000 0.005
9 0.255 0.000 0.000 0.000 0.000 0.000 0.000
10 0.255 0.096 0.000 0.000 0.000 0.001 0.000
11 0.207 0.000 0.000 0.000 0.000 0.128 0.000
12 0.188 0.000 0.000 0.000 0.000 0.000 0.001

Computed values for the isolator displacements due to

a longitudinal earthquake are as follows (numbers in
parentheses are those used to calculate the initial
properties to start iteration from the Simplified
Method):

o disol,1 = 7.80 (8.00) in

o disol,2 = 7.02 (7.60) in
o disol,3 = 7.02 (7.60) in
o disol,4 = 7.80 (8.00) in

B2.7 Convergence Check B2.7 Convergence Check, Example 2.2

Compare the resulting displacements at the
superstructure level (d) to the assumed displacements.
These displacements can be obtained by examining
the joints at the top of the isolator spring elements. If
in close agreement, go to Step B2.9. Otherwise go to
Step B2.8.
The results for isolator displacements could be
considered close enough (better than 4% at abutments
and 9% at piers). But for illustrative purposes a
second cycle of iteration is performed.
Go to Step B2.8 and update properties.

263

B2.8 Update Kisol,i Keff,j, and BL B2.8 Update Kisol,i Keff,j, and BL, Example 2.2
Use the calculated displacements in each isolator Updated values for Kisol,i are given below (previous
element to obtain new values of Kisol,i for each isolator values are in parentheses):
as follows:
, o Kisol,1 = 1.27 (1.26) k/in
, , (B-24) o Kisol,2 = 5.24 (5.06) k/in
,
o Kisol,3 = 5.24 (5.06) k/in
Recalculate Keff,j : o Kisol,4 = 1.27 (1.26) k/in

Eq. , ,
7.1-6 ,
(B-25) Since the isolator displacements are relatively close to
, ,
GSID previous results no significant change in the damping
ratio is expected. Hence Keff,j and are not
Recalculate system damping ratio, : recalculated and BL is taken at 1.70.

Eq. 2 , , ,
7.1-10 , , ,
(B-26)
GSID
Recalculate system damping factor, BL:

Eq. .
0.3
.
7.1-3 ( B-27)
1.7 0.3
GSID
Since the change in effective period is very small
Obtain the effective period of the bridge from the (2.224 to 2.183 sec) and no change has been made to
multi-modal analysis and with the revised damping BL, there is no need to construct a new composite
factor (Eq. B-27), construct a new composite response response spectrum in this case. Go back to Step B2.6
spectrum. Go to Step B2.6. (see immediately below).

B2.6 Multimodal Analysis Second Iteration,

Example 2.2
Reanalysis gives the following values for the isolator
displacements (numbers in parentheses are those
from the previous cycle):

o disol,1 = 7.65 (7.80) in

o disol,2 = 6.86 (7.02) in
o disol,3 = 6.86 (7.02) in
o disol,4 = 7.65 (7.80) in

Go to Step B2.7

B2.7 Convergence Check B2.7 Convergence Check, Example 2.2

Compare results and determine if convergence has Satisfactory agreement has been reached on this
been reached. If so go to Step B2.9. Otherwise Go to second cycle (better than 2% at the abutments and 3%
Step B2.8. at the piers). Go to Step B2.9

B2.9 Superstructure and Isolator Displacements B2.9 Superstructure and Isolator Displacements,
Once convergence has been reached, obtain Example 2.2
o superstructure displacements in the longitudinal From the above analysis:
(xL) and transverse (yL) directions of the bridge, o superstructure displacements in the
and longitudinal (xL) and transverse (yL) directions
o isolator displacements in the longitudinal (uL) are:
and transverse (vL) directions of the bridge, for xL= 7.69 in
each isolator, for this load case (i.e. yL= 0.0 in
longitudinal loading). These displacements

264

 may be found by subtracting the nodal o isolator displacements in the longitudinal (uL)
displacements at each end of each isolator and transverse (vL) directions are:
spring element. o Abutments: uL = 7.65 in, vL = 0.00 in
o Piers: uL = 6.86 in, vL = 0.00 in

All isolators at same support have the same

displacements.

B2.10 Pier Bending Moments and Shear Forces B2.10 Pier Bending Moments and Shear Forces,
Obtain the pier bending moments and shear forces in Example 2.2
the longitudinal (MPLL, VPLL) and transverse (MPTL, Bending moments in single column pier in the
VPTL) directions at the critical locations for the longitudinal (MPLL) and transverse (MPTL) directions
longitudinally-applied seismic loading. are:
MPLL= 0
MPTL= 3,897 kft

Shear forces in single column pier the longitudinal

(VPLL) and transverse (VPTL) directions are
VPLL= 163.79 k
VPTL= 0

The above column shear (163.8 k) is larger than the

calculated shear in the Simplified Method (115.3 k)
because inertia load from the pier cap is not included
in the Simplified Method. The pier cap weighs 92 k.

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 2.2
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1

Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Long. Transv. Axial
shear shear forces
due to due to due to
long. long. long.
EQ EQ EQ
Isol. 1 9.72 0 1.93
Abut
Isol. 2 9.72 0 1.91
ment
Isol. 3 9.72 0 1.93
Isol. 1 35.93 0 0.68
Pier Isol. 2 35.97 0 0.98
Isol. 3 35.93 0 0.68

265

STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake
Repeat the above process, starting at Step B1, for
earthquake loading in the transverse direction of the
bridge. Support flexibility in the transverse direction
is to be included, and a composite response spectrum
is to be applied in the transverse direction. Obtain
isolator displacements in the longitudinal (uT) and
transverse (vT) directions of the bridge, and the biaxial
bending moments and shear forces at critical locations
in the columns due to the transversely-applied seismic
loading.
C1. Analysis for Transverse Earthquake, Example
2.2
Key results from repeating Steps B1 and B2
(Simplified and Mulitmode Spectral Methods) for
transverse loading , are as follows:
o Teff = 2.23 sec
o Superstructure displacements in the longitudinal
(xT) and transverse (yT) directions due to
transverse load are as follows:
xT = 0 in
yT = 7.88 in
o Isolator displacements in the longitudinal (uT)
and transverse (vT) directions due to transverse
loading are as follows:
Abutments uT = 0.00 in, vT = 7.93 in
Piers uT = 0.00 in, vT = 6.25 in
o Pier bending moments in the longitudinal (MPLT)
and transverse (MPTT) directions due to
transverse load are as follows:
MPLT = 3,345 kft
MPTT = 0 kft
o Pier shear forces in the longitudinal (VPLT) and
transverse (VPTT) directions due to transverse
load are as follows:
VPLT = 0 k
VPTT = 127.1 k
o Isolator shear and axial forces are summarized in
Table C1-1.

266

 Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Long. Transv. Axial
shear shear forces
due to due to due to
transv. transv. transv.
EQ EQ EQ
Isol. 1 0.0 9.91 26.67
Abut
Isol. 2 0.0 9.92 0
ment
Isol. 3 0.0 9.91 26.67
Isol. 1 0.0 34.41 50.0
Pier Isol. 2 0.0 34.51 0
Isol. 3 0.0 34.41 50.0

267

STEP D. CALCULATE DESIGN VALUES

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements at Pier,

Following the provisions in Art. 2.1 GSID, and Example 2.2
illustrated in Fig. 2.1-1 GSID, calculate the total Load Case 1:
design displacement, dt, by combining the u1 = uL + 0.3uT = 1.0(6.86) + 0.3(0) = 6.86 in
displacements from the longitudinal (uL and vL) and v1 = vL + 0.3vT = 1.0(0) + 0.3(6.25) = 1.88 in
transverse (uT and vT) cases as follows: R1 = = 6.86 1.88 = 7.11 in
u1 = uL + 0.3uT (D-1)
v1 = vL + 0.3vT (D-2) Load Case 2:
R1 = (D-3) u2 = 0.3uL + uT = 0.3(6.86) + 1.0(0) = 2.06 in
v2 = 0.3vL + vT = 0.3(0) + 1.0(6.25) = 6.25 in
u2 = 0.3uL + uT (D-4) R2 = = 2.06 6.25 = 6.58 in
v2 = 0.3vL + vT (D-5)
R2 = (D-6) Governing Case:
Total design displacement, dt = max(R1, R2)
dt = max(R1, R2) (D-7) = 7.11in

D2. Design Moments and Shears D2. Design Moments and Shears at Pier, Example
Calculate design values for column bending moments 2.2
and shear forces using the same combination rules as Load Case 1:
for displacements. VPL1= VPLL + 0.3VPLT = 1.0(163.79) + 0.3(0)=163.79 k
VPT1= VPTL + 0.3VPTT = 1.0(0) + 0.3(127.10) = 38.13 k
Alternatively this step may be deferred because the R1 = = 163.79 38.13 = 168.17 k
above results may not be final. Upper and lower
bound analyses are required after the isolators have Load Case 2:
been designed as described in Art 7. GSID. These VPL2= 0.3VPLL + VPLT = 0.3(163.79) + 1.0(0) = 49.14 k
analyses are required to determine the effect of VPT2= 0.3VPTL + VPTT = 0.3(0) + 1.0(127.10) =127.10 k
possible variations in isolator properties due age,
R2 = = 49.14 127.10 = 136.27 k
temperature and scragging in elastomeric systems.
Accordingly the results for column shear in Steps
Governing Case:
B2.10 and C are likely to increase once these analyses
Design column shear = max (R1, R2)
are complete.
= 168.17 K

Note that this column shear force (168.17 k) is larger

than the plastic shear (128 k) and fully elastic
behavior might not be achievable with this seismic
demand (S1=0.6g) and column size.

268

STEP E. DESIGN OF LEAD-RUBBER (ELASTOMERIC) ISOLATORS

A lead-rubber isolator is an elastomeric
bearing with a lead core inserted on its vertical
centreline. When the bearing and lead core are
deformed in shear, the elastic stiffness of the
lead provides the initial stiffness (Ku).With
increasing lateral load the lead yields almost
perfectly plastically, and the post-yield
stiffness Kd is given by the rubber alone. More
details are given in Buckle et.al, 2006.

While both circular and rectangular bearings

are commercially available, circular bearings
are more commonly used. Consequently the
procedure given below focuses on circular
bearings. The same steps can be followed for
rectangular bearings, but some modifications will be necessary. When sizing the physical dimensions of the
bearing, plan dimensions (B, dL) should be rounded up to the next 1/4 increment, while the total thickness of
elastomer, Tr, is specified in multiples of the layer thickness. Typical layer thicknesses for bearings with lead
cores are 1/4 and 3/8. High quality natural rubber should be specified for the elastomer. It should have a shear
modulus in the range 60-120 psi and an ultimate elongation-at-break in excess of 5.5. Details can be found in
rubber handbooks or in Buckle et.al., 2006.

The following design procedure assumes the isolators are bolted to the masonry and sole plates. Isolators that use
shear-only connections (and not bolts) require additional design checks for stability which are not included below.
See Buckle et al, 2006.

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

E1. Required Properties E1. Required Properties, Example 2.2

Obtain from previous work the properties required of The design of one of the exterior isolators on a pier is
the isolation system to achieve the specified given below to illustrate the design process for a lead-
performance criteria (Step A1). rubber isolator.
required characteristic strength, Qd / isolator
required post-elastic stiffness, Kd / isolator From previous work
total design displacement, dt, for each isolator Qd / isolator = 19.72 k
maximum applied dead and live load (PDL, PLL) Kd / isolator = 2.43 k/in
and seismic load (PSL) which includes seismic Total design displacement, dt = 7.11 in
live load (if any) and overturning forces due to PDL = 187 k
seismic loads, at each isolator, and PLL = 123 k and PSL = 50 k (Table C1-1)
maximum wind load, PWL PWL = 8.21 k < Qd OK

E2. Isolator Sizing

E2.1 Lead Core Diameter E2.1 Lead Core Diameter, Example 2.2
Determine the required diameter of the lead plug, dL,
using:
19.72
(E-1) 4.68
0.9 0.9 0.9
See Step E2.5 for limitations on dL.

269

E2.2 Plan Area and Isolator Diameter
Although no limits are placed on compressive stress in
the GSID, (maximum strain criteria are used instead,
see Step E3) it is useful to begin the sizing process by
assuming an allowable stress of, say, 1.6 ksi.
E2.2 Plan Area and Isolator Diameter, Example
2.2
Looking ahead (and based on experience from
previous examples) the isolator must be stable at
= 1.5 dt = 1.5(7.11) = 10.67 in (Step E4.2).

Then the bonded area of the isolator is given by: This is a much larger displacement than in the
(E-2) benchmark example (where it was only 2.34 in). It is
1.6 therefore likely that in this example, this value for
will dictate the size of the isolator and a rule of thumb
and the corresponding bonded diameter (taking into is to choose a diameter, B, between 1.5 and 2 times ,
account the hole required to accommodate the lead in order to provide sufficient vertical load capacity
core) is given by: when the isolator is deformed to 1.5 dt.
4
(E-3) For this reason choose B = 1.75 = 18.6 ~ 19.0 in.

Round the bonded diameter, B, to nearest quarter inch,

and recalculate actual bonded area using

(E-4)
4
Then
Note that the overall diameter is equal to the bonded
diameter plus the thickness of the side cover layers 4
(usually 1/2 inch each side). In this case the overall
diameter, Bo is given by: 19.00 4.68 266.32
4

1.0 (E-5)

Bo = 19.0 + 2(0.5) = 20.0 in

E2.3 Elastomer Thickness and Number of Layers E2.3 Elastomer Thickness and Number of Layers,
Since the shear stiffness of an elastomeric bearing is Example 2.2
given by:
(E-6)
where G = shear modulus of the rubber, and
Tr = the total thickness of elastomer,
it follows Eq. E-5 may be used to obtain Tr given a
required value for Kd
0.1 266.32
(E-7) 10.96
2.43
A typical range for shear modulus, G, is 60-120 psi.
Higher and lower value are available and used in
special applications.

If the layer thickness is tr, the number of layers, n, is 10.96

43.8
given by: 0.25
(E-8)
rounded to the nearest integer. Round up to nearest integer, i.e. n = 44

Note that because of rounding the plan dimensions

and the number of layers, the actual stiffness, Kd, will
not be exactly as required. Reanalysis may be

270

necessary if the differences are large.

E2.4 Overall Height E2.4 Overall Height, Example 2.2

The overall height of the isolator, H, is given by:

1 2 ( E-9) 44 0.25 43 0.125 2 1.5 19.375

where ts = thickness of an internal shim (usually

about 1/8 in), and
tc = combined thickness of end cover plate (0.5
in) and outer plate (1.0 in)

E2.5 Lead Core Size Check E2.5 Lead Core Size Check, Example 2.2
Experience has shown that for optimum performance Since B = 22.00 check
of the lead core it must not be too small or too large.
The recommended range for the diameter is as
follows: 20.0 20.0
(E-10) 3 6
3 6
i.e., 6.67 3.33

Since dL = 4.68, lead core size is acceptable.

E3. Strain Limit Check E3. Strain Limit Check, Example 2.2
Art. 14.2 and 14.3 GSID requires that the total applied Since
shear strain from all sources in a single layer of 187.0
elastomer should not exceed 5.5, i.e., 0.702
266.32

, + 0.5 5.5 (E-11) and

266.32
17.85
where , , , are defined below. 19.0 0.25
(a) is the maximum shear strain in the layer due to
compression and is given by: then
1.0 0.702
(E-12) 0.393
0.1 17.85
where Dc is shape coefficient for compression in
circular bearings = 1.0, , G is shear
modulus, and S is the layer shape factor given by:
(E-13)

(b) , is the shear strain due to earthquake loads and

is given by: 7.11
, 0.646
(E-14) 11.0
,

(c) is the shear strain due to rotation and is given

by:
0.375 19.00 0.01
(E-15) 0.492
0.25 11.0
where Dr is shape coefficient for rotation in circular

271

bearings = 0.375, and is design rotation due to DL, Substitution in Eq E-11 gives
LL and construction effects. Actual value for may
not be known at this time and a value of 0.01 is , 0.5 0.39 0.65 0.5 0.49
suggested as an interim measure, including 1.28
uncertainties (see LRFD Art. 14.4.2.1). 5.5

E4. Vertical Load Stability Check E4. Vertical Load Stability Check, Example 2.2
Art 12.3 GSID requires the vertical load capacity of
all isolators be at least 3 times the applied vertical
loads (DL and LL) in the laterally undeformed state.

Further, the isolation system shall be stable under

1.2(DL+SL) at a horizontal displacement equal to
either
2 x total design displacement, dt, if in Seismic Zone 1
or 2, or
1.5 x total design displacement, dt, if in Seismic Zone
3 or 4.

E4.1 Vertical Load Stability in Undeformed State E4.1 Vertical Load Stability in Undeformed State,
The critical load capacity of an elastomeric isolator at Example 2.2
zero shear displacement is given by

4
1 1 3 3 0.1 0.3
2 (E-16)
0.3 1 0.67 17.85 64.03
where
19.0
6397.1
Ts = total shim thickness 64
64.03 6397
37,239 /
1 0.67 11.0
E = elastic modulus of elastomer = 3G
0.1 266.32
64 2.42 /
11.0
It is noted that typical elastomeric isolators have high
shape factors, S, in which case:
4
1 (E-17)

and Eq. E-16 reduces to:

(E-18) 2.42 37,239 943.1

Check that:

3 (E-19) 943
3.04 3
187 123

E4.2 Vertical Load Stability in Deformed State E4.2 Vertical Load Stability in Deformed State,
The critical load capacity of an elastomeric isolator at Example 2.2
shear displacement may be approximated by: Since bridge is in Zone 4,

272

 1.5 1.5 7.11 10.65

(E-20)
where
Ar = overlap area between top and bottom plates 10.65
2 1.95
of isolator at displacement (Fig. 2.2-1 19.0
GSID)
= 4
2 1.95 1.95
0.325

Agross = 4

It follows that:
0.325 943.1 306.5
(E-21)

Check that:

1 (E-22)
1.2
306.5
1.12 1
1.2 1.2 187 50.0

E5. Design Review E5. Design Review, Example 2.2

The basic dimensions of the isolator designed above
are as follows:

20.0 in (od) x 19.375 in (high) x 4.68 in dia. lead core

and the volume, excluding the steel end and cover

plates = 5,144 in3

Although this design satisfies all the required criteria,

the total rubber shear strain (1.28) is much less than
the maximum allowable (5.5), as shown in Step E3. In
other words, the isolator is not working very hard and
a redesign appears to be worth exploring to see if a
more optimal design can be found. Since the plan
dimension was set at the beginning (Step E2.2) to
satisfy vertical load stability requirements
(successfully as it turned out) the only way to
optimize the design is to reduce its height. But
reducing the height will increase the stiffness, Kd,
unless the shear modulus of the elastomer is likewise
reduced. In the redesign below, the plan dimensions
remain the same but the shear modulus is reduced
from 100 to 60 psi.

19.72
4.68
0.9 0.9

Choose B = 19.00 in as before, then

273

 19.0 4.68 266.32

4

Bo = 19.0 + 2(0.5) = 20.0 in

0.06 266.32
6.58
2.43

6.58
26.3
0.25

Round to nearest integer: n = 26

26 0.25 25 0.125 2 1.5 12.625

Since B = 19.0 check

19 19
3 6

6.33 3.17

Since dL = 4.68 in, size of lead core is acceptable.

187.0
0.702
266.32

266.32
17.85
19.0 0.25

1.0 0.702
0.66
0.06 17.85

7.11
, 1.09
6.5

0.375 19.0 0.01

0.833
0.25 6.5

, 0.5 0.66 1.09 0.5 0.83

2.17 5.5

3 3 0.06 0.18

0.18 1 0.67 17.85 38.42

19.0
6,397.1
64

274

 38.42 6,397.1
37,812 /
6.5

0.06 266.32
2.46 /
6.5

2.46 37,812 957.8

957.8
3.09 3
187 123

10.67
2 1.95
19.0

1.95 1.95
0.325

0.325 957.8 311.1

311.1
1.13 1
1.2 1.2 187 50.0

The basic dimensions of the redesigned isolator are as

follows:

20.0 in (od) x 12.625in (high) x 4.68 in dia. lead core

and the volume, excluding steel end and cover plates,

= 3,024 in3

This is a more optimal design. It is smaller than the

previous design (3,024 in3 vs 5,144 in3) while still
satisfying all the design criteria. Being smaller it
works harder to satisfy these requirements, as can be
seen by an increase in the total shear strain from 1.28
to 2.17 (but still less than the maximum allowable of
5.5).

E6. Minimum and Maximum Performance Check
Art. 8 GSID requires the performance of any isolation
system be checked using minimum and maximum
values for the effective stiffness of the system. These
values are calculated from minimum and maximum
values of Kd and Qd, which are found using system
property modification factors, as indicated in Table
E6-1.
Table E6-1. Minimum and maximum values
for Kd and Qd.
E6. Minimum and Maximum Performance Check,
Example 2.2
Minimum Property Modification factors are
min,Kd = 1.0
min,Qd = 1.0
which means there is no need to reanalyze the bridge
with a set of minimum values.
Maximum Property Modification factors are
max,a,Kd = 1.1

275

 Eq. max,a,Qd = 1.1

8.1.2-1 Kd,max = Kd max,Kd (E-23)
GSID max,t,Kd = 1.1
Eq. max,t,Qd = 1.4
8.1.2-2 Kd,min = Kd min,Kd (E-24)
GSID max,scrag,Kd = 1.0
Eq. max,scrag,Qd = 1.0
8.1.2-3 Qd,max = Qd max,Qd (E-25)
GSID
Eq.
8.1.2-4 Qd,min = Qd min,Qd (E-26)
GSID

Determination of the system property modification

factors shall include consideration of the effects of
temperature, aging, scragging, velocity, travel (wear)
and contamination as shown in Table E6-2. In lieu of
tests, numerical values for these factors can be
obtained from Appendix A, GSID.

Table E6-2. Minimum and maximum values for

system property modification factors.

Eq. min,Kd = (min,t,Kd) (min,a,Kd)

8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-27)
GSID (min,scrag,Kd)
Eq. max,Kd = (max,t,Kd) (max,a,Kd)
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-28)
GSID (max,scrag,Kd)
Eq. min,Qd = (min,t,Qd) (min,a,Qd)
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-29)
GSID (min,scrag,Qd)
Eq. max,Qd = (max,t,Qd) (max,a,Qd)
8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd) (E-30)
GSID (max,scrag,Qd)

Adjustment factors are applied to individual -factors

(except v) to account for the likelihood of occurrence
of all of the maxima (or all of the minima) at the same
time. These factors are applied to all -factors that
deviate from unity but only to the portion of the -
factor that is greater than, or less than, unity. Art.
8.2.2 GSID gives these factors as follows:
1.00 for critical bridges
0.75 for essential bridges
0.66 for all other bridges
As required in Art. 7 GSID and shown in Fig. C7-1
GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Qd,max. As indicated in Fig C7-1 GSID, maximum
displacements will probably be given by the first case
(Kd,min and Qd,min) and maximum forces by the second
Applying a system adjustment factor of 0.66 for an
other bridge, the maximum property modification
factors become:
max,a,Kd = 1.0 + 0.1(0.66) = 1.066
max,a,Qd = 1.0 + 0.1(0.66) = 1.066
max,t,Kd = 1.0 + 0.1(0.66) = 1.066
max,t,Qd = 1.0 + 0.4(0.66) = 1.264
max,scrag,Kd = 1.0
max,scrag,Qd = 1.0
Therefore the maximum overall modification factors
max,Kd = 1.066(1.066)1.0 = 1.14
max,Qd = 1.066(1.264)1.0 = 1.35

276

case (Kd,max and Qd,max).

Since the possible variation in upper bound properties
exceeds 15% a reanalysis of the bridge is required to
determine performance with these properties.

The upper-bound properties of pier isolators are:

Qd,max = 1.35 (19.72) = 26.62 k
and
Kd,ma x=1.14(2.43) = 2.77 k/in

E7. Design and Performance Summary E7. Design and Performance Summary
E7.1 Isolator dimensions E7.1 Isolator dimensions, Example 2.2
Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E7.1-1.
Overall diameter (includes cover layer)
Overall height Table E7.1-1 Isolator Dimensions
Diameter lead core Isolator Overall size Overall size Diam.
Bonded diameter Location including without lead
Number of rubber layers mounting mounting core
Thickness of rubber layers plates (in) plates (in) (in)
Total rubber thickness Under
Thickness of steel shims edge 24.0 x 24.0 20.0 dia.
Shear modulus of elastomer 4.68
girder x 12.625(H) x 11.125(H)
on Pier
Check all dimensions with manufacturer.

Isolator No. of Rubber Total Steel

Location rubber layers rubber shim
layers thick- thick- thick-

ness ness ness
(in) (in) (in)
Under
edge
26 0.25 6.50 0.125
girder
on Pier

Shear modulus of elastomer = 60 psi

E7.2 Bridge Performance E7.2 Bridge Performance, Example 2.2

Summarize bridge performance
Maximum superstructure displacement
(longitudinal)
Maximum superstructure displacement
(transverse)
Maximum superstructure displacement
(resultant)
Maximum column shear (resultant)
Maximum column moment (about transverse
axis)
Maximum column moment (about longitudinal
axis)
Maximum column torque
Check required performance as determined in Step
A3, is satisfied.
Bridge performance is summarized in Table E7.2-1
where it is seen that the maximum column shear is
175 k. This is more than the column plastic shear (128
k) and therefore the required performance criterion is
not satisfied (fully elastic behavior). Clearly the
seismic demand (S1 = 0.6g) is too high and the
column too small for isolation to give fully elastic
response. The next step might be to a conduct
pushover analysis of the column to determine if the
ductility demand at 8.21 in is acceptable. If not, and
this is an existing bridge, jacket the column (as well
as isolate the bridge). If a new design, increase the
size of the column and thereby increase its strength. It
is noted that the maximum longitudinal displacement
(7.69 in) exceeds the available clearance, and as noted
in Section B1.12, this gap needs to be increased to say

277

 9.0 in to meet the requirements for isolation. An

alternative is to accept pounding at the abutments.
The consequential damage is not likely to be life-
threatening and easily repaired.

Table E7.2-1 Summary of Bridge Performance

Maximum superstructure
7.69 in
displacement (longitudinal)
Maximum superstructure
7.88 in
displacement (transverse)
Maximum superstructure
8.21 in
displacement (resultant)
Maximum column shear
175 k
(resultant)
Maximum column moment
4049 k-ft
about transverse axis
Maximum column moment
3814 k-ft
about longitudinal axis
Maximum column torque 14 k-ft

278

SECTION 2
Steel Plate Girder Bridge, Long Spans, Single-Column Pier

DESIGN EXAMPLE 2.3: Spherical Friction Isolators

Design Examples in Section 2

Site
ID Description S1 Column height Skew Isolator type
Class
2.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
2.1 Change site class 0.2g D Same 0 Lead rubber bearing
2.2 Change spectral acceleration, S1 0.6g B Same 0 Lead rubber bearing
Spherical friction
2.3 Change isolator to SFB 0.2g B Same 0
bearing
2.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
2.5 Change column height 0.2g B H1=0.5H2 0 Lead rubber bearing
0
2.6 Change angle of skew 0.2g B Same 45 Lead rubber bearing

279

DESIGN EXAMPLE 2.3 (Spherical Friction

DESIGN PROCEDURE
STEP A: BRIDGE AND SITE DATA
A1. Bridge Properties
Determine properties of the bridge:
number of supports, m
number of girders per support, n
angle of skew
weight of superstructure including railings,
curbs, barriers and other permanent loads,
WSS
weight of piers participating with
superstructure in dynamic response, WPP
weight of superstructure, Wj, at each
support
stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge. The calculation of these
quantities requires careful consideration of
several factors such as the use of cracked
sections when estimating column or wall
flexural stiffness, foundation flexibility,
and effective column height.
column shear strength (minimum value).
This will usually be derived from the
minimum value of the column flexural
yield strength, the column height, and
whether the column is acting in single or
double curvature in the direction under
consideration.
allowable movement at expansion joints
isolator type if known, otherwise to be
selected
Isolators)
A1. Bridge Properties, Example 2.3
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 3
Angle of skew = 00
Number of columns per support = 1
Weight of superstructure including permanent
loads, WSS = 1651.32 k
Weight of superstructure at each support:
o W1 = 168.48 k
o W2 = 657.18 k
o W3 = 657.18 k
o W4 = 168.48 k
Participating weight of piers, WPP = 256.26 k
Effective weight (for calculation of period),
Weff = Wss + WPP = 1907.58 k
Stiffness of each pier in the both directions
(assume fixed at footing and single curvature
behavior) :
o Ksub,pier1 = 288.87 k/in
o Ksub,pier2 = 288.87 k/in
Minimum column shear strength based on
flexural yield capacity of column = 128 k
Displacement capacity of expansion joints
(longitudinal) = 2.5 in for thermal and other
movements
Spherical Friction Isolators

A2. Seismic Hazard A2. Seismic Hazard, Example 2.3

Determine seismic hazard at site: Acceleration coefficients for bridge site are given in
acceleration coefficients design statement as follows:
site class and site factors PGA = 0.40
seismic zone S1 = 0.20
Plot response spectrum. SS = 0.75

Use Art. 3.1 GSID to obtain peak ground and spectral
acceleration coefficients. These coefficients are the
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (e.g. Figures
3.10.2.1-1 to 3.10.2.1-21 LRFD).
Use Art. 3.2 to obtain Site Class and corresponding Site
Bridge is on a rock site with shear wave velocity in
upper 100 ft of soil = 3,000 ft/sec.
Table 3.10.3.1-1 LRFD gives Site Class as B.
Tables 3.10.3.2-1, -2 and -3 LRFD give following
Site Factors:
Fpga = 1.0
Fa = 1.0

280

Factors (Fpga, Fa and Fv). These data are the same as for Fv = 1.0
conventional bridges and Art 3.2 refers the designer to
the corresponding articles in the LRFD Specifications,
i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3,
LRFD.

Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give

modified spectral acceleration coefficients that include
site effects as follows:
As = Fpga PGA As = Fpga PGA = 1.0(0.40) = 0.40
SDS = Fa SS SDS = Fa SS = 1.0(0.75) = 0.75
SD1 = Fv S1 SD1 = Fv S1 = 1.0(0.20) = 0.20

Seismic Zone is determined by value of SD1 in Since 0.15 < SD1 < 0.30, bridge is located in Seismic
accordance with provisions in Table5-1 GSID. Zone 2.

These coefficients are used to plot design response Design Response Spectrum is as below:
spectrum as shown in Fig. 4-1 GSID.
0.9

0.8

0.7

0.6
Acceleration 0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Period(s)

A3. Performance Requirements A3. Performance Requirements, Example 2.3

Determine required performance of isolated bridge In this example, assume the owner has specified full
during Design Earthquake (1000-yr return period). functionality following the earthquake and therefore
the columns must remain elastic (no yield).
Examples of performance that might be specified by
the Owner include: To remain elastic the maximum lateral load on the
Reduced displacement ductility demand in pier must be less than the load to yield any one
columns, so that bridge is open for emergency column (128 k).
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in The maximum shear in the column must therefore be
columns or yield elsewhere in bridge), so that less than 128 k in order to keep the column elastic
bridge is fully functional and open to all vehicles and meet the required performance criterion.
immediately following earthquake.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.

281

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for

Isolator Force, F
non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most
have bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use Ku
equivalent linear springs and viscous damping to
represent the isolators, so that linear methods of analysis
may be used to determine response. Since equivalent dy disol
Isolator
Ku Ku
properties such as Kisol are dependent on displacement Displacement, d

(d), and the displacements are not known at the beginning

of the analysis, an iterative approach is required. Note Kd
that in Art 7.1, GSID, keff is used for the effective stiffness
of an isolator unit and Keff is used for the effective disol = Isolator displacement
stiffness of a combined isolator and substructure unit. To dy = Isolator yield displacement
Fisol = Isolator shear force
minimize confusion, Kisol is used in this document in Fy = Isolator yield force
place of keff. There is no change in the use of Keff and Keff,j, Kd
Kisol
= Post-elastic stiffness of isolator
= Effective stiffness of isolator
but Ksub is used in place of ksub. Ku = Loading and unloading stiffness (elastic stiffness)
Qd = Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement (disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 2.3
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.20 2.0
GSID

(2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not

282

occur under non-seismic loads (e.g. wind) but

low enough that yield will occur during an
earthquake. Experience has shown that
taking Qd to be 5% of the bridge weight is a
good starting point, i.e.
0.05 (B-2)

(3) Post-yield stiffness, Kd 0.05 0.05 1651.32 82.56

Art 12.2 GSID requires that all isolators
exhibit a minimum lateral restoring force at
the design displacement, which translates to a
minimum post yield stiffness Kd,min given by
Eq. B-3.

Art.
0.025
12.2 , (B-3)
GSID 1651.32
Experience has shown that a good starting 0.05 0.05 41.28 /
2.0
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 2.3
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength, ,

Qd, and stiffness, Kd, values in proportion to the dead o Qd, 1 = 8.42 k
load applied at that support: o Qd, 2 = 32.86 k
o Qd, 3 = 32.86 k
o Qd, 4 = 8.42 k

, (B-4)
and
and
,
, (B-5) o Kd,1 = 4.21 k/in
o Kd,2 = 16.43 k/in
o Kd,3 = 16.43 k/in
o Kd,4 = 4.21 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 2.3
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of , ,
the isolators at support j (Kisol,j) and the stiffness of , ,
the substructure Ksub,j. See figure below for
definitions (after Fig. 7.1-1 GSID). o 1 = 8.43x10-4
o 2 = 1.21x10-1
An expression for Keff,j, is given in Eq.7.1-6 GSID, but o 3 = 1.21x10-1
a more useful formula is as follows (MCEER 2006): o 4 = 8.43x10-4
,
, (B-6)
1 ,
where ,
1
, ,
(B-7) o Keff,1 = 8.42 k/in
, , o Keff,2 = 31.09 k/in
o Keff,3 = 31.09 k/in
and Ksub,j for the piers are given in Step A1. For the o Keff,4 = 8.42 k/in

283

abutments, take Ksub,j to be a large number, say 10,000

k/in, unless actual stiffness values are available. Note
that if the default option is chosen, unrealistically high
values for Ksub,j will give unconservative results for
column moments and shear forces.
F
Kd
Qd Kisol

dy disol
Superstructure
Isolator Effective Stiffness, Kisol

F
Isolator(s), Kisol
Ksub

Substructure, Ksub
dsub

Substructure Stiffness, Ksub

dsub disol F

d
Keff

d = disol + dsub

Combined Effective Stiffness, Keff

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 2.3
Calculate the total effective stiffness, Keff, of the
bridge:

Eq. , 79.02 /
7.1-6 , ( B-8)
GSID

B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 2.3
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1
, (B-9)
1 o disol,1 = 2.00 in
o disol,2 = 1.79 in
o disol,3 = 1.79 in
o disol,4 = 2.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the stiffness of the isolation system at Example 2.3
support j, Kisol,j, for all supports:
,
, ,
,
,
(B-10) o Kisol,1 = 8.43 k/in
, ,
, o Kisol,2 = 34.84 k/in
o Kisol,3 = 34.84 k/in
o Kisol,4 = 8.43 k/in

284



B1.7 Substructure Displacement at Each Support B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j, Example 2.3
for all supports:
, ,

, , (B-11) o dsub,1 = 0.002 in

o d sub,2 = 0.215 in
o d sub,3 = 0.215 in
o d sub,4 = 0.002 in
B1.8 Lateral Load in Each Substructure B1.8 Lateral Load in Each Substructure, Example
Calculate the lateral load in substructure j, Fsub,j, for 2.3
all supports:
, , , (B-12) , , ,

where values for Ksub,j are given in Step A1. o F sub,1 = 16.84 k
o F sub,2 = 62.18 k
o F sub,3 = 62.18 k
o F sub,4 = 16.84 k

B1.9 Column Shear Force at Each Support B1.9 Column Shear Force at Each Support,
Calculate the shear force in column k at support j, Example 2.3
Fcol,j,k, assuming equal distribution of shear for all
columns at support j:
,
, ,
#
,
, ,
# (B-13) o F col,2,1 = 62.18 k
o F col,3,1 = 62.18 k

Use these approximate column shear forces as a check These column shears are less than the plastic shear
on the validity of the chosen strength and stiffness capacity of each column (128k) as required in Step
characteristics. A3 and the chosen strength and stiffness values in
Step B1.1 are therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 2.3
damping ratio, , of the bridge:

Eq. 1907.58
7.1-5 2 (B-14) 2 2
GSID 386.4 79.02

and = 1.57 sec

Eq. 2 , , , and taking dy,j = 0:

7.1-10 (B-15)
, , ,
GSID 2 , , 0
0.30
, , ,
where dy,j is the yield displacement of the isolator. For
friction-based isolators, dy,j = 0. For other types of
isolators dy,j is usually small compared to disol,j and has
negligible effect on , Hence it is suggested that for
the Simplified Method, set dy,j = 0 for all isolator
types. See Step B2.2 where the value of dy,j is revisited

285

for the Multimode Spectral Analysis Method.

B1.11 Damping Factor B1.11 Damping Factor, Example 2.3

Calculate the damping factor, BL, and the Since 0.30 0.3
displacement, d, of the bridge:

Eq. .
, 0.3 1.70
.
7.1-3 (B-16)
1.7, 0.3
GSID and

9.79 9.79 0.2 1.57

Eq. 9.79 (B-17) 1.81
1.70
7.1-4
GSID
B1.12 Convergence Check B1.12 Convergence Check, Example 2.3
Compare the new displacement with the initial value Since the calculated value for displacement, d (=1.81)
assumed in Step B1.1. If there is close agreement, go is not close to that assumed at the beginning of the
to the next step; otherwise repeat the process from cycle (Step B1.1, d = 2.0), use the value of 1.81 as the
Step B1.3 with the new value for displacement as the new assumed displacement and repeat from Step
assumed displacement. B1.3.

This iterative process is amenable to solution using a
spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
After three iterations, convergence is reached at a
superstructure displacement of 1.65 in, with an
effective period of 1.43 seconds, and a damping factor
of 1.7 (30% damping ratio). The displacement in the
isolators at Pier 1 is 1.44 in and the effective stiffness
of the same isolators is 42.78 k/in.
See spreadsheet in Table B1.12-1 for results of final
iteration.
Ignoring the weight of the hammerhead, the column
shear force must equal the isolator shear force for
equilibrium. Hence column shear = 42.78 (1.44) =
61.60 k which is less than the maximum allowable
(128 k) if elastic behavior is to be achieved (as
required in Step A3).
Also the superstructure displacement = 1.65 in, which
is less than the available clearance of 2.5 in.

Therefore the above solution is acceptable and go to

Step B2.

Note that available clearance (2.5 in) is greater than

minimum required which is given by:

8 8 0.20 1.43
1.35
1.7

286

Table B1.12-1 Simplified Method Solution for Design Example 2.3 Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
1651.32 256.26 1907.58 0.2 3

StepB1.1 d 1.65 Assumeddisplacement

Qd 82.57 Characteristicstrength
Kd 50.04 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 168.48 8.424 5.105 10,000.00 0.001022 10.206 1.648 10.216 0.002 16.839 13.885 27.785
Pier1 657.18 32.859 19.915 288.87 0.148088 37.260 1.437 42.778 0.213 61.480 47.224 101.441
Pier2 657.18 32.859 19.915 288.87 0.148088 37.260 1.437 42.778 0.213 61.480 47.224 101.441
Abut2 168.48 8.424 5.105 10,000.00 0.001022 10.206 1.648 10.216 0.002 16.839 13.885 27.785
Total 1651.32 82.566 50.040 K eff,j 94.932 156.638 122.219 258.453
Step B1.4

StepB1.10 T eff 1.43 Effectiveperiod

0.30 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.71

B L 1.70 DampingFactor
d 1.65 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 2.808 1.702 3.405 1.69 3.363
Pier1 10.953 6.638 14.259 1.20 15.766
Pier2 10.953 6.638 14.259 1.20 15.766
Abut2 2.808 1.702 3.405 1.69 3.363

287

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and then
applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 2.3

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 (see
elastic stiffness, Kd,i, of each isolator i as follows: Table B1.12-1) by the number of isolators at each
, support (n = 3), the following values for Qd /isolator
, (B-19)
and Kd /isolator are obtained:
and o Qd, 1 = 8.42/3 = 2.81 k
, o Qd, 2 = 32.86/3=10.95 k
, ( B-20)
o Qd, 3 = 32.86/3 = 10.95 k
where values for Qd,j and Kd,j are obtained from the o Qd, 4 = 8.42/3 = 2.81 k
final cycle of iteration in the Simplified Method (Step
B1. 12 and
o Kd,1 = 5.10/3 = 1.70 k/in
o Kd,2 = 19.92/3 = 6.64 k/in
o Kd,3 = 19.92/3 = 6.64 k/in
o Kd,4 = 5.10/3 = 1.70 k/in

Note that the Kd values per support used above are

from the final iteration given in Table B1.12-1. These
are not the same as the initial values in Step B1.2,
because they have been adjusted from cycle to cycle,
such that the total Kd summed over all the isolators
satisfies the minimum lateral restoring force
requirement for the bridge, i.e. Kdtotal = 0.05 W/d. See
Step B1.1. Since d varies from cycle to cycle, Kd,j
varies from cycle to cycle.

B2.2 Initial Stiffness and Yield Displacement B2.2 Initial Stiffness and Yield Displacement,
Calculate the initial stiffness, Ku,i, and the yield Example 2.3
displacement, dy,i, for each isolator i as follows: Since the isolator type has been specified in Step A1
(1) For friction-based isolators Ku,i = and dy,i = 0. to be an elastomeric bearing (i.e. not a friction-based
(2) For other types of isolators, and in the absence of bearing), calculate Ku,i and dy,i for an isolator on Pier 1
isolator-specific information, take as follows:
, 10 , 10 6.64 66.4 /
, 10 , ( B-21)
and then and
, 10.95
,
( B-22) , 0.18
,
, , 66.4 6.64
, ,

As expected, the yield displacement is small

compared to the expected isolator displacement (~2

288

in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.

B2.3 Isolator Effective Stiffness, Kisol,i B2.3 Isolator Effective Stiffness, Kisol,i, Example 2.3
Calculate the isolator stiffness, Kisol,i, of each isolator Dividing the results for Kisol (Step B1.12) among the 3
i: isolators at each support, the following values for Kisol
/isolator are obtained:
, o Kisol,1 = 10.22/3 = 3.41 k/in
, (B -23)
o Kisol,2 = 42.78/3 = 14.26 k/in
o Kisol,3 = 42.78/3 = 14.26 k/in
o Kisol,4 = 10.22/3 = 3.41 k/in

B2.4 Three-Dimensional Bridge Model B2.4 Three-Dimensional Bridge Model, Example

Using computer-based structural analysis software, 2.3
create a 3-dimensional model of the bridge with the Although the bridge in this Design Example is regular
isolators represented by spring elements. The stiffness and is without skew or curvature, a 3-dimensional
of each isolator element in the horizontal axes (Kx and finite element model was developed for this Step, as
Ky in global coordinates, K2 and K3 in typical local shown below.
coordinates) is the Kisol value calculated in the
previous step. For bridges with regular geometry and
minimal skew or curvature, the superstructure may be .
represented by a single stick provided the load path
to each individual isolator at each support is explicitly
modeled, usually by a rigid cap beam and a set of rigid
links. If the geometry is irregular, or if the bridge is
skewed or curved, a finite element model is
recommended to accurately capture the load carried
by each individual isolator. If the piers have an
unusual weight distribution, such as a pier with a
hammerhead cap beam, a more rigorous model is
recommended.

B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in Step
A2 is for a 5% damped system. It is modified in this
step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the effective
period of the bridge, Teff, by the damping factor, BL.
B.2.5 Composite Design Response Spectrum,
Example 2.3
From the final results of Simplified Method (Step
B1.12), BL = 1.70 and Teff = 1.43 sec. Hence the
transition in the composite spectrum from 5% to 30%
damping occurs at 0.8 Teff = 0.8 (1.43) = 1.14 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.14 sec by 1.70.
0.8

0.7

0.6

0.5
Csm (g)

0.4

0.3

0.2

0.1

0
0 0.5 1 1.5 2 2.5 3 3.5 4
T (sec)

289

B2.6 Multimodal Analysis of Finite Element Model B2.6 Multimodal Analysis of Finite Element
Input the composite response spectrum as a user- Model, Example 2.3
specified spectrum in the software, and define a load Results of modal analysis of the example bridge are
case in which the spectrum is applied in the summarized in Table B2.6-1 Here the modal periods
longitudinal direction. Analyze the bridge for this and mass participation factors of the first 12 modes
load case. are given. The first three modes are the principal
transverse, longitudinal, and torsion modes with
periods of 1.60, 1.46 and 1.39 sec respectively. The
period of the longitudinal mode (1.46 sec) is very
close to that calculated in the Simplified Method. The
mass participation factors indicate there is no
coupling between these three modes (probably due to
the symmetric nature of the bridge) and the high
values for the first and second modes (92% and 94%
respectively) indicate the bridge is responding
essentially in a single mode of vibration in each
Mode Period Mass Participation Ratios
No Sec UX UY UZ RX RY RZ
1 1.604 0.000 0.919 0.000 0.952 0.000 0.697
2 1.463 0.941 0.000 0.000 0.000 0.020 0.000
3 1.394 0.000 0.000 0.000 0.000 0.000 0.231
4 0.479 0.000 0.003 0.000 0.013 0.000 0.002
5 0.372 0.000 0.000 0.076 0.000 0.057 0.000
6 0.346 0.000 0.000 0.000 0.000 0.000 0.000
7 0.345 0.000 0.001 0.000 0.010 0.000 0.000
8 0.279 0.000 0.003 0.000 0.013 0.000 0.002
9 0.268 0.000 0.000 0.000 0.000 0.000 0.000
10 0.267 0.058 0.000 0.000 0.000 0.000 0.000
11 0.208 0.000 0.000 0.000 0.000 0.129 0.000
12 0.188 0.000 0.000 0.000 0.000 0.000 0.001

direction. Similar results to that obtained by the

Simplified Method are therefore expected.

Table B2.6-1 Modal Properties of Bridge

Example 2.3- First Iteration
Computed values for the isolator displacements due to
a longitudinal earthquake are as follows (numbers in
parentheses are those used to calculate the initial
properties to start iteration from the Simplified
Method):

o disol,1 = 1.69 (1.65) in

o disol,2 = 1.20 (1.44) in
o disol,3 = 1.20 (1.44) in
o disol,4 = 1.69 (1.65) in

B2.7 Convergence Check B2.7 Convergence Check, Example 2.3

Compare the resulting displacements at the
superstructure level (d) to the assumed displacements.
These displacements can be obtained by examining
the joints at the top of the isolator spring elements. If
in close agreement, go to Step B2.9. Otherwise go to
Step B2.8.
The results for isolator displacements are close but
not close enough (15% difference at the piers)
Go to Step B2.8 and update properties for a second
cycle of iteration.

290

B2.8 Update Kisol,i, Keff,j, and BL B2.8 Update Kisol,i, Keff,j, and BL, Example 2.3
Use the calculated displacements in each isolator Updated values for Kisol,i are given below (previous
element to obtain new values of Kisol,i for each isolator values are in parentheses):
as follows:
, o Kisol,1 = 3.36 (3.41) k/in
, , (B-24) o Kisol,2 = 15.77 (14.26) k/in
,
o Kisol,3 = 15.77 (14.26) k/in
Recalculate Keff,j : o Kisol,4 = 3.36 (3.41) k/in

Eq. , ,
7.1-6 ,
(B-25) Since the isolator displacements are relatively close to
, ,
GSID previous results no significant change in the damping
ratio is expected. Hence Keff,j and are not
Recalculate system damping ratio, : recalculated and BL is taken at 1.70.

Eq. 2 , , ,
7.1-10 , , ,
(B-26)
GSID

Recalculate system damping factor, BL:

Eq. .
0.3
.
7.1-3 ( B-27)
1.7 0.3 Since the change in effective period is very small
GSID
(1.43 to 1.46 sec) and no change has been made to BL,
Obtain the effective period of the bridge from the there is no need to construct a new composite
multi-modal analysis and with the revised damping response spectrum in this case. Go back to Step B2.6
factor (Eq. B-27), construct a new composite response (see immediately below).
spectrum. Go to Step B2.6.
B2.6 Multimodal Analysis Second Iteration,
Example 2.3
Reanalysis gives the following values for the isolator
displacements (numbers in parentheses are those
from the previous cycle):

o disol,1 = 1.66 (1.69) in

o disol,2 = 1.15 (1.20) in
o disol,3 = 1.15 (1.20) in
o disol,4 = 1.66 (1.69) in

Go to Step B2.7

B2.7 Convergence Check B2.7 Convergence Check, Example 2.3

Compare results and determine if convergence has Satisfactory agreement has been reached on this
been reached. If so go to Step B2.9. Otherwise Go to second cycle. Go to Step B2.9
Step B2.8.

B2.9 Superstructure and Isolator Displacements B2.9 Superstructure and Isolator Displacements,
Once convergence has been reached, obtain Example 2.3
o superstructure displacements in the longitudinal From the above analysis:
(xL) and transverse (yL) directions of the bridge, o superstructure displacements in the
and longitudinal (xL) and transverse (yL) directions
o isolator displacements in the longitudinal (uL) are:
and transverse (vL) directions of the bridge, for xL= 1.69 in

291

each isolator, for this load case (i.e. yL= 0.0 in

longitudinal loading). These displacements
may be found by subtracting the nodal
displacements at each end of each isolator
spring element.
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions are:
o Abutments: uL = 1.66 in, vL = 0.00 in
o Piers: uL = 1.15 in, vL = 0.00 in

All isolators at same support have the same

displacements.

B2.10 Pier Bending Moments and Shear Forces B2.10 Pier Bending Moments and Shear Forces,
Obtain the pier bending moments and shear forces in Example 2.3
the longitudinal (MPLL, VPLL) and transverse (MPTL, Bending moments in single column pier in the
VPTL) directions at the critical locations for the longitudinal (MPLL) and transverse (MPTL) directions
longitudinally-applied seismic loading. are:
MPLL= 0
MPTL= 1602 kft

Shear forces in single column pier the longitudinal

(VPLL) and transverse (VPTL) directions are
VPLL=67.16 k
VPTL=0

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 2.3
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1

Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Sub- Long. Transv. Axial
Isol-
struc shear due shear due forces due
ator
ture to long. to long. to long.
EQ EQ EQ
1 5.63 0 1.29
Abut
2 5.63 0 1.30
ment
3 5.63 0 1.29
1 18.19 0 0.77
Pier 2 18.25 0 1.11
3 18.19 0 0.77

The difference between the longitudinal shear force in

the column (VPLL = 67.16k) and the sum of the
isolator shear forces at the same Pier (54.63 k) is
about 12.5 k. This is due to the inertia force
developed in the hammerhead cap beam which
weighs about 128 k and can generate significant
additional demand on the column (about a 23%
increase in this case).

292

STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake
Repeat the above process, starting at Step B1, for
earthquake loading in the transverse direction of the
bridge. Support flexibility in the transverse direction
is to be included, and a composite response spectrum
is to be applied in the transverse direction. Obtain
isolator displacements in the longitudinal (uT) and
transverse (vT) directions of the bridge, and the biaxial
bending moments and shear forces at critical locations
in the columns due to the transversely-applied seismic
loading.
C1. Analysis for Transverse Earthquake, Example
2.3
Key results from repeating Steps B1 and B2
(Simplified and Mulitmode Spectral Methods) are:
o Teff = 1.52 sec
o Superstructure displacements in the longitudinal
(xT) and transverse (yT) directions are as follows:
xT = 0 and yT = 1.75 in
o Isolator displacements in the longitudinal (uT)
and transverse (vT) directions as follows:
Abutments uT = 0.00 in, vT = 1.75 in
Piers uT = 0.00 in, vT = 0.71 in
o Pier bending moments in the longitudinal (MPLT)
and transverse (MPTT) directions are as follows:
MPLT = 1548.33 kft and MPTT = 0
o Pier shear forces in the longitudinal (VPLT) and
transverse (VPTT) directions are as follows:
VPLT = 0 and VPTT = 60.75 k
o Isolator shear and axial forces are in Table C1-1.

Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Long. Transv. Axial
Sub-
Isol- shear d shear due forces due
struct
ator ue to to transv. to transv.
ure
transv. EQ EQ
EQ
1 0.0 5.82 13.51
Abut
2 0.0 5.83 0
ment
3 0.0 5.82 13.51
1 0.0 15.40 26.40
Pier 2 0.0 15.57 0
3 0.0 15.40 26.40

The difference between the transverse shear force in

the column (VPLL = 60.75k) and the sum of the
isolator shear forces at the same Pier (46.37 k) is
about 14.4 k. This is due to the inertia force
developed in the hammerhead cap beam which
weighs about 128 k and can generate significant
additional demand on the column (about 31% ).

293

STEP D. CALCULATE DESIGN VALUES

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements at Pier 1,

Following the provisions in Art. 2.1 GSID, and Example 2.3
illustrated in Fig. 2.1-1 GSID, calculate the total To illustrate the process, design displacements for the
design displacement, dt, for each isolator by outside isolator on Pier 1 are calculated below.
combining the displacements from the longitudinal (uL
and vL) and transverse (uT and vT) cases as follows: Load Case 1:
u1 = uL + 0.3uT = 1.0(1.15) + 0.3(0) = 1.15 in
u1 = uL + 0.3uT (D-1) v1 = vL + 0.3vT = 1.0(0) + 0.3(0.71) = 0.21 in
v1 = vL + 0.3vT (D-2) R1 = = 1.15 0.21 = 1.17 in
R1 = (D-3)
Load Case 2:
u2 = 0.3uL + uT (D-4) u2 = 0.3uL + uT = 0.3(1.15) + 1.0(0) = 0.35 in
v2 = 0.3vL + vT (D-5) v2 = 0.3vL + vT = 0.3(0) + 1.0(0.71) = 0.71in
R2 = (D-6) R2 = = 0.35 0.71 = 0.79 in

dt = max(R1, R2) (D-7)

Governing Case:
Total design displacement, dt = max(R1, R2)
= 1.17 in

D2. Design Moments and Shears D2. Design Moments and Shears in Pier 1,
Calculate design values for column bending moments Example 2.3
and shear forces for all piers using the same Design moments and shear forces are calculated for
combination rules as for displacements. Pier 1 below, to illustrate the process.

Alternatively this step may be deferred because the Load Case 1:

above results may not be final. Upper and lower VPL1= VPLL + 0.3VPLT = 1.0(67.16) + 0.3(0) = 67.16 k
bound analyses are required after the isolators have VPT1= VPTL + 0.3VPTT = 1.0(0) + 0.3(60.75) = 18.23 k
been designed as described in Art 7. GSID. These R1 = = 67.16 18.23 = 69.59 k
analyses are required to determine the effect of
possible variations in isolator properties due age, Load Case 2:
temperature and scragging in elastomeric systems. VPL2= 0.3VPLL + VPLT = 0.3(67.16) + 1.0(0) = 20.15 k
Accordingly the results for column shear in Steps VPT2= 0.3VPTL + VPTT = 0.3(0) + 1.0(60.75) = 60.75 k
B2.10 and C are likely to increase once these analyses
R2 = = 20.15 60.75 = 64.00 k
are complete.
Governing Case:
Design column shear = max (R1, R2)
= 69.59 k

294

STEP E. DESIGN OF SPHERICAL FRICTION ISOLATORS

The Spherical Friction Bearing (SFB) isolator has
an articulated slider to permit rotation, and a POLISHED STAINLESS STEEL SURFACE

spherical sliding interface. It has lateral stiffness

due to the curvature of this interface. These
isolators are capable of carrying very large axial
loads and can be designed to have long periods of SEAL
vibration (5 seconds or longer).

R
The main components of an SFB isolator are a STAINLESS STEEL
stainless steel concave spherical plate, an ARTICULATED SLIDER COMPOSITE LINER MATERIAL
(ROTATIONAL PART)
articulated slider and a housing plate as illustrated
in figure above. In this figure, the concave spherical plate is facing down. The bearings may also be installed with
this surface facing up as in the figure below. The side of the articulated slider in contact with the concave
spherical surface is coated with a low-friction composite material, usually PTFE. The other side of the slider is
also spherical but lined with stainless steel and sits in a spherical cavity coated with PTFE.

Spherical friction bearings are described by the same equation

of motion as conventional pendulums. As a consequence their
period of vibration is directly proportional to the radius of
curvature of the concave surface. See figure at right. Long
period shifts are therefore possible with surfaces that have large
radii of curvature. Friction between the articulated slider and the R
concave surface dissipates energy and the weight of the bridge
W
acts as a restoring force, due to the curvature of the sliding
surface. Restoringforce

The required values for Qd and Kd determine the coefficient of Friction

friction at the sliding interface and the radius of curvature. D

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

E1. Required Properties E1. Required Properties, Example 2.3

Obtain from previous work the properties required of The design of one of the pier isolators is given below
the isolation system to achieve the specified to illustrate the design process for spherical friction
performance criteria (Step A1). isolators.
required characteristic strength, Qd / isolator
required post-elastic stiffness, Kd / isolator From previous work
the total design displacement, dt, for each Qd / isolator = 2.34 k
isolator Kd / isolator = 1.17 k/in
maximum applied dead load, PDL Total design displacement, dt = 1.55 in
maximum live load, PLL and PDL = 45.52 k
maximum wind load, PWL PLL = 15.50 k
PWL = 1.76 k < Qd OK

E2. Isolator Dimensions
E2.1 Radius of Curvature E2.1 Radius of Curvature, Example 2.3
Determine the required radius of curvature, R, using:

187
(E-1) 27.66 27.75
6.76

295

E2.2 Coefficient of Friction E2.2 Coefficient of Friction, Example 2.3

Determine the required coefficient of friction, ,
using:

10.95
(E-2) 0.0585 5.85%
187

E2.3 Material Selection E2.3 Material Selection, Example 2.3

Based on the required coefficient of friction select an Select 25GF Teflon and size disc to achieve required
appropriate PTFE compound and contact pressure, c, contact pressure of 6,500 psi (Step E2.4).
from Table E2.3-1.

Table E2.3-1 Material Properties

PTFE
Compound Contact
(Filled and Pressure, c (%)
Unfilled (psi)
Teflon)
1,000 11.93
Unfilled 2,000 8.70
(UF) 3,000 7.03
6,500 5.72
Glass-filled 1,000 14.61
15% by 2,000 10.08
weight 3,000 8.49
(15GF) 6,500 5.27
Glass-filled 1,000 13.20
25% by 2,000 11.20
weight 3,000 9.60
(25GF) 6,500 5.89

E2.4 Disk Diameter E2.4 Disk Diameter, Example 2.3

Determine the required contact area, Ac, and disk
diameter, d, using:

187
(E-3) 28.77
6.5
and
4
(E-4) 4 7.00
6.05 6.00

E2.5 Isolator Diameter E2.5 Isolator Diameter, Example 2.3

Determine the required diameter of the concave As the bridge is in Seismic Zone 2,
surface, Lchord, and overall isolator width, B, using: = 2(dt) = 2(1.17) = 2.34 in

2 (E-5) 2 2.34 3.0 10.68

2

296

and Select s = 1.5 in:

2 (E-6)
10.68 2 1.5 13.68 13.75
where:
= 2 x total design displacement, dt, if in Seismic
Zone 1 or 2, or
1.5 x total design displacement, dt, if in Seismic
Zone 3 or 4.
s = width of shoulder of concave plate.

E2.6 Isolator Height E2.6 Isolator Height, Example 2.3

E2.6.1 Rise E2.6.1 Rise, Example 2.3

Determine the rise of the concave surface, h, using:

10.68
(E-7) 0.52
8 8 27.66

E2.6.2 Throat Thickness E2.6.2 Throat Thickness, Example 2.3

Determine the required throat thickness, t, based on Assume safe bearing stress below isolator:
the minimum required bearing area, Ab, such that the
maximum allowable bearing stress, bearing, is not bearing = 2.0 ksi.
exceeded on either the sole plate above or the
masonry plate below, depending on whether the
isolator is installed with concave surface facing up or
down.
187 123
(E-8) 155.0
2.0

4 4 155
14.05
(E-9)

0.5 (E-10) 0.5 14.05 6.0 4.02 4.0

This assumes a 45 distribution of compressive stress

through the throat to the support plates.

E2.6.3 Total Height E2.6.3 Total Height, Example 2.3

Determine the thickness of concave plate, T1, using:

(E-11) 0.52 4.0 4.52 4.50

Thickness of slider plate (T2) will vary with detail for

socket that holds articulated slider and rotation
requirement. Check with manufacturer for value. For
estimating purposes take T2 = T1. 4.50

Then total height of isolator:

(E-12) 4.50 4.50 9.00

E3. Design Summary E3. Design Summary, Example 2.3

Overall diameter = 13.75 in
Overall height = 9.0 in (est.)

297

Radius concave surface = 27.75 in

PTFE is 25% GF; contact pressure = 6,500 psi
Diameter PTFE sliding disc = 6.00 in
E4. Minimum and Maximum Performance Check E4. Minimum and Maximum Performance Check,
Art. 8 GSID requires the performance of any Example 2.3
isolation system be checked using minimum and For spherical friction isolators, property modification
maximum values for the effective stiffness of the factors are applied to Qd only.
system. These values are calculated from minimum
and maximum values of Kd and Qd, which are found Minimum Property Modification factors are:
using system property modification factors, as min = 1.0
indicated in Table E4-1.
which means there is no need to reanalyze the bridge
Table E4-1. Minimum and maximum values with a set of minimum values.
for Kd and Qd.
Maximum Property Modification factors are (GSID
Eq. Appendix A.1):
8.1.2-1 Kd,max = Kd max,Kd (E-13)
GSID max,a = 1.1
Eq. max,c = 1.0
8.1.2-2 Kd,min = Kd min,Kd (E-14) max,tr = 1.2
GSID max,t = 1.2
Eq.
8.1.2-3 Qd,max = Qd max,Qd (E-15)
GSID
Eq.
8.1.2-4 Qd,min = Qd min,Qd (E-16)
GSID

Determination of the system property modification

factors should include consideration of the effects of
temperature, aging, scragging, velocity, travel (wear)
and contamination as shown in Table E4-2. In lieu of
tests, numerical values for these factors can be
obtained from Appendix A, GSID.

Table E4-2. Minimum and maximum values for

system property modification factors.

Eq. min,Kd = (min,t,Kd) (min,a,Kd)

8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-21)
GSID (min,scrag,Kd)
Eq. max,Kd = (max,t,Kd) (max,a,Kd)
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-18)
GSID (max,scrag,Kd)
Eq. min,Qd = (min,t,Qd) (min,a,Qd)
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-19)
GSID (min,scrag,Qd)
Eq. max,Qd = (max,t,Qd) (max,a,Qd)
8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd) (E-20)
GSID (max,scrag,Qd)

Adjustment factors are applied to individual -factors Applying a system adjustment factor of 0.66 for an
(except v) to account for the likelihood of other bridge, the maximum property modification
occurrence of all of the maxima (or all of the factors become:
minima) at the same time. These factors are applied max,a = 1.0 + 0.1(0.66) = 1.066

298

to all -factors that deviate from unity but only to the
portion of the -factor that is greater than, or less
than, unity. Art. 8.2.2 GSID gives these factors as
follows:
1.00 for critical bridges
0.75 for essential bridges
0.66 for all other bridges
As required in Art. 7 GSID and shown in Fig. C7-1
GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Qd,max. As indicated in Fig C7-1 GSID, maximum
displacements will probably be given by the first case
(Kd,min and Qd,min) and maximum forces by the second
case (Kd,max and Qd,max).
max,c = 1.0
max,t = 1.0 + 0.2(0.66) = 1.132
max,a = 1.0 + 0.2(0.66) = 1.132
Therefore the maximum overall modification factors
max = 1.066(1.0)(1.132)(1.132) = 1.37
Since the possible variation in upper bound properties
exceeds 15% a reanalysis of the bridge is required to
determine performance with these properties.
The upper-bound properties are:
Qd,max = 1.37 (10.95) = 15.0 k
Kdmax = Kd = 6.76 k/in

E5. Design and Performance Summary, Example

E5. Design and Performance Summary
2.3
E5.1 Isolator dimensions E5.1 Isolator dimensions, Example 2.3
Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E5.1-1.
Overall diameter of isolator
Overall height Table E5.1-1 Isolator Dimensions
Radius of curvature of concave plate
Isolator Overall size Overall size Radius
Diameter of PTFE disc Location including without (in)
PTFE Compound mounting mounting
PTFE contact pressure plates (in) plates (in)
Under
Check all dimensions with manufacturer. edge 17.75x17.75 13.75 dia.
27.75
girder x 9.0(H) x 7.0(H)
on Pier 1

PTFE is 25% Glass-filled; 6,500 psi contact pressure;

disc diameter is 6.00 in.

E5.2 Bridge Performance E5.2 Bridge Performance, Example 2.3

Summarize bridge performance
Maximum superstructure displacement
(longitudinal)
Maximum superstructure displacement
(transverse)
Maximum superstructure displacement
(resultant)
Maximum column shear (resultant)
Maximum column moment (about transverse
axis)
Maximum column moment (about
longitudinal axis)
Maximum column torque
Bridge performance is summarized in Table E5.2-1
where it is seen that the maximum column shear is
71.74k. This less than the column plastic shear (128k)
and therefore the required performance criterion is
satisfied (fully elastic behavior). Furthermore the
maximum longitudinal displacement is 1.69 in which
is less than the 2.5in available at the abutment
expansion joints and is therefore acceptable.

Check required performance as determined in Step

A3, is satisfied.

299

Table E5.2-1 Summary of Bridge Performance

Maximum superstructure
1.69 in
displacement (longitudinal)
Maximum superstructure
1.75 in
displacement (transverse)
Maximum superstructure
1.82 in
displacement (resultant)
Maximum column shear
71.74 k
(resultant)
Maximum column moment
1,657 kft
about transverse axis
Maximum column moment
1,676 kft
about longitudinal axis
Maximum column torque 21.44 kft

300

SECTION 2
Steel Plate Girder Bridge, Long Spans, Single-Column Pier

DESIGN EXAMPLE 2.4: Eradiquake Isolators

Design Examples in Section 2

Site
ID Description S1 Column height Skew Isolator type
Class
2.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
2.1 Change site class 0.2g D Same 0 Lead rubber bearing
2.2 Change spectral acceleration, S1 0.6g B Same 0 Lead rubber bearing
Spherical friction
2.3 Change isolator to SFB 0.2g B Same 0
bearing
2.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
2.5 Change column height 0.2g B H1=0.5H2 0 Lead rubber bearing
0
2.6 Change angle of skew 0.2g B Same 45 Lead rubber bearing

301

DESIGN PROCEDURE DESIGN EXAMPLE 2.4 (EQS Isolators)

STEP A: BRIDGE AND SITE DATA

A1. Bridge Properties
Determine properties of the bridge:
number of supports, m
number of girders per support, n
angle of skew
weight of superstructure including railings,
curbs, barriers and other permanent loads,
WSS
weight of piers participating with
superstructure in dynamic response, WPP
weight of superstructure, Wj, at each
support
stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge. The calculation of these
quantities requires careful consideration of
several factors such as the use of cracked
sections when estimating column or wall
flexural stiffness, foundation flexibility,
and effective column height.
column shear strength (minimum value).
This will usually be derived from the
minimum value of the column flexural
yield strength, the column height, and
whether the column is acting in single or
double curvature in the direction under
consideration.
allowable movement at expansion joints
isolator type if known, otherwise to be
selected
A1. Bridge Properties, Example 2.4
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 3
Angle of skew = 00
Number of columns per support = 1
Weight of superstructure including permanent
loads, WSS = 1651.32 k
Weight of superstructure at each support:
o W1 = 168.48 k
o W2 = 657.18 k
o W3 = 657.18 k
o W4 = 168.48 k
Participating weight of piers, WPP = 256.26 k
Effective weight (for calculation of period),
Weff = Wss + WPP = 1907.58 k
o Stiffness of each pier in the both directions
Ksub,pier1 = 288.87 k/in
o Ksub,pier2 = 288.87 k/in
Minimum column shear strength based on
flexural yield capacity of column = 128 k
Displacement capacity of expansion joints
(longitudinal) = 2.5 in for thermal and other
movements
Eradiquake isolators

A2. Seismic Hazard A2. Seismic Hazard, Example 2.4

Determine seismic hazard at site: Acceleration coefficients for bridge site are given in
acceleration coefficients design statement as follows:
site class and site factors PGA = 0.40
seismic zone S1 = 0.20
Plot response spectrum. SS = 0.75

Use Art. 3.1 GSID to obtain peak ground and spectral Bridge is on a rock site with shear wave velocity in
acceleration coefficients. These coefficients are the upper 100 ft of soil = 3,000 ft/sec.
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD Table 3.10.3.1-1 LRFD gives Site Class as B.
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (e.g. Figures Tables 3.10.3.2-1, -2 and -3 LRFD give following
3.10.2.1-1 to 3.10.2.1-21 LRFD). Site Factors:
Fpga = 1.0
Use Art. 3.2 to obtain Site Class and corresponding Site Fa = 1.0
Factors (Fpga, Fa and Fv). These data are the same as for Fv = 1.0
conventional bridges and Art 3.2 refers the designer to

302

the corresponding articles in the LRFD Specifications,

i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3,
LRFD.

Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give

modified spectral acceleration coefficients that include
site effects as follows:
As = Fpga PGA As = Fpga PGA = 1.0(0.40) = 0.40
SDS = Fa SS SDS = Fa SS = 1.0(0.75) = 0.75
SD1 = Fv S1 SD1 = Fv S1 = 1.0(0.20) = 0.20

Seismic Zone is determined by value of SD1 in Since 0.15 < SD1 < 0.30, bridge is located in Seismic
accordance with provisions in Table5-1 GSID. Zone 2.

These coefficients are used to plot design response Design Response Spectrum is as below:
spectrum as shown in Fig. 4-1 GSID.
0.9

0.8

0.7

0.6
Acceleration 0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Period(s)

A3. Performance Requirements A3. Performance Requirements, Example 2.4

Determine required performance of isolated bridge In this example, assume the owner has specified full
during Design Earthquake (1000-yr return period). functionality following the earthquake and therefore
the columns must remain elastic (no yield).
Examples of performance that might be specified by
the Owner include: To remain elastic the maximum lateral load on the
Reduced displacement ductility demand in pier must be less than the load to yield any one
columns, so that bridge is open for emergency column (128 k).
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in The maximum shear in the column must therefore be
columns or yield elsewhere in bridge), so that less than 128 k in order to keep the column elastic
bridge is fully functional and open to all vehicles and meet the required performance criterion.
immediately following earthquake.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.

303

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for

Isolator Force, F
non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most
have bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use Ku
equivalent linear springs and viscous damping to
represent the isolators, so that linear methods of analysis
may be used to determine response. Since equivalent dy disol
Isolator
Ku Ku
properties such as Kisol are dependent on displacement Displacement, d

(d), and the displacements are not known at the beginning

of the analysis, an iterative approach is required. Note Kd
that in Art 7.1, GSID, keff is used for the effective stiffness
of an isolator unit and Keff is used for the effective disol = Isolator displacement
stiffness of a combined isolator and substructure unit. To dy = Isolator yield displacement
Fisol = Isolator shear force
minimize confusion, Kisol is used in this document in Fy = Isolator yield force
place of keff. There is no change in the use of Keff and Keff,j, Kd
Kisol
= Post-elastic stiffness of isolator
= Effective stiffness of isolator
but Ksub is used in place of ksub. Ku = Loading and unloading stiffness (elastic stiffness)
Qd = Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement (disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 2.4
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.20 2.0
GSID

304

(2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not
occur under non-seismic loads (e.g. wind) but
low enough that yield will occur during an
earthquake. Experience has shown that
taking Qd to be 5% of the bridge weight is a
good starting point, i.e.
0.05 (B-2) 0.05 0.05 1651.32 82.56

(3) Post-yield stiffness, Kd

Art 12.2 GSID requires that all isolators
exhibit a minimum lateral restoring force at
the design displacement, which translates to a
minimum post yield stiffness Kd,min given by
Eq. B-3.

Art. 1651.32
0.025 0.05 0.05 41.28 /
12.2 , (B-3) 2.0
GSID
Experience has shown that a good starting
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 2.4
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength, ,

Qd, and stiffness, Kd, values in proportion to the dead o Qd, 1 = 8.42 k
load applied at that support: o Qd, 2 = 32.86 k
o Qd, 3 = 32.86 k
o Qd, 4 = 8.42 k

, (B-4)
and
and
,
, (B-5) o Kd,1 = 4.21 k/in
o Kd,2 = 16.43 k/in
o Kd,3 = 16.43 k/in
o Kd,4 = 4.21 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 2.4
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of , ,
the isolators at support j (Kisol,j) and the stiffness of , ,
the substructure Ksub,j. See figure below for
definitions (after Fig. 7.1-1 GSID). o 1 = 8.43x10-4
o 2 = 1.21x10-1
An expression for Keff,j, is given in Eq.7.1-6 GSID, but o 3 = 1.21x10-1
a more useful formula is as follows (MCEER 2006): o 4 = 8.43x10-4
,
, (B-6)
1 ,
where ,
1
, ,
(B-7) o Keff,1 = 8.42 k/in
, , o Keff,2 = 31.09 k/in

305

o Keff,3 = 31.09 k/in

and Ksub,j for the piers are given in Step A1. For the o Keff,4 = 8.42 k/in
abutments, take Ksub,j to be a large number, say 10,000
k/in, unless actual stiffness values are available. Note
that if the default option is chosen, unrealistically high
F
Kd
Qd Kisol

dy disol
Superstructure
Isolator Effective Stiffness, Kisol

F
Isolator(s), Kisol
Ksub

Substructure, Ksub
dsub

Substructure Stiffness, Ksub

dsub disol F

d
Keff

d = disol + dsub

Combined Effective Stiffness, Keff

values for Ksub,j will give unconservative results for

column moments and shear forces.

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 2.4
Calculate the total effective stiffness, Keff, of the
bridge:

Eq. , 79.02 /
7.1-6 , ( B-8)
GSID

B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 2.4
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1
, (B-9)
1 o disol,1 = 2.00 in
o disol,2 = 1.79 in
o disol,3 = 1.79 in
o disol,4 = 2.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the stiffness of the isolation system at Example 2.4
support j, Kisol,j, for all supports:
,
, ,
,
,
(B-10) o Kisol,1 = 8.43 k/in
, ,
, o Kisol,2 = 34.84 k/in

306



o Kisol,3 = 34.84 k/in

o Kisol,4 = 8.43 k/in
B1.7 Substructure Displacement at Each Support B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j, Example 2.4
for all supports:
, ,

, , (B-11) o dsub,1 = 0.002 in

o d sub,2 = 0.215 in
o d sub,3 = 0.215 in
o d sub,4 = 0.002 in
B1.8 Lateral Load in Each Substructure B1.8 Lateral Load in Each Substructure, Example
Calculate the lateral load in substructure j, Fsub,j, for 2.4
all supports:
, , , (B-12) , , ,

where values for Ksub,j are given in Step A1. o F sub,1 = 16.84 k
o F sub,2 = 62.18 k
o F sub,3 = 62.18 k
o F sub,4 = 16.84 k

B1.9 Column Shear Force at Each Support B1.9 Column Shear Force at Each Support,
Calculate the shear force in column k at support j, Example 2.4
Fcol,j,k, assuming equal distribution of shear for all
columns at support j:
,
, ,
#
,
, ,
# (B-13) o F col,2,1 = 62.18 k
o F col,3,1 = 62.18 k

Use these approximate column shear forces as a check These column shears are less than the plastic shear
on the validity of the chosen strength and stiffness capacity of each column (128k) as required in Step
characteristics. A3 and the chosen strength and stiffness values in
Step B1.1 are therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 2.4
damping ratio, , of the bridge:

Eq. 1907.58
7.1-5 2 (B-14) 2 2
GSID 386.4 79.02

and = 1.57 sec

Eq. 2 , , , and taking dy,j = 0:

7.1-10 (B-15)
, , ,
GSID 2 , , 0
0.30
, , ,
where dy,j is the yield displacement of the isolator. For
friction-based isolators, dy,j = 0. For other types of
isolators dy,j is usually small compared to disol,j and has
negligible effect on , Hence it is suggested that for

307

the Simplified Method, set dy,j = 0 for all isolator

types. See Step B2.2 where the value of dy,j is revisited
for the Multimode Spectral Analysis Method.

B1.11 Damping Factor B1.11 Damping Factor, Example 2.4

Calculate the damping factor, BL, and the Since 0.30 0.3
displacement, d, of the bridge:

Eq. .
, 0.3 1.70
.
7.1-3 (B-16)
1.7, 0.3
GSID and

9.79 9.79 0.2 1.57

Eq. 9.79 (B-17) 1.81
1.70
7.1-4
GSID
B1.12 Convergence Check B1.12 Convergence Check, Example 2.4
Compare the new displacement with the initial value Since the calculated value for displacement, d (=1.81)
assumed in Step B1.1. If there is close agreement, go is not close to that assumed at the beginning of the
to the next step; otherwise repeat the process from cycle (Step B1.1, d = 2.0), use the value of 1.81 as the
Step B1.3 with the new value for displacement as the new assumed displacement and repeat from Step
assumed displacement. B1.3.

This iterative process is amenable to solution using a
spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
After three iterations, convergence is reached at a
superstructure displacement of 1.65 in, with an
effective period of 1.43 seconds, and a damping factor
of 1.7 (30% damping ratio). The displacement in the
isolators at Pier 1 is 1.44 in and the effective stiffness
of the same isolators is 42.78 k/in.
See spreadsheet in Table B1.12-1 for results of final
iteration.
Ignoring the weight of the hammerhead, the column
shear force must equal the isolator shear force for
equilibrium. Hence column shear = 42.78 (1.44) =
61.60 k which is less than the maximum allowable
(128 k) if elastic behavior is to be achieved (as
required in Step A3).
Also the superstructure displacement = 1.65 in, which
is less than the available clearance of 2.5 in.

Therefore the above solution is acceptable and go to

Step B2.

Note that available clearance (2.5 in) is greater than

minimum required which is given by:

8 8 0.20 1.43
1.35
1.7

308

Table B1.12-1 Simplified Method Solution for Design Example 2.4 Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
1651.32 256.26 1907.58 0.2 3

StepB1.1 d 1.65 Assumeddisplacement

Qd 82.57 Characteristicstrength
Kd 50.04 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 168.48 8.424 5.105 10,000.00 0.001022 10.206 1.648 10.216 0.002 16.839 13.885 27.785
Pier1 657.18 32.859 19.915 288.87 0.148088 37.260 1.437 42.778 0.213 61.480 47.224 101.441
Pier2 657.18 32.859 19.915 288.87 0.148088 37.260 1.437 42.778 0.213 61.480 47.224 101.441
Abut2 168.48 8.424 5.105 10,000.00 0.001022 10.206 1.648 10.216 0.002 16.839 13.885 27.785
Total 1651.32 82.566 50.040 K eff,j 94.932 156.638 122.219 258.453
Step B1.4

StepB1.10 T eff 1.43 Effectiveperiod

0.30 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.71

B L 1.70 DampingFactor
d 1.65 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 2.808 1.702 3.405 1.69 3.363
Pier1 10.953 6.638 14.259 1.20 15.766
Pier2 10.953 6.638 14.259 1.20 15.766
Abut2 2.808 1.702 3.405 1.69 3.363

309

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and then
applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 2.4

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 (see
elastic stiffness, Kd,i, of each isolator i as follows: Table B1.12-1) by the number of isolators at each
, support (n = 3), the following values for Qd /isolator
, (B-19)
and Kd /isolator are obtained:
and o Qd, 1 = 8.42/3 = 2.81 k
, o Qd, 2 = 32.86/3=10.95 k
, ( B-20)
o Qd, 3 = 32.86/3 = 10.95 k
where values for Qd,j and Kd,j are obtained from the o Qd, 4 = 8.42/3 = 2.81 k
final cycle of iteration in the Simplified Method (Step
B1. 12 and
o Kd,1 = 5.10/3 = 1.70 k/in
o Kd,2 = 19.92/3 = 6.64 k/in
o Kd,3 = 19.92/3 = 6.64 k/in
o Kd,4 = 5.10/3 = 1.70 k/in

Note that the Kd values per support used above are

from the final iteration given in Table B1.12-1. These
are not the same as the initial values in Step B1.2,
because they have been adjusted from cycle to cycle,
such that the total Kd summed over all the isolators
satisfies the minimum lateral restoring force
requirement for the bridge, i.e. Kdtotal = 0.05 W/d. See
Step B1.1. Since d varies from cycle to cycle, Kd,j
varies from cycle to cycle.

B2.2 Initial Stiffness and Yield Displacement B2.2 Initial Stiffness and Yield Displacement,
Calculate the initial stiffness, Ku,i, and the yield Example 2.4
displacement, dy,i, for each isolator i as follows: Since the isolator type has been specified in Step A1
(1) For friction-based isolators Ku,i = and dy,i = 0. to be an elastomeric bearing (i.e. not a friction-based
(2) For other types of isolators, and in the absence of bearing), calculate Ku,i and dy,i for an isolator on Pier 1
isolator-specific information, take as follows:
, 10 , 10 6.64 66.4 /
, 10 , ( B-21)
and then and
, 10.95
,
( B-22) , 0.18
,
, , 66.4 6.64
, ,

As expected, the yield displacement is small

compared to the expected isolator displacement (~2

310

in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.

B2.3 Isolator Effective Stiffness, Kisol,i B2.3 Isolator Effective Stiffness, Kisol,i, Example 2.4
Calculate the isolator stiffness, Kisol,i, of each isolator Dividing the results for Kisol (Step B1.12) among the 3
i: isolators at each support, the following values for Kisol
/isolator are obtained:
, o Kisol,1 = 10.22/3 = 3.41 k/in
, (B -23)
o Kisol,2 = 42.78/3 = 14.26 k/in
o Kisol,3 = 42.78/3 = 14.26 k/in
o Kisol,4 = 10.22/3 = 3.41 k/in

B2.4 Three-Dimensional Bridge Model B2.4 Three-Dimensional Bridge Model, Example

Using computer-based structural analysis software, 2.4
create a 3-dimensional model of the bridge with the Although the bridge in this Design Example is regular
isolators represented by spring elements. The stiffness and is without skew or curvature, a 3-dimensional
of each isolator element in the horizontal axes (Kx and finite element model was developed for this Step, as
Ky in global coordinates, K2 and K3 in typical local shown below.
coordinates) is the Kisol value calculated in the
previous step. For bridges with regular geometry and
minimal skew or curvature, the superstructure may be .
represented by a single stick provided the load path
to each individual isolator at each support is explicitly
modeled, usually by a rigid cap beam and a set of rigid
links. If the geometry is irregular, or if the bridge is
skewed or curved, a finite element model is
recommended to accurately capture the load carried
by each individual isolator. If the piers have an
unusual weight distribution, such as a pier with a
hammerhead cap beam, a more rigorous model is
recommended.

B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in Step
A2 is for a 5% damped system. It is modified in this
step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the effective
period of the bridge, Teff, by the damping factor, BL.
B.2.5 Composite Design Response Spectrum,
Example 2.4
From the final results of Simplified Method (Step
B1.12), BL = 1.70 and Teff = 1.43 sec. Hence the
transition in the composite spectrum from 5% to 30%
damping occurs at 0.8 Teff = 0.8 (1.43) = 1.14 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.14 sec by 1.70.
0.8

0.7

0.6

0.5
Csm (g)

0.4

0.3

0.2

0.1

0
0 0.5 1 1.5 2 2.5 3 3.5 4
T (sec)

311

B2.6 Multimodal Analysis of Finite Element Model B2.6 Multimodal Analysis of Finite Element
Input the composite response spectrum as a user- Model, Example 2.4
specified spectrum in the software, and define a load Results of modal analysis of the example bridge are
case in which the spectrum is applied in the summarized in Table B2.6-1 Here the modal periods
longitudinal direction. Analyze the bridge for this and mass participation factors of the first 12 modes
load case. are given. The first three modes are the principal
transverse, longitudinal, and torsion modes with
periods of 1.60, 1.46 and 1.39 sec respectively. The
period of the longitudinal mode (1.46 sec) is very
close to that calculated in the Simplified Method. The
mass participation factors indicate there is no
coupling between these three modes (probably due to
the symmetric nature of the bridge) and the high
values for the first and second modes (92% and 94%
respectively) indicate the bridge is responding
essentially in a single mode of vibration in each
Mode Period Mass Participation Ratios
No Sec UX UY UZ RX RY RZ
1 1.604 0.000 0.919 0.000 0.952 0.000 0.697
2 1.463 0.941 0.000 0.000 0.000 0.020 0.000
3 1.394 0.000 0.000 0.000 0.000 0.000 0.231
4 0.479 0.000 0.003 0.000 0.013 0.000 0.002
5 0.372 0.000 0.000 0.076 0.000 0.057 0.000
6 0.346 0.000 0.000 0.000 0.000 0.000 0.000
7 0.345 0.000 0.001 0.000 0.010 0.000 0.000
8 0.279 0.000 0.003 0.000 0.013 0.000 0.002
9 0.268 0.000 0.000 0.000 0.000 0.000 0.000
10 0.267 0.058 0.000 0.000 0.000 0.000 0.000
11 0.208 0.000 0.000 0.000 0.000 0.129 0.000
12 0.188 0.000 0.000 0.000 0.000 0.000 0.001

direction. Similar results to that obtained by the

Simplified Method are therefore expected.

Table B2.6-1 Modal Properties of Bridge

Example 2.4 First Iteration
Computed values for the isolator displacements due to
a longitudinal earthquake are as follows (numbers in
parentheses are those used to calculate the initial
properties to start iteration from the Simplified
Method):

o disol,1 = 1.69 (1.65) in

o disol,2 = 1.20 (1.44) in
o disol,3 = 1.20 (1.44) in
o disol,4 = 1.69 (1.65) in

B2.7 Convergence Check B2.7 Convergence Check, Example 2.4

Compare the resulting displacements at the
superstructure level (d) to the assumed displacements.
These displacements can be obtained by examining
the joints at the top of the isolator spring elements. If
in close agreement, go to Step B2.9. Otherwise go to
Step B2.8.
The results for isolator displacements are close but
not close enough (15% difference at the piers)
Go to Step B2.8 and update properties for a second
cycle of iteration.

312

B2.8 Update Kisol,i, Keff,j, and BL B2.8 Update Kisol,i, Keff,j, and BL, Example 2.4
Use the calculated displacements in each isolator Updated values for Kisol,i are given below (previous
element to obtain new values of Kisol,i for each isolator values are in parentheses):
as follows:
, o Kisol,1 = 3.36 (3.41) k/in
, , (B-24) o Kisol,2 = 15.77 (14.26) k/in
,
o Kisol,3 = 15.77 (14.26) k/in
Recalculate Keff,j : o Kisol,4 = 3.36 (3.41) k/in

Eq. , ,
7.1-6 ,
(B-25) Since the isolator displacements are relatively close to
, ,
GSID previous results no significant change in the damping
ratio is expected. Hence Keff,j and are not
Recalculate system damping ratio, : recalculated and BL is taken at 1.70.

Eq. 2 , , ,
7.1-10 , , ,
(B-26)
GSID

Recalculate system damping factor, BL:

Eq. .
0.3
.
7.1-3 ( B-27)
1.7 0.3 Since the change in effective period is very small
GSID
(1.43 to 1.46 sec) and no change has been made to BL,
Obtain the effective period of the bridge from the there is no need to construct a new composite
multi-modal analysis and with the revised damping response spectrum in this case. Go back to Step B2.6
factor (Eq. B-27), construct a new composite response (see immediately below).
spectrum. Go to Step B2.6.
B2.6 Multimodal Analysis Second Iteration,
Example 2.4
Reanalysis gives the following values for the isolator
displacements (numbers in parentheses are those
from the previous cycle):

o disol,1 = 1.66 (1.69) in

o disol,2 = 1.15 (1.20) in
o disol,3 = 1.15 (1.20) in
o disol,4 = 1.66 (1.69) in

Go to Step B2.7

B2.7 Convergence Check B2.7 Convergence Check, Example 2.4

Compare results and determine if convergence has Satisfactory agreement has been reached on this
been reached. If so go to Step B2.9. Otherwise Go to second cycle. Go to Step B2.9
Step B2.8.

B2.9 Superstructure and Isolator Displacements B2.9 Superstructure and Isolator Displacements,
Once convergence has been reached, obtain Example 2.4
o superstructure displacements in the longitudinal From the above analysis:
(xL) and transverse (yL) directions of the bridge, o superstructure displacements in the
and longitudinal (xL) and transverse (yL) directions
o isolator displacements in the longitudinal (uL) are:
and transverse (vL) directions of the bridge, for xL= 1.69 in

313

each isolator, for this load case (i.e. yL= 0.0 in

longitudinal loading). These displacements
may be found by subtracting the nodal
displacements at each end of each isolator
spring element.
o isolator displacements in the longitudinal (uL)
and transverse (vL) directions are:
o Abutments: uL = 1.66 in, vL = 0.00 in
o Piers: uL = 1.15 in, vL = 0.00 in

All isolators at same support have the same

displacements.

B2.10 Pier Bending Moments and Shear Forces B2.10 Pier Bending Moments and Shear Forces,
Obtain the pier bending moments and shear forces in Example 2.4
the longitudinal (MPLL, VPLL) and transverse (MPTL, Bending moments in single column pier in the
VPTL) directions at the critical locations for the longitudinal (MPLL) and transverse (MPTL) directions
longitudinally-applied seismic loading. are:
MPLL= 0
MPTL= 1602 kft

Shear forces in single column pier the longitudinal

(VPLL) and transverse (VPTL) directions are
VPLL=67.16 k
VPTL=0

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 2.4
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1

Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Sub- Long. Transv. Axial
Isol-
struc shear due shear due forces due
ator
ture to long. to long. to long.
EQ EQ EQ
1 5.63 0 1.29
Abut
2 5.63 0 1.30
ment
3 5.63 0 1.29
1 18.19 0 0.77
Pier 2 18.25 0 1.11
3 18.19 0 0.77

The difference between the longitudinal shear force in

the column (VPLL = 67.16k) and the sum of the
isolator shear forces at the same Pier (54.63 k) is
about 12.5 k. This is due to the inertia force
developed in the hammerhead cap beam which
weighs about 128 k and can generate significant
additional demand on the column (about a 23%
increase in this case).

314

STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake
Repeat the above process, starting at Step B1, for
earthquake loading in the transverse direction of the
bridge. Support flexibility in the transverse direction
is to be included, and a composite response spectrum
is to be applied in the transverse direction. Obtain
isolator displacements in the longitudinal (uT) and
transverse (vT) directions of the bridge, and the biaxial
bending moments and shear forces at critical locations
in the columns due to the transversely-applied seismic
loading.
C1. Analysis for Transverse Earthquake, Example
2.4
Key results from repeating Steps B1 and B2
(Simplified and Mulitmode Spectral Methods) are:
o Teff = 1.52 sec
o Superstructure displacements in the longitudinal
(xT) and transverse (yT) directions are as follows:
xT = 0 and yT = 1.75 in
o Isolator displacements in the longitudinal (uT)
and transverse (vT) directions as follows:
Abutments uT = 0.00 in, vT = 1.75 in
Piers uT = 0.00 in, vT = 0.71 in
o Pier bending moments in the longitudinal (MPLT)
and transverse (MPTT) directions are as follows:
MPLT = 1548.33 kft and MPTT = 0
o Pier shear forces in the longitudinal (VPLT) and
transverse (VPTT) directions are as follows:
VPLT = 0 and VPTT = 60.75 k
o Isolator shear and axial forces are in Table C1-1.

Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Long. Transv. Axial
Sub-
Isol- shear d shear due forces due
struct
ator ue to to transv. to transv.
ure
transv. EQ EQ
EQ
1 0.0 5.82 13.51
Abut
2 0.0 5.83 0
ment
3 0.0 5.82 13.51
1 0.0 15.40 26.40
Pier 2 0.0 15.57 0
3 0.0 15.40 26.40

The difference between the transverse shear force in

the column (VPLL = 60.75k) and the sum of the
isolator shear forces at the same Pier (46.37 k) is
about 14.4 k. This is due to the inertia force
developed in the hammerhead cap beam which
weighs about 128 k and can generate significant
additional demand on the column (about 31% ).

315

STEP D. CALCULATE DESIGN VALUES

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements at Pier 1,

Following the provisions in Art. 2.1 GSID, and Example 2.4
illustrated in Fig. 2.1-1 GSID, calculate the total To illustrate the process, design displacements for the
design displacement, dt, for each isolator by outside isolator on Pier 1 are calculated below.
combining the displacements from the longitudinal (uL
and vL) and transverse (uT and vT) cases as follows: Load Case 1:
u1 = uL + 0.3uT = 1.0(1.15) + 0.3(0) = 1.15 in
u1 = uL + 0.3uT (D-1) v1 = vL + 0.3vT = 1.0(0) + 0.3(0.71) = 0.21 in
v1 = vL + 0.3vT (D-2) R1 = = 1.15 0.21 = 1.17 in
R1 = (D-3)
Load Case 2:
u2 = 0.3uL + uT (D-4) u2 = 0.3uL + uT = 0.3(1.15) + 1.0(0) = 0.35 in
v2 = 0.3vL + vT (D-5) v2 = 0.3vL + vT = 0.3(0) + 1.0(0.71) = 0.71in
R2 = (D-6) R2 = = 0.35 0.71 = 0.79 in

dt = max(R1, R2) (D-7)

Governing Case:
Total design displacement, dt = max(R1, R2)
= 1.17 in

D2. Design Moments and Shears D2. Design Moments and Shears in Pier 1,
Calculate design values for column bending moments Example 2.4
and shear forces for all piers using the same Design moments and shear forces are calculated for
combination rules as for displacements. Pier 1 below, to illustrate the process.

Alternatively this step may be deferred because the Load Case 1:

above results may not be final. Upper and lower VPL1= VPLL + 0.3VPLT = 1.0(67.16) + 0.3(0) = 67.16 k
bound analyses are required after the isolators have VPT1= VPTL + 0.3VPTT = 1.0(0) + 0.3(60.75) = 18.23 k
been designed as described in Art 7. GSID. These R1 = = 67.16 18.23 = 69.59 k
analyses are required to determine the effect of
possible variations in isolator properties due age, Load Case 2:
temperature and scragging in elastomeric systems. VPL2= 0.3VPLL + VPLT = 0.3(67.16) + 1.0(0) = 20.15 k
Accordingly the results for column shear in Steps VPT2= 0.3VPTL + VPTT = 0.3(0) + 1.0(60.75) = 60.75 k
B2.10 and C are likely to increase once these analyses
R2 = = 20.15 60.75 = 64.00 k
are complete.
Governing Case:
Design column shear = max (R1, R2)
= 69.59 k

316

STEP E. DESIGN OF ERADIQUAKE ISOLATORS

An EradiQuake Isolator (EQS) is a sliding
isolation bearing composed of a multi-
directional sliding disc bearing and lateral
springs. Each spring assembly consists of a
cylindrical polyurethane spring and a spring
piston. The piston keeps the spring straight as
the isolator moves in different directions. The
disc bearing and springs are housed in a
mirror-finished stainless steel lined box.

The required values for Qd and Kd determine

the coefficient of friction at the sliding
interface and the properties of the springs.

The sliding interface is typically comprised of

stainless steel and PTFE. Energy is dissipated
during sliding while the springs provide a
restoring force. PTFE is an attractive material
in that at slow sliding speeds it has a low
coefficient of friction, which is ideal for
accommodating thermal effects, and at higher
speeds the friction becomes greater and acts as
an effective energy dissipator during seismic events. The polyurethane springs are designed such that they are
never in tension. Their basic design and composition is derived from the die-spring industry.

Design and materials conform to the LRFD Specifications. Steel components are designed in accordance with
Section 6, while the disc bearing is designed and constructed per Section 14.

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

Notation
A Area Kd Stiffness when sliding (Total spring rate)
A1 Area based on dead load k1 Stiffness (spring rate) for one spring
A2 Area based on total load L Length
AS Spring area LGB Guide bar length
BBB Bearing block plan dimension LS Spring length
BBox Guide box plan dimension (out to out LSI Installed spring length
dimension of guide bars) LL Live Load
BBP Base plate length L1 Spring length based on max long term
BPTFE PTFE dimension displacement
BSP Slide plate (guide box top) length L2 Spring length based on max short term
DD Disc outer diameter displacement
DPTFE PTFE diameter M Moment
DS Spring outer diameter MN Factored moment
dL Service (long term) displacement PDL Dead load
dT Total seismic displacement PLL Live load
E Elastic modulus PSL Seismic live load
F Spring force PWL Wind load
FY Yield stress Qd Characteristic strength
H Isolator height SG Gross shape factor of disc
IDS Spring inner diameter TBB Bearing block thickness

317

TBP Base plate thickness Bearing rotation

TD Disc thickness Inner to outer diameter ratio
TGB Guide bar thickness W Wind displacement
TSP Slide plate thickness C Maximum average compression strain
W Isolator weight Coefficient of friction
W Plan width of isolator

WBP Base plate length
WL Wind load
Z Plastic modulus

E1. Required Properties E1. Required Properties, Example 2.4

Obtain from previous work the properties required of The design of one of the exterior isolators on a pier is
the isolation system to achieve the specified given below to illustrate the design process for an
performance criteria (Step A1). EQS isolator.
required characteristic strength, Qd, per
isolator From previous work:
required post-elastic stiffness, Kd, per isolator Qd/isolator = 10.95 k
total design displacement, dt, for each isolator Kd/isolator = 6.76 k/in
maximum applied dead and live load, PDL, Total design displacement, dt = 1.17 in
PLL, and seismic load, PSL, which includes PDL = 187.0 k
seismic live load (if any) and overturning PLL = 123.0k
forces due to seismic loads, at each isolator PSL = 26.4 k
wind load per isolator, PWL, and
thermal displacement of the superstructure at Calculated for this design:
each isolator, dL. PWL = 8.21 k
dL = +/- 0.53 in
E2. Isolator Sizing
E2.1 Size the Disc E2.1 Size Disc, Example 2.4
Estimate the disc outer diameter based on an average
compressive stress of 4.5 ksi using the gross plan
area.
4
4.5

(E-1)

0.53
0.53310 9.33 9.50
Estimate disc thickness based on a gross shape factor
of 2.4:
9.5
(E-2) 0.99 1.00
4 4 2.4

Check rotational capacity and adjust disc thickness if

required. Use standard disc design rotation, , of

0.02 radians, and a maximum compression strain, c,
of 0.10 for this calculation.

0.02 9.50
(E-3)
0.95
2 2 2 0.10

318

E2.2 Size the Springs E2.2 Size the Springs, Example 2.4

E2.2.1 Calculate Installed Spring Length E2.2 Calculate Installed Spring Length, Example
Assume 60% max compressive strain on the MER 2.4
spring for short term loading, 40% max compressive
strain for long term loading. Add 20% of long term
loading strain for elastomer compression set.
Then
2.5 (E-4) 2.5 0.53 1.33

And using a load combination of two times the total

seismic design displacement (for Seismic Zone 2)
plus 50% service (thermal):

2 0.5 2 1.17 0.5 0.53

(E-5) 4.34
0.60 0.60

Then required spring length is given by:

max , 0.2 (E-6) 4.34 0.2 0.53 4.45 5.0

E2.2.2 Check Wind Displacements E2.2.2 Check Wind Displacements, Example 2.4
Calculate displacement due to wind as follows:

0.25 8.2 0.25 11.0

(E-7) 0.81
6.76
If the displacement due to wind is too large, add
spring precompression equal to the wind In this example, the wind displacement is acceptable
displacement to the spring length in the transverse and no adjustment of spring length is required.
direction. Precompression doubles the stiffness over
the precompressed displacement. If the wind
displacements are still too large, consider increasing
the spring stiffness in the transverse direction, or
using sacrificial shear keys.

E2.2.3 Calculate Spring Diameter E2.2.3 Calculate Spring Diameter, Example 2.4
Assume only one spring per side is used to meet
spring rate requirements, i.e. let k1=Kd, and take the
elastic modulus for polyurethane spring to be 6.0 ksi.
Since Since Kd = 6.76 k/in

6.0 6.76 5.00

(E-8) 5.63
6.0
it follows that
(E-9) Take initial value for = 0.20 and then
6.0
and
4 5.63
4 2.73 2.75
(E-10) 1 0.2
1

319

E2.2.4 Adjust Spring Length Using Nominal E2.2.4 Adjust Spring Length Using Nominal
Diameters Diameters, Example 2.4
For manufacturing purposes it is advantageous to use Use 2-3/4 in for the spring OD, and 0.50 in for the
standard diameters and adjust the spring length spring ID then
according to the actual value of to fine tune the
stiffness (spring rate).

(E-11) 0.50
0.18
2.75
1 (E-12)
4 2.75 1 0.18 5.75
4
(E-13)
6.0 5.75
5.10
6.76

Check that Ls is greater than either L1 and L2 max ,

Note that LS is the installed spring length. The actual

size of the springs may be slightly different than the
installed size. Springs are pre-compressed to provide
additional wind resistance if needed and account for
compression set in the elastomer.

E2.3 Size the PTFE Pad E2.3 Size the PTFE Pad, Example 2.4

E2.3.1 Calculate Coefficient of Friction E2.3.1 Calculate Required Coefficient of Friction,

Calculate the required coefficient of friction from: Example 2.4

(E-14)
Select PTFE and polished stainless steel as the
sliding surfaces. Low coefficients of friction are
possible with these materials at high contact stresses.
In general the friction coefficient decreases with
increasing pressure.
11.0
0.059
187
A value of 0.059 is at the low end of the spectrum for
virgin PTFE/stainless steel materials and will require
use of the highest contact stresses allowed in the
GSID and LRFD Specifications to achieve this value;
i.e. 3.5 ksi under dead load and 4.5 ksi under (dead +
live) load.

E2.3.2 Calculate Required Area of PTFE E2.3.2 Calculate Required Area of PTFE, Example
Calculate required area of PTFE using allowable 2.4
contact stresses in GSID Table 16.4.1-1. For service
loads (i.e. dead load) allowable average stress is 3.5
ksi and then:

187
(E-15) 53.4
3.5 3.5
Check area required under dead plus live load using
an allowable average stress of 4.5 ksi (as permitted in
LRFD Sec 14.)
187 123
(E-16) 68.9
4.5 4.5
Then required area is
max , (E-17) max , 68.9

Since the structure design rotation of 0.01 radians is

320

only one-half of the disc design rotation, the limits on

the PTFE edge contact stresses (GSID Table 16.4.1-
1) do not govern.

E2.3.3 Calculate Size of PTFE Pad E2.3.3 Calculate Size of PTFE Pad, Example 2.4
For a circular PTFE pad, the diameter is given by:

4 4
(E-18) 68.9 9.36 9.375

E2.4 Size the Bearing Block E2.4 Size the Bearing Block, Example 2.4

E2.4.1 Calculate Bearing Plan Dimension E2.4.1 Calculate Bearing Plan Dimension,
Two criteria must be checked to determine the Example 2.4
bearing block plan dimension. The disc must fit
under the block with some clearance, and the PTFE
must fit on top of the block with at least 1/8 in edge
clearance.
1.15 (E-19) 1.15 9.50 10.9

2 0.125 (E-20) 9.375 2 0.125 9.625

max , (E-21) max , 10.9 11.00

E2.4.2 Calculate Bearing Block Thickness E2.4.2 Calculate Bearing Block Thickness,
The thickness of bearing block must be sufficient to Example 2.4
ensure that the springs can be attached on each side
of the block, allowing for a 30% increase in diameter
upon spring compression.

1.3 (E-22) 1.3 2.75 3.58 3.50

Note that if TBB is too large, reduce the diameter of

the springs and increase their number. Size is ok. No need to resize spring diameters.

321

E2.5 Size the Box E2.5 Size the Box, Example 2.4

E2.5 .1 Calculate Guide Bar Thickness E2.5.1 Calculate Guide Bar Thickness, Example
(a) Guide Bar Force 2.4
Guide bars resist the spring forces. They are
modeled as cantilever beams, with the fixed end of
the cantilever located where the guide bar meets the
slide plate. Assume the resisting length of guide bar
to be three times the diameter of the spring. The
moment arm is one-half of the bearing block
thickness, plus 0.20 in. Forces corresponding to two
times the seismic displacement, imposed during
prototype testing, are used to design the guide bar.

2 0.5 (E-23) 6.76 2 1.17 0.5 0.53 17.6

(b) Guide Bar Moment

0.5 0.20 (E-24) 0.5 3.50 0.20 17.6 34.33

Since the effective length of guide bar resisting this

moment is assumed to be 3Ds, the bending moment
per inch of guide bar is:
34.33
(E-25) 4.16 /
3 3 2.75
(c) Guide Bar Thickness
Using a load factor of 1.75, a resistance factor of
1.00, and assuming 50 ksi steel:

1.75 1.00 (E-26)

then
1.75 1.75 4.16
(E-27) 0.146 /
50
But since
(E-28)
4
then
4 (E-29) 4 0.146 0.76 0.75

E2.5.2 Calculate Guide Bar Length E2.5.2 Calculate Guide Bar Length, Example 2.4

2 (E-30) 0.75 11.00 2 5.10 21.95 22.0

E2.5.3 Calculate Guide Bar Width E2.5.3 Calculate Guide Bar Width, Example 2.4

3.50 0.5 1.00 4.0

0.5 (E-31)
E2.5.4 Calculate Size of Box and Slide Plate E2.5.4 Calculate Size of Box and Slide Plate,
Calculate side dimension of box Example 2.4
(E-32)
22.0 0.75 22.75
Choose plan dimension of slide plate equal to, or
slightly larger than, the box, i.e.
Take

322

23.0
(E-33)

E2.5.5 Calculate Box Top (Slide Plate) Thickness E2.5.5 Calculate Box Top (Slide Plate) Thickness,
Example 2.4
Make the slide plate (guide box top) the same
thickness as the guide bars, with a minimum value of 0.75
in.

E2.6 Size the Lower Plate E2.6 Size the Lower Plate, Example 2.4
(a) Thickness
Use inch minimum thickness unless otherwise
required by State DOT specifications. 0.75
(b) Width
Since GSID provisions for prototype testing require
the isolator to be displaced to twice the design
displacement (for Seismic Zone 2), the base plate
must be wide enough to allow such movement
without interference from the anchor bolts.
4 8 (E-34) 23.0 4 1.17 8 35.68 36.00

(c) Length
Take
(E-35) 23.00

(d) Anchor Bolts

Design anchor bolts per LRFD specifications,
increase WBP if necessary.

E3. Design Summary E3 Design Summary, Example 1.4

Overall dimensions of isolator are:
Width = WBP Width = WBP = 36.00 in
Length = BSP Length = BSP = 23.00 in
Height is given by: Height is given by:
0.20 (E-36) 0.75 1.00 3.50 0.75 0.20 6.20

323

E4. Minimum and Maximum Performance Check E4. Minimum and Maximum Performance Check,
Art. 8 GSID requires the performance of any Example 2.4
isolation system be checked using minimum and For Eradiquake isolators, Modification Factors are
maximum values for the effective stiffness of the applied to both Qd and Kd , because both frictional and
system. These values are calculated from minimum elastomeric (urethane) elements are used in these
and maximum values of Kd and Qd, which are found isolators.
using system property modification factors, as
indicated in Table E4-1. Minimum Property Modification factors are:
min,Kd = 1.0
Table E4-1. Minimum and maximum values min,Qd = 1.0
for Kd and Qd.
which means there is no need to reanalyze the bridge
Eq. with a set of minimum values.
8.1.2-1 Kd,max = Kd max,Kd (E-37)
GSID Maximum Property Modification factors are (GSID
Eq. Appendix A):
8.1.2-2 Kd,min = Kd min,Kd (E-38) max,a,Kd = 1.0
GSID max,a,Qd = 1.2
Eq.
8.1.2-3 Qd,max = Qd max,Qd (E-39) max,t,Kd = 1.3
GSID max,t,Qd = 1.5
Eq.
8.1.2-4 Qd,min = Qd min,Qd (E-40) max,tr,Kd = 1.0
GSID max,tr,Qd = 1.0

max,c,Kd = 1.0
Determination of the system property modification
max,c,Qd = 1.1
factors should include consideration of the effects of
temperature, aging, scragging, velocity, travel (wear)
and contamination as shown in Table E4-2. In lieu of
tests, numerical values for these factors can be
obtained from Appendix A, GSID.

Table E4-2. Minimum and maximum values for

system property modification factors.

Eq. min,Kd = (min,t,Kd) (min,a,Kd)

8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-41)
GSID (min,scrag,Kd)
Eq. max,Kd = (max,t,Kd) (max,a,Kd)
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-42)
GSID (max,scrag,Kd)
Eq. min,Qd = (min,t,Qd) (min,a,Qd)
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-43)
GSID (min,scrag,Qd)
Eq. max,Qd = (max,t,Qd) (max,a,Qd)
8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd) (E-44)
GSID (max,scrag,Qd)

Adjustment factors are applied to individual -factors
(except v) to account for the likelihood of
occurrence of all of the maxima (or all of the
minima) at the same time. These factors are applied
to all -factors that deviate from unity but only to the
portion of the -factor that is greater than, or less
than, unity. Art. 8.2.2 GSID gives these factors as
Applying a system adjustment factor of 0.66 for an
other bridge, the maximum property modification
factors become:
max,a,Kd = 1.0 + 0.0(0.66) = 1.00
max,a,Qd = 1.0 + 0.2(0.66) = 1.13

324

follows: max,t,Kd = 1.0 + 0.3(0.66) = 1.20

1.00 for critical bridges
0.75 for essential bridges
0.66 for all other bridges
As required in Art. 7 GSID and shown in Fig. C7-1
GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Qd,max. As indicated in Fig C7-1 GSID, maximum
displacements will probably be given by the first case
(Kd,min and Qd,min) and maximum forces by the second
case (Kd,max and Qd,max).
max,t,Qd = 1.0 + 0.5(0.66) = 1.33
max,tr,Kd = 1.0 + 0.0(0.66) = 1.00
max,tr,Qd = 1.0 + 0.0(0.66) =1.00
max,c,Kd = 1.0 + 0.0(0.66) = 1.00
max,c,Qd = 1.0 + 0.0(0.66) =1.00
Therefore the maximum overall modification factors
max,Kd = (1.00)(1.20)(1.00)(1.00) = 1.20
max,Qd = (1.13)(1.33)(1.00)(1.00) = 1.50

Since the possible variation in upper bound properties

exceeds 15% a reanalysis of the bridge is required to
determine performance with these properties.

The upper-bound properties are:

Qd,max = 1.50(11.0) = 16.5 k
and
Kd,ma x=1.20(6.76) = 8.11 k/in

E5. Design and Performance Summary, Example

E5. Design and Performance Summary 2.4

E5.1 Isolator dimensions E5.1 Isolator dimensions, Example 2.4

Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E5.1-1.
Overall size of lower plate
Overall size of box (top plate) Table E5.1-1 Isolator Dimensions
Overall height
Isolator Overall size Overall size Diam.
Size of disc Location including without Disc
Size of PTFE pad mounting mounting (in)
Number of polyurethane springs plates (in) plates (in)
Diameter of polyurethane springs Under
edge 36.0 x 23.0 23.0 x 23.0
9.50
girder x 6.20(H) x 6.20 (H)
on Pier 1
Isolator Diam. No. Diam.
Location PTFE poly- poly-
pad urethane urethane
(in) springs springs (in)
Under
edge
9.375 4 2.75
girder
on Pier 1

E5.2 Bridge Performance E5.2 Bridge Performance, Example 2.4

Summarize bridge performance
Maximum superstructure displacement
(longitudinal)
Maximum superstructure displacement
(transverse)
Maximum superstructure displacement
Bridge performance is summarized in Table E5.2-1
where it is seen that the maximum column shear is
71.74k. This less than the column plastic shear (128k)
and therefore the required performance criterion is
satisfied (fully elastic behavior). Furthermore the
maximum longitudinal displacement is 1.69 in which

325

(resultant) is less than the 2.5in available at the abutment

Maximum column shear (resultant) expansion joints and is therefore acceptable.
Maximum column moment (about transverse
axis) Table E5.2-1 Summary of Bridge Performance
Maximum column moment (about
longitudinal axis) Maximum superstructure
1.69 in
Maximum column torque displacement (longitudinal)
Maximum superstructure
1.75 in
Check required performance as determined in Step displacement (transverse)
A3, is satisfied. Maximum superstructure
1.82 in
displacement (resultant)
Maximum column shear
71.74 k
(resultant)
Maximum column moment
1,657 kft
about transverse axis
Maximum column moment
1,676 kft
about longitudinal axis
Maximum column torque 21.44 kft

326

 SECTION 2
Steel Plate Girder Bridge, Long Spans, Single-Column Pier

DESIGN EXAMPLE 2.5: H1=0.5H2

Design Examples in Section 2

Site
ID Description S1 Column height Skew Isolator type
Class
2.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
2.1 Change site class 0.2g D Same 0 Lead rubber bearing
2.2 Change spectral acceleration, S1 0.6g B Same 0 Lead rubber bearing
Spherical friction
2.3 Change isolator to FPS 0.2g B Same 0
bearing
2.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
2.5 Change column height 0.2g B H1=0.5H2 0 Lead rubber bearing
0
2.6 Change angle of skew 0.2g B Same 45 Lead rubber bearing

327

 DESIGN PROCEDURE
STEP A: BRIDGE AND SITE DATA
A1. Bridge Properties
Determine properties of the bridge:
number of supports, m
number of girders per support, n
weight of superstructure including railings,
curbs, barriers and other permanent loads,
WSS
weight of piers participating with
superstructure in dynamic response, WPP
weight of superstructure, Wj, at each
support
stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge. The calculation of these
quantities requires careful consideration of
several factors such as the use of cracked
sections when estimating column or wall
flexural stiffness, foundation flexibility,
and effective column height.
column shear strength (minimum value).
This will usually be derived from the
minimum value of the column flexural
yield strength, the column height, and
whether the column is acting in single or
double curvature in the direction under
consideration.
allowable movement at expansion joints
isolator type if known, otherwise to be
selected
A2. Seismic Hazard
Determine seismic hazard at site:
acceleration coefficients
site class and site factors
seismic zone
Plot response spectrum.
Use Art. 3.1 GSID to obtain peak ground and spectral
acceleration coefficients. These coefficients are the
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (e.g. Figures
3.10.2.1-1 to 3.10.2.1-21 LRFD).
Use Art. 3.2 to obtain Site Class and corresponding Site
Factors (Fpga, Fa and Fv). These data are the same as for
conventional bridges and Art 3.2 refers the designer to
the corresponding articles in the LRFD Specifications,
i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3,
DESIGN EXAMPLE 2.5 (H1= 0.5H2)
A1. Bridge Properties, Example 2.5
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 3
Number of columns per support = 1
Weight of superstructure including permanent
loads, WSS = 1651.32 k
Weight of superstructure at each support:
o W1 = 168.48 k
o W2 = 657.18 k
o W3 = 657.18 k
o W4 = 168.48 k
Participating weight of piers, WPP = 256.26 k
Effective weight (for calculation of period),
Weff = Wss + WPP = 1907.58 k
Stiffness of each pier in the both directions:
o Ksub,pier1 = 288.87 k/in
o Ksub,pier2 = 36.58 k/in
Minimum column shear strength based on
flexural yield capacity of column = 128 k
Displacement capacity of expansion joints
(longitudinal) = 2.5 in (required to accommodate
thermal expansion and other movements)
Lead rubber isolators
A2. Seismic Hazard, Example 2.5
Acceleration coefficients for bridge site are given in
design statement as follows:
PGA = 0.40
S1 = 0.20
SS = 0.75
Bridge is on a rock site with shear wave velocity in
upper 100 ft of soil = 3,000 ft/sec.
Table 3.10.3.1-1 LRFD gives Site Class as B.
Tables 3.10.3.2-1, -2 and -3 LRFD give following
Site Factors:
Fpga = 1.0
Fa = 1.0
Fv = 1.0
328

LRFD.

Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give

modified spectral acceleration coefficients that include
site effects as follows:
As = Fpga PGA As = Fpga PGA = 1.0(0.40) = 0.40
SDS = Fa SS SDS = Fa SS = 1.0(0.75) = 0.75
SD1 = Fv S1 SD1 = Fv S1 = 1.0(0.20) = 0.20

Seismic Zone is determined by value of SD1 in Since 0.15 < SD1 < 0.30, bridge is located in Seismic
accordance with provisions in Table 5-1 GSID. Zone 2.

These coefficients are used to plot design response Design Response Spectrum is as below:
spectrum as shown in Fig. 4-1 GSID.
0.9

0.8

0.7

0.6

Acceleration
0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Period(s)

A3. Performance Requirements A3. Performance Requirements, Example 2.5

Determine required performance of isolated bridge In this example, assume the owner has specified full
during Design Earthquake (1000-yr return period). functionality following the earthquake and therefore
the columns must remain elastic (no yield).
Examples of performance that might be specified by
the Owner include: To remain elastic the maximum lateral load on the
Reduced displacement ductility demand in pier must be less than the load to yield any one
columns, so that bridge is open for emergency column (128 k).
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in The maximum shear in the column must therefore be
columns or yield elsewhere in bridge), so that less than 128 k in order to keep the column elastic
bridge is fully functional and open to all vehicles and meet the required performance criterion.
immediately following earthquake.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.

329

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for

Isolator Force, F
non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most
have bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use Ku
equivalent linear springs and viscous damping to
represent the isolators, so that linear methods of analysis
may be used to determine response. Since equivalent Ku
dy disol
Isolator
Ku Displacement, d
properties such as Kisol are dependent on displacement
(d), and the displacements are not known at the
beginning of the analysis, an iterative approach is Kd
required. Note that in Art 7.1, GSID, keff is used for the
effective stiffness of an isolator unit and Keff is used for disol = Isolator displacement
the effective stiffness of a combined isolator and dy
Fisol
= Isolator yield displacement
= Isolator shear force
substructure unit. To minimize confusion, Kisol is used in Fy = Isolator yield force
Kd = Post-elastic stiffness of isolator
this document in place of keff. There is no change in the Kisol = Effective stiffness of isolator
use of Keff and Keff,j, but Ksub is used in place of ksub. Ku = Loading and unloading stiffness (elastic stiffness)
Qd = Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement (disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 2.5
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.20 2.0
GSID

(2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not

330

 occur under non-seismic loads (e.g. wind) but

low enough that yield does occur under
earthquake. Experience has shown that
taking Qd to be 5% of the bridge weight is a
good starting point, i.e.
0.05 (B-2) 0.05 0.05 1651.32 82.56
(3) Post-yield stiffness, Kd
Art 12.2 GSID requires that all isolators
exhibit a minimum lateral restoring force at
the design displacement, which translates to a
minimum post yield stiffness Kd,min given by
Eq. B-3.

Art.
0.025
12.2 , (B-3) 1651.32
GSID 0.05 0.05 41.28 /
2.0
Experience has shown that a good starting
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 2.5
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength, ,

Qd, and stiffness, Kd, values in proportion to the dead o Qd, 1 = 8.42 k
load applied at that support: o Qd, 2 = 32.86 k
o Qd, 3 = 32.86 k
o Qd, 4 = 8.42 k

, (B-4)
and
and
,
, (B-5) o Kd,1 = 4.21 k/in
o Kd,2 = 16.43 k/in
o Kd,3 = 16.43 k/in
o Kd,4 = 4.21 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 2.5
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of
the isolators at support j (Kisol,j) and the stiffness of
the substructure Ksub,j. See figure below (after Fig.
7.1-1 GSID). , ,

, ,
An expression for Keff,j, is given in Eq.7.1-6 GSID, but
a more useful formula is as follows (MCEER 2006): o 1 = 8.43x10-4
,
(B-6) o 2 = 1.21x10-1
,
1 o 3 = 1.63
where o 4 = 8.43x10-4
, ,
(B-7)
, ,
and Ksub,j for the piers are given in Step A1. ,
,
1
For abutments, take Ksub,j to be a large number, say

331

10,000 k/in , unless actual stiffness values are

available. Note that if the default option is chosen, o Keff,1 = 8.42 k/in
unrealistically high values for Ksub,j will give o Keff,2 = 31.09 k/in
unconservative results for column moments and shear o Keff,3 = 22.67 k/in
forces. o Keff,4 = 8.42 k/in

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 2.5
Calculate the total effective stiffness, Keff, of the
bridge:

Eq. , 70.61 /
7.1-6 , ( B-8)
GSID

B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 2.5
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1

(B-9) o disol,1 = 2.00 in

,
1 o disol,2 = 1.79 in
o disol,3 = 0.76 in
o disol,4 = 2.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the stiffness of the isolation system at Example 2.5
support j, Kisol,j, for all supports:
,
, ,
,
,
(B-10) o Kisol,1 = 8.43 k/in
, ,
, o Kisol,2 = 34.84 k/in
o Kisol,3 = 59.65 k/in
o Kisol,4 = 8.43 k/in

B1.7 Substructure Displacement at Each Support B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j, Example 2.5
for all supports: , ,

o dsub,1 = 0.002 in
, , (B-11) o d sub,2 = 0.215 in
o d sub,3 = 1.240 in
o d sub,4 = 0.002 in

B1.8 Substructure Shear at Each Support B1.8 Lateral Load in Each Substructure, Example
Calculate the shear at support j, Fsub,j, for all 2.5
supports:
, , , (B-12) , , ,

where values for Ksub,j are given in Step A1. o F sub,1 = 16.84 k
o F sub,2 = 62.18k
o F sub,3 = 45.35 k
o F sub,4 = 16.84 k

332

B1.9 Column Shear at Each Support B1.9 Column Shear Forces at Each Support,
Calculate the shear in column k at support j, Fcol,j,k, Example 2.5
assuming equal distribution of shear for all columns at
support j:
,
, ,
#
,
, ,
# (B-13) o F col,2,1 = 62.18 k
o F col,3,1 = 45.35 k

Use these approximate column shears as a check on These column shears are less than the plastic shear
the validity of the chosen strength and stiffness capacity of each column (128k) as required in Step
characteristics. A3 and the chosen strength and stiffness values in
Step B1.1 are therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 2.5
damping ratio, , of the bridge:

Eq. 1907.58
7.1-5 2 (B-14) 2 2
GSID 386.4 70.61

and = 1.66 sec

Eq. 2 , , , and taking dy,j = 0:

7.1-10 (B-15)
, , ,
GSID 2 , , 0
0.26
, , ,
where dy,j is the yield displacement of the isolator and
assumed to be small compared to disol,j with negligible
effect on , i.e., take dy,j = 0.

B1.11 Damping Factor B1.11 Damping Factor, Example 2.5

Calculate the damping factor, BL, and the Since 0.26 0.3
displacement, d, of the bridge:

Eq. .
.
, 0.3 0.26
7.1-3 .
(B-16) 1.65
1.7, 0.3 0.05
GSID
and

Eq. 9.79 (B-17) 9.79 9.79 0.2 1.66

7.1-4 1.97
1.65
GSID

B1.12 Convergence Check B1.12 Convergence Check, Example 2.5

Compare the new displacement with the initial value
assumed in Step B1.1. If there is close agreement, go
to the next step; otherwise repeat the process from
Step B1.3 with the new value for displacement as the
assumed displacement.
This iterative process is amenable to solution using a
Since the calculated value for displacement, d (=1.97)
is close to that assumed at the beginning of the cycle
(Step B1.1, d = 2.0), this solution could be taken as
final. However to see the effect of iteration, Step B1.3
was repeated using 1.97 as the new assumed
displacement.

333

spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
After two iterations, convergence is reached at a
superstructure displacement of 1.95 in, with an
effective period of 1.64 seconds, and a damping factor
of 1.65 (26% damping ratio). The displacement in the
isolators on Pier 1 (19 ft column) is 1.73 in and the
effective stiffness of the same isolators is 35.79 k/in.
For Pier 2 (38 ft column), these values are 0.72 in and
62.49 k respectively.
See spreadsheet in Table B1.12-1 for results of final
iteration.
Since the column shear must equal the isolator shear
for equilibrium, the column shear in Pier 1 = 35.79
(1.73) = 61.92 k. Likewise the column shear in Pier 2
is 62.49(0.72) = 44.99 k. Both values are less than the
maximum allowable (128k) for elastic behavior in the
columns as required in Step A3. It is seen that the
taller pier attracts less shear because of its greater
flexibility.
Also the superstructure displacement = 1.95 in, which
is less than the available clearance of 2.5 in.
Therefore the above solution is acceptable and go to
Step B2.
Note that available clearance (2.5 in) is greater than
minimum required
8 8 0.20 1.64
1.59
1.65
334

 Table B1.12-1 Simplified Method Solution for Design Example 2.5 Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
1651.32 256.26 1907.58 0.2 3

StepB1.1 d 1.95 Assumeddisplacement

Qd 82.57 Characteristicstrength
Kd 42.34 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 168.48 8.424 4.320 10,000.00 0.000864 8.636 1.948 8.644 0.002 16.841 16.413 32.839
Pier1 657.18 32.859 16.851 288.87 0.123894 31.844 1.735 35.789 0.215 62.096 57.012 121.087
Pier2 657.18 32.859 16.851 36.58 1.708203 23.073 0.720 62.486 1.230 44.992 23.660 87.735
Abut2 168.48 8.424 4.320 10,000.00 0.000864 8.636 1.948 8.644 0.002 16.841 16.413 32.839
Total 1651.32 82.566 42.342 K eff,j 72.189 140.769 113.496 274.500
Step B1.4

StepB1.10 T eff 1.64 Effectiveperiod

0.26 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.65

B L 1.65 DampingFactor
d 1.95 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 2.808 1.440 2.881 1.69 3.102
Pier1 10.953 5.617 11.930 1.20 14.744
Pier2 10.953 5.617 20.829 1.20 14.744
Abut2 2.808 1.440 2.881 1.69 3.102

335

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and then
applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 2.5

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 by
elastic stiffness, Kd,i, of each isolator i as follows: the number of isolators at each support (n = 3), the
, following values for Qd /isolator and Kd /isolator are
, (B-19)
obtained:
and o Qd, 1 = 8.42/3 = 2.81 k
, o Qd, 2 = 32.86/3=10.95 k
, ( B-20)
o Qd, 3 = 32.86/3 = 10.95 k
where values for Qd,j and Kd,j are obtained from the o Qd, 4 = 8.42/3 = 2.81 k
final cycle of iteration in the Simplified Method (Step
B1. 12 and
o Kd,1 = 4.32/3 = 1.44 k/in
o Kd,2 = 16.85/3 = 5.62 k/in
o Kd,3 = 16.85/3 = 5.62 k/in
o Kd,4 = 4.32/3 = 1.44 k/in

Note that the Kd values per support used above are

from the final iteration given in Table B1.12-1. These
are not the same as the initial values in Step B1.2,
because they have been adjusted from cycle to cycle,
such that the total Kd summed over all the isolators
satisfies the minimum lateral restoring force
requirement for the bridge, i.e. Kdtotal = 0.05 W/d. See
Step B1.1. Since d varies from cycle to cycle, Kd,j
varies from cycle to cycle.

B2.2 Initial Stiffness and Yield Displacement B2.2 Initial Stiffness and Yield Displacement,
Calculate the initial stiffness, Ku,i, and the yield Example 2.5
displacement, dy,i, for each isolator i as follows: For an isolator on Pier 1:
, 10 , 10 5.62 56.2 /
In the absence of isolator-specific information take
and
, 10 , ( B-21) , 10.95
and then , 0.22
, , 56.2 5.62
,
, ( B-22) As expected, the yield displacement is small
, ,
compared to the expected isolator displacement (~2
in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.

336

B2.3 Isolator Effective Stiffness, Kisol,i
Calculate the isolator stiffness, Kisol,i, of each isolator
i:
(B -23)
B2.4 Finite Element Model
Using computer-based structural analysis software,
create a finite element model of the bridge with the
isolators represented by spring elements. The stiffness
of each isolator element in the horizontal axes (Kx and
Ky in global coordinates, K2 and K3 in typical local
coordinates) is the Kisol value calculated in the
previous step.
B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in Step
A2 is for a 5% damped system. It is modified in this
step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the effective
period of the bridge, Teff, by the damping factor, BL.
B2.6 Multimodal Analysis of Finite Element Model
Input the composite response spectrum as a user-
specified spectrum in the software, and define a load
case in which the spectrum is applied in the
longitudinal direction. Analyze the bridge for this
load case.
B2.3 Isolator Effective Stiffness, Kisol,i, Example 2.5
Dividing the results for Kisol (Step B1.12) among the 3
isolators at each support, the following values for Kisol
/isolator are obtained:
o Kisol,1 = 8.64/3 = 2.88 k/in
o Kisol,2 = 35.79/3 = 11.93 k/in
o Kisol,3 = 62.498/3 = 20.83 k/in
o Kisol,4 = 8.64/3 = 2.88 k/in
B2.4 Finite Element Model, Example 2.5
.
B.2.5 Composite Design Response Spectrum,
Example 2.5
From the final results of Simplified Method (Step
B1.12), BL = 1.65 and Teff = 1.64 sec. Hence the
transition in the composite spectrum from 5% to 26%
damping occurs at 0.8 Teff = 0.8 (1.64) = 1.31 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.31 sec by 1.65.
B2.6 Multimodal Analysis of Finite Element
Model, Example 2.5
Results of modal analysis of this bridge are
summarized in Table B2.6-1 Here the modal periods
and mass participation factors of the first 12 modes
are given. The first three modes are the principal
isolation modes with periods of 1.85, 1.68 and 1.52
sec respectively. The mass participation factors in
Table B2.6-1 gives the following information about
these three modes:
The first mode (1.85 sec) is a strongly coupled mode
in the transverse and torsional directions (rotation
337

 about z-axis) due to a lack of symmetry in the column

stiffness in this direction.

The second mode is the longitudinal mode and its

period (1.68 sec) is very close to that calculated in the
Simplified Method (1.64 sec). It is not coupled with a
torsional mode because the column stiffness is
symmetric in this direction, and a single degree-of-
freedom model (as assumed in the Simplified
Method) should give good answers.

The third mode is also a coupled mode (transverse

and torsional) but not as strongly coupled as the first
mode.

It follows that in the longitudinal direction spectral

modal analysis should give similar answers to the
Simplified Method and converge quickly to a final
solution. But the same might not be true for loading in
the transverse direction (See Step C).

Table B2.6-1 Modal Properties of Bridge

Example 2.5 First Iteration
Mode Period MassParticipatingRatios
No Sec UX UY UZ RX RY RZ
1 1.848 0.000 0.838 0.000 0.863 0.000 0.911
2 1.681 0.950 0.000 0.000 0.000 0.020 0.000
3 1.516 0.000 0.088 0.000 0.091 0.000 0.037
4 0.467 0.000 0.001 0.000 0.013 0.000 0.002
5 0.400 0.005 0.000 0.003 0.000 0.002 0.000
6 0.372 0.000 0.001 0.000 0.000 0.000 0.003
7 0.371 0.000 0.000 0.071 0.000 0.054 0.000
8 0.343 0.000 0.002 0.000 0.012 0.000 0.000
9 0.286 0.000 0.008 0.000 0.009 0.000 0.006
10 0.261 0.043 0.000 0.000 0.000 0.001 0.000
11 0.223 0.000 0.015 0.000 0.003 0.000 0.023
12 0.208 0.000 0.000 0.001 0.000 0.143 0.000
Computed values for the isolator displacements due to
a longitudinal earthquake are as follows (numbers in
parentheses are those used to calculate the initial
properties to start iteration from the Simplified
Method):

o disol,1 = 2.10 (1.5) in

o disol,2 = 1.62 (1.73) in
o disol,3 = 0.53 (0.72) in
o disol,4 = 2.10 (1.95) in

B2.7 Convergence Check B2.7 Convergence Check, Example 2.5

Compare the resulting displacements at the The results for isolator displacements are not close.
superstructure level (d) to the assumed displacements.
These displacements can be obtained by examining Go to Step B2.8 and update properties for a second
the joints at the top of the isolator spring elements. If cycle of iteration.
in close agreement, go to Step B2.9. Otherwise go to
Step B2.8.

338

B2.8 Update Kisol,i, Keff,j, and BL B2.8 Update Kisol,i, Keff,j, and BL, Example 2.5
Use the calculated displacements in each isolator Updated values for Kisol,i are given below (previous
element to obtain new values of Kisol,i for each isolator values are in parentheses):
as follows:
, o Kisol,1 = 2.73(2.88) k/in
, , (B-24) o Kisol,2 = 12.38 (11.93) k/in
,
o Kisol,3 = 26.28 (20.83) k/in
Recalculate Keff,j : o Kisol,4 = 2.78 (2.88) k/in

Eq. , ,
7.1-6 ,
(B-25) Since the isolator displacements are relatively close to
, ,
GSID previous results no significant change in the damping
ratio is expected. Hence Keff,j and are not
Recalculate system damping ratio, : recalculated and BL is taken at 1.65.

Eq. 2 , , ,
7.1-10 , , ,
(B-26)
GSID

Recalculate system damping factor, BL:

Eq. .
0.3
.
7.1-3 ( B-27)
1.7 0.3
GSID

Obtain the effective period of the bridge from the
multi-modal analysis and with the revised damping
factor (Eq. B-27), construct a new composite response
spectrum. Go to Step B2.6.
Since the change in effective period is very small
(1.64 to 1.68 sec) and no change has been made to BL,
there is no need to construct a new composite
response spectrum in this case. Go back to Step B2.6
(see immediately below).

B2.6 Multimodal Analysis Second and Third

Iteration, Example 2.5
Reasonable convergence was not obtained after the
second iteration and a third cycle was performed.
Results for the isolator displacements at the end of
the third cycle (numbers in parentheses are those from
the first cycle):

o disol,1 = 1.96 (2.10) in

o disol,2 = 1.56 (1.62) in
o disol,3 = 0.33 (0.53) in
o disol,4 = 1.96 (2.10) in

Go to Step B2.7

B2.7 Convergence Check B2.7 Convergence Check, Example 2.5

Compare results and determine if convergence has Satisfactory agreement has been reached on the third
been reached. If so go to Step B2.9. Otherwise Go to cycle. Go to Step B2.9
Step B2.8.

B2.9 Superstructure and Isolator Displacements B2.9 Superstructure and Isolator Displacements,
Once convergence has been reached, obtain Example 2.5
o superstructure displacements in the longitudinal From the above analysis:
(xL) and transverse (yL) directions of the bridge, o superstructure displacements in the

339

 and longitudinal (xL) and transverse (yL) directions

o isolator displacements in the longitudinal (uL) are:
and transverse (vL) directions of the bridge, for xL= 1.98 in
each isolator, for this load case (i.e. yL= 0.0 in
longitudinal loading). These displacements
may be found by subtracting the nodal o isolator displacements in the longitudinal (uL)
displacements at each end of each isolator and transverse (vL) directions are:
spring element. o Abutment 1: uL = 1.96 in, vL = 0.00 in
o Pier 1: uL = 1.56 in, vL = 0.00 in
o Pier 2: uL = 0.33 in, vL = 0.00 in
o Abutment 2: uL = 1.96 in, vL = 0.00 in

All isolators at same support have the same

displacements.

B2.10 Pier Bending Moments and Shear Forces B2.10 Pier Bending Moments and Shear Forces,
Obtain the pier bending moments and shear forces in Example 2.5
the longitudinal (MPLL, VPLL) and transverse (MPTL, Bending moments in single column pier in the
VPTL) directions at the critical locations for the longitudinal (MPLL) and transverse (MPTL) directions
longitudinally-applied seismic loading. are:
Pier 1: MP1LL= 0
MP1TL= 2020 kft
Pier 2: MP2LL= 0
MP2TL= 1789 kft

Shear forces in single column pier the longitudinal

(VPLL) and transverse (VPTL) directions are:
Pier 1: VP1LL = 84.79 k
VP1TL = 0
Pier 2: VP2LL = 41.69 k
VP2TL = 0

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 2.5
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1

340

 Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Long. Transv. Axial
shear shear forces
due to due to due to
long. long. long.
EQ EQ EQ
Isol.1 5.63 0 1.31
Abut 1 Isol.2 5.63 0 1.33
Isol.3 5.63 0 1.33
Isol.1 19.62 0 0.99
Pier 1 Isol.2 19.65 0 1.36
Isol.3 19.62 0 0.99
Isol.1 11.61 0 0.84
Pier 2 Isol.2 11.67 0 1.14
Isol.3 11.61 0 0.84
Isol.1 5.63 0 1.04
Abut 2 Isol.2 5.63 0 1.05
Isol.3 5.63 0 1.02

341

STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake C1. Analysis for Transverse Earthquake, Example
Repeat the above process, starting at Step B1, for 2.5
earthquake loading in the transverse direction of the Key results from repeating Steps B1 and B2
bridge. Support flexibility in the transverse direction (Simplified and Mulitmode Spectral Methods) for
is to be included, and a composite response spectrum transverse loading , are as follows:
is to be applied in the transverse direction. Obtain
isolator displacements in the longitudinal (uT) and o Teff = 1.86 sec
transverse (vT) directions of the bridge, and the biaxial o Superstructure displacements in the longitudinal
bending moments and shear forces at critical locations (xT) and transverse (yT) directions due to
in the columns due to the transversely-applied seismic transverse load are as follows:
loading. xT = 0
yT = 1.81 in
o Isolator displacements in the longitudinal (uT)
and transverse (vT) directions due to transverse
loading are as follows:
o
Abutment1 uT = 0.15 in, vT = 2.91 in
Pier1 uT = 0.14 in, vT = 1.04 in
Pier2 uT = 0.08 in, vT = 0.22 in
Abutment2 uT = 0.14 in, vT = 3.40 in

o Pier bending moments in the longitudinal (MPLT)

and transverse (MPTT) directions due to
transverse load are as follows:
Pier 1: MPLT = 1906 kft
MPTT = 2 kft
Pier 2: MPLT = 1722 kft
MPTT = 1 kft

o Pier shear forces in the longitudinal (VPLT) and

transverse (VPTT) directions due to transverse
load are as follows:
Pier 1: VPLT = 0.33k
VPTT = 70.13 k
Pier 2: VPLT = 0.09k
VPTT = 45.85 k

o Isolator shear and axial forces are summarized in

Table C1-1.

342

 Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Long. Transv. Axial
shear shear forces
due to due to due to
transv. transv. transv.
EQ EQ EQ
Isol. 1 0.45 8.84 15.36
Abut 1 Isol. 2 0 8.85 1.31
Isol. 3 0.45 8.84 15.27
Isol. 1 2.61 19.45 26.77
Pier 1 Isol. 2 0 19.63 1.58
Isol. 3 2.61 19.45 26.87
Isol.1 4.18 11.55 26.32
Pier 2 Isol.2 0 11.82 1.71
Isol.3 4.18 11.55 26.58
Isol.1 0.32 7.70 16.26
Abut 2 Isol.2 0 7.71 1.78
Isol.3 0.32 7.70 16.44

343

STEP D. CALCULATE DESIGN VALUES (Exterior Isolator at Pier 1)

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements, Example 2.5
Following the provisions in Art. 2.1 GSID, and Load Case 1:
illustrated in Fig. 2.1-1 GSID, calculate the total u1 = uL + 0.3uT = 1.0(1.57) + 0.3(0.14) = 1.61 in
design displacement, dt, by combining the v1 = vL + 0.3vT = 1.0(0) + 0.3(1.04) = 0.31 in
displacements from the longitudinal (uL and vL) and R1 = = 1.61 0.31 = 1.64 in
transverse (uT and vT) cases as follows:
u1 = uL + 0.3uT (D-1) Load Case 2:
v1 = vL + 0.3vT (D-2) u2 = 0.3uL + uT = 0.3(1.57) + 1.0(0.14) = 0.61 in
R1 = (D-3) v2 = 0.3vL + vT = 0.3(0) + 1.0(1.04) = 1.04 in
R2 = = 0.61 1.04 = 1.21 in
u2 = 0.3uL + uT (D-4)
v2 = 0.3vL + vT (D-5)
R2 = (D-6)
Governing Case:
dt = max(R1, R2) (D-7) Total design displacement, dt = max(R1, R2)
= 1.64in
This is the design displacement for an exterior isolator
at Pier 1.

D2. Design Moments and Shears D2. Design Moments and Shears (Pier 1), Example
Calculate design values for column bending moments 2.5
and shear forces using the same combination rules as Load Case 1:
for displacements. VPL1= VPLL + 0.3VPLT = 1.0(19.62) + 0.3(2.61) = 20.40
k
Alternatively this step may be deferred because the VPT1= VPTL + 0.3VPTT = 1.0(0) + 0.3(19.45) = 5.84 k
above results may not be final. Upper and lower R1 = = 20.40 5.84 = 21.22 k
bound analyses are required after the isolators have
been designed as described in Art 7. GSID. These Load Case 2:
analyses are required to determine the effect of VPL2= 0.3VPLL + VPLT = 0.3(19.62) + 1.0(2.61) = 8.5 k
possible variations in isolator properties due age, VPT2= 0.3VPTL + VPTT = 0.3(0) + 1.0(19.45) = 19.45 k
temperature and scragging in elastomeric systems.
R2 = = 8.5 19.45 = 21.22 k
Accordingly the results for column shear in Steps
B2.10 and C are likely to increase once these analyses
Governing Case:
are complete.
Design column shear = max (R1, R2)
= 21.22 k

344

STEP E. DESIGN OF LEAD-RUBBER (ELASTOMERIC) ISOLATORS

A lead-rubber isolator is an elastomeric
bearing with a lead core inserted on its vertical
centreline. When the bearing and lead core are
deformed in shear, the elastic stiffness of the
lead provides the initial stiffness (Ku).With
increasing lateral load the lead yields almost
perfectly plastically, and the post-yield
stiffness Kd is given by the rubber alone. More
details are given in MCEER 2006.

While both circular and rectangular bearings

are commercially available, circular bearings
are more commonly used. Consequently the
procedure given below focuses on circular
bearings. The same steps can be followed for
rectangular bearings, but some modifications will be necessary. When sizing the physical dimensions of the
bearing, plan dimensions (B, dL) should be rounded up to the next 1/4 increment, while the total thickness of
elastomer, Tr, is specified in multiples of the layer thickness. Typical layer thicknesses for bearings with lead
cores are 1/4 and 3/8. High quality natural rubber should be specified for the elastomer. It should have a shear
modulus in the range 60-120 psi and an ultimate elongation-at-break in excess of 5.5. Details can be found in
rubber handbooks or in MCEER 2006.

The following design procedure assumes the isolators are bolted to the masonry and sole plates. Isolators that use
shear-only connections (and not bolts) require additional design checks for stability which are not included below.
See MCEER 2006.

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

E1. Required Properties E1. Required Properties, Example 2.5

Obtain from previous work the properties required of The design of one of the exterior isolators on a pier is
the isolation system to achieve the specified given below to illustrate the design process for lead-
performance criteria (Step A1). rubber isolators.
required characteristic strength, Qd / isolator
required post-elastic stiffness, Kd / isolator From previous work
total design displacement, dt, for each isolator Qd / isolator = 10.95 k
maximum applied dead and live load (PDL, PLL) Kd / isolator = 5.62 k/in
and seismic load (PSL) which includes seismic Total design displacement dt = 1.64 in
live load (if any) and overturning forces due to PDL = 187 k
seismic loads, at each isolator, and PLL = 123 k
maximum wind load, PWL PSL = 27 k (Table C1-1)
PWL = 8.21 k < Qd OK

E2. Isolator Sizing

E2.1 Lead Core Diameter E2.1 Lead Core Diameter, Example 2.5
Determine the required diameter of the lead plug, dL,
using:
10.95
(E-1) 3.49
0.9 0.9 0.9
See Step E2.5 for limitations on dL

345

E2.2 Plan Area and Isolator Diameter
Although no limits are placed on compressive stress in
the GSID, (maximum strain criteria are used instead,
see Step E3) it is useful to begin the sizing process by
assuming an allowable stress of, say, 1.6 ksi.
E2.2 Plan Area and Isolator Diameter, Example
2.5
Based on the final design of the isolators for Example
2.0, increase the allowable stress to 3.2 ksi.

Then the bonded area of the isolator is given by:

187 123
(E-2) 96.88
1.6 3.2 3.2

and the corresponding bonded diameter (taking into

account the hole required to accommodate the lead
core) is given by:
4 4 4 96.88
(E-3) 3.49

= 11.64 in
Round the bonded diameter, B, to nearest quarter inch,
and recalculate actual bonded area using Round B up to 12.5 in (based on experience with
Example 2.0) and the actual bonded area is:
(E-4) 12.50 3.49 113.16
4 4

Note that the overall diameter is equal to the bonded

diameter plus the thickness of the side cover layers
(usually 1/2 inch each side). In this case the overall
diameter, Bo is given by:

1.0 (E-5) Bo = 12.50 + 2(0.5) = 13.50 in

E2.3 Elastomer Thickness and Number of Layers E2.3 Elastomer Thickness and Number of Layers,
Since the shear stiffness of the elastomeric bearing is Example 2.5
given by:
(E-6)
where G = shear modulus of the rubber, and
Tr = the total thickness of elastomer,
it follows Eq. E-5 may be used to obtain Tr given a
required value for Kd
0.1 113.16
(E-7) 2.01
5.62

If the layer thickness is tr, the number of layers, n, is

given by:
2.01
(E-8) 8.05
0.25
rounded up to the nearest integer.
Round to nearest integer, i.e. n = 8
Note that because of rounding the plan dimensions
and the number of layers, the actual stiffness, Kd, will
not be exactly as required. Reanalysis may be
necessary if the differences are large.

346

E2.4 Overall Height E2.4 Overall Height, Example 2.5

The overall height of the isolator, H, is given by:

1 2 ( E-9) 8 0.25 7 0.125 2 1.5 5.875

where ts = thickness of an internal shim (usually

about 1/8 in), and
tc = combined thickness of end cover plate (0.5
in) and outer plate (1.0 in)

E2.5 Lead Core Size Check E2.5 Lead Core Size Check, Example 2.5
Experience has shown that for optimum performance Since B=12.5 check
of the lead core it must not be too small or too large.
The recommended range for the diameter is as 12.5 12.5
follows: 3 6
(E-10)
3 6 i.e., 4.16 2.08

Since dL = 3.49, lead core size is acceptable.

E3. Strain Limit Check E3. Strain Limit Check, Example 2.5
Art. 14.2 and 14.3 GSID requires that the total applied Since
shear strain from all sources in a single layer of 187.0
elastomer should not exceed 5.5, i.e., 1.65
113.16

, + 0.5 5.5 (E-11) and

113.16
11.53
where , , , are defined below. 12.5 0.25
(a) is the maximum shear strain in the layer due to
compression and is given by: then
1.0 1.65
(E-12) 1.43
0.1 11.53
where Dc is shape coefficient for compression in
circular bearings = 1.0, , G is shear
modulus, and S is the layer shape factor given by:
(E-13) 1.64
, 0.82
2.0
(b) , is the shear strain due to earthquake loads and
is given by:

, (E-14)
0.375 12.5 0.01
1.17
0.25 2.0
(c) is the shear strain due to rotation and is given
by: Substitution in Eq E-11 gives
(E-15)
, 0.5 1.43 0.82 0.5 1.17
where Dr is shape coefficient for rotation in circular 2.84
bearings = 0.375, and is design rotation due to DL, 5.5
LL and construction effects . Actual value for may
not be known at this time and a value of 0.01 is
suggested as an interim measure, including
uncertainties (see LRFD Art. 14.4.2.1).

347

E4. Vertical Load Stability Check E4. Vertical Load Stability Check, Example 2.5
Art 12.3 GSID requires the vertical load capacity of
all isolators be at least 3 times the applied vertical
loads (DL and LL) in the laterally undeformed state.

Further, the isolation system shall be stable under

1.2(DL+SL) at a horizontal displacement equal to
either
2 x total design displacement, dt, if in Seismic Zone 1
or 2, or
1.5 x total design displacement, dt, if in Seismic Zone
3 or 4.

E4.1 Vertical Load Stability in Undeformed State E4.1 Vertical Load Stability in Undeformed State,
The critical load capacity of an elastomeric isolator at Example 2.5
zero shear displacement is given by

3 3 0.1 0.3
4
1 1
2 (E-16) 0.3 1 0.67 11.53 26.89

where 12.50
1198.4
64
Ts = total shim thickness
26.89 1198.4
16,110 /
2.0
1 0.67
E = elastic modulus of elastomer = 3G 0.1 113.16
5.66 /
2.0
64

It is noted that typical elastomeric isolators have high

shape factors, S, in which case:
4
1 (E-17)

and Eq. E-16 reduces to:

(E-18) 5.66 16,110 948.5

Check that:

3 (E-19) 948.5
3.06 3
187 123

E4.2 Vertical Load Stability in Deformed State E4.2 Vertical Load Stability in Deformed State,
The critical load capacity of an elastomeric isolator at Example 2.5
shear displacement may be approximated by: Since bridge is in Zone 2, 2 2 1.64 3.28

(E-20) 3.28
2 2.61
where 12.50
Ar = overlap area between top and bottom plates
of isolator at displacement (Fig. 2.2-1
GSID)

348

 2.61 2.61
= 4 0.669
2

Agross = 4
0.67 948.5 635.3
It follows that:
(E-21)

635.3
Check that: 2.53 1

1.2 1.2 187 27
1 (E-22)
1.2
E5. Design Review E5. Design Review, Example 2.5
The basic dimensions of the isolator designed above
are as follows:

13.50 in (od) x 5.875in (high) x 3.49 in dia. lead core

and the volume, excluding steel end and cover plates,

= 412 in3

This design is considered satisfactory since both the

total strain (Eq E-11) and the vertical load stability
factors are reasonable values (not excessively low or
excessively high).

E6. Minimum and Maximum Performance Check
Art. 8 GSID requires the performance of any isolation
system be checked using minimum and maximum
values for the effective stiffness of the system. These
values are calculated from minimum and maximum
values of Kd and Qd, which are found using system
property modification factors, as indicated in Table
E6-1.
Table E6-1. Minimum and maximum values
for Kd and Qd.
E6. Minimum and Maximum Performance Check,
Example 2.5
Minimum Property Modification factors are
min,Kd = 1.0
min,Qd = 1.0
which means there is no need to reanalyze the bridge
with a set of minimum values.
Maximum Property Modification factors are

max,a,Kd = 1.1
Eq.
max,a,Qd = 1.1
8.1.2-1 Kd,max = Kd max,Kd (E-23)
GSID
max,t,Kd = 1.1
Eq.
max,t,Qd = 1.4
8.1.2-2 Kd,min = Kd min,Kd (E-24)
GSID
max,scrag,Kd = 1.0
Eq.
max,scrag,Qd = 1.0
8.1.2-3 Qd,max = Qd max,Qd (E-25)
GSID
Eq.
8.1.2-4 Qd,min = Qd min,Qd (E-26)
GSID

Determination of the system property modification

349

factors shall include consideration of the effects of

temperature, aging, scragging, velocity, travel (wear)
and contamination as shown in Table E6-2. In lieu of
tests, numerical values for these factors can be
obtained from Appendix A, GSID.

Table E6-2. Minimum and maximum values for

system property modification factors.

Eq. min,Kd = (min,t,Kd) (min,a,Kd)

8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-27)
GSID (min,scrag,Kd)
Eq. max,Kd = (max,t,Kd) (max,a,Kd)
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-28)
GSID (max,scrag,Kd)
Eq. min,Qd = (min,t,Qd) (min,a,Qd)
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-29)
GSID (min,scrag,Qd)
Eq. max,Qd = (max,t,Qd) (max,a,Qd)
8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd) (E-30)
GSID (max,scrag,Qd)

Adjustment factors are applied to individual -factors
(except v) to account for the likelihood of occurrence
of all of the maxima (or all of the minima) at the same
time. These factors are applied to all -factors that
deviate from unity but only to the portion of the -
factor that is greater than, or less than, unity. Art.
8.2.2 GSID gives these factors as follows:
1.00 for critical bridges
0.75 for essential bridges
0.66 for all other bridges
As required in Art. 7 GSID and shown in Fig. C7-1
GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Qd,max. As indicated in Fig C7-1 GSID, maximum
displacements will probably be given by the first case
(Kd,min and Qd,min) and maximum forces by the second
case (Kd,max and Qd,max).
Applying a system adjustment factor of 0.66 for an
other bridge, the maximum property modification
factors become:
max,a,Kd = 1.0 + 0.1(0.66) = 1.066
max,a,Qd = 1.0 + 0.1(0.66) = 1.066
max,t,Kd = 1.0 + 0.1(0.66) = 1.066
max,t,Qd = 1.0 + 0.4(0.66) = 1.264
max,scrag,Kd = 1.0
max,scrag,Qd = 1.0
Therefore the maximum overall modification factors
max,Kd = 1.066(1.066)1.0 = 1.14
max,Qd = 1.066(1.264)1.0 = 1.35
Since the possible variation in upper bound properties
exceeds 15% a reanalysis of the bridge is required to
determine performance with these properties.

The upper-bound properties are:

Qd,max = 1.35 (10.95) = 14.78 k
and
Kd,ma x=1.14(5.62) = 6.41 k/in

350

E7. Design and Performance Summary E7. Design and Performance Summary
E7.1 Isolator dimensions E7.1 Isolator dimensions, Example 2.5
Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E7.1-1.
Overall diameter (includes cover layer)
Overall height Table E7.1-1 Isolator Dimensions
Diameter lead core Isolator Overall size Overall size Diam.
Bonded diameter Location including without lead
Number of rubber layers mounting mounting core
Thickness of rubber layers plates. (in) plates (in) (in)
Total rubber thickness Under
Thickness of steel shims edge 17.5 x17.5 x 13.5 dia
3.49
Shear modulus of elastomer girder on 5.875(H) x 4.375(H)
Pier
Check all dimensions with manufacturer. Isolator No. of Rubber Total Steel
Location rubber layers rubber shim
layers thick- thick- thick-
ness ness ness
(in) (in) (in)
Under
edge
8 0.25 2.0 0.125
girder
on Pier

Shear modulus of elastomer = 100 psi

E7.2 Bridge Performance E7.2 Bridge Performance, Example 2.5

Summarize bridge performance Bridge performance is summarized in Table E7.2-1
Maximum superstructure displacement where it is seen that the maximum column shear is
(longitudinal) 87.36k (Pier1) and 47.53k (Pier2). This less than the
Maximum superstructure displacement column plastic shear (128k at Pier 1 and 72k at Pier2)
(transverse) and therefore the required performance criterion is
Maximum superstructure displacement satisfied (fully elastic behavior). Furthermore the
(resultant) maximum longitudinal displacement is 1.98 in which
Maximum column shear (resultant) is less than the 2.5in which is available at the
Maximum column moment (about transverse abutment expansion joints and therefore acceptable.
axis)
Maximum column moment (about longitudinal Table E7.2-1 Summary of Bridge Performance
axis)
Maximum column torque Maximum superstructure
1.98 in
displacement (longitudinal)
Check required performance as determined in Step Maximum superstructure
1.81 in
A3, is satisfied. displacement (transverse)
Maximum superstructure
2.05 in
displacement (resultant)
Maximum column shear 87.36k (Pier1)
(resultant) 47.53k (Pier2)
Maximum column moment 2020 kft (Pier1)
about transverse axis 1789 kft (Pier2)
Maximum column moment 1906 kft (Pier1)
about longitudinal axis 1722 kft (Pier2)
72 kft (Pier1)
Maximum column torque
99 kft (Pier2)

351

 SECTION 2
Steel Plate Girder Bridge, Long Spans, Single-Column Pier

DESIGN EXAMPLE 2.6: Skew = 450

Design Examples in Section 2

Site
ID Description S1 Column height Skew Isolator type
Class
2.0 Benchmark bridge 0.2g B Same 0 Lead-rubber bearing
2.1 Change site class 0.2g D Same 0 Lead rubber bearing
2.2 Change spectral acceleration, S1 0.6g B Same 0 Lead rubber bearing
Spherical friction
2.3 Change isolator to SFB 0.2g B Same 0
bearing
2.4 Change isolator to EQS 0.2g B Same 0 Eradiquake bearing
2.5 Change column height 0.2g B H1=0.5H2 0 Lead rubber bearing
0
2.6 Change angle of skew 0.2g B Same 45 Lead rubber bearing

352

DESIGN PROCEDURE
STEP A: BRIDGE AND SITE DATA
A1. Bridge Properties
Determine properties of the bridge:
Number of supports, m
Number of girders per support, n
Angle of skew
Weight of superstructure including
railings, curbs, barriers and other
permanent loads, WSS
Weight of piers participating with
superstructure in dynamic response, WPP
Weight of superstructure, Wj, at each
support
Stiffness, Ksub,j, of each support in both
longitudinal and transverse directions of
the bridge. The calculation of these
quantities requires careful consideration of
several factors such as the use of cracked
sections when estimating column or wall
flexural stiffness, foundation flexibility,
and effective column height.
Column shear strength (minimum value).
This will usually be derived from the
minimum value of the column flexural
yield strength, the column height, and
whether the column is acting in single or
double curvature in the direction under
consideration.
Allowable movement at expansion joints
Isolator type if known, otherwise to be
selected
A2. Seismic Hazard
Determine seismic hazard at site:
acceleration coefficients
site class and site factors
seismic zone
Plot response spectrum.
Use Art. 3.1 GSID to obtain peak ground and spectral
acceleration coefficients. These coefficients are the
same as for conventional bridges and Art 3.1 refers the
designer to the corresponding articles in the LRFD
Specifications. Mapped values of PGA, SS and S1 are
given in both printed and CD formats (e.g. Figures
3.10.2.1-1 to 3.10.2.1-21 LRFD).
Use Art. 3.2 to obtain Site Class and corresponding Site
Factors (Fpga, Fa and Fv). These data are the same as for
conventional bridges and Art 3.2 refers the designer to
the corresponding articles in the LRFD Specifications,
DESIGN EXAMPLE 2.6 (450 skew)
A1. Bridge Properties, Example 2.6
Number of supports, m = 4
o North Abutment (m = 1)
o Pier 1 (m = 2)
o Pier 2 (m = 3)
o South Abutment (m =4)
Number of girders per support, n = 3
Number of columns per support = 1
Angle of skew = 450
Weight of superstructure including permanent
loads, WSS = 1651.3 k
Weight of superstructure at each support:
o W1 = 168.48 k
o W2 = 657.18 k
o W3 = 657.18 k
o W4 = 168.48 k
Participating weight of piers, WPP = 256.3 k
Effective weight (for calculation of period),
Weff = Wss + WPP = 1907.6 k
Stiffness of each pier in the both directions:
o Ksub,pier1 = 288.87 k/in
o Ksub,pier2 = 288.87 k/in
Minimum column shear strength based on
flexural yield capacity of column = 128 k
Displacement capacity of expansion joints
(longitudinal) = 2.5 in (required to accommodate
thermal expansion and other movements)
Lead rubber isolators
A2. Seismic Hazard, Example 2.6
Acceleration coefficients for bridge site are given in
design statement as follows:
PGA = 0.40
S1 = 0.20
SS = 0.75
Bridge is on a rock site with shear wave velocity in
upper 100 ft of soil = 3,000 ft/sec.
Table 3.10.3.1-1 LRFD gives Site Class as B.
Tables 3.10.3.2-1, -2 and -3 LRFD give following
Site Factors:
Fpga = 1.0
Fa = 1.0
353

i.e. to Tables 3.10.3.1-1 and 3.10.3.2-1, -2, and -3, Fv = 1.0

LRFD.
Art. 4 GSID and Eq. 4-2, -3, and -8 GSID give
modified spectral acceleration coefficients that include
site effects as follows:
As = Fpga PGA As = Fpga PGA = 1.0(0.40) = 0.40
SDS = Fa SS SDS = Fa SS = 1.0(0.75) = 0.75
SD1 = Fv S1 SD1 = Fv S1 = 1.0(0.20) = 0.20

Seismic Zone is determined by value of SD1 in Since 0.15 < SD1 < 0.30, bridge is located in Seismic
accordance with provisions in Table 5-1 GSID. Zone 2.

These coefficients are used to plot design response Design Response Spectrum is as below:
spectrum as shown in Fig. 4-1 GSID.
0.9

0.8

0.7

0.6

Acceleration
0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Period(s)

A3. Performance Requirements A3. Performance Requirements, Example 2.6

Determine required performance of isolated bridge In this example, assume the owner has specified full
during Design Earthquake (1000-yr return period). functionality following the earthquake and therefore
the columns must remain elastic (no yield).
Examples of performance that might be specified by
the Owner include: To remain elastic the maximum lateral load on the
Reduced displacement ductility demand in pier must be less than the load to yield any one
columns, so that bridge is open for emergency column (128 k).
vehicles immediately following earthquake.
Fully elastic response (i.e., no ductility demand in The maximum shear in the column must therefore be
columns or yield elsewhere in bridge), so that less than 128 k in order to keep the column elastic
bridge is fully functional and open to all vehicles and meet the required performance criterion.
immediately following earthquake.
For an existing bridge, minimal or zero ductility
demand in the columns and no impact at
abutments (i.e., longitudinal displacement less
than existing capacity of expansion joint for
thermal and other movements)
Reduced substructure forces for bridges on weak
soils to reduce foundation costs.

354

STEP B: ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN LONGITUDINAL

DIRECTION

In most applications, isolation systems must be stiff for

Isolator Force, F
non-seismic loads but flexible for earthquake loads (to
enable required period shift). As a consequence most
have bilinear properties as shown in figure at right. Fy Fisol Kisol
Qd Kd
Strictly speaking nonlinear methods should be used for
their analysis. But a common approach is to use Ku
equivalent linear springs and viscous damping to
represent the isolators, so that linear methods of analysis
may be used to determine response. Since equivalent Ku
dy disol
Isolator
Ku Displacement, d
properties such as Kisol are dependent on displacement
(d), and the displacements are not known at the
beginning of the analysis, an iterative approach is Kd
required. Note that in Art 7.1, GSID, keff is used for the
effective stiffness of an isolator unit and Keff is used for disol = Isolator displacement
the effective stiffness of a combined isolator and dy = Isolator yield displacement
Fisol = Isolator shear force
substructure unit. To minimize confusion, Kisol is used in Fy = Isolator yield force
Kd = Post-elastic stiffness of isolator
this document in place of keff. There is no change in the Kisol = Effective stiffness of isolator
use of Keff and Keff,j, but Ksub is used in place of ksub. Ku = Loading and unloading stiffness (elastic stiffness)
Qd = Characteristic strength of isolator

The methodology below uses the Simplified Method (Art 7.1 GSID) to obtain initial estimates of displacement for
use in an iterative solution involving the Multimode Spectral Analysis Method (Art 7.3 GSID).

Alternatively nonlinear time history analyses may be used which explicitly include the nonlinear properties of the
isolator without iteration, but these methods are outside the scope of the present work.

B1. SIMPLIFIED METHOD

In the Simplified Method (Art. 7.1, GSID) a single degree-of-freedom model of the bridge with equivalent linear
properties and viscous dampers to represent the isolators, is analyzed iteratively to obtain estimates of
superstructure displacement (disol in above figure, replaced by d below to include substructure displacements) and
the required properties of each isolator necessary to give the specified performance (i.e. find d, characteristic
strength, Qd,j, and post elastic stiffness, Kd,j for each isolator j such that the performance is satisfied). For this
analysis the design response spectrum (Step A2 above) is applied in longitudinal direction of bridge.

B1.1 Initial System Displacement and Properties B1.1 Initial System Displacement and Properties,
To begin the iterative solution, an estimate is required Example 2.6
of :
(1) Structure displacement, d. One way to make
this estimate is to assume the effective
isolation period, Teff, is 1.0 second, take the
viscous damping ratio, , to be 5% and
calculate the displacement using Eq. B-1.
(The damping factor, BL, is given by Eq.7.1-3
GSID, and equals 1.0 in this case.)

Art
9.79
C7.1 10 (B-1) 10 10 0.20 2.0
GSID

355

 (2) Characteristic strength, Qd. This strength

needs to be high enough that yield does not
occur under service loads (e.g. wind) but low
enough that yield does occur under
earthquake. Experience has shown that
taking Qd to be 5% of the bridge weight is a 0.05 0.05 1651.32 82.56
good starting point, i.e.
0.05 (B-2)
(3) Post-yield stiffness, Kd
Art 12.2 GSID requires that all isolators
exhibit a minimum lateral restoring force at
the design displacement, which translates to a
minimum post yield stiffness Kd,min given by
Eq. B-3.
1651.32
Art. 0.05 0.05 41.28 /
0.025 2.0
12.2 , (B-3)
GSID
Experience has shown that a good starting
point is to take Kd equal to twice this
minimum value, i.e. Kd = 0.05W/d

B1.2 Initial Isolator Properties at Supports B1.2 Initial Isolator Properties at Supports,
Calculate the characteristic strength, Qd,j, and post- Example 2.6
elastic stiffness, Kd,j, of the isolation system at each
support j by distributing the total calculated strength, ,

Qd, and stiffness, Kd, values in proportion to the dead o Qd, 1 = 8.42 k
load applied at that support: o Qd, 2 = 32.86 k
o Qd, 3 = 32.86 k
o Qd, 4 = 8.42 k

, (B-4)
and
and
,
, (B-5) o Kd,1 = 4.21 k/in
o Kd,2 = 16.43 k/in
o Kd,3 = 16.43 k/in
o Kd,4 = 4.21 k/in

B1.3 Effective Stiffness of Combined Pier and B1.3 Effective Stiffness of Combined Pier and
Isolator System Isolator System, Example 2.6
Calculate the effective stiffness, Keff,j, of each support
j for all supports, taking into account the stiffness of
the isolators at support j (Kisol,j) and the stiffness of
the substructure Ksub,j. See figure below (after Fig.
7.1-1 GSID). , ,

, ,
An expression for Keff,j, is given in Eq.7.1-6 GSID, but
a more useful formula as follows (MCEER 2006): o 1 = 8.43x10-4
,
(B-6) o 2 = 1.21x10-1
,
1 o 3 = 1.21x10-1
where o 4 = 8.43x10-4
, ,
(B-7)
, ,
and Ksub,j for the piers are given in Step A1.

356

 F ,
Kd ,
1
Qd Kisol

dy disol
Superstructure o Keff,1 = 8.42 k/in
Isolator Effective Stiffness, Kisol
o Keff,2 = 31.09 k/in
F o Keff,3 = 31.09 k/in
Isolator(s), Kisol
Ksub o Keff,4 = 8.42 k/in

Substructure, Ksub
dsub

Substructure Stiffness, Ksub

dsub disol F

d
Keff

d = disol + dsub

Combined Effective Stiffness, Keff

For abutments, take Ksub,j to be a large number, say

10,000 k/in, unless an actual stiffness values are
available. Note that if the default option is chosen,
unrealistically high values for Ksub,j will give
unconservative results for column moments and shear
forces.

B1.4 Total Effective Stiffness B1.4 Total Effective Stiffness, Example 2.6
Calculate the total effective stiffness, Keff, of the
bridge:

Eq. , 79.02 /
7.1-6 , ( B-8)
GSID

B1.5 Isolation System Displacement at Each B1.5 Isolation System Displacement at Each
Support Support, Example 2.6
Calculate the displacement of the isolation system at
support j, disol,j, for all supports:
,
1

(B-9) o disol,1 = 2.00 in

,
1 o disol,2 = 1.79 in
o disol,3 = 1.79 in
o disol,4 = 2.00 in

B1.6 Isolation System Stiffness at Each Support B1.6 Isolation System Stiffness at Each Support,
Calculate the stiffness of the isolation system at Example 2.6
support j, Kisol,j, for all supports:
,
, ,
,
,
(B-10) o Kisol,1 = 8.43 k/in
, ,
, o Kisol,2 = 34.84 k/in
o Kisol,3 = 34.84 k/in
o Kisol,4 = 8.43 k/in

357

B1.7 Substructure Displacement at Each Support B1.7 Substructure Displacement at Each Support,
Calculate the displacement of substructure j, dsub,j, Example 2.6
for all supports:
, ,

, , (B-11) o dsub,1 = 0.002 in

o d sub,2 = 0.215 in
o d sub,3 = 0.215 in
o d sub,4 = 0.002 in

B1.8 Substructure Shear at Each Support B1.8 Lateral Load at Each Support, Example 2.6
Calculate the shear at support j, Fsub,j, for all
supports: , , ,
, , , (B-12)
o F sub,1 = 16.84 k
where values for Ksub,j are given in Step A1. o F sub,2 = 62.18k
o F sub,3 = 62.18 k
o F sub,4 = 16.84 k

B1.9 Column Shear at Each Support B1.9 Column Shear Forces at Each Support,
Calculate the shear in column k at support j, Fcol,j,k, Example 2.6
assuming equal distribution of shear for all columns at
support j: ,
, ,
#

, o F col,2,1 = 62.18 k
, ,
# (B-13) o F col,3,1 = 62.18 k

These column shears are less than the plastic shear

Use these approximate column shears as a check on capacity of each column (128k) as required in Step
the validity of the chosen strength and stiffness A3 and the chosen strength and stiffness values in
characteristics. Step B1.1 are therefore satisfactory.

B1.10 Effective Period and Damping Ratio B1.10 Effective Period and Damping Ratio,
Calculate the effective period, Teff, and the viscous Example 2.6
damping ratio, , of the bridge:

Eq. 1907.58
7.1-5 2 (B-14) 2 2
GSID 386.4 79.02

and = 1.57 sec

Eq. 2 , , , and taking dy,j = 0:

7.1-10 (B-15)
, , ,
GSID 2 , , 0
0.30
, , ,
where dy,j is the yield displacement of the isolator and
assumed to be small compared to disol,j with negligible
effect on , i.e., take dy,j = 0.

358

B1.11 Damping Factor
Calculate the damping factor, BL, and the
displacement, d, of the bridge:
Eq. .
, 0.3
.
7.1-3 (B-16)
1.7, 0.3
GSID
Eq. 9.79 (B-17)
7.1-4
GSID
B1.12 Convergence Check
Compare the new displacement with the initial value
assumed in Step B1.1. If there is close agreement, go
to the next step; otherwise repeat the process from
Step B1.3 with the new value for displacement as the
assumed displacement.
This iterative process is amenable to solution using a
spreadsheet and usually converges in a few cycles
(less than 5).
After convergence the performance objective and the
displacement demands at the expansion joints
(abutments) should be checked. If these are not
satisfied adjust Qd and Kd (Step B1.1) and repeat. It
may take several attempts to find the right
combination of Qd and Kd. It is also possible that the
performance criteria and the displacement limits are
mutually exclusive and a solution cannot be found. In
this case a compromise will be necessary, such as
increasing the clearance at the expansion joints or
allowing limited yield in the columns, or both.
Note that Art 9 GSID requires that a minimum
clearance be provided equal to 8 SD1 Teff / BL. (B-18)
B1.11 Damping Factor, Example 2.6
Since 0.30 0.3
1.70
and
9.79 9.79 0.2 1.57
1.81
1.70
B1.12 Convergence Check, Example 2.6
Since the calculated value for displacement, d (=1.81)
is not close to that assumed at the beginning of the
cycle (Step B1.1, d = 2.0), use the value of 1.81 as the
new assumed displacement and repeat from Step
B1.3.
After three iterations, convergence is reached at a
superstructure displacement of 1.65 in, with an
effective period of 1.43 seconds, and a damping factor
of 1.7 (30% damping ratio). The displacement in the
isolators at Pier 1 is 1.44 in and the effective stiffness
of the same isolators is 42.78 k/in.
See spreadsheet in Table B1.12-1 for results of final
iteration.
Since the column shear must equal the isolator shear
for equilibrium, the column shear = 42.78 (1.44) =
61.60 k which is less than the maximum allowable
(128k) if elastic behavior is to be achieved (as
required in Step A3).
Also the superstructure displacement = 1.65 in, which
is less than the available clearance of 2.5 in.
Therefore the above solution is acceptable and go to
Step B2.
Note that available clearance (2.5 in) is greater than
minimum required
8 8 0.20 1.43
1.35
1.7
359

 Table B1.12-1 Simplified Method Solution for Design Example 2.6 Final Iteration

SIMPLIFIEDMETHODSOLUTION

StepA1,A2 W SS W PP W eff S D1 n
1651.32 256.26 1907.58 0.2 3

StepB1.1 d 1.65 Assumeddisplacement

Qd 82.57 Characteristicstrength
Kd 50.04 Postyieldstiffness

Step A1 B1.2 B1.2 A1 B1.3 B1.3 B1.5 B1.6 B1.7 B1.8 B1.10 B1.10
K eff,j (d isol,j
Wj Q d,j K d,j K sub,j j K eff,j d isol,j K isol,j d sub,j F sub,j Q d,j d isol,j +d sub,j )
2

Abut1 168.48 8.424 5.105 10,000.00 0.001022 10.206 1.648 10.216 0.002 16.839 13.885 27.785
Pier1 657.18 32.859 19.915 288.87 0.148088 37.260 1.437 42.778 0.213 61.480 47.224 101.441
Pier2 657.18 32.859 19.915 288.87 0.148088 37.260 1.437 42.778 0.213 61.480 47.224 101.441
Abut2 168.48 8.424 5.105 10,000.00 0.001022 10.206 1.648 10.216 0.002 16.839 13.885 27.785
Total 1651.32 82.566 50.040 K eff,j 94.932 156.638 122.219 258.453
Step B1.4

StepB1.10 T eff 1.43 Effectiveperiod

0.30 Equivalentviscousdampingratio

StepB1.11 B L (B15) 1.71

B L 1.70 DampingFactor
d 1.65 ComparewithStepB1.1

Step B2.1 B2.1 B2.3 B2.6 B2.8

Q d,i K d,i K isol,i d isol,i K isol,i
Abut1 2.808 1.702 3.405 1.69 3.363
Pier1 10.953 6.638 14.259 1.20 15.766
Pier2 10.953 6.638 14.259 1.20 15.766
Abut2 2.808 1.702 3.405 1.69 3.363

360

B2. MULTIMODE SPECTRAL ANALYSIS METHOD

In the Multimodal Spectral Analysis Method (Art.7.3), a 3-dimensional, multi-degree-of-freedom model of the
bridge with equivalent linear springs and viscous dampers to represent the isolators, is analyzed iteratively to
obtain final estimates of superstructure displacement and required properties of each isolator to satisfy
performance requirements (Step A3). The results from the Simplified Method (Step B1) are used to determine
initial values for the equivalent spring elements for the isolators as a starting point in the iterative process. The
design response spectrum is modified for the additional damping provided by the isolators (see Step B2.5) and then
applied in longitudinal direction of bridge.
Once convergence has been achieved, obtain the following:
longitudinal and transverse displacements (uL, vL) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

B2.1 Characteristic Strength B2.1Characteristic Strength, Example 2.6

Calculate the characteristic strength, Qd,i, and post- Dividing the results for Qd and Kd in Step B1.12 by
elastic stiffness, Kd,i, of each isolator i as follows: the number of isolators at each support (n = 3), the
, following values for Qd /isolator and Kd /isolator are
, (B-19)
obtained:
and o Qd, 1 = 8.42/3 = 2.81 k
, o Qd, 2 = 32.86/3=10.95 k
, ( B-20)
o Qd, 3 = 32.86/3 = 10.95 k
where values for Qd,j and Kd,j are obtained from the o Qd, 4 = 8.42/3 = 2.81 k
final cycle of iteration in the Simplified Method (Step
B1. 12 and
o Kd,1 = 5.10/3 = 1.70 k/in
o Kd,2 = 19.92/3 = 6.64 k/in
o Kd,3 = 19.92/3 = 6.64 k/in
o Kd,4 = 5.10/3 = 1.70 k/in

Note that the Kd values per support used above are

from the final iteration given in Table B1.12-1. These
are not the same as the initial values in Step B1.2,
because they have been adjusted from cycle to cycle,
such that the total Kd summed over all the isolators
satisfies the minimum lateral restoring force
requirement for the bridge, i.e. Kdtotal = 0.05 W/d. See
Step B1.1. Since d varies from cycle to cycle, Kd,j
varies from cycle to cycle.

B2.2 Initial Stiffness and Yield Displacement B2.2 Initial Stiffness and Yield Displacement,
Calculate the initial stiffness, Ku,i, and the yield Example 2.6
displacement, dy,i, for each isolator i as follows: For an isolator on Pier 1:
, 10 , 10 6.64 66.4 /
In the absence of isolator-specific information take
and
, 10 , ( B-21) , 10.95
and then , 0.18
, , 66.4 6.64
,
, ( B-22) As expected, the yield displacement is small
, ,
compared to the expected isolator displacement (~2
in) and will have little effect on the damping ratio (Eq
B-15). Therefore take dy,i = 0.

361

B2.3 Isolator Effective Stiffness, Kisol,i
Calculate the isolator stiffness, Kisol,i, of each isolator
i:
, o Kisol,1 = 10.22/3 = 3.41 k/in
, (B -23)
B2.4 Finite Element Model
Using computer-based structural analysis software,
create a finite element model of the bridge with the
isolators represented by spring elements. The stiffness
of each isolator element in the horizontal axes (Kx and
Ky in global coordinates, K2 and K3 in typical local
coordinates) is the Kisol value calculated in the
previous step.
B2.5 Composite Design Response Spectrum
Modify the response spectrum obtained in Step A2 to
obtain a composite response spectrum, as illustrated
in Figure C1-5 GSID. The spectrum developed in Step
A2 is for a 5% damped system. It is modified in this
step to allow for the higher damping () in the
fundamental modes of vibration introduced by the
isolators. This is done by dividing all spectral
acceleration values at periods above 0.8 x the effective
period of the bridge, Teff, by the damping factor, BL.
B2.6 Multimodal Analysis of Finite Element Model
Input the composite response spectrum as a user-
specified spectrum in the software, and define a load
case in which the spectrum is applied in the
longitudinal direction. Analyze the bridge for this
load case.
B2.3 Isolator Effective Stiffness, Kisol,i, Example 2.6
Dividing the results for Kisol (Step B1.12) among the 3
isolators at each support, the following values for Kisol
/isolator are obtained:
o Kisol,2 = 42.78/3 = 14.26 k/in
o Kisol,3 = 42.78/3 = 14.26 k/in
o Kisol,4 = 10.22/3 = 3.41 k/in
B2.4 Finite Element Model, Example 2.6
.
B.2.5 Composite Design Response Spectrum, Ex
2.6
From the final results of Simplified Method (Step
B1.12), BL = 1.70 and Teff = 1.43 sec. Hence the
transition in the composite spectrum from 5% to 30%
damping occurs at 0.8 Teff = 0.8 (1.43) = 1.14 sec.
The spectrum below is obtained from the 5%
spectrum in Step A2, by dividing all acceleration
values with periods > 1.14 sec by 1.70.
0.8
0.7
0.6
0.5
Csm (g)
0.4
0.3
0.2
0.1
0
0 0.5 1 1.5 2 2.5 3 3.5 4
T (sec)
B2.6 Multimodal Analysis of Finite Element
Model, Example 2.6
Results of the modal analysis of this bridge are
summarized in Table B2.6-1 Here the modal periods
and mass participation factors of the first 12 modes
are given. The first three modes are the principal
isolation modes with periods of 1.57, 1.39 and 1.38
sec respectively. Mode shapes corresponding to these
three modes are plotted in Figure B2.6-1. The first
362

 and second modes are seen to be coupled translational

modes whereas the third mode is a pure torsional
mode (rotation about the Z-axis). It is clear in Figure
B2.6-1 that the first mode is predominantly transverse
with some longitudinal displacement, and the second
mode is predominantly longitudinal with some
transverse displacement. This observation is
confirmed by the relative sizes of the mass
participation factors in Table B2.6-1. Because the
coupling is not strong, the results from the Simplified
Method are considered to be a good starting point for
the iterative Multimodal Analysis.

Table B2.6-1 Modal Properties of Bridge

Example 2.6 First Iteration

Mode Period MassParticipatingRatio

No Sec UX UY UZ RX RY RZ
1 1.573 0.057 0.772 0.000 0.878 0.002 0.591
2 1.385 0.791 0.062 0.000 0.056 0.018 0.048
3 1.375 0.000 0.000 0.000 0.000 0.000 0.212
4 0.522 0.001 0.007 0.000 0.012 0.001 0.006
5 0.403 0.000 0.000 0.001 0.000 0.001 0.000
6 0.372 0.000 0.000 0.064 0.000 0.048 0.000
7 0.340 0.000 0.001 0.000 0.002 0.000 0.001
8 0.303 0.000 0.000 0.000 0.000 0.000 0.000
9 0.296 0.000 0.000 0.000 0.000 0.000 0.001
10 0.285 0.064 0.061 0.000 0.024 0.000 0.047
11 0.283 0.000 0.000 0.001 0.000 0.001 0.009
12 0.255 0.001 0.009 0.000 0.013 0.007 0.007

Mode 1, T=1.573s
Transverse translational mode

UY

UX Longitudinal translational mode Mode 2, T=1.385s

RZ

In-plane rotational mode

Mode 3, T=1.375s

Figure B2.6-1 First Three Mode Shapes for Isolated Bridge with 450 Skew (Example 2.6)

363

 Computed values for the isolator displacements due to

a longitudinal earthquake are as follows:

Loc. Isol.# uL vL R
Isol.1 1.57 0.45 1.63
Abut1 Isol.2 1.56 0.46 1.63
Isol.3 1.55 0.46 1.62
Isol.1 1.23 0.37 1.29
Pier1 Isol.2 1.23 0.38 1.29
Isol.3 1.22 0.38 1.28
Isol.1 1.22 0.38 1.28
Pier2 Isol.2 1.23 0.38 1.29
Isol.3 1.23 0.37 1.29
Isol.1 1.55 0.46 1.62
Abut2 Isol.2 1.56 0.46 1.63
Isol.3 1.57 0.45 1.63

Because of coupling between modes, there is

displacement in both the longitudinal and transverse
directions even when the earthquake is applied in the
longitudinal direction only.

The resultant isolator displacements which will be

used to calculate the effective isolator stiffness are
(numbers in parentheses are those used to calculate
the initial properties to start iteration from the
Simplified Method):
o disol,1 = 1.63 (1.65) in
o disol,2 = 1.29 (1.44) in
o disol,3 = 1.29 (1.44) in
o disol,4 = 1.63 (1.65) in

B2.7 Convergence Check B2.7 Convergence Check, Example 2.6

Compare the resulting displacements at the
superstructure level (d) to the assumed displacements.
These displacements can be obtained by examining
the joints at the top of the isolator spring elements. If
in close agreement, go to Step B2.9. Otherwise go to
Step B2.8.
The results for isolator displacements are close but
not close enough (10% difference at the piers)
Go to Step B2.8 and update properties for a second
cycle of iteration.

B2.8 Update Kisol,i, Keff,j, and BL B2.8 Update Kisol,i, Keff,j, and BL, Example 2.6
Use the calculated displacements in each isolator Updated values for Kisol,i are given below (previous
element to obtain new values of Kisol,i for each isolator values are in parentheses):
as follows:
, o Kisol,1 = 3.43 (3.41) k/in
, , (B-24) o Kisol,2 = 15.17 (14.26) k/in
,
o Kisol,3 = 15.17 (14.26) k/in
Recalculate Keff,j : o Kisol,4 = 3.43 (3.41) k/in

Eq. , ,
7.1-6 ,
(B-25) Since the isolator displacements are relatively close to
, ,
GSID previous results no significant change in the damping
ratio is expected. Hence Keff,j and are not

364

Recalculate system damping ratio, : recalculated and BL is taken at 1.70.

Eq. 2 , , ,
7.1-10 , , ,
(B-26)
GSID

Recalculate system damping factor, BL: Since the change in effective period is very small
(1.38 to 1.36 sec) and no change has been made to BL,
Eq. .
0.3 there is no need to construct a new composite
. response spectrum in this case. Go back to Step B2.6
7.1-3 ( B-27)
1.7 0.3 (see immediately below).
GSID

Obtain the effective period of the bridge from the

multi-modal analysis and with the revised damping
factor (Eq. B-27), construct a new composite response
spectrum. Go to Step B2.6.

B2.6 Multimodal Analysis Second Iteration,

Example 2.6
Reanalysis gives the following values for the isolator
displacements (numbers in parentheses are those
from the previous cycle):

Loc. Isol.# uL vL R
Isol.1 1.54 0.46 1.60
Abut1 Isol.2 1.53 0.47 1.60
Isol.3 1.52 0.47 1.59
Isol.1 1.19 0.37 1.25
Pier1 Isol.2 1.19 0.37 1.25
Isol.3 1.18 0.38 1.24
Isol.1 1.18 0.38 1.24
Pier2 Isol.2 1.19 0.37 1.25
Isol.3 1.19 0.37 1.25
Isol.1 1.52 0.47 1.59
Abut2 Isol.2 1.53 0.47 1.60
Isol.3 1.54 0.46 1.60

The resultant isolator displacements which will be

used to calculate the effective isolator stiffness are:
o disol,1 = 1.60 (1.63) in
o disol,2 = 1.25 (1.29) in
o disol,3 = 1.25 (1.29) in
o disol,4 = 1.60 (1.63) in

Go to Step B2.7

B2.7 Convergence Check B2.7 Convergence Check, Example 2.6

Compare results and determine if convergence has Satisfactory agreement has been reached on this
been reached. If so go to Step B2.9. Otherwise Go to second cycle. Go to Step B2.9
Step B2.8.

365

B2.9 Superstructure and Isolator Displacements B2.9 Superstructure and Isolator Displacements,
Once convergence has been reached, obtain Example 2.6
o superstructure displacements in the longitudinal From the above analysis:
(xL) and transverse (yL) directions of the bridge, o superstructure displacements in the
and longitudinal (xL) and transverse (yL) directions
o isolator displacements in the longitudinal (uL) are:
and transverse (vL) directions of the bridge, for xL= 1.53 in
each isolator, for this load case (i.e. yL= 0.48 in
longitudinal loading). These displacements
may be found by subtracting the nodal o isolator displacements in the longitudinal (uL)
displacements at each end of each isolator and transverse (vL) directions are:
spring element. o Abutments: uL = 1.54 in, vL = 0.47 in
o Piers: uL = 1.19 in, vL = 0.37 in

All isolators at same support have the same

displacements.

B2.10 Pier Bending Moments and Shear Forces B2.10 Pier Bending Moments and Shear Forces,
Obtain the pier bending moments and shear forces in Example 2.6
the longitudinal (MPLL, VPLL) and transverse (MPTL, Bending moments in single column pier in the
VPTL) directions at the critical locations for the longitudinal (MPLL) and transverse (MPTL) directions
longitudinally-applied seismic loading. are:
MPLL= 884
MPTL= 1508 kft

Shear forces in single column pier the longitudinal

(VPLL) and transverse (VPTL) directions are
VPLL = 70.63 k
VPTL = 43.85 k

B2.11 Isolator Shear and Axial Forces B2.11 Isolator Shear and Axial Forces, Example
Obtain the isolator shear (VLL, VTL) and axial forces 2.6
(PL) for the longitudinally-applied seismic loading. Isolator shear and axial forces are summarized in
Table B2.11-1

Table B2.11-1. Maximum Isolator Shear and Axial

Forces due to Longitudinal Earthquake.

VLL (k) VTL (k) PL (k)

Long. Transv. Axial
shear shear forces
due to due to due to
long. long. long.
EQ EQ EQ
Isol. 1 5.27 1.58 7.42
Abut
Isol. 2 5.26 1.60 9.99
ment
Isol. 3 5.21 1.61 14.52
Isol. 1 18.05 5.57 20.77
Pier Isol. 2 18.03 5.68 15.28
Isol. 3 17.86 5.69 16.21

366

STEP C. ANALYZE BRIDGE FOR EARTHQUAKE LOADING IN TRANSVERSE

DIRECTION
Repeat Steps B1 and B2 above to determine bridge response for transverse earthquake loading. Apply the
composite response spectrum in the transverse direction and obtain the following response parameters:
longitudinal and transverse displacements (uT, vT) for each isolator
longitudinal and transverse displacements for superstructure
biaxial column moments and shears at critical locations

C1. Analysis for Transverse Earthquake
Repeat the above process, starting at Step B1, for
earthquake loading in the transverse direction of the
bridge. Support flexibility in the transverse direction
is to be included, and a composite response spectrum
is to be applied in the transverse direction. Obtain
isolator displacements in the longitudinal (uT) and
transverse (vT) directions of the bridge, and the biaxial
bending moments and shear forces at critical locations
in the columns due to the transversely-applied seismic
loading.
C1. Analysis for Transverse Earthquake, Example
2.6
Key results from repeating Steps B1 and B2
(Simplified and Mulitmode Spectral Methods) for
transverse loading , are as follows:
o Teff = 1.51 sec
o Superstructure displacements in the longitudinal
(xT) and transverse (yT) directions due to
transverse load are as follows:
xT = 0.49 in
yT = 1.63 in
o Isolator displacements in the longitudinal (uT)
and transverse (vT) directions due to transverse
loading are as follows:

Loc. Isol.# uT vT R
Isol.1 0.51 1.59 1.67
Abut1 Isol.2 0.49 1.62 1.69
Isol.3 0.46 1.64 1.70
Isol.1 0.38 0.87 0.95
Pier1 Isol.2 0.38 0.87 0.95
Isol.3 0.38 0.85 0.93
Isol.1 0.38 0.85 0.93
Pier2 Isol.2 0.38 0.87 0.95
Isol.3 0.38 0.87 0.95
Isol.1 0.46 1.64 1.70
Abut2 Isol.2 0.49 1.62 1.69
Isol.3 0.51 1.59 1.67

Because of coupling between modes, there is

displacement in both the longitudinal and transverse
directions even when the earthquake is applied in the
transverse direction only.

o Pier bending moments in the longitudinal (MPLT)

and transverse (MPTT) directions due to
transverse load are as follows:
MPLT = 1645 kft
MPTT = 953 kft
o Pier shear forces in the longitudinal (VPLT) and
transverse (VPTT) directions due to transverse
load are as follows:

367

 VPLT = 43.01 k
VPTT = 63.67 k
o Isolator shear and axial forces are summarized in
Table C1-1.

Table C1-1. Maximum Isolator Shear and Axial

Forces due to Transverse Earthquake.

VLT (k) VTT (k) PT(k)

Long. Transv. Axial
shear shear forces
due to due to due to
transv. transv. transv.
EQ EQ EQ
Isol. 1 1.70 5.34 8.77
Abut
Isol. 2 1.65 5.43 14.68
ment
Isol. 3 1.55 5.49 23.32
Isol. 1 6.87 15.48 19.78
Pier Isol. 2 6.74 15.50 29.00
Isol. 3 6.71 15.25 31.33

368

STEP D. CALCULATE DESIGN VALUES (For Isolator 1 at Pier 1)

Combine results from longitudinal and transverse analyses using the (1.0L+0.3T) and (0.3L+1.0T) rules given in
Art 3.10.8 LRFD, to obtain design values for isolator and superstructure displacements, column moments and
shears.

Check that required performance is satisfied.

D1. Design Isolator Displacements D1. Design Isolator Displacements, Example 2.6
Following the provisions in Art. 2.1 GSID, and Load Case 1:
illustrated in Fig. 2.1-1 GSID, calculate the total u1 = uL + 0.3uT = 1.0(1.19) + 0.3(0.38) = 1.30 in
design displacement, dt, by combining the v1 = vL + 0.3vT = 1.0(0.37) + 0.3(0.87) = 0.63 in
displacements from the longitudinal (uL and vL) and R1 = = 1.30 0.63 = 1.44 in
transverse (uT and vT) cases as follows:
u1 = uL + 0.3uT (D-1) Load Case 2:
v1 = vL + 0.3vT (D-2) u2 = 0.3uL + uT = 0.3(1.19) + 1.0(0.38) = 0.74 in
R1 = (D-3) v2 = 0.3vL + vT = 0.3(0.37) + 1.0(0.88) =0.98 1in
R2 = = 0.74 0.98 = 1.23 in
u2 = 0.3uL + uT (D-4)
v2 = 0.3vL + vT (D-5)
R2 = (D-6)
Governing Case:
dt = max(R1, R2) (D-7) Total design displacement, dt = max(R1, R2)
= 1.44in

D2. Design Moments and Shears D2. Design Moments and Shears, Example 2.6
Calculate design values for column bending moments Load Case 1:
and shear forces using the same combination rules as VPL1= VPLL + 0.3VPLT = 1.0(70.63) + 0.3(43.01) =
for displacements. 83.53 k
VPT1= VPTL + 0.3VPTT = 1.0(43.85) + 0.3(63.67) =
Alternatively this step may be deferred because the 62.95 k
above results may not be final. Upper and lower R1 = = 83.53 62.95 = 104.59 k
bound analyses are required after the isolators have
been designed as described in Art 7. GSID. These Load Case 2:
analyses are required to determine the effect of VPL2= 0.3VPLL + VPLT = 0.3(70.63) + 1.0(43.01) =
possible variations in isolator properties due age, 64.20 k
temperature and scragging in elastomeric systems. VPT2= 0.3VPTL + VPTT = 0.3(43.85) + 1.0(63.67) =
Accordingly the results for column shear in Steps 76.83 k
B2.10 and C are likely to increase once these analyses
R2 = = 64.20 76.83 = 100.12 k
are complete.
Governing Case:
Design column shear = max (R1, R2)
= 104.59 K

369

STEP E. DESIGN OF LEAD-RUBBER (ELASTOMERIC) ISOLATORS

A lead-rubber isolator is an elastomeric
bearing with a lead core inserted on its vertical
centreline. When the bearing and lead core are
deformed in shear, the elastic stiffness of the
lead provides the initial stiffness (Ku).With
increasing lateral load the lead yields almost
perfectly plastically, and the post-yield
stiffness Kd is given by the rubber alone. More
details are given in MCEER 2006.

While both circular and rectangular bearings

are commercially available, circular bearings
are more commonly used. Consequently the
procedure given below focuses on circular
bearings. The same steps can be followed for
rectangular bearings, but some modifications will be necessary. When sizing the physical dimensions of the
bearing, plan dimensions (B, dL) should be rounded up to the next 1/4 increment, while the total thickness of
elastomer, Tr, is specified in multiples of the layer thickness. Typical layer thicknesses for bearings with lead
cores are 1/4 and 3/8. High quality natural rubber should be specified for the elastomer. It should have a shear
modulus in the range 60-120 psi and an ultimate elongation-at-break in excess of 5.5. Details can be found in
rubber handbooks or in MCEER 2006.

The following design procedure assumes the isolators are bolted to the masonry and sole plates. Isolators that use
shear-only connections (and not bolts) require additional design checks for stability which are not included below.
See MCEER 2006.

Note that the procedure given in this step is intended for preliminary design only. Final design details and
material selection should be checked with the manufacturer.

E1. Required Properties E1. Required Properties, Example 2.6

Obtain from previous work the properties required of The design of one of the exterior isolators on a pier is
the isolation system to achieve the specified given below to illustrate the design process for lead-
performance criteria (Step A1). rubber isolators.
required characteristic strength, Qd / isolator
required post-elastic stiffness, Kd / isolator From previous work
total design displacement, dt, for each isolator Qd / isolator = 10.95 k
maximum applied dead and live load (PDL, PLL) Kd / isolator = 6.64 k/in
and seismic load (PSL) which includes seismic Total design displacement dt = 1.44 in
live load (if any) and overturning forces due to PDL = 187 k
seismic loads, at each isolator, and PLL = 123 k
maximum wind load, PWL PSL = 29 k (Table C1-1)
PWL = 8.21 k < Qd OK

E2. Isolator Sizing

E2.1 Lead Core Diameter E2.1 Lead Core Diameter, Example 2.6
Determine the required diameter of the lead plug, dL,
using:

370

 10.95
(E-1) 3.49
0.9 0.9 0.9
See Step E2.5 for limitations on dL

E2.2 Plan Area and Isolator Diameter
Although no limits are placed on compressive stress in
the GSID, (maximum strain criteria are used instead,
see Step E3) it is useful to begin the sizing process by
assuming an allowable stress of, say, 1.6 ksi.
E2.2 Plan Area and Isolator Diameter, Example
2.6
Based on the final design of the isolators for Example
2.0, increase the allowable stress to 3.2 ksi.

Then the bonded area of the isolator is given by:

187 123
(E-2) 96.88
1.6 3.2 3.2

and the corresponding bonded diameter (taking into

account the hole required to accommodate the lead
core) is given by:
4 4 4 96.88
(E-3) 3.49

= 11.64 in
Round the bonded diameter, B, to nearest quarter inch,
and recalculate actual bonded area using Round B up to 12.5 in (based on experience with
Example 2.0) and the actual bonded area is:
(E-4)
4
12.50 3.49 113.16
4
Note that the overall diameter is equal to the bonded
diameter plus the thickness of the side cover layers
(usually 1/2 inch each side). In this case the overall
diameter, Bo is given by:

1.0 (E-5) Bo = 12.50 + 2(0.5) = 13.50in

E2.3 Elastomer Thickness and Number of Layers E2.3 Elastomer Thickness and Number of Layers,
Since the shear stiffness of the elastomeric bearing is Example 2.6
given by:
(E-6)
where G = shear modulus of the rubber, and
Tr = the total thickness of elastomer,
it follows Eq. E-5 may be used to obtain Tr given a
required value for Kd
0.1 113.16
(E-7) 1.70
6.64

If the layer thickness is tr, the number of layers, n, is

given by:
(E-8) 1.70
6.8
rounded up to the nearest integer. 0.25

Round up to nearest integer, i.e. n = 7

371

Note that because of rounding the plan dimensions

and the number of layers, the actual stiffness, Kd, will
not be exactly as required. Reanalysis may be
necessary if the differences are large.

E2.4 Overall Height E2.4 Overall Height, Example 2.6

The overall height of the isolator, H, is given by:

1 2 ( E-9) 7 0.25 6 0.125 2 1.5 5.50

where ts = thickness of an internal shim (usually

about 1/8 in), and
tc = combined thickness of end cover plate (0.5
in) and outer plate (1.0 in)

E2.5 Lead Core Size Check E2.5 Lead Core Size Check, Example 2.6
Experience has shown that for optimum performance Since B=16.25 check
of the lead core it must not be too small or too large.
The recommended range for the diameter is as 12.50 12.50
follows: 3 6
(E-10)
3 6 i.e., 4.16 2.08

Since dL = 3.49, lead core size is acceptable.

E3. Strain Limit Check E3. Strain Limit Check, Example 2.6
Art. 14.2 and 14.3 GSID requires that the total applied Since
shear strain from all sources in a single layer of 187.0
elastomer should not exceed 5.5, i.e., 1.65
113.16

, + 0.5 5.5 (E-11) and

113.16
11.53
where , , , are defined below. 12.50 0.25
(a) is the maximum shear strain in the layer due to
compression and is given by: then
1.0 1.65
(E-12) 1.43
0.1 11.53
where Dc is shape coefficient for compression in
circular bearings = 1.0, , G is shear
modulus, and S is the layer shape factor given by:
(E-13) 1.44
, 0.82
1.75
(b) , is the shear strain due to earthquake loads and
is given by:

, (E-14)
0.375 12.50 0.01
1.34
0.25 1.75
(c) is the shear strain due to rotation and is given
by: Substitution in Eq E-11 gives
(E-15)

372

where Dr is shape coefficient for rotation in circular , 0.5 1.43 0. 82 0.5 1.34
bearings = 0.375, and is design rotation due to DL, 2.92
LL and construction effects. Actual value for may 5.5
not be known at this time and a value of 0.01 is
suggested as an interim measure, including
uncertainties (see LRFD Art. 14.4.2.1).

E4. Vertical Load Stability Check E4. Vertical Load Stability Check, Example 2.6
Art 12.3 GSID requires the vertical load capacity of
all isolators be at least 3 times the applied vertical
loads (DL and LL) in the laterally undeformed state.

Further, the isolation system shall be stable under

1.2(DL+SL) at a horizontal displacement equal to
either
2 x total design displacement, dt, if in Seismic Zone 1
or 2, or
1.5 x total design displacement, dt, if in Seismic Zone
3 or 4.

E4.1 Vertical Load Stability in Undeformed State E4.1 Vertical Load Stability in Undeformed State,
The critical load capacity of an elastomeric isolator at Example 2.6
zero shear displacement is given by

4
1 1 3 3 0.1 0.3
2 (E-16)
0.3 1 0.67 11.53 26.89
where
12.50
1198.4
Ts = total shim thickness 64
26.89 1198.4
18,412 /
1 0.67 1.75
E = elastic modulus of elastomer = 3G
0.1 113.16
64 6.47 /
1.75
It is noted that typical elastomeric isolators have high
shape factors, S, in which case:
4
1 (E-17)

and Eq. E-16 reduces to:

(E-18) 6.47 18,412 1084

Check that:

3 (E-19) 1084
3.49 3
187 123

E4.2 Vertical Load Stability in Deformed State E4.2 Vertical Load Stability in Deformed State,
The critical load capacity of an elastomeric isolator at Example 2.6
shear displacement may be approximated by: Since bridge is in Zone 2, 2 2 1.44 2.88

373

 (E-20) 2.88
2 2.68
where 12.50
Ar = overlap area between top and bottom plates
of isolator at displacement (Fig. 2.2-1
GSID)
2.68 2.68
= 0.711
4
2

Agross = 4 0.711 1084 769

It follows that:
(E-21)

Check that:
769
1 (E-22) 3.03 1
1.2 1.2 1.2 187 29

E5. Design Review E5. Design Review, Example 2.6

The basic dimensions of the isolator designed above
are as follows:

13.50 in (od) x 5.50 in (high) x 3.49 in dia. lead core

and the volume, excluding steel end and cover plates,

= 358 in3

This design is considered satisfactory since both the

total strain (Eq E-11) and the vertical load stability
factors are reasonable values (not excessively low or
excessively high).

E6. Minimum and Maximum Performance Check
Art. 8 GSID requires the performance of any isolation
system be checked using minimum and maximum
values for the effective stiffness of the system. These
values are calculated from minimum and maximum
values of Kd and Qd, which are found using system
property modification factors, as indicated in Table
E6-1.
Table E6-1. Minimum and maximum values
for Kd and Qd.
E6. Minimum and Maximum Performance Check,
Example 2.6
Minimum Property Modification factors are
min,Kd = 1.0
min,Qd = 1.0
which means there is no need to reanalyze the bridge
with a set of minimum values.
Maximum Property Modification factors are

Eq. max,a,Kd = 1.1

8.1.2-1 Kd,max = Kd max,Kd (E-23) max,a,Qd = 1.1
GSID
Eq. max,t,Kd = 1.1
8.1.2-2 Kd,min = Kd min,Kd (E-24) max,t,Qd = 1.4
GSID
Eq. max,scrag,Kd = 1.0
Qd,max = Qd max,Qd (E-25) max,scrag,Qd = 1.0
8.1.2-3

374

 GSID
Eq.
8.1.2-4 Qd,min = Qd min,Qd (E-26)
GSID

Determination of the system property modification

factors shall include consideration of the effects of
temperature, aging, scragging, velocity, travel (wear)
and contamination as shown in Table E6-2. In lieu of
tests, numerical values for these factors can be
obtained from Appendix A, GSID.

Table E6-2. Minimum and maximum values for

system property modification factors.

Eq. min,Kd = (min,t,Kd) (min,a,Kd)

8.2.1-1 (min,v,Kd) (min,tr,Kd) (min,c,Kd) (E-27)
GSID (min,scrag,Kd)
Eq. max,Kd = (max,t,Kd) (max,a,Kd)
8.2.1-2 (max,v,Kd) (max,tr,Kd) (max,c,Kd) (E-28)
GSID (max,scrag,Kd)
Eq. min,Qd = (min,t,Qd) (min,a,Qd)
8.2.1-3 (min,v,Qd) (min,tr,Qd) (min,c,Qd) (E-29)
GSID (min,scrag,Qd)
Eq. max,Qd = (max,t,Qd) (max,a,Qd) Applying a system adjustment factor of 0.66 for an
8.2.1-4 (max,v,Qd) (max,tr,Qd) (max,c,Qd) (E-30) other bridge, the maximum property modification
GSID (max,scrag,Qd) factors become:

Adjustment factors are applied to individual -factors max,a,Kd = 1.0 + 0.1(0.66) = 1.066
max,a,Qd = 1.0 + 0.1(0.66) = 1.066
(except v) to account for the likelihood of occurrence
of all of the maxima (or all of the minima) at the same
max,t,Kd = 1.0 + 0.1(0.66) = 1.066
time. These factors are applied to all -factors that
max,t,Qd = 1.0 + 0.4(0.66) = 1.264
deviate from unity but only to the portion of the -
factor that is greater than, or less than, unity. Art.
max,scrag,Kd = 1.0
8.2.2 GSID gives these factors as follows:
max,scrag,Qd = 1.0
1.00 for critical bridges
0.75 for essential bridges
Therefore the maximum overall modification factors
0.66 for all other bridges
max,Kd = 1.066(1.066)1.0 = 1.14
As required in Art. 7 GSID and shown in Fig. C7-1
max,Qd = 1.066(1.264)1.0 = 1.35
GSID, the bridge should be reanalyzed for two cases:
once with Kd,min and Qd,min, and again with Kd,max and
Since the possible variation in upper bound properties
Qd,max. As indicated in Fig C7-1 GSID, maximum
exceeds 15% a reanalysis of the bridge is required to
displacements will probably be given by the first case
determine performance with these properties.
(Kd,min and Qd,min) and maximum forces by the second
case (Kd,max and Qd,max).
The upper-bound properties are:
Qd,max = 1.35 (10.95) = 14.78 k
and
Kd,ma x=1.14(6.64) = 7.57 k/in

E7. Design and Performance Summary E7. Design and Performance Summary

375

E7.1 Isolator dimensions E7.1 Isolator dimensions, Example 2.6

Summarize final dimensions of isolators: Isolator dimensions are summarized in Table E7.1-1.
Overall diameter (includes cover layer)
Overall height Table E7.1-1 Isolator Dimensions
Diameter lead core Isolator Overall size Overall size Diam.
Bonded diameter Location including without lead
Number of rubber layers mounting mounting core
Thickness of rubber layers plates (in) plates (in) (in)
Total rubber thickness Under
Thickness of steel shims edge 17.5 x 17.5 13.5 dia. x
3.49
Shear modulus of elastomer girder x 5.5(H) 4.0(H)
on Pier
Check all dimensions with manufacturer. Isolator No. of Rubber Total Steel
Location rubber layers rubber shim
layers thick- thick- thick-
ness ness ness
(in) (in) (in)
Under
edge
7 0.25 1.75 0.125
girder
on Pier

Shear modulus of elastomer = 100 psi

E7.2 Bridge Performance E7.2 Bridge Performance, Example 2.6

Summarize bridge performance Bridge performance is summarized in Table E7.2-1
Maximum superstructure displacement where it is seen that the maximum column shear is
(longitudinal) 106.8 k. This less than the column plastic shear
Maximum superstructure displacement (128k) and therefore the required performance
(transverse) criterion is satisfied (fully elastic behavior).
Maximum superstructure displacement Furthermore the maximum longitudinal displacement
(resultant) is 1.53 in which is less than the 2.5 in which is
Maximum column shear (resultant) available at the abutment expansion joints and
Maximum column moment (about transverse therefore acceptable.
axis)
Maximum column moment (about longitudinal Table E7.2-1 Summary of Bridge Performance
axis)
Maximum column torque Maximum superstructure
1.53 in
displacement (longitudinal)
Check required performance as determined in Step Maximum superstructure
1.63 in
A3, is satisfied. displacement (transverse)
Maximum superstructure
1.69 in
displacement (resultant)
Maximum column shear
106.8 k
(resultant)
Maximum column moment
1,621 kft
about transverse axis
Maximum column moment
1,692 kft
about longitudinal axis
Maximum column torque 14.2 kft

376

2012 AASHTO BRIDGE COMMITTEE AGENDA ITEM: 49

SUBJECT: LRFD Bridge Design Specifications: Section 14, Articles C14.6.5.2, C14.6.5.3 &
14.10 (WAI-1)

TECHNICAL COMMITTEE: T-2 Joints and Bearings

REVISION ADDITION NEW DOCUMENT

DESIGN SPEC CONSTRUCTION SPEC MOVABLE SPEC

MANUAL FOR BRIDGE SEISMIC GUIDE SPEC COASTAL GUIDE SPEC
EVALUATION OTHER

DATE PREPARED: 2/14/12

DATE REVISED: 5/7/12

AGENDA ITEM:
Item #1

Revise Article C14.6.5.2 as follows:

Provisions for the design, specification, testing, and acceptance of isolation bearings are given in AASHTO
(1999) the current AASHTO Guide Specifications for Seismic Isolation Design.

Item #2

Revise the 9th paragraph in Article C14.6.5.3 as follows:

Seismic isolation-type bearings are not within the scope of these provisions, but they should also be
considered, and designed in accordance with the requirements of the current AASHTO Guide Specifications for
Seismic Isolation Design.

Item #3

Revise the following reference in Article 14.10:

AASHTO 1999 2010Guide Specifications for Seismic Isolation Design. Second Third Edition 2010, GSID-23.
American Association of State Highway and Transportation Officials, Washington, DC.

OTHER AFFECTED ARTICLES:

None

BACKGROUND:
Reference is made to the latest AASHTO Guide Specifications for Seismic Isolation Design, Third Edition 2010.
ANTICIPATED EFFECT ON BRIDGES:
None

REFERENCES:
Lead State: WA
Industry:
FHWA:

OTHER:
None