You are on page 1of 484

DREDGER

FAMILIARIZATION

Short Term Course Organized by:

INDIAN MARITIME UNIVERSITY


VISAKHAPATNAM CAMPUS
JANUARY 2011

PREFACE
The Course notes is being prepared as a part of conduct of Short Term Course titled
Dredger Familiarization conducted by Indian Maritime University, Visakhapatnam
Campus.

It is intended that the extent of subject covered through this Course shall give a decent
exposure to the participants on various important aspects of dredgers and their operations.
However, for advanced knowledge level, they shall refer to additional reading material,
referred to in the notes. Perhaps, being on a dredger involved in the dredging operations
would be the best thing!!!

The Course Coordinators acknowledge the contributions made by various resource


personnel. The material in this Course notes is compiled and edited from various sources
including internet resources and intended for education purpose only. The lecture notes by
Prof W.J Vlasblom are felt to be excellent in getting familiar with the dredger. The
equipment by dredger giants like IHC, Holland is referred to at places. Material contribution
from several other organizations related to the field of dredging is duly acknowledged. Any
inadvertent copyright violations are sincerely regretted.

Course Coordinators

CONTENTS

INTRODUCTION TO DREDGING EQUIPMENT


TRAILING SUCTION HOPPER DREDGER
CUTTER SUCTION DREDGER
PLAIN SUCTION DREDGER
BARGE UNLOADING OR RECLAMATION DREDGER
BUCKET OR LADDER DREDGER
DREDGING ENGINEERING PUMPS AND SYSTEMS
DISPOSAL OF SOIL
OPTIMUM LOADING PRACTICES
DREDGER INSTRUMENTAION AND AUTOMATION
DREDGING AND RECLAMATION: TRENDS AND FUTURE

Chapter 1 Introduction

1. Introduction to Dredging Equipment


1.

Introduction to Dredging Equipment .......................................................................... 1


1.1.
Introduction......................................................................................................... 1
1.2.
Types of dredging equipment ............................................................................. 2
1.3.
Mechanical dredgers ........................................................................................... 3
1.3.1.
The bucket ladder dredge................................................................................ 3
1.3.1.1.
General........................................................................................................ 3
1.3.1.2.
Working method ......................................................................................... 5
1.3.1.3.
Area of application...................................................................................... 6
1.3.2.
Grab or Clamshell dredger.............................................................................. 7
1.3.2.1.
General........................................................................................................ 7
1.3.2.2.
Working method ......................................................................................... 7
1.3.2.3.
Area of application...................................................................................... 9
1.3.3.
Hydraulic cranes (Backhoe and front shovel)............................................... 10
1.3.3.1.
Working method ....................................................................................... 11
1.3.3.2.
Area of application.................................................................................... 12
1.4.
Hydraulic dredgers............................................................................................ 13
1.4.1.
Plain suction dredger..................................................................................... 13
1.4.1.1.
General...................................................................................................... 13
1.4.1.2.
Working method ....................................................................................... 15
1.4.1.3.
Area of application.................................................................................... 16
1.4.2.
Barge unloading dredger............................................................................... 17
1.4.2.1.
General...................................................................................................... 17
1.4.3.
The cutter suction dredger ............................................................................ 18
1.4.3.1.
General...................................................................................................... 18
1.4.3.2.
Working Method....................................................................................... 19
1.4.3.3.
Applied working area................................................................................ 21
1.4.4.
The bucket wheel dredger............................................................................. 22
1.4.5.
Trailing Suction Hopper Dredger ................................................................. 23
1.4.5.1.
General...................................................................................................... 23
1.4.5.2.
Working method ....................................................................................... 24
1.4.5.3.
Applied working area................................................................................ 26
1.5.
Conclusion ........................................................................................................ 27

1.1. Introduction
Definition: A dredgers is a piece of equipment which can dig, transport and dump a
certain amount of under water laying soil in a certain time.
The quantity of soil moved per unit of time is called Production.
Dredgers can dig hydraulically or mechanically. Hydraulic digging make use of the
erosive working of a water flow. For instance, a water flow generated by a dredge
pump is lead via suction mouth over a sand bed. The flow will erode the sand bed and
forms a sand-water mixture before it enters the suction pipe. Hydraulic digging is
Page1 of 27

Wb3408b

Designing Dredging Equipment

mostly done with special water jets. Hydraulic digging is mostly done in cohesionless
soils such as silt, sand and gravel.
Mechanical digging by knives, teeth or cutting edges of dredging equipment is apply
to cohesive soils.
The transport of the dredged soil can be done hydraulically or mechanically too, ether
continuously or discontinuously.
Continuously

Hydraulically
Transport via pipeline

Discontinuously

Mechanically
Transport via conveyor
belts
Transport via grab, ship,
car

Deposition of soil can be done in simple ways fi by opening the grab, turning the
bucket or opening the bottom doors in a ship. Hydraulic deposition happens when the
mixture is flowing over the reclamation area. The sand will settle while the water
flows back to sea or river.
Dredging equipment can have these three functions integrated or separated.
The choice of the dredger for executing a dredging operation depends not only on the
above mentioned functions but also from other conditions such as the accessibility to
the site, weather and wave conditions, anchoring conditions, required accuracy and so
on.

1.2. Types of dredging equipment


Dredging equipment can be divided in Mechanical Dredgers and Hydraulic Dredgers.
The differences between these two types are the way that the soil is excavated; either
mechanical or hydraulic.

Mechanical dredgers are

Bucket ladder dredge

Page 2 van 27

Grab dredge

Chapter 1 Introduction

Dipper and backhoe dredge


Hydraulic dredgers are:

Plain suction dredge

Cutter dredge

Trailing suction hopper dredge


All dredgers except the trailing suction hopper dredgers are stationary dredgers, which
means that they are anchored by wires or (spud)poles.

1.3. Mechanical dredgers


1.3.1.

The bucket ladder dredge

1.3.1.1.

General

Page 3 of 27

Wb3408b

Designing Dredging Equipment

The bucket ladder dredge Big Dalton


The bucket ladder dredge or bucket chain dredger is a stationary dredger, which has an
endless chain of buckets carried by the so-called ladder, positioned in the well of a Ushape pontoon. The chain is driven by the upper tumbler, a pentogonal, at the upper
part of the ladder and fixed at the bottom with lower tumbler, mostly a hectagonal.
Under the ladder the chain hangs freely, while on the upper site of the ladder the chain
is supported and guided by rollers. The buckets filled during their rotation over the
lower tumbler are emptied by the rotation over upper tumbler. The soil from there
guided via shutes to an alongside layer barge.
Bucket sizes vary from 30 liters to 1200 liters. Rock bucket dredgers do have a double
set of buckets; a small rock bucket and a bigger soft soil bucket.

Page 4 of 27

Chapter 1 Introduction

1.3.1.2.
Working method
The bucket ladder dredge is positioned on 6 wires. Under working conditions the
dredge swings around her bow anchor. The bow anchor line or headline can have
length longer than 1000 m. In order to avoid dragging of the wire over the soil, which
results in a smaller radius, the wire is supported by a headline pontoon. As a result of
this long headline the cut width can be large as well (200 m or more). The sideline
winches take care of the swinging of the dredge as well as the power necessary for the
cutting process. The swing speed depends on the spoil condition, the layer thickness
cut and forward step (pawl length)

Page 5 of 27

Wb3408b

Designing Dredging Equipment


Bow anchor

Headwire
Forward ground anchor PS
Headwire pontoon

Forward side anchor SB

"Pawl" length

Swing over

Aft ground anchor PS


Aft side anchor SB

Groundlevel

Dry
excavation

Dredge
profile

Spillage

Cutwidth

Stern anchor

1.3.1.3.
Area of application
A bucket dredgers can be applied in almost all soils, from soft silt and clays to soft
rock depending on the power on and the strength of the bucket chain. They are use in
blasted rock as well.
The maximum dredging depth depends on the size of the dredger. Bucket ladder
dredgers with a maximum dredging depth of over the 30 m are built.
However for such dredgers the minimum dredging depth is almost 8 m.
Page 6 of 27

Chapter 1 Introduction
Nowadays they are often used for dredging contaminated mud, because the can dig the
soil under in situ density conditions.
The bucket ladder dredge can not applied under offshore conditions and is certainly an
obstruction for shipping. Compared to hydraulic dredgers he production is rather low.

1.3.2.

Grab or Clamshell dredger

1.3.2.1.
General
The grab dredger is the most common used dredger in the world, especially in North
America and the Far East.
It is a rather simple and easy to understand stationary dredger with and without
propulsion. In the latter the ship has a hold (hopper) in which it can store the dredge
material, otherwise the material is transported by barges. The dredgers can be moored
by anchors or by poles (spuds)
The capacity of a grab dredger is expressed in the volume of the grab. Grab sizes
varies between less than 1 m3 up to 200 m3.
The opening of the grab is controlled by the closing and hoisting wire or by hydraulic
cylinders.
1.3.2.2.
Working method
For grab dredgers the method of anchoring and the positioning system plays an
important role for the effectiveness of the dredger.
At every pontoon position an area as wide as possible will be dredged. Looking from
the centerline the volume to be dredged at the position decreases with the angle to the
centerline.
The positioning is important to localize the bit of the grab. This helps the dredge
master to place the next bit after the fore going.
Page 7 of 27

Wb3408b

Designing Dredging Equipment

Releasing the aft wires and pulling the fore wires does the movement of the pontoon.
When the dredgers have spud poles, this movement is done by a spud operation, which
is more accurate than executed by wires.
15 %

37 %

48 %

1* step

ste

Center line

tep
7s
0.8

0.5

30
60

Dredge pattern

The dredging process is discontinuously and cyclic.


1.
Lowering of the grab to the bottom
2.
Closing of the grab by pulling the hoisting wire
3.
Hoisting starts when the bucket is complete closed
4.
Swinging to the barge or hopper
5.
Lowering the filled bucket into the barge or hopper
6.
Opening the bucket by releasing the closing wire.
The principle of this hoisting operation is given in the figure below. In order to avoid
spinning of the clamshell a so-called taught wire is connected to the clamshell.

Page 8 of 27

Chapter 1 Introduction
Hoist winch
Top shieves

Hoist wires

Gear segments

Upper sheave block

Closing winch
Closing wires

Bucket

Gear segments

Lower sheave block

1.3.2.3.
Area of application
The large grab dredgers are used for bulk dredging. While the smaller ones are mostly
used for special jobs, such as:
Difficult accessible places in harbors
Small quantities with strongly varying depth.
Along quay walls where the soil is spoiled by wires and debris
Borrowing sand and gravel in deep pits
Etc.
The production of a grab depends strongly on the soil. Suitable materials are soft clay,
sand and gravel. Though, boulder clay is dredged as well by this type of dredger. In
soft soils light big grabs are used while in more cohesive soils heavy small grabs are
favorable.
The dredging depth depends only on the length of the wire on the winches. However
the accuracy decreases with depth.

Page 9 of 27

Wb3408b

1.3.3.

Designing Dredging Equipment

Hydraulic cranes (Backhoe and front shovel)

Hydraulic cranes are available in two models the backhoe and the front shovel. The
first is used most. The difference between those two is the working method. The
backhoe pulls the bucket to the dredger, while the front shovel pushes. The last
method is only used when the water depth is insufficient for the pontoon.
These stationary dredgers are anchored by three spud poles; two fixes to the front side
of the pontoon and one movable at the aft side. This means that the dredging depth is
limited to about 15 m. (maximum 25 m). At the front of the pontoon is normally a
standard cranes mounted. Here pontoon deck is lower to increase the dredging depth.
Bucket sizes vary from a few m3 to 20 m3.

Backhoe dredge

Page 10 of 27

Chapter 1 Introduction

Front shovel
1.3.3.1.
Working method
During dredging the pontoon is lifted a few out of the water by wires running over the
spud poles. A part of the weight of the dredger is now transferred via the spuds to the
bottom, resulting a sufficient anchoring to deliver the required reaction for the digging
forces. Besides that the dredger is in this case less sensible for waves. The bucket is
placed and filled by hydraulic cylinders on the boom and the bucket arm.
Due to the small radius of the boom and arm is the cut width limited to 10 to 20 m, see
figure below.
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

Page 11 of 27

Wb3408b

Designing Dredging Equipment

The effective dredging area depends on the swing angle and the forward step per
pontoon position. A small step results in a large width and a large step in a small
width, however the total area is almost the same.

1.3.3.2.
Area of application
This is roughly the same as for the clamshell dredgers with the exception dredging
depth over the 25 m

Page 12 of 27

Chapter 1 Introduction

1.4. Hydraulic dredgers


1.4.1.

Plain suction dredger

1.4.1.1.
General
A plain suction dredger is a stationary dredger that position on one ore more wires,
with at least one dredge pump, which is connected to the suction pipe and the delivery
pipe. The suction pipe is situated in a well in front of the pontoon.
Good production can only achieved by this kind of dredgers either the soil is free
running sand or the cut or breach height is sufficient (at least 10 m)
The discharge of the soil sucked is done either by pipeline or by barges.
Most suction dredgers are equipped with jet water pump(s) to assist either the
beaching process or to improve the mixture forming process near the suction mouth.
Types of plain suction dredgers
There are different types to be distinguished.
1.

Barge Loading suction dredger

Page 13 of 27

Wb3408b

Designing Dredging Equipment

28

18
24

16

22

11

31

21
27

13

25 19

10

23

17

26

20

14
15

Used when the transport distances are too large for direct pumping
2. Standard plain suction dredger

Discharged the material direct via pipeline to the reclamation area.


3.

Deep suction dredger

Pag 14 of 27

Chapter 1 Introduction

22
23
24

5
1

17/18

11

28

27

19/20

16

10

30

29

34

11

31

14

12

12

35

33

35

33

13/15

32

32

2
2

This dredger is equipped with an underwater pump and have two appearances; the
standard or from the barge loading type. When dredging depth exceeds the 30 m this
dredgers is more appropriate than the standard one.
4.

Dustpan dredger

A suction dredger with a wide suction mouth, which makes it possible to dredge with
reasonable productions low cut heights.
1.4.1.2.
Working method
The working method is based on the breaching process and the erosion created by
the flow near the suction mouth, generated by the dredge pump.
Breaching is a process of soil shearing on a slope caused by local instabilities or by
erosion of the density current running along the slope to the suction mouth

Page 15 of 27

Wb3408b

Designing Dredging Equipment


z

Instabilities

H br

Sand-water mixture
(density current)
Suction tube
Vz

Breach
This process is essential for this type of dredger and is fully determined by the soil
conditions of the slope, from which the permeability and the relative density re the
most important parameters. The dredge patron made by a plain suction dredger is
shown below.

The length of the cut depends, inside the borrow area, on the position of the anchors.
Mostly the anchors are laid down in such a way that more cuts can be made without
repositioning the anchors. However this depends not only of the length of the
anchoring wires but also from the breachebility of the soil.
1.4.1.3.
Area of application
Due to the lack of cutting devices this type of dredger is only suitable in non-cohesive
soils. Further more this method exclude accurate dredging work. Dredging under
Page 16 of 27

Chapter 1 Introduction
offshore conditions is possible with special equipment. As already said borrowing in
deep pits of over 100 m depth is possible.
These types of dredgers are frequently used in borrow pits for reclamation areas as
well as for the borrowing of sand for the concrete industry.

1.4.2.

Barge unloading dredger

1.4.2.1.
General
Barge unloading dredgers are used for emptying loaded barges either by suction
dredgers or by bucket ladder dredgers and cranes.
The barge-unloading dredger is a stationary special suction dredger anchored by spuds
near the shore, where the water depth is sufficient for the loading barges to come along
side the dredger.
The water for the unloading and the transport is supplied into the barge by a jet.

Page 17 of 27

Wb3408b

1.4.3.

Designing Dredging Equipment

The cutter suction dredger

1.4.3.1.
General
The cutter suction dredger is a stationary dredger equipped with a cutter device (cutter
head) which excavate the soil before it is sucked up by the flow of the dredge pump(s).
During operation the dredger moves around a spud pole by pulling and slacking on the
two fore sideline wires. This type of dredger is capable to dredge all kind of material
and is accurate due to their movement around the spud. The spoil is mostly
hydraulically transported via pipeline, but some dredgers do have barge-loading
facilities as well.
Sea going cutter suction dredgers have their own propulsion, however this is only used
during (de) mobilization.
Cutter power ranges from 50 kW up to 5000 kW, depending on the type of soil to be
cut.

Custom build dredger


Page 18 of 27

Chapter 1 Introduction
The more powerful dredgers are capable to dredge rock
The small and medium size cutter suction dredgers are deliverable in a demountable
application. In that case the hull consists out of five or more pontoons. The central
pontoon contains the machinery.

Standard Beaver dredger


1.4.3.2.
Working Method
The rotating cutter excavates the soil during their movement, generated by the side
winches, form port side to starboard and vise versa.
The necessary side winch force depends not only on the type of soil but also on:

The rotation direction of the cutter head; (over cutting) rotation in the direction
of the swing movement or (under cutting) opposite to that.

Ds

Ds
Under cutting mode

Over cutting mode

In the over cutting mode the cutter head tries to drag the cutter dredger in the direction
of the pulling winch. Braking with the opposite winch may be necessary.

The position of the anchors in relation to the path of the cutter head. The more
the anchor lies in the direction of the moving cutter head the less the required side
winch force will be.
External forces, such as wind, current and waves.
Page 19 of 27

Wb3408b

Designing Dredging Equipment

The thickness of the layer, which can be cut in one swing, depends besides on the soil
conditions also on the size of the cutter head. At the end of the swing will either the
ladder be lowered and the dredger is swung in the opposite direction or the dredger
will make a step forwards.

As said earlier the dredgers swings around a pole the working spud, which is
positioned mostly in a carriage. The spud carriage can be moved over a distance of 4
to 6 m. by a hydraulic cylinder. When the working spud is set on the ground the
dredger is pushed forward when the cylinder pushes against the carriage. This forward
movement is called step and depends also on the soil conditions and the size of the
cutter head.
During a step the breach is cut in one or more cuts.
Cut width

Workspud
in carriage

Spud carriage
length
Auxilary spud

Vertical swing pattern

Page 20 of 27

Chapter 1 Introduction
Because the spud stays on the same spot the dredger makes concentric circles during
swinging.
Is the stroke of the hydraulic cylinder is maximum the dredger is moved to the
centerline of the cut where a second spud at the aft side of the pontoon, the step spud,
is lowered. Where after the working spud is hoisted and the carriage is pulled back, the
working spud lowered to the ground and the step spud hoisted again. The dredger can
make a new cycle again.
1.4.3.3.
Applied working area
Cutter suction dredgers are applied for dredging harbors, channels, reclamation areas
and so on. The transport distance of the mixture is limited to maximum 10 km. She is
very useful when the accuracy of the works is important. As said already the cutter
dredger can dredge all kinds of soil.

clay cutter

Rock Cutter

For dredging under offshore conditions is this dredger less suitable.

Page 21 of 27

Wb3408b

1.4.4.

Designing Dredging Equipment

The bucket wheel dredger


This dredger is, with the exception
of the cutter head, is comparable
with the cutter suction dredger. The
rotation axe of the bucket wheel is
perpendicular with the ship axe. The
wheel contains 10 14 open or
closed buckets.
Due to the construction of the drive
the wheel is difficult to replace and
therefore less universal than the
cutter suction dredger.
Is application area is the same as the
cutter dredger with the exception of
hard rock. This dredger is often used
in areas with constant conditions,
such as the sea mining.

Page 22 of 27

Chapter 1 Introduction

1.4.5.

Trailing Suction Hopper Dredger

1.4.5.1.
General
A Trailing Suction Hopper Dredger (TSHD) is a self-propelled sea-going or inland
vessel equipped with a hold, called hopper, and a dredging installation by which it can
fill and/or empty the hopper.
The basic options of a THSD are:
One or more suction tubes provided with suction mouths (dragheads) which are
dragged over the seabed during dredging.
One or more dredge pumps to suck the material from the seabed.
A hopper in which the dredged material can settle.
Easy operational bottom doors or valves in the hopper to dump the dredge
material
Gantries and winches to operate the suction tubes.
A swell compensator to control the contact between the suction mouth and the
seabed when dredging in waves.
The size of a TSHD is expressed in the hopper volume and varies between a few
hundred m3 up to 33000 m3

Page 23 of 27

Wb3408b

Designing Dredging Equipment

A 23350 and 700 m3 hopper dredger


1.4.5.2.

Working method
When arrived at the dredging area, the speed
of the vessel is reduced to about 2 to 3 knots
(1 to 1.5 m/s), where after the suction tubes
are lowered till the seabed and the dredge
pumps started. When the suction tubes reach
the seabed the swell compensator reacts, easy
to see by the movement of the hydraulic
cylinder.
Nowadays electronic charts and screens
shows where and how much there is to
dredge.
During dredging a mixture of soil and water
is dumped into the hopper. When dredging
non-settling slurries dredging is stopped when
the mixture reach the overflow; a device to
discharge fluids from the hopper above a
certain level.

When dredging settling slurries dredging is continue after the mixture has reached the
top of the overflow. Now the majority of the soil will settle in the hopper, while the
fine particles together with the water will leave the hopper via the overflow.
Page 24 of 27

Chapter 1 Introduction

Overflows
After the overflow is reached, the dredging procedure depends either the overflow
level is fixed or variable.
With a fixed overflow level the loading is continued till the ship has reached the
allowed draught. The mixture volume in the hopper stays constant during this part
of the loading process. Depending on the bulk density of the settled material there
will be a certain volume of water above the settled material. (constant volume
system)
Is de THSD provided with a variable overflow system, the overflow may be
lowered when the ship has reached the allowed draught, on order to replace the
water volume by settled material. (constant tonnage system)

Rods for opening


and closing
Suction channel for
self-discharching
Upper
door
Rubber seal

Rubber seal
Bottom door

Bottom door
Page 25 of 27

Pivot

Wb3408b

Designing Dredging Equipment

When the hopper is filled, dredging is stopped and the suction tubes placed on the
deck of the ship, where after she is ready to sail to the unloading area. The THSD can
be unloaded either by opening the bottom doors or to pump the load via a pump ashore
equipment to the reclamation area.

Pumping ashore (rain bowing)

1.4.5.3.
Applied working area
The THSD is a free sailing vessel and does not hinder other shipping during dredging
and is therefore ideal for dredging in harbors and shipping channels inshore as well as
offshore. The seagoing vessels are very suitable for borrowing sand under offshore
conditions (wind and waves) and large sailing distances. The dredged material is
dredged, transported and discharged by the vessel without any help from other
equipment.
(De)mobilization is very easy for this type of dredger. It can sail under its own power
to every place in the world.
Suitable materials for the THSD to dredge are soft clays, silt sand and gravel. Firm
and stiff clays are also possible but can give either blocking problem in the draghead
and/or track forming in the clay. In that case the draghead slips into foregoing tracks,
resulting in a very irregular clay surface. Dredging rock with a TSHD is in most cases
not profitable. It requires very heavy dragheads with rippers and the productions are
rather low.

Page 26 of 27

Chapter 1 Introduction

Modern 9000 m3 Hopper dredger with one dredge pipe

1.5. Conclusion
Summarized it can be stated that every type of dredger has its own applied working
area in which its production is optimal in a technical way as well as in an economical
way. It will be clear that the boundaries of these applied working areas are not strictly
determined, but are also determined by other working conditions, which can differ
from lob to job.
In the table below the possibilities of the different types are shortly summarized.
Bucket Grab
Backhoe
Dredger Dredger Dredger

Suction Cutter
Trailer
Hopper
Dredger Dredger Dredger Dredger

Dredging sandy materials

yes

yes

yes

yes

yes

yes

Dredging clayey materials

yes

yes

yes

no

yes

yes

yes
no

Dredging rocky materials

yes

no

yes

no

yes

no

no

anchoring wires

yes

yes

no

yes

yes

no

yes

Maximum dredging depth [m]

30

> 100

20

70

25

100

50

accurated dredging possible

yes

no

yes

no

yes

no

no

working under offshore conditions possible

no

yes

no

yes

no

yes

yes

Transport via pipeline

no

no

no

yes

yes

no

no

Dredging in situ densities possible

yes

yes

yes

no

limited

no

no

Page 27 of 27

Chapter 2 Trailing suction hopper dredger


2

Trailing suction hopper dredger............................................................................................1


2.1
General description ..................................................................................................2
2.1.1 Characteristics .............................................................................................2
2.1.2 Application area ..........................................................................................3
2.1.3 History.........................................................................................................3
2.1.4 Work method...............................................................................................5
2.2
The design ................................................................................................................8
2.2.1 The productive capacity ..............................................................................8
2.2.2 The main dimensions ..................................................................................10
2.2.3 The dredge installation ................................................................................15
2.2.4 The propulsion power .................................................................................32
2.2.5 Power balance .............................................................................................38
2.2.6 Main layout .................................................................................................41
2.3
Technical Construction ............................................................................................47
2.3.1 The dredge installation ................................................................................47
2.3.2 The hopper ..................................................................................................63
2.3.3 The propulsion ............................................................................................75
2.3.4 The maneuverability....................................................................................75
2.4
Strength and stability ...............................................................................................77
2.4.1 Strength .......................................................................................................77
2.4.2 Stability .......................................................................................................78
2.5
The dredging process ...............................................................................................80
2.5.1 The loading process.....................................................................................80
2.5.2 Sailing from and to the discharging area.....................................................99
2.5.3 The discharge ..............................................................................................100
2.5.4 The cycle production...................................................................................102
2.5.5 The instrumentation ....................................................................................103
2.6
Special designs of trailing suction hopper dredgers.................................................104
2.6.1 The gravel suction dredger..........................................................................104
2.6.2 The stationary suction hopper dredger ........................................................106
2.6.3 Boom dredgers ............................................................................................107
2.7
Literature..................................................................................................................109

Page 1 of 109

wb3408B

Designing Dredging Equipment

2 Trailing suction hopper dredger

Figure 2-1 Trailing Suction Hopper Dredger (TSHD)

2.1 General description


2.1.1

Characteristics

The characteristics of the trailing suction hopper dredger are that it is a self-propelled sea or
inland waterway vessel, equipped with a hold (hopper) and a dredge installation to load and
unload itself.
In a standard design the trailing suction hopper dredger is equipped with:

One or more suction pipes with suction mouths, called dragheads that are dragged over the
seabed while dredging.
One or more dredge pumps to suck up the loosened soil by the dragheads.
A hold (hopper) in which the material sucked up is dumped.
An overflow system to discharge the redundant water.
Closable doors or valves in the hold to unload the cargo.
Suction pipe gantries to hoist the suction pipes on board.
An installation, called the swell compensator, to compensate for the vertical movement of
the ship in relation with the sea-bed.

Page 2 of 109

Chapter 2 Trailing suction hopper dredger

2.1.2

Application area

The trailing suction hopper dredger has a very wide application area and is therefore called the
workhorse of the dredging industry.
Because it needs no anchorage system to position the vessel when dredging, which can be an
obstacle for passing ships, in the early days the trailing suction hopper dredger (TSHD) was
mainly used for the deepening and maintaining of waterways. Nowadays the trailing suction
hopper dredger is also used for land reclamation. Examples of that type of jobs are the large
reclamation works executed in the Far East. Here the non-bearing soil was first removed by the
trailing suction hopper dredger, after which the same area was filled again with sand. The
reason for a preference of the trailing suction hopper dredger above other types of equipment
for this type of work is mainly the fact that the distances to the dump areas for the non-suitable
material and distance from the sand pits are too large for a direct discharge and supply with
pipelines.
The main advantages of a trailing suction hopper dredger are:

The ship does not dredge on a fixed position. It has no anchors and cables, but it moves
freely, which is especially important in harbor areas.
The trailing suction hopper dredger is quite able to work under offshore conditions.

The materials that can be sucked are mainly silt and sand. Clay is also well possible, but can
give some trouble with congestions in the draghead and rutting. Rutting is the slipping back of
the dragheads in their old rut or trail. Dredging rock with a trailing suction hopper dredger is
in most cases not economical. It requires very heavy dragheads, also called ripper-heads, and
the productions are usually very low.

2.1.3

History

The first TSHD General Moultry with a hopper size of 155 cu yard (118.5 m3) was built in
1855 in the United States. Few years later 1959 a trailing suction hopper dredger was build in
France for maintenance work in the harbor of St. Nazaire.

Figure 2-2 French trailing suction hopper dredger from 1859


Page 3 of 109

wb3408B

Designing Dredging Equipment

The ship had two drag suction pipes, which were connected at the bottom by a tube with holes
(Figure 2.2). The dredging material, silt, was sucked through the holes in the connection tube
by a steam-driven centrifugal pump. The size of the hopper was 240 m3.
In 1962 a dredger was built according to this layout at the yard Fijenoord at Rotterdam,
Netherlands. Those types were able to dredge only very light silty material.
The real development of the trailing suction hopper dredger emanated from the stationary
suction hopper dredger, one of the few Dutch dredge inventions. This self-propelled ship has
a hopper and a forward pointing suction pipe. The dredge method is like a stationary suction
dredger, working stationary on anchors and cables. At first with a pipe in the well, but the
suction pipe was mounted on the side during the excavation of the Nieuwe Waterweg as it
appeared not the right solution in waves.
The change from an anchored to a self-propelled dredging ship was a big step ahead. At first
the suction pipe on board of a trailing suction hopper dredger was placed in a well behind the
ship, but was soon moved to the side. The trailing suction hopper dredger has mainly developed
in the USA and reintroduced in the Netherlands in the fifties and improved till it state of today.

Figure 2-3 Artist impression of TSHD

Page 4 of 109

Chapter 2 Trailing suction hopper dredger

2.1.4

Work method

When arriving on the dredging area the


speed of the trailing suction hopper dredger
is reduced to approximately 3 knots ( 1.5
m/s) and the suction pipes are swung
outboard. The suction pipes are initially
lowered approximately horizontally until
the trunnion slide is positioned in front of
the suction intake (Figure 2.4).

Main deck

Middle gantry wire


Base of ship

Draghead wire

Next the intermediate gantry and the


draghead winch gantry are lowered such
that the pipe rotates like a straight line
around the trunnion.

Figure 2-4 Suction pipe lowered

When the suction mouth arrives a few meters


above the sea bottom the sand pumps are started,
the dragheads are lowered onto the seabed (which
can be seen by the rise of the swell compensators
cylinders (Figure 2.5) and the dredging can start.
Where and how much needs to be dredged is
nowadays shown on electronic maps (computer
screens). It also shows the position, direction and
course of the ship.

Figure 2-5 The swell compensator

The trailing suction hopper


dredger sucks the soil from the
seabed at a sailing speed of 1
to 1.5 m/s (2 to 3 knots) and
deposits it in the hopper. For
non- or bad-settling soils the
dredging is stopped when the
surface of the mixture in the
hopper reaches the upper edge
of the overflow (Figure 2.6).

Adjustable overflow

Dredging mark

Figure 2-6 Justable overflow


Page 5 of 109

wb3408B

Designing Dredging Equipment

The hopper filling is at maximum or the fill rate is 100%. Usually pumping continues for five
minutes more to remove floating water on the mixture through the overflow. When dredging
settling soils the dredging continues when the maximum level of the overflow is reached. Most
of the solids will settle and the remainder is discharged with the water through the overflow.

Fixed overflow

Fixed overflow
This water is not removable

Dredging mark

Constant Volume hopper

If the trailing suction


hopper dredger is
equipped with a fixed
overflow
(not
adjustable) than the
ship is loaded until it
reaches its dredge
mark
(a
fixed
allowed
draught)
after
which
the
suction is stopped.

Figure 2-7

That case it is said that the ship is designed as a Constant Volume System (CVS).

Adjustable overflow

Dredging mark

Constant Tonnage system

Figure 2-8

If the ship however


has
a
height
adjustable overflow
system, than it is
possible, when the
hopper is full and the
ship is on its mark, to
lower the overflow
level such that the
total weight of the in
the hopper present
water
and
soil
remains constant.

This is called a Constant Tonnage System (CTS).

The dredging is stopped when:

The hopper is full. Overflow not allowed.


The maximum allowable draught is reached and the overflow can not be lowered usefully
anymore.
The economical filling rate is reached.

When dredging stops, the suction pipes are pumped clean to prevent settling of the sand or
gravel during the hoisting of the pipes causing an extra load for the winches. When the pipes
Page 6 of 109

Chapter 2 Trailing suction hopper dredger


are cleaned the pumping stops and the pipes are raised. When the dragheads are out of the
water the ships velocity is increased to sail to the discharge area.
The discharge area can:

Be in its most simple shape a natural deepening of the seabed, the dumping area (shortly
dump), to store redundant material. If the storage capacity is large, there is no concern
about the way of dumping. This hardly happens nowadays. The client demands usually a
dump plan to fill the dump as efficiently as possible. At all times the draught on the dump
needs to be sufficient to open the bottom doors or valves (Figure 2.9).
Be a storage location for contaminated silt, like for instance the Slufter (Rotterdam harbor).
Here the material is pumped ashore using a pump ashore discharge system.
An area that has to be reclaimed.
An oil or gas pipe that has to be covered.

Rods for opening


and closing
Suction channel for
self-discharching
Upper
door
Rubber seal

Rubber seal

Pivot

Bottom door

Figure 2-9 Bottoms doors operated by rods

In case of the discharge area is a dump,


opening the doors or valves in the base
of the hopper does the unloading.
This is usually done with an almost
non-moving ship, certainly when
accurate dumping is required. During
the dumping water is pumped onto the
load by means of the sand pumps. The
eroding water stimulates the dumping
process. If the trailing suction hopper
dredger is equipped with jet pumps
connected to a jet nozzle system in the
hopper, those will be used too. The jets
more or less fluidize the load and
improve the dumping process.

If the load is pumped ashore using the sand pumps than only these jets are available to fluidize
or erode the load.
The shore connection, being the
connection between the board pipeline
and the shore pipeline is currently
mostly positioned just above the bow
(Figure 2.10). The connection between
the ship and the shore piping is this
case a rubber pipeline. The ship
remains in position by maneuvering
with its main propellers and bow
thruster(s).

Figure 2-10 Pump ashore connection

Page 7 of 109

wb3408B

Designing Dredging Equipment

When the load is either


dumped or pumped ashore
the ship will return to its
suction area and a new cycle
starts. In general the ship
sails empty, in a non-ballast
way, back to its suction
section. There is only some
residual water and/or load
left in the hopper
Figure 2-11 TSHD J.J.F. de NUL picking up the floating pipeline
to the shore connection

2.2 The design


2.2.1

The productive capacity

When a dredging company wants to order a new trailing suction hopper dredger usually a
market study is performed that about the required production capacity of the new dredger.
The required production capacity is expressed in m3/week or m3/month or even cubic meters
per year. Besides that insight required about the expected average cycle time of the trailing
suction hopper dredger on the different jobs, as well as the type of soils to be dredged. Then the
production capacity can be translated to:

The required payload in ton mass.


The maximum hopper volume in m3.

If the ship is used for a single purpose, for instance the maintenance of a harbor area, than the
required production capacity is usually known and therefore the above mentioned ship data.
For an international operating dredging contractor this is different and far more complicated.
Answers have to be given to the question how the average cycle and the required production
capacity will evolve in the future. For these contractors there is in fact only one requirement
and that is dredging cheaper than their competitors. This leads quickly to a demand for large
dredgers, which dredge cheaper and therefore more competitive.

Page 8 of 109

Chapter 2 Trailing suction hopper dredger


y = 3.0656Ln(x) - 19.711

Load - Draught relation

R = 0.8888

Draught [m]

14
12
10
8
6
4
2
0
0

5000

10000

15000

20000

25000

30000

35000

Payload [ton]

Figure 2-12 Payload - draught relation

The only decelerator on the building of larger vessels is the draught of the ship. When the
draught increases, the usability of the ship decreases. The contractor can, dependent on the
expected amount of work as function of the (initial) dredging depth, determine the availability
of the ship for a certain draught.

Cumulative
frequency [%]

Cumulative frequency distribution of initial dredging


depth
120
100
80
60
40
20
0
0

10

20

30

40

50

Initial dredging depth [m]


Figure 2-13

Unfortunately it is possible that market expectations of today are totally out-of-date in 5 years.
The management chooses for a certain production capacity and later one wills just if this choice
was right.
The design is usually made a co-operation between the builder and the client is often scaled-up
from successful ships. Of course the proper scale rules have to be obeyed when scaling-up.
At this moment five classes of trailing suction hopper dredgers can be distinguished:
Small hoppers
Medium size hoppers
Large hoppers
Jumbo hoppers
Mega hoppers
Page 9 of 109

deadweight capacity to
deadweight capacity
deadweight capacity
deadweight capacity
deadweight capacity

50 MN (to 5000 ton mass)


50-100 MN (5000-10000 ton mass)
100-150 MN (10000-15000 ton mass)
150 250 MN (15000-25000 ton mass)
>250 MN (above 25000 ton mass)

wb3408B

Designing Dredging Equipment

Figure 2-14 Different scales Fairway (23.347 m3) and the Sospan (700 m3)

2.2.2

The main dimensions

When the choice for the production capacity of the trailing suction hopper dredger to be built is
made, the hopper volume is known too. The main dimensions of the trailing suction hopper
dredger are determined, as by other ships, by the required payload, draught and speed. It will be
clear that a straight correlation exists between these quantities to satisfy the shipbuilding
demands. After all a large hopper volume with a limited draught gives wide long ships with
possible disadvantages like a poor behavior in swell or problems to obtain the required speed.
Trailing suction hopper dredgers are therefore build according to certain ship ratio, such as
L/B, B/H and B/T ratio's (L=length, B=width, H=depth and T=draught). Those ratioss depend
on market requirements too and therefore change in time (Figure 2.15)
With the remark that a large B/T ratio:

Results in a large initial stability, resulting in heavy ship motions in swell.


Has an adverse effect on the resistance of the ship.

With a large L/B ratio a lean ship is obtained with the advantages of:

A simple construction as a result of the long equal mid-section (cheap).


A relative low resistance, therefore a higher velocity with the same installed propulsion
power.

Page 10 of 109

Chapter 2 Trailing suction hopper dredger

L/B, B/H, B/T

Ships Numbers
8
7
6
5
4
3
2
1
0
1965

L/B
B/H
B/T

1970

1975

1980

1985

1990

1995

2000

Year of Construction
Figure 2-15

On the other hand a small L/B gives a good stability and longitude strength and demands
therefore less material, which is also cheaper.
In general a smaller B/H and a larger L/B result in less building costs. So demands for the
draught (smaller T) will cost extra money and will have to be earned with a higher usability.

T
L

Cb =

LBT

Figure 2-16 Definition Block coefficient

Definition Block coefficient


Of course the required block coefficient Cb =
Displacement
B
L
T

Page 11 of 109

=
=
=
=

displacement
is involved too.
=
L B T
L B T

In m3
Width of ship at the main section I m
Length between perpendiculars in m
Draught at International mark in m

wb3408B

Designing Dredging Equipment

The lower Cb, the longer the ship will be with the same displacement. For trailing suction
hopper dredger Cb lies between 0,78 and 0,85.
Also the required maximum dredging depth can have an influence on the length of the ship.
Naturally, the long suction pipe has to be stored on the deck and that requires length.

Specific Ships Weight


1

W_spec

0.8
0.6
0.4
0.2
0
0

10000

20000

30000

40000

50000

60000

Displacement [t]

Figure 2-17

A good measure to see if the trailing suction hopper dredger is well placed in the market is to
compare its specific weight with that of its competitors. The specific weight can be defined as
the ratio between the ships weight and payload. The weight is directly related to the costs and
the payload to the profits. In Figure 2.17 the specific weight for a large number of ships is
given.
2.2.2.1 The load
As aid, the payload in tons and the maximum hopper volume in m3 determine the amount of
soil that a trailing suction hopper dredger is able to carry each voyage. These are of great
importance. The payload is the weight of the paying load that the ship may carry on the
maximum allowed draught. The payload is often a cause for misunderstandings. As a definition
the payload is the ship weight of the loaded ship subtracted with the weight of the empty ship
ready for service. This is shown in the hereunder shown chart.

Dutch term
Scheepsgewicht

Toegevoegde gewichten Added weights

1+2

Gewicht leeg schip

Page 12 of 109

English term
Ships weight

Weight light ship

Explanation
Construction weight and necessary
equipment like: anchors, chains,
moor cables, rescue equipment,
nautical equipment and inventory of
the cabins, galley, engine-room and
tool-room of the boatswain
This is the liquid filling of all
systems on board including the
water in the inlets. Also the outside
water situated above the bottom
deck for instance under and around
the bottom doors is included.

Chapter 2 Trailing suction hopper dredger


3

Toelading

Dead weight

1+2+3 Gewicht van het


geladen schip
4
Gewicht lading
1+2+3 Gewicht bedrijfsklaar
+4
schip

Weights of:
Crew and their possessions,
consumer goods, spare parts, and
ballast water and load.

Weight of loaded
Vessel
Weight cargo
Ships weight ready for
Service

Weight of the paying load.

Figures below gives some information about light weight and dead weight of TSHDs
70,000
60,000

y = 0.6827x
R2 = 0.9929

Weight [t]

50,000
40,000

G Light weight

30,000

Dead weight

20,000

y = 0.3173x
R2 = 0.9622

10,000
0
0

20,000 40,000 60,000 80,000 100,000


Displacement [t]

Figure 2-18
Light weight as function of deadweight

Light weight [t]

25,000
20,000
15,000

y = -3E-06x 2 + 0.5586x
R2 = 0.9607

10,000
5,000
0
0

10,000

20,000 30,000 40,000 50,000

60,000 70,000

Deadweight [t[

Figure 2-19

Except that there are different names for the payload, it is also apparent that it varies in time
and often decreases. The reason is that when the ship has been in use for a while things will be
added or reinforced, which causes an increase in the ships weight. Spare parts also tend to
remain on board that should be stored onshore. In fact there is only one way to determine the
payload correctly:
1. Clear the hopper such that no remaining soil is present.
2. Determine the displacement of the ship with the draught and the trim of the ship, the
displacement is the weight of the ship including the water in the hopper.
3. Determine the weight of the water present in the hopper by determining its volume and the
specific gravity
4. Subtract the weight of this water the ships weight determined under point 2. This is the
weight of the ship ready for service.
Page 13 of 109

wb3408B

Designing Dredging Equipment

5. The payload is obtained by subtracting the ships mass (displacement x water density) in
tons on the maximum allowed draught with the weight of the ship ready for service.
It will b clear that the payload is never constant, but varies with the weight of the consumer
goods like fuel, lubricants, drinking water etc.
In case of light soils, such as silt and soft clay, the maximum hopper volume can be decisive
for production instead of the payload.
2.2.2.2 The hopper density.
As mentioned earlier, the production capacity of a trailing suction hopper dredger is indicated
with the quantities:

Pay-load
Maximum hopper volume

The quotient

pay load
[kg / m3 ] is called the hopper density and is a
maximum hopper volume

measure for the average density that a dredging contractor expects to dredge during the
economical lifetime of the ship. It also says something over the purpose for which the dredger
is designed. Is this for instance maintenance of a fairway in a sandy soil, than the dredges sand
in the hopper will have a density of approximately 1900 kg/m3. Unfortunately no hopper can be
filled to a 100% but approximately to maximum 90%. The maximum hopper density required
is 1900 * 0.9 = 1710 kg/m3
For a gravel trailing suction hopper dredger this is for instance: 2000 * 0,9 = 1800 kg/m3. And
for a silt trailing suction hopper dredger this could be even 1300 kg/m3. In Figure 2.20 the
hopper density of international operating dredging contractors is shown as function of time. It
stabilizes at the end of the eighties and early nineties around 1500 kg/m3, but due to the big
reclamation works it is increasing again.

Hopper denisty as function of time


2.50

Hopper density [t/m3]

2.00
1.50
1.00
0.50
0.00
1950

1960

1970

1980

1990

Construction year

Figure 2-20

Pag 14 of 109

2000

2010

Chapter 2 Trailing suction hopper dredger

2.2.3

The dredge installation

The design of a dredge installation includes the determination of the required main dimensions
and required powers of the following dredging components:

Number of suction pipes


Pump capacity [m/s]
Suction and discharge pipe diameter [m]
Type dredge pump
Sand pump drive and power [W]
Type and size of the draghead(s)
Hopper shape
Jet pump power and drive [W]
Discharge systems

For the subjects the production should be corrected in a certain way from the average cycle
production of the dredger.
For instant, assume that the dredger is designed for a payload of 16000 ton and a hopper
volume of 10000 m3 and a average loading time in sand with a d50 of 200 of 90 minutes. De
density of the soil in the hopper is 1900 kg/m3. When the hopper is loaded the volume of sand
will be 8421 m3. The average load rate is in this case 8421/90=93 m3/min=1.56 m3/s.
When cumulative overflow losses of 20% are to be expected, then the dragheads should
excavate 1.56/0.8=1.95 m3/s as an average. Every m3 of sand contains 1-(1900-1025)/(26501025)= 1-0.538=0.462 m3 water in the pores. (water=1025 kg/m3, sand is 2650 kg/m3). So a
production of 1.95 m3/s equals a sand mass of 1.95*0.538*2650=2780 kg/s
2.2.3.1 Number of suction pipes
A trailing suction hopper dredger is usually equipped with two suction pipes. For smaller and
medium size trailing suction hopper dredgers it is cheaper to use only one suction pipe. With
two suction pipes the total efficiency is often better because it is still possible to dredge when
one of the pipes fails.
There are also examples of large trailing suction hopper dredger with one suction pipe: the
ANTIGOON of Dredging International with a hopper volume of 8.400 m3 and the VOLVOX
TERRA NOVA of Van Oord ACZ with 18.000 m3 hopper volume. In principal it is an
economical consideration, but looking from the process technical side there are some questions.
For example: is one draghead as efficient as two dragheads with the same width?

Page 15 of 109

wb3408B

Designing Dredging Equipment

Figure 2-21 Volvox Terra Nova and HAM 316, both with one suction pipe

2.2.3.2 Pump capacity


The sand pump capacity can be determined using several criteria:

1. In a particular type of soil a certain load time is demanded. (for instance 1 hour for
sand with a d50 of 200-300 m)
T
T

V
C
Q
C
Q
=
The volume pure sand as function of time is: sand
i i o o dt
0

0
Co
Ci
Qo
Ci
T
n0
ov
Vsh
s

=
=
=
=
=
=
=
=
=

Volumetric concentration at overflow


Volumetric concentration at intake
Discharge at overflow
Flow rate at intake
Loading time
Porosity
Cumulative overflow losses
Volume sand in the hopper
Density of sand in the hopper

This sand occupied in volume in the hopper of Vsh =

[-]
[-]
[m3/s]
[m3/s]
[s]
[-]
[-]
[m3]
[kg/m3]

Vsand
; n is porosity
1 - n0 0

For TSHDs having a constant volume system Qm=Qi=Qo, with Qm is the pump
capacity; so the mass of the load becomes:
T

s
s
L mass =
Qm (Ci -Co ) dt=
Qm Ci (1 ov) T
1 n0
1

n
0
0

Page 16 of 109

Chapter 2 Trailing suction hopper dredger


T

CQ

CQ

With ov being the cumulative overflow losses defines as

ov=

0
T

For 1 hour loading the flow rate is:

Qm =

Vsand
(1-ov ) Ci 3600

Ci and Co are delivered concentrations; so Ci=Cdv


The expected Cvd depends on the particle size, the permeability of the soil and the
available jet water momentum. (see 2.5.5.1.3)
If the TSHD is designed as a constant tonnage dredger the incoming mass equals
the outgoing mass; so m=mi=mo.
mi = Qi mi and mo = Qo mo so Qi mi = Qo mo or Qo = Qi

mi
mo

mi andmo are respectively the mixture densities at the intake and overflow.
The load becomes now:
T

L mass =

s
Qi
1 n 0 0

C -C mi dt= s Q C (1 ov) T
i
o
i
m

1 n0
mo

Although the formula is the same as for the constant volume system hopper
dredgers it doesnt mean that the cumulative overflow losses are the same for
both types.

2.

In an ascertain type of sand the load rate in m/s or in t/s must have a minimum value.
If there would be no overflow losses than the load rate is directly proportional to the
flow rate. However, the overflow losses increase with an increasing flow rate, which
result in an increasing deviation from the linear relation. (Figure 2.22& 2.23)

Page 17 of 109

wb3408B

Designing Dredging Equipment

Loadrate=F{Q} d50=.15 mm
=1100

=1200

=1300

10

12

180
160

Loadrate [m3/min]

140
120
100
80
60
40
20
0
0

14

16

18

20

Capacity [m3/s]

Figure 2-22 Loadrate as function of pump capacity

It can be proven that for certain particle sizes there is an optimum loadrate.

Loadrate=F{Q} d50=.1 mm
=1.1

=1.2

=1.3 [t/m3]

Load rate m /s
350
300
250
200
150
100
50
0
0

10

15

20

Capacity [m3/s]
Figure 2-23

The increase of a higher suction production (load rate) must be considered against the
higher sand and water pump power, larger suction pipe diameter and dragheads etc.
Page 18 of 109

Chapter 2 Trailing suction hopper dredger


Remark: In Figure 2.23 the step in the load rate is caused by the fact that for high
densities and high flow rates the loading after the overflow is not necessary since the
optimal production for the dredge cycle has been reached.
3.

When apart from the soil the cycle time is known too, than the flow rate can be
chosen such that the cycle production is maximal. The cycle production is defined as
the quotient between loading and cycle time, so: Pc =

load
tsuction + tnon suction

If there are no overflow losses than this formula can be written as:

Pc =

load

load
+ tnon suction

Q Cvd k g

Q load

load

+ Q tnon suction

Cvd k g

This is a monotone ascending function. However the overflow losses cause an


optimal flow rate for which the cycle production has a maximum. (Figure 2.24)

Cycle Production d50=.15 mm


=1.1

Pcycle [m3/c]

=1.2

=1.3 [t/m3]

2000
1500
1000
500
0
0

10

15

20

Capacity [m3/s]
Figure 2-24

4.

Also the pump capacity can be scaled from existing "well working" trailing suction
hopper dredgers, by using the scale rule from Froude. However overflow losses will
not be on scale when using this scale rule.

Above mentioned criterions lead to a design flow rate and a design density.
2.2.3.3 Suction pipe diameters
Old trailing suction hopper dredgers are equipped with relatively large suction pipe diameters.
In the past the size of the diameter was mainly based on minimizing the pressure loss in the
suction pipe to avoid cavitation of the dredge pump. However it was understood that the
concentration distribution was homogeneous over the diameter, which is not always the case.

Page 19 of 109

wb3408B

Designing Dredging Equipment

For a homogenous flow it can be shown


that the suction production is maximum
for a certain suction velocity. This is
done with the so-called suction formula,
a force balance over the suction pipe.
For a pump that is positioned k meters
under the surface The pressure at the
suction mouth is mgH. The pressure in
front of the pump p is equal to the
allowable underpressure, vacuum, so
p=-VAC.
The pressure difference over the suction
pipe equals the weight of the mixture
and the losses in the pipe.

hz

Mixture velocity vs
Mixture density m

Figure 2-25

water g H + Vac = mixture g hz + 12 mixture v 2 = mixture g ( H k ) + 12 mixture v 2

g H + Vac
mixture = water
2
g (H k ) +

Pr = Q Cvd k = v A

mixture water
grain
grain water

This function appears to have, dependent on H, k, Vac and , an optimum for a certain suction
velocity v, which is independent of the suction pipe diameter.
can be written as = + L with;
D

=entre loss coefficient [-]


=Darcy-Weisbach resistance coefficient [-]
L=length of suction pipe in m
D=suction pipe diameter in m

Page 20 of 109

Chapter 2 Trailing suction hopper dredger

D=750 mm

D=1000 mm

1350

1200

1300

1000

1250

800

1200

600

1150

400

1100
1050

200

1000

0
0

2
4
6
Suction velocity [m/s]
vacuum=80kPa

Production [kg/s]

Mixture density [kgm3]

rho_m

Figure 2-26

Application of the suction formula has several disadvantages:


1. The mixture density, the resistance factor and the suction velocity are not independent of
each other, but are determined by the erosion process and the pump characteristics.
2. The flow is only homogeneous for sand types with a d50 < 0.15 mm.
For coarser materials the flow becomes heterogeneous. As a result the volumetric concentration
(the amount of sand in the pipe) increases and therefore also the pressure loss in the pipe. In
other words the decrease of the pressure loss by the lower velocity is cancelled out by the
increase as a result of the higher volumetric concentration. Therefore the pressure loss in the
pipe does no longer behave according: p = 12 v 2 .
For this reason modern trailing suction hopper dredgers do have relative smaller suction pipe
diameter then in the past. Besides that heavier pipes demand heavier winches, gantries and their
foundations. This leads to a lower useful deadweight capacity and more investment cost.
Figure 2.27 below shows the relation between the maximum hopper volume and the suction
pipes diameters for trailing suction hopper dredgers with two suction pipes. (diameters above
800 mm are round off to 100 mm and under 800 mm to 50 mm)
As can be seen in the Figure 2.27 the spread in the used suction pipe diameters is considerable.
This could lead to the conclusion that design process is not yet unambiguous. At present
however modern TSHDs have smaller in suction pipe diameter at the same flow rate. This is
especially affected by the better insights in the two-phase flow at relative low velocities for
inclined pipes.

Page 21 of 109

wb3408B

Designing Dredging Equipment

One pipe vessels

Two pipe vessels

1.20

Pipe diameter [m]

Pipe diameter [m]

1.40
1.00
0.80
0.60
0.40
0.20
0.00
0

5,000

10,000

15,000

20,000

1.60
1.40
1.20
1.00
0.80
0.60
0.40
0.20
0.00

25,000

5000

10000

15000

20000

25000

30000

35000

Hopper volume [m3]

Hopper volume [m3]

Figure 2-27

From many researches it appears that the velocity for which all soil particles in the pipe are still

v2
in motion is dependent on the Froude-value:
. (v=velocity and D pipe diameter)
gD
Depending on the grain size and concentration the Froude-value may not become less than a
certain value FI,H. Adding the maximum average velocities for which no stationary bed is

formed in a horizontal pipeline can be calculated using Vsm = Fl 2 g ( Ss 1) D or with the


demi-McDonald of Wilson, which can be estimated with the formula:

s ( Ss S f )
1.75
8.8
D 0.7 d50
0.66

Vsm =
d502 + 0.11 D 0.7
0.55

With d50 in mm and the diameter D in meters.

In Figure 2.28 both formulas are drawn (Durant, Fl=1.4). For inclined suction pipes Vsm has to
be raised with a value D dependent of the incline. According Wilson and Tse D reaches a
maximum for approximately 30 and is then D=0.333 (Matousek, 1997).
In the design of trailing suction hopper dredgers usually Fl = 1.00 is assumed and D is not
considered. This implies that the dredger is designed for materials with a d50 between 100 and
300 m and that for coarser materials a stationary bed is accepted.

Page 22 of 109

Chapter 2 Trailing suction hopper dredger

V_stationary deposition for horizontal transport


mm
Wilson

Durant

d50=.5

Practice

V_deposit [m/s]

10
8
6
4
2
0
0

0.2

0.4

0.6

0.8

1.2

Pipe diameter [m]

Figure 2-28

The use of suction pipe with a


submerged pump (Figure 2.29)
has a direct influence on the
choice of the diameter of the
suction pipe. Is this the case
then it is possible to choose the
suction pipe diameter a little
smaller and so lighter and
cheaper,
against
the
disadvantage of a little
additional pressure loss in the
pipeline..

Figure 2-29 Dredge pump incorperate in the suction pipe

Page 23 of 109

wb3408B

Designing Dredging Equipment

2.2.3.4 The pressure pipe diameter


The diameter of the pressure pipe should have a larger diameter than the suction pipe, because
the factor 0.333 for the inclined transport. However often, depending of the value of the factor
Fl,H, the pressure pipe diameter is chosen 50-100 mm smaller for costs reasons. Particular when
the casted elbows and valves are used. The diameter of the pump ashore installation will
generally be chosen smaller than the suction pipe. Normally the hopper is unloaded with
considerable higher concentrations than loaded. This allow for a lower flow rate when
discharge time equals the suction time.
2.2.3.5

The dredge pump

The main dimensions of the ship


and the dredge installation are now
known, so an estimate can be made
to the required manometric head of
the dredge pump for the different
(un)loading conditions.
The required pump pressure during
loading is determined by the static
head from hart pump to the
discharge in the hopper and the
losses in the discharge line.
The manometric head is the sum of
required pressure and the allowable
vacuum at the suction side of the
pump.

Figure 2-30 Pump room with 2 pumps

Because the impeller diameter is approximately known ( minimum 2 times suction pipe
diameter) and there is a relation between the required manometric pressure and the peripheral
velocity of the pump impeller, also the specific pump speed is approximately known.
The dimensionless specific pump speed is defined as:
1

Ns =

2
3

With:

Q
=
Db

p
p
=
=
2
u
2 r 2

In these is:
Q = flow rate
p = pressure
Page 24 of 109

dimensionless capacity

dimensionless pressure
[m3/s]
[Pa]

Chapter 2 Trailing suction hopper dredger


D
b
r

=
=
=
=
=

diameter pump impeller


width pump impeller
D
density fluid
angular velocity pump impeller

[m]
[m]
[m]
[kg/m3]
[rad/s]

Filling in and results in


1

Ns =

3
4

3
4

Q
p

3
4

D
4 b

(1)

Figure 2-31

The specific speed is assessed to the maximum efficiency point and is a characteristic number
to compare pumps with their dimensions like the b/D ratio, inlet and outlet diameter ratio Di/Du
and impeller shapes (Figure 2.31). Equation (1) shows that for a constant number of revolutions
() the specific number of revolutions increases with an increasing flow rate and decreasing
pressure. Since the pressure is proportional to the square of the peripheral velocity, the pressure
will decrease at a constant number of revolutions with a decreasing diameter. A higher flow
rate requires a larger diameter in the impeller, therefore a larger b/D ratio. Besides the b/D ratio
especially a wider passage in the impeller has a large influence.
Figure 2.32 shows the relation between the dimensionless capacity and pressure as function of
the number of revolutions for all types of hydraulic suction dredgers. Left in the chart are the
standard centrifugal pumps and on the right the modern half-axial or mixed flow pumps,
usually used as submerged pump in the suction pipe pump of trailing suction hopper dredgers
and cutter suction dredgers. In general the dimensionless pressure for hopper pumps is slightly
higher for the same specific flow rate than for the pressure pumps of cutter suction dredgers
and suction dredgers.
From formula (1) it follows that when Q, p, and Ns are known, the pump speed can be
determined, so that the pump and impeller type can also be chosen. (note: When the dredger
will be equipped with a pump ashore installation, there will be two pump speeds.)
Page 25 of 109

wb3408B

Designing Dredging Equipment

For relative small trailing suction hopper dredgers and suction depths a fixed pump speed for
the dredging mode (suction) is often sufficient. When the difference between minimum and
maximum dredging depth is large, a variable pump speed may be required.

All Dredgers

Specific Capacity

Ca p a c ity

0.8

0.14

0.7

0.12

0.6

0.1

0.5

0.08

0.4

0.06

0.3

0.04

0.2

0.02

0.1

Specific Head

Hea d

0.16

0
0

0.2

0.4

0.6

0.8

Specific Speed

Figure 2-32

With increasing size and particular for increasing depth the question may rise if this can lead to
large flow rate variations during the dredging process. Large flow rate variations often lead to
water-hammer problems in the pipelines. If this risk exists than an adjustable pumpspeed is
necessary.
There are more factors involved in the choice of a pump, such as:

3, 4 or 5 impeller vanes. Dependent on the required minimal opening area between the
blades.
Single- or double-walled pump (wear considerations).
Inboard or submerged pump or both. If great suction depths are expected, it has to be
considered if the installation of submerged pumps is more economical. The limit where this
economical point is reached is closely connected with depth of the inboard pump below
water level under service conditions, so roughly with the draught of the ship. This break
point is therefore different for every ship.
The operation of the pump during pumping ashore (if necessary).

When the dredger is provided with a pump ashore installation attention shall be given to the
pumps working under both conditions. During pumping ashore it becomes more and more a
custom that all available power of the main engines are used. This implies that the maximum
pump speed when pumping ashore differs significantly from the pump speed during dredging.
As a consequence the best efficiency point of the pump when pumping ashore shifts to a
considerable higher flow rate than during dredging. This shift is in reality even larger because
the pump ashore capacity is usually smaller than the flow rate during dredging (why?).
It has to be realized however that a pump working under conditions far above or below the best
efficiency point, will wear faster. A good research of the position of the best efficiency points
under the different service conditions is therefore necessary to obtain the optimal installation.
Page 26 of 109

Chapter 2 Trailing suction hopper dredger


Also the required pump power for both modes can now be calculated. However, the maximum
available pump power during pumping ashore is with a combined drive (one engine for pump +
propulsion) determined by the required propulsion power.

Pumpcharacteristics for dredging and pump ashore


Q-p/165 rpm

Eff/280 rpm

Eff/165 rpm
120

1000

100

800

80

600

60

400

40

200

20

Efficiency [%]

Manometric pressure [kPa]

Q-p/280 rpm
1200

0
0

capacity [m3/s]

Figure 2-33

2.2.3.6 The dredge pump drive


Before choosing a drive the question should be answered whether continuous pump speed
control is required or speed control by a gearbox is sufficient.

The following factors are involved:

The expected range of the flow rate variation between the pumping of the water and of the
slurry. This range is larger with an increasing suction depth, provided no cavitation takes
place. Limitation of this variation can be necessary to reduce the risk of water-hammer. In
that case a constant pump speed or a stepped control is insufficient.
When a constant flow rate control is desired. The flow rate is regulated by a variation of the
pump speed. An electric drive is necessary. A constant flow rate control by varying the
number of revolutions is not suitable to prevent water-hammer (too slow).
If the ship is equipped with a pump ashore installation and the propulsion power can be
used totally or partly when pumping ashore. To use this additional power a higher pump
speed than use in the dredging mode is required.

Dependent on these demands the sand pump can be driven directly by the main engine through
a, if necessary, a stepped gearbox or directly by an electric engine through a generator. Of
course there are several intermediate solutions that are treated in the chapter "Main
arrangement".

Page 27 of 109

wb3408B

Designing Dredging Equipment

2.2.3.7 The dragheads


Dragheads are designed to excavated the soil and mix it with water for hydraulic transport.
Excavation can be done hydraulically or mechanically or combined. Hydraulic excavation is
either by erosion of the dredge pump flow, by pressurized water jets or both

Pure mechanical excavation is mainly done


in cohesive soils, such as clays and very
soft rock. For that case teeth or blades are
mounted in the draghead (Figure 2.34).
The width of the draghead is now
dependent on the expected cutting forces in
the particular soil in relation to the
available cutting force from the propulsion.
The length of the visor of the draghead
should be chosen such the flow pattern for
the transport of the excavated material
suites the excavation process.

Visor

Figure 2-34 Draghead with blade

Modern dragheads have water jets assisted


with knives or teeth.
A reasonable assumption is that the jetproduction is linear with the total momentum
flux of the jet system and independent of the
trail speed. The momentum I=wQu.

M sand = I = w Qu = w Q

2 p jet
w

Figure 2-35 Draghead with jets (not working)

With:
I
Msand
pjet
Q
u

=
=
=
=
=
=

Momentum in N
Eroded sand mass in kg/s per jet
Jet pressure at the nozzle in Pa
Jet capacity in m3/s
Jet velocity at the nozzle in m/s
Coefficient depending on the particle size, jet pressure, jet capacity and trailspeed.
A reasonable assumption for alpha is =0.1
Water density in kg/m3.

When the nozzle are divided well over the width of the draghead the mass M should fulfill the
relation:

sand

= B d v trail

all jets

Page 28 of 109

situ water
particle
particle water

Chapter 2 Trailing suction hopper dredger


B
D
vtrail
situ
particle

=
=
=
=
=

Width draghead in m.
Eroded layer thickness in m
Trailspeed in m/s
Density soil in situ kg/m3
Particle density in kg/m3

length/width [mm]

When the trailspeed is said to 1.5 m/s, which equals 3 knots and using the relation between pipe
diameter and draghead width of Figure 2.36, d can be calculated.
In general the effective of the jet decreases somewhat with increasing pressure at constant
momentum. This means that low pressure- high capacity jets are more effective than high
pressure-low capacity jets. They use more specific energy too. On the other hand however,
much jetwater dilutes the mixture density (Figure 2.128). So the designer has to search for the
optimum solution between cost (power) en production

4000
3500
3000
2500
2000
1500
1000
500
0

width

Length

500

1000

1500

Suctionpipe diameter [mm]


Figure 2-36 Dimensions Dutch draghead

2.2.3.8 The water pumps


Jet-water is used for loosening the soil within the dragheads, as well as to assist the process
during discharging the load, either by dumping or by pumping ashore. The flow rate of the
water pump is between 20 to 30 % of the sand pump flow rate and the pressure is usually
between 5 and 15 bar. The required pressure can be calculated using the same basic formulas
as mention in the forgoing chapter.

M sand = C vd Q m sand = w Q jet

LM
MN

1 sand C vd Q m
p=
2 w Q jet

OP
PQ

2p
w

In general there is no requirement for speed control of the type of pump

Page 29 of 109

wb3408B

Designing Dredging Equipment

2.2.3.9 The hopper


As mentioned before ships are built according certain L/B, B/T and B/H ratios. This also
accounts for trailing suction hopper dredgers.

Some insight in the effect of these ratios on the overflow losses is got from the Camps
Diagram (Figure 2.132)
The removal Ratio R, the percentage of the incoming material that settles in the hopper is een
function of:

R=f

FG S , S IJ = R = f FG SbBLg , SbBHgIJ
HS V K
H Q Q K
0

The following conclusion from Figure 2.132 can now be drawn when keeping the hopper
volume constant:
1. The width B is kept constant and L2L and H0.5H
1st term of the removal ratio shall increase and 2e term shall decrease. This results in
the conclusion:
More sedimentation at long shallow hoppers or less in short deep hoppers
2. The height H is kept constant and L2L and B0.5B
1st term of the removal ratio stays constant and 2e term shall decrease. This results in:
A little less sedimentation at long small hoppers or little better sedimentation
in short wide hoppers.
3. The length L is constant and H2H and B0.5B1st term of the removal ratio shall
decrease and 2e term stays constant. This results in:
Less sedimentation in small deep hoppers or better sedimentation in wide
shallow hoppers.
4. The height H and the width B are kept constant, while L0.5L and Q0.5Q
1st term of the removal ratio stays constant and 2e term shall increase. This results in:
Central intake or a TSHD with 2 hoppers is a little better.
From the theory of the overflow losses (chapter 2.5.1.3) can be derived that long, shallow
hoppers are favorable for the settlement process. Unfortunately such a shape leads to long
relatively narrow ship with a limited depth that result in certain design problems for engine
room en deckhouse. Therefore a compromise has to be found between the price and the
performance.

When scaling-up the hopper shape to larger dimensions one should be aware for an undesirable
increase of the overflow losses. After all for all new to build trailing suction hopper dredgers it
is often demanded that the load time, independent of the size of the hopper, has to be 1 hour for
a sand type with a d50 of 250 m. This implies that the flow rate will be proportional to the
volume of the hopper when the concentration is assumed constant.
Therefore the capacity scale is: Q = ( L )

b g

b g

Both the terms S BL and S BH shall decrease and this implies that the overflow loss for
larger trailing suction hopper dredgers will be higher than for smaller trailing suction hopper
dredgers, even if the hoppers are similar. Dependent on the magnitude of this increase this
Page 30 of 109

Chapter 2 Trailing suction hopper dredger


could still be acceptable, since the cycle production can still be higher with higher overflow
losses.
A design requirement directly related to the hopper shape is that the sand level at restricted
loads needs to be higher than the sealevel.
Such a requirement is of
importance in situations
where it is not possible to
dredge to the dredge mark
because of the waterdepth.
If the sea level is higher
than the sand level, the
water cannot flow out and
the ship cant be loaded
economically.

This water is not removable


Adjustable overflow

Dredging mark

Constant Tonnage system

Figure 2-37

For modern ships this requirement can be satisfied for a 50-60% of the maximal load.
2.2.3.10 The discharge system
From the theory of the flow of bulk material from silos follows that a plane symmetrical flow
will occur for discharge openings where length L 3B (width) and that this flow type, is
preferred above an axial symmetrical flow. Unfortunately most discharge systems, except for
the split hopper (Figure 2.38) don't satisfy this requirement, while the building of a split hopper
suction dredgers is considerably more expensive than "single hull" ships.

Figure 2-38 The split TSHD

As a rule of thumb the following ratios between the discharge opening and the well surface are
used, dependent on the discharge material:
Page 31 of 109

wb3408B

Designing Dredging Equipment

for silt
for clay
average

10%
50%
30%

Instead of a large door or valve surface there are also systems that discharge the load with a
limited amount of doors or valves by partly fluidizing or eroding the load. Experience showed
that these systems function usually well for the fine sand types.
A design requirement for
discharge system may be the
necessity of dumping in shallow
water. Is this the case than sliding
doors or a splithopper are
options. Also cone valves
function well when discharging
in shallow water. With a small
opening they already provide a
good discharge. If doors are used
shallow dumping doors have to
be considered

Figure 2-39

2.2.4

The propulsion power

Except for the propulsion there are also requirements for the maneuverability of the trailing
suction hopper dredger. For this purpose extra bow thrusters are often used.
2.2.4.1

The propulsion power

Trailing suction hopper dredgers are real workships. They have a high block coefficient, no
high ship velocities and they often sail in shallow waters, which make them "feel" the bottom.
The velocities in knots do not exceed 1.4L (Figure 2.40).

Page 32 of 109

Chapter 2 Trailing suction hopper dredger


1.4*L^1/2

20

Loaded speed [kn]

18
1.22*L^1/2

16
14
12
10
8
6
8.5

9.5

10.5

11.5

12.5

13.5

SQRT(L) [m^1/2]

Figure 2-40 Maximum speed TSHD's

The ships resistance is composed of a number of components:

Rtotal = R f (1 + k ) + Rapp + Rw + RTR + RA


with

Further is:
Rf =

Rfl
1+k
Rw
Rapp
Rb

friction resistance according the ITTC-1957 formula


[N]
shape factor for the hull
[-]
wave resistance
[N]
resistance as a result of the appendage
[N]
resistance as a result of the additional pressure difference [N]

1
V 2 C f S total
2

with
Cf =

blog

0.075
10

Rn 2

Determination of the resistance demands a lot of experience. The average sailing speed in knots
for TSHDs is 1.22Length (0.63L for v in m/s) Figure 2.40. That means that the wave
resistance part is small and the total resistance can be estimated by a polynomial of the second
order.
Nevertheless the ships resistance of a trailing suction hopper dredger is considerably higher
under sailing conditions compared to normal ships with the same block coefficient. This is
caused by the bottom valves or doors and the suction pipe guides in the hull.

Page 33 of 109

wb3408B

Designing Dredging Equipment

R2

Fimpuls

R3

Rship
G1

pdraghead

R1

Rpipe

Fcutting+Rfriction
Rdraghead

G2

Figure 2-41 Forces working on a TSHD

The required propulsion power appears to be decisive under the trailing condition, in particular
when a combined drive is used. For this condition requirements are set regarding the trail
speed, expected counter current and effective cutting forces at the draghead.
For the trail speeds a normal value is 1.5 m/s with a counter current of 1 m/s. At these
velocities the resistance of the hull, as could be expected, is little. The largest resistance arises
from the dragging of the suction pipes over the seabed.
This suction pipe resistance is composed of several components:
The first, the hydro-visco components.
In the direction perpendicular of the pipe:
R pipe = C D

1
w v sin v sin LD
2

In the direction parallel with the pipe:


1
w v cos v cos LD
2

R pipe6 = C L
In which:
=
CD
=
CL
D
=
L
=
R pipe =

Drag coefficient
Lift coefficient
Pipe diameter
Pipe length
Drag force

Page 34 of 109

[-]
[-]
[m]
[m]
[N]

Chapter 2 Trailing suction hopper dredger


R pipe6

=
=
=
=

Lift force
Relative water velocity to the ship
Pipe angle
Density water

[N]
[m/s]
[]
[kg/m3]

The dimensionless coefficients CD and CL are apart from dependent on Reynolds number, also
dependent on the appendages on the suction pipe. For a more accurate calculation it is better to
divide the pipeline in different section with different projected areas. This has the advantage
that the relative velocity of the water can be dependant of the waterdepth
Another force that the propulsion has to generate, which is often forgotten, is the force needed
to accelerate the dredge mixture to the trail velocity of the ship, this momentum force.

FMom = Q mix v trail


with:
FMom
Q
vtrail
mix

=
=
=
=

Momentum force
Pump capacity
Trail speed
Density mixture

[N]
[m3/s]
[m/s]
[kg/m3]

The resistance of the draghead over the seabed.


This force is more difficult to determine, but it can be derived as follows:
During dredging erosion water shall enter the draghead at the backside and the sides. (See
chapter 2.5.1.1.3) This pressure difference depends on the type of soil and the amount of jetwater used to loosen the soil (chapter 2.5.1.1). An average value for this pressure difference is
50 kPa. Multiplying the suction area of the draghead with the pressure difference gives the
force that push the draghead to the seabed.
Additional to this is the weight of the draghead on the bottom, which can be determined with a
simple equilibrium equation. The coefficient of friction of steel on wet sand is 0.3 to 0.5.
Additionally it is known that the draghead "bulldozers". Therefore a coefficient of friction of at
least 0.5 must be used.
Teeth or blades mounted in the draghead with intension to cut a significant part of the soil do
increase the trail force significant. Effective trailing forces of 250 to 500 kN per pipe are
common for the big dredgers
If the total resistance of the suction pipe is known than this can be roughly converted to other
diameters using:

W1 D1
=

W2 D2

with = 2.2 2.4

In conclusion the required effective trail force(s) are strongly dependent on the expected type
of the dredging work and therefore to consider in detail during design.
The above consideration can be visually clarified in the resistance-propulsion power chart:

Page 35 of 109

wb3408B

Designing Dredging Equipment

Thrust-Resistance Diagram
Thrust / Resistance [kN]

Opeation point when trailing


2000
Operation point when sailing
1500
1000

Pipe only
Sailing

500

Trailing

0
0

Speed [m/s]

10

Figure 2-42

In Figure 2.42 the effective propulsion force (trust), T_sailing (corrected for wake) as the ships
resistance, R_sailing, are shown as a function of the ships speed. In the operating point
"sailing" the supplied power is equal to the ships resistance. Under this condition the main
engines are usually only driving the screws and the thrust curve is determined by the power of
the main engines. This propulsion force curve can be described by a second-order polynomial:

Tsailing = a 0 + a 1 v s + a 2 v 2s
During dredging the main engines usually drive, besides the screws, also the sand-pump
installation (sand- and water-pump) either directly or through a generator/electric motor set.
This means that less propulsion is available for the propulsion in this mode. Because the
propulsion force is proportional to the propulsion power as:

T2
= constant, the propulsion
P3

force curve is approximated under dredging (trailing) conditions by:

Ttrailing

FP I
=a G
H P JK
trailing

2
3

+ a 1 v s + a 2 v s2

sailing

The sum of the ships resistance (R_ship) and the suction pipe resistance (R_pipe) has to be
equal with this propulsion force curve (operating point "trailing"). Usually this condition
appears to be decisive for the to be installed power of the main engines. If no combined drive is
used than the "sailing" condition is normative for the required propulsion power.
2.2.4.2

The bow thruster power

Page 36 of 109

Chapter 2 Trailing suction hopper dredger


Maneuverability of THSDs has
improved much compare to the past. In
the sixties and the seventies the socalled bow jets (Figure 2.43) were
used. These made it possible to
generate a transverse force with the
sand-pumps. But for practical reasons
this was done only when the pumproom was positioned in the bow. The
effectiveness of these jets is pretty
good, certainly for 2 to 3 knots. The
construction costs are only a fraction of
those for a bow thruster.
Figure 2-43 Bow jet

However continuous use during dredging is not possible and so not economical.
Therefore this idea is abandoned and one or more bow thrusters are used. However bow
thrusters have the disadvantage of hardly any transverse force above 3 knots. There are
different types on the market.
A propeller mounted in a tunnel with a speed or pitch control, which means that the flow
direction and capacity is control by the revolutions and speed direction or by changing the pitch
of the propeller vanes. A axial flow pump by which the direction of the flow is control by
valves and the capacity by the speed of the impeller.

Figure 2-44 Thruster types

With the increase of the jet-pump power one could consider to use these, totally o partly, for
the bow jets.
The required bow thruster power depends strongly on the expected type of work for which the
trailing suction hopper dredger has to be designed.
Page 37 of 109

wb3408B

2.2.5

Designing Dredging Equipment

Power balance

From the above mentioned it shows that a lot of power is installed in a trailing suction hopper
dredger, that is:

the dredge-pump power


the jet-pump power
the propulsion power
the bow thruster power

and of course the power for the electrical circuit on board. After all the suction pipes have to be
lowered and raised. The valves and other auxiliary equipment must operate, etc. Powers of
15000 kW or more are no exception. Therefore it makes sense to take a close look to the power
balance. For instance, separate drives for the propulsion and the sand-pumps are not always
necessary or desirable. Most of the time several objects can be combined. The following will
show that this is strongly related to the suction pipe configuration.

Figure 2-45 Direct drive

The most common combination is to drive both the propeller as well as the dredge-pump with
one engine (Figure 2.45). The total installed power will not be much less than these units are
separate as shown in Figure 2.46 but during sailing more power is available for a higher sailing
speed and resulting in a higher production. If the units are driven directly, there will be no loss
in generators, cables and electric motors. The speed control of the sand-pump is however poor.
The engines run on constant speed, while adjustable propellers control the speed of the vessel,
while the configuration of Figure 2.46 has fixed propellers (Why?).

Figure 2-46 Separate propulsion and dredge pump engines

When the trailing suction hopper dredger needs pump ashore installation than generally an
extra transmission is installed in the gear-box to use the total available power for this
installation. The same engine supplies the jet-pump power usually. In that case the gear box is
fitted with an extra axis. The only disadvantage for this arrangement is the limitation in the
suction pipe length. Of course this is not totally black-and-white. Extending of the inboard
Page 38 of 109

Chapter 2 Trailing suction hopper dredger


suction pipe offers the possibility to place a longer pipe on the deck, but this results in a lower
production when dredging at large depths. Such a ship is put into service in 1992 and the
concerned company (J.F.J. de Nul) took this decision intentionally.

Figure 2-47 TSHD with dredge pumps in the fore ship

If limitation of the suction pipe length is not desired both powers can be combined with the
arrangement of Figure 2.47. In the engine room the main engines drive the adjustable screw,
but on the other side a generator is placed that supplies the dredge-pump placed in the fore ship
with energy. This is attended by an energy loss of 10 to 15 % of the power required. So for a
sand-pump power of 2000 kW times two, there is a loss of approximately 400 to 500 kW!
This also accounts for jet-pumps installed in the fore ship too. If the pump ashore installation
needs the total power of the main engines this solution will require a considerable larger
investment than the previous case. The speed control of the dredge pump can of course be well
adjusted with an electrical drive.
Between these two solutions there are of course all kinds of variants possible, which have been
built in the past too. (See chapter 2.26 Main Layout)
25000

y = 0.4641x - 510.11
2

R = 0.8741

Pp [kW]

20000
15000
10000
5000
0
0

10000

20000

30000

Displacement [t]

Figure 2-48 Propulsion power

Page 39 of 109

40000

50000

wb3408B

Designing Dredging Equipment

Bow trust power [kW]

3,500
3,000

y = 0.1758x - 19.495
R2 = 0.8036

2,500
2,000
1,500
1,000
500
0
0

2,000

4,000

6,000

8,000 10,000 12,000 14,000 16,000

Propulsion power during trailing [kW]

Figure 2-49 Bow thrust power

30000

y = 0.5806x

25000

R = 0.8931

Pi [kW]

20000
15000
10000
5000
0
0

10000

20000

30000

Displacement [t]

Figure 2-50 Total installed power

Page 40 of 109

40000

50000

Chapter 2 Trailing suction hopper dredger

2.2.6

Main layout

Now the main dimensions of the ship and the dredging equipment are known, the layout of the
ship has to be determined.
2.2.6.1 Single well ships
Most currently built trailing suction hopper dredgers are of the single cargo-hold type. The
hopper, also called well, is positioned somewhat forward of the middle of the ship. This is also
the case when the bridge is on the foreship. The engine-room is always positioned in the stern.
The trailing suction hopper dredgers used by the dredging industry are usually equipped with
two adjustable screws.

The position of the pump-room, a with watertight bulkheads sealed space in which the sandpumps are located, also has a large influence on the layout of the trailing suction hopper
dredger. The simplest and most efficient layout is the one where the pump-room is positioned
just before the engine-room (Figure 2.45).
In this case the main engines drive both the adjustable screws as the sand-pumps. Adjustable
screws are necessary in this case because if the sailing velocity of the trailing suction hopper
dredger is controlled by varying the number of revolutions of the engine then also the
production of the pump changes which can lead to production loss.
Since the sand-pumps on a trailing suction hopper dredger usually run on a fixed number of
revolutions (variation of the suction depth has only a limited influence on the required head)
the ships velocity can be easily adjusted by varying the pitch angle of the adjustable screws.
Of course adjustable screws are more expensive and vulnerable than fixed screws. If fixed
screws are desired than the layout shown in Figure 2.46 is appropriate with different engines
for the sailing and dredging.
An alternative for Figure 2.46 is Figure 2.51

Figure 2-51

It will be clear that in the first solution the total installed power is better used. After all during
sailing the full power of the engines is available for the propulsion. However these solutions
are also seen with adjustable screws.
In both cases the limitation of the arrangement is the suction pipe length and therefore the
suction depth. After all the suction pipes still need to be stored on board. If large dredging
depths are also required (until 70 m) than the layout of Figure 2.47 and 2.52 are automatically
Page 41 of 109

wb3408B

Designing Dredging Equipment

obtained. Figure 2.52 is called the All Electric Ship, an development of nowadays. All power
needed is delivered by the main engines via high efficient generators and motors.

Figure 2-52 The all electric ship

Of course there are may combinations possible with of these main layouts. The number of
suction pipes may have some influence. Many smaller trailing suction hopper dredgers have
only one suction pipe. Nevertheless these small trailing suction hopper dredgers are equipped
with twin screws for two reasons:
1. The empty draught determines the maximum allowed propeller-diameter. Transferring a
certain amount of power to one screw leads to a high revolutions, heavy loaded propeller
with a relatively low efficiency.
2. A twin screw ship has a much higher maneuverability than a single screw ship
Nevertheless, special trailing suction hopper dredgers such as gravel dredgers, are equipped
with a single screw (see special applications)
2.2.6.2 Twin Hopper Trailers
In the end of the sixties and starting seventies several trailing suction hopper dredgers were
build with two separate hoppers. In these ships the engine-room and/or pump-room is
positioned between the two hoppers. The main advantage of the twin hopper type is the smaller
longitudinal ships bending moment that arises from the mid-ships connection of the engineroom and/or pump-room bulkheads.

Figure 2-53

Page 42 of 109

Chapter 2 Trailing suction hopper dredger

Figure 2-54

The disadvantage of such ships that on one hand the hopper ratios are unfavorable for the
settling process and to the other hand the total capacity is dived over both hoppers which will
improve the sedimentation process somewhat. Besides several extra valves are needed to trim
the ship sufficiently. These layouts are shown in Figure 2.53 and Figure 2.54. The
accommodation is also positioned amidships. In both cases the main engines drive propellers
and dredge-pumps. Besides the longer pipes for large dredging depth can be installed. Of
course an electrical driven dredging installation is possible too.
2.2.6.3 Single well ships with a submerged-pump
For larger suction depths, more than 50 m, the installation of a submerged-pump becomes
economical. The submerged-pump, also called the suction pipe pump, can be driven electrical
or hydraulically. The hydraulic drive exists on smaller trailing suction hopper dredgers.

On larger trailing suction hopper dredgers the pump and the electrical drive with bearings are
accommodated in a compact compartment, directly mounted in the suction pipe. The number of
revolutions of the electrical drive is chosen such that it corresponds with the required number
of revolutions of the submerged-pump. This solution provides a compact and relative light
construction.
The submerged-pump related possible layouts of the engine rooms and/or pump-rooms are
shown in Figure 2.55.

Figure 2-55 TSHD with inboard (direct driven) and submerged pumps

Page 43 of 109

wb3408B

Designing Dredging Equipment

Figure 2-56 TSHD with inboard en submerged electrical driven pumps

Figure 2-57 Electrical driven pumps and direct driven propulsion

For smaller, simpler trailing suction hopper dredgers and converted barges submerged pumps
can be used to. For such ships the dredge installation is composed of modules (Figure 2.58).
The drive unit of a dredge installation is now positioned on the fore-deck. The (existing) engine
room is located in the stern. Therefore adjustable propellers are not necessary..

Page 44 of 109

Chapter 2 Trailing suction hopper dredger

Figure 2-58 Pump module on barges

2.2.6.4 Split hopper suction dredgers


Split hopper suction dredgers can in principle also be divided as shown in Figure 2.59 and 2.60.

Figure 2-59

Figure 2-60

With the observation that both the engine-room and the pump-room are divided in the
longitudinal direction (Figure 2.50 and 2.60).

Page 45 of 109

wb3408B

Designing Dredging Equipment

The engine driver has to ascend to a height higher than sea level when he wants to go from
starboard to portside.

Figure 2-61 Split TSHD

2.2.6.5 The position of the pump-room


Positioning of the pumproom near the engine-room instead of in the fore-ship has the following
advantages:

the control and the maintenance of the pump installation can be done in a simple way by
the engine-room personnel.
with an empty ship the suction intake is submerged deeper than in the fore-ship, as a result
of the trim.
as a result of the shape of the stern the dragheads will move less frequently under the ship
base, when working in shallow waters or on slopes.
the direct drive of the sand pump by the main engine is considerable more efficient than the
transport of energy from the stern to the fore-ship.
the total propulsion power can used easily for the pump ashore installation. With a foreship pumproom this requires considerable investments.

Of course there are also disadvantages:


the main disadvantage of the pump-room near the engine-room in the stern is the limitation
in the dredging depth of the suction pipe, something that has become more important in the
last few years.
the distribution of the weight is less ideal than with a pump-room in the front. For this
reason the bridge is positioned on the bow nowadays.
because the dragheads are nearer to the screws there is an increased chance for cables
picked up to get entangled in the propellers.

Page 46 of 109

Chapter 2 Trailing suction hopper dredger

2.3 Technical Construction


The technical construction of the trailing suction hopper dredger will be discussed in the flow
direction of the dredging process.

2.3.1

The dredge installation

2.3.1.1

The dragheads

Figure 2-62 Modern draghead (Vasco da Gamma)

The draghead is the suction mouth of the trailing suction hopper dredger and is, with the sandpump, one of the most important components of the dredge installation. Looking at the amount
of patent applications on the area of dragheads the conclusion can be made that there is a lot of
knowledge of the operation of this device. Unfortunately this is not the case, the last 5 years the
remarkable progress made about the understanding of excavation process in the draghead.
Dragheads must be able to break up the coherence of varied soil types. The excavation process
is done erosive, mechanical or by both methods.
Dragheads are designed to resist the forces, needed to loosen and suck up the soil. They also
need to be strong enough to withstand collisions with unknown objects in the dredge area. This
especially gives high demands on the reliability of the equipment mounted on the draghead to
control the water supply and/or cutting blade depth.
Page 47 of 109

wb3408B

Designing Dredging Equipment

In general draghead consist of a fixed part that is connected to the suction pipe, sometimes
helmet mentioned and a one or two pivoting part(s), the visor, which is mounted in the fixed
part. The last part is (self) adjusting to keep in fully touch with the seabed.
In the dredging industry different types of dragheads are used. The most known dragheads are:

the Hollandse (Dutch) draghead, also called IHC draghead (Figure 2.63 and 2.65)

the Californian draghead (Figure 2.64 and 2.66)

Figure 2-63 Dutch draghead

Figure 2-65

Figure 2-64 Californian draghead

Figure 2-66

Both type are developed based on the principal of erosion generated by the dredge pump flow.
Nowadays these dragheads can be equipped with water jets too (Fig 2.65 and 2.66)
In addition to the excavation of the soil, the jets are also important for the forming of the
mixture in the draghead.
The dragheads rest on the seabed by means of replaceable, so-called, heel-pads of wear
resistant material.
When dredging cohesive materials the dragheads are provided with blades or cutting teeth
mounted in the visors. The position of the visor is fixed relative to the helmet corresponding
with the average dredging depth Sometimes this position is controlled by hydraulic cylinders.
Page 48 of 109

Chapter 2 Trailing suction hopper dredger


When using tooth and/or blades one has to realize that different items can hook on to the
draghead, causing high longitudinal forces in the suction pipe. This can be prevented by
dividing the fixed part, the helmet, (Figure 2.67) in two parts, connected with a hinge on the
top and breaker bolts at the bottom. The strength of the breaker bolts has to be slightly weaker
than the weakest link of the several components of the suction pipe.

Figure 2-67

Page 49 of 109

wb3408B

Designing Dredging Equipment

However, if they are designed such that they fail regularly than soon the two parts are welded
together with the danger that the next link fails.
Modern dragheads have one visor with jet nozzles over full width. At the backside of the visor
replaceable teeth are fitted. The purposes of these teeth are to remove not eroded sand bands
and to guide the flow in the direction of the suction pipe. Some of those dragheads do have
movable water flaps to control the diluting water to the draghead. Visors can be adjusted either
by bars or by hydraulic cylinders.

Figure 2-68 Modern dragheads

Furthermore fenders are mounted


on the draghead, to prevent damage
caused by the bumping of the
draghead against the hull. By
mounting these fenders on both
sides, the draghead can be used
both on starboard and port.

Figure 2.69 fenders of the draghead


of the One piper TSHD Volvox
Terra Nova
Figure 2-69 Fender for protection

The connection between the movable visors and the fixed helmet is usually sealed with a
rubber strip. This prevents the entering of "strange" water and it decreases the wear caused by
the sand picked up by this "strange" water.
2.3.1.1.1 Other types of dragheads
In the last 25 years a lot of experiments are performed with several types of dragheads, like:
Page 50 of 109

Chapter 2 Trailing suction hopper dredger

Figure 2-70 Silt draghead

The silt head (Figure 2.70). A draghead specially designed for dredging silt and soft clays. The
silt is pushed in the draghead, while the propulsion delivers the required force.
The active draghead (Figure 2.71)
A draghead with a hydraulic driven roller with
cutting tools, able to cut firm clay or compact
sand.
The disadvantage of this dragheads was the
ability to pick up cables and wires

Figure 2-71 Active draghead

And
The venturi head (2.72). A draghead that would be
hydraulically better shaped than the Hollandse and
the Californian draghead and therefore would reach
higher productions.
The advantage of this draghead was the high trailing
force due to the pressure difference over the draghead
.

Figure 2-72 Venturi draghead

All these dragheads were not successful. Mostly the idea behind was good, but secondary
reasons. like wear, sensitive for dirt, difficult to handle, etc. etc. Resulting in lower average
productions than the earlier mentioned dragheads.
Page 51 of 109

wb3408B

Designing Dredging Equipment

Aside from IHC there are also other companies that supplies dragheads. Usually these draghead
are named after the company since they differ somehow from the standard dragheads.
Examples are the "Van de Graaf-heads" and VOSTA. heads
Furthermore every dredging company with self-respect has developed its own draghead,
whether or not used.
2.3.1.2

The suction pipe

Figure 2-73 Suction pipe

The purpose of the suction pipe (Figure 2.73) is to make a connection between the seabed and
the ship in order to make transport of dredge slurry possible. Because a fixed connection is not
possible due to a varying water depth and the forces in size and direction, they have to comply
with a number of important requirements:

the dredging depth must be adjustable.


there must be enough freedom of movement to maintain the connection with the seabed as
good as possible.
the bending moments due to the forces acting on the pipe should be kept as small as
possible for reasons of strength and weight
hit- and shock load resistant.
a small pipeline resistance for the mixture flow.

Page 52 of 109

Chapter 2 Trailing suction hopper dredger


The trunnion slide (Figure 2.74) that slides between the hull guides during the raising and
lowering of the suction pipe, is fitted with tapered cams that push the trunnion slide against the
hull when the suction pipe is in front of the suction intake.

Figure 2-74 Trunnion slide with elbow

Mounted on this trunnion


piece is a casted elbow,
which can rotate around a
horizontal
axis,
perpendicular to the hull.
This hinge construction
allows the suction pipe to
be lowered to the desired
depth. The elbow has two
arms, positioned in the
vertical plane of the
suction pipe. On these
arms, the upper or short
piece pipe is mounted
with hinges.This upper
hinge makes the bending
moments
small,
for
example for the case

where the ship is swayed aside by the current.


Between the elbow and
the upper pipe a rubber
suction
hose
is
mounted that can move
40 to both sides. Steel
rings are vulcanized in
this suction sack to
prevent a collapse of the
suction sack by the
subpressure as a result
of the suction. The
upper pipe is connected
with the lower pipe by
Figure 2-75 Universal joint
the gimbal (Figure
2.75) and a second
suction sack. This gimbal allows the two pipes to move independently, which is necessary in
heavy weather and/or an irregular sea bottom.
A turning gland (Figure 2.76) is mounted, usually directly behind the gimbal, in the lower
pipe. This allows the lower pipe to rotate around its longitudinal axis, so that the draghead can
also follow the bottom profile in the transverse direction.

Page 53 of 109

wb3408B

Inner pipe

Wear ring

Designing Dredging Equipment

Outer pipe

Wearing ring

Wearing ring

Inner pipe

Lip seal

For small diameters (<900 mm)

Outer pipe
For large diameters (>900 mm)

Figure 2-76 Turning glands

If the draghead is fitted for jet-water, a jet-water pipeline is mounted along the suction pipe
(Figure 2.77).

Figure 2-77 Suction pipe with a jet water pipe

Because this pipeline also needs to


follow all suction pipe motions, a
lot of pressure hoses and elbows
are needed, causing additional
pressure losses in the jet-pipeline.
The connection of the suction pipe
with the ship becomes now more
complicated.

Figure 2-78 Jet pipeline passing the universal joint


Page 54 of 109

Chapter 2 Trailing suction hopper dredger

It becomes even more


complicated
when
a
submerged pump is mounted
together with the suction pipe
(Figure 2.79). Except the
pipelines, a lot of cables for
power supply and to control
the pump speed are necessary.
For the powerful pumps a
special frame is necessary to
carry the loads.

Figure 2-79 Submerged sand pump frame

2.3.1.3

The suction pipe gantries

Figure 2-80 Suction pipe gantries

The three suction pipe gantries serve to move the suction pipe either inboard or outboard.

The draghead gantry and the middle gantry are carried out mostly as an A-frame, connected
with the main deck by a hinge-construction (Figure 2.81 and 2.82). A hydraulic cylinder or the
hoisting wires controls the motion when moving the suction pipe in- or outboard.

Page 55 of 109

wb3408B

Designing Dredging Equipment

Figure 2-81 3 different types of suction pipe gantries

Figure 2-82 Suction pipe elbow gantry

The suction elbow gantry consists of a fixed and a moveable part. The fixed part is welded to
the main deck and is fitted with tracks for the wheels of the moveable part. (Figure 2.82). When
the moveable part has reached the lowest [position than the trunnion slide can be lowered into
the guides in the hull

Page 56 of 109

Chapter 2 Trailing suction hopper dredger


2.3.1.4

The swell compensator

Figure 2-83 Swelll compensator

Elektrical driven draghead winch

Switching relays

Draghead
winch
controller

Air-oil vessel

Swelll compensator

Suction pipe
Dragheadp

Figure 2-84 Swell compensator with


draghead winch controller

Page 57 of 109

The swell compensator has contributed to the


success of the trailing suction hopper dredger
too. The most important goal of the swell
compensator is to maintain the contact between
the seabed and the ship, due to either both ship
motions or the irregularities of the bottom
contour. The swell compensator is positioned in
the hoist-cable system of the draghead winch
gantry. The swell compensator prevents the
uncontrolled slackening and re-tensioning of the
hoist cables. (Figure 2.83):
Furthermore it maintains almost a constant
pressure of the draghead on the seabed. A swell
compensator system consists of the following
components:
An hydraulic cylinder, of which the head is fitted
with one or two pulleys that guide the hoist cable
of the draghead.
One or more pressure vessels, of which the lower
part is filled with oil and the upper part with air.
A oil pump and reservoir.
An air compressor.
A pipeline system that connects the hydraulic or
pneumatic components.

During an ascending motion of the ship the


piston rod of the compensator is pushed
downward as a result of the increasing force in
the cable. The plunger then compresses the air in
the pressure vessel. During the following
descending motion of the ship the piston is
pushed out again as a result of the increased
pressure in the pressure vessels. This assures a
tight cable at all times.
The average pressure in the pressure vessels is
determined by the weight with which the
draghead may rest on the bottom, or better: how
much the swell compensator has to compensate
this weight. It will be clear that the compensation
in silt will be higher than in sand. In table 1
values are given as a guideline by IHC for a
certain configuration.

wb3408B

Designing Dredging Equipment

Table 1.

Suction depth

Draghead weight on bottom


Fill air pressure
P in
P midstroke
P out
Draghead weight on bottom
Fill air pressure
P in
P midstroke
P out
Draghead weight on bottom
Fill air pressure
P in
P midstroke
P out

25 m

17.5 m

10 m

2.3.1.5

kg
bar
bar
bar
bar
kg
bar
bar
bar
bar
kg
bar
bar
bar
bar

Compensation
Mud
Sand
80% 50% 50% 20%
1800 4500 4500 7200
15.0
15.0
8.0
8.0
26.2
17.9
18.6
9.8
24.7
17.1
17.1
9.4
23.3
16.4
15.8
9.0
2080 5200 5200 8320
15.0
15.0
8.0
8.0
30.0
20.1
21.0 10.8
27.9
19.1
19.1 10.3
26.1
18.2
17.5
9.8
2190 5475 5475 8760
15.0
15.0
8.0
8.0
31.4
20.8
21.8 11.2
29.1
19.8
19.8 10.6
27.1
18.9
18.1 10.1

The suction pipe winches


Suction pipe winches have a grooved
winding drum, with a length and /or
diameter such that the there are 5
windings left on the drum (Figure
2.85) when the suction pipe is in its
lowest position. When the suction pipe
is out of the water. The load of the
winches becomes heavier. To
overcome this problem the wire is
transport to a drum with a smaller
diameter, which results in a lower
torque for the winch drive.

The winch drives is either electrical or


hydraulically.

Figure 2-85 Suction pipe winch

Page 58 of 109

Chapter 2 Trailing suction hopper dredger

2.3.1.6 The dredge pump


The dredge pump is the heart of the trailing suction hopper dredger.

The position of the dredge or sand


pump has to meet certain requirements,
certainly for the case without a suction
pipe pump:

Figure 2-86 Dredge pump

1. The inboard placed dredge pump


must be installed as low as possible.
The deeper the pump is under the
water level, the higher the
concentration of the mixture can be.
2. The resistance of the pipeline must
as low as possible. So short suction
pipes, wide bends and no
constrictions.
3. The direction of rotation of the
pump has to comply with the
rotation direction of the mixture
caused by the bends in the piping
system.

The second requirement cannot always be met because of demands for maintenance or the
accessibility for inspection or removal of debris.
There are also some practical objections concerning the third requirement. To comply with it
the direction of rotation of the starboard and port pumps has to be opposite. This means more
different spare parts like pump casings, impellers etc.
Speed control of the dredge pumps is highly dependent on the type of drive. If the main engine
directly drives the sand pump then speed regulation is not possible or only by stepwise control
using a gearbox. Is the dredgepump driven by a separate diesel engine then speed control is
possible, but the best control is obtained by an electric drive. It has to be mentioned that
currently new developments in variable transmissions come available for diesel engine driven
pumps.
In most cases the requirements regarding the cavitation properties of the dredgepump are more
important than the pressure properties. After all, even if the trailing suction hopper dredger has
a pump ashore system, operations in dredging mode are considerably more frequent than the
pump ashore mode.
Both single walled and double walled pumps (Figures 2.87 and 2.88) are used in trailing
suction hopper dredgers, dependent on the view and strategy of the dredging company. Double
walled pumps have a separate inner pump casing that can be worn out without necessary
repairs. This is achieved by pressure compensation. The pressure in a running pump is equal
inside and outside the inner pump casing. To do this the space between inner and outer pump
casing is filled with water and pressurized. Besides the advantage of a longer lifetime for the
inner pump, this type of pumps gives a higher security in case of explosions.

Page 59 of 109

wb3408B

Designing Dredging Equipment

Figure 2-87 Single wall dredge pump

Figure 2-88 Double wall dredge pump

Page 60 of 109

Chapter 2 Trailing suction hopper dredger


2.3.1.7 The jet-water pump
The water- or jet pumps are usually also positioned in the pump room. If these pumps are
implemented with "clean-water pumps" than attention has to be paid to the position of the
water inlet. After all contaminated water causes extra wear. Because the water surrounding the
trailing suction hopper dredger is usually very muddy due to the overflow water, nowadays
dredge pumps or weir resistant water pumps are used jet pumps.
2.3.1.8 The discharge pipeline
The discharge pipeline connects the dredge pump and the hopper loading system, or the dredge
pump and the shore pipeline. Every trailing suction hopper dredger has the possibility to
discharge the dredge mixture directly. Previously this was done above the waterlevel, but with
increasing environmental protection demands, the so-called poor mixture installation (Figure
2.89) is connected with an always submerged pipe-end.
FLOW

doorsnede

doorsnede

FLOW

Figure 2-89 Poor mixture overboard systems

Page 61 of 109

wb3408B

Designing Dredging Equipment


Trailing suction hopper dredgers with one
suction pipe do have one delivery pipe
constructed over the middle of the hopper
and connected the discharge side of the
dredge pump. Trailing suction hopper
dredgers with two suction pipes can also
have one central delivery pipe (Figure
2.90) on which the discharges of both
dredgepumps are connected, or two
separate delivery pipes (Figure 2.91).

Figure 2-90 TSHD with one delivery line

In
this
last
configuration it must
be possible to use both
delivery pipes with
both pumps. When one
of the suction pipes
cuts of, whatever the
cause may be, the ship
still must be loaded
equally athwart-ships
to prevent listing. This
requires more valves
than for one central
loading gully.
Figure 2-91 TSHD with 2 delivery pipes

A similar complexity of the piping system arises also when shore pumping must be possible
over starboard, port and over the bow. In a shore pumping installation the pressure pipe usually
ends in a ball on which the shore piping can be connected. The bends are usually from cast
steel for maintenance reasons.
Page 62 of 109

Chapter 2 Trailing suction hopper dredger


On every trailing suction hopper dredger it must be possible, whether or not it is equipped with
a shore pump installation, to suck the water from the well. If poor settling material is sucked
than it is strongly recommended to discard the water that is left in the well when dumping,
before suction to prevent dilution of the sucked up mixture.

2.3.2

The hopper

2.3.2.1 The loading system


The goal of the loading system is to dump the sucked up sand-water mixture as quiet and even
as possible in the well. Three systems can be distinguished:
the diffuser system (Figure 2.92).
the central loading system (Figure 2.93).
the deep loading system (Figure 2.94).

All with several variants on which many have explored their creativity.
In the diffuser system an open
diffuser is positioned at the end
of the delivery pipe, which
discharges just under the
highest overflow level. With
such a system a good width
distribution can be achieved. A
disadvantage of the open
diffuser is the reasonable
amount of air that is taken in,
which can obstruct the settling.
Therefore closed diffusers are
used sometimes that always
discharge under the overflow
level.
The
system
is
maintenance friendly of the
system, compare to deep
loading systems

overvloei

Waterniveau

Figure 2-92 Diffuser system

Via closed diffuser the mixture


is
dumped
through
a
distribution box in the middle of
the hopper. The mixture flows
to both sides of the hopper,
where adjustable overflows are
fitted. Theoretically the hopper
load remains equal, if the flow
remains 2D. The turbulence
degree will decrease due to the
distribution of the flow rate to
two sides. An additional
advantage of this system is that
due to the overflows on both
sides of the hopper the ship can
achieve even keel more easely.
Page 63 of 109

overflow
Discharge pipe

centrale Discharge

Figure 2-93 Central loading system

wb3408B

Designing Dredging Equipment

overvloei

Figure 2-94 deep loading system

In a deep loading system the mixture is discharged deep in the well, whether or not with a
vertical diffuser. The advantage of such a system is the energy reduction that is achieved as a
result of the contact of the mixture with the already settled material. Another advantage
mentioned the energy profit as a result of the siphon effect. In principle this is true, but there
are quite a number of trailing suction hopper dredgers with a deep loading system for which it
doesn't count because the delivery pipe is not airtight. Fitting of a simple kind of heavy loading
or distribution valves in the delivery line causes this. These valves are necessary dredging
coarse sand coarse or gravel. Than the settlement is that good that when these valves are not
fitted the material settles immediately at the inlet and it becomes impossible to fill the hopper
evenly (Figure 2.95). This results in a uneven trim vessel with water on their load

Diffusor

Distribution
valves

Delivery pipeline

water
Sand

Figure 2-95 Distribution valves in the delivery pipeline, necessary for coarse material

Apart from that the take-in of air largely reduces the advantage of the deep loading system.
Another disadvantage is that it is very hard to discharge the mixture evenly distributed over the
width of the hopper. This causes jets with turbulence production with as a result possible
disturbance of the already settled material.
A combination of the diffuser system and the
deep load system is the diffuser box, which is
placed half way the hopper height

Water level

Figure 2-96 Box diffuser system


Page 64 of 109

Chapter 2 Trailing suction hopper dredger


2.3.2.2 The well shape
The well shape has to comply with the following requirements:

the least as possible obstructions in the well to keep the turbulence degree as low as
possible in connection with the settling.
as straight as possible side walls, preferably angling inward to improve the discharge of the
load.
easy accessible for maintenance.
sand level above outside water level at least when the ship is in maximal draught, but
preferable also at restricted draught (50-60% of maximum pay load).

The goal of the well or hopper is that the dredged material settles while the surplus water leaves
the hopper through the overflow.
These overflow losses are largely dependent on the parameter Q/(L*B)/w and less on
Q/(B*H)/w. The first parameter is the ratio between the time a particle needs to settle and the
time it stays in the hopper. The second parameter is the ratio between the horizontal velocity in
the well and settle velocity of the particle and is a measure for the turbulence degree in the
hopper. For a good settling a long narrow and shallow hopper shape is therefore favorable.
A danger is however that no equal distributed load over the length of the hopper can be
obtained which results in a need for distribution valves in the delivery pipe. These valves
decrease the settle length the final result can become worse. Besides, long small ships with a
limited depth results in small engine room(s). A compromise between price and performance
has to be found.
In the years past the obstructions in the hopper became less and less, as can be seen in the
following cross sections (Figure 2.97):

Page 65 of 109

wb3408B

Designing Dredging Equipment

f
Figure 2-97 Different hopper cross sections

Page 66 of 109

Chapter 2 Trailing suction hopper dredger

The last years hoppers


with a V-shape become
more and more popular

Figure 2-98 V-shape cross section

A well-shaped hopper (Figure 2.99) without any obstacle is formed by the split hopper suction
dredger. There are no bars or obstacles, because the ship has no doors or valves but splits in
two parts. The largest split hopper suction dredger built, has a deadweight of 7000 ton.

Figure 2-99 Split TSHD

Page 67 of 109

wb3408B

Designing Dredging Equipment

The installation of pump ashore systems, as well as the requirement for easy maintenance have
caused that, in general, closed hoppers hardly build, although they have certain advantages.
(Figure 2.97e)

In heavy seas rolling and pitching of the ship with a open hopper causes water movements
and splashing over the deck of the mixture. A ship having with a closed hopper and a small
overpressure, the water displacements during the rolling and pitching will be much less,
which improves settling.

The free space on the deck of a closed hopper is also seen as an advantage. Especially
during mobilization, the trip from one job to another, when all kinds of equipment can be
stored on the deck. During dredging these have to be removed to increase the deadweight
of the ship.

2.3.2.3 The overflow type


At present almost all trailing suction hopper dredgers are built with a continuous adjustable
overflow (Figure 2.101 & 2.102). Besides that most trailing suction hopper dredgers are of the
so-called Constant Tonnage System, which requires a continuous adjustable overflow system.

Figure 2-101 Adjustable overflow over the full width of the


hopper

Figure 2-100 Overflow with environmental


valve

Page 68 of 109

Figure 2-102 Standard adjustable overflow.

Chapter 2 Trailing suction hopper dredger

There are however differences in the shape and place of the overflow to in order to increase the
effective settling length (Figure 2.103 and 2.)

Figure 2-103 Flow of two round overflows on the side

Figure 2-104 Flow of straight overflow at the end

A requirement that gets increasing attention is the environmental friendly overflow.


Environmentalists do not liked a beautiful silt-jet behind the dredger. Dredging is often
associated with polluted silt, so everything visible behind the dredger is polluted. A method
to reduce the visibility of the overflow losses is to prevent the intake of air by the flow. This
means that the overflow has to work as a non-free fall spillway instead of a free fall spillway.
This can be done by building a so-called environmental valve (Figure 2.100) in the overflow.
However, it is of course much better to design the overflow such that it works as an imperfect
weir. This leads to a higher head (the height of the fluid surface above the upper side of the
overflow).

2.3.2.4 The discharge system


As said earlier, discharging the load can be done in two completely different ways, either by
dumping or by pumping.
2.3.2.4.1 Dumping systems
The goal of the dumping system is to discharge as quick as possible the material dredged with
great effort.

All kind of systems are available. Expensive conical valves (Figure 2.105a), simple bottom
doors (Figure 2.105b), horizontal sliding doors or valves (Figure 2.105c) or a ship that splits
totally in two halves (fig 2.105d). There are also several exotic systems (fig 2.105f to 2.105h)
all with their specific advantages and disadvantages. The lijster valve (Figure 2.105f) is very
Page 69 of 109

wb3408B

Designing Dredging Equipment

expensive and takes a significant loss of hopper space. Recesses valves (Figure 2.105g)
influence the stability unfavorable and necessitate a larger hull.

Figure 2-105 Different discharge systems

Requirements for his dumping systems

First of all the ship has to be able to discharge the load in a short time, as completely as
possible (so without any load left) and for all types of soil. This means that the discharge
area has to be large enough. Dependent on the dredged material the discharge-area ratio
(the ratio total discharge-area/ horizontal hopper area) increases from 10% for slurries to
50% for the cohesive soil types. For general useable ships this will be about 30% of the
hopper area. As already mentioned in chapter 2.2.3.10 the discharge is better as the outflow behaves like a plane symmetrical flow. The length/width ratio of the discharge
opening has to apply to L 3B.
Furthermore as few as possible protruding parts are allowed in the hopper, they can cause
bridging of the material. Additionally they have the disadvantage of forming an obstruction
for the settling too.
An proper sealing under all circumstances. This demand increases in importance when
(polluted) silt is dredged.

Page 70 of 109

Chapter 2 Trailing suction hopper dredger

Little or no influence on the ships resistance.


Maintenance friendly. Places where wear can occur have to be easily accessible.
Possibility for discharging the load in shallow waters and grounded ship.
Regarding the first
requirement
the
doors have the
advantage over the
others and for the
last four demands
the conical valve or
the
split-hopper.
Dumping
in
shallow water can
also be achieved
with
so-called
shallow dumping
doors.
(Figure
2.106).
Figure 2-106 Shallow dumping doors

The operation of the dumping system is mainly done by a hydraulic system. For the doors and
the valves the cylinders are positioned vertical. The doors or valves in this system can be
operated in groups, usually three. In every group the hydraulic system controls both the
starboard and the corresponding port cylinder.

For the horizontal sliding bottom


valves two cylinders positioned in
the longitudinal direction of the
ship activate those. Both cylinders
move simultaneously, so all doors
are open at the same moment.

Hopper

Cilinder for closing


the valves
valves

Bottom plating

Cilinder for closing


the valves

closed
open
Discharge apertures

Figure 2-107 Sliding bottom valves

The split hopper dredger has a hopper without obstacles and in opened position one large
discharge opening (plane symmetrical flow) and therefore a high discharge velocity, especially
Page 71 of 109

wb3408B

Designing Dredging Equipment

useful to dump submerged dams. The split hopper dredger can under grounded conditions
discharge well. The frequently mentioned advantage of well discharging cohesive soils is
disappointing in practice. Usually the bottom plates in the hopper, even in opened position are
insufficiently steep to be assured of a good discharge (Figure 2.108).
For a split hopper dredger dumping is done
by the splitting of the ship in the longitudinal
direction. The two halves are connected with
hydraulic cylinders and hinges. Of course
the deckhouse and the accommodation
remain upright during the splitting, because
it is connected with the deck by hinges and
hinge rods.
A

Figure 2-108 Split hopper dredger


C

Figure 2-109 Different mechanism

2.3.2.4.2 The pump ashore system


Except for direct discharge or dumping, it can be desirable for certain works to pump straight to
shore, not only for technical reasons but sometimes also for financial reasons. In principal
direct discharge and re-handling with a cutter suction dredger is cheaper, however several
important financial conditions have to be met:

The work must have a sufficient size to earn back the mobilization costs of an extra cutter
suction dredger.
This also counts for the re-handling pit, from which the cutter suction dredger pumps the
dumped sand to the reclamation area. This can be positive if such a dump can be situated
within the work.

If the work is done with more trailing suction hopper dredgers it is in many cases beneficial to
discharge directly and re-handle the sand. Because, even having two identical trailing suction
hopper dredgers on the job, the stochastic behavior of the dredging process causes that at a
certain time that the two ships arrive at the same time at the connection point for pumping
ashore, causing waiting for one of the dredgers.

Page 72 of 109

Chapter 2 Trailing suction hopper dredger


But there are also works where the direct pumping ashore or so-called rainbowing has large
advantages. For example works at sea like beach nourishments. For that goal small trailing
suction hopper dredgers are equipped with pump ashore equipment. There are also jobs without
space for a re-handling pit.

Figure 2-110 Rainbowing

Besides, there are jobs requiring controlled dumping of their load at a certain depth and in a
relative small area. Then the material is pumped back through the suction pipe. This has been
the case at the Oosterschelde works and is done too when covering pipelines.

Upper doors
Valve
Valve

Dredge pump

Self discharge channel


Water intake

Flow direction

Bottum valves

Figure 2-111 Longitude cross section pump ashore system

The decision to equip a ship with a pump ashore system is not taken just before the work needs
it. Except for the fact that the preparation and the fitting time can be more than half a year, it is
also much more expensive than when it is fitted directly during the construction of the ship.
Ships initially not fitted with a pump ashore system dont have mostly today. Nowadays the
European dredging contractor usually chooses for a pump ashore system.
A pump ashore discharge system consists of one or two suction channels, situated at both sides
of the center-keelson (Figure 2.112 under) or a pipe centrally placed in the center-keelson
(Figure 2.112 upper). In the first case the top of this suction or self-emptying channel is fitted
with so-called top-doors, by which the sand can be supplied into the channel. Transport water is
mostly supplied in two ways, first through the channel, which is connected in some way with
Page 73 of 109

wb3408B

Designing Dredging Equipment

the outside water and second by the jetpumps that fluidize or erode the sand in the surrounding
where the sand has to enter the channel.
The mixtures pumped ashore with a well-designed installation do have very high densities. For
example 7500 m3/h in a 800 mm pipe. Of course this is also dependent on the type of sand.
The rest load, the load that cannot or hardly be removed, is a measure for the design of the
shore pump discharge system. For the mono-hull ships it may not be more than 5% of the total
load.

Rods for opening


and closing
Upper
door
Suction channel for
self-discharching

Rubber seal

Rubber seal

Pivot

Bottom door

Figure 2-112 central discharge pipe line (above) and


channels on both sides of the keelson (under)

In split hopper dredger the self


discharge channel(Figure 2.113) is
situated exactly in the middle,
between the connection of the two
halves.
For split hopper dredgers this rest
load is zero, except for cohesive
materials.

Self empty channel


Figure 2-113

Page 74 of 109

Chapter 2 Trailing suction hopper dredger


Except for trailing suction hopper dredgers having besides bottom doors or valves, a pump
ashore discharge system, there are also trailing suction hopper dredgers without a bottom
discharge system, but with a self-discharge system. This is usually seen on aggregates hopper
dredgers. The self-discharging happens mechanically, either with a dredging wheel (Figure
2.114) or with a clamshell that grabs over the full width of the hopper.

Figure 2-114 Dredging wheel unloader (Left) and clamshell unloader (right)

2.3.3

The propulsion

Trailing suction hopper dredgers in general two controllable pits propellers. (see also chapter
2.25) Only in the sea mining industry trailing suction hopper dredgers with only one screw can
be found, whether or not controllable pitch. The advantage of controllable pitch propellers has
to do with the method of operation of the ships. On one hand the ship needs enough propulsion
power at relative slow speed of 2 to 3 knots to drag the suction pipes over the seabed. On the
other side the sailing speed from and to the borrow area should be as high as possible, normally
between 12 and 15 knots. TO fulfill both requirements the propellers are placed within nozzles.
Additionally the concept of double and adjustable screws strongly improves the
maneuverability.
A trailing suction hopper dredger needs surely good maneuverability. For instance dredging
along a quay wall with a ship with a length of 100m or more on a distance of less than 10m.
When maintaining harbors trailer dredgers always moves in shipping lanes. This in contrast
with merchant shipping stays in the harbor as short as possible. The maneuverability has
strongly improved over the years. Not only by installing more powerful bow thrusters and in
some cases even aft thrusters, but also by (special) rudders with large angles

2.3.4

The maneuverability

The trailing speed of trailing suction hopper dredger dredges is 2 to 3 knots (1 to 1.5 m/s). At
this velocity the maneuverability needs to be high. After all the higher the maneuverability the
less the over-dredging (outside the tolerances) and the less a chance on collisions there will be.
Therefore most trailing suction hopper dredgers are equipped with double propellers and one or
more bow thrusters. If Dynamic Positioning/Tracking (DP/DT) is stern thrusters are sometimes
installed too. To maneuver the following options are available on a trailing suction hopper
dredger:

Just rudders
Just the adjustable screws
Just the bow screw and/or stern screw

Page 75 of 109

wb3408B

Designing Dredging Equipment

A combination of these

Which possibility will be used depends strongly on the direction in which the ship has to sail
and the effectiveness of the various options under certain circumstances. The thrusters are only
effective for very slow forward velocities. Above 2 to 3 knots the effect is mostly gone, the
combination of propellers and the rudders are in that case a better option. However, the
maneuverability is also strongly dependent on the center position of the rudders in relation to
the propellers. On trailing suction hopper dredgers these are usually positioned more inboard in
relation to the direction of the propeller shafts to be able to exchange the propellers without
removing the rudders. Turning with one propeller forward (port) and one backward (starboard)
with both rudders fully starboard is now less effective than the starboard propeller full ahead.
After all in the first case the port propeller will hardly exert any force on the rudder.
Is a transverse movement desired and the ship
is equipped with both a bow and stern thruster
than it is logical to use these. If there is no
stern thruster available the transverse
movement can be generated by rotating the
adjustable screws opposite (Figure 2.115).

Figure 2-115 Opetration of adjustable screws

Page 76 of 109

Also the effects of flow during dredging have


to be compensated either by the bow thruster
or delivering more power to one of the
propellers than the other.

Chapter 2 Trailing suction hopper dredger

2.4 Strength and stability


2.4.1

Strength

Every sea-going vessel longer than 24 m, and


therefore also a trailing suction hopper
dredger must have a load line assigned
according international agreed rules. The free
board is the distance from the load line to the
top of the main deck. The size of this free
board is indicated on the vessel both on port
and starboard by the Plimsoll-line1 (Figure
2.116) (Samuel Plimson let the English
Parliament approve an act in 1876 that had to
prevent the overloading of ships).
This line indicates, except for the allowed
loading level in several different waters, also
the initials of the registering agency of the
ship.

Top of main deck

TFW
FW
L

T
S
W

or:

WNA

Figure 2-116 Plimsoll mark

Every seaship loaded to the International Free Board Line, has to comply with certain
demands for strength. In principal there are two demands:
1. demands of strength concerning the loading of the ship until the allowed draught on flat
water.
2. demands of strength concerning the wave forces on the ship
For this last condition a distinction is made of the working areas of the ship. The so-called
classification:
1. Deep sea ( haute mer). Is assigned to ships capable for transoceanic navigation.
2. Great coasting trade (grand cobotage). Assigned to ships deemed suitable to perform deep
sea voyages but not transoceanic navigation.
3. Small coasting trade (petit cabotage). Assigned to ships that may not sail further from the
coast than a distance from the coast that they can reach a save harbor or mooring place
within six hours.
4. sheltered waters (eaux arbitres). This class is assigned to ships that are allowed to sail,
usually under good circumstances, at most at a small distance from the coast (mostly less
than 15 miles).
Above mentioned classification, of the Bureau Veritas, is international acknowledged, as well
as those of other classification bureaus (Lloyds Register, Germanische Lloyd, Norske Veritas,
American Bureau of Shipping and others).
In the dredging industry there is a by local authorities allowed draught, known as the dredging
mark. That is the allowed draught that is usually set in the middle between the international free
Page 77 of 109

wb3408B

Designing Dredging Equipment

board and the top of the main deck of the ship. The ship must of course be able to carry the
loads that can arise under such circumstances.
Trailing suction hopper dredgers that are loaded to the dredging mark are not allowed to make
international trips.
Except for classifications there are also notations that are related to the rules for building
specialty ships. Both the trailing suction hopper dredger as the stationary suction dredgers are
assigned to those rules.

2.4.2

Stability

Except demands regarding the strength, a ship has to comply too with a minimum stability. For
sea-going ships the international demands apply, dependent on the type of the ship. For trailing
suction hopper dredger in principal the same rules apply as for sea-going cargo vessels.
Definition:

Stability is the ability of a floating construction (ship) to return to its


original equilibrium position when it is disturbed from its equilibrium
position by external effects.

The stability of a ship is determined by a lot of factors, like the shape, the weight, the weight
distribution and particular for a trailing suction hopper dredger all so-called free liquid surfaces
in relation with the "wet surface". Wind, waves, movement of the cargo, movements of liquid
cargo, sharp turns, etc can cause forces or moments that can bring the ship out of equilibrium.
When a ship tilts, the position of the mass center of gravity doesn't change as long as the cargo
doesn't move. This is in contradiction with the center of buoyancy that shifts to the side to
which the ship tilts (Figure 2.117).
M0 , N

FB
G
B

a
B
FG

Figure 2-117 Recovering moment

The upward force remains, of course, the same but opposite to the weight, but their worklines
are now shifted apart over a distance a. They form a moment that tries to bring the ship back in
equilibrium. This moment is called the static stability. The work-line of the upward force cuts
the symmetry plane in a point that is called the meta-center M. For small angles of heel (<6)
this point can be considered as fixed (initial meta-center). The distance between the center of
gravity and the meta-center is also called meta-center height MG. For larger angles of heel
the meta-center is dependent on the angle of heel (false meta-center).

Page 78 of 109

Chapter 2 Trailing suction hopper dredger


From the Figure 2.117 can be directly derived that :

The arm of the static stability is equal to MG*sin .


There is only an equilibrium recovering moment when the meta-center is above the center
of gravity of the ship.

If the arm of the static stability is set out as a function of the angle of heel than a curve is
obtained that looks globally like Figure 2.118.

B
C
A

10

20

30

40

50

60

70

80

angles of heel (degrees)

Figure 2-118

Every ship has to comply with the minimum stability curve (Figure 2.119).

0,30
0,20
MG = 0,15
0,10
0
10

20

30

40

50

angle of heel (degrees)

Figure 2-119

This is determined with the following requirements:


The surface under the curve to a angle of heel of 30 has to be at least 0.055 radial.
The surface under the curve to a angle of heel of x has to be at least 0.09 radial.
The surface under the curve between the angles of heel of 30 to x has to be at least 0.03
radial.
The arm of the static stability has to be at least 0.2 m.
The initial meta-center height has to be at least 0.15 m.
In the above mentioned requirements x is equal to 40 or a smaller angle that is indicated by
openings in the hull or deckhouse that cannot be closed watertight. With the above mentioned
Page 79 of 109

wb3408B

Designing Dredging Equipment

stability curves it has been assumed that the mass center of gravity does not shift but remains in
the symmetry plane.
If a fuel or water tank is not completely filled, the fluid will try to maintain a horizontal level
independent of the tilt of the ship. This so-called free water surface is the cause, however, of a
shift of the mass center of gravity outside of the symmetry plane. As a result the arm of the
raising couple becomes smaller. It is clear that the effects of a free liquid surface in all possible
storage tanks have to be taken into account in a stability calculation.
The free liquid surface is not only important for the tanks of common ships, but particular
important for ships with a relative large free liquid surface like a trailing suction hopper
dredger.

2.5 The dredging process


As already described in paragraph 2.1.4, the dredging process of a trailing suction hopper
dredger consists of the cycle of dredging, sailing to the discharge area, discharging and sailing
back to the dredging area. Every part of this cycle contributes more or less to the production.
So the less malfunctions occur in the separate processes the higher is the cycle production. In
the following chapter these cycle parts and the connected dredging processes are discussed.

2.5.1

The loading process

The loading process can be divided in excavation, the transport and the deposit of the
material in the hopper.
2.5.1.1 The excavation
Though other working methods exist, in principal the trailing suction hopper dredger deepens a
large area entirely. The different layers of soil are removed horizontally. This in contrast to the
cutter suction dredger and surely the suction dredger, that first deepen locally and than slowly
expand horizontally. This working method has consequences for the determination of the
material to be removed. Usually the horizontal variation, for instance the grain size or the
chance of soil type, is considerably less than the vertical variation. This also implies that the
mixture of the several layers is considerably less, which gives less meaning to an average
material in the dredging area.

The trailing suction hopper dredger can in principal be deployed in nearly all soil types. Only
the efficiency is strongly dependent on the soil type and the power and means to break up the
coherency of the soil type.
When excavating with dragheads the soil type is very important. In the excavating process the
following materials can be distinguished:

Liquid soil types (silt and soft clay).


Cohesive soil types (firm clay, soft rock).
Non-cohesive soil types (sand and gravel).

2.5.1.1.1 Excavating of liquid soil types


When dredging silt or soft clay the Attenbergs limits (plasticity-index and the liquid-index) are
important. The first index determines if the soil type behaves clayey or sandy. For a plasticityindex < 7 the material behaves sandy. The second index determines if the material behaves like
Page 80 of 109

Chapter 2 Trailing suction hopper dredger


a fluid and thus easy to dredge or firm and has to be cut. A soil type behaves like a fluid when
the water content is close to the liquid limit.
For a fluid-like behavior the liquid-index must be like:
Water content

0
Firm

w wp
Ip

> 0.9
100 %
LiquidWater

Plastic
Plastic limit

Liquid limit

Plasticity index
Figure 2-120

When dredging a liquid-like soil the volumetric concentration, Cv =

mixture water
, is almost
situ water

independent of the in situ density. Also the dimensions and type of the draghead have hardly
matters. This means that the fill rate also is almost constant. For virginal fluid silt this is around
70 to 75 %. Then the ship is loaded "until overflow". The nett suction time is totally
determined by the rheological behavior of the silt.
If there is a lot of contamination, like stones, wires, old bikes, etc. in the silt or if the length of
the dredging area is small, requiring frequently turning, the fill rate will reasonably decrease.
When debris clogs the draghead, the dredge-master will dilute the mixture. Besides that regular
stops for removing the debris in the draghead as well the restarts of the process, dilutes the
mixture too. Fill rates of 40 % or less are easily reached. When the silt gets a more consistent
behavior, thus a lower liquid-index, the fill rate to the overflow decreases. But because the silt
is more consistent it will behave less like a homogeneous fluid and more like a mixture of
pieces silt/clay in a heavy transport fluid. The loading after the overflow is reached, with a lot
of overflow losses, becomes interesting again; therefore the fill rate can still be reasonable.
However the suction time will increase.
In silt, as a result of the decay processes of organic material, gas can exist in the form of
bubbles. Besides it is possible too that this gas is dissolved in the pore-water. When dredging
silt, the gas-bubbles will grow when moving upwards caused by the pressure drop in the
suction-pipe. (p*V=constant) Regarding physics this situation is almost equal to the forming of
vapor bubbles in water during a pressure drop, however than it is called cavitation. Because
cavitation decreases the performance of the dredge pump, this will also be the case with gas
bubbles. The advantage with gas bubbles is that it happens in the pipeline system before the
pump. This creates the possibility to take away a part of the gas bubbles before they implode in
the pump. For this reasons a de-gassing installation is mounted in the pipeline just before the
pump. A well-designed de-gassings installation does not or hardly decrease the performance of
the pump. Two systems are used: a de-gassings installation with accumulator (Figure 2.121) or
a de-gassings installation with a gas-extractor tank (Figure 2.122).

Page 81 of 109

wb3408B

Designing Dredging Equipment


Atm air

CV
Waterlevel at
empty ship

Valve

ejector

accumulator
High

Valve

low
water
pump

Remote controled valve

Water intake

filter

(B goes open when A is closed)


(CV = controlable valve)

Figure 2-121 Degassing installation with accumulator

Gas discharge
VW = water supply pump
BT = buffertank
GA = gas-suction mouth
VA = vacum-control valve
RA = control valve
LP = drain pump
WRP = water-ring pump
MRA = mixture return-valve
MRP = mixture return-pump

WRP

BT

RA

Water discharge
to drain
MRP
MRA

VA

LP
VW
GA

max
min

Figure 2-122 Degasssing installation with gas-extractor tank

2.5.1.1.2 Excavating in cohesive soil types


In cohesive soil types, like very soft rocks, clay and to a less extend in silt, the cutting
dominates the excavating process. In the dragheads blades, chisels or teeth are mounted (Figure
2.123). A well-shaped design is important to prevent clogging. Besides this improves the
mixture forming too.

Page 82 of 109

Chapter 2 Trailing suction hopper dredger

Figure 2-123 Modular draghead with a teeth beam

The linear cutting theories for rock cutting and undrained clay cutting apply here. In this case
the cutting forces for the applied trail-velocities are only slightly speed-dependent. Besides the
cutting forces increase linear with the depth. This means that the specific energy is almost
constant for this cutting process. The pressure difference over the draghead plays not or hardly
a role for the cutting forces. To keep the blades pushed into the soil the pressure difference over
the draghead is usually insufficient and the visor has to be fixed to the helmet. The cutting
depth is adjusted either by placing a stopper on the helmet related to the dredging depth or by
hydraulic cylinders. As described in chapter 2.2.5.1 these cutting forces has to be provided by
the propulsion.
For the calculation of the cutting forces for design purposes it is the custom to use the specific
energy concept. The specific energy Es is the energy needed to cut one m3. In formula:

Es =
Es
Ns
Ps

=
=
=

Ns
Ps
Specific energy
cutting power
cutting production

[J]
[W]
[m3/s]

For the force applies:

Ps = v d b
and for the power:

N s = v Fs

Fs =
with:
v
Fs
d
B

=
=
=
=

Page 83 of 109

Es v d B
= Es d B
v
drag velocity
cutting force
cutting depth
draghead width

[m/s]
[N]
[m]
[m]

wb3408B

Designing Dredging Equipment

The specific energy of different soil types is known within the dredging companies, but can be
calculated also the linear cutting theories.
From the available thrust of the propellers the maximal available pulling force can be
determined. For the calculation of the excavation production of the draghead, however, the
average available force must be used. This depends among other things on the variation in the
cutting depth.

Fpiek
Faverage

is usually between 1.25 and 1.5 and sometimes even 2.

The production is totally determined by the cutting process and is independent of the
pump flow rate, if it does not interfere too much with the mixture forming.
2.5.1.1.3 Excavating in non-cohesive soil types
In non-cohesive soils, like sand and gravel, the excavation process within the draghead is
physically complicated. If no jets are used to excavate the soil, the working of the draghead is
totally based upon the erosion by the flow underneath the rims of the draghead generated by the
dredgepump. The pressure difference over the draghead generated by this flow causes a
groundwater flow underneath the draghead (Figure 2.124 and 2.125).

Excavating profile and grondwater flow


underneath a draghead without jets

Groundwater flow underneath a draghead


in longitudal direction

Figure 2-124

1/2b

1/2b

For the 2-D stationary situation


this groundwater potentials can be
describe accoding to :

LM
MM
MN

F x + 1 bI
F 1 IO
GG 2 JJ arctanGG x 2 b JJ PP
GH y JK
GH y JK PPQ

H
=
arctan

Figure 2-125

The vertical groundwater flow under the draghead generated by this pressure difference causes
a decrease of the effective stress in the sand. The critical hydraulic gradient for moving the
particles follows from the equilibrium of the flow force with submerged weight of the particles.
This leads to the equation:
Page 84 of 109

Chapter 2 Trailing suction hopper dredger

LM
OP
1
1
x+ b
x b
2
2
MM
PP = 100 n 1
_
MN y + FGH x + 21 bIJK y + FGH x 21 bIJK PQ 100
L
OP
d H M
b
Hb
y
H 1
1
x=0
>1
>y + b
<

= M
1 P
4
dy M

b
b
4
y + b P
N 4 Q

d
H
=
dy

For

d H
=
dy

For y = 0

b
Hb
1
>1
> x2 b2
1 2

4
x b
4

x
<
b

H 1
+
b 4

For y=0 this condition is always fulfilled. The term (100-n)/100 is the ratio sand particles over
the total volume. For Y=0 the condition is always fulfilled because X/b is always smaller than
or equal to
Critical depth
Pressure differance H/b [-]
0

Critical depth for X=0 is shown


in the Figure next and shows
relatively very high critical
depth!

0
0.2
y/b [-]

0.4
0.6
0.8
1
1.2
1.4

Figure 2-126

However, by the erosive action of the water entraining into the draghead, the grains want to
move from each other (dilatancy) and a pore pressure drop, which increases the effective
stresses of the grains. Which process is dominant depends on a number of factors. The question
is if the ground water flow is able to keep up with the increase of pore volume of the sand. If
that is not the case than a further decrease of the water pressures arises, with a decreased
erosion process as a result.
The ratio between the mixture flow rate Qm and the erosion flow rate Qe as function of the Cvd
is:

Qmixture = Qerosion + Q pores + Qsand 1 =

With:

Qerosion Q pores
Q
+
+ sand
Qmixture Qmixture Qmixture

1=

Qerosion
n
Qsand
Q
Q
n
+ 0
+ sand 1 = erosion + 0 Cvd + Cvd
Qmixture 1 n0 Qmixture Qmixture
Qmixture 1 n0

or

Qerosion
C
= 1 vd
1 n0
Qmixture

Qmixture = The mixture or suction pump flow rate.


[m3/s]
Qerosion = the erosion flow rate, sucked from underneath the rims of the [m3/s]
draghead

Page 85 of 109

wb3408B
Qsand
Qpores
Cvd
n0

Designing Dredging Equipment


[m3/s]
[m3/s]
[-]
[-]

= The sand flow rate.


= The flow rate of the pore water present in the sand.
= transport concentration
pore ratio

This volume balance is shown in Figure 2.127. From a physical point of view, the
concentration will increase as well when the erosion or crack velocity underneath the draghead
increases (erosion line in the Figure 2.127) when Qmixture remains constant). From experience it
is known that for a certain type of draghead without jets, the concentration Cvd is only slightly
dependent on the mixture flow rate, which points out that the quotient

Qerosion
remains almost
Qmixture

constant. As a rule of thumb for the average erosion depth can be written: d =

k 0.3
.
vt0.9

In this k is the water permeability of the sand and vt the trail speed of the draghead, both in m/s.
The factor is dependent on the dimensions of the draghead.

Draghead without jet water


0.6
Volume balance
0.5

Cvd

0.4
Erosion
0.3

0.2

0.1

0
0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Qe/Qm

Figure 2-127

With increasing width of the draghead the average depth will decrease, looking to the erosion
process around the draghead. Unfortunately there is yet insufficient knowledge of this process
to determine an optimum width of the draghead. The maximum concentration Cvd for the
dragheads without jets remains limited to 15 % in loose sand. In a lot of cases however Cvd is
smaller than 10 %.
If jets are used to excavate the sand, this decreases the erosion flow rate, because the volume
balance should be fulfilled:

Qmixture = Qerosion + Q jet + Qsand + Q pores


With:

Page 86 of 109

[m3/s]

Chapter 2 Trailing suction hopper dredger


Qmixture = the mixture or suction pump flow rate.
[m3/s]
Qerosion = the erosion flow rate, sucked from underneath the rims of the [m3/s]
draghead
Qjet
= the jet pump flow rate.
[m3/s]
[m3/s]
Qsand = the sand flow rate.
[m3/s]
Qpores = the flow rate of the pore water present in the sand.
= transport concentration
[-]
Cvd
n
pore ratio
[-]
Furthermore:

Qsand
= Cvd
Qmixture
n
Q pores =
Qsand
1 n
With:

(transport concentration) and:

pore ratio [-]

From the above mentioned continuity condition now follows:

Q jet
Cvd
Q
= erosion +
1 n Q mixture Q mixture

This is a bundle of lines under 45 in a

Qerosion Q jet
C
,
diagram for constant values of vd
Qmixture Qmixture
1 n

(Figure 2.128).

1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0

0.1

0.2

0.3

0.4

0.5

Qjet / Qmixture

0.6

0.7

0.8

0.9

Cvd/(1-n)=0
Cvd/(1-n)=0.2
Cvd/(1-n)=0.4
Cvd/(1-n)=0.6
Cvd/(1-n)=0.8

Figure 2-128 Relation between capacities to fulfill the volume balance in the draghead

This picture shows that high concentration or mixture densities can be reached only for low
Q jet
Q
values of erosion and
Q mixture
Q mixture
Page 87 of 109

wb3408B

Designing Dredging Equipment

In case of a large jetpump capacity the erosion flowrate can get negative value resulting in
spillage behind the draghead.
With jets well devided over the width
of the draghead an erosion-profile can
reached with an almost constant
depth over the full width

Erosion profile for a draghead with a well designed jetsystem

Figure 2-129

As said earlier, a reasonable assumption is that the jet- production is linear with the total
momentum flux of the jet system and independent of the trail speed. The momentum I=wQu.

M sand = I = w Qu = w Q
I
Msand
pjet
Q
U

=
=
=
=
=
=

2 p jet

Momentum in N
Eroded sand mass in kg/s per jet
Jet pressure at the nozzle in Pa
Jet capacity in m3/s
Jet velocity at the nozzle in m/s
Coefficient depending on the particle size, jet pressure, jet capacity and trailspeed.
A reasonable assumption for alpha is =0.1
Water density in kg/m3.

When the nozzle are divided well over the width of the draghead the mass M should fulfill the
relation:

sand

= B d v trail

all jets

B
D
vtrail
situ
particle

=
=
=
=
=

situ water
particle
particle water

Width draghead in m.
Eroded layer thickness in m
Trailspeed in m/s
Density soil in situ kg/m3
Particle density in kg/m3

When the trailspeed is said to 1.5 m/s, which equals 3 knots and the product B.d can be
calculated.
In general the effective of the jet decreases somewhat with increasing pressure at constant
momentum. This means that low pressure- high capacity jets are more effective than high
pressure-low capacity jets. They use more specific energy too. On the other hand however,
much jetwater dilutes the mixture density (Figure 2.128). So the designer has to search for the
optimum solution between cost (power) and production
Page 88 of 109

Chapter 2 Trailing suction hopper dredger

M sand = C vd Q m sand = w Q jet


1 L
p= M
2 MN

sand
w

C vd Q m
Q jet

OP
PQ

2p
w

The results are give in fig 2-130

2
1.8

10000
9000

1.6
1.4
1.2

8000
7000
6000

1
0.8
0.6
0.4

5000
4000
3000
2000

0.2
0

1000
0
0

0.2

0.4

0.6

0.8

Cvd
p [kPa]

The required pressure can be


calculated using the same basic
formulas as mention in the forgoing
chapter.

Density [t/m2], Qj/Qm,Cvd

Jet-water is used for loosening the soil within the dragheads, as well as to assist the process
during discharging the load, either by dumping or by pumping ashore. The flow rate of the
water pump is between 20 to 30 % of the sand pump flow rate and the pressure is usually
between 5 and 15 bar.

density
Qjet/Qm
p {kPa}

Cvd/(1-n)

Figure 2-130

The breaking up of the coherence of the


soil, which is done in the draghead
either by the erosion or by jets, can also
be done by the gravity under certain
circumstances. When the sand layer has
sufficient thickness a narrow path is
deepened to full depth as quickly as
possible. Next the trailing suction
hopper dredger keeps on dredging at
the base of the embankment. By the
breaches process the embankment will
slowly move perpendicular to the trail
direction (Figure 2.131). Besides the
breach causes the sand to be looser
packed at the bottom of the
embankment. Also mixing of various
materials takes place.

Movement of slope

Figure 2-131

The disadvantage of this method is, of course, that the material has to be obtained at greater
depth. If the "horizontal" or "vertical" method is preferred depends therefore on the grain
distribution of the various layers, the suction depth and how far the pump of the trailing suction
hopper dredger is below the waterlevel.
The dredging soft rock by trailing suction hopper dredgers is only done in exceptional cases. In
fact only in those cases where the operating hours of a cutter suction dredger are so limited by
the weather conditions that it is not profitable or where the amounts to be dredged are so
limited that the mobilization of a cutter suction dredger is not profitable.
Dredging rock with a trailing suction hopper dredger is not just done. The dredger has to be
equipped for that. This means that the dragheads, the suction pipes and hull attachments able to
resist the forces that during the ripping of rock.

Page 89 of 109

wb3408B

Designing Dredging Equipment

2.5.1.2 The transport of the slurry


In the course Dredging Processes II (Wb 3414) the pumping of sand-water mixtures will be
discussed extensively, so that only specific cases will be discussed here with regard to the
transport and deposition in the hopper of the dredged material.

If the trailing suction hopper dredger is limited for its dredging depth to a dredging depth of
30 m than one fixed pump-speed is sufficient. If the ship has to dredge over a deeper range of
depth or equipped with an additional submerged pump, than the question rises whether the flow
rate variations are not too high between the suction in shallow waters and at the maximum
dredging depth. The maximum suction depth determines the highest pump speed, if the pump is
sufficiently under water. If this pump-speed is fixed than the flow rate when dredging in
shallow water will significantly larger than dredging at the maximum depth. Since overflow
losses increase linear with the flow rate it must be considered if it is economical to equip the
dredgepump with a speed control to keep the flow rate constant at different depth.
Furthermore the pump will have to be optimized for either the dredging operations or pumping
ashore, depending on the total expected time of operations under these modes.
When no submerged pump is fitted, it might better to pursuit for straight a piping system in the
suction line, even if lead to an extra elbow in the discharge line.
2.5.1.3 The loading
In order to obtain the highest possible fill rate during the loading the hopper with nonsettling
slurries, the poor mixture (mixture with a too little density) van be pumped straight overboard.
An automated valve controller can easily do this. However, with the increase of environmental
requirements this is banned nowadays.

For settling mixtures like pieces of clay, sand and gravel, a part will settle and a part will leave
the hopper through the overflow. A rule of thumb sometimes followed is that all with a d50 < 75
m flows overboard.
A measure for the quality of the settling process is the relative cumulative overflow loss. This
is defined as the ratio between the total amount of solids that leave the hopper through the
overflow and the total amount of solids pumped in the hopper. This relative cumulative
overflow loss is, except for the material properties as grain size, the grain distribution, shape
and specific mass, also dependent on the loading conditions like the flow rate, concentration,
turbulence intensity, temperature and the hopper geometry.
These overflow losses are, like mentioned above, largely dependent on the parameter

( s Q) = s
BL

term

s0

and less of

( s Q) = s
BH

v0

(see reader: Dredging Processes I (Wb3413). The

Q
is called the surface load.
BL

In these:
Q
L
B
H
s
s0
v0

=
=
=
=
=
=
=

Page 90 of 109

the total in-going mixture flow rate


the length of the hopper
the width of the hopper
the settling height in the hopper
the settling velocity
Surface load
Horizontal velocity

[m3/s]
[m]
[m]
[m]
[m/s]
[m/s]
[m/s]

Chapter 2 Trailing suction hopper dredger

The first parameter is the ratio between the time the particle needs to settle and the time that it
stays in the hopper. The second parameter is the ratio between the horizontal velocity in the
well and the settle velocity of the particle and is a measure for the degree of turbulence in the
hopper.

The overflow losses as function of the earlier mentioned terms:

( s Q) = s
BH

v0

( s Q) = s
BL

s0

and

are reasonably approximated by the theory of Camp, although the

sedimentation process in the hopper is quite different as assumed by Camp. For a real
understanding of the sedimentation process the reader is referred to the thesis of Dr.Ir. C. van
Rhee .
In Figure 2.132 the settled part (removal ratio), so Rt = (1-overflow losses), is shown as
function of these two parameters.

Page 91 of 109

wb3408B

Designing Dredging Equipment


2

1,0

S
So
0,9

2,0
1,5

0,8

1,2
1,1

0,7

1,0
0,9
0,6

0,8
0,7

0,5

0,6
0,5

0,4
0,4
0,3
0,3

0,2

0,2

0,1

0,1

0
0,001

0,01

0,1

S 1
Vo

Figure 2-132 Camps diagram

By calculating the settling process in a number of steps the relative cumulative overflow losses
can be determined as function of time or load rate. From the theory of Camp can be de derived
that the influence of the bed height is marginally. This implies that during the loading process
the overflow losses are almost constant. Although in practice loading curves are almost
straight. The overflow rate is not.
2.5.1.3.1 Loading curve
Dependent on the way of payment, in cubic meters or in Tons Dry Solids (TDS), the contractor
will like to know the development of the volume in m3 or of the TDS in the hopper during
loading. To do this it is necessary to measure the volume of the total load (sand and water).
Acoustic silo indicators usually do this. The weight of the (useful) load is measured by
determining the development of the draught as function of the time (chapter 2.2.2.1). From the
volume and the weight of the useful load the volume in m3 or the TDS can be determined if the
volume weight z of the sand and the specific weight k of the sand and the water w are
known.

Page 92 of 109

Chapter 2 Trailing suction hopper dredger


The loading curve can be divided in three phases:
1. Before the overflow is reached:

Vload ( t ) = Qi t
Gload ( t ) = Vload ( t ) i = Qi i t

i w
t
z w
w


k t = Qi i w z w k t = Qi i w k t
Gsand ( t ) = Vsand ( t ) z
k w
z w k w
k w

Vsand ( t ) = Qi

In this:
Gload and Vload, the weight and the volume of the total load, so sand and water. Vsand the
sand volume (including the pores) in the hopper and Gsand the weight of the sand (excluding
the pore water), so TDS.
Qi and Qu are the in- and out-going flow rate. i, k, z and w are the volume weights ( =
g) of the mixture, the sand grains, the sand volume with the pores and the water.
In this it is silently assumed that the hopper is totally empty before the start of the suction.
If this is not the case than volume must be increased with the value V0 and the weight with
G0.
2. When the overflow is reached tov, but the ship is not yet on its dredge mark, the hopper
volume remains constant (constant volume loading).

Qi = Qu
Gi = Qi i and Gu = Qu u with i > u > w
and therefore:

Vload ( t ) = Vhopper ( t ) = constant = Vload ( tov )


Gload ( t ) = Gov + Qi ( i u )( t tov )

( i u ) t t
( ov )
( z w )
( )
ov
Gsand ( t ) = Gsand
+ Qi i u ( t tov )
( k w )

ov
Vsand ( t ) = Vsand
+ Qi

ov
ov
are the volume of the sand and the weight of the grains at the moment the
Vsand
and Gsand

overflow is reached.
3. The overflow is reached and the ship is on the dredge mark.
In this case the weight of the total load (water and sand) remains constant (constant
tonnage loading).

Gi = GU = Qi i = Qu u and therefore Qu = Qi

i
u

Page 93 of 109

wb3408B

Designing Dredging Equipment

mark
mark
Vload ( t ) = Vload
Qu ( t tmark ) = Vload
Qi

Gload ( t ) = Gmark = constant

i
( t tmark )
u

i u
( t tmark )
z w

+ Qi i u k ( t tmark )
k w

mark
Vsand ( t ) = Vsand
+ Qi
mark
Gsand ( t ) = Gsand

mark
mark
are the volume of the sand with pores and the weight of the sand grains
Vsand
and Gsand

(TDS) on the moment the hopper reaches the valid dredge mark.
The total load curve is now known in mass and volume if Qi, i, u, k, h and w are known. u
can be determined from the overflow losses and v depends on the type of soil.

Loading curve for hopper density =1450 kg/m3


V_mixture

V_sand

Load

W_sand

16000

Volume [m3] / Load [ton]

14000
12000
10000
8000
6000
4000
2000
0
0

20

40

60

80

100

120

Loading time [min]

Figure 2-133

For pure constant volume hoppers the weight of the load is proportional to the draught of the
ship. This increases in time, though the mixture-volume in the hopper remains constant.

Page 94 of 109

Chapter 2 Trailing suction hopper dredger

Loading curves for constant volume hopper


V_mixture

V_sand

Load

W_sand

20000

Volume [m3] / Load [ton]

18000
16000
14000
12000
10000
8000
6000
4000
2000
0
0

20

40

60

80
Loading time [min]

100

120

140

160

Figure 2-134

This does not account for the pure constant tonnage hoppers. Then the draught remains
constant after reaching the overflow (Figure 2.135).

Loading curves for constant tonnage hopper


V_mixture

V_sand

Load

W_sand

12000

Volume [m3] / Load [ton]

10000
8000
6000
4000
2000
0
0

10

20

30

40
50
Loading time [min]

60

70

80

90

Figure 2-135

To calculate the weight of the load extra data is needed: the volume of the mixture and the
volume-weight (or density) of the sand in the hopper. The first quantity is measured with silo
indicators and the second by probing on several trips the volume of the sand.
Now the determination of the load during the dredging process is done as follows:

Page 95 of 109

wb3408B

Designing Dredging Equipment

Before the start of the dredging the displacement and the weight of the water in the hopper
is determined. The displacement by measuring the draught of the vessel and the watervolume by the silo indicators.

displacement empty ship =

displacement
water g
volume water in hopper

During dredging the fore and aft draught of the ship is measured continuously and so the
displacement as well as the mixture volume by means of silo indicators.
By subtracting the start values from the momentary values of the displacement and the
mixture volume, the weight of the dry load (TDS) can be determined with the following
formula.

Gload
w
Vload
TDS =
kVload
k w
Gload
= load is the volume weight of the mixture in the hopper.
Vload
Though the load nowadays usually is expressed in TDS, it does not imply that payment is also
dependent on the amount of TDS. This can be:
1. ton dry solid (TDS)
2. m3 in the hopper (means of transport)
3. m3 in the excavation
The mutual relation between these quantities is:
TDS with volume load in the hopper:

Vload =

TDS grains water

grains load water

Therefore the conversion factor of TDS to m3:

fv =

Vload
1
=
TDS grains

grains water

load water

And for m3 to TDS:

fTDS =

water
TDS
= grains load

Vload
grains water

Shown in Figure 2.136.

Page 96 of 109

Chapter 2 Trailing suction hopper dredger

Multiplica tion fa ctors for TDS to m 3


2

25

1.8
1.6

20

1.2

15

1
0.8

10

m3 to TDS

TDS to m3

1.4

0.6
0.4

0.2
0
1000

1200

1400

1600

1800

2000

0
2400

2200

Situ density [kg/m3]

Figure 2-136

An aspect that also takes place during loading is the change in the volume weight of the
dredged material, the bulking, which can be positive, so more, as well as negative, so less. The
production unit in the dredging industry is the cubic meter per time unit. Unfortunately this is
not an unambiguous unit. A m3 in excavation appears to be a "different" m3 after settlement in
the well. Because sand grains in the hopper are usually stacked looser than in situ. The volume
weight in the hopper is lower than the situ volume weight. Also, as a result of overflow losses,
more fine sand particles will flow overboard than coarse particles. If these particles are located
in a matrix of coarser particles than the volume weight will decrease even if the stacking of the
matrix remains the same. If this phenomena happens in the dredged material can be simply
shown by comparing the sand curve with the Fller-distribution (Figure 2.137).

FLLER'S METHOD

FLLER'S METHOD
100

% by weight passing

overmaat fijn

overmaat grof

Fller

100

90

90

80

80

70

70

60

60

50

50

40

40

30

30

20

20

10

10

% by weight passing

overmaat fijn

overmaat grof

Fller

0
0

.1

.2

.3

.5
.6
.4
SQRT (d/dmax)

.7

.8

.9

0.01

0.1

10

1
SQRT (d/dmax)

Figure 2-137

In a Fller-distribution the cumulative grain distribution, given as function of

d
d max

, is a pure

straight line. Such a distribution appears to give a maximum volume weight, which implies that
the pores are constantly filled with the smaller particles. If the gradient of the smaller particles
is above the Fller-distribution than there is a surplus of fine material and the above mentioned
phenomenon would not show. If the gradient of the fine material is below the Fller-

Page 97 of 109

wb3408B

Designing Dredging Equipment

distribution than the fine material is embedded in the coarser material and the phenomenon
shows.
The volume weight in the hopper is usually lower than in situ. Dependent on the grain
distribution, a situ m3 takes the same or more space in the hopper, caused by the increase of the
ratio, which are filled with water. So the water takes this larger volume.
Example:

Assume the in situ density of the sand 1 and the density in the hopper 2. The specific weight
of the sand is k and of the water w. The cumulative overflow losses are ov and according the
Fller distribution there is a surplus of fine material. If the situ volume is V1, then the volume
in the hopper with in-situ density (1-ov) V1. The weight of solids of this volume must be equal
to the solid weight of the volume V2.
Weight of the volume V1 for 1:

G1 = (1 ov ) V1

1 w
with = g
k w k

Weight of the volume V2:

G2 = V2

2 w
k
k w

Since G1 = G2 :

V2

w
= (1 ov ) 1 w = (1 ov ) 1
V1
2 w
2 w
Example:

1
2
water
ov

=
=
=
=

2000 kg/m3
1900 kg/m3
1020 kg/m3
10 %

V1
2000 1020
= (1 ov )
= 0.9 1.11 = 1.0
V2
1900 1020

So the volume in the hopper occupies the same space as the in the excavation. It has been
silently assumed that the overflow losses do not flow back into the winning area. If that is the
case than the term (1-ov) is discarded and the delivery becomes 11 %.
If the fine sand particles are situated in a matrix of coarser particles than, for a similar stack of
the coarser particles, G2 = 0.9 G1 with V1 = V2. This leads to:

(1 ov ) V1

1 w
w
k = V2 2
k (1 ov ) ( 1 w ) = 2 w
k w
k w

This gives in the example:

2 = 0.9 1 + 0.1 w = 1800 + 102 = 1902 kg/m 3


Page 98 of 109

Chapter 2 Trailing suction hopper dredger


If all overflow losses remain in the winning area than this still holds but as a result the original
layer will be covered with 10% fine material at the end of the work.
When sucking very loose sand the bulking can be smaller than 1. The bulking is than called
negative. When dredging firm clay the bulking in the hopper is substantial, as is proven in the
following example:
Assume the situ density of the clay as 2000 kg/m3. After cutting the pore percentage of the clay
fragments is 40 %. The volume weight is than 2 = 0.6*2000 + 0.4*1020 = 1608 kg/m3. And
the bulking than will be:
V1 2000 1020
=
= 1.67
V2 1608 1020

This can be seen directly as the new volume is only 60 % of the original.
During pumping ashore to a reclamation area, usually a negative bulking takes place, since the
volume weight of the dump material is often higher than the volume weight of the material in
the hopper and losses can occur at the reclamation.

2.5.2

Sailing from and to the discharging area

It will be clear that the sailing speed determined during the sea trials, for an empty as well as
for a fully loaded ship, cannot be used as the average speed during the lifespan of the trailing
suction hopper dredger. Between the dry dock periods the hull of the ship becomes overgrown
with barnacles and seaweed and the propulsion engines and propellers are subjected to wear.
This leads to a 5 to 10 percent lower average or operational speed in deep water than the sea
trial speed. In general the trailing suction hopper dredger sails in seaways with a depth which
gives the ship extra resistance. The trailing suction hopper dredger "feels" the bottom. The
influence of the less deep seaway on the operational velocity is calculated with Lackenby's
formula (Figure 2.138).

v shallow = v deep

R| L
F A
|S1 MM01242
.
|| MM GH bd + Dg
T N

I
0.05J + 1
K

1
c
1
c+
c
c

OPU|
PP|V
PQ||W

with:

c=e

g FGH v 4 IJK

g d+D

deep

in this:
d
D
A

=
=
=

keel clearance
draught of the ship
wet cross-section of mid ship

[m]
[m]
[m2]

Page 99 of 109

wb3408B

Designing Dredging Equipment

Sailingspeed according Lackerby


Loaded

Empty

15
14.5

Sailing speed [knots]

14
13.5
13
12.5
12
11.5
11
10.5
10
0

10

15

20
25
keel clearance [m]

30

35

40

Figure 2-138

The sail time can now be determines with:


N

Tvh =
n +1
N

Tvt =
n +1

sn

with draugth full

sn

with draugth empty

( vvol )n

(v )

leeg n

Another facet that has to be accounted for, are the sail-limitations in certain areas like harbors
and narrow fairways. Furthermore the fairway has always to be checked for sufficient depth. In
case of doubt it might even be wise to carry out a hydrographic survey

2.5.3

The discharge

As described in the chapter Technical Construction the trailing suction hopper dredger may be
able to discharge its load in two ways, either by direct dumping or by means of the selfemptying installation by rainbowing or pumping to the shore.
If the load can be dumped directly it has to be known if the depth of the dump area is always
sufficient to sail with opened doors or valves, even with extremely low water. The increasing
lack of dump areas it happens regularly that the depth of the dump is limited. In such a case it is
advised to make a dump plan to use the dump as efficient as possible.
For land reclamation works for which the first layer of the sand body can be dumped directly, a
dump plan has to be made too, in order to dump directly as much material as possible, so that
less material needs to be pumped ashore.
The discharge of the load through the bottom doors or valves usually costs little time. For free
flowing soils this is done within several minutes. The discharge time increases when the
material becomes finer and more cohesive. For plastic clays this can increase to half an hour.
For such a material it has to be checked that no load, the rest load, remains in the hopper.
There is a possibility that this rest load increases with the number of trips. It appears that the
longer the clay remains in the hopper the more difficult it is to flush it out.
Page 100 of 109

Chapter 2 Trailing suction hopper dredger

Discharge through the hopper self-emptying installation is done to:


pump the load, through pressure piping to the shore.
to heighten, for example, submerged dumps that are too shallow to dump; the so-called
rainbowing (Figure 2.139).
to accurate fill submerged dumps or to cover pipelines with the use of pipe dumping.
After the pumps are started and the water comes out of the pipe the discharge of the load is
started on the side of the hopper that is the furthest away from the pump. This assures that the
pump is always as deep under water as possible. Because the material in the hopper is in
general pretty loose packed, the process looks a lot like the process of a stationary suction
dredger. The sand breaches to the opening of the suction pipe.

Figure 2-139 Rainbowing

If the hopper is not equipped with an installation that improves the breaching by means of
water-jets, than, as a rule of thumb, the discharge time is equal to the suction time. If the
hopper is equipped with water-jets to fluidize or loosen the load, than the discharge time can be
shortened considerably.
The discharge process through the hopper self-emptying installation behaves clearly like an Scurve. The discharge process is started usually slowly, because a quick start often leads to a
blocked suction pipe. After that there is for 75 to 80 % of the time an almost constant high
production. At the end of the unloading process the decreases almost linear zero (Figure 2.140).

Page 101 of 109

wb3408B

Designing Dredging Equipment

Unloading proces with time


Pr_time

Pr_ave

100
90
Prod/max. production

80
70
60
50
40
30
20
10
0
0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

t/t_unloading

Figure 2-140 Production of the unloading process

In almost all self-emptying installation a rest load remains of around 5 %. By the fluidization
process the rest loads of rocks and dirt accumulate, so that regularly the rest-load needs to be
dumped on a dump.

2.5.4

The cycle production

The cycle consisting of: loading, sailing to, discharging, sailing back can be optimized simply.
The cycle production is defined as:

Pcycle =

L (t )
tsuction + tsail + tdischarge

If tsuction and tdischarge are considered constant than this production is optimal when the following
is condition is met:

dPcycle
dtsuction

=0

This is the tangent to loading curve L(t) that also crosses the negative y-axis in the point tsailing +
tdischarge (Figure 2.141).

Page 102 of 109

Chapter 2 Trailing suction hopper dredger

Load [m3]
max. cycle production

Loading time [minutes]

no loading time

Figure 2-141 Optimal cycle production

This loading process can be made visible on board of the dredgers to determine the optimal
load. However it should be noticed that the overflow losses increase sufficient at the end of the
loading process to determine the optimal point.

2.5.5

The instrumentation

To support the dredge master instruments are available. Modern trailing suction hopper
dredgers are equipped with suction pipe position indicators both in the longitudinal as in the
transverse direction. Not only the position in relation to the bottom is indicated but also the
position of the suction pipe and the draghead in relation to the ship and sometimes even the
soil. Furthermore the dredge master has a direct view on the swell-compensators to judge if the
dragheads are on the bottom. If this is not the case than indicators are necessary. For the suction
process there are besides the vacuum and pressure indicators, also velocity and concentration
indicators. With the aid of these instruments the suction chief will optimize the suction process
by trial and error.

Figure 2-142 Instrumentation panels

Page 103 of 109

wb3408B

Designing Dredging Equipment

2.6 Special designs of trailing suction hopper dredgers


2.6.1

The gravel suction dredger

Trailing suction hopper


dredgers that collect
aggregates for the
concrete industry and
road construction differ
in several aspects from
the "standard" trailing
suction hopper dredger.
These
differences
usually arise from
economical
considerations. Items
that are of less use are
left out, while others
are added.
Figure 2-143 Gravel dredger Charlemagne

These include:

The maneuverability. A lot of gravel suction dredgers are built to collect aggregates at
sea. These are relative wide concessions where accurate dredging is of no or small concern.
Furthermore there are long transportation distances. Therefore the requirements for the
maneuverability are less strict than for the trailing suction hopper dredger that has to
dredge frequently in busy fairways or ports.For this reason the gravel suction hopper
dredger is equipped with only one screw.

The longer dredge cycle. The longer sail distances mean that the suction time is only a
small percentage of the total cycle time. Therefore it is much more economical to equip the
ship with only one suction pipe and one dredge pump.

Since the quality of the material determines the price, these ships are equipped with a
creening installation. The "bad" material can than be put overboard. Of course it is also
possible to load all the material (called all-in or tout-venant).

A discharge installation with which it is possible to unload "dry" in every arbitrary port.
Seldom a gravel suction hopper dredger has bottom doors or valves.

Page 104 of 109

Chapter 2 Trailing suction hopper dredger

Figure 2-144 Screening installation

Since the concessions are increasingly further away from the land and therefore in deeper
waters, submerged pumps on the suction pipe are also used on modern gravel suction hopper
dredgers.
The discharge systems are of the drag system, clamshell or excavation wheel (Figure 2.114)
principle that delivers the material from the hopper to a silo from which the material is
distributed further via a conveyor belt. The way of operation does not differ much from the
"classical" trailing suction hopper dredger. Instead of pumping the material straight into the
hopper, it is now pumped into the screening installation, where it is separated into the required
class(es). When sailing to the discharge area the drain installation is turned on to bring the load
as dry as possible ashore.

Trailing suction hopper dredger for


inland waters provides also sand and
gravel to the concrete industry as well
as sand for reclamation purposes. They
do also maintenance dredging in river
harbours Their design is much simpler
than ordinary trailer suction hopper
dredgers (Figure 2.145).

Figure 2-145 Trailing suction hopper dredger for


inland waters

Page 105 of 109

wb3408B

2.6.2

Designing Dredging Equipment

The stationary suction hopper dredger

The stationary suction hopper


dredger is the predecessor of the
trailing suction hopper dredger.
In the most well known design
the stationary suction hopper
dredger has a hopper and behind
it the pump room with one
dredge pump. The suction pipe is
directed however forward.
Stationary suction hopper
dredgers are single-screw ships.
The propulsion engine directly
drives the dredge pump.
Figure 2-146 Stationary suction hopper dredger

The method of operation differs significantly from the trailing suction hopper dredger and is in
principle equal to the suction dredger.
When dredging the vessel anchored in its borrow area. The amount of anchors needed depends
strongly on the operational circumstances, like current and wind velocity, current and wind
direction and shipping. If the circumstances are well than one or two front anchors are
sufficient. If the dredging takes place in a tidal area where the current change direction
depending on the tide, than also one or two aft anchors are placed. A second anchor is needed if
the ship must be hauled frequently.
As with suction dredgers the stationary hopper dredger is used in free running sand. Dependent
on the breach height the ship is slowly hauled in the direction of the suction direction. The
loading of the hopper is similar to the process of the trailing suction hopper dredger.

Page 106 of 109

Chapter 2 Trailing suction hopper dredger

Figure 2-147 Trailing suction hopper dredger for stationary dredging

Sometimes the trailing suction hopper dredger is used as a stationary dredgers" for certain
works. To do this the dragheads are removed and if not already present an aft anchor is
mounted. When arriving at the winning area first the aft anchor is placed. Dependent on the
weather conditions the front anchor is also placed. Since the pipes put backwards the trailing
suction hopper dredger works itself while dredging backwards. There are also trailing suction
hopper dredgers that have the possibility to bring their suction pipe forward and are than able to
work on the bow anchor (Figure 2.147). With well-breaching sand trailing suction hopper
dredgers can also suck profiles with the drag suction method. The embankment must than be at
all times more gentle than the suction pipes of the trailing suction hopper dredger. The trailing
suction hopper dredger forces its way into the embankment with a velocity of 0.25 to 0.5 knots.
The main advantage of this method is that no anchors are needed which gives more freedom of
movement and a quicker leave in case of an emergency.

TSHD working as PS Dredger


TSHD dredging to the Face

Figure 2-148 Trailer suction hopper dredger working in a plain suction mode

2.6.3

Boom dredgers

The boom dredger (Figure 2.149) is a special design of the trailing suction hopper dredger.

Page 107 of 109

wb3408B

Designing Dredging Equipment

Figure 2-149 Boom dredger

It is equipped with a 50 to 60 meter long construction, the boom, that makes it possible to
pump the dredged material immediately sideways back (side casting). This method of dredging
is used in silt rich fairways, where it is cheaper to spray the material to the side, a hundred
meters from the bank of the fairway instead of bringing it to a dump far away. Approach
channels at the lake of Maricaibo in Venezuela are dredged in this manner

Page 108 of 109

Chapter 2 Trailing suction hopper dredger

2.7 Literature
1. Trailing Suction Hopper Dredging Handbook.
Issued by The Training's Institute for Dredging.
2. Coastal and Deep Ocean Dredging, John B. Herbich, Gulf Publishing Company, Houston,
Texas, USA, 1975.
3. Dredging and Dredging Equipment, R.J. de Heer and Rochmanhadi, part 1 and 2, IHE,
Delft, 1989.
4. Baggertechniek, collegedictaat f14, G.L.M. van der Schrieck, TU Delft, Civiele Techniek,
1996 (in Dutch).
5. Constant Tonnage Loading System of Trailing Suction Hopper Dredgers, J. de Koning,
Proceedings International Course Modern Dredging, 1977.
6. Nassbaggertechnik, A. Welte, Institut fr Machinenwessen in Baubetrieb, Universitt
Fridericiana, Karlsruhe, 1993.
7. Proceedings of the dredging days, Europort 1980, CEDA, 1980.
8. Technical aspects of large Trailing Suction Hopper Dredgers, P.J. Koert, IHC Holland.
9. Further development of loading and unloading processes for Trailing Suction Hopper
Dredgers, S. Steinkhler, 14 World Dredging Congress, Amsterdam, 1995.
10. Several articles from Port & Dredging of IHC Holland.

P&D

Split trailer suction hopper dredgers


106 + 107 + 110
128
VOLVOX SCALDIA, Trailing Dredgers with built-in booster unit
CORONAUT, the sixth IHC Eurotrail
130
AGRONAUT, the seventh IHC Eurotrail
134
New Trailing Suction Hopper Dredger for Dredging International
134
3
Trailer VOLVOX IBERIA, 5700 m
140
142
TSHD J.F.J. DE NUL, Versatile Leviathan
Trailing Dredger, HAM 311
143
Trailing Dredger, CRISTOFORO COLOMBO
143
PEARL RIVER, Trailing Dredger of 17000 m3
144
TSHD Ham 311 and Ham 312
148
151
TSHD Queen of Penta Ocean
TSHD Ham 317
153
TSHD Rotterdam
155+156
TSHD Ham 318
157
Gravel Dredger Cambeck and Charlemange
133 + 157
Dragheads
124 + 137+157

Page 109 of 109

Chapter 3: Cutter Suction Dredger

3.

The Cutter Suction Dredger .........................................................................2


3.1.
General description........................................................................................3
3.1.1. Areas of application......................................................................................4
3.1.2. History ..........................................................................................................5
3.1.3. Working method...........................................................................................6
3.2.
The design.......................................................................................................8
3.2.1. The production capacity ...............................................................................9
3.2.2. The dredging depth.......................................................................................9
3.2.2.1. The maximum dredging depth..................................................................9
3.2.2.2. The minimum dredging depth ..................................................................10
3.2.3. The width of the cut......................................................................................12
3.2.4. The type of soil.............................................................................................14
3.2.5. The transport distance...................................................................................14
3.2.6. Access to the dredging site ...........................................................................15
3.3.
The dredging equipment ...............................................................................15
3.3.1. The cutter head .............................................................................................16
3.3.1.1. The dimensions of the cutter head............................................................16
3.3.1.2. The cutting power.....................................................................................16
3.3.1.3. The cutter speed........................................................................................17
3.3.2. The reaction forces on the cutter ..................................................................18
3.3.2.1. The horizontal and vertical cutting force..................................................18
3.3.2.2. The axial force..........................................................................................20
3.3.2.3. The ladder weight .....................................................................................21
3.3.3. The side-winch power and speed..................................................................21
3.3.4. The ladder winch speed and power ..............................................................24
3.3.5. The dredge pumps ........................................................................................24
3.3.6. The jet pump.................................................................................................25
3.4.
The drives .......................................................................................................25
3.4.1. The cutter head drive ....................................................................................25
3.4.2. The side winch drives ...................................................................................27
3.4.3. The ladder drive............................................................................................27
3.4.4. The sand pump drives...................................................................................27
3.5.
Spudsytems .....................................................................................................28
3.5.1. The spud carriage system .............................................................................28
3.5.2. The fixed spud system ..................................................................................30
3.5.3. The spud door system...................................................................................32
3.5.4. The walking spud system .............................................................................32
3.5.5. The rotor spud system ..................................................................................33
3.5.6. The Christmas tree........................................................................................34
3.6.
The general layout .........................................................................................35
3.7.
Technical construction
Fout! Bladwijzer niet gedefinieerd.
3.7.1. The Hull........................................................................................................40
3.7.2. The cutter head ladder ..................................................................................41
3.7.3. The cutter head .............................................................................................43
3.7.4. Tooth and cutting edge systems ...................................................................46
3.7.5. The side wires...............................................................................................50
3.7.6. The anchor booms ........................................................................................51
3.7.7. The spuds......................................................................................................52
3.7.8. The spud lifting system ................................................................................52
3.7.9. Pumps and pipelines .....................................................................................53
3.7.9.1. The suction pipeline .................................................................................53
3.7.9.2. The pumps ................................................................................................54

Page 1 of 79

Wb3408b

Designing Dredging Equipment

3.7.10.
The winches..............................................................................................55
3.7.10.1.
The ladder winch ..................................................................................55
3.7.10.2.
The side winces ....................................................................................55
3.7.10.3.
Other winces.........................................................................................56
3.7.11.
Hoisting equipment ..................................................................................56
3.7.12.
Auxiliary equipment.................................................................................56
3.8.
The dredging process.....................................................................................57
3.8.1. The spillage ..................................................................................................57
3.8.2. The production in breach-forming soils .......................................................59
3.8.3. The production by non-breach forming soils ...............................................61
3.8.4. Specific energy .............................................................................................63
3.8.5. The cutting production .................................................................................65
3.8.6. The spillage ..................................................................................................67
3.9.
Enclosures.......................................................................................................68
3.9.1. The relation between swing speed and side winch speed.............................68
3.9.2. The side winch force and power...................................................................69
3.9.3. The shape and cutting geometry of cutter heads ..........................................70
3.9.4. Cutting by teeth or chisels ............................................................................74
3.9.5. Conditions for cutting clearance...................................................................75
3.9.5.1. The effect of warping on the clearance angles .........................................77
3.10.
References.......................................................................................................79

3. The Cutter Suction Dredger

Figure 3. 1

Page 2 of 79

The Mashhour, at present the biggest cutter suction dredger in the world,

Chapter 3: Cutter Suction Dredger


3.1. General Considerations

Discharge pipe
Dredge pump
Suction pipe

Cutter ladder

Cutter head
Dredge pump
Auxiliary spud

Ladder winch
Port side winch.

Spud carriage
Working spud
Starboard winch

Figure 3. 2 Lay-out

snijkopzuiger.

The cutter suction dredger is a stationary dredger equipped with a cutter device (cutter head)
which excavate the soil before it is sucked up by the flow of the dredge pump(s).
During operation the dredger moves around
a spud pole by pulling and slacking on the
two fore sideline wires. This type of
dredger is capable to dredge all kind of
material and is accurate due to their
movement around the spud pole. The
stationary cutter suction dredger is to
distinguished easily from the plain suction
dredger by its spud poles, which the last
dont have.
The spoil is mostly hydraulically
transported via pipeline, but some dredgers
do have barge-loading facilities as well.
Cutter power ranges from 50 kW up to
5000 kW, depending on the type of soil to
be cut.

Cut width

Workspud
in carriage

Spud carriage
length
Auxilary spud

Vertical swing pattern

Figure 3. 3 Swing pattern


The ladder, the construction upon which the cutter head, cutter drive and the suction pipe are
mounted, is suspended by the pontoon and the ladder gantry wire.
Seagoing cutter suction dredgers have their own propulsion that is used only during
mobilization. The propulsion is situated either on the cutter head side or on the spud poles
side.

Page 3 of 79

Wb3408b

Designing Dredging Equipment

Figure 3. 4 Seagoing CSD Aquarius sailing in the Beaufort Sea

3.1.1. Areas of application


Cutter suction dredgers are largely used in the dredging of harbours and fairways as well as
for land reclamation projects. In such cases the distance between the dredging and disposal
areas is usually smaller than the distances covered by trailing suction hopper dredgers. The
cutter suction dredger also has the advantage when an accurate profile has to be dredged.
The cutter suction dredger can tackle almost all types of soil, although of course this depends
on the installed cutting power. Cutter suction dredgers are built in a wide range of types and
sizes, the cutting head power ranges between 20 kW for the smallest to around 4,000 kW for
the largest. The dredging depth is usually limited; the biggest suction dredger can reach
depths between 25 and 30 m. The minimum dredging depth is usually determined by the
draught of the pontoon.
In the late seventies and early
eighties of the previous
century two offshore cutter
suction dredgers have been
build for applications
offshore. The All Wassl
(Figure3.5) build by
Mishubitsi, Japan for Gulf
Cobla Ltd. Has dredged the
approach channel to the
harbour Jebel Ali in Dubai,
Unit Arab Emirates.
Figure 3. 5 All Wassl Bay

Page 4 of 79

Chapter 3: Cutter Suction Dredger


After 2 years working the
dredger is sold and scrapped.
The Simon Stevin (Figure
3.6) build for Volker Stevin
Dredging has even never
worked.
Boths dredgers appeared too
specialised to be economical.

Figure 3. 6 Simon Stevin


As said the cutter suction dredger is a stationary dredger with at least two side anchors that
are necessary for the dredging process. Because of these anchors they may obstruct shipping
movements. Self propelled cutter suction dredgers uses their propulsion system no only
during mobilisation but also during shifting from one place to the other or when the dredging
area has to be left, breaking up when bad weather is expected.
The small to medium sized cutter suction dredgers can be supplied in a demountable form.
This makes them suitable for transport by road to inland sites that are not accessible by water,
for example to lay a sand foundation for a road or to dredge sand and gravel for the building
industry.
When working under offshore condition with waves or swell cutter suction dredgers clearly
have more limitations than trailing suction hopper dredgers even if equipped with swell
compensators
3.1.2. History
The cradle of the cutter suction dredgers stood in the United States. In 1884 a cutter suction
dredger was used in the port of Oakland, California. This dredger had a cylindrical cutter head
and was used to dredge layers of sandstone. It had a pipeline of 500 mm diameter and a pump
with an impeller of 1.8 m! The disadvantage of this design was that the suction mouth was
frequently blocked. At the end of the 19th century and beginning of the 20th century there was
a major development in suction dredgers

Page 5 of 79

Wb3408b

Designing Dredging Equipment

Figure 3. 7 Layout

of the cutter suction dredger RAM

For example, in the fall of 1893 the cutter suction dredger RAM was built by the Bucyrus
Steam Shovel and dredged company for use on the lower Mississippi river. This dredger was
already equipped with an rotating cutter head. (Figure 3.7).
The cutter suction dredger became the workhorse of the dredging industry in America, as did
the bucket dredger in Europe at that time.

3.1.3. Working method


After the ladder of the cutter suction dredger has been lowered under water, the dredge
pump(s) started and the cutter head set in motion. The ladder is then moved down until it
touches the bottom, or until it reaches the maximum depth. The movement of the dredger
round the spud pole is initiated by slacking the starboard anchor cable and pulling in the port
side anchor cable or reverse. These anchor cables are connected via sheaves close to the cutter
head to winches (dredging side winches) on deck. The pulling winch is called the hauling
winch. The paying out winch ensures the correct tension in both cables, this being particularly
important when dredging in hard rock.

Ds

Ds
Under cutting mode

Page 6 of 79

Over cutting mode

Chapter 3: Cutter Suction Dredger


Figure 3. 8 Different cutting modes

In addition to the type of soil, the required side winch force also depends on:
Whether the rotation of the cutter head is in the same direction as or the opposite direction
to that of the swing movement. In the first case the reaction force of the cutter head on the
soil will pull the dredger with it, as a result of which the side winch forces are smaller
than when rotation is in the opposite direction
It is also necessary to ensure the correct pre-tensioning of the cables when the cutter head
rotates in the same direction as swing. If the cutter head forces propel the cutter head
more quickly than the hauling winch does there is a very real danger that the cable of the
hauling winch will be picked up and cut through by the cutter head.
The position of the anchors has a big influence on the force needed to swing the dredger.
The closer the path of the cutter head is to the direction of the side cable, the smaller the
required force.
Naturally the side winch force is also affected by external influences such as wind,
current and waves.

Figure 3. 9 Steps and cuts

Of course, the thickness of the layer that can be removed by one swing (cut thickness
Figure3.9) depends on both the diameter of the cutter head and the type of soil. When the
required dredging depth has not been reached at the end of a swing, the ladder is set more
deeply and the ship will move in the opposite direction.
As previously mentioned, the cutter suction dredger describes an arc round a fixed point, the
spud pole or working pole. In many cutter suction dredgers this pole is mounted on a movable
carriage, the spud carriage. A second pole, the auxiliary spud, is set out of the centreline,
usually on the starboard side of the stern of the pontoon.
The spud carriage can be moved over a distance of 4 6 m by means of a hydraulic cylinder.
Because the spud is standing on the bottom, pressing the spud carriage towards the stern can
move the cutter suction dredger forward. The size of the cutter head and the hardness of the
soil determine the size of this step. During each step one or more layers of the face are cut
away by lowering the ladder one cutting thickness at the end of the swing.

Page 7 of 79

Wb3408b
With each step the
cutter head
describes the arc
of a concentric
circle round the
spud, the radius of
which increases
with the step
length.
(Figure 3.10)
a) = step length
b) = length of
carriage
If the spud
carriage cylinder
has reached the
end of its path the
spuds must be
moved. Before
stepping, the
cutter moves to
the centre line of
the cut.

Designing Dredging Equipment

D
D
a

a = steplength
b = length of carriage

a b

Vertical swing pattern

Figure 3. 10Vertical swing pattern

The auxiliary spud is then placed on the bottom, the working spud is lifted and the spud
carriage is moved forward. After this the work spud is again lowered and the auxiliary spud is
lifted. The dredger can then resume working. The first cut made after stepping is not an arc of
a concentric circle!

3.2. The design


When designing cutter suction dredgers, the following basic design criteria are important:
Production capacity
Dredging depth
Working conditions which affect the size of the dredger
Type of soil
Transport distance(s)
Access to the side
As mentioned earlier, the cutter suction dredger can be used in all types of soil, from soft clay
to hard rock. The soil to be dredged has a great influence on the design and construction.
Considerable forces are generated when working in rock. They are generated by the cutter
head and returned to the ground partly via the ladder and side winches and partly via the
pontoon and the spud pole. The design of cutter suction dredgers is also determined by the
required amount of installed cutting power.

Page 8 of 79

Chapter 3: Cutter Suction Dredger


3.2.1. The production capacity
As in the case of other types of dredger, the production capacity is determined by the market
demand with regard to the projects for which the dredger can be used. Because many cutter
suction dredgers must dredge various types of soil during their lifetime, design parameters are
set with regard to the types of soil the dredger must be able to dredge. A dredger designed to
dredge rock will also be able to dredge sand, but a sand cutter suction dredger will not be
able to dredge rock. On the other hand a sand cutter suction dredger will be able to dredge
sand more cheaply than a rock cutter suction dredger. In other words the design production
capacity of a cutter suction dredger is related to the hardness of the material that it must be
able to dredge. For example, 100 m3/hr in a rock of 10 MPa.
It is important that the production capacity is defined m3 per week, hour or second. The
smaller the unit of time chosen, the greater the production capacity. (As a result of averaging
the long term production capacity is less.)
When the requirement with regard to the production capacity in the design-soil is known, this
can be translated into a production to be cut by the cutter head. This so called cutter
production is considerably higher than the dredged production because not all the material
that has been cut enters the suction mouth. Often 20 30 % remains behind as spillage. This
must be taken into account when determining the production to be cut.
The maximum cutter production is also higher for reasons such those described above as a
result of the unit of time. With a cutter suction dredger this appears primarily in the mode of
work employed. Production is usually highest in the middle of a cut. In the corners of the cut
where manoeuvres are often carried out with the ladder or spud carriage, the production is
low or zero. This results in the fact that the cutter production when expressed in m3/s is 20
30% higher than the cutter production in m3/hr.
In order to maintain a high degree of usability cutter suction dredgers designed for rock
dredging should be equally as good in other types of soil. This implies that although the
cutting equipment is designed for rock dredging with regard to the other parts of the dredging
equipment, the other types of soil must not be forgotten.
3.2.2. The dredging depth
When designing cutter suction dredgers both the maximum and the minimum dredging depths
must be taken into consideration, since these both influence the usability of the dredger. Often
the need for a greater dredging depth leads to a pontoon with deeper draught and thus to a
reduction in the minimum dredging depth. So on one hand the usability of the dredger
increases with increasing dredging depth, while on the other hand it decreases as a result of
the related smaller minimum dredging depth. Here too the market demand plays a role in the
best choice.

The maximum dredging depth


The maximum dredging depth is an important design parameter. Because in a cutter suction
dredger the pontoon and the spud pole transfer part of the interplay of forces to the soil, the
magnitudes of the moments that occur are proportional to the dredging depth. Thus with
increasing dredging depth, not only is the dredger larger and broader (for stability), it must
also have a heavier construction. Moreover the dredging depth has a great influence on the
design of the ladder construction and thus on the pontoon. After all it must be possible to raise
the ladder above water for inspection.

Page 9 of 79

Wb3408b

Designing Dredging Equipment

Figure 3. 11 Different cutter dredgers with ladder above waterlevel

From the point of view of production, the suction depth determines whether an underwater
pump is needed to obtain the required production capacity. It is obvious that mounting an
underwater pump will increase the weight of the ladder.
If no underwater pump is considered, the diameter of the suction pipe and the head of the
pump must be increased and the concentration of the mixture reduced in order to avoid
creating a vacuum. This may lead to the pumping of low concentrations and thus much water,
which is uneconomic.
With the aid of the vacuum formula (see also lecture notes Dredging processes), from a
given limiting vacuum and the maximum concentration to be dredged it is possible to
determine whether or not an underwater pump is necessary, and if so how far under water it
must be placed. Whether or not an underwater pump is fitted is, of course, also a question of
economics, since cost of the fitting of an underwater pump is considerable.

8000

Light weight [t]

7000
2

6000

y = 9.0577x - 101.29x
2

R = 0.757

5000
4000
3000
2000
1000
0
0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

Maximum dredging depth [m]

Figure 3. 12

The minimum dredging depth


The minimum dredging depth makes demands with regard to the draught of the pontoon, the
position of the cooling water inlet and the shape and construction of the cutter ladder. It will
be clear that even when dredging at minimum depths the pontoon must have sufficient bottom
clearance. For heavy duty cutter suction dredgers this leads to deep draughts or wide vessels
(Figure3.13). The minimum dredging depth must be at least 1 m deeper than the maximum
draught of the vessel. The design of the cooling water inlet must be adapted to prevent the
intake of material from the bottom

Page 10 of 79

Chapter 3: Cutter Suction Dredger

Maximum draught [m]

0
0

500

1000

1500

2000

2500

3000

3500

4000

Cutterpower [kW]

Figure 3. 13

When dredging at depths, which are shallow in comparison to the draught of the vessel, the
shape of the ladder must also be adapted to avoid dragging of the ladder. To prevent dragging
the angle between the underside of the ladder and the horizontal must be at least 50 (Figure
3.14).

Removable wedge

Figure 3. 14

In order to obtain a better rate of filling


when dredging free running material is
desirable that the axis of the cutter head
shaft should make a steeper angle with
the horizontal than the ladder. The filling
of the cutter is determined by the sum of
the angles of the slope gradient and the
ladder (+ ) (Figure 3.15).

Figure 3. 15
Page 11 of 79

Wb3408b

Designing Dredging Equipment

3.2.3. The width of the cut


The usefulness of a cutter suction dredger is also determined by the minimum width of cut
that the equipment can dredge, and to a lesser degree on the maximum width of the cut.
Minimum width of cut is taken to mean the width that the dredger needs to dredge a channel
for itself in an area where the surface of the ground is higher than the water level; a problem
that occurs is during dredging the onshore end of pipeline trenches.

Figure 3. 16 Minimum cut width

The minimum width of the cut is determined by the line that meets the contour surface of the
cutter head at the front of the pontoon (Figure 3.16) or at the outer side of the side winch
sheaves. To reduce the minimum cutting width each side of the front of the pontoon is often
chamfered as shown in Figure 3.17 and 3.19. Figure 3.18 also shows that the further the cutter
head projects in front of the pontoon, the smaller is the minimum cutting width. Such a
solution is particularly common in American and Japanese dredgers.

Ballast tank
Lubricating
olie
Ballast tank

Fuel

ballast
tank
ballast
tank

Spare parts

Spare parts

Dry
tank

ballast
tank

Drinking
water
Engine room
Ballast tank
Ballast tank

Fuel

ballast
tank
ballast
tank

Spare parts

Figure 3. 17 Chamferred pontoon

Page 12 of 79

Spare parts

Dry
tank

ballast
tank

Chapter 3: Cutter Suction Dredger

Figure 3. 18

Figure 3. 19

The distance between the spud and the cutter head determines the maximum cutting width. To
ensure the efficiency of the side winches the maximum swing angle is restricted to 450 ; so
that the maximum width B = 2L*sin(450) +Dcutter, in which L is the distance between the spud
and the cutter head. The length L depends on the depth of the water and the position of the
spud pole.
From the point of view of
production a broad cutting
width is desirable, since per
m3 dredged the downtime for
stepping, anchoring and other

manoeuvres is shorter.
However long cutter suction
T
S
dredgers have a big minimum
cutting width, so the
advantages must be weighed
L=S+Tcos
against the disadvantages.
The maximum cutwidth
depends on the maximum side
winch force too. This will be
explained in chapter 3.2.2.3

B=2Lsin +Dcutter

Figure 3. 20 Cut width

Page 13 of 79

Wb3408b

Designing Dredging Equipment

3.2.4. The type of soil


The type of soil to be dredged has a strong influence on the installed cutter head and side
winch power, the strength of the ladder, pontoon and spuds. To some degree the type of soil
also influences the choice of suction pipe and discharge pipeline diameters. With the same
cutting power a cutter suction dredger dredging rock will have a lower production rate than
when dredging sand. In view of this, a rock-cutting cutter suction dredger should have
pipelines of smaller diameter, because it becomes more economical to pump solids with
higher concentrations. With the same production rate it is possible to increase the
concentration by reducing the pump flow. Because a minimum velocity is required to
transport solids this can only be achieved by reducing the diameter of the pipelines. It must be
noted that reduction of the pump flow may lead to a higher percentage of spillage resulting
caused by a bad mixture forming-process in the cutter head. (See Dredging Processes, Spill.)
3.2.5. The transport distance
The transport distance makes demands in relation to the installed sand pump power and the
need to load barges. The requirement to load barges is determined by the question of whether
the required transport distance is too great to be economically bridged by using a hydraulic
pipeline. It is also possible that the use of a hydraulic pipeline is impossible from the point of
view of hindrance to navigation. Cutter suction dredgers are seldom equipped to load barges
only. Figure 3.21 shows the CD Marco Polo barge loading in the busy waters of Singapore
If the cutter suction dredger is equipped with an underwater pump, the pump power can be
such that during the loading of barges this pump is used only. The pipeline system and valves
should be designed to fulfil this requirement.

Figure 3. 21 CD Marco Polo

It is also possible to choose an underwater pump with a higher power than is needed for barge
loading. The surplus capacity can then be used during discharging.
The grain size and the discharge length of the pipeline determine the required pump pressure,
while this determines the number of dredgepumps required.
The maximum allowable pump pressure that a dredger can supply depends on the quality of
the shaft sealing of the last pump. Often values exceeding 25 - 30- bar are not permitted.
Page 14 of 79

Chapter 3: Cutter Suction Dredger

3.2.6. Access to the dredging site


Dredging sites are not always easy accessible via water. The access can be very shallow and
have to be dredges deeper before the actual dredging can start. If there is no access via water
at all, the dredger have to be mobilised to the site by road. This is only possible with small
demountable dredgers.
In case of long contracts,
such as for the tin and gold
mining the dredgers can be
constructed on the dredging
site. Both cases do influence
the design of the dredger.
Figure 3.17 shows a general
plan of a demountable
dredger consisting of one
main middle pontoon and
two side pontoons.

Figure 3. 22
Another point in relation
with access to the site is the
possible restriction height of
the dredger.
High ladder and spud
gantries can be a problem by
passing bridges or electrical
cables. Compare the different
designs of the dredgers in
Figure3.22 and Figure 3.23

Figure 3. 23

3.3. The dredging equipment


For the design of the dredging equipment the following dredging parts will be considered:
The cutter head
The bow side-winch power
The axial cutting force
The vertical cutting force
The ladder winch power
The drives
The dredge pump
The sand pump drive
The water pump
The spud system

Page 15 of 79

Wb3408b

Designing Dredging Equipment

3.3.1. The cutter head


The cutter head is the most important part for this type of dredger, because it determines the
production in may cases that shall be excavated and transported. For the production is besides
the required cutting power also the cutter head speed and the dimensions important. The
cutting power to be able to cut the soil. The cutter head speed is important for the mixture
forming process and the dimensions should be in relation to the cutting power and the
production.
Further it is important to know the reaction of the cutting process working on the cutter head
for determining the side winch forces, speed and power; the ladder weight, ladder inch forces,
ect.
The dimensions of the cutter head
The production capacity is affected not only by the cutting power, the side winch power and
the velocity, but may also depend on the diameter of the cutter head. This is the case when the
side winch force, the side winch velocity and the cutting torque are not limiting factors.
Production can only be increased by increasing the cut thickness and step size, thus
increasing the cutter head dimension. The dimensions should be in relation with the theory
described in chapter 3.3.2
The cutting power
The required cutting power can be determined either from the cutting theories (Lecture notes
Wb 3413) or from the required specific energy that is needed to cut the design-soil. The
specific cutting energy SPE is defined as the work that is needed to cut m3 of soil, that is the
power P that is needed to cut a production Qcutter of m3/s, thus

SPE =

PCutter
Q Cutter

[N/m]

The cutting power is therefore:

PCutter = Q Cutter SPE

[W]

When cutting soil the cutting force is seldom constant due to the inconstancy of the soil.
Therefore the terms average cutting force and peak forces are used. The peak forces for
rock may well be a factor 2 higher than the average forces (Verhoef, 1997)
The following may be used as rules of thumb:
.
for rock; depending whether the cutting process is ductile or
F peak
= 1.5 2
brittle.

Fmean

F peak
Fmean
F peak
Fmean

= 1.25 1.5
= 1.1 1.5

for sand
for clay, depending whether the cutting process is flow, tear
or shear type.

The theoretical cutting power must also be multiplied by these factors. The revolution
velocity of the cutter head is also dependent on the type of soil.
Note: This factor should be included in the work coefficient as mentioned in chapter 1

Reference to be made

Page 16 of 79

Chapter 3: Cutter Suction Dredger


The cutter speed
Specific energy decreases as the rock size increases. In rock a nominal cutter head speed of 30
revolutions per minute is often used. Lower nominal revolution rates leads to bigger rock
pieces and so to lower specific energy but also to higher torques and cutting forces. Higher
cutting torques and forces can also be achieved by reducing the diameter of the cutter head.
Except that the rock size does not increases in this case the maximum thickness of the cut
decreases and thus the maximum production will reduce.
Both cutter head speed and pump capacity have big influence on the spillage of the cutter.
Spillage is the material that is cut but no sucked up by the dredged pump.
Den Burger (1999) showed from his research on laboratory scale that the optimal cutter head
speed in rock depends a little with the pump capacity (Figure 3. 24)
Translation of the optimum results for the different mixture velocities or pump capacities to
prototype values leads to Figure 3. 25 when using the scale laws as describe by den Burger.
It should be noticed that for a cutter head with a diameter of 3 m the pump capacity should be
more than 5 m3/s (mixture speed 5m/s) to get a relative production of a little more than 70%
(30% spillage). Reducing the cutter head diameter with a half a meter results in more
acceptable practical values for the pump capacity with a cutter head speed of a little less than
40 rpm. Higher speed will give in rock smaller particles and therefore less spillage.

Figure 3. 24
Vm=4 m/s

Vm=5 m/s

120

12

100

10

80

60

40

20

0
0.0

0.5

1.0

1.5

2.0

2.5

Cutter diameter [m]

Figure 3. 25

Page 17 of 79

3.0

3.5

4.0

Pump capa city [m3/s

Vm=2.67 m/s

Cutter head speed [rpm]

As could be expected the


results for dredging sand are
quite different from dredging
rock. In Figure 3. 25. The
results for rock and sand are
plotted against the
dimensionless flow number:
Q
. The difference
R 3
between two soil types is
tremendously.

Wb3408b

RELATIVE PRODUCTION
Gravel 10 mm

Gravel 15 mm

Ladder 25 deg.

Sand

80
70
60
Pr [%]

The productivity depends


except on the capacity and
the cutter head speed on the
particle size and the ladder
angle too (Figure 3. 26) The
flow numbers with the same
productivity for sand at the
(ladder angle also 25) are a
factor 1.5 smaller than for
gravel (10 mm). This allows
the use of cutter heads with a
large diameters and with
higher production results.

Designing Dredging Equipment

50
40
30
20
10
0
0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

Flow number [-]

Figure 3. 26
If the cutter suction dredger is designed for dredging sand a speed of 20 revolutions per
minute is adequate (see also Figure 3. 26). In silt or soft clay even lower revolutions are
sufficient, provided that the cutter head does not become blocked.
3.3.2. The reaction forces on the cutter
Forces acting on the cutter suction dredge are shown in Figure 3.27. All reaction forces from
the cutter head have to be transferred in a certain way the surroundings, either by the side
winch forces or the spud poles to the soil or via the ladder wires and the pontoon to water.
Besides that these cutting forces determines the weight of the dredger, while the forces to
move the dredger through the water can have influences on the design of the dredging parts.
In a ladder related co-ordinate system he cutting forces can be decomposed in the 3
dimensions; horizontal, vertical and axial.
There is a general linear relation between the 3D-cutting forces and the cutting power
(Vlasblom, 1998). Furthermore the cutting forces in cavitating sand, clay and rock are almost
independent for the cutting speed.
Therfore:
Fhor.R cutter
F R
F R
= c h , vert cutter = c v , axial cutter = c a can be taken as constant for a specific soil
M cutter
M cutter
M cutter
type and relative cutting thickness D s .
2Rc
The horizontal and vertical cutting force

M
. cutter is the tangential force T as
R cutter
shown in Figure 3. 28;
Both the cutting force as well as the
normal force can be decomposed in the
horizontal force Fh and the vertical force
Fv. Fh is delivered by the side winch and
Fv by the weight of the ladder or the extra
draught of the pontoon. The axial force is
partly taken up sideline forces, depending
on the directions of those wires and partly
via the thrust bearing of the cutter shaft

Page 18 of 79

Chapter 3: Cutter Suction Dredger


via the ladder trunnion transferred to the
seabed via the spud pole.
Design values are for cv=0.9, ca=0.4 and
ch=1for under cutting and ch=0.6 for over
cutting.
The relative thickness of the cut (d/Dc)
has a considerably greater influence on
the hauling force than on the vertical and
axial forces.)

Rsh

Gs

Mc

Fl

Fv

Wp
WL

Rs

Fa

Fpsw

Rs

Fa+Fv

Fh
Fsbw
Rw

Figure 3. 27

The horizontal component of the cutting force changes in direction when it passes the rotation
centre of the cutter head. (Figure 3. 28, Left
Center
of
Cutter

Tangential Force T
Ra
di

us

Vertical
Force V

Cutting Force C

Tooth

Cutting Force C

Tooth
Normal Force Ncos
Path of Tooth

Horizontal Force H


Radial Force R
Normal Force Ncos

Forces in a plane perpendicular to the cutter shaft

Decomposition of the Forces working on a Tooth

Figure 3. 28

Page 19 of 79

Wb3408b

Designing Dredging Equipment

The axial force


Generally cutter heads have profiles
as given in Figure 3. 29. This profile
is determined by a plane through the
cutter axes and the surface of
revolution shaped by the teeth
positions. Cutter teeth are positioned
such that the centerline of the tooth
is perpendicular to the contour line.
This can easily be understand when
the break out pattern is considered.
(Figure 3. 30,right)
The normal force N can be
de-composed in 2 perpendicular
forces : Nsin and Ncos ,which are
respectively parallel and
perpendicular with the cutter axes.

Nsin

Ncos

Axial and Normal Force


Figure 3. 29

Minimum distance
= cut depth
Break out pattern

Fv

Fa

Fh

R
Break out Pattern
Figure 3. 30

Page 20 of 79

Cutter heads with plain or serrated edges (Chapter 3.4.4) develop axial force by the helix.
angle of the cutter head blade, which causes the so-called snow plough effect (Miedema.
1995).
In that case is the leading edge of
the knife not perpendicular to
direction of the movement (Figure
3. 30, left) The cutting process
have to be considered in 2
perpendicular directions; one
perpendicular with the cutting
edge and the other parallel with it.
The last one takes care for the
transport of the soil in the
direction of the knife. Furthermore
the component of the side winch
forces also gives a force in the
axial direction (Figure 3.

Figure 3. 31

Chapter 3: Cutter Suction Dredger


31),depending on the position of
the anchor.
As with the cutting force, the maximum forces are higher than the average forces.
The ladder weight

Fvert R cutter
= c v = 0.9 the minimum weight of the ladder can
M cutter
be determined in order to fulfil the requirement that over cutting have to be possible.
0.9 Pcutter
Rewriting the condition and multiplying with the rotational speed gives Fvert =
;
Rcutter
R is in the order of 4 m/s, which means that Fvert 0.225 Pcutter
Following from de condition that

If the load on and the weight of the ladder are divided equally over the length of the ladder
than the weight of the ladder W 0.45 Pcutter
The mass of existing ladders is somewhat lower as shown in figure 3.39. This might be
caused by an uneven distribution of the load.

Mladder/Pcutter

Ladder mass over cutter power


0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
0

1000

2000

3000

4000

5000

Cutter power [kW]

Figure 3. 32

3.3.3. The side-winch power and speed


If the relation between the horizontal force and the tangential force is assumed to be constant,
then for the net side winch power:

Pc Fc
=
Ps Fh
Symbol
Fc
Fh
Pc
Ps
Rc
N
vh

=
=
=
=
=
=
=

Page 21 of 79

2 nRc
60 = Fc nRc
vw
Fh 30vw
Parameter
Tangential force
Swing force
Cutter power
Swing Power
Radius Cutter
Cutter head speed
Swing speed

Dimension
[N]
[N]
[W]
[W]
[m]
[rpm]
[m/s]

Wb3408b

Designing Dredging Equipment

For a dredger with a cutter head of radius Rc=1 m, a swing speed v of 20 m/min (.333m/s) and
a cutter speed of 30 revolutions per minute, this gives a relation between the capacities of:

Pc Fc nRc Fc 30 1
F
P
=
=
= 9.4 c with ch=1 follows c = 9.4
Ph Fh 30vh Fh 30 0.333
Fh
Ph
For a cutter head of half this size the relation is:

Pc Fc nRc Fc 30 0.5
=
=
= 4.7 1 = 4.7
Ph Fh 30vh Fh 30 0.333
Ds

is the same for both cutter heads.


2R c
This relative increase in side winch power with reducing cutter head radius is also shown in
the installed power in existing cutter suction dredgers (Figure 3. 33.) Small dredgers have
small cutter head radius and less cutter power.

Here it is assumed that the relative cut thickness

14

ratio CP/SW

12
10
8
6
4
2
0
0

1000

2000

3000

4000

5000

Cutter power [kW]


Figure 3. 33.

Ratio Cutter power over Swing Power

In (Vlasblom, 1998) it is shown that the ratio of the normal force to the cutting force
influences the required ratio of cutter power over sidewinch power too. For sharp teeth this
ratio is 33 but decreases rapidly with increasing wear flat to a ratio of 5 for worn cutter teeth.
In addition to the soil type and the revolutions of the cutter head, both the side winch power
and the side winch (wire) speed depend on the dimensions of the dredger and the position of
the anchor.
It should be noted that the swing force is not equal to the side winch force and the swing
velocity not to sideline velocity. If Fh is the horizontal swing force to move the cutter with a
speed vc and the force in the sideline wire is Fw and de speed vw
It can be proven that in a horizontal plane the power needed to swing the cutter head

Pswing = Fh vh = Fw vw = Pwinch
under the assumption that the friction in the winches, blocks and motors are small.

Page 22 of 79

Chapter 3: Cutter Suction Dredger


Moreover the power required to swing the dredger around its spud depends not only on the
cutting forces but also on the ladder angle and the resistance force W to rotate the pontoon.
The influence of the ladder angle is because the torque on the cutter has a de-component in
the horizontal plane (Figure 3. 27). The moment to swing the dredger around the spud pole
is:

M h = Fh Rsp M c sin + W Rw
in which Rsp and Rw are respectively the distance from the spud to the working point of Fh
and W.
Mc may be either positive or negative, depending on the direction in which the cutter head is
turning.
Therefore the swing power is: Ps = M h

vh
v
= ( Fh Rsp M c sin + W Rw ) h
Rsp
Rsp

For dredging rock the influence of the force W is in order smaller than that of the cutting
reaction forces.
The swingspeed vh should be taken in relation to the production Q, because Q = S Dc vh ,
with S the stepsize in m. and Dc the layer thickness in m.
In the position of the side winch sheave on the ladder (Figure 3. 34, Left) , the relation band
velocity Vz to warping direction of the side winch sheave Vp is equal to:
k

vz
vp

= l

sin

cos
l
(Figure 3. 34, right)
2
2
k
b

cos + sin
l
l

Figure 3. 34

For the cutting of rock the maximum wire velocity is 20 tot 25 m/minute. For cutting sand
values of 30 tot 35 m/minute are taken.

Page 23 of 79

Wb3408b

Designing Dredging Equipment

3.3.4. The ladder winch speed and power


If it is necessary for the cutter suction dredger to dredge slopes completely automatically the
ladder winch speed must be in accordance with the nominal side winch velocity. If this is not
necessary the ladder winch speed may be chosen freely, bearing in mind that at low ladder
winch speeds the production may be significantly affected. When, for example, the teeth of
the cutter head must be frequently changed it will be necessary to raise the ladder many times.
For medium large cutter suction dredgers a value of 10 m/minutes is often used. The required
power is determined by the weight of the ladder and the vertical reaction forces during slope
dredging in the under cutting mode.
3.3.5. The dredge pumps
To decide which pump type is appropriate for the dredger the working range of the
pumpcapacity and pump pressure have to be assessed. Therefore the production capacity in
various types of soil must be translated into:
1. The mixture capacity
2. The mixture concentration
Because:

Q = Q mixture

C vd
1 n

with:
Q
Qmixtur
Cvd
N

=
=
=
=

Production
Pumpcapacity
Transport concentration
Void ratio

[m3/s]
[m3/s]
[-]
[-]

The mixture capacity is determined by the mixture forming process in the cutter (see
chapter 3.3.1.1)
The critical velocity required to keep the material in motion determines the minimum flow
velocity and thus the pipe diameter. v crit = Fl ,H 2 g (S s 1) D in which the value of Fl,H is
determined by the material to be pumped (see Section 2.2.3.3. Suction pipe diameters of
lecture notes Dredging Processes). Ss is the relative density of the solids and D the pipe
diameter in m
Figure Figure 3. 35 from MTI
shows practical values used in the
dredging industry for the critical
velocity in horizontal pipelines
The expected production is
determined by the cutting power,
the side winch power or the side
winch velocities, depending on
which is the limiting factor in the
various types of soil. Using the
equation Q = Q mixture

C vd
1 n

together with vcr gives the pipe


diameter and Cvd

Figure 3. 35

Page 24 of 79

Chapter 3: Cutter Suction Dredger

Jet Pump Power [kW]

3.3.6. The jet pump


To promote mixture forming when dredging sand some cutter suction dredgers are equipped
with water jet installations. One or more jets are mounted on the sides of the ladder close to
the cutter ring.

2500
2000
1500
1000
500
0
0

500

1000

1500

2000

2500

Cutter Power [kW]


Figure 3. 36.Jetpump

power versus cutter power

The power needed for the jets depends strongly on the insight of the designer as.Figure 3. 36
shows. For more theoretical insight into this phenomena the chapter jet pumps for plain
suction dredgers should be consulted.

3.4. The drives


The drives of the cutter head, the side winches and the ladder winch are either electric or
hydraulic drives. Formerly the ladder winch and the side winches were combined to form a
tree drum winch with one drive, which made simultaneous operation of the ladder and side
winches impossible. With hydraulic systems various drives can run on the same hydraulic
circuit and for this reason they can influence each other. The best choice of what may or may
not run on the same circuit is important for the operation and thus finally for the production of
the dredger.

3.4.1. The cutter head drive


The cutter head drive is mounted on the ladder either near the hinge side (the trunnion) or
close to the cutter head. In the first case the drive and the gearbox are above water and in the
second case these may be in a box under water.

Page 25 of 79

Wb3408b

Designing Dredging Equipment

Figure 3. 37

If the drive of the cutter


head is mounted near the
hinge the shaft must be both
long and heavy because of
the high torque. This long
shaft needs several ladder
bearings. When the drive is
mounted close to the cutter
head there is more freedom
to adapt the direction of the
cutter head axle to the
required angle, especially
when dredging in shallow
water.

Figure 3. 38
The choice between hydraulic and electric drive depends primarily on the expected relation
between the average load and the peak load. Electric drives are especially suitable because
they can take overloading up to 150% without stalling (Figure 3. 39, right). This is possible
because of the considerable rotation energy of the rapidly turning electric motor. As a result a
flywheel effect is created. The long driving shaft also plays a role in this.
However, due to the strong dynamic character of the dredging process, gearboxes for cutter
drives have to resist heavier loads than gearboxes for the all drives on board of the dredge.
The dynamic cutting process and as consequence the torsion vibrations cause remarkable
increase of the torque. It is even possible that due to these vibrations negative torques occur in
the shaft and gearboxes with a result hammering of the gears. Such situation decrease the
live time of the gears. Therefore gearboxes for heavy duty cutter dredgers are designed to
resist a torque of 3.5 of the nominal torque. (Hiersig, 1981)

Page 26 of 79

Chapter 3: Cutter Suction Dredger

Speed [%]

Speed [%]

TORQUE - SPEED CHARACTERISTIC


(Simple Hydraulic Drive)

TORQUE - SPEED CHARACTERISTIC


(Electric Drive)

100

100

100

100

Torque [%]

Torque [%]

150

Figure 3. 39

With hydraulic drives the torque is determined by the piston displacement of the engine and
the pressure in the system. When overloading occurs a safety valve which limits the pressure
operates, stopping the engine. This means that the average pressure c.q torque is usually
considerably lower than the maximum in the order of 60-70 % (Figure 3. 39, left). Hydraulic
drives do have the advantages of being completely watertight and of driving the cutter head
directly without a gearbox. Often several hydraulic drives are used simultaneously to provide
the cutter head with the desired power.

3.4.2. The side winch drives


Here too, the drives may be electric or hydraulic. This choice is based on the same line of
reasoning as that followed for the cutter head drive. It is not necessary that when the cutter
head drive is electric the side winch drives must also be electric. The required power for the
side winch drives is roughly a factor 5-10 smaller, so often secondary matters such a
standardisation and price play a different role
3.4.3. The ladder drive
Because the depth of the cutter head is set with the aid of the ladder winches, the drives must
be easy to regulate and must not slip when the ladder drive is not activated. The latter happens
frequently with hydraulic drives as a result of leakage of the hydraulic fluid, resulting in
changes of the cutting depth the dredging operation. To prevent this slipping the winch must
be equipped with a break or ratchet.
3.4.4. The sand pump drives
Underwater pumps are often electrically
driven. If barge loading is required with
the underwater pump, it is necessary to use
drives with speed control. With a fixed rate
of revolution, f.i. an asynchrony ac-current
motor, the variations in flow resulting from
differences in concentration and grain size
are often too big for the efficient loading
of the barges or leads to overload of the
motor.
Nowadays underwater pumps for small
dredgers can also be driven by diesel
engines via a pivoting gearbox. (Figure 3.
40)

Page 27 of 79

Wb3408b

Designing Dredging Equipment

Figure 3. 40
Diesel drives are most suitable for the discharge pumps. The choice between one or more
pumps and thus diesels depends on the total required pump pressure and the requirements in
relation to the speed control of the diesel engines. It will be clear that when only one large
pump is installed it is not so easy to control the pumping system for long and short pumping
distances. Very important when using diesel drives is the type of governor. Modern governors
limit the fuel injection at low revolution to avoid incomplete burning of the fuel. These
governors increase increases the speed control of the diesel engines.
For jet pumps diesel engines or an asynchrony ac-current motor are used often. Speed control
is less important for jet pumps than for dredge pumps, because of the almost fixed layout of
the pipeline and the constant fluid density.

3.5. Spudsytems
The choice of the spud system plays an important part in the design of the cutter suction
dredger. The spud system influences not only the layout of the pontoon, but also the
efficiency of the cutter suction dredger. The most frequently used systems are the spud
carriage system and fixed spuds (several other systems have been mentioned in the section on
technical construction).

3.5.1. The spud carriage system


With the spud carriage system the
work spud is placed in a carriage
which, with the aid of a hydraulic
cylinder, can travel over several
metres (4 - 6 m) (Figure 3. 41) in
longitudinal direction in a well at
the stern of the dredger. The
carriage is generally positioned in
the centre of the dredger (Figure 3.
42) and is support by four wheels
on rails for the vertical forces and
by guide rollers or bearing strips
for the lateral forces. The cylinder
is a double acting hydraulic ram.
Spud carriage

Figure 3. 41. Spud carriage

A second spud, the auxiliary spud is mounted at the stern of the pontoon, which is used to
move the carriage back to its start position.

Page 28 of 79

Chapter 3: Cutter Suction Dredger


The initiation of a new cut is
obtained by moving the spud
carriage one step forwards. After
stepping, the cutter head describes
concentric circles until the spud
carriage reaches the end of the
stroke of the hydraulic cylinder.
The return of the carriage usually
takes place in the middle of a cut in
the following sequence of actions.
The auxiliary spud is lowered and
the work spud is lifted, the carriage
is moved back and then the spuds
again changed.
After each single swing the dredge
master is free either to step
forwards or to lower the ladder till
the final is reached.

Cut width

Workspud
in carriage

Spud carriage
length
Auxilary spud

Vertical swing pattern

Figure 3. 42

In addition to the spud carriage in the stern


well of the main pontoon of the dredger, it is
also possible to have a separate spud carriage
pontoon. This pontoon is fixed to the cutter
suction dredger by a stiff link, usually by
making use of the existing auxiliary spud
carriage
. This is done to change the existing, less
efficient spud system or to make a wider
swing (Figure 3. 43and Figure 3. 44). It is
also necessary to move the pivoting bend on
the stern of the dredger to the rear of the spud
pontoon.

Figure 3. 43

Page 29 of 79

Wb3408b

Designing Dredging Equipment

Spud carriage

Auxiliary spud

Spud carriage pontoon

Figure 3. 44

3.5.2. The fixed spud system


When using fixed spuds both the work
spud and the auxiliary spud are in
fixed positions on the stern of the
pontoon at equal distance from the
centre line of the dredger (Figure 3.
45).

Figure 3. 45
The step or start of the cut is now
initiated by letting the dredger make an
angle from the centre line, then
lowering the auxiliary spud and lifting
the work spud.
The dredger is then swung into a
symmetrical position with regard to the
centre line where both spuds are
changed again (Figure 3. 46).
After each single swing the ladder is
lowered till the final depth is reached.
It will be clear that stepping with fixed
spuds takes considerably longer than
with a spud carriage, due to the down
time of the swing movements.

Figure 3. 46
Note that the arc is not symmetrical with regards to the centre line of the cut.

Page 30 of 79

Chapter 3: Cutter Suction Dredger

As an example the difference in effective dredging time has been worked out for a spud
system with fixed spuds and one with a spud carriage. Both dredgers are the same with regard
to size and power. The following boundary conditions are taken for the work:
B
vs

Width of cut
Swing velocity

75
15

S
Lsc

Step size
Effective spud carriage
length
Distance between fixed
spud and cutter head
Distance between fixed
spuds
Number of cut layers
Number of steps per
carriage movement

1
5

Nc
Ns

Time
Spud carriage travel

[m]
[m/
s]
[m]

80

[m]
[m]

10

[m]

Spud changing
2 min.
Change in swing direction incl. lifting 2 min.
and lowering ladder minute

[-]
[-]

Fixed spuds

2 min.

=Lsc/Ns

Spud Carriage

Fixed / carriage
0.9

Efficiency [-]

0.8
0.7

0.7

0.6
0.5

0.5

0.4
0.3

E_fixed/E_carr

0.9

0.3
0

10

12

Number of cuts [-]

Figure 3. 47 Effectiveness of spud systems

The above example (Figure 3. 47) clearly shows the superiority of the spud carriage system
over a fixed spud system.

Page 31 of 79

Wb3408b

Designing Dredging Equipment

3.5.3. The spud door system


For small dredger a cheaper system than the spud carriage
is developed by IHC-Holland; the so called Spud Door
In A heavily constructed door, pivoting around the
auxiliary spud, is placed the working spud. The dredge
pattern is the same as for the spud-carriage system,
however spuds have to be changed more frequently and the
accuracy is less because the working spud stays not exactly
in the centerline of the dredger. The system is much
cheaper than the spud carriage system.

Figure 3. 48

3.5.4. The walking spud system


The walking spud system is similar to the spud carriage system with regard to the movement
of the cutter head during swinging and stepping. The working spud is not in a carriage but
swivels round a horizontal axis (Figure 3. 49). The step is now taken by allowing the spud to
tilt to the requisite angle.
The disadvantage is immediately apparent; the maximum step depends on the depth of the
water and so walking spuds are difficult to use in shallow water. The disadvantage is that it is
very little or not at all cheaper than a spud carriage. The dredging pattern is similar to that
with a spud carriage, while the number of spud movements is considerably larger.

Walking spud

Figure 3. 49

Page 32 of 79

Chapter 3: Cutter Suction Dredger

3.5.5. The rotor spud system


This system was already invented in the early years of 20th century. With the rotor spud
system both spuds are in a rotor and stand on the ground diametrically opposite each other.
(Figure 3. 50 ).

Figure 3. 50 Rotor spuds

During dredging the midpoint of the rotor


remains in the centreline of the cut, so the
dredger turns round the rotor. Stepping is
accomplished by lifting the rear spud and
turning the rotor until the rear spud becomes
the front spud. The step S=2*L*sin(2), in
which L is the distance between the spuds
and the angle through which the rotor
turns. Using this system the dredger makes a
pattern of concentric circles. The advantage
of this type of system is that when stepping,
only one spud has to be raised and lowered.
This disadvantage is that it is very expensive,
certainly for the large cutter suction dredgers.
Moreover the spuds cannot be placed
horizontally.

Figure 3. 51
From the point of view of efficiency, here defined as the actual dredging time in relation to
total time per spud cycle, the spud wagon is the best. The number of spud changes per metre
of progress is minimal. With a well-chosen cutting pattern no partly or entirely unproductive
swings (warping without cutting) are needed.
Likewise the rotor spud and tilting spud systems have advantages over the fixed spud
systems.

Page 33 of 79

Wb3408b

Designing Dredging Equipment

3.5.6. The Christmas tree Page 31 of 79


There are situations in which anchoring by means of spuds is not possible. Such a situation
arises when working at sea if the forces that waves or swell can exert on the spuds are too
large. In that case one changes to working on wires. For this a Christmas tree (Figure 3. 52), a
construction with wire leads, is mounted in one of the auxiliary spud carriages. With this the
anchor wires meet at one point under the under the hull. However, in order to keep the cutter
head well into the face throughout the entire swing the laterally directed anchors of the
Christmas tree must stand well forward. with the disadvantage that they must be moved
frequently. For this reason a bow anchor is often used.
One of the advantages is the possibly to work in deep water, but this can only be done in
special cases. In a well designed cutter suction dredger the spuds are so long that they can
reach the maximum dredging depth at all times, so dredging in deep water is only possible
with an extension by means of a special ladder construction.
A very real advantage of working on anchors is that a considerably bigger cutting width can
be achieved..
Obviously the disadvantages overweigh the advantages, otherwise the system would be more
widely used. These are:
At least three anchors must be moved.
The freedom of movement when working on anchors is so great that it is almost
impossible to dredge accurately
This is equally true for dredging in hard soil. A star system is needed for this.

middelste lier "Christmas tree"

BB lier "Christmas tree"

SB lier "Christmas tree"


acht

SB achterzijanker

erzijd
e

zuige
r

50

75
50

30

BB achterzijanker
achteranker

Figure 3. 52

Page 34 of 79

Chapter 3: Cutter Suction Dredger

3.6. The general layout

Figure 3. 53 CD EDAX

Depending on the spud system the hull may consist of a simple U-shapes pontoon (with fixed
spuds) or an H-shaped pontoon (with a spud carriage system). The main dimensions; length,
beam and draught of the pontoon derive from the requirements in relation to the above
mentioned design parameters and the associated requirements in relation to stability and
strength. Figures 3.54 and 3.55 gives design information for the pontoon.

Page 35 of 79

Wb3408b

Designing Dredging Equipment

10,000
y = 0.4664x
R2 = 0.9597

9,000

y = 0.3485x

8,000

8,000

R = 0.925

6,000

7,000

Light weight [t]

Light weight [t]

10,000

4,000
2,000

6,000
5,000
4,000
3,000
2,000

1,000

5,000

10,000

15,000

20,000

25,000

5,000

10,000

Total installed power [kW]

15,000

20,000

25,000

BLD [m3]

Figure 3. 54
L/B

B/T

12.00

L/B & B/T

10.00
8.00
6.00
4.00
2.00
0.00
0

1000

2000

3000

4000

5000

6000

7000

8000

9000

Light weight [t]

Figure 3. 55

The engine room, the pump room and sometimes in larger cutter suction dredgers, also the
control room for the machinery, are located in the pontoon. In smaller cutter suction dredgers
the sand pump is sometimes located on the engine room directly in front of the engine, with
all the well-known disadvantages of such an arrangement.

Figure 3. 56

Page 36 of 79

Chapter 3: Cutter Suction Dredger


A frequently used layout is shown in Figure 3. 56. Here the pump room is directly aft of the
bow well; aft of which is the engine room. The fuel and ballast tanks are located in the side
pontoons of the fore and aft wells.
The storerooms are located in the side pontoons of the forward well. The hydraulic system
drives, workshops and a galley for the *local crew are often located in the side pontoon next
to the well for the spud carriage. Mess rooms, toilet facilities and possibly also crew quarters
are above deck.

Figure 3. 57

If the cutter suction dredger has been designed to work in the tropics the generators are
separated from the engine room to assist in the cooling of these machines (Figure 3. 57).
Opmerking [T1]: Ook deze
figuur moet vergroot worden

Figure 3. 58

Figure 3.56 shows a dredger with the spud carriage out of the centre line of the dredger, while
the cutter lead axes is the the centre line . This means that the teeth position is not optimal for
both sides and as a consequence this will result in more teeth wear.
Self propelled cutter suction dredgers have a more complicated layout resulting from the two
possible modes of working; dredging and sailing. The propulsion mechanism can be located
at the ladder end (CD Taurus, CD Marco Polo, CD da Vinci) or at the spud end (CD Ursa, CD
Oranje). In the second case the dredger sails with the ladder at the front and port and
starboard is the same for both sailing and dredging. Moreover the propellers are directly
driven by the main engines. This is not possible in the first case, so the propellers are powered
by electric motors. The layouts described are therefore self explanatory (Figure 3. 59).

Page 37 of 79

Wb3408b

Designing Dredging Equipment


Opmerking [T2]: Tekening
van een zelfvarende snijkopzuiger

Figure 3. 59

Self propelled cutter suction dredger Ursa, built in 1986

Small to medium sized (to 3500 kW) cutter suction dredgers are often used to make roadbeds.
To permit overland transport to the sand extraction area these dredgers are demountable.
Because of the need for strength, the main pontoon in which the pump and diesel engine are
located is usually constructed as a single unit. When designing demountable dredgers it is
necessary to consider how the parts of the dredger will be transported by road or over water.
In the first case the maximum size of the pontoons is determined by the permitted size and
weight for road transport. For smaller dredgers the pontoons are made up of 40 or 20-foot
containers, while the other parts are of such size that they can be carried in containers.

Page 38 of 79

Chapter 3: Cutter Suction Dredger

Figure 3. 60 Beaver 1600 on 5 trucks

Opmerking [T3]: Deze figuur


is veel te onduidelijk

Ballast tank
Lubricating
olie
Ballast tank

Fuel

ballast
tank
ballast
tank

Spare parts

Spare parts

Dry
tank

ballast
tank

Drinking
water
Engine room
Ballast tank
Ballast tank

Fuel

ballast
tank
ballast
tank

Spare parts

Spare parts

Dry
tank

ballast
tank

Figure 3. 61

In demountable dredgers also, the pump room and the engine room are located one behind the
other in the main pontoon and the ballast tanks and storerooms are in the side pontoons
(Figure 3. 61). With containerized dredgers the entire vessel is built up out of containers. In
this case the pump and motor are often in a container on deck (Figure 3. 62).

Page 39 of 79

Wb3408b

Designing Dredging Equipment

Figure 3. 62 Containerised Dredger

3.6.1. The Hull


The floating capacity of a stationary cutter suction dredger derives from the pontoon that is
constructed as a single unit (mono-hull or mono-pontoon) for most large cutter suction
dredgers and, for demountable cutter suction dredgers, consists of several pontoons. The
pontoons beside the ladder well are often chamfered to form trapezoids in order to limit the
minimum width of cut.
It is essential that there is a separate pump room: if the pumps were located in the engine
room a leakage or an error during inspection of pumps might result in the flooding of the
engine room with a good chance of the dredger sinking. The pump room should be designed
in such a way that, when flooded, the dredger doesnt sink. Furthermore the pipeline system
must be designed in such a way that the flooding of the pump room can be kept to a
minimum.
Consider therefore:
a remote controlled valve behind the well bulkhead. This is necessary for the changing of
the rubber suction hose
a bilge alarm.

Page 40 of 79

Chapter 3: Cutter Suction Dredger


In designing the hull it is necessary to take into account that a part of the reaction forces from
the dredging process must be transferred to the work spud via the hull. For this reason the
main pontoon of demountable dredgers is constructed as a single unit. This means that the
ladder hinge and spuds are mounted on the main pontoon, so the side pontoons as well as the
links to the main pontoon are not so heavily loaded.
The ladder gantry spans over the forward well as a simple A-frame, a frame construction or a
frame in the form of a box girder construction. When dredging in undercut the vertical
forces are transferred to the pontoon via the gantry.

Figure 3. 63 Different ladder gantries

3.6.2. The cutter head ladder


Originally the cutter ladder, or cutter ladder was constructed as a frame girder with two
longitudinal girders consisting of steel beams connected to each other by many transverse
beams and struts. The name cutter ladder derives from this structure. The transverse beams
were used as supports for the cutter shaft bearings.
The ladder that is located in the forward well is hinged (the trunnion) on one end to the
pontoon and a tackle and ladder wire to the ladder gantry suspends the other end. The ladder
wire runs via the ladder gantry and various sheaves to the ladder winch to adjust the desired
depth.

Page 41 of 79

Wb3408b

Designing Dredging Equipment

Because owing to the transverse


forces it is essential for the ladder
of a cutter suction dredger to be
stiff, for the large cutter suction
dredgers
a
double
box
construction is used, strengthened
by longitudinal and transverse
links. Furthermore this has the
advantage that the ladder is given
sufficient weight. This weight is
needed in order to swing the
cutter head to both sides. If the
ladder is not heavy, as in the case
of small cutter suction dredgers,
extra arrangements must be made.
For example the cutter head drive
can be mounted as close as
possible to the cutter head. Lead
is often added close to the cutter
head. For very heavy cutter
suction dredgers the requirement
of the stiffness may exceed the
demand for sufficient underwater
weight. In this case the ladder is
equipped with floats.

Figure 3. 64 Boxtype cutter ladder

In small cutter suction dredgers the ladder is often built up from basic elements. The ladder is
supported by pins that are fixed to the ladder and rest in bearing houses that are rigidly fixed
to the pontoon.
The drive of the cutter head is either at the top of the ladder, thus at the hinge side or below
near the cutter head. In the first case the drive and the gearbox remain above water and the
cutter head is driven by a long shaft, sometimes tens of metres long. Because of the high
torque demanded by the cutter head this shaft has a considerable diameter. The shaft has
supported at various points and must, especially in the case of heavy cutter suction dredgers,
be on the centreline of the ship.
The end bearing, (Figure 3. 66 and Figure 3. 66) close to the cutter head is made of rubber
and lubricated by water. The axial forces are taken up by a pressure bearing that is mounted in
the gearbox.

Page 42 of 79

Chapter 3: Cutter Suction Dredger

Figure 3. 65 Rubber end Bearing

Cutter ring
Gland water

Cutter shaft
Bearing bush

Release ring
Rubber bearing
Suction mouth
Cutter hub
Cutter blade
Figure 3. 66

3.6.3. The cutter head


The production of the cutter suction dredger is largely determined by the cutter head. Its type
and size depend not only on the technical specifications of the cutter suction dredger,
including cutting and side winch power, cutter revolutions and the weight of the ladder, but
also on type of soil to be dredged. With relatively high side winch forces and a small cutter
diameter, higher cutting forces can be generated and thus harder soil can be cut. In contrast,
with the same cutter power in soft ground it is necessary to use a bigger cutter diameter and
exchange the high side winch forces for a higher speed by changing the gears of the side
winch drive. When cohesive soil is being cut different boundary conditions play a role, for
example, the need to avoid blocking the cutter head.
General guidelines for cutter heads for various types of soil.(Figure 3. 68):
for hard soil. Suitable to withstand impact forces on one or more teeth, thus heavy and
robust. Small in contour with replaceable teeth. Can withstand extreme wear on both the
cutter head itself and on the teeth and adapters. Good, accurate tooth positions. The size
of the fragments may not exceed the minimum passage of the pump.
for non-cohesive soil. Suitable for very high production rates Good mixture formation
required. Many replaceable chisels (wide or narrow) or cutting edges. Wide though
flattened contour (little pumping action). Well able to withstand wear, especially of the
cutting elements. Here also good, accurate tooth positions are needed.

Page 43 of 79

Wb3408b

Designing Dredging Equipment

cohesive soil. The cutter head may not become blocked, so is ample and round in
contour. Open near the hub. Often with one less blade (thus 5 blades). Good cutting
properties in clay, small fragments. Plain or serrated edges or many small teeth.

HUB

Ring

Figure 3. 67

Sticky soils
Non sticky soils
Rock

Elements of a cutter head

Contours

- open to prevent blockage


- multi purpose
- high torque

Figure 3. 68 Cutter head contours


Although it is better to use a different type of cutter head for each type of soil, cutter heads
are marketed that can be used in more than one type of soil. The so-called multipurpose
cutter is a compromise with regard to contour.
A cutter head is comprised of the following parts (Figure 3. 67).
The back ring, that is the ring on the underside of the cutter head. The inside diameter of
the ring is such that this fits the suction mouth and or the cutter shield (Figure 3.66).
The hub by which the cutter head is mounted via an Acme or three threaded screw onto
the cutter shaft. The distance between the underside of the ring and the underside of the
hub is termed the set height.

Figure 3. 69

Page 44 of 79

Chapter 3: Cutter Suction Dredger

The cutter arms or blades, usually 5 or 6. The number is related to the required strength
and/or space between the arms. The cutter arms form a screw shape and link the ring to
the hub. The cutter head is termed a normal helical cutter head if the chosen screw shape
is such that the dredged material is transported to the ring. (Figure 3. 69 left) If the thread
of the screw runs in the other direction the cutter head is termed a reverse helical cutter
(Figure 3. 69 right).
Edges (knives) or replaceable teeth or chisels are mounted on the cutter arms. The tooth is
attached by means of a locking pin to an adapter that is fastened to one of the blades. In
hard soil a six bladed cutter head is often used with teeth on the even blades that are offset
in relation to those on the uneven blades. This is termed staggered mounting.
The turning direction of a cutter head is defined when looking from the control cabin
towards the cutter head; that is against the underside of the ring.
The passage through the cutter head increases towards the ring. This may cause blockages
in the pump if fragments that are too large for the pump can be taken up. The passage
through the cutter head is sometimes reduced by the addition of skirts, which are welded
onto the blades to extend the cutter arms(Figure 3. 70).
The passage can also be reduced by the welding of plates perpendicular to the blades
(Figure 3. 70).

Figure 3. 70

Besides the turning


direction the height H
between the under side
of the hub and
underside of the ring,
the internal ring
diameter Di and the type
of tread in the hb are the
important data for
mounting the cutter well
on the shaft and
ladder.(Figure 3. 71)

Double
ACME
Tread

Hub

Cutter blade
H
Cutter ring
Inner diameter Di

Figure 3. 71

Page 45 of 79

Cutter (teeth)
contour

Protection plate

Opmerking [T4]: Page: 37


foto de Heer II blz 121

Wb3408b

Designing Dredging Equipment

3.6.4. Tooth and cutting edge systems


There are various tooth and cutting edge systems on the market, each with its own advantages
and disadvantages. They are all based on the principle that it must be possible to quickly
replace the parts that are subject to heavy wear. In addition to the property mentioned above,
a tooth must satisfy the following requirements:
There must be a good transfer of the cutting force to the cutter arm.
The positioning of the teeth and adapters must be such that there is little or no wear on the
cutter arms. The blades must therefore run freely.
Mixture formation in the cutter head is promoted.
.

Figure 3. 72

As shown in Figure 3. 72, there is a wide range of types of tooth and chisel.
The use of the specific type of tooth depends on the strength of the soil.

pick points short

Page 46 of 79

: hard rock

Chapter 3: Cutter Suction Dredger

chisels

long
trapezoid
narrow
wide
flared

: rock
: soft rock
:cemented sand
:sand and loose soil
: clay

Figure 3. 73 Tooth Systems

A*

Cutting angle
Cutting angle

Rake angle

CONVENTIONAL

Figure 3. 74 Vosta tooth System

Page 47 of 79

Rake angle

VOSTA D

Wb3408b

Designing Dredging Equipment

The best known systems are:


Esco (Figure 3. 73 left)
Florida (Figure 3. 73 right)
Vosta (Figure 3. 74)
The first two types are very similar to each other.

3.48 Verschillende adapter typen.

3.49, Spherilok systeem.

Figure 3. 75 Adapter systems

The difference lies in the fitting of the tooth and the adapter (Figure 3. 73
Four types of adapter can be distinguished of both systems, these being:

the weld-on adapter


the single-leg adapter
the double-leg adapter
the Spherilock adapter

From above downwards these adapters have a reduced grade of freedom in positioning. On
the other hand the chance of incorrect positioning during repairs also decreases.

Page 48 of 79

Chapter 3: Cutter Suction Dredger


There is a wide variation in the types of teeth and chisels used by these systems, depending on
the material to be dredged. The adapters take up the cutting force, which implies that there
must be a good fit between the tooth and the adapter, in other words the tooth must not be
loose. The joint is secured with a locking pin, which is prevented from falling out by a
flexible rubber locking keeper.
The Vosta system is clearly different from the Esco and Florida systems (Figure 3.73).

ADAPTER EDGE

TOOTHED EDGE

SERRATED EDGE

PLAIN EDGE

Types of cutter knives.

Figure 3. 76

In addition to cutter heads with replaceable teeth or chisels there are also cutter heads with
cutting edges. The edges welded directly onto the cutter arm of the cutter head, with or
without a fitting lip (see Figure 3. 76) Such types of cutting edge are suitable for various
types of. edges.
The main shapes are :
plain edges
: for various types of soil
serrated edges
: for clay
: for hard clay
toothed edges
adapter edges
: for hard clay
These edges can also be obtained as projecting offset edges. In this case the plane of the edge
forms an angle with the cutter head arm. This prevents material such as clay from sticking to
the arm.

Page 49 of 79

Wb3408b

Designing Dredging Equipment

3.6.5. The side wires


As said, the dredger is moved over the width of the cut by hauling on one of the side wires
while at the same time paying out the other. The side wires run from the side winches via the
side wire sheaves to the anchors
The side wire sheaves, which are fastened at
the lower end of the ladder must be able to
adjust to the angle that the side wire makes
with the plane of the horizontal, because the
anchor is not usually at the same level as the
point of attachment of the side wire to the
ladder. The position of the side wire sheaves
and the anchor determines not only the force in
the side wire, but also the speed at which the
cutter head moves. (Figure 3. 77)

Figure 3. 77Side wire sheaved in upwards position


The side line winches can either be placed on
the ladder or on the pontoon. Some heavy duty
cutter suction dredgers have double drum
winches (Figure 3. 78). The side line wire is
first laid over a grooved drum with a relative
small diameter to a drum with a bigger
diameter. On the grooved drum sufficient wire
length can be stored to swing over a full cut
width On the big drum additional wire can be
stored.

Figure 3. 78

Figure 3. 79
Figure 3. 80
Figure 3. 79shows the sheaves on the ladder to guide the side wires to the winches on the
pontoon and Figure 3. 80hydraulic winches on a Beaver Dredger.

Page 50 of 79

Chapter 3: Cutter Suction Dredger

3.6.6. The anchor booms


Anchors can be moved by a floating crane, assisted by a flatboat. To keep anchoring
movements to the minimum, they are dropped as far as possible from the dredger. Modern
cutter suction dredgers are often equipped with anchor booms, which makes it possible for the
skipper to move the anchors without outside assistance.
Top wire

re
y wi
iliar
x
u
A

Buoy wire
Ancher boom

Figure 3. 81 Anchor boom

The anchor booms are placed on the bow pontoons at the point where the chamfering starts
(Figure 3. 82) and fastened to the deck by a pivoting construction. Each anchor boom is
fastened by one or more wires to a frame or, as if often seen, to the ladder gantry.

Figure 3. 82 Al Mirfa changing her anchor position

Page 51 of 79

Wb3408b

Designing Dredging Equipment

The anchor boom can turn on its pivoting construction by means of the anchor wires which
are fixed to the top of the anchor boom and which run via a series of sheaves to the anchor
winches. The anchor wire, which is used to pull up the anchor, runs from the anchor to the top
of the anchor boom via the anchor boom downward and then via a set of sheaves to the
anchor winch.

3.6.7. The spuds


The spuds are fastened via spud doors to
the spud carriage or the pontoon.
Because the spuds are loaded on a bending
moment the wall thickness increases with
the stress level (Figure 3. 83 right). To
obtain a good penetration into the soil, the
lower ends of the spuds are pointed. In hard
soil the spud is often dropped in free fall
and needs therefore a massive point (Figure
3. 83 left)

Figure 3. 83

In soft ground, on the other hand, the spuds are set down to prevent them from sagging too far
into the ground.
During transport the spuds must be carried horizontally, so most cutter suction dredgers have
special equipment for this purpose.

3.6.8. The spud lifting system


In order to move the dredger, the spuds must be lifted and various systems for this are in use.
The simplest method is one in which the spud is hoisted by means of a wire attached to the
upper end(Figure 3. 84 .a). This method is often used by American cutter suction dredgers
and has the advantage of simplicity and accessibility when wires break.
.

Page 52 of 79

Chapter 3: Cutter Suction Dredger

sling

a
Figure 3. 84 Spud Lifting systems

The great disadvantage is the high construction height needed to lift the spud in this way. It is
also difficult to extend the spuds, should this be necessary. In order to avoid this disadvantage
the spud can be hoisted on a wire that runs through a pulley mounted on the underside of the
spud (Figure 3. 84.b). Although this is still a simple construction it has the disadvantage that
when a wire breaks it is not easy to thread the new wire through the pulley and it is necessary
to use either a diver or a crane.
Many cutter suction dredgers lift their spuds by means of a sling, which is clamped round the
spud by the tension in the hoisting wire. The hoisting wire runs over a sheave that is attached
to a double action cylinder above and which runs down to a fixed position on deck. The spud
is then hoisted by extending the cylinder (Figure 3. 84.c). This construction has the advantage
that all the parts are easily accessible and it is not a high structure. Moreover the spud can fall
freely because the sling is self releasing. The disadvantage is that the lifting height is
restricted by the stroke of the cylinder. In that case the spud must be taken over. For this
reason the spud has holes through which pins can be pushed so that the spud remains
suspended on the auxiliary carriage.

3.6.9. Pumps and pipelines


The suction pipeline
The suction mouth is mounted under
the end bearing and opens into the
cutter plate/shield (Figure 3. 85).
The area of the suction mouth is
usually a little bigger than the area of
the suction pipe (1.2/suction pipe). In
some cases the suction mouth is not
symmetrical but somewhat turned in
the turning direction of the cutting
head. This gives less spillage when
over-cutting (cutter head turning in
the direction of swing).
The suction pipe must be mounted in
or under the ladder in such a way that
parts can be easily changed.
Figure 3. 85 view on suction mouth of CSD Ursa
Page 53 of 79

Wb3408b

Designing Dredging Equipment

The connection of the suction pipe on the ladder to the pipeline in the ship must be flexible
because of the pivoting movements of the ship. Often a suction hose is used. This is a heavy
cylindrical rubber hose with steel rings embedded in the rubber to prevent it from collapsing
when under pressure occurs.
When dredging in coral or coral-like types of rock, suction hoses cannot be used owing to
the sharpness of the fragments of coral that cut the rubber. In such cases a ball joint from a
floating pipeline forms the link. The angle through which the ladder rotates is then usually
more restricted than when a suction hose is used. It is also recommended that an extra suction
pipe be placed in front of the first on board pump through the bottom of the hull. When using
long discharge pipelines this extra suction pipeline makes it possible to raise the ladder, for
example to inspect the teeth, while the pumps are still being used to clean out the discharge
pipeline.

The pumps
For cutter suction dredgers without an underwater pump the suction pipelines should be kept
as short as possible and the position of the first pump should be as low as possible under the
waterline. Where the suction pipe emerges above water the chance of air being sucked into
must be minimized. (The taking in of air has the same effect as cavitation.) Besides good
discharge characteristics the first pump must also have good suction characteristics. In other
words a high vacuum limit and/or low NPSH-value.
If the dredger is equipped with an underwater pump the layout is less critical and factors
such as accessibility for inspection and repair play a more important role. The inboard pump
requires only good discharge characteristics. If there is more than one inboard pump on board
the layout must be such that, if desired, the ladder pump and one of the inboard pump can be
used. All pumps must have an inspection hatch so that the pump and impeller can be
inspected and, if necessary, to remove debris.

3.4.4.1 The discharge pipeline


The pipeline runs from the pump room high above the deck to the stern (Figure 3.57). In the
pipeline on board are:
an expansion joint to take up possible changes in length.
a gate valve in case it is necessary to prevent water from running back from a higher-level
disposal site.
an air release valve
a suspension bracket from which lower bend can be suspended and still rotate.
a lower bend with a ball joint to which the floating pipeline can be attached. A suction
hose may be used instead of a ball joint.

Page 54 of 79

Chapter 3: Cutter Suction Dredger

Figure 3. 86 Pipeline layout on a dredger

3.6.10. The winches


The ladder winch
As previously stated, the depth of the cutter head is adjusted by means of the ladder winch.
This variable speed winch may be an electric or a hydraulic drive. For heavy ladder
constructions, with consequent high forces on the wires, the winch drums are grooved to
prevent wire weir. The size of the drums needs a diameter to accommodate the entire wire in
the groove. During repairs and transport the ladder is kept in a fixed position (Figure 3.87),
often by slings or rods that are directly fastened to the ladder gantry.

Figure 3. 87

The side winces


The dredging process is controlled with the aid of the side winches. To a large extent the
production of a cutter suction dredger is determined by the swing speed. The hauling winch
takes care of the feeding of the cutter head, while the paying out winch ensures that wire
remains taught. The side winches may also have electric or hydraulic drives.

Page 55 of 79

Wb3408b

Designing Dredging Equipment

Modern cutter suction dredgers are often equipped with an automated cutter control system
which controls the side winch speed on a number of values such as the cutting power, side
winch force (amps), the concentration and the velocity of the mixture.
Older cutter suction dredgers sometimes have side winches that are combined with the ladder
winch to form one central winch, thus three drums and one drive. The paying out of the side
winch then takes place by freeing it from the drive shaft. Braking is then entirely mechanical.
It will be clear that in this case the ladder winch and the side winch cannot be operated
independently of each other, which is necessary when dredging slopes.

Other winces
If the dredger is equipped with anchor booms, it needs anchor winches and buoy line
winches. Depending on the spud hoist system there may also be spud winches and if the cutter
suction dredger must be able to work on a Christmas tree, stern winches and perhaps also a
bow winch will be needed. All these winches may be found in either electric or hydraulic
form.
3.6.11. Hoisting equipment
On board cutter suction dredgers cranes are necessary to lift heavy parts such as pump houses,
impellers and cutter heads. On large dredgers they can often travel over the length of the
pontoon.

Figure 3. 88 Mobile and fixed cranes

3.6.12. Auxiliary equipment


Cutter suction dredgers require the following auxiliary equipment:
A flatboat to move the dredger. By this it is understood the towing of the dredger from
dredging point to dredging point.
A work barge with a crane to carry supplies to the dredger. This can also be used to move
anchors if there are no anchor booms and to set out or move parts of a floating pipeline. It
may also be used to change the cutter head.
Some cutter suction dredgers even have a special cutter head pontoon. The cutter head
rests on this support. The pontoon sails under the raised ladder. (There are also special
cutter suction dredgers equipped with cutter manipulators with which the cutter can be
removed from the shaft in an easy way and placed on deck, after which a new cutter head
can be fitted.)

Page 56 of 79

Chapter 3: Cutter Suction Dredger

3.7. The dredging process


When dredging with the cutter suction dredger the three main phases of excavation, transport
and disposal can be distinguished too, however in this chapter only the excavation will be
considered.
In the process of excavation by cutter suction dredgers an important part is played by the
breach-forming characteristics of the soil to be dredged. In good breach-forming soil, which
will be defined later, the flow of soil to the underside of the breach is so good that little or no
further cutting is required. With soil that does not breach easily, the cutter head must cut the
entire face of the bank. This takes more time and thus the production rate will be lower.
In addition to the type of soil and its properties, it appears that the cutter production also
depends on a number of the ships characteristics such as the cutting power, the swing speed
and swing force, the spud system, and the position of the anchors during the cutting process.
The boundary conditions set by the work, such as the cutting pattern, possible slopes that
must be dredged, hydraulic pipeline transport distances, weather conditions and shipping
movements also have a big influence on the production.

3.7.1. The spillage


In both breach-forming and non-breach-forming
soil, spillage plays an important role. Spillage is
defined as the material in the dredging area that
comes to rest above the cutting area of the cutter
head. In other words spillage is the material that is
not taken up by the suction mouth.. (Figure 3. 89)

Lowest cutting level


Spillage

Figure 3. 89
There are two reasons why such material is not recovered by the dredger
1.
The method of working is such that not all
the material comes into contact with the
cutter head and thus it cannot be taken up.
Such a situation arises when the thickness
of the material that the cutter head removes
with one cut is greater than the diameter of
the cutter head. The material which lies
above the cutter head falls behind it and
thus cannot be taken up. (Figure 3. 90).
This phenomenon occurs mainly in
cohesive soils such as clay and in rock.

Figure 3. 90

Page 57 of 79

spillage

Wb3408b

Designing Dredging Equipment

2.
All the dredged ground is not taken up.
The reason for this is more complex.
Owing to its shape a cutter head has some
pumping power. It pumps water in an axial
direction to the rear. When the dredge
pump is out of action the water taken in by
the cutter head leaves the pump close to
the ring. As in the case of dredge pumps,
the size of the flow that is sucked in by the
cutter head is proportional to the revolution
speed of the cutter head.

Figure 3. 91
If the dredge pump is also running, the amount of water that leaves the cutter head close to the
ring is reduced. In principle it is possible to use such a pump flow rate that no outflow takes
place near the ring.
It appears that the percentage of the material cut by the cutter head that is taken up is linearly
dependent on the relation::

Production=1-Spillage=

Q pump
Pump capacity
v
= F z =
Cutterhead capacity
RR3cutter
R

The value of the angle depends on the direction of rotation of the cutter head, swing
direction and on the material to be dredged. For sand with a d50 < 500, can be taken as
0.4. For soils such as clay and rock the process is much more complicated because the
interaction of the separate soil particles with the cutter head play an important role. As stated
in chapter 3.2.2.2. may be a factor 3 higher in that case. Often in this type of case a constant
spillage factor of 0.3 - 0.4 is used.
As mentioned earlier, the spillage also depends on the work method.
When breach-forming soil (Figure 3. 92)
that forms an angle of slope with the
horizontal is cut by a cutter head, the
spillage depends only on the above
I
mentioned relation of the velocity as long
as the underside of the slope passes
through the cutter and area I equals area II

The maximum cutter head filling by an


+
unchanging spillage factor is obtained if
II
the cutter head is at right angles to the

slope. That is when + = 900, in which


For no addtional spillage Arae I = Area II
is the angle that the cutter ladder makes
with the horizontal.
Figure 3. 92
If the underside of the slope runs behind the cutter ring the material will not be cut but will be
transported further from the cutter head by the action of the pump. Moreover there is now a
good chance that that part have to be shifted by the ladder. See chapter 3.2.2.2 minimum
dredging depth.
The further the underside of the ladder comes behind the slope, the greater will be the chance
of a dragging ladder. On the other hand the filling of the cutter head is better.
Page 58 of 79

Chapter 3: Cutter Suction Dredger

Whether the underside of the slope passes through the cutter ring depends on the breach
forming behaviour of the sand, the swing velocity and the size of the step of the cutter head.

3.7.2. The production in breach-forming soils


The breach-forming characteristic of a slope depends on the permeability, thus the grain size
and pore volume of the sand layer.
If a suction pipe is quickly lowered vertically
Suction Tube
into the sand, a pit with almost vertical sides
240
is formed. The dimension of the pit increases
210
Slope
180
with time because the sand grains and
150
Vz
fragments of sand slide from the slope and
120
0
100
flow to the suction pipe. The bank of the
80
15
60
slope moves away from the suction mouth at
50
20
an almost constant velocity. The velocity is
40
30
also called the bank velocity Vwal. This Vwal
is roughly 30 * the permeability.
Time in seconds
B

Suction Velocity Vz = 2.5 m/s

Figure 3. 93

In the lecture notes lecture of Wb3413, part the Breaching Process the following theoretical
k k w 1 n 1
value for Vwal is derived: v wal =
which leads to the above-mentioned
n w
n tan
value of vwal30k .
The angle of slope in
front of the suction pipe
follows directly from the
relation between the bank
velocity Vw and the
velocity Vh at which the
suction pipe moves
forward (Figure 3. 94.).
tan

v h = v w 1
tan

vh

vw

Figure 3. 94
(3.12)
From this relation it follows that is equal to 90 when Vh = Vw.
The maximum angle of slope , the angle at which no more soil runs down to the suction
mouth, is for small breach heights the angle of internal friction. In most cases however, and
certainly with deep extraction pits, this angle is smaller. With bank heights of 15 m or more,
angles of slope of 1:10 to 1:20 occur. The erosion of the sand flowing over the slope causes
these.

Page 59 of 79

Wb3408b

Designing Dredging Equipment

When dredging good breach-forming soil, with a permeability 1*10-4 en = 10 at such a


depth that the axis of the cutter head makes an angle of 30 with the horizontal, the maximum
cutter head filling = 60.
The maximum progress of the dredger is then:
tan 10 o
4
m/s
v h = 30 10 4 1
o = 27 0
60
tan

The breach production is: Pb = v h B H [m3/s]


In which:
B = width of the cut
H = height of the face

[m]
[m]

The bank production for a width of 80 m and a face height of 5 m is now:

Qb =27*10-4 *80*5=1.08

[m3/s]

For an average cutter head radius of 1 m, a cutter head speed of 30 revolutions per minute and
a suction velocity of 4 m/s in an 800 mm suction pipe, the percentage that can be taken up is:
v
4
[-]
Pf = z = 0.4 = 0.51
R

The suction production is therefore:

Qs =0.511.08=0.55

[m3/s]

The spillage is thus 49 % of the face height, that is 2.45 m.


If the revolution of the cutter head is reduced from 30 to 15 revolutions per minute because no
cutting process develops in breach forming soil, then :
Pf = 0.4

4
= 1.02
0.5

[m3/s]

Because there is always some loss, for example due to the variation in the permeability of the
sand layer, Qs is given an upper threshold Pf = 0.9
The suction production is then:

Qs =0.91.08=0.97

[m3/s]

The spillage is now only 45 cm.


In breach-forming soil the ladder is almost at maximum depth and only swings from port to
starboard and back.

If a specified depth must be dredged it is always necessary to make a clean-up sweep: a final
swing, which removes all irregularities.
The question that now arises is how quickly must the cutter head swing in order to remove
this material..
If the area of the cutter contour is assumed to be Ac = 3 m2 , the cutter head must move at a
swing velocity of:

vt =

Q b 1.08
=
=0.36 m/s = 21.6 m/min
Ac
3

Whether or not the side winches are able to deliver this velocity in one way or another must
be ascertained. (see chapter 3.2.2.3)
Page 60 of 79

Chapter 3: Cutter Suction Dredger

The area Ac that the cutter head cuts while swinging across the face also determines the step
size that the dredger must make in the corners. After all the face production must be equal to
the cutting production, thus:
A
A c v t = H S v t S = c [m/s]
H
vt = translation velocity of the cutter in [m/s]
The average production reached during a full dredging cycle, that is the time between two
movements of the spuds, is in fact lower. This is because stepping, moving the spuds and, if
necessary, raising the ladder, all take time. These factors are entirely dependent on the spud
system and the time needed to perform the various procedures.

3.7.3. The production by non-breach forming soils


If the soil forms an inadequate breach or does not breach at all, as is the case with cohesive
soils such as clay and rock, and to a lesser degree fine sand, the cutter head must do what it is
designed for, that is cut the soil loose.

Depending on the
type of soil, the
spud system, the
suction depth and
the insight of the
dredge master, the
breach may be cut
in various ways.
Figure 3. 95 gives
an example for a
cutter
suction
dredger with fixed
spuds.

Dredging in cohesive soil

Swing number

5+6

11+12

10

17+18

13

14

79

15

21

16

22

23+24

29+30

28

Page 61 of 79

20

27

cut 4

cleaning up

Figure 3. 95

25

19

26

cut 3

cut 2

cut 1

Wb3408b

Designing Dredging Equipment

If the dredger has a


spud
carriage
the
variety of ways in
which the breach can be
cut is even greater
(Figure 3. 96). This
pattern is used when the
cut is to be made to the
desired depth in a single
cut. The numbering
gives the order of
cutting.

Dredging in cohesive soil

Swing number

11

16

17

22

21

13

18

19

10

cut 2

cut 3

15

20

cut 4

cleaning up

24

23

14

12

cut 1

Figure 3. 96
Dredging in cohesive soil

If the breach rises above


the water level, in order
to prevent a spillage
problem. The pattern
shown in Figure 3.97 or
Figure 3.98 must be
used.

19

12

Swing number

27

cut 1

cut 2
2

13

20

14

21

29

28
cut 3

34

cleaning up

18
+

26
+

25
+

33

32
24
+

+
31

17
+

+
23

35

11
+

+
16

5+
10

4+

9+
15

+
22

+
30

Figure 3. 97
Dredging in cohesive soil

14+15

1+2

16

24

33

10

17

25

34

cut 3

5+
11
+
18
+
26
6+
+
12
35
+
19
+
27
+
36
13
+
20
+
28
+
37
21
+
29
+
38

35

30
+
39

Swing number

Figure 3. 98

Page 62 of 79

22+23 31+32 cut 1

cleaning up

cut 2

Chapter 3: Cutter Suction Dredger


3.7.4. Specific energy
The number of layers over which the breach is cut, the step size and the swing velocity are
closely related to the specific energy that is required to cut the soil. The energy consumption
per unit of production is called the specific energy and is thus, by definition, the energy that is
needed to cut loose one m3 of soil. Although it is often thought that the specific energy is
independent of the cutting process, it is certainly not, since the finer the material that must be
cut, the greater the energy consumption.
The cutting method also exerts a big influence. When cutting rock, the specific energy
increases strongly as the teeth are worn away. Furthermore the influences of the radius and
the revolutions of the cutter head are limited, so no account can be taken of the possible
dependence of cutting force on the velocity or of the permissible torque.

To obtain some insight into this subject, the specific energy is calculated from a general
cutting theory or a straight cutting edge on a rotating cutter head.
With a linear movement the cutting force of a straight cutting edge can be characterised by
the following power equation:

Fc =cd vt W
in which:
c

d
Vt
W

=
=
=

[N]

a constant that is dependent on the soil type


and on the boundary conditions such as water
depth, cutting edge angle, cutting edge height,
etc
the cutting depth or slice thickness
the cutting velocity
the width of the cutting edge

[m]
[m/s]
[m]

The production of a straight cutting edge is:


Q = dVtW
[m3/s]
Therefore the specific energy is:

Es =

Fc v t cd vt Wv t
=
=cd -1vt
Q
dv t W

[J/m3]

From this it follows that the specific energy is only constant if the cutting process is entirely
linear, thus when:

Fc =cdv t W
If this theory is applied to
cutting with a cutter the
chip thickness is:

2v t
d=
z
d=psin

sin

Tooth path

Ra
diu
sr

p
d

Tooth path

Figure 3. 99

= the angular velocity of the cutter head


z = the number of cutter arms
Page 63 of 79

[rad/s]
[-]

Wb3408b

Designing Dredging Equipment

vt = the swing velocity


[m/s]
= the angle between the cutter radius and the tooth path [radian]
2 v t
The maximum chip thickness is: d max =

z
Because the peripheral velocity of the cutter is equal to R, the cutting force of a cutter

2v t
is: Fc =c
z

sin ( R ) L

2v t

L is proportional to the step size S thus: Fc =c


sin ( R ) S
z

Moreover the cutting power is equal to: Pc =Fc v t =Fc R


'

2v t
Pc =c
z

+1

sin ( R ) S

'

With increasing step size the average radius of the cutter head increases; thus R = f ( S) = S .

2v t

+1
From this the cutting force can be reduced to: Pc =c
sin ( S ) S
z

'

The cutting production is: Qc =Sv t D


and thus the specific power:

2v t

+1
' 2v t
c
sin ( S ) S c
z
z

Es =
=
Sv t D
'

+1
sin ( S )

v t D

From this equation it follows directly that the specific cutting power is constant only under
very exceptional conditions. These conditions are:

A cylindrical cutter head = 0

The cutting force must increase linearly with increasing chip thickness.

This gives

The average chip thickness must be linear with the layer thickness. Thus

v t
vt

= v t 1 is constant

[sin]

is

constant

The cutting force must be independent if the constant = 0

Then:
Es = c''

R
z

From this it follows that the specific cutting energy is always dependent on the type of cutter
head.

Page 64 of 79

Chapter 3: Cutter Suction Dredger


Because there are often big variations in the types and strength of the soil and many factors
that cannot be determined in advance play a part in the cutting process, the specific energy
appears to be a good parameter for estimating the production of cutter suction dredgers.

3.7.5. The cutting production


The specific energy required for a particular type of soil can be estimated with the aid of
existing cutting theories or from production estimates from previous work with the same type
of soil.
If the specific energy Esp, is known, it follows from the definition of the cutting process:
w Nc
Pc =
Es

in which Nc is the cutter power.


The value w, a work coefficient, gives an indication of the average maximum percentage of
the installed cutting power that can be used. This value is dependent, not only on the type of
soil (relation between peak forces and average forces), but also on the man-machine relation.
The dredge master and the automated cutter control regulate the cutting speed on the basis of
the amperage (torque) of the cutter head engine.
Types of soil the hardness or strength of which vary greatly from place to place will give a
torque or amperage signal that varies greatly over time in which Nc is the cutter power (Figure
3. 100.)

TORQUE SIGNAL

mean value
150

torque [%]

120

90

60

30

0
0

10

15

20

25

30

35

40

45

50

Time [s]

Figure 3. 100.

This may quickly lead to overloading of the cutter head engine, with the result that, for
example, for the torque-revolution characteristic shown below, the cutter head will stall at a
torque of 150% (Figure 3.101)

Page 65 of 79

Wb3408b

Designing Dredging Equipment

Speed [%]

100

100

Torque [%]

150

Figure 3. 101 Torque speed Characteristics of an electrical drive

If this occurs frequently the dredge master will reduce the swing speed of the dredger to
ensure that the peak loads do not cause the cutter to cease turning.
It will be clear that the type of drive plays a big part in this. An electric drive can take up the
variation in torque better than a hydraulic drive. (See chapter 3.4.2.)
The skill of the dredge master also plays a part. Dependence on his skill can be reduced to
some extent by the use of an automated cutter control. This regulates the swing velocity, for
example in relation to the torque of the cutter head. In many cases such an automated control
system can react more quickly than the dredge master can, certainly at times when his watch
is almost over.
It will also be clear that only rough estimates can be given for such a factor as the work
coefficient.
For rock :

w = 0.5 - 0.65

For sand :

w = 0.65 - 0.8

For clay :

w = 0.8 - 0.9

An automated cutter control increases these values by 10% to 20%.


With the information given above, the cutting process can be found and also the warping
speed of the cutter head. Because:

Qc =Sv t D
With
Pc =
D =
S =
Vh =

[m3/s]

cutting production
layer thickness
step size
swing velocity

Page 66 of 79

[m3/s]
[m]
[m]
[m/s]

Chapter 3: Cutter Suction Dredger

Dredging in cohesive soil

Dredging in cohesive soil


3

cut 1

10

12

13

18

19

14

21

22

17

20

14

79

15

21

16

22

23+24

29+30

20

26

cut 2

27

cut 3

cut 4

cleaning up

24

23

cut 3

15

3
16

25

19

cut 2

Swing number
11

13

1
6

Swing number

5+6

10

11+12

17+18

28

cut 4

cleaning up

Figure 3. 102

In this case the spillage can be calculated as follows:


Assume that the spillage is M % of the cut surface. (M can be determined in the same way as
in breach forming soil.). If the thickness of the layer and the step do not greatly exceed the
dimensions of the cutter head, the spillage is M % of the layer thickness. Thus:
-

for layer 1:

for layer 2:

for layer k:

Z1 = M D

D = layer thickness

1 M k

Z k = M + M + M +.......... M

After simplification it follows that:

Zk = M D

)
)D

Z 2 = M ( D + M D )1 = M + M 2 D

M 1 Mk

1 M
k (1 M )
The part taken up is thus:
M 1 Mk
Sk = H Z k = H1
k (1 M )

)H

Clearly, when the thickness of the


layer or the size of the step exceeds
the dimensions of the cutter head
the part of the material that has no
chance of entering the cutter head
must immediately be considered as
spillage. Figure 3. 103 shows a
breach, which projects above
water.

Figure 3. 103

Because the suction mouth must remain sufficiently under water to prevent the taking in of
air, the dredge master must make the first cut thicker than the diameter of the cutter head.
Page 67 of 79

cut 1

Wb3408b

Designing Dredging Equipment

The direction in which the bank is


stripped now affects the spillage,
although not in the cut, which is
being dredged, but in the cut that has
already been dredged. If the first cut
has been made with a reverse turning
cutter working towards the already
dredged cut, because of the failure to
raise the necessary reaction force, it
is possible that at the end of the cut,
some of the material from this new
cut is pushed into the already
dredged area.

Swing direction

Former cut

Over cutting mode

Figure 3. 104

The result is that a ridge of soil is formed on the boundary between the cuts. In such a case it
is better to make the uppermost cut in the same direction as the rotation of the cutter head.
If the spillage is known the average dredging production over one spud cycles is:

Qs =

Sk W L
ts + ta

in which:
Sk = the thickness of the layer which has been taken up
W = the width of the cut
L = effective advance of the spud carriage
ts = net cutting time during a spud cycle
ta =the sum of the times during the spud cycle when no cutting
occurs, such as ladder raising, stepping, spud moving ,etc.

[m3/s]

[m]
[m]
[m]
[s]
[s]

In non-breach forming soil, if a specified depth has to be delivered a clean-up swing must also
be made. The production of this swing is calculated separately. The cutting energy that is
required in this layer can only be determined from the part that has not been cut. It is therefore
possible that because of a thin layer, the clean-up production is high.

3.8. Enclosures
3.8.1. The relation between swing speed and side winch speed.
The swing speed of the cutter head must
not be confused with the side wire speed.
The latter is the speed with which the side
wire is hauled in and which controls the
swing velocity. Although there is a clear
relation between these two velocities, they
are certainly not equal. The position of the
anchors in relation to the cut plays an
important part in this. By the correct
positioning of the anchors it is possible to
reach a high swing velocity with a small
side winch velocity.

Page 68 of 79

Chapter 3: Cutter Suction Dredger


Figure 3. 105

In Figure 3.105 the distance between the work spud and the sheaves of the side winch on the
ladder is equal to L and the distance between the sheaves and the anchor is equal to S. If the
angle between the centreline of the cut and the line linking the spud-side winch sheaves is
equal to , then:

x = l cos
y = l sin
z = k x = k l cos
t = b y = b l sin
s = z2 + t2 =

( k l cos ) + ( b l sin )
2

ds ds d 2 ( k l cos ) ( l sin ) + ( b l sin ) ( l cos ) 


=
=
2
2
dt d dt
2 ( k l cos ) + ( b l sin )

Since l is the swing velocity, the previous equation can also be written:

Opmerking [T5]: Page: 65


Hier de afleiding toevoegen.

ds
k sin l sin cos b cos + l sin cos
= l
2
dt
k 2 k l cos + l 2 cos 2 + b 2 2 b l sin + l 2 sin 2

of
ds
k
b
sin cos
dt =
l
l
2
2
l
k
b

cos
sin

l
l

Since the side winch force do not act on the ladder at the same distance from the spud as the
cutter head, the swingspeed have to be corrected according:

vs l
=
vc lc

3.8.2. The side winch force and power


The swing force Fh takes effect at right angles to the centreline of the dredger, thus in the
direction of the movement of the cutter head. The chance that the anchor is positioned in
exactly the same direction as this track of the cutter head is valid for only one point. If the
angle made by the tangent at one point of the track of the cutter head with the line joining this
point to the anchor position is , the required side winch power is Fz = Fh/cos ().

Cos() can also be expressed in the units given in.

sin

cos = cos + arctan l


k cos
2
l

The side winch force is thus:


Page 69 of 79

(3.45)

Wb3408b

Fz =

Designing Dredging Equipment


Fh
cos

Fh R c
M c cos

Fh R c

sin

M c cos + arctan l
k cos
2
l

(3.46)

If the diameter of the side winch drum is equal to Dw, the required side winch torque is:

Dw
D
Fh Rc w
2 =
2 =
Mw =
M c cos
cos
Fh

Fh Rc

Dw
2

l sin
M c cos + arctan

k
2
cos
l

Both the side winch velocity and the side winch torque are now known as functions of the
position of the anchors and the position of the cutter head in the cut. Neither the necessary
side winch velocity, nor the necessary torque may exceed the maximum value of the side
winch characteristic. If this does happen, the side winch velocity must be reduced until this
condition is met.
Because during the progress of the dredger the positions of the anchors in relation to the track
of the cutter head must be continually changed, if the side winch velocity or the side winch
force is the limiting factor for the dredging process, the dredge master must continuously
adjust the side winch velocity until the point is reached where it seems wiser to move the
anchors.
From the above it will be clear that the further away the anchors are positioned from the ship,
the longer the force will be effective, thus the anchors will have to be moved less often. On
the other hand the longer the side wires, the weaker the system will be. This is a disadvantage
when dredging hard soil such as rock.
From the relation between the swing velocity vh or the angular velocity , together with
required side winch electric current, dredge master can see whether or not the anchor is
holding or dragging.
3.8.3. The shape and cutting geometry of cutter heads
Because the cutting process plays an important role in excavation, this section will give more
detailed consideration to the shape and cutting geometry of cutter heads.
Definitions:

The base plane is the plane that passes through the underside of the cutter ring.
The cutting point P may be a point on a cutting edge of a plain edge, the cutting point of a
serrated edge or the edge or point of a tooth. The position of the cutting point determined by
the cylinder coordinates Rp, Hp, and p.
Here:
Rp = the radius from the cutting point to the cutter axis.
Hp = the distance between the cutting point and the base plane.
p = the angle between the projection of the cutting point onto the base plane and the
cutting point (Rp,0,0)

Page 70 of 79

Chapter 3: Cutter Suction Dredger


The cutting edge of a cutter blade is the smooth curve passing through the cutting points.
The contour or outline of the cutter head is the section made by the cutting edge (the contour
plane) though a plane perpendicular to the axis of the cutter head.
The contour tangent touches point P on the contour.
The contour angle is defined as the angle between the line in the contour plane passing
through P at right angles to the contour tangent and the line through P parallel to the base
plane.
The cutting plane is at right angles to the contour plane and the contour tangent..
In the dredging world both Florida and Esco cutters are used. The positions of the tooth points
of both systems are determined by using cylinder coordinates.
The direction of the tooth axis given by Esco differs from that given by Florida.
Tooth axis direction according to ESCO
ESCO gives the direction of the tooth axis in two ways:

1.

By giving the tooth point and the tooth base of the tooth axis in cylinder coordinates.

2e. By giving the tooth point and two angles of the tooth axis.
These angles are defined as follows:

The pitch out angle . This is the angle between the tooth axis projection in the plane
parallel to the base plane and the tangent on the circle passing through the tooth point
projection.

The pitch up angle this is the angle between the tooth axis and its projection in the
plane parallel to the base plane..
Thus in Figure 3.77.:
P' B'
= arctan

BB'

en
PP'
= arctan

P' B'
In addition ESCO give the roll angle (rho) of a tooth. This is the position of a tooth in
relation to the tooth axis.
The roll angle is the angle between the edge of a chisel (flared or chisel leading edge) and
the line parallel to the cutter axis as seen along the tooth axis. This angle is equal to the
centreline of the locking pin and the line parallel to the base plane seen along the tooth axis.
Tooth axis direction according to FLORIDA.
FLORIDA gives the tooth axis by the giving coordinates of the tooth point with two angles.
FLORIDA defines these angles as follows:

The tooth axis angle (tooth angle).This is the angle between the tooth axis and the
tangent on the circle passing through the tooth pint. This is the tangent to the line of the
movement during rotation.

Page 71 of 79

Wb3408b

Designing Dredging Equipment

The contour angle t (Kappa=Profile angle) of the tooth. This is the angle between the
tooth axis projection in the contour plane and the line parallel to the base plane (P'B').

FLORIDA has a fixed roll angle (rho) because the cutting edge or blade edge of the
tooth always lies in the contour plane. This makes the roll angle a function of the tooth
axis angle and the contour angle t

Florida = arctan[ tan t cos ]


When working, in most cases a piece of auxiliary equipment, the so-called ALFE is used in
order to ensure that adapters are correctly positioned on the cutter head arm when these have
to be replaced owing to breakage or loss (Figure 3. 106.). The plane of the ALFE is thus a
contour plane.
In that case the FLORIDA instruction is more simple than the ESCO. With ESCO cutter
heads the angles must be recalculated to the FLORIDA instruction.

Page 72 of 79

Chapter 3: Cutter Suction Dredger

Figure 3. 106

Tooth axis angle

= arccos( cos cos )

Contour angle t
tan
t = arctan

sin
Roll angle
Florida = Esco mal

Page 73 of 79

Wb3408b

Designing Dredging Equipment

Here mal is the angle over which the adapter must be turned on its axis to get the cutting edge
in the contour plane, thus against the ALFE. _mal may be positive or negative.
3.8.4. Cutting by teeth or chisels

CUTTERAXIS

Hp

(Rp ,Hp, p)
CUTTING EDGE

R
p

(Rp,0,0)

OUTLINE OF
THE CUTTER

Hc
CUTTING EDGE

(HELIX ANGLE)
BASE

Figure 3. 107

For the definitions of the various angles see Figure 3.107.


-

Cutting edge/rake angle

Tooth axis angle

Clearance angle

Wedge angle
In addition to a clearance angle on the rear of a chisel there are also side clearance angles.

Page 74 of 79

Chapter 3: Cutter Suction Dredger


3.8.5. Conditions for cutting clearance
The front and rear edges of the arms of cutter heads, edges, teeth and chisels follow different
tacks during the cutting process (Figure 3.108). The most unfavourable point for the cutting
clearance is the point where the velocity vector s of both the front and rear edges are parallel.
In that case there is a maximum and minimum distance between the two paths. This happens
when the velocity component in the X-direction is vx=0.

Figure 3. 108

The path of a point on a cutter head can be described by the two following equations in
parameter form (Figure 3.81.):
x p = v h t + R p cos t
y t = R p sin t
= t
Here:
Xp, Yp = the coordinates of the point P with regard to the cutter head axis.
vh
= the swing velocity of the cutter head
= the angular velocity of the cutter head
Rp
= the radius of the cutter head
t
= the the time of passagede

The direction of the velocity is the tangent to the path:


R p cos t
dy dy dt
=

=
dx dt dx v h R p sin t
The velocity in the x-direction is zero when the deriviative is infinite, thus as:
v h R p sin t = 0
Further:
y = R p sin t
so that:
vh y = 0

y=

vh

and the associated angle :

Page 75 of 79

Wb3408b

Designing Dredging Equipment

p = arcsin

vh
Rp

Now when:
l
= distance between the front of the tooth and the rear of the arm
Rv = the radius of the tooth point and Ra, the radius of the rear of the arm .
then:
v = arcsin

vh
Rv

and
a = arcsin

vh
Ra

Furthermore if l is the distance between the front of the tooth and the rear of the arm, it
follows from Figure 3.80 with =0 that the angle between the two pointy mentioned is equal
to:

( R v cos 0 R a ) 2 + ( R v sin 0 ) 2
2
2
l 2 = ( R v cos 0 R a ) + ( R v sin 0 )
l=

l 2 = R 2v + R 2a 2 R v R a cos 0
R 2 + R 2 l2
a
0 = arccos v

2 Rv Ra
The tooth and arm now run clear if the horizontal distance between the paths at the distance y
is greater than the distance the cutter head moves as a result of the following the swing
velocity round the 0 + a v .
Thus when
v
R v cos v R a cos a ( 0 + a v ) h

Example:
Rv = 1.50 m, Ra=1.45 m, l=0.7 m vh=0.3 m/s en
= (n=30 t/min)
then:
R v cos v = 1497
.

R a cos a = 1447
.
v = 0.064
a = 0.066
0 = 0.478
y = 0.095
The maximum side winch velocity may then be:
vh

( R v cos v R a cos a )

thus v h 0.33 m/s


Page 76 of 79

0 + a v

Chapter 3: Cutter Suction Dredger

It will be clear that when designing cutter heads this exercise must be carried out for a number
of points on the cutter head, since cutter arm length and radius are a function of the height of
the cutter head, measured from the ring.
This also determines the maximum thickness of the cut. When the rear of the arm touches the
path of the front of the tooth, the maximum cut thickness is equal to:
d max =

60 ( v h ) max

nz
in which z is the number of arms.
From the example it thus follows that:
d max =

60 ( v h ) max
nz

60 0.33
30 6

= 011m
.

A ch te rzijd e arm

T an d p un t

1.5
1
0.5
0
-1.5

-1

-0.5

-0.5

0.5

1.5

-1
-1.5
Xp
Figure 3. 109

Finally the same example, but now with n=10 t/m and Ra=1.36 m. dmax=0.30 m and
vmax=0.30 m/s.
The path of the two points is shown in Figure 3. 109..
If parts of the tooth or arm project through the line passing between the tooth point and the
rear of the arm, it is necessary to carry out a check for more points.
The effect of warping on the clearance angles
The direction of the movement of the tooth point is (see Figure 3. 110):
R p cos t
R p cos
dy
=
=
dx baan v h R p sin t v h R p sin

Page 77 of 79

Wb3408b

Designing Dredging Equipment

VH
VT

nC

corr.

RP
VH
V
A

corr.

VH
VT + VH

Figure 3. 110

The rear plane of the tooth makes an angle A with the circumference of the cutter head, thus
with the tangent on the circle:
R p cos t
R p cos
dy
1
=
=
=

dx circel R p sin t R p sin tan
The clearance angle between the path of the tooth and the back of the tooth thus varies with
the rotation.
The difference between the two tangents is the varying clearance angle:
R p cos
R p cos
1
corr = arctan
arctan

= arctan
tan
v h sin
v h sin 2

For Rp = 1.0m,

= ,

=0 and vh = 0.3 m/s it follows that:


corr = arctan
= 0.0095 rad=-5.27'
0.3 0 2
In other words, the cutting angle is 5 27' smaller.

Page 78 of 79

Chapter 3: Cutter Suction Dredger

3.9. References
1. calculation of the cutting forces when cutting in fully saturated sand, S.A. Miedema,
Thesis TU-Delft, 1987 (in Dutch)
2. Coastal and Deep Ocean Dredging, John B. Herbich, Gulf Publisching Company,
Houston, Texas, U.S.A., 1975
3. Dredging and Dredging Equipment, R.J. de Heer and Rochmanhadi, Parts 1 and 2,
IHE, Delft, 1989
4. Dredging technology, lecture notes, G.L.M. van der Schrieck, TU-Delft, Civiele
techniek, 1996 (in Dutch)
5. Concept, design and construction of the World's first self elevating offshore heavy
cutter suction dredger: "Al Wassl Bay", D.A. Gaasterland, Proceedings 3e
International Symposium on Dredging Technology, BHRA 198?
6. Nassbaggertechnik, A. Welte, Institut fr Machinenwesen in Baubetrieb, Universitt
Fridericiana, Karlsruhe, 1993.
7. Proceedings of the CEDA Dredging Days, Europort 1980, CEDA, 1980
8. Technical aspects of large cutter suction dredgers, P.J. Koert, IHC Holland
9. Dredgers of the World, 3rd edition, Oilfield Publications Ltd (OPL). England, 2001
10. Various articles from Port & Dredging from IHC Holland
ARTICLE
Spudsystemen van cutterzuigers
Demonteerbare cutterzuiger/baggerwielzuiger SCORPIO
IHC Beaver cutterzuigers
Cutterzuiger NOORDZEE
Automatisering van cutterzuigers
Zelfvarende cutterzuiger van 27000 PK
LEONARDO DA VINCI: een nieuw record
Nieuwe serie IHC Beaver cutterzuigers
The IHC Beaver container dredger
Cutter suction dredger ABU AL ABYADH for NMDC
Sensative environmental cutter dredger for Samsung
Mighty MASHHOUR for Suez Canal
Dismountabe IHC Beaver dredgers
CD Al Mirfa
CD Kattouf

Page 79 of 79

P & D no
108
108
109
118
119
119
124
126
134
145
146
147
153
154
157

Chapter 4 Plain Suction Dredgers

4.

The Plain Suction Dredger ...........................................................................................2


4.1
General considerations ........................................................................................2
4.2
Areas of application ............................................................................................3
4.3
Types of plain suction dredgers ..........................................................................3
4.4
History.................................................................................................................5
4.5
Working method .................................................................................................6
4.6
The design ...........................................................................................................8
4.6.1
The production capacity.............................................................................9
4.6.2
The suction depth.......................................................................................10
4.6.3
The transport distance ................................................................................12
4.6.4
The dredging installation ...........................................................................12
4.6.4.1
Suction and discharge pipe diameter .....................................................12
4.6.5
The dredge pump .......................................................................................13
4.6.5.1
Pump types..........................................................................................13
4.6.5.2
The sand pump drives .........................................................................14
4.6.6
Jetpumps ....................................................................................................14
4.6.6.1
Pump type ...........................................................................................14
4.6.6.2
Jetpump drives. ...................................................................................17
4.7
General layout .....................................................................................................18
4.8
Technical construction ........................................................................................20
4.8.1
The hull ......................................................................................................20
4.8.2
The dredging equipment ............................................................................21
4.8.2.1
The suction mouth...............................................................................21
4.8.2.2
The suction pipe..................................................................................22
4.8.2.3
The sand pumps ..................................................................................23
4.8.2.4
The sandpump drives ..........................................................................25
4.8.2.5
The discharge pipeline ........................................................................25
4.8.2.6
Sprayers...............................................................................................25
4.8.2.7
Jet-pipeline and pump .........................................................................26
4.8.2.8
The winches ........................................................................................26
4.8.2.10
The bow winch .................................................................................27
4.8.2.11
The side winches ..............................................................................27
4.8.2.12
The stern winch ................................................................................27
4.8.2.13
The auxiliary winches.......................................................................27
4.8.2.14
The fairlead.......................................................................................27
4.9
The dredging process ..........................................................................................29
4.9.1
The production of the breach .....................................................................29
4.9.2
The production of the pumps .....................................................................32
4.9.3
The production of the barges .....................................................................33
4.10
The dustpan dredger............................................................................................34
4.11
References ...........................................................................................................35

Page 1 of 35

Wb w3408b

4.

Designing Dredging Equipment

The Plain Suction Dredger

Figure 4. 1 A plain Suction Dredger

4.1 General considerations


The characteristic of a plain suction dredger is that it is a stationary dredger, consisting of a
pontoon anchored by one or more wires and with at least one sand pump, that is connected to
a suction pipe. The discharge of the dredged material can take place via a pipeline or via a
barge-loading installation. The suction tube is positioned in a well in the bows of the pontoon
to which it is hinged. The other end of the suction pipe is suspended from a gantry or Aframe by the ladder hoist. The ladder hoist is connected to the ladder winch in order to
suspend the suction pipe at the desired depth. Excavation of material to dredge is by the
erosion of a jetstream and/or the suction flow of the dredge pump and the breaching process
(see lecture notes wb3413 the Braching process)During sand dredging the dredger is moved
slowly forwards by a set of winches. To increase the amount of sand flowing towards the
suction mouth, a water jet is often directed onto the breach/bank. In this case the jet-pipe is
often mounted above the suction pipe.

Figure 4. 2 Plan view of a PSD

Page 2 of 35

Chapter 4 Plain Suction Dredgers

4.2 Areas of application


Plain suction dredgers are only used to extract non-cohesive material. Moreover these
dredgers are less suitable for accurate work such as the making of specified profiles. Suction
dredgers are very suitable for the extraction of sand, certainly when this occurs in thick
layers. Suction dredgers can be seen in working in many sandpits.
If the dredger is equipped with an underwater pump, it is possible to dredge at depths
exceedin
g 80 m. Depending on the pumping capacity; it is possible to transport material over
considerable distances via hydraulic pipelines.
Because suction dredgers are often demountable they can also be used in excavation pits
which are not on navigable waterways. In general, suction dredgers are relatively light
vessels and, although anchored on wires, are usually unsuitable for dredging in open waters
(unless specially adapted).

4.3 Types of plain suction dredgers


Different type of plain suction dredgers can be distinguished.
1. The barge loading plain suction dredger
A dredger which loads the barges which lie alongside it by means of a spraying
system. This type is used when the transport distance is too long for hydraulic
transport to be economic (Error! Reference source not found.).

28

18
24

16

22

11

23

17

31

26

13

25 19

10
21

20

27

14
15

Figure 4. 3 Barge loading PSD

2. The reclamation dredger


This dredger pumps the sand ashore via a pipeline and, if necessary, further away to a
disposal site or treatment plant (Figure 4. 4).
.

Page 3 of 35

Wb w3408b

Designing Dredging Equipment

Figure 4. 4 Reclamation dredger

3. The deep suction dredger


The deep suction dredger. A dredger equipped with an underwater pump. It may take
the form of a barge loader or a reclamation dredger. (Figure 4. 5)

22
23
24

17/18

11

28

6
27

19/20

16

10

30

29

34

12

11

31

14

12

35
33

35

33

13/15

32

32

2
2

Figure 4. 5 The deep suction dredger

4. The dustpan dredger


A suction dredger with a relatively wide suction mouth. This dredger is suitable for
extracting sand at a reasonably high production rate with a low breach or bank
height. With regard to production the cutter suction dredger (Figure 4. 6) has
superseded this type.

Page 4 of 35

Chapter 4 Plain Suction Dredgers

Figure 4. 6 Dustpan dredger

In many cases these types can easily be transformed to another type. The barge loading
dredger shown in figure 4.2 can be transformed to a reclamation dredger by connecting a
booster just behind this dredger. The same might be possible with reclamation dredgers by
placing a sprayer pontoon after the dredger.

4.4 History
In 1851, more than a century after their invention, the first centrifugal pumps were used to
excavate sand with hopper dredgers. A few years later (1856) the first attempts were already
being made to transport the material onshore via pipelines. Ten years later this idea was
demonstrated in the Netherlands during the excavation of the North Sea Canal. (Figure 4.7)

Figure 4. 7 The wooden Hutton Dredger dredging the North Sea Canal

Meanwhile, in 1864, Freeman and Burt patented a flexible floating-pipeline.

Page 5 of 35

Wb w3408b

Designing Dredging Equipment

From this history it appears clear that the development of the suction dredger was closely
linked with the development of the dredge pump. Because at that time little power was
available to drive the dredge pump, the reclamation dredger was only used when the
distances to the disposal site were short. In the other cases barges were used or the dredger
was modified. As the sand pumps became able to withstand higher pressures, the transport
distances and pump capacities were increased.

4.5 Working method


The working method of the suction
dredger depends on both the
progressive collapsing of the
breach/bank and the loosening of the
sand near the suction mouth by
eddies created by the flow of water
caused by the sand pump (Figure 4.
8). The progressive collapse of the
breach/bank resulting from the
dislodgement of particles of soil or
of masses of soil as a result of
localised instabilities is termed
breaching.

Instabilities

H br

Sand-water mixture
(density current)
Suction tube
Vz

Figure 4. 8 Breaching

This process is essential for the production of a suction dredger and is entirely determined by
the soil mechanical properties of the slope, the most important factors being its permeability
to water and relative density.
When a suction dredger starts on a new work there is no dredge pit, slope or breach and the
angle between the suction pipe and the horizontal is usually very small. The sand that is
carried towards the suction pipe lies entirely within the area influenced by the water flowing
to the suction mouth. This process causes a small pit to develop in the soil.
The dredger is now drawn forwards a little
by means of the bow winch and the suction
pipe is set deeper, after which the process is
repeated. As the small pit becomes deeper
and the angle of the suction tube becomes
steeper (more effective for the swirling up
and transporting of the sand) the production
increases. (Figure 4. 9) This process is
continued until the suction mouth is deep
enough or until the production is so high
that the pump can no longer cope with a
further increase. This slow forward
movement with the dredger, with
simultaneous lowering of the suction pipe is
termed breaking in or commencing.

Figure 4. 9 Breaking in

Page 6 of 35

Chapter 4 Plain Suction Dredgers

The time that is needed to reach a state of equilibrium thus depends on the previously
mentioned soil mechanical properties, the height of the slope and the pump capacity of the
dredger.
When a state of equilibrium has been reached it is the task of the dredge master to maintain
this situation by letting the dredger follow the breach/bank, by regularly hauling the dredger
forwards and by continuing to lower the suction pipe for as long as this remains possible.
If the movement of the dredger is too slow, a less steep slope forms and the production is
reduced.
If, on the other hand, the forward movement is faster than the transport of the sand, the angle
of slope will increase and there is an increasing chance that large scale shearing will occur.
The sand concentration may then become so high that the pump cannot cope with it and the
mixture ceases to flow. The shearing can be so great that even the suction pipe becomes
fast/firmly embedded and, if it cannot be pulled free, another dredger must be used to free it
by using suction or must cut it free.
The dredging pattern that is made with a suction dredger generally appears like that shown in
Figure 4. 10. As long as it lies within the dredging area, the length of the cut is determined
by the positions of the anchors. The anchors are usually placed in such a way that more cuts
can be made beside each other from the same position. In addition to the length of the anchor
wires, this possibility also depends on the width over which the sand is being excavated.
This, in turn, depends on the shear characteristics of the sand layers.

Figure 4. 10 Dredge pattern of a PSD

For suction dredgers equipped with an underwater pump the excavation depth no longer
determines the production. This also makes it possible to exploit the dredging area in the
vertical sense. In other words, production can be maintained by continuing to lower the
suction pipe until the maximum suction depth has been reached. If the production falls below
an economic minimum, the pit is abandoned and dredging recommences to pit diameter
away from it. It will be clear that this dredging method produces a pockmarked excavation
area and that considerable amount of sand that cannot be economically excavated remain
behind in the dredging area. This is a situation that the managers of the dredging sites prefer
not to see.

Page 7 of 35

Wb w3408b

Designing Dredging Equipment

This method of dredging does provide the possibility to obtain sand from directly beneath a
clay layer, but it must be realised that the removal of the sand will cause the clay to lose its
stability. In the most favourable case the clay will fall onto the slope in fragments that will be
taken up with the sand. If the clay falls in large pieces there is a good chance that these will
become fast and block the suction pipe, with all the disadvantages that this can bring. It is
difficult for the water needed for mixture formation to flow, especially in the beginning phase
when the clay layer has not yet been penetrated.

Figure 4. 11 PSD with suction pipe of 2 sections

Water must be brought to the suction pipe via the jet pipe. For the above described
excavation method the suction pipe is made in two parts, (Figure 4.11) the lowest section
being hinged onto the upper section so that the lowest part is always first suspended almost
vertically. With such a suction pipe, moments that occur during horizontal movements can be
taken up only to a small extent.

4.6 The design


When designing suction dredgers the following parameters are important:

Production capacity
Suction depth
Transport distance
Type of soil

Because suction dredgers are only suitable for the dredging of non-cohesive material, the last
parameter plays an important role only in the determination of the diameters of the suction
pipe and hydraulic pipeline and the required sand pump capacity.

Page 8 of 35

Chapter 4 Plain Suction Dredgers

4.6.1 The production capacity


As in other dredgers, the market forces in relation to the sites where the dredger can be used
determine the production capacity. As mentioned earlier, the plain suction dredger is much
used in the extraction of sand for landfill sites and for the concrete industry.
For this too, it is important to know the production capacity per week or per hour. In the
Netherlands, to a limited extent, the labour agreements between the trade unions and industry
permit a working week of 168 hrs, thus an entirely continuous operation.
Often this is restricted to only four nine-hour days (36 hrs). The percentage of hours during
which effective dredging can take place, however, is not equal. With a 36-hr week, major
repairs are often carried out during overtime. When using barge transport, for example, the
percentage of downtime resulting from the absence of barges is lower during a 36-hr week
than during a continuous working week, since part of the downtime is made up when the
dredger has stopped work at the end of the day.
If, during a 168 hr working week, the number of effective working hours is 0.75*168=126
and during a 36 hr working week the effective hours are 0.86*36 = 30.6, the production ratio
is 126/36.6 = 4.1 instead of 168/36 = 4.7.
For the design of the dredging installation, and thus for the vessel also, the production per
hour is more important than the daily, weekly or monthly production. In many cases, in order
to prevent overloading of the drives, even shorter time intervals are considered. If the
production capacity is known, this requirement can be translated into:
1. A sand flow rate
2. A sand concentration
Since: Q = Qmixture

Cvd
1 n

(4.2)

with
Symbol
Q
Qmixture
Cvd
n

=
=
=
=

describtion
Production
Flow rate
Delivered concentration
Porosity

dimension
[m3/s]
[m3/s]
[-]
[-]

The anticipated average concentration depends on the behaviour of the soil in the
breach/bank (see lecture notes Dredging Processes). The maximum suction concentration is
determined on the basis of the types of soil and the insight of the designer.
The maximum average concentration that can be transported by a pipeline depends on the
ratio maximum grain diameter/pipe diameter and the length of the pipeline. In long pipelines
aggregation (increased concentration) may occur as a result of density variations during
dredging (Matousek, 1995).
As rule of thumb, a maximum average density of 1500 kg/m3 (Cvd = 30%) is often used for
sand. On the basis of this assumption the flow rate is now fixed because the production
capacity is taken as a given value.

Page 9 of 35

Wb w3408b

Designing Dredging Equipment

4.6.2 The suction depth


A second important design parameter is the suction depth. This determines whether an extra
underwater pump is needed to achieve the required production. When the suction depth
increases, if the use of an underwater pump is not considered the suction pipe diameter and
also the pump flow must be increased. At the same time the concentration must be reduced to
avoid reaching the vacuum limit (under-pressure at which cavitation occurs). This can lead to
the pumping of low concentrations and thus much water, which is uneconomic.
With the aid of the suction formula one can
determine if a submerged pump is useful and
hoe deep below the waterlevel the pump has to
be fitted on the suction tube. The suction
formula is a force balance over the suction tube.
The pressure difference over the suction tube
equals the weight of the mixture in the suction
tube and the friction due to the flow.

k
rw

H
rm

hz

rp

hp

Figure 4. 12 scheme for suction formula

1
L

w g (H h p ) + p gh p p pomp = m ghz + m v 2 1 + +
2
D

with
w = density water
p = density suspended sand in the pit
m = mixture density in the suction tube
H = waterdepth
hp = depth of pit
hx = suction height
ppump = pressure in front of the pump
v
= mixture velocity

= entree loss factor

= Darcy Weisbach headloss factor


L = total length suction tube
D = diameter suction tube

[kg/m3]
[kg/m3]
[kg/m3]
[m]
[m]
[m]
[N/m2]
[m/s]
[-]
[-]
[m]
[m]

Because hz = H k the equation can be written as:

1
L

w g (H h p ) + p ghp p pomp = m g (H k ) + m v 2 1 + +
D
2

This results in:

m =

w g (H hp ) + p gh p p pomp
L
1
g (H k ) + v 2 1 + +
D
2

For the boundaries given in Figure 4.13 the maximum dredgeable mixture density is
calculated for different depth of the dredge pump below thw waterlevel
Page 10 of 35

Chapter 4 Plain Suction Dredgers

Mixture density as funktion depth pump below water line


hp=3 m, Vac=75 kPa, Vz=5 m/s, rho_water=1000 km/m3, G_p=1600 km/m3, Zeta=2 , Lambda=0.02, L/(H-k)=1.5, D=0.8 m

1800

Mixture density [kN/m3]

1700
1600
1500

k=0 m

1400

k=5 m
k=10 m

1300
1200
1100
1000
0

10

20

30

40

50

60

70

80

90

100

Dredging depth [m]

Figure 4. 13

The above graph (Figure 4. 13) is derived from this equation


In order to dredge, from a depth of 30 m, a density of 1500 kg/m3 the dredge should be place
almost 8.5 m below the waterline.
A pump on the waterline can pump a density of1120 kg/m3. In the second case, if the same
production is required, the flow should be:

Q0 5 w 1500 1000
=
=
= 4.17 as great.
Q5 0 w 1120 1000

With the same pumping velocity this leads to a suction pipe of a diameter that is 2 times as
big.
For a given decisive vacuum and a maximum suction concentration it is possible to determine
whether an underwater pump is necessary and, if so, how far under water this pump must be
positioned, as a function of the required suction depth.

Rho_mixutre=1500 kg/m3
hp=0 m, Vac=75 kPa, Vz=5 m/s, rho_water=1000 km/m3, G_p=1600 km/m3, Zeta=2 , Lambda=0.02, L/(H-k)=1.5, D=0.8 m

Depht pump below water level [m]

35.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
0

10

20

30

40

50

60

70

80

90

Dredging depth [m]

Figure 4. 14

From the above graph (Figure 4.14) it appears that to pump a mixture density of 1500 kg/m3
at a depth of 50 metres the pump must be positioned 17 metres under water.

Page 11 of 35

Wb w3408b

Designing Dredging Equipment

Of course whether or not an underwater pump is mounted is a question of economics. The


cost of fitting an underwater pump is considerable and, moreover, the suction depth can have
a great influence on the ladder construction and thus on the pontoon construction. It is also
necessary to hoist the suction pipe above water for inspection.
4.6.3 The transport distance
The transport distance makes demands with regard to the installed sand pump capacity and/or
the need to load barges. The need for barge loading depends whether the required transport
distance is too long to be economically covered by the use of a hydraulic pipeline. It is also
possible that the use of a pipeline may not be feasible from the point of view of hindrance to
shipping. Suction dredgers may also be designed exclusively for barge unloading. In general,
if material does have to be transported by a hydraulic pipeline there is still the option to place
a booster station with the necessary capacity behind the plain suction dredger.
If the suction dredger is equipped with an underwater pump the chosen discharge pressure
(and thus capacity) can be such that during the loading of barges only the underwater pump is
used. The pipeline system and valves can also be designed for this. The grain size and the
distance over which the material must be transported determine the required manometric
pressure for the discharge pump(s). It is also possible to choose an underwater pump of
higher capacity than is needed to unload the barges. The surplus capacity can then be used
during discharging. The maximum discharge pressure that a dredger can supply depends on
the quality of the shaft sealing of the last pump. Often values exceeding 25 - 30 bar are not
permitted.

4.6.4 The dredging installation


Under the dredging the following components are included
Suction and discharge pipe
The dredge pumps
The dredge pumps drives
The jet pumps
The jet pump drives
4.6.4.1 Suction and discharge pipe diameter
The critical velocity that is necessary to keep the dredged material in motion determines the
maximum suction and pressure pipe diameters.

Thus: v kritiek = Fl ,h + Fl ,v

2 g ( S s 1) D in which the value of Fl,h is determined by the

material to be pumped. (See lecture notes Dredging Processes) Fl,v is the correction for
sloping transport and has a maximum value of .333
(See also the relevant Section 2.2.4.3. of Hopper dredgers).
If both the critical velocity and the average concentration have been determined, the relation
between pipeline diameters and production is:

Q = Qmixture

Cvd
D 2 Cvd
D 2 Cvd
C
= vkrit
= Fl 2 g ( S s 1) D
1.5 D 2.5 vd [m/s]
1 n
4 1 n
4 1 n
1 n

with
Page 12 of 35

Chapter 4 Plain Suction Dredgers

Symbol

describtion

Q
Qmixture
D
Cvd
Ss
n
g
vcr

=
=
=
=
=
=
=
=

Production
Flow rate
Pipe diameter
Delivered concentration
Relative density of the solids=s/w
Porosity
Gravity
Critical velocity

dimensio
n
[m3/s]
[m3/s]
[m]
[-]
[-]
[-]
[m/s2]
[m/s]

Figure below give the results of the equation for Cvd=30%

Discharge diameter [m[

0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

1.1 1.2 1.3

Production [m3/s]

Figure 4. 15 Minimum discharge diameter

4.6.5

The dredge pump

4.6.5.1 Pump types


Now that the capacity, the required pressures on both sides of the pump and the power are
known under the various transport conditions, the type(s) of pump can be selected.
The pump types, centrifugal, semi axial or axial are determined by the specific speed of the
pump; defined as:

ns =

( gH )

3
4

4 Q

( p)

3
4

For discharge pumps the specific speed ns is in the interval between 0.25 and 0.50 (Figure
4.16). With the aid of this figure the type of pump and impeller can be chosen.

Page 13 of 35

Wb w3408b

Designing Dredging Equipment

0.08

0.8

0.07

0.7

0.06

0.6

0.05

0.5

0.04

0.4

0.03

0.3

0.02

0.2

0.01

0.1
0

0
0.15

Specific Head

Specific Capacity

Inboard Pumps

0.2

0.25

0.3

0.35

0.4

0.45

0.5

Specific Speed

Figure 4. 16

For the underwater pump usually a higher specific speed is taken than for the discharge
pumps, but for the sake of standardisation the same pump is often selected. One should ask
oneself whether the position of the maximum efficiency point could still reasonably satisfy
the stipulated demands with regard to the flow. This is also valid when no underwater pump
is fitted. In such a case stipulations must be made with regard to the suction properties
(NPSH value) of the inboard pump.
Other factors also play a part in the selection of a pump and impeller:

A three, four or five blade impeller. Depending on the required minimum passage
between the blades.
Single or double walled pump. (considerations relating to wear.)

If long transport distances have to be covered the question arises of whether one large pump
or two smaller ones will be needed. In addition to the specific revolution speed the peripheral
velocity of the impeller also plays a part. To limit wear, the peripheral velocity of the
impeller is limited to 35 to 40 m/s. This also limits the maximum manometric pressure.
Whether or not one or more delivery pumps are needed depends on the total require delivery
pressure and delivery pump power.

4.6.5.2 The sand pump drives


Underwater pumps often have electric drives, but hydraulic drives and even direct diesel
drives may be encountered. If barge loading is required, a controllable drive is necessary.
With a fixed revolution speed the variations in flow resulting from differences in
concentration and grain size are often too big for the efficient loading of the barges.
Diesel drives are often used for the delivery pumps, but of coarse electrical drives are
possible too

4.6.6

Jetpumps

4.6.6.1 Pump type


The flow of the water pumps depends on the required functions of these pumps.
Two functions can be distinguished:
Page 14 of 35

Chapter 4 Plain Suction Dredgers

1. The activation of the breach process of the bank.


Suction dredgers are usually equipped with a water jet for this purpose. The speed of the jet
flowing from the water jet decreases hyperbolically with the distance from the water jet in
accordance with:

vL =

6D
v
L 0

See Figure 4. 17

Here:

v L = Velocity of the jet at distance L in m/s.


D = Diameter of the jet nozzle in m.
L = Distance to the jet nozzle in m.
v 0 = Velocity of the jet at the nozzle in m/s.

Vr
r
D
Jet

VL

V0
L

Figure 4. 17 Flow establishment of a jet

Example.
If the pressure at the nozzle is 500 kPa and the jet nozzle has a diameter of 0.3 m e and a
minimum velocity in the centre of the jet **at the breach/bank of 3 m/s is needed to activate
the breach/bank, the maximum distance to the breach/bank is:

v0
L = 6D
= 6D
vL

2p

vL

0.6
= 60.3

2500
1
= 11 m
3
r 2

90
vr
= e L .
The decrease in velocity towards the edge of the jet can be calculated with:
vL
Here v r = the velocity of the jet at distance r from the centre.

Page 15 of 35

Wb w3408b

Designing Dredging Equipment

v_r=v_L*exp(-90*(r/L)^2)
0.25

r/L

0.2
0.15
0.1
0.05
0
0

0.2

0.4

0.6

0.8

1.2

V_r/V_L
Figure 4. 18 jet velocity as function of the radius r.

At a distance of 11 m and with a relation of vr = 0.4 the diameter of the jet is as shown in the
vL

r
graph below D = 2 L = 201
. 11 = 2.2 m
L
In other words, the influence of the water jet is only very local.

The jet flow is: Q j =

D 2
0.32
18.9 = 134
. m/s
v0 =
4
4

and the power at the water pump: Pj =

Qj p

134
. 500
= 838 KWatt
.8

2. The maintenance and control of mixture forming.


In this case, when it is assumed that no water from the environment can be sucked in because
the suction mouth is completely embedded in the soil, it is necessary to satisfy the volume
balance:

Qj
Qm

= 1

Cvd
1 n

Here:

Q j = the jet flow m/s


Qm = the sand flow in m/s
Cvd = the transport concentration [-]

= the pore number [-]

Figure 4. 19 gives a graphical representation of the equations.

Page 16 of 35

Chapter 4 Plain Suction Dredgers

Verband Qj/Qm - Cvd


0.8
n=.35
n=.4
n=.45
n=.5

Cvd

0.6
0.4
0.2
0
0

0.5

1.5

Qj/Qm
Figure 4. 19

Example:
If Cvd = 0.25 and n=0.5 (loose packed sand), then

Qj
Qm

=.5

The area of influence by the jet is now less important, as long as the water that is added
benefits mixture formation.
The water pumps are chosen in the same way as the sand pump
4.6.6.2 Jetpump drives.
In case of activation the breaching process required pressure and capacity will always be
constant. So separate diesel engines are frequently used.
In the other case, the mixture forming process a speed control engine is required to control
the density.

Page 17 of 35

Wb w3408b

Designing Dredging Equipment

4.7 General layout

3,000

3000

2,500

2500

Light weight [t]

Light weight [t]

The hull consists of a simple U-shape pontoon. De width of the pontoon is determined by
stability and sometimes by the distribution of the loads. (Figure 3.1.7) The length of the
pontoon is in certain way determined by the length of the suction pipe, the number dredge
inboard pumps or by the requirements for mooring barges along side.
Loads on the suction pipe resulting from the dredging process are relatively small, so are the
loads on the pontoon. For small plain suction dredgers the dredgepump is situated in the
engine-room, however a separate pump room is certainly advisable from safety point of view,
in particular for the bigger dredgers. Nowadays even small dredgers do have a submerged
pump.

2,000
1,500

y = 0.2712x

1,000

R = 0.712

y = 0.4074x
2

2000

R = 0.8715

1500
1000
500

500

0
0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

1000

2000

3000

4000

5000

6000

LBD [m3]

Total installed power [kW]

Figure 4. 21

Figure 4. 20

The lightweight of the plain suction dredgers depend on the total power installed. (Figure
4.20), while the volume of the pontoon is 2.5 times the light weight (Figure 4.21).
The main ships parameters vary widely; L/B between 3 and 8 and B/T between 7 and 3.5,
because the length is mainly determined by the factors mentioned above. (Figure 4.22)

L/B & B/T

L/B

B/T

9.00
8.00
7.00
6.00
5.00
4.00
3.00
2.00
1.00
0.00
0

500

1000

1500

2000

2500

3000

Light weight [t]

Figure 4. 22

Figure 4.23 shows the dredger Seeland, with a total installed power of 3200 kW and a
maximum dredging depth of 40 m. The dredger is build under the classification of the
Germanische Lloyd GL + 100 A 4 dredger.
The length of the suction pipe often determines the length of the well. With very long suction
pipes or two-part suction pipes the catamaran principle is often used. The suction pipe is then
hinged onto the stern of the pontoon (Figure 4.2) This is certainly not essential.
Sometimes special gantries are designed to carry the long suction tube (figure 2.23).

Page 18 of 35

Chapter 4 Plain Suction Dredgers

Figure 4. 23

In deep dredgers with an articulated pipe, the lower pipe is fastened to the upper pipe by
hydraulic cylinders, in which case it is not necessary to have a long well (Figure 4.24).

Figure 4. 24 Plain Suction _Dredge Seeland, Yard Orestein and Koppel


Lbeck Germany

Page 19 of 35

Wb w3408b

Designing Dredging Equipment

In other cases an additional


pontoon is connected to the main
pontoon by means of a special
construction (Figure 4.24 PSD
Weesperkaspel).
The engine room, pump room, fuel
tanks, water tanks and storeroom
are all located in the pontoon. On
small suction dredgers the sand
pump is located in the engine
room, while large suction dredgers
have a separate pump room. The
control cabin, and if required, crew
quarters are above deck. The
anchor winches are also on deck

Figure 4. 25

Figure 4.25 shows an offshore plain suction dredgers designed for significant wave heights of
2.75 m and a total installed power of 7425 kW. The coupling with the floating pipeline is in
the middle of the port side where the movements of the pontoon are minimum when working
in waves. This is in contradiction with dredgers for inland waters. They do have the
connection on the aft of the pontoon.

Figure 4. 26 General arrangement of an offshore plain suction dredger,


Yard IHC Holland

4.8 Technical construction


4.8.1 The hull
As previously mentioned, the hull usually consists of a simple U-shaped pontoon.
The width of the pontoon is determined by stability considerations and varies from 6 m for
small to 20 m for large deep dredgers. The length of the dredger is usually determined by the
requirements relating to the length of the suction pipe and/or the need to accommodate barges
alongside and by the warping of the barges.

Page 20 of 35

Chapter 4 Plain Suction Dredgers

The ladder gantry, which usually


takes the form of an A-frame,
provides the link between the
pontoons, which are separated
by the well. By deep dredgers,
having a suction pipe in the
raised position pointed very far
ahead of the pontoon, the gantry
is a relatively heavy structure
(Figure 4.23 and 4.27).

Figure 4. 27

4.8.2
The dredging equipment
The dredging equipment will be discussed according the flow o f the mixture.
4.8.2.1 The suction mouth
Suction mouths of plain suction dredgers are in many cases very simple. The end of the pipe
is just covered by a screen to avoid pump blockage by boulders and debris (Figure 4.1, 4.28
and 4.29)

Figure 4. 28

Figure 4. 29

In many cases jet nozzle are situated around the suction mouth to activated either the
breaching process and/or the mixture forming (Figure 4.30)
When the suction mouth is fully penetrated in the sand, water jets are necessary the fulfil the
requirements for the mixture forming. In that case jets are situated around the suction mouth
(figure 4.31)

Page 21 of 35

Wb w3408b

Designing Dredging Equipment

Figure 4. 30

Figure 4. 31 Suction mouth of the sea going


PSD DECIMA

4.8.2.2 The suction pipe


For many suction dredgers the suction pipe, together with the jet water pipe, forms a strong
construction (Figure 4.32). To strengthen the suction pipe this it also equipped with a jacket
pipe through which the jet water flows to the suction mouth. If this jacket pipe is divided into
sections, these can also be used as float tanks to reduce the underwater weight of the suction
pipe.

Figure 4. 32

With bigger dredgers, and certainly at


greater
suction
depths,
these
constructions are too weak and it is
necessary to turn to the use of a ladder
(Figure 4.19). If an underwater pump
is used, the upper part of the suction
pipe must certainly be constructed as a
ladder in order to transfer the heavy
weight to the hull.

Figure 4. 33

Page 22 of 35

Chapter 4 Plain Suction Dredgers

On the suction pipe there is often


a water admitting valve or
breaching valve. If, as a result of
irregular
shearing
of
the
breach/bank the vacuum becomes
so high that the pump starts to
cavitate and threatens to cut out,
water can be admitted through
this valve to keep the process
going. This valve, which was
formerly operated manually, is
currently regulated automatically
by the under pressure in front of
the pump.

Valve open
Valve closed

Cylinder

to pump

From suction mouth

Figure 4. 34

To ensure good control it is advisable to provide the valve with two openings, a big one for
sudden emergencies and a second smaller valve that can be used for fine control with a
continuously high vacuum.
A rubber suction hose forms the link between the suction pipe and the pipelines on board.
This rubber hose is equipped with vulcanised steel rings, which prevent it from collapsing
when under pressure occurs in it. The centreline of the suction hose is at the same height as
the hinge and often lies beneath the waterline (Figure 4.35).

Figure 4. 35

To prevent water from flowing in during pump inspections a so-called outboard valve must
be fitted onboard before the pump
PSDs without a submerged pumps have to be designed in such away that the suction pipeline
is as short as possible. Where the suction pipeline comes above water, the chance of taking in
air must be reduced to the minimum. (Taking in air has the same effect as cavitation.)
4.8.2.3 The sand pumps
Barge-loading suction dredgers usually have only one pump, even when the dredger is
equipped with an underwater pump, while reclamation dredgers have one or more inboard
pumps independent if provided with an underwater pump.

Page 23 of 35

Wb w3408b

Designing Dredging Equipment

Figure 4. 36 View on ladder with pump for a PSD

When suction dredgers do not have an underwater pump, efforts must be made the position of
the first pump must be as deep as possible below the water line. This means on the base of
the pontoon. As well as good discharge characteristics, the first pump must also have good
suction characteristics, thus a high decisive vacuum and/or a low NPSH value.
If the dredger is equipped with an underwater pump the layout is less critical. In that case
aspects such as accessibility for inspection and repairs play a more important role. The
onboard pump is then only required to possess discharge characteristics. For the required
specific speed for these pumps referred is to chapter 2.2.3.5 Dredge pump.
Submerged pumps have mainly a single
wall, while inboard pumps have either a
single or a double wall.
If there is more than one inboard pump the
layout must be chosen in such a way that,
if desired, it is also possible to work with
the ladder pump and one inboard pump.
An inspection hatch must be provided for
every pump, so that the pump and the
impeller can be inspected and, if
necessary, debris can be removed.
.

Figure 4. 37 Double wall pump

Page 24 of 35

Chapter 4 Plain Suction Dredgers

4.8.2.4 The sandpump drives


The underwater pump often has an electric drive while the inboard pumps are powered by
diesel engines. Diesel direct driven submerged pumps is till today in use for relative low
powered pumps. See also chapter 3.2.3.4
4.8.2.5 The discharge pipeline
Reclamation dredgers pump the dredged material ashore by means of a floating pipeline and,
if necessary, to a more distant disposal site via the land pipeline.
Because the movement of the suction dredger is considerably less than that of a cutter suction
dredger, it is not necessary to connect the discharge pipeline of the vessel to the floating
pipeline by means of a swivel on the stern of the vessel. Often the discharge pipeline is
connected to the floating pipeline by means of a delivery hose/pressure hose (a floating
rubber hose). This can be mounted either on the stern of the vessel or on the port or starboard
side.

Figure 4. 38 Ths sea-going PSD AURORA with the discharge pipeline connected on starboard

4.8.2.6 Sprayers
If the dredged material has to be loaded into barges alongside because the transport distance
is too long for pipeline transport to be economic, sprayers which are connected to the
discharge pipeline are fitted on both sides of the dredger. The number of sprayers that is
fitted on each side of the dredger depends on the capacity of the dredger and the size of the
barges and varies between one and four per side.

Figure 4. 39 Two different types of sprayers

Page 25 of 35

Wb w3408b

Designing Dredging Equipment

To prevent barges from being unevenly loaded, the sprayers must be positioned as closely as
possible to the centreline of the barge (Figure 4.39). Sometimes extra measures are necessary
for this. For example, when it is necessary that to satisfy the demand that free fall of the
dredged material must be prevented, the sprayers must be positioned as low as possible.
The capacity of the pump and the pipeline plan must be designed in such a way that on each
side a barge can be loaded simultaneously. The sprayers are moved by means of winches or
by a hydraulic system.

Figure 4. 40 barge loading with movable sprayers

4.8.2.7 Jet-pipeline and pump


The jet pipeline is of such a size that the pipeline loss remains within acceptable boundaries.
It is advisable to design the bends, valves, crossovers etc. as large as possible in order to keep
the losses within acceptable limits.
Often a sand pump is used as a jet pump to keep the wear between limits. This is certainly
advisable when the dredger is a barge loading suction dredger. The water surrounding the
dredger due to the overflow of the barges is diluted by fine sand particles, and thus the water
taken in by the water pump.

4.8.2.8 The winches


Besides the ladder winch and the auxiliary winches, the Suction dredger is equipped with six
winches for mooring:
one bow winch
two forward side winches
two after side winches
one stern winch to maintain tension on the bow winch
4.8.2.9 The ladder winch
Page 26 of 35

Chapter 4 Plain Suction Dredgers

The ladder winch that serves to adjust to the


correct dredging depth is usually mounted on
deck. If the hoisting wire runs through one or
more blocks, the lowest block is fastened to
the suction pipe by a rod (Figure 4.41). This
is to prevent the block from being fouled by
sand when dredging an irregularly shearing
breach/bank. At present slow running electric
or hydraulic drives are used.

Rod

Figure 4. 41

4.8.2.10The bow winch


With the aid of the bow winch the suction pipe is held against the breach or bank. For the
optimum control of the suction process good control of the bow winch is essential. It must be
possible to pay out the bow winch quickly when moving the bow anchor. Bow winches are
mounted on or below deck. Because of the great length of the bow wire, the bow winch
usually has a large drum.
4.8.2.11The side winches
The side winches control the position and direction of the dredger in both the cut and in the
dredging area. Side winches are usually mounted on deck and are electrically or hydraulically
driven.
4.8.2.12The stern winch
The stern winch has a secondary function, namely to maintain tension on the bow wire, and it
does not determine the production. Like the side winches it mainly comes into action when
the dredger is being moved to another cut. The stern winch is usually mounted on the stern
deck and electrically or hydraulically driven.
4.8.2.13The auxiliary winches
The moving of the sprayers and the warping of the barges is usually done by separate
winches.
One or more jib cranes may be fitted and used to lift heavy parts during repairs.
4.8.2.14The fairlead
To sail the barge from and to the dredger fairleads are used to bring the side line wires on a
sufficient depth below the water level that the barge can sail over the wires.

Page 27 of 35

Wb w3408b

Designing Dredging Equipment

Pin to change the height


of the fairlead

Fairlead
guide

Side wire

Figure 4. 42 Fairlead

Page 28 of 35

Chapter 4 Plain Suction Dredgers

4.9 The dredging process


The dredging process of a suction dredger can be subdivided into
1. The behaviour of the breach/bank during dredging also termed the breach/bank
production.
2. The suction production of the dredger.
3. The discharge production of the dredger.
The last two productions will not be considered in these lecture notes. They will be treated in
a course on dredge pumps and pipeline transport because the calculations involved are
similar for all types of dredger.
4.9.1

The production of the breach

When a vertical suction pipe is


lowered into a sand layer quickly,
narrow pit forms with almost
vertical side slopes (Figure 4.43).
The diameter of the pit decreases
from the top downward with time
so the sand grains and sand
fragments glide down under the
force of gravity.
The velocity at which the
instability of the slope moves
depends on the permeability and
the relative density of the sand
layer and is roughly 20 to 40 times
the permeability, depending on the
slope and the angle of internal
friction of the breach.

Suction tube

240
210
180

vwall

Slope
150

vs

120
100

0
80

Time in seconds

15

60
50

20
40

30

Suction velocity vs = 2.5 m/s

Figure 4. 43

Detailed information about this process can be found in the lecture note wb3413 the
Breaching process .

When, under laboratory


conditions, a 2-D suction
mouth is moved forward
with a constant speed at
the base of a breach, a
slope with an angle will
occur which is much
steeper than the angle of
internal friction. (Figure
4.4)

vh

vw
a

b
B

Figure 4. 44

The relation between vw and vz follows from the similarity of shape after a time t.

Page 29 of 35

Wb w3408b

Designing Dredging Equipment

vw t =

H
and vh t =
tan

H
1
1

tan tan

From this it thus follows that:

tan

v h = v w 1
tan
Production per metre wide:

tan
Qsand = vh H = vw H 1

tan
Here H is the height of the breach/bank.
The cause of the steeper slope is cause by the dilantancy (an increase of porosity) due to the
shearing of the sand matrix. When the porosity increases pore water has to flow to the these
large pores. When this happens slowly a decrease in pore pressure will occur and a increase
in the effective stresses causing an more stability. When sufficient water has flowed into the
pores the under pressure and additional stability will vanish.
When a 3D suction pipe is moved
forward horizontally at a constant
speed a pit forms the slope of which is
at its steepest directly in front of the
suction pipe (Figure 4.45). The slope
decreasing at the sides to a value that
is determined by the eroding effect of
the density current flowing towards the
suction mouth. The angle between
the slope just in front of the suction
pipe and the horizontal can be derived
according above. If all the material is
removed, the production will be:

Q =W

H
tan

a)

FG 1 1 IJ
H tan tan K

H
tan
Symmetry plane

H
tan

Vh

H
H2
vh = vh
2
tan

However, due to the movement of the


suction tube not all the material from
the side slopes will reach the suction
mouth and spillage will occur.

b)

Spillage from breach

Figure 4. 45

This spillage can be calculated with the following production balance can be set up:

(H S) v
2S S
H S
Hvh
vw =
h
tan
2 tan
tan
2

with:
Symbol

Declaration
Page 30 of 35

Dimension

Chapter 4 Plain Suction Dredgers

H
S
vh
Vw

Maximum pit depth


Height of spillage
Horizontal velocity suction mouth
Distortion (Wall) velocity
Minimum slope angle angle of internal friction

M
M
m/s
m/s

The first term is the volume per unit of time passing through area of the plane TAR, the
second term is the production from the face BAT and BRA with S being the average height
retrogressive erosion or wall over the area considered and the term on right side of the
equation is the volume per unit of time passing through a plane with the final cross section.

0.5b

H
tan

H
H

tan tan

S
tan

0.5b

H
S

Figure 4. 46

This leads to:

Page 31 of 35

Wb w3408b

Designing Dredging Equipment

H 2 HS
S 2 vw H 2 2 HS + S 2

=
tan
tan vh
tan
H 2 HS S 2

tan vw
= H 2 2 HS + S 2
tan vh

S = 0 and S =

H
tan vw
1+
tan vh

The theoretical production without spillage, according equation Q = vh

H2
,
tan

v
H
v
w
the real production Q = h
tan
, and
tan
vh + vw
tan

the spillage production Q


spillage

tan

v
vh H 2 tan h
=

tan tan v + v
h
w
tan

Laboratory measurements have shown that tan = 4.77 .


tan
However, in practice appeared that the angle is small too. Taking = results in a
production of:

vh H 2 vw
vh H 2 1

Q=

=
tan vh + vw
tan vh + 1
vw

4.9.2 The production of the pumps


The sand flowing towards the suction mouth will be taken up by the dredger and must be
transported away by means of barges or pumped to the disposal site via a pipeline.
Depending on the pipeline system and the position(s) of the sand pump(s) the following
situations may occur.

More sand flows to the suction mouth than the pumps can handle. The pump is the limiting
factor and this criterion can be subdivided as follows:

The under-pressure/vacuum in front of the pump is the limiting factor. The under
pressure in front of the pump is so high that cavitation occurs, resulting in the loss of the
discharge pressure. The pump then cuts out. The only good remedy is to position the
underwater pump deeper.
The discharge pressure is the limiting factor. The discharge distance is so long that the
pressure required for the critical velocity of the mixture is higher than the pump can
deliver. A stationary deposit will be formed in the pipeline, with the chance of a totally
blocked pipeline. Depending on the loading on the engine, consideration can be given to
the installation of a pump with a larger impeller or to changing the transmission ratio in
the gearbox. If the loading of the engine is already maximal the maximum concentration
has been reached.
The pump torque is the limiting factor. This is the contrary situation to the above
mentioned limiting pressure situation. The remedy is to use a smaller impeller.
Page 32 of 35

Chapter 4 Plain Suction Dredgers

4.9.3 The production of the barges


The pump production of a barge loading stationary suction dredger is not the same as the
amount of material transported by means of barges. This is caused by the overflow losses that
occur during the loading and also the bulking that occurs because the sand in the barges often
has a lower density than the in situ density. These two factors must be taken into account
when determining how many barges are required.

The number of barges follows from:

n=

P( 1 ov )
P( 1 ov )
P(1 ov )
=
=
t cyclus
Lbak
Pbak
Lbak
t cyclus

(4.24)

Here:
N
P
Ov

Lbarge
Tcycle

= number of barges
= pump production
= overflow loss [-]
= bulking factor
= load of barge [m]
= cycle time

[-]
[m/s]
[-]
[s]

As a rule of thumb the percentage smaller than 100 m can be taken as overflow losses.
The bulking is determined by the difference volume weight in situ and in the barge. With
strongly graded material the volume weight in the barge is 19 kN/m and with uniform
material this can decrease to 18 kN/m. For the calculation of the bulking reference should
be made to Section 2.6.3.1.
The cycle time of the barge is composed of:
the loading time
the sailing time
the discharge time
the return sailing time
waiting times for bridges, locks etc.
In addition to the fact that the pit or the pump can be *maatgevend, with a barge-loading
dredger, a situation may occur in which the barges are *maatgevend. In other words there are
not enough barges. A situation that may have a variety of causes such as:
weather and wave conditions
shipping
Bridges and lock
Unequal speeds of the barges
Loss of time by the barge
Delays on the dredger
Loss of time at the discharge site
It will be clear that when using a barge-loading dredger there is always a chance of delays
due to the absence of a barge.

Page 33 of 35

Wb w3408b

Designing Dredging Equipment

Because the above mentioned delays can be reasonably well estimated with regard to their
average values and standard deviations, the Monte Carlo Simulation can provide insight into
the probability of delay resulting from the absence of barges.

4.10

The dustpan dredger

As appears in chapter 4.9.1, the


production of the suction
dredger is proportional to the
square of the breach height.
With low breach heights the
production remains lower than
the discharge capacity of the
pump. In order to compensate
this to some extent, a broad
suction mouth, the dustpan
head, is mounted on the suction
pipe. The width of the dustpan
head is 10 - 15 times the
diameter of the suction pipe. In
addition a large number of
spray nozzles are mounted on
this suction head, which by
means for water jets stimulate
breaching process. Moreover
they are necessary to prevent
the
suction
head
from
becoming
blocked.
The
working effect of the spray
nozzles can be calculated in the
same way as is given in chapter
4.5.6.1.
In fact, the dustpan dredger has
been superseded by the cutter
suction dredger, which, with a
considerable larger width of
cut, can attain a much higher
production
on
low
breaches/banks.

zuigmond

Figure 4. 47 Dustpan heads

Dustpan dredgers are now only used for small projects or on special dredgers such as the
Cardium. The Cardium is equipped with 6 suction pipes and suction pumps, each with
two suction mouths, in order to ensure that the bottom is at the correct depth (the foremost
suction mouth is in dustpan mode) and is flat and clean immediately before a block mattress
is laid down (clean up model).

Page 34 of 35

Chapter 4 Plain Suction Dredgers

Figure 4. 48 Dustpan haed with pump and pipel ine sceme of the matress laying vessel CARDIUM

4.11

References

1. Offshore soil mechanics, Verruit, 1992


2. Investigations to the spillage of the horizontal suction process, W.J. Vlasblom, to be
published in May 2003.
3. Hydraulic excavation of sand, H.N.C. Breusers, Proceedings International course Modern
Dredging, June 1977, The Hague
4. Neue Erkentnisse beim Gewinne und Transport von Sand im Splproject Venserpolder,
J. de Koning
5. Coastal & Ocean Dredging, J.B. Herbich, Gulf Publishing Company, Texas
6. Lecture notes wb 3413 The Breaching Process
7. Lecture notes additional to wb 3414 Dredge pumps

Page 35 of 35

Chapter 5 Barge Unloading Dredgers

5. The barge unloading/reclamation Dredger


5.

The barge unloading/reclamation Dredger........................................................................ 1


5.1.
General considerations...................................................................................................2
5.1.1. Characteristics ............................................................................................................3
5.1.2. The areas of application..............................................................................................3
5.1.3. The history..................................................................................................................3
5.1.4. Work method ..............................................................................................................4
5.2.
The design......................................................................................................................6
5.2.1. The production capacity .............................................................................................6
5.2.2. The transport distance.................................................................................................6
5.2.3. The dredge installation. ..............................................................................................7
5.3.
Main layout ....................................................................................................................13
5.4.
Technical construction ...................................................................................................16
5.4.1. The hull.......................................................................................................................16
5.4.2. The pipelines ..............................................................................................................18
5.4.3. The shore connection..................................................................................................18
5.5.
The dredging process .....................................................................................................19

Figure 5-1 Barge unlading dredger HOLLAND

A specialized dredging tool that can be categorized in the section of stationary plane suction
dredgers is the barge unloading/reclamation suction dredger.

Page 1 of 21

Wb3408b

5.1.

Designing Dredging Equipment

General considerations

Barges that are used for the transport of dredged material can be divided in self-unloading and
non-self-unloading. The self-unloading barges, called hopper dump barges or bottom unloaders,
are usually equipped with doors (valves) that one way or the other can be opened to dump the
dredged material under water.
Non-self-unloading barges need to be unloaded either mechanically or hydraulically. Mechanical
unloading can be done with a grab, backhoe, excavating wheel or bucket elevator. Non-selfunloading barges are therefore often called elevator barges.

Figure 5-2 A Japanese BUD with backhoes and belt conveyors.

Hydraulic unloading can be done using a shore pump discharge system, usually installed in
trailing suction hopper dredgers or by means of a barge unloading suction dredger.
For the last 20 years the transport with barges is strongly reduced and because, as mentioned, the
barge unloader is a specialized dredge tool, it is hard to use the tool for other purposes. Hence the
amount of barge unloading suction dredgers has decreased considerably in this period. At present
many barge unloading suction dredgers are in service that can also be used as plain suction
dredger or cutter suction dredger.

Page 2 of 21

Chapter 5 Barge Unloading Dredgers

5.1.1.

Characteristics
28

28

13a

15

17

14

22
18

8
20
22

4
3

16
2

13

12
6

13a

10

11

6
3

9
25

19

13

10

24

13

11
9
9
25

25
7
8
13a
16

13

19

15

14

17
30
31
31

18
21

20

13
3

13a

18

29

30

22

31

31

23

Figure 5-3

The barge unloading suction dredger is a stationary dredge tool, moored along mooring piles or
anchored with spuds. (Figure 5-3) The barges are moored along the tool for unloading. The tool
is equipped with one or more sand pumps and a jet pump. The suction pipe sticks out at the side
of the tool and can be lowered in the barge lying next to the dredger. The water needed for the
mixture and the transport is jetted into the barge using one or more nozzles.

5.1.2. The areas of application


The barge unloading suction dredger is able to unload barges hydraulically. These barges are
filled one way or the other, for instance with a plain suction dredger or a bucket ladder dredger.
The material in the barge is diluted with water and sucked up (figure 5.1). This immediately
implies that the barge unloading suction dredger can only handle materials that fluidize quickly
like silt and sand. Cohesive materials, of which the forming of a mixture is too slow, will cause
the barge unloading suction dredger a lot of problems.

5.1.3. The history


The barge unloading suction dredger is a Dutch development. During the excavation of the North
Sea Canal a stationary plain suction dredger was transformed to a barge unloading suction
dredger ( 1875). Before the barges were unloaded using a bucket elevator. Next the material was
transported to the dump with small sand trains. With the arrival of the barge unloading suction
dredger these trains, which were very labor-extensive became redundant. Besides it was now
possible to transport weak soils simply. The first pressure pipes were mad of wood but soon these
were replaced by iron pipes.

Page 3 of 21

Wb3408b

Designing Dredging Equipment

Figure 5-4 The steam driven BUD Sliedrecht I

5.1.4. Work method


In Figure 5-1 the working method is schematically explained
At the start of unloading process, the suction pipe is lowered to the sand level in the barge, while
the jet pump is connected to the suction pipe. The speed of the dredge pump on board of the
dredgers is reduced in such away that the jet water flows via the suction tube on the sand in the
barge, where its erodes a pit under the suction mouth. The dredge master lowers the suction
mouth below the water level in this pit. When no air is released via the suction mouth, the
butterfly valve between the jet pump and suction pipeline is slowly closed, causing an outflow of
jet water via the jet nozzle. (Figure 5-5.A.). Meanwhile the speed of the dredge pump is
increased
When the dredging process is running well, the jet nozzle erodes the breach while the sand is
removed via the suction mouth. During this process the pit under the suction mouth becomes
larger and the suction mouth is lowering until she reaches the bottom of the barge. (Figure
5-5.B). Sand flowed behind the suction mouth has to be jetted back to the suction mouth regularly
(Figure 5-5.C). Therefore modern BUDs have either a jet installation around the suction mouth
or additional jet pipe to overcome this problem The concentration in the discharge line is
controlled by hauling the barge (Figure 5-5.D).

Page 4 of 21

Chapter 5 Barge Unloading Dredgers

Jet pjpe
Suction pipe

Figure 5-5 Working method of barge unloading

During the exchange of the barges the pressure side of the jet pump is connected with the suction
side of the sand pump. This keeps the sand pump moving in the discharge line. The more the
sand-water mixture is exchanged for clean water in the discharge pipeline, the velocity increases
and if necessary the number of revolutions of the sand pump can be reduced. Apart from the
continuation of the dredging process, this construction is necessary to prevent the suction in of air
through the suction mouth of the suction pipe, with all consequences (think of submerged
pipelines).
When the next barge is moored along the barge unloading suction dredger, the number of
revolutions of the sand pump is decreased such that it just can handle the flow rate of the jet
pump. The surplus water is run away through the jet piping and the suction pipe and a new dredge
cycle can start.

Page 5 of 21

Wb3408b

Designing Dredging Equipment

Figure 5-6 Unloading a barge

5.2.

The design

The barge unloading suction dredger has to fulfill in principal two functions:
1. the material in the barges must be diluted such that a mixture develops that can be sucked up
in high concentrations.
2. the dredge pumps in the dredger have to take care that the sucked up material can be pumped
to the reclamation area with enough velocity and production.
5.2.1.
The production capacity
Like with the other tools the required production capacity plays a crucial role in the design. The
production capacity is however determined by the supply of the sand by barges and therefore by
the tool that loads the barges. This can be, for instance, a barge loading plain suction dredger, a
backhoe dredger or a bucket dredger.
For the design of the barge unloading suction dredger the required production for each barge is
the criterion, so the required discharge time for each barge. After all the non-presence of barges
by external causes has nothing to do with the required production capacity. Besides that the size
of the barges is of course of influence on the required production capacity.
5.2.2.
The transport distance
The transport distance gives requirements for the installed dredge pump power and the necessity
for the installation of one or more pumps. For further details with regard to the choice of the
pumps see Chapter 4 Plain suction dredgers.

Page 6 of 21

Chapter 5 Barge Unloading Dredgers

5.2.3.

The dredge installation.

5.1.1.1 General
When the dredge capacity is known, this requirement, like with the plain suction dredger, is
translated in:
1. a sand flow rate Q
2. a sand concentration Cvd
After all:

P = Q

Cvd
1 n

with:

=
=
=
=

production
flow rate
transport concentration
porosity

P
Q
Cvd
n

[m3/s]
[m3/s]
[-]
[-]

The minimum flow rate is determined by the critical velocity that is required to keep the material
in motion. So vcritical = Fl , h 2 g ( S s 1) D in which the value of Fl,h is determined by the to be
pumped material (see wb3414, Dredging processes).
The maximum concentration that can be sucked depends on many factors, like:

the breach behavior of the soil.


the design of the suction mouth in comparison with the width of the barge.
the maximum mixture forming that can be reached with the water nozzles and the jets at the
suction mouth and the flow rate of the jet pump.
the height of the suction pipeline. Because the maximum under pressure is created here, it
determines for a large part the maximum concentration. As a value a concentration of 1400
kg/m3 is maintained.

This last factor can be checked with the vacuum formula (see also Figure 5-7):

H + k v2
H b g + vac = +
+ H + k m g

D sin 2 g

H b g + vac
m =

H + k v2

+
+ H + kg

D sin 2 g

In which:
H
k
vac
w
m

Depth suction mouth below water level in barge


height discharge piping above the water in the barge
maximum allowable vacuum in the discharge piping
density water
density mixture

Page 7 of 21

[m]
[m]
[kPa]
[k/m3]
[k/m3]

Wb3408b

[k/m3]
[]
[-]
[-]
[m]
[m/s]
[m/s2]

density mixture of the water in the barge


angle of suction pipe with horizontal.
loss coefficient
friction coefficient Darcy Weissbach
diameter suction pipe
suction velocity
Gravity

D
v
g

Designing Dredging Equipment

Figure 5-7

For H=2.5 m; b=1050 [k/m3]; vac = 90 kPa; +

H +k
D sin

= 1.5 + 0.01( hz + k ) and v= 4

m/s the below shown graph is obtained.

19
18
17
16
15
14
13
12
11
10
0

H e ig h t su c tio n lin e a b o ve w a te rle ve l in b a rg e [m ]

Figure 5-8

This graph shows (Figure 5-8) that the upper side of the suction pipe may lay hardly more than 3
m above the water level in the barge to meet the earlier mentioned requirement of m=1400
[kN/m3]. This height needs than to be sufficient to haul the barge underneath the suction pipe.
Page 8 of 21

Chapter 5 Barge Unloading Dredgers

The expected average concentration during the suction of the barge is dependent on:
the time necessary to start the process, see the chapter the dredging process 5.4.
the availability of a barge hauling installation. The production is mainly determined by the
haul speed of the barge.
the whether or not present of additional bulkheads in the barge, for which extra breaking in
necessary.
When both the critical velocity as the average and maximal concentration are determined, both
the pump flow rate and the diameter of the pressure piping are also fixed (see chapter 4.2.1).
5.1.1.2 The suction mouth and pipe
Nowadays the suction mouth of a BUD is
provided with jets to improve the mixture
forming and to hindered the settling of
material behind the suction mouth (Figure
5-9)
The width of the suction mouth is based on
the smallest hopper width of the barge. Are
barges used with different sizes it is
advisable to design a flexible suction pipe
(Figure 5-9).
Sometimes the suction mouth is provided
with bars to avoid debris and boulders
entering the suction mouth.

Figure 5-9 Suction mouth and pipe

5.1.1.3 The jet pumps


All the water necessary to transport the sand over the required distance must be supplied to the
barge by the jet pump.
The flow rate of the jet pumps depends on the functions of these pumps. Usually two functions
are considered:
1. The activation of the breach. By way of a water nozzle before the suction mouth the breach is
activated. Usually a second water nozzle is present that jets loose the sand behind the suction
mouth so that it still is sucked up by the suction mouth.
2. The mixture forming. The flow rate of the jet pump must be related to the average
concentration that can be sucked. Here also that the following condition must be met:

Qj
Qm

= 1

Cvd
1 n

In this:
Qj
Qm
Cvd
n

=
=
=
=

the jet flow rate


the sand flow rate
the transport concentration
pore percentage

Page 9 of 21

[m3/s]
[m3/s]
[-]
[-]

Wb3408b

Designing Dredging Equipment

Relation Qj/Qm - Cvd


0.
0.

n=.

0.
0.

n=.

0.

n=.

0.
n=.

0.
0
0

0.

0.

0.

0.

Qj/Q

Figure 5-10

Looking at the above mentioned boundary conditions (Figure 5-10) the flow rate of the jet pump
needs to be 0.4 to 0.5 times the flow rate of the sand pump. With a decrease in the concentration,
like when the suction mouth reaches the end of the barge, the flow rate of the jet pump will have
to increase to maintain the desired velocity in the pressure piping. If this is not possible the water
level in the barge will drop. If there is however enough water in the barge to maintain the velocity
there is no problem. If this is not the case water have be supplied in another way to maintain the
velocity in the discharge line. F.i. an additional water inlet connected to the suction side of the
discharge pump

Jetpump

jetpump engine

suction
strainer

2
1
Nozzle

3
4

dredgepump engine

Dredgepump

Turning gland

Suction mouth
valve

Figure 5-11 Pump-pipeline layout on board of a barge unloading dredger

This is possible by installing a pipe from the suction side of the pump to the bottom of the
pontoon or the suction strainer or weed box (Figure 5-11). In such a design enough water can be
sucked up at all times to maintain the dredge pump process , also when the unloading of the barge
is stopped completely.

Page 10 of 21

Chapter 5 Barge Unloading Dredgers

5.1.1.4 The jet pump drive


The drive of the jet pump may be electrical or diesel driven. The dredge master controls the
process visually by keep the water level in the barge at a constant height. Increasing or decreasing
of the water level determines that there is no equilibrium between the volumes water pump into
the barge and the mixture pump out of the barge. Therefore speed control is necessary to control
the unloading process well.
5.1.1.5 The sand pump.
The dredge pump should be chosen on basis of discharge properties and less on suction
properties, because the last properties are mainly determined by the highest point of suction
pipeline.
The required manometric pressure of the pump is determined by the transport distance. When
large pumping distance is large, more than one dredge pump may be necessary.
The use of submerged pumps close to the suction mouth to increase to design density of the
mixture is also possible but expensive. For Dutch dredging environment it seems not useful due
to the shallow and relatively small barges. However in Japan where large sea-going barges are
frequently use, there is a need for a submerged pump as shown inFigure 5-12.

Figure 5-12 Japanese BUD

5.1.1.6 The sand pump drive


In the process of barge unloading suction dredging the control of the sand pump(s) plays an
important part. After all, when the sand pump is not connected to a suction strainer, the flow rate
must drop to the value of the jet pump when exchanging the barges. This is done by decreasing
the number of revolutions of the dredge pump drive. By the decrease in flow rate this will usually
not cause any trouble for the allowable couple of the drive.

Page 11 of 21

Wb3408b

Designing Dredging Equipment

5.1.1.7 The barge hauling installation


Modern BUDS have an installation to move the barge along the dredger by means of a so-called
barge hauling installation.
The installation consist of a
steel wire or rail along the full
length of the mooring side of
the BUD. (Figure 5-13)
On a pulley or a movable part
on the rail two slings are
connected. These sling are on
the side connected to the
bollards on the barge. (Figure
5-14)
This construction has the
advantage that the barge is
kept along side of the BUD,
The pulley or slide is
connected via a wire to a
winch, which makes it possible
for the dredge master to
control the haul speed by
himself.

Figure 5-13 Sliding part of the Barge Hauling installation

Figure 5-14 Barge hauling installation with pulley and wire


Page 12 of 21

Chapter 5 Barge Unloading Dredgers

5.3. Main layout


The layout of the barge unloading suction dredgers is quite simple. The hull consists of a simple
rectangular pontoon, usually anchored by spuds at the ends (Figure 5-16Figure 5-16). Centrally
in the pontoon the pumps (dredge and jet pump) and engine room are located. Furthermore fuel
and water tanks and storage rooms are situated in the pontoon. The control of the dredger is done
from a cabin at the side of the deck from which the suction operator has a good view on the
alongside moored barge. Present accommodations are also situated above decks (Figure 5-15 and
Figure 5-16). Instead of spuds the barge unloading dredger might be moored on wires. Suction
pipe, discharge pipe are supported by booms or A-frames. The jet pipe or nozzle by hydraulic
cylinders to control the direction of the jet water.

Figure 5-15 BUD Rozkolec

Figure 5-16 View of the BUD Rozkolec

Page 13 of 21

Wb3408b

Designing Dredging Equipment

Figure Figure 5-17 shows the top view of the BUD Sliedrecht 14 and Figure 5-18 the side view
of the same dredger.

Figure 5-17 Top view of BUD Sliedrecht 14

Figure 5-18 Side view of BUD Sliedrecht 14

Figure Figure 5-19 shows a barge unloading dredger that can be used as a plain suction dredger
too.

Page 14 of 21

Chapter 5 Barge Unloading Dredgers

Figure 5-19 BUD Hercules

Page 15 of 21

Wb3408b

Designing Dredging Equipment

5.4. Technical construction

The pontoon volume in cubic


meters is almost 2.5 times the
light weight in tons (Figure
5.13).
Length of width have values
between 4 and 4.5 while width
over draught have values
between 3 and 6.

1400
y = 0.2496x
R2 = 0.7486

Light wieight [ton]

1200
1000
800
600
400
200
0
0

1,000

2,000

3,000

4,000

5,000

6,000

Total installed power [kW]

Figure 5-20 Light weight versus installed power.


3500
y = 2.4534x

3000

R = 0.8951

2500
BLD [m3]

5.4.1.
The hull
The main dimensions length,
width and depth of the pontoon
depend
totally
on
the
requirements for the above
mentioned design parameters
and the from these following
demands for stability and
strength.
The light weight of the pontoon
in tons is roughly 25 % of the
total power installed (Figure
5.12)

2000
1500
1000
500
0
0

200

400

600

800

1000

1200

1400

Light weigth [tons]

Figure 5-21 pontoon volume versus light weight


L/B

B/T

7
6
L/B and B/T

The fuel and water tanks are


distributed such over the
pontoon that a good trim of the
ship is obtained.
The winches for hauling the
barges during the suction
process are located on the deck.
The barge unloading suction
dredger is in general equipped
with spuds for anchorage.

5
4
3
2
1
0
0

200

400

600

800

Light weight [tons]

Figure 5-22

Page 16 of 21

1000

1200

1400

Chapter 5 Barge Unloading Dredgers

Figure 5-23 Plain Suction and BargeUunloading Dredge Seeland


Besides plain suction dredgers Figure 5-1 and Figure 5-23 also cutter suction dredgers can be
converted into a barge unloading dredger. ( Figure 5-24), although the last conversion will be
more expensive.

Figure 5-24 The CSD VICKSBURG converted to a Barge Unloading Dredger

Page 17 of 21

Wb3408b

Designing Dredging Equipment

5.4.2. The pipelines


The suction pipe that sticks out of the construction on
the side where the barges are moored, must on the one
hand be located as low as possible for the pump
process and on the other hand be high enough to let
the empty barges through underneath. The lower part
of the suction pipe, the haul pipe, runs parallel and
approximately in the centerline of the barge. This part
can rotate around a horizontal axis by way of a
rotation gland mounted in the horizontal part of the
suction pipe. Since this construction causes a under
pressure in the suction pipe during dredging, a lot of
attention must be given to the air tightness of the
piping. The necessary movability of the suction pipe is
obtained by hanging this pipe in a boom with a hoist
cable. For good movability the suction pipe can swing
in a horizontal plane by a hinge mounted in the
suction tube. (Figure 5-25)
The suction mouth is in general widened to obtain a
lower height of the suction mouth with a similar
opening surface. This reduces the chance of sucking in
air. (Figure 5-9)

Figure 5-25 Movable suction tube

The supply of the necessary dilution water to the barge is done with one or two water nozzles. In
case of one nozzle the suction mouth is usually equipped with jets, while the movability of the
main nozzle is than so large that it can also spray behind the suction mouth. To present sand well
to the suction mouth it is necessary to have moveable water nozzles. This is done using hydraulic
cylinders.
For the dredging process the pressure side of the jet pump is, except for the water nozzles, also
connected with suction side of the dredge pump.

5.4.3. The shore connection


The connection of the dredger to the shore needs to be flexible at all times, due to the movements
of the barge unloading suction dredger by:

trim during dredging


difference in draught by supplies
tides or water levels
hits of the barges against the dredger

Page 18 of 21

Chapter 5 Barge Unloading Dredgers

Figure 5-26 Shore connection for a barge unloading suction dredger

The shore connection must therefore consist of enough hinges. A flexible hose can also possibly
give enough flexibility, if this doesn't get stuck on the slope of the embankment. For large
differences extra attention must be paid to this movement (Figure 5-26).

5.5.

The dredging process

The dredge process is a hydraulic


transport process with a clear nonstationary character as a result of
the exchange of the barges. After
all this results that on regular
intervals the production reduces to
nil.
In Figure 5-27 the concentration
and the sand pump speed and jet
pump flow rate are shown as
function of time.
The first phase is characterized by
an increasing concentration during
the process to bring the suction
mouth to the bottom of the barge.
During the second phase the
concentration is approximately
constant. The barge is hauled
under the suction pipe with
constant velocity. The last phase
consists
of
a
decreasing
concentration because the suction
mouth reaches the end of the
barge, resulting in a decreasing
face height

Conc.

Time

Speed
dredge pump

Time

Qjet

To
dredge pump

To jet nozzle

To
dredge pump

Time

Figure 5-27

This phase is lengthened if the barge have to be cleaned. (The barge is pulled back and the
remaining sand is dredged.) Such a process might be necessary when the barge is relative wide
compare to the suction mouth and the suction mouth cant swing in the horizontal plane.

Page 19 of 21

Wb3408b

Designing Dredging Equipment

The production is determined by the breachebility of the sand in the barge and the erosion by the
jet water. This dredging process is mainly determined by the minimum NPSH value on top of the
suction pipeline and the time necessary to change the barge and to start the dredging process
again, as mentioned above.
A complication however is that during the emptying of the barge the sand pump flow rate
corresponds to the jet pump flow rate and the amount of sucked up sand. If this is not the case
than the flow rate in the barge will raise or drop. In a good tuned up process the suction operator
maintains the water level in the barge by hauling the barge slower or faster underneath the suction
mouth. If there is a continuous increase or decrease of the water level in the barge than the
number of revolutions of the sand pump must be adjusted. To obtain the highest possible
concentration the water level in the barge must be as high as possible. Unfortunately the
breaching of the sand behaves different under water than above water. If the water level in the
barge is high the dredge master cant see if sand flows behind the suction mouth and prefers a low
water level in the barge.
During the exchange of the barges the velocity in the pressure piping needs to be maintained to
avoid sanding up. For this the suction side of the sand pump can be connected to the weed box
(figure 5.7). This is not directly necessary. Since the pressure side of the jet pump is in
connection with the suction side of the dredge pump a situation with two pumps in series is
obtained. The required sand pump flow rate can now be reached by the control of the number of
revolutions of the sand pump engine.

pressure
Pipeline resistance
for mixture

Dredge pump curve for


mixture at high rpm

W
D

Dredge pump curve for


water at high rpm

E
Jet pump curve II

Jet pump curve I


Pipeline resistance for water

Dredge pump curve for


water at low rpm

QC

QA QB

Capacity

Figure 5-28 Pump pipeline interaction for a barge unloading system

Page 20 of 21

Chapter 5 Barge Unloading Dredgers

In Figure 5-28 the pipe and pump characteristics are drawn for the pumping of water and mixture.
If it is desired to maintain the minimal flow rate QA during the exchange of the barges, than this is
possible, when the suction side of the sand pump is connected with the weed box, by reducing the
number of revolutions regularly. This makes the operating point W shift to A. Without a
reduction of the number of revolutions of the sand pump, in the last phase of the emptying
process, the operating point W will shift over the dotted line to point E, so to a reasonable higher
flow rate. If the suction side of the sand pump is connected to the pressure side of the jet pump
than the operating point will be in A or B for the same low number of revolutions of the sand
pump and dependent on the pump characteristics of the jet pump.
For the calculation of the hydraulic process one can refer to the course Wb3414 Dredging
Processes 2.

Page 21 of 21

Chapter 6 Bucket (Ladder) Dredger

6. The bucket dredger

Figure 6- 1
6.

The bucket dredger ----------------------------------------------------------------------------------------------------1


6.1.

General Considerations ------------------------------------------------------------------------------------------2

The bucket dredger is one of the mechanical dredgers.----------------------------------------------------------------2


6.2.

Area of application------------------------------------------------------------------------------------------------3

6.3.

The history ---------------------------------------------------------------------------------------------------------4

6.4.
The method of working -------------------------------------------------------------------------------------------5
When a bucket dredger is working the anchoring system plays an important role in both positioning the
dredger in the cut and in the excavation by the buckets.------------------------------------------------------------5
6.5.

The design ----------------------------------------------------------------------------------------------------------6

6.5.1.

The production capacity --------------------------------------------------------------------------------------7

6.5.2.

The dredging depth --------------------------------------------------------------------------------------------7

6.5.2.1. De maximum dredging depth ---------------------------------------------------------------------------------8


5.2.3. The soil ---------------------------------------------------------------------------------------------------------- 10
5.2.4. The transport of the dredged material------------------------------------------------------------------------ 10
5.2.5. The main drive -------------------------------------------------------------------------------------------------- 10
Page 1 of 25

Wb3408b

Designing Dredging Equipment

For small dredging depths this may increase to a factor 4!--------------------------------------------------------- 11


5.2.6. The winches ----------------------------------------------------------------------------------------------------- 13
6.6.

5.3. The general layout ----------------------------------------------------------------------------------------- 15

6.7.
5.4. The technical construction -------------------------------------------------------------------------------- 15
5.4.1. The hull ---------------------------------------------------------------------------------------------------------- 15
5.4.2. The main gantry ------------------------------------------------------------------------------------------------ 16
5.4.3. The bucket ladder ---------------------------------------------------------------------------------------------- 16
5.4.4. Dredge buckets ------------------------------------------------------------------------------------------------- 18
5.4.5 The ladder gantry------------------------------------------------------------------------------------------------ 20
5.4.6 The main drive--------------------------------------------------------------------------------------------------- 20
5.4.7 The winches------------------------------------------------------------------------------------------------------ 21
6.8.

5.5 The stability-------------------------------------------------------------------------------------------------- 21

6.9.

5.6. The dredging process -------------------------------------------------------------------------------------- 21

This is the machine characteristic.--------------------------------------------------------------------------------------- 24

6.1.

General Considerations

The bucket dredger is one of the mechanical dredgers.


A bucket dredger is a stationary dredger that is equipped with a continuous chain of buckets, which
are carried through a structure, the ladder (Figure 6- 2. This ladder is mounted in a U-shaped pontoon.
The drive of the bucket chain is on the upper side. The bucket dredger is anchored on six anchors.
During dredging, the dredger swings round the bow anchor by taking in or paying out the winches on
board. The buckets, which are filled on the underside, are emptied on the upper side by tipping their
contents into a chute along which the dredged material can slide into the barges moored alongside.
The chain is driven by the so called upper tumbler at top of ladder frame, which is connected either
via a belt to the diesel or directly to an electro motor or hydro-motor.

Figure 6- 2

Page 2 of 25

Chapter 6 Bucket (Ladder) Dredger

Since 1960, bucket dredgers ( also called bucket line dredge(r) or bucket chain dredge(r)) that were
much used before the Second World War, have been almost entirely replaced by Backhoe dredgers or
trailing suction hopper dredgers and cutter suction dredgers. The reason for this is that the bucket
dredger, with its six anchors, is a big obstacle to shipping. Moreover maintenance costs are high and
the bucket dredger requires many highly skilled operatives. But above all, their production has not
kept pace with the increase in scale that has taken place in the suction dredgers.

6.2. Area of application


Bucket dredgers are only used in new or maintenance dredging projects when the initial depth of the
area to be dredged is too shallow for trailing suction hopper dredgers and the distances involved are
too long for hydraulic transport. For environmental projects, which require the dredging of in situ
densities, the bucket dredger is suitable peace of equipment.
When dredging for construction materials such as sand and gravel, or for minerals such as gold and
tin ores, bucket dredgers are still frequently used.
Bucket dredgers also come in a variety
of types. For example:
Dredgers with or without the
means of propulsion (Figure 6- 3)

Dredgers with a conveyor belt


system (Figure 6- 4)

Dredgers with equipped with


pumps

Figure 6- 3

Figure 6- 4

Page 3 of 25

Wb3408b

Designing Dredging Equipment

The maximum dredging depth is highly dependent on the size of the dredger. There are dredgers with
a maximum dredging depth of more than 30 metres. Such large dredgers the minimum dredging depth
is often 8 metres. Dredging in shallow water is certainly not the strongest point of the bucket dredger.
Bucket dredgers can be used in almost every type of soil, from mud to soft rock. When rock has been
fragmented by blasting, bucket dredgers are often used, because of their relative lack of sensitivity to
variations in the size of the stones.
Bucket dredgers cannot be used in areas with waves and swell. Furthermore, because of the amount of
noise they produce, in urban areas they are often subject to restrictions in relation to the working time
or the permitted number of decibels measured at a specific distance from them.
The capacity of a bucket dredger is expressed in terms of the content of the buckets. The capacity of a
bucket can vary between 50 and 1200 litres. Rock bucket dredgers often have a double set of buckets,
the small rock buckets and the large mud buckets. This is in order to make better use of the power of
the dredger and to widen the range of its use.

6.3. The history


From a historical point of
view, the bucket dredger
derives from the mud mill
that was invented in the
Netherlands in 1589. In the
earliest days this mill was
powered by a treadmill driven
by manpower.
(Figure 6-5)
In 1622 the drive system was
improved and horses could
replace the men. Around the
beginning of the 19th century
the first steam driven bucket
Figure 6- 5 MUD Mill Dredging Museum at Sliedrecht
dredgers came into existence.
Still, it was not until the second half of the century that steam dredgers had replaced those powered by
horses. Over the course of the years preceding 1915, both the power of the dredgers and the capacity
of the buckets increased. There was no further increase after that time. The great advantage of the
bucket dredger is that it can attain a reasonable production in most types of soil from soft clay to soft
rock.
For this reason, by about 1900 the bucket dredger had grown to be the most important type of dredger
in Europe; a position that it maintained until just after the Second World War. The two last steam
powered bucket dredgers were built in the Netherlands in 1956. At the end of the fifties and beginning
of the sixties, because of the big increase in the tonnage of oil and ore tankers, large deep-water ports
were needed. This led to large dredging contracts, which created a need for bigger production units
that, moreover, could dredge to a greater depth. Increasing the capacity of bucket dredgers is no
longer the solution because deeper dredging with larger buckets leads to a very heavy bucket chain.
Stationary suction dredgers and cutter suction dredgers could solve this problem in a considerably less
expensive way. Besides their bigger production capacity, these suction dredgers also have the
advantage that their maintenance costs are much lower. For these reasons buckets dredgers are now
only used for the types of work mentioned above.

Page 4 of 25

Chapter 6 Bucket (Ladder) Dredger

6.4. The method of working


When a bucket dredger is working the anchoring system plays an important role in both
positioning the dredger in the cut and in the excavation by the buckets.

Figure 6- 6 Positioning of the dredger in the cut


As mentioned previously, the dredger swings round the bow anchors (Figure 6- 6) The bow wire has a
length of 1 to 2 times the bucket capacity in litres. This means that for large dredgers it may be 1 to 2
km long. It will be clear that with such great lengths, measures must be taken to prevent the radius of
the swing circle from being reduced by the bow wire being dragged over the bottom. Over water,
therefore, one or more pontoons/floats/bow barges are positioned under the bow wire. If the bow wire
runs mainly over land it is placed on a drum roller.
Page 5 of 25

Wb3408b

Designing Dredging Equipment

The swinging of the dredger and the provision of the excavation forces is mainly carried out by the
side winches. The side winch velocity used depends on the type of soil and also on the step length
and the height of the cut. For the most effective possible transition of forces the side wires must make
an angle with the bow wire that is a little smaller than 90 .
When swinging round the bow anchor the swing angle () that the dredger makes with the swing
circle (Figure 6- 6), must be kept as constant as possible. The choice of the swing angle is related to
the clearance between the buckets on the lower part of the chain over the bottom or the slope. If this is
not done it is possible that the bucket chain will run off the bottom tumbler as a result of the lateral
forces that act on it. At the beginning of a new cut the swing angle is brought to the desired value as
quickly s possible. If there is a current in the dredging area the swing angle must be kept as large as
possible, that is at 90. The stern winch controls the swing angle. The stern anchor is used to obtain
the required tension in the bow wire. When dredging in tidal waters the stern anchor is usually used as
a flood anchor if the winch and the wire are strong enough for this.
The step length, the cut thickness and the swing velocity along the cut determine the amount of soil
that is cut per unit of time. This amount must be at least in balance with the number of buckets per
unit of time multiplied by the capacity of the buckets. In other words the bucket capacity and the
bucket speed are related to the factors mentioned above, Some dredgers have more than one type of
bucket, so that, depending on the soil type, the capacity can be adapted to the expected production.
Because with high excavation forces the dredger will not be able to completely fill the buckets, so that
they are partly filled with water. This is of course not economical.
The position of the ladder, particular the ladder angle, also affects the maximum filling degree of the
buckets. If the bucket rim is not horizontal, fluid soil will partly flow out of the bucket.
After being carried upwards, the buckets are turned upside-down as they pass over the upper tumbler
or the pentagon and, depending on the time, the material will fall out of the buckets. In order to
accommodate to this time effect the discharge chute into which the dredged material falls, is
adjustable in relation to the upper tumbler. Depending on the type of soil, extra measures may be
necessary to promote the emptying of the buckets.
From the discharge chute the material slides directly into the barge that is moored alongside the
dredger or it is transported to it via conveyor belts. To obtain the most even possible filling of the
barge it must be frequently warped along the side of the dredger.

6.5. The design


When designing bucket dredgers the following design parameters are important:

Production capacity

Dredging depth (minimum and maximum)

Soil type

The discharge of the dredged material (barges or via pipeline)

As previously mentioned, the bucket dredger can be used in all types of soil from clay to soft rock
which has not been blasted and hard rock which has been fragmented by blasting. The type of soil to
be dredged has a big influence on the design and the construction of the dredger. Considerable forces
arise during the dredging of rock. For all types of soil it is necessary to know the required cutting
capacity and the energy that is needed to transport the dredged material via the bucket chain to the
upper tumbler.

Page 6 of 25

Chapter 6 Bucket (Ladder) Dredger

6.5.1.

The production capacity

The production capacity of a bucket dredger cannot be increased indefinitely. Increasing the
production capacity of bucket dredgers implies increasing the bucket capacity. This means that the
forces in the bucket chain resulting from the weight of the buckets and links themselves is also greatly
increased. This in turn demands an even heavier construction. The production capacity of bucket
dredgers therefore seldom rises above 100.000 m/week. The same goes, to an even greater degree, for
the dredging depth, because greater dredging depths demand longer bucket ladders and thus more
buckets.
In principle, the product of the bucket capacity and the bucket velocity determines the production
capacity, thus: Qb = I b vb ; with: Q the production capacity in m3/s, Ib the effective volume of the
bucket and vb the bucket speed in buckets per second.
The maximum bucket size is 1200 litres and the maximum bucket velocity approximately 30 buckets
per minute or .5 buckets per second. Often this bucket velocity can only be reached with empty
buckets. With full buckets and when some excavation force is needed, the bucket velocity is quickly
reduced to values of 15 to 20 buckets per minute. Moreover factors such as the filling rate of the
bucket and the bulking factor of the soil play a part.
For a bank height h [m], a step size s [m] and a lateral or swing speed vs [m/s], the insitu production
Qs dredged is:

Qs = h s vs [m/s]
This insitu production must be in balance with the bucket production Qb corrected for the filling
degree FDb and the bulking factor B, thus:

Qs = h s vz =

I b FDb vb
B

Note: The filling degree FDb<1 and B>1


Because it is impossible to fill every bucket for 100% it is advisable to take as first assumption the
filling degree a value of 0.85 and bulking factor depending on the soil to be dredge:
Type of soil
Very soft silts and clay
Clay
sand
Rock

6.5.2.

Bulking factor
1.05
1.3-1.5
1.05- 1.25
1.3-1.4

The dredging depth

As with other dredgers both the maximum and minimum dredging depths are very important in
relation to the use of the dredger. Requirements in relation to these values are closely related to
market demands. The difference between the maximum and minimum dredging depth determine the
change of the angle of the bucket rim with the horizon.

Page 7 of 25

Wb3408b

6.5.2.1.

Designing Dredging Equipment

De maximum dredging depth

For large bucket dredgers the maximum dredging depth is about 25 m. and exceptional 30 m. By
adjusting the height of the mounting of the ladder on the ladder gantry or by lengthening the ladder, it
is possible to dredge to a maximum depth of 35 m (see 5.4.3). It will be apparent that by adjusting the
setting of the ladder or lengthening it, the number of buckets will increase. The figure below gives a
general view of the dredging depths used. For the smaller bucket dredgers the dredging depth is
around 10 m.

maximum ladder depth [m]

Dredging depth
40
35
30
25
20
15
10
5
0

Normal
extended

0.0

0.2

0.4

0.6

0.8

1.0

Bucket capacity [m3]


Figure 6- 7

Page 8 of 25

1.2

1.4

1.6

Chapter 6 Bucket (Ladder) Dredger

5.2.2.2. The minimum dredging depth


The minimum draught is, on one hand, determined by the required clearance including
*navigational/keel clearance and, on the other hand, by the *filling degree of the buckets at the
minimum dredging depth. In Figure 5.4 below, the maximum draught of the bucket dredger is shown
as a function of the bucket capacity. From the graph it can be seen that for bucket dredgers with a
bucket capacity of 300 litres the minimum dredging depth must lie between 3 and 4 metres.

4
3.5
3
2.5
2
1.5
1
0.5
0
0.2

0.4

0.6

0.8

1.2

1.4

Bu c ke t c a p ac ity [m 3 ]

5. 1
With small dredging depths, depending on the ladder angle, because the buckets are tilted so far back
*the filling degree may well be so low so that dredging in this situation becomes uneconomic.
In the figure below (Figure 5.5), the *filling degree of the buckets is given as a function of the
maximum dredging depths. The shape of the buckets is such that the maximum filling degree is
obtained at the maximum dredging depth. Naturally the buckets can also be designed for the average
dredging depth.

5. 2

Page 9 of 25

Wb3408b

Designing Dredging Equipment

5.2.3. The soil


The influence of the soil to be dredged is seen in the power of the upper tumbler, the strength of the
ladder, links and buckets and also in the bucket capacity and shape. If a bucket dredger is equipped
with buckets for both soft soil and rock, the capacity of the rock buckets is roughly 60 to 70% of that
of the soft soil buckets. Naturally, the length of the links must be the same for both types of bucket.
The length of the link must be the same. (Why?)
Moreover rock buckets are usually cast and soft soil buckets are often welded.
5.2.4. The transport of the dredged material
Usually barges that are loaded while moored alongside the dredger are used to transport the dredged
material. The height of the main gantry must be such that the soil falling from the buckets can slide
down into the barges moored alongside via the chute.
5.2.4.1. The bucket dredger with a pipeline discharge system
Sometimes the dredged material is carried away directly. In these cases it is collected in a hopper and
mixed with the right amount of water to be transported by means of a dredged pump and pipeline. As
in the case of a cutter suction dredger, the floating pipeline is attached to the stern of the dredger.
Naturally a barge with a dredge pump can also be moored alongside the dredger for this purpose. This
option is increasingly rarely used; indeed, unless the work stipulates the use of a bucket dredger the
contractor will employ the much cheaper cutter suction dredger.

5.2.4.2. Discharge by conveyor belts

5. 3
Conveyor belts are frequently used to discharge the dredged material when excavating sand and
gravel for the cement industry. This type of discharge system can be easily fitted to the normal bucket
dredger.
The conveyor belts are mounted on floats that are attached to the stern of the dredge. Because no
discharge chutes are used the main gantry can be lower.
5.2.5. The main drive
The choice of the source of power for the drive of the bucket chain is now limited to a diesel with a
direct belt drive, a diesel-electric drive or a diesel-hydraulic drive. When electricity can be obtained
from landlines, for example during sand or gravel dredging, it is also possible to use an electric drive.
The power/energy needed for the excavation, lifting of the soil, the friction of the buckets over the
guiding rollers and the tumblers, the friction of the tumblers, resulting from tension in the bucket
chain are transferred to the upper tumbler via the bucket chain.
The required cutting power can be determined in a way similar to that described for the cutter suction
dredger. Thus with the aid of the specific energy. If the desired cutting production is Qs and the
specific cutting energy Es, the required cutting power is:
Ps = Q s E s
(5.1)
Page 10 of 25

Chapter 6 Bucket (Ladder) Dredger

The required cutting power must be multiplied by a factor the represents the relation between the
average and peak loads.
When lifting the soil the number of buckets under or above water plays a role. Since:

P0 = Q e g ( e w )H bw + z H ow
With:
Qe =
g =
e =
How =
Haw =

the bucket production


acceleration due to gravity
the density of the dredged material in the bucket
the dredging depth
the height above water that the soil must be lifted.

(5.2)
[m3/s]
[m/s2]
[kg/m3]
[m]
[m]

In principle, the cutting production cannot exceed the production of the bucket chain, thus:
I E v
Qs e v e = Qe
(5.3)
B
Here:
Ev = the bucket *filling
ve = the bucket velocity
Ie = the bucket capacity
B = the bulking factor

If it is assumed that the quotient Ev is equal to 1 and Qs=Qe, the power required to lift the soil, is
known. With a filling degree lower than 1 the weight of the water above the soil must also be
included.
Because the number of buckets that goes upwards is equal to the number of buckets that goes
downwards it is not necessary to take into account lifting the weight of the buckets themselves.
Naturally the friction of the guide rollers over which the buckets slide must be taken into account.
The effect of the tensile forces also makes an extra contribution to the required drive power, with the
exception of the friction in the bearings of the lower tumblers.
To calculate the reactions and the tensile forces see Section 5.7
The total power required is thus:
Pt = Ps + Po + Pwl + PwT
Pt
Ps
Po
Pwl
PwT

=
=
=
=
=

(5.4)

the power to be installed


the cutting power
the lifting power
the friction power/work of the guide rollers/pulleys
the friction power/work of the tumblers

The friction forces that, as described above, can arise are the cause of the fact that the gross energy
requirement to lift the soil with a ladder angle of 45, are roughly two times as high as the nett energy
requirement. At small dredging depths this can increase to a factor 4!
So the relation between the length of the lower/under-bend of the bucket chain and the length of the
ladder has a big influence on the horizontal force (Figure 5.7).
For small dredging depths this may increase to a factor 4!
Thus the relation between the length of the lower bend and the ladder S/L a big influence upon the
horizontal tensile force (Figure 5. 5)

Page 11 of 25

Wb3408b

Designing Dredging Equipment

5. 4

As a guideline it can be assumed that the installed power in kW for the drive of the chain in soft soil is
roughly 1/2 and for heavy soil at 2/3 of the bucket capacity in litres. (Figure 5. 6)

1200
1000
800
600
400
200
0
0

0.2

0.4

0.6

0.8

Bu c ke t c a p a c ity [m 3 ]

5. 5

Page 12 of 25

1.2

1.4

Chapter 6 Bucket (Ladder) Dredger

5.2.6. The winches

5. 6 Simplified diagram of a barge loading bucket dredger

The winches on a bucket dredger have various functions and therefore various requirements with
regard to the power, the forces and band velocity, which differ from winch to winch.

5.2.6.1. The ladder winch


The ladder winch (letter i in Figure 5. 7), which is used to adjust the required dredging depth is
usually mounted on the ladder gantry of the larger bucket dredgers, while the smaller demountable
dredgers usually have the ladder winch mounted on deck.
Owing to the great weight of the ladder and the buckets this is the strongest winch on the bucket
dredger. The installed power is often in the order of magnitude of of the bucket drive. The ladder
winch velocity is roughly between 6 and 10 m/min. Currently the drive is usually a slow running
electric or hydraulic engine. Because of the need to set the dredging depth it is necessary to have an
adjustable winch.
5.2.6.2 The bow side winches

As in the cutter suction dredger, the side winches (see Figure 5. 7) make a major contribution to the
excavation process. The installed bow side winch power is between 10% and 20% of the main drive.
The side winch velocity of the bucket dredger is generally lower than that of the cutter suction
dredger. Nominal side winch velocities lie between 10 and 15 m/min. It will be clear that the
excavation process requires a winch that can be well controlled and adjusted. The control must be
such that any desired velocity can be set and remain as constant as possible, even when side winch
forces vary.
As in the cutter suction dredger, when paying out, the wire being loosened must be kept under control
by braking while paying out. The winches are mounted on the fore deck.

Page 13 of 25

Wb3408b

Designing Dredging Equipment

.
5.2.6.3 The stern side winches
The stern side winches have a secondary function and do not determine the production. The stern
winches control the dredger with regard to the cut (swing angle , (see Figure 5.2). The requirements
relating to the control and force are thus considerably less than for the bow side winches. The power
is roughly half that of the bow side winches. The nominal side winch velocities are of course equal.
The stern side winches are usually mounted on the afterdeck. To avoid hindering the arrival and
departure of barges, as well as the warping of the barges alongside the dredger, the side wires are led
down to a sufficient depth directly beside the dredger in vertical guides, also called wire spuds (Figure
5. 8).

5. 7 The wire spud construction

5.2.6.4 The bow winch


The bow winch is used to pull the dredger forwards when a new cut is started. The required force for
this lies in the same order of magnitude as for the side winch. The required velocity, however, is
considerably lower (nominally 2 - 3 m/min). Higher speeds are, of course, necessary when positioning
the bow anchor.

5.2.6.5. The stern winch


The function of the stern winch is to ensure the required tension in the bow wire. This consideration
demands that the required force is roughly equal to that of the bow wire, however, the need to move
the bucket dredger backward quickly to the adjacent cut places higher demands on the velocity (5-10
m/min).

Page 14 of 25

Chapter 6 Bucket (Ladder) Dredger

6.6.

5.3. The general layout

The hull consists of a U-shaped pontoon with long forward pontoons. The dimensions of the pontoon
are primarily determined by the required dredging depth and the necessary stability. The well is rather
long compared to that of a cutter suction dredger, roughly 60 % *of the length of the dredger.
The pontoon is divided into a number of compartments for the engine room, crew accommodation,
stores, and fuel and ballast tanks. The latter are often located for and aft in the pontoon. The engine
room is located in the pontoon aft of the main gantry and its layout depends on the type of main drive.
To satisfy the need for longitudinal stability the bottom of the dredger slopes upward at the stern or
the forward end may be wider (Figure

5. 9.). The main gantry is roughly in the middle of the pontoon. Although formerly the crew quarters
were often located in the pontoon, in modern dredgers they are now often situated on deck.

5. 9 Tekening IHC

6.7.

5.4. The technical construction

5.4.1. The hull


The hull consists of a U-shaped pontoon with almost horizontal deck and bottom plates. Often the
bottom plate slopes up at the stern to ensure the correct longitudinal weight distribution of the ship.
The corners of the pontoon are rounded off to make it easier for the barges to come alongside.
Page 15 of 25

Wb3408b

Designing Dredging Equipment

5.4.2. The main gantry


Because of the way in which the buckets are emptied and the need to load barges that are moored
alongside, the main gantry is high and heavy. The construction of the main gantry is often carried
through to the bottom ribs.
In modern bucket dredgers the drive of the tumblers is mounted on the main gantry. The *stort
wagons/fixed chutes are located on each side of the main gantry. They catch the dredged material
from the buckets that have been turned over by the tumbler and convey it to the movable chutes,
which discharge into the barges,
5.4.3. The bucket ladder

5. 10 Bucket ladder of the demountable bucket dredger Big


Dalton

The vertically rotating upper end of the ladder is suspended from two axle boxes which are mounted
on the sloping legs of the main gantry (Figure 5. 11). If necessary, these axle boxes, which are
attached by bolts, can be moved along the legs of the main gantry in order to dredge more deeply.
When they are in the lowest position it is necessary to add an auxiliary ladder to support the bucket
guides. /If they were in the lowest position the upper part of the bucket guides would come to be
suspended in the air. To prevent this from happening an auxiliary ladder is added. The shape of
the auxiliary ladder is such that the bucket chain is also carried over the upper part. (Figure 5. 12) and
is suspended at the lower end via the ladder wire which runs from the ladder gantry.

Page 16 of 25

Chapter 6 Bucket (Ladder) Dredger

5. 11

The weight of the full buckets is transferred to the ladder by rollers. These rollers are mounted at a
distance of twice the link length apart. To guide the buckets these rollers are fitted with flanges, hence
the name *ladder rollers/guide rollers
De bucket *chain/leiding is driven by the upper tumbler (often five-sided) and pulled round the
underside by the lower tumbler (often six-sided). As a rule of thumb the total tensile force exercised
by the upper tumbler on the bucket chain is 700 kN per 100 litre bucket capacity. The weight of the
descending buckets that form a chain provides the tensile force in the tumblers. *These tensile forces,
are dependent not only on the ladder angle, but also on the relation between the arc and the chord,
which generally amount to 1.1 to 1.15 and if necessary can be changed by adding or removing
buckets. /These tensile forces, excepting the ladder angle are dependent on the relation between the
arc and the chord, which generally amount to 1.1 to 1.15 and if necessary can be changed by adding
or removing buckets. See Section 5.7. **NB not included in Dutch version).
Summarising, the following forces act on the ladder:
1. The weight of the ladder itself, including the guide rollers.
2.

The weight of the bucket chain, including the links and bolts.

3.

The weight of the contents of the buckets.

4.

The tensile forces generated in the under bend.

5.

The excavation forces in both longitudinal and transverse directions if necessary multiplied by a
factor for impact loading.

Page 17 of 25

Wb3408b

Designing Dredging Equipment

5.4.4. Dredge buckets


Dredge buckets may be either welded or cast. Welded buckets are most often used on small dredgers
or dredgers that are suitable only for soft types of soil. The buckets are either welded onto the links or
cast as one unit with the links. The weight is then very high; 30 to 40 times the bucket capacity in kN.
For welded buckets the weight is 13 to 15 times the bucket capacity.
The front of the upper edge of the buckets is equipped with a cutting edge or with cutting teeth
(Figure 5. 14). The latter are most often found on rock buckets.

The shape of the bucket is always a compromise.

Because a good shape for excavation and the required strength do not give the optimum content.

The shape of the buckets is also determined by the required swing force (Figure 5. 13).

The theoretical filling degree, the amount of water that the bucket can contain in relation to the
total bucket capacity, is highly dependent on the dredging depth (Figure 5. 5).

A bucket shape from which the soil readily falls is equally difficult to combine with a good
excavation shape.

The price of the bucket.

5. 12

Rock buckets are small heavy buckets, somewhat egg-shaped, which must be able to resist impact
loads. Soft soil buckets, termed mud buckets, are much bigger and lighter. The relation rock bucket
capacity to mud bucket capacity lies between 60 and 70 %.
The so-called *pan buckets have good soil discharging properties; their disadvantage is that the
*filling degree is very sensitive to the angle of the bucket.

Page 18 of 25

Chapter 6 Bucket (Ladder) Dredger

5. 13

The links are fastened to each other by bucket bolts. The holes in the links, through which the bucket
bolts pass are equipped with wearing bushes, termed, bucket bushes. These are forged steel
*bushes/sleeves that are hydraulically pressed into the link. This simple means of attachment makes
these bucket bushes very prone to wear and so they must be frequently replaced. (Figure). *The
lubrication of the guide rollers and tumblers is now carried out centrally. Nowadays *caterpillar tracks
are sometimes used instead of links and bushes (Figure 5. 15).

Page 19 of 25

Wb3408b

Designing Dredging Equipment

5. 14 Undercarriage van Caterpillar

5.4.5 The ladder gantry


The ladder gantry straddles the outer end of the well. On it are found:

The ladder winch that is used to set the dredging depth.

The control cabin of the dredge master. From this it is now possible to operate all the winches.

The crane.
The free height of the ladder gantry is determined by the height required to rotate the entire ladder
above water.
Because of the large well, in order to give sufficient stiffness to the dredger the ladder gantry
construction must be very heavy.

5.4.6 The main drive


Although in the past many steam powered dredgers were built, nowadays the choice is limited to:

Diesel-direct driven via belt

Diesel-electric drives.

Diesel-hydraulic drive.

Direct power supplies from the shore; sometimes used for sand and gravel extraction.

This means that the upper tumbler may be electric or driven by a hydraulic engine.
In steam powered dredgers or those powered by diesel engines with a direct drive the energy is
transferred to the upper tumbler by driving belts.
The control of the revolutions of the upper tumbler and thus of the bucket velocity is simple when
using the above mentioned modern control systems. With an upper tumbler that is directly driven by a
diesel engine control is limited and switchable or hydrodynamic gears are needed.
The drives of auxiliary equipment such as winches and chutes present no problems when modern
drives are used.

Page 20 of 25

Chapter 6 Bucket (Ladder) Dredger

5.4.7 The winches


5.4.7.1 The ladder winch
Because of the great weight of the ladder two wires are usually used to hoist it. For this purpose the
winch drum is grooved on both sides in such a way that when the ladder is raised the wires are on the
outer sides of the drum (Figure 5. 16).

5. 15

5.4.7.2 The bow winch


With the aid of the bow winch the dredger is held against the cut. This winch also serves to pull the
dredger forward to the following cut during stepping. The revolution speed of this winch is very
important.
When moving the bow anchor this winch is paid out. Bow winches may be mounted above or below
the deck. Because of the great length of the bow wire the bow winch has a very large drum.

5.4.7.5 The auxiliary winches


Separate winches are used to operate the discharge chutes and for the warping of the barges.
A jib crane is needed to lift out stones and debris that has been dredged, and also when changing the
buckets during repairs. The winches used by this crane must satisfy the stipulations that apply to
lifting cranes.

6.8.
5.5 The stability
Under working conditions the stability of the bucket dredger is seldom in question. After all, the
greatest weight is always under water.
If the ladder is raised, however, the situation is entirely different. The great weight of the ladder is
then entirely above water.
For this reason, when a bucket dredger is being towed at sea it must be unrigged. The entire bucket
chain must be dismantled and, if possible, stowed below deck.

6.9.
5.6. The dredging process
The dredging process of the bucket dredger includes only the excavation and lifting of the dredged
material. Barges carry out the transport.

Page 21 of 25

Wb3408b

Designing Dredging Equipment

As previously mentioned, the bucket dredger swings on the bow anchor along the arc of a circle
*following a curving path. The axis of the dredger makes an angle , the swing angle with the tangent
to this arcuate path. The size of the swing angle depends primarily on the clearance between the lower
bend and the bottom and on the slope of the breach/bank.
At the end of the cut the dredge master will allow the swing angle to slowly increase to 90. After this
a step will be taken or, if necessary, the cutting of the following layer will be started.
By means of this movement back and forth, the bucket dredger makes concentric arcs/curves that lie
at a distance of one step length from each other. During this swinging back and forth the dredge
master closely observes/keeps an eye on the tension in the bow wire and the loading of the bucket
chain. The tension of the bow wire is controlled with the aid of the stern winch.
The amount of soil that is cut per unit of time depends on:

The thickness of the cut. This is the thickness of the layer that can be dredged in one swing.

The step length; the forward motion of the dredger during one swing.

The warping velocity of the dredger along the cut.

To prevent spillage, the cutting production must be less than or equal to the product of the bucket
velocity and the bucket capacity.
The cutting thickness depends on the total thickness of the layer to be dredged. If this is not too thick,
generally less than 5 m, the dredge master will try to dredge it in a single cut. If the layer exceeds 5 m
thick the entire breach/bank will be dredged by making several cuts. In any case the first cut must be
so thick that the dredger can create sufficient draught for itself.
The step length is roughly equal to the length of the links. As rule of thumb, 0.6 to 0.8 times the cube
root of the bucket capacity may also be taken. For both cases the swing velocity must be sufficiently
high (> 5m/min).
The warping velocity selected is such that either the buckets are full with a minimum spillage or that
the loading on the bucket chain is the limiting factor.
If possible, a width of the cut is selected that is so wide that the total width of the work can be covered
in one swing. The wider the cut the fewer the anchor movements. If that is not possible the total width
is divided into a number of equal cutting widths.
There is also a minimum cutting width for every bucket dredger. The required depth for the dredger
and the space for manoeuvring the barges play a role in determining this (Figure 5.2). This is roughly
1.5 times the length of the bucket dredger.
The dredging depth also determines the position of the buckets on the ladder and thus for the *filling
degree.
The available excavation energy of a bucket dredger is highly dependent on the energy needed to
carry/lift up the dredged material. This depends on:
1. The nett weight of the bucket contents. Part of this is under water and part is above water. The
weight of the buckets themselves plays no role because there is an equal number of buckets under
and above the ladder.
2.

The friction resistance in the ladder/guide rollers results from the weight of the buckets and their
contents.

3.

The friction resistance in the axles of the tumblers results from the tensile forces of the bucket
chain.

4.

The impact loads that develop as a result of the bumping of the buckets.
Page 22 of 25

Chapter 6 Bucket (Ladder) Dredger

The cutting production of the buckets is:


Qs = h s v

[m/s]

with:
h = cutting thickness usually < 5m
s
= step length
v = swinging velocity

[m]
[m]
[m/s]

(5.5)

The cutting production must balance with the amount that can be transported by the buckets per unit
of time thus:
I E v
Q
Q s = hsv = e v e = e
[m3/s]
(5.6)
B
60 B
Ie = bucket capacity
[m]
ve = bucket velocity ev
[buckets/min]
[-]
Ev = filling degree
B = bulking factor
[-]
[m/s]
Qe = bucket production
On the basis of the specific energy concept, the cutting energy for this production is:
I E v
Psnij = Q s E sp = e v e E sp
(5.7)
60 B
The energy needed to lift sand and water is:
I E v
Popv = e v e g ( e w )H ow + e + (1 E v ) w ) H bw
60B

e
w
Ee
How
Hbw

=
=
=
=
=

(5.8)

density of the soil in the bucket


density of water
bucket filling
lifting height under water
lifting height above water

[kg/m]
[kg/m]
[-]
[m]
[m]

If the friction in the ladder/guide rollers and tumblers is assumed to be a linear function of the weight
and the velocity then:

Pwr =

Q e A( n e , )v e
60

Ie E v ve
v
A( n e , ) e
60B
60

(5.9)

Here A ( n e , ) is the influence of the friction force on the ladder/guide rollers and the tumblers. Thus
here the influence of the tensions is *taken into account /verdisconteerd.

The total power required is thus:


Ptot = Psnij + Popv + Pwr

Ptot =

Ie E v ve
60B

(5.10)

ve

E sp + gB ( e w )H b + e + (1 E v ) w ) H o + BA( n e , )
60

(5.11)

Because the installed power must be higher than the average required power, it must be true that:
Page 23 of 25

Wb3408b

Designing Dredging Equipment

Ptot = Pinst w

(5.12)

Here w is the relation between the average and the peak power.
The relation between installed power and production is therefore:
Pinst =

IeE vve
60wB

ve

E sp + gB ( e w )H b + e + (1 E v ) w ) H o + BA( n e , )
60

(5.13)

If the bucket chain is driven by a top tumbler the relation between and ve is:
v e = 5n = 5

M = M

M=

60 150
=
2

(5.14)

ve
ve Ie E v ve
=
E sp B + g ( e w )H b + e + (1 E v ) w ) H o + A( n e , )
150
60B
60

ve
2.5I e E v
E sp B + g ( e w )H b + e + (1 E v ) w ) H o + A( n e , )
B
60

(5.15)
This is the machine characteristic.
When the drive characteristic is known, the bucket velocity and the associated torque are known and
thus the production.

5. 16

The filling degree is determined by the equation:


60hvsB
Ev =
Ie ve

(5.16)

So, for a given step length and cutting thickness the desired warping velocity is also known.
As long as Qe>=Qs is valid the spillage during cutting will be limited. The spillage that occurs during
the turning of the buckets is an entirely different question. Here factors such as cohesion, adhesion,
the shape of the buckets and the position of the fixed chute all play a part.
Page 24 of 25

Chapter 6 Bucket (Ladder) Dredger

Cohesive soil and also fine sands can give great problems on this point. In principle, this is a problem
of timing. Although the fixed chute is indeed adjustable, the range over which it is adjustable is
closely linked with the dredging depth and the shape of the lower bend. With soil that is not easily
loosened the bucket velocity must be reduced, as otherwise there will be too much spillage behind the
dredger. Measures are also taken to get rid of the under-pressure, which develop in the buckets when
discharging cohesive soils.
As with the barge-loading dredger/reclamation dredger, a situation may also arise in which the supply
of barges is the limiting factor. This situation may be caused by many different factors, such as:

Weather and wave conditions

Shipping movement

Bridges and locks

Differences in the speed of the barges.

Differences in the size of the barges.

Delays of the barge

Delays of the *reclamation dredger/barge unloading dredger

Delays at the discharge site

Clearly, with a bucket dredger, there is always a chance that sometimes there will be no barge
available.
Because the above mentioned delays can be reasonably well estimated with regard to their average
values and standard deviations, the Monte Carlo Simulation can provide insight into the probability of
delay resulting from the absence of barges. Clearly, when using a barge-loading dredger there is
always a chance of delays due to the absence of a barge.

Page 25 of 25

DREDGE PUMPS

DREDGING ENGINEERING
Wb 3413
CONTENTS:

Pumps and Systems

1.

INTRODUCTION

2.

DEFINITIONS

3.

SET OF PUMP CHARACTERISTICS

3.

SET OF PUMP CHARACTERISTICS

4.

EULERS EQUATION FOR CENTRIFUGAL PUMPS

4.1.

VELOCITY DISTRIBUTION BETWEEN THE BLADES

4.2.

THE EXISTENCE OF SLIP.

12
13

5.

CORRECTION ON THE THEORETICAL CHARACTERISTICS.

16

6.

DIMENSIONLESS PUMP CONSTANTS (SIMILARITY CONSIDERATIONS) 18

7.

AFFINITY LAWS:

19

VARIATION OF EFFICIENCY

21

8.

DIMENSIONLESS PUMP CHARACTERISTICS

23

9.

SPECIFIC SPEED

24

10.
INFLUENCE OF ENGINE CHARACTERISTIC ON THE PUMP
CHARACTERISTICS

29

10.1.

EQUATION OF THE CONSTANT POWER LINE.

30

10.2.

CONSTANT TORQUE LINE.

32

10.3.

VARIABLE TORQUE LINE.

34

11.

CAVITATION

11.1.

NET POSITIVE SUCTION HEAD (NPSH)

36

11.2.

THE DELIVERED (or produced) NPSH OF A PUMP

37

11.3.

RELATION BETWEEN (NPSH)d AND DECISIVE VACUUM

38

35

_______________________________________________________________________________________
Page 2 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

INFLUENCE OF DENSITY AND VISCOSITY ON THE PUMP CHARACTERISTICS FOR


12.
NEWTONIAN FLUIDS
39
12.1.

Fluids having the same viscosity but another density than water.

39

12.2.

Fluids having the same density but another viscosity than water. (Stepanoff 1967)

39

13.

INFLUENCE OF SOLIDS ON THE PUMP CHARACTERISTICS.

42

13.1.

PUMP CHARACTERISTICS FOR MIXTURES

14.

INFLUENCE OF SOLIDS ON CAVITATION

44

15.

PUMP PIPELINE COMBINATION

46

15.1.

PUMPING AT CONSTANT SPEED

48

15.2.

PUMPING AT CONSTANT TORQUE OR POWER

49

16.

RELATION BETWEEN PRODUCTION PUMPING DISTANCE

51

17.

SERIES OPERATION:

53

17.1.

THE LOCATION OF THE BOOSTER

18.

PARALLEL OPERATION OF PUMPS AND PIPES

18.1.

PUMP CHARACTERISTICS OF PARALLEL OPERATION

57

18.2.

PARALLEL PIPELINES

60

19.

INFLUENCE OF WEAR ON THE PERFORMANCE OF PUMPS.

19.1.

WEAR AT THE SUCTION INLET

62

19.2.

WEAR AT THE OUTLET.

62

19.3.

WEAR AT THE LINING PLATES

62

19.4.

WEAR AT THE CUTWATER

63

20.

BIBIBLIOGRAPHY

ENCLOSURE A

42

55

56

62

63
64

_______________________________________________________________________________________
Page 3 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

1. INTRODUCTION
Centrifugal pumps are particular suitable for pumping solids due to a small number of moveable
part.
More advantages of this pump type are:

a continuous pump capacity

the possibility of a direct drive

relatively cheap and maintenance friendly


Centrifugal pumps (dredge pumps) as used in the dredging industry are to distinguished by
"ordinary water pumps" by:

a large bore in the impeller as well as in the pump casing, without any restriction in the
direction of the flow.
at the impeller inlet the bore is most small

a small number (3, 4 or 5) and short vanes in the impeller as a compromise between a
large bore and an efficient pump action

a large clearance between the cutwater (Dutch puntstuk) and the impeller (10 to 20% of
the impeller diameter

An easy replacement of wear parts

the use of gland water for flushing the space between the impeller shrouds and the
wearing plates on the pump cover, in order to prevent particles to enter the shaft seals

1
10

Passage at
cutwater

12

PASSAGE IN DREDGE PUMP

_______________________________________________________________________________________
Page 4 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

3D VIEW OF DREDGE PUMP

PUMPROOM VIEW

_______________________________________________________________________________________
Page 5 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

2. DEFINITIONS
CAPACITY Q:
The volume liquid pumped per second; dimension [m/s]
MANOMETRIC PRESSURE p m :
The total pressure which can be delivered by the pump, dimension [N/m],
is defined as:

pm = p p ps + g h p hs +

( v 2p v s2 )
2

EFFICIENCY:

Qpm
PUMPPOWER
100% or =
100%
ENGINEPOWER
P

_______________________________________________________________________________________
Page 6 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

3. SET OF PUMP CHARACTERISTICS

Pressure p [kN/m2]

pressure [kPa]
1000
800
600
400
200
0
0

0.5

1.5

2.5

3.5

4.5

3.5

4.5

3.5

4.5

Flow Q [m3/s]

Power P [kW]

Power [kW]
3500
3000
2500
2000
1500
1000
500
0
0

0.5

1.5

2.5
Flow Q [m3/s]

Chart Title

Efficiency [%]

100
80
60
40
20
0
0

0.5

1.5

2.5
Flow Q [m3/s]

speed [rpm]
Dimp [m]
Bimp. [m]

400

Dens [t/m3]

1.65

Power [kW]

3000

0.4

DEIRA BAY

_______________________________________________________________________________________
Page 7 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

4. EULERS EQUATION FOR CENTRIFUGAL PUMPS


The changes of momentum for rotating bodies is:

T=

d ( mvr )
dt

m = mass [kg]
v = rotational velocity [m/s]
r = radius [m]
T = torque [Nm]
t= time [s]

_______________________________________________________________________________________
Page 8 of 66

DREDGING ENGINEERING
Wb 3413

T=

Pumps and Systems

d
mr2 c2 cos 2' mr1c1 cos 1' ]
[
dt

For a stationary flow the mass: m = Q


Q = capacity [m/s]
= density [kg/m]
c1 and c2 are the absolute velocities and 1 and 2 true directions of the liquid particles.
If all losses in the pump are disregarded, the required power equals delivered power

P = T = Q [ r2 c2 cos 2' r1c1 cos 1' ] = Qpth


pth the theoretical delivered pump pressure [n/m] ;
Resulting in:

p th = r2 c 2 cos '2 r1c1 cos 1' = [u 2 cu 2 u1cu1 ]


With r = u the peripheral velocity of the impeller and c cos ' = cu the component of the
absolute velocity on the peripheral velocity.
_______________________________________________________________________________________
Page 9 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

When expressed in the vane angles 1 and 2 the equation becomes.

uc
u c
pe = u22 u12 2 r 2 1 r1
tan 2 tan 1

Q
Q
and cr 2 =
2r1b
2r2b

Q u2
u1
pe = u22 u12

2b r2 tan 2 r1 tan 1

and because cr1 =

This is Euler's pump equation


If the liquid enters impeller without a tangential component thus radial for centrifugal pumps
then cu1 = 0 and Euler's equation becomes

uc
pe = u22 2 r 2
tan 2

Note:
pth is based on actual velocities and directions. Unfortunately those are in practice unknown.
therefore pe is based on the known velocities and vane angles.
For constant speed (u=constant) the equation reduced to: pe = A B Q
with A =

u2
u1

[ u22 u12 ] and B =

2b r2 tan 2 r1 tan 1

Which is an equation of a strait line.


Comform pe, pth can be written as: pth = A ' B ' Q
_______________________________________________________________________________________
c
Page 10 of 66

DREDGING ENGINEERING
Wb 3413
Because the power can be written as: P = pe Q = AQ B Q 2

Pumps and Systems

Which is the equation of a parabola.

When the liquid has prerotation before approaching the impeller eye the cu1<>0 and Eulers had
will be lower by the term u1cu1

_______________________________________________________________________________________
Page 11 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

4.1. VELOCITY DISTRIBUTION BETWEEN THE BLADES


Already stated true velocity angles 1' and 2' may not be the same as the blade angles 1 and 2 .
However the last ones are used in impeller design because it is easier to calculate flow velocities
based on those angles than the actual flow velocities.
Derivation of the fluid from the vane direction reduces the peripheral component of the absolute
velocity c u2 . This causes a reduction in head. This phenomenon is called slip and is a
consequence of the non-uniform velocity distribution across the impeller channel. The input
power keeps roughly the same because the capacity doesnt change.
Note:
1 = 1' no-shock condition at entry

2 = 2' no fluid slip at exit


The difference in head between those angles is called the head reduction factor .

pth cu2
=
pe cu2

cu2

u2

cu2

cu2
W2

W2

C2

Cr2

2
ideal flow

2
+ + ++
++++
+++

low pressure

++
++

||||
|||||
||||||

actual flow

high
pressure

Actual and ideal velocities at pump outlet


[Jonker 1995]

Slip velocity

cu2 is defined as: cu2 = cu2 cu2

_______________________________________________________________________________________
P Page 12 of 66

DREDGING ENGINEERING
Wb 3413

cu2
cu2 cu2
cu2
With =
this gives =
= 1
cu2
cu2
cu2

Pumps and Systems

4.2. THE EXISTENCE OF SLIP.


To transmit power to the liquid the pressure on the leading front of the vane should be higher
than on the back.
For any force exerted by the vane to the fluid has an equal and opposite reaction.
this means that the relative velocities at the back of the vane are higher than at the front.
This velocity profile in the impeller can be regarded as the through flow on which a relative eddy
is superimposed
Such a relative circulation can also be explained by the orientation of fluid particles through the
impeller

Fluid particles moving through the impeller fails to turn around their axes.
So the eddy has the same but opposite angular velocity as the impeller
These two flows cause that the direction of the flow at the outlet is inclined
Stodola has estimated

cu2 =

e
2

(mean velocity in the channel)

e = channel width at outlet and is:

e=

2r2 sin 2
z

z = number of vanes of thickness zero


So

cu2 =

r2 sin 2
z

u2 sin 2
z

_______________________________________________________________________________________
Page 13 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

FURTHER

cr2
tan 2

cu2 = u2

The relative eddy between impeller blades [Jonker 1995]

cr2

= is the component of the absolute velocity normal to the peripheral velocity

This results in

= 1

u2 sin 2

cr2

zu2
tan 2

with

uc
pe = u22 2 r 2
tan 2

gives

= 1

u2 sin 2
z

= 1

pe
u2

u22 sin 2
zpe

pth
pe

pth
pe

or

_______________________________________________________________________________________
P
Page 14 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

p
u22 sin 2
1 th =
pe
zpe
pe pth =

u22 sin 2
z

Being a line parallel with pe


Tests with 3 and 4 vane impeller do show a shift in the pressure curve.

4-Vane

3-Vane

700
600
500
400
300
200
100
0
0

0.5

1.5

2.5

Flow [m3/s]

Input power was the same for both impellers.


Another formula to calculate the slipfactor is proposed by Pfleiderer:
=

F I 2
1 + aG1 + J
H 60 K 1 br / r g

with a between 0.65 and 0.85 for volute type pumps and r1 and r2

respectively the radius at entrance and discharge.

_______________________________________________________________________________________
Page 15 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

5. CORRECTION ON THE THEORETICAL CHARACTERISTICS.


pressure

secundary losses, leakage


and recirculation
friction losses

shock
losses

Flow
1.
FRICTION LOSSES:
In pump and impeller friction loss is can be written as:

ph = c1Q 2

2.
SHOCK LOSSES:
Impact losses at the impeller blades because direction of flow differs from the blade angles. At
best efficiency point these losses are zero; so

ps = c2 ( Q Qs )

3.
SECONDARY LOSSES
Leakage, recirculation in pump casing
ACTUAL OR MANOMETRIC PRESSURE:

pm = pi c1Q 2 c2 ( Q Qs )

pm = A ' B ' Q c1Q 2 c2 ( Q Qs )


More general:

pm = A0 + A1Q + A2 Q 2

The equation is only valid for centrifugal pumps and not for axial flow pumps!

_______________________________________________________________________________________
P Page 16 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

EFFICIENCIES

HYDRAULIC EFFICIENCY:

h =

phydraulic
phydraulic + plosses

ACTUAL PUMP PRESSURE


THEORETICAL PUMP PRESSURE

VOLUMETRIC EFFICIENCY:
Q
Q
FLOW RATE TROUGH PUMP
=
=
Q + Qloss Qimp FLOW RATE TROUGH IMPELLER
MECHANICAL EFFICIENCY:

Q =

m =

Qi pth POWER SUPPLIED TO IMPELLER


=
P
POWER INPUT TO SHAFT

OVERALL EFFICIENCY:
Qp FLUID POWER DEVELOPED BY PUMP
=
=
p
SHAFT POWER INPUT

= h v m

_______________________________________________________________________________________
P
Page 17 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

6. DIMENSIONLESS PUMP CONSTANTS (SIMILARITY


CONSIDERATIONS)
Flows conditions in two geometrically similar systems are called similar if all fluid velocities
change with a constant ratio.
So in hydraulic machines similarity of flow requires a constant ration between fluid velocities
and peripheral velocities.

c
= constant
u

So

c m2 =

cm
Q
Q
Q
and u 2 = r 2 =
=
=
u2
2rb
2rbu 2 2rb r

Or
=

Q
nD
.(Db )

60

Full similarity is only obtained if the width b changes with the same ratio as D, so:

Q
nD 3

For similarity of the pressure


When p is devided by

= const u 2

u22 the term

p
p
=
==
2
u 2 2 r22

p
nD

60

= const

p
n D 2
2

Becomes dimensionless and is called the dimensionless pressure.

Dimensionless power can be defined as:

P
P

= const 3 5
=
3

n D
nD
( Db)

60

From the momentum follows that full similarity is only got when viscous effects do not change.
_______________________________________________________________________________________
Page 18 of 66

DREDGING ENGINEERING
Wb 3413
this is only the case for high Reynolds numbers!

Pumps and Systems

EULERS EQUATION:

Q u2
u1
pe = u22 u12

2b r2 tan 2 r1 tan 1

can now be rewritten in:

Q
u12
u

1
1

1
2
u2 2b u2 r2 tan 2 u2 r1 tan 1
u1 r1
because
=
u2 r2
1
r2
Q
1
= 1 12

tan 1
r2 2b u2 r2 tan 2

= 1

1
r12
1
= 1 2

tan 1
r2
tan 2

Comment: According Stepanoff

(page 80 and 168). Dit geeft r1=0.644r2


of r2=1.552r1

In case of radial flow into the impeller, this equation reduce to

= 1

1
tan 2

So the dimensionless Euler equation is only determined by the discharge angle

7. AFFINITY LAWS:
=

2 r22

p
nD

60

p1 n12 D12
=
=
p2 n22 D22

gives:

Q
Q
=
2
2 b r2 nD

. ( Db )
60
P
3

nD
(Db )
60

r12
= 0.585 = constant
r22

Q1 n1 D12
= =
Q2 n2 D22

gives:

gives:

1
=1
2

_______________________________________________________________________________________
P
Page 19 of 66

Comment: Page: 18
The following formulae are
not correct! The real values
to be changed with the non
dimensionless values.

DREDGING ENGINEERING
Wb 3413
Pumps and Systems
This condition is strictly only true for the impeller action and for the location of the best
efficiency point.
However it can be stated:
If pump tests of centrifugal pumps do not fulfil these laws, check the results or the measuring
devices.
If there is prerotation the affinity law regarded to the diameter is less applicable .
(see: dimensionless Eulers equation)

For variable speed and constant impeller diameter, lines of constant efficiencies are parabolas
going through the origin.
The condition =

1
= 1 is strictly only true for the impeller action.
2

The influence of the impeller casing results in an optimum speed with the highest efficiency,
however the best efficiency point at different speed are still located at a parabola through the
origin.

Location of efficiencies

400 rpm

1000

375 rpm

350 rpm

Location of equal efficiencies

800
600
400
200
0
0

0.5

1.5

2.5

3.5

Flow [m3/ s]

_______________________________________________________________________________________
P
Page 20 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems


400 rpm

375 rpm

350 rpm

2500
2000
1500
1000
500
0
0

0.5

1.5

2.5

3.5

3.5

Flow [m3/ s]

Efficiency [%]

400 rpm

375 rpm

350 rpm

90
80
70
60
50
40
30
20
10
0
0

0.5

1.5

2.5

Flow [m3/s]

Due to the flow in the volute there is a small deviation of this theory. instead of parabola of
constant efficiency it appeared to be more or less ellipses
BEP line

80%

VARIATION OF EFFICIENCY
______________________________________________________________________________________
P
Page 21 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

_______________________________________________________________________________________
Page 22 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

8. DIMENSIONLESS PUMP CHARACTERISTICS


For centrifugal pumps and to a lesser extend for half-axial flow pumps as well, the dimensionless
pump characteristics can be written as a power serie of the second degree
So for the pressure:

= 0 + 1 + 2 2
and for the power:

= 0 + 1 + 2 2
Dimensionles s Characteristics
Dimensioless Head

Dim_head= -7.9516x 2 - 0.1632x + 0.5034

0.6
0.5
0.4
0.3
0.2
0.1
0

0.06
0.05
0.04
0.03
0.02
0.01
0
0

0.05

0.1

0.15

Dimensioless Capacity

Dim_head
Dim_Power

0.2
Dim_cap= -0.5418x 2 + 0.3031x + 0.022

DIMENSIONLESS PUMP CHARACTERISTICS


Calculating the actual pump characteristics from the dimensionless gives for the pressure:
2

1
p
1
2

Db
+
Q

2 = 0 + 1
2
Q
nD
nD
nD
60

60 Db
60
or

nD 2
1 2 2
nD 1

Q + 2
Q
+ 1
p = 0
60 Db
Db
60

FOR THE POWER CHARACTERISTIC:

3
2

nD
nD
nD
1 2

(
)
P = 0 ( Db)

Db
Q
+
+
1
2

60
60
60

_______________________________________________________________________________________
P
Page 23 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

9. SPECIFIC SPEED
In the selection of pumps the discharge Q, the pressure p and the pump speed n are usually
known.
A dimensionless combination of these variables at the best efficiency point is known as the
specific speed:

ns =

( gH )

3
4

Comment: Page: 23

4 Q

( p)

3
4

In literature

ns =

n Q
3

( H) 4

is

frequently use, however this


number is not dimensionless!

The specific speed is used as a "type" number and to compare different impeller designs and
dimensions such as b/D and inlet over outlet diameters

D1
D2

By defining Q = 2br22 and p = 2 r22 in which and are based on the best efficiency
point.
3

ns =

4 r2 2 b
3
4

3
4

( r )

1.5

3
4

b
b
2 = 2ns'
D
D

or
1

Because for simualar impellers the ratio b/D is constant the ratio

number or another form of specific speed: ns' =

1
2
3
4

ns

b
2
D

3
4

can be used as a type

D
1
ns
2
b

An increase in specific speed requires a wider impeller and/or a smaller impeller.


A change in the diameter results in a shift of the specific speed.
Figure below shows typical impeller shapes with their specific speeds

_______________________________________________________________________________________
Page 24 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

HYDRO DYNAMIC ROTORS OF DIFFERENT SPECIFIC SPEEDS [JONKER 1995]


The pump types have different characteristics in a well-defined region of head en flow as shown
in the next graph.

Radial
Mixed flow
Axial

_______________________________________________________________________________________
P Page 25 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

EFFECT OF SPECIFIC SPEED ON IMPELLER SHAPE [JONKER 1995]


Experiments have shown that for each type of impeller shape the maximum efficiency is in a
narrow range.
In dredging practice only centrifugal and half-axial flow (mixed) pumps are used.
The first in all type of dredgers and the latter mainly as additional submerged pumps on board
of trailing suction hopper dredgers when equipped for dredging over the 50 m depth.
In that case low head and low head and high capacity is required.
Submerged pumps used on cutter dredgers or plain suction dredgers are mainly from the
centrifugal type. Because there head is mostly much more than required to pump the mixture to
the inboard pump. The additional head is used for overcome the pipeline resistance of the
discharge line.
Figure below shows specific head and capacity as function of specific speed of pumps used in
the dredging industry.

_______________________________________________________________________________________
Page 26 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems


All dredgers

0.16

0.8

0.7

0.6

0.5

0.08

0.4

0.3

Specific Head

Specific Capacity

0.12

0.2

0.04

Capacity
Head

0.1

0
0

0.2

0.4

0.6

0.8

Specific Speed

On basis of figure ??? the dimensions of the impeller and the pumpspeed can be determined.
Example:
Assume Q=2 m3/s and p=750 kPa; Determine pumpspeed and diameter.
For Ns= 0.3, and can be estimated from the graph above; =0.042 and =0.6.
The rotational speed r can be calculated from and impeller internal width from .
1

N s' =

0.3

3
4

r 2b

0.042

Q
b=
r 2
2
=
= 0.214 [m]
35.35 2 0.042

b g

wr

0.6

r =

750
1 * 0.6

= 1250 = 35.35 [m / s]

With the figures ???? the ratio b/D can be estimated.


Note that in these figure the specific speed is ns while in figure ??? this is ns'

_______________________________________________________________________________________
P
Page 27 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

_______________________________________________________________________________________
Page 28 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

10. INFLUENCE OF ENGINE CHARACTERISTIC ON THE PUMP


CHARACTERISTICS
At characteristic of electrical engines types one can distinct:

constant speed

constant power
variable torque

Note:
Constant power condition is also possible with diesel engines with special gearboxes (f.i. hydrodynamic)
For diesel engines this

constant speed

constant torque

_______________________________________________________________________________________
Page 29 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

10.1. EQUATION OF THE CONSTANT POWER LINE.

The equation of the actual power can be rewritten as:

P = A0 n 3 D 4 + A1n 2 D 2 Q + A2 nQ 2
with:

A0 =

4b
60

0 , A1 =

2
60

1 AND A2 =

60b 2

Are for a certain pump 0 , 1 , 2 , D and b given, then the pump speed can be determined as
function of the capacity q. (the solution of a cubic equation or numerical solution by Newton
Raphson)
Substituting the results in the pressure equation gives the so-called constant power line. (see
enclosure a)

_______________________________________________________________________________________
Page 30 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

Pressure p [kN/m2]

CHARACTERISTICS FOR CONSTANT POWER

1000
800
600
400
200
0
0

0.5

1.5

2.5

3.5

4.5

Flow [m3/s]

Power [kW]

2500
2000
1500
1000
500
0
0

0.5

1.5

2.5

3.5

4.5

3.5

4.5

Flow [m3/s]

Efficiency [%]

100
80
60
40
20
0
0

0.5

1.5

2.5
Flow [m3/s]

speed [rpm]
Dimp [m]
Bimp. [m]

400

Dens [t/m3]

1.65

Power [kW]

2000

0.4

DEIRA BAY

_______________________________________________________________________________________
Page 31 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

For a for a given pump speed and capacity, optimum impeller diameter can be determined by
solving the equation:

A0 n 3 D 4 + A1n 2 D 2 Q + A2 nQ 2 P = 0

this gives:

A1n 2 Q +

Dopt =

( A1n 2 Q) 2 4 A0n 3 ( A2 nQ 2 P)
2 A0 n 3

10.2. CONSTANT TORQUE LINE.

The same technique can be applied for the case of constant torque.
The torque can be written as:

T=
with

3
nD 2

0
nD
nD

Q + 2
( Db) 1 Q 2
(
)
Db

1

60
60

60
2n

60

or with the simplified equation as:

T=
or:

1
[ A n 3 D 4 + A1n 2 D 2 Q + A2 nQ 2 ]
2n 0
60

T = B0 n 2 D 4 + B1n D 2 Q + B2 Q 2

here in is:

Bn = An

30

The line of constant torque can be found by solving the equation:

B0 n 2 D 4 + B1n D 2 Q + B2 Q 2 T = 0
giving:

n=

B1 D 2 Q +

( B1 D 2 Q) 2 4 B0 D 4 ( B2 Q 2 T )
2 B0 D 4

_______________________________________________________________________________________
Page 32 of 66

DREDGING ENGINEERING
Wb 3413
CHARACTERISTICS FOR CONSTANT TORQUE

Pumps and Systems

pressure [kPa]

Pressure p [kN/m2]

1000
800
600
400
200
0
-200 0

0.5

1.5

2.5

3.5

4.5

3.5

4.5

3.5

4.5

Flow Q [m3/s]

Power [kW]

Power P [kW]

2500
2000
1500
1000
500
0
0

0.5

1.5

2.5
Flow Q [m3/s]

Chart Title

Efficiency [%]

100
80
60
40
20
0
-20 0

0.5

1.5

2.5
Flow Q [m3/s]

speed [rpm]
Dimp [m]
Bimp. [m]

400

Dens [t/m3]

1.65

Power [kW]

2000

0.4

DEIRA BAY

_______________________________________________________________________________________
Page 33 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

10.3. VARIABLE TORQUE LINE.

Is the available torque a function of the speed, such as in the case of electric motors, then
T = C0 C1n . In that case the solution is:

n=

( B1 D 2 Q C1 ) +

( B1 D 2 Q C1 ) 2 4 B0 D 4 ( B2 Q 2 C0 )
2 B0 D 4

_______________________________________________________________________________________
Page 34 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

11. CAVITATION
Cavitation is a condition in a liquid in which the local pressure has dropped below the vapour
pressure corresponding to the temperature of the water. (boiling)
Cavitation can occur at:

high points in a pipeline f.i. siphons


high velocities (Bernoulli)
large suction heights or long suction lines
high fluid densities.
high altitudes (reservoirs) or low atmospheric pressure

Results:
1.
Collapse of the vapour bubbles when they enter the high-pressure zone
2.
Drop of the manometric pressure- and efficiency curves
3.
Pitting and corrosion
In dredge pumps low pressure is on the entrance side and cavitation start between the vanes
Start of cavitation

Cavitation bubbles

Full cavitation

CAVITION BETWEEN THE VANES

_______________________________________________________________________________________
Page 35 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

11.1. NET POSITIVE SUCTION HEAD (NPSH)

The NPSH is defined as total (energy) head available to the pump above the vapour pressure in
front of the pump.

(NPSH )a =

p s pv v 2

+
[m]
g g 2 g

or

(NPSH )a = ps pv + 1 v 2
2

[Pa]

ps = absolute pressure in front of pump


pv = vapour pressure of liquid
v = velocity
This can be written as:

( NPSH )a = pa pv ghs L

pa = Atmospheric pressure
hs = suction height
= fluid density
L= all pipeline losses

[Pa]

[Pa]
[Pa]
[kg/m3]
[Pa]

VAPOR PRESSURE
LOSSES

NPSH
AVAILABLE

NPSH AT A SYPHON

_______________________________________________________________________________________
Page 36 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems


vapor pressure

NPSH available
1 2
v
2

psuction

Hydraulic losses

suction lift

NPSH IN FRONT OF THE PUMP

11.2. THE DELIVERED (or produced) NPSH OF A PUMP

The minimum NPSH delivered by a pump is a function of the capacity at which the pressure
drop due to cavitation with a certain value f.i. 5 % . It can only determined by testing the
pressure drop by trottling progressively the pump inlet.

5% REDUCTION IN PRESSURE

Flow

MINIMUM NPSH

Flow

MINIMUM NPSH AS FUNCTION OF CAPACITY

The pressure- and efficiency drop are measured as function of net positive suction head.
_______________________________________________________________________________________
Page 37 of 66

DREDGING ENGINEERING
Wb 3413
No cavitation if (NPSH)d < (NPSH)a

Pumps and Systems

To estimate (NPSH)d around the best efficiency point use can be made of
Specific NPSH number:

S =

( g NPSH ) 4

For dredge pumps S = 3 - 3.5


Because (NPSH)d is proportional with liquid velocity squared, it also means that NPSH u 2
and so with n
So affinity law:

( NPSH ) 1 n12
Q n
= 2 and 1 = 1
( NPSH ) 2 n2
Q2 n2

11.3. RELATION BETWEEN (NPSH)d AND DECISIVE VACUUM

( NPSH ) d =

pa
p
v2
v +
g g 2 g

pa = pb (Vac) d

(NPSH )d =

pb p v v 2

+
(Vac )d
g g 2 g

or

(ps)decisive

(Vac) d = ( NPSH ) d +

vapor pressure
NPSH
delivered

NPSH available
1 2
v
2

Hydraulic losses

psuction

Decisive vacuum

margin

1 2
v
2

atmospheric pressure

pb
p
v2
v +
g g 2 g

suction lift

Relation vaccum and NPSH


_______________________________________________________________________________________
Page 38 of 66

DREDGING ENGINEERING
Wb 3413
Pumps and Systems
12. INFLUENCE OF DENSITY AND VISCOSITY ON THE PUMP
CHARACTERISTICS FOR NEWTONIAN FLUIDS
12.1. Fluids having the same viscosity but another density than water.
The manometric pressure for a fluid other than water relates to that of water by:

p fluid = pwater

fluid
water

and for the power

Pfluid = Pwater

fluid
water

12.2. Fluids having the same density but another viscosity than water.
(Stepanoff 1967)

Due to the viscous effects affinity laws hold with less accuracy than for water,
capacity varies with speed. Because efficiency is mostly higher at higher specific speeds,
power increases less than the cube of the speed and the pressure more than the square of the
speed
When speed varies specific speed at the bep-points remains the same.
ns =

Q
3

p4

ns1

= 1
ns2 n2

Q1 p24 n1 n12 n22


=
=1
3
1
3
Q2 4 n2 2 2
p1
n2 n1

This relation stands irrespective of the deviation of the affinity laws.


At constant speed pressure curve decreases as viscosity increases in such a way that the specific
speed at "bep" remains constant

ns =

Q1
3
4
1

Q2
3
4
2

so for the same speed at different viscocities the relationship

FG IJ
H K

Q1
p
= 1
Q2
p2

3
2

Is valid.
At constant speed pressure curve decreases as viscosity increases, but head at zero capacity
remains the same. However the influence of the pump casing on the characteristics is higher
more than when pumping water.

o
_______________________________________________________________________________________
Page 39 of 66

DREDGING ENGINEERING
Wb 3413
Pumps and Systems
For constant viscosity and variable speed. Efficiency at "bep" increases at higher speeds. (higher
Reynolds numbers give less resistances so higher efficiencies.

Influence viscosity on pump performance (Stepanoff, 1957)


A change in Reynolds number due to a change in viscosity causes a change in the hydraulic
losses.
If
hydraulic losses are estimated for water the hydraulic losses for another viscosity can be
calculated according:

1 hydr .

fluid

= 1 hydr .

water

fluid
water

_______________________________________________________________________________________
Page 40 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

In which is the Darcy-Weisbach resistance coefficient?

pressure

secundary losses, leakage


and recirculation
friction losses

shock
losses

Flow

_______________________________________________________________________________________
Page 41 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

13. INFLUENCE OF SOLIDS ON THE PUMP CHARACTERISTICS.


Solids in suspension cannot posses or transmit any pressure energy. Solids can only acquire
kinetic energy. When a particle is accelerated the required energy is taken from the liquid phase.
When a particle is de-accelerated by the fluid, the kinetic energy is transformed to turbulent
energy from which only a part is transformed to pressure energy.

13.1. PUMP CHARACTERISTICS FOR MIXTURES

For homogeneous flows the required power is proportional with the density of the fluid.
(see page 8 P = T = Q r2 c2 cos 2' r1c1 cos 1' = Qpth )

Pmixture = Pwater

mixture
water

Solids transform their kinetic energy partially to pressure energy(potential)


According to Stepanoff:

pm = pw

m
f
w c

and because

Pm = Pw m it follows that:

m
= fc
w
and

f c = 1 Ccd [.8+.6 log( d 50 )]

_______________________________________________________________________________________
Page 42 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems


Influence of particle size on pump performance (Stepanoff)

1
0.9
0.8

Cvd=5 %

0.7

Cvd=10 %

0.6

Cvd=15 %

fc 0.5

Cvd=20 %

0.4

Cvd=25 %

0.3

Cvd=30 %

0.2

Cvd=35 %

0.1
0
0.1

10

100

d50 [mm]

Research in the laboratory of Dredging Technology TUD have shown the following:
For fine and medium sand efficiency is less than according Stepanoff but increase more than
linear at high concentrations
For course sand efficiency is lower than according Stepanoff
For fine and medium sand power is proportional with the density but for coarse sand the required
power increases strongly with delivered concentration.
A more general solution can be obtained with a distinction between the different effects.:
fp
p
P
m
= f and m w = f p m w =
w
pw m
Pw m f
Wilson has published a more generalised solids-effect diagram for slurry pumps. He concludes
that the sloids effect on pressure, efficiency, and power may be strongly influenced by the size of
the dredge pump (Scale effects).

_______________________________________________________________________________________
Page 43 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

GENERALISED SOLIDS EFFECT DIAGRAM BY WILSON


Note:
Wilsons consideration it is only based on limited experimental data.

14. INFLUENCE OF SOLIDS ON CAVITATION


In principal a negative influence.
The presents of solids in the flow will incept cavitation earlier.
Silt and clay can cause a higher vapour pressure.
However the most important aspect of pumping solids is the higher-pressure drop in vertical
lines due to the higher density.
As a consequence the decisive vacuum is reached earlier.
In order to avoid cavitation in suction lines there are in principle three possibilities:
1.
Reduce the concentration of the mixture.
2.
Put the pump (further) below the water level.
3.
Reduce the velocity

_______________________________________________________________________________________
Page 44 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

This can easily proved by the so-called vacuum formulae for homogeneous transport

1
1

v 2 = mengsel g( H k ) + mengsel v 2
2 mengsel
2
1
Vac = water gH + mengsel g( H k ) + mengsel v 2
2

water gH + Vac = mengsel ghz +

_______________________________________________________________________________________
Page 45 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

15. PUMP PIPELINE COMBINATION


When pumping water under a constant boundary conditions, there is only one operating point,
but
when the operating conditions are variable, there is a operating area.

OPERATING AREA DURING FILLING A WATER TOWER BASIN


Under constant speed condition
pumping through short pipelines
requires more power then pumping
through long lines.

_______________________________________________________________________________________
Page 46 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

When the operating point shifts to the constant power or constant torque line the engine speed
will decrease in order to avoid overloading of the motor.

For diesel engines this speed reduction is limited by the smoke limit.
This is the point where insufficient air is available for a complete combustion.
At lower speed the available torque will drop sharply and heavily polluted gasses are emitted
resulting in higher wear.
The position of the smoke limit depends mainly on the degree of supercharging.
Rule of thumb 90% of the nominal speed.
In case of normally aspirated engines speed drops of 60-70% of nominal speed are possible.
The allowable torque at speeds lower than at the smoke limit depends on the type of engine.
When the allowable torque results in a decreasing capacity with decreasing head the operating
point can easily come below the critical capacity resulting in a blockage of the pipe.
Installing an impeller with a smaller diameter is now the only solution to get a normal operating
condition.
As already said cavitation causes a drop of the manometric head. Working under high cavitation
condition can reduce the available pump pressure remarkable.

_______________________________________________________________________________________
Page 47 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

15.1. PUMPING AT CONSTANT SPEED


Pressure [kPa]
mixture
3
water

4
mixture

water

Flow [m3/s]
_______________________________________________________________________________________
Page 48 of 66

DREDGING ENGINEERING
Wb 3413
Pumps and Systems
For a pump-pipeline combination with a short suction line compared by the discharge line
( LSUCTION LINE << LDISCHARGE LINE ) the operating points are:
1.
When the complete line (suction and discharge line) are filled with water
2.
Suction line filed with mixture and pump and discharge line filled with water.
3.
The complete system filled with mixture
4.
Suction line and pump filled with water, discharge line with mixture.

15.2. PUMPING AT CONSTANT TORQUE OR POWER


The numbering is now clockwise
Pressure [kPa]

WATER
MIXTURE

4
3

MIXTURE

2
CONSTANT POWER
OR TORQUE LINES

WATER

Flow [m3/s]

_______________________________________________________________________________________
P Page 49 of 66

DREDGING ENGINEERING
Wb 3413
In case of operating area around the nominal torque point

Pumps and Systems

mixture
Pressure [kPa]
mixture
3

water

water
4

water
mixture

Flow [m3/s]

_______________________________________________________________________________________
Page 50 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

16. RELATION BETWEEN PRODUCTION PUMPING DISTANCE


In case of pump speed is maximum, the maximum output of solids per unit of time (production)
depends on the pumping distance.
Using the expression for empirical correlation for the pressure gradient between mixture and
water: =

Im I f
I m = I f (1 + Cvd )
Cvd I f

The pressure loss can be written as:

pl = AQ 2 1 +
with A =

B
3 Cvd
Q

w L

and b depending on the particle size and pipe diameter.

1120

25
00

manometrische pressure (kPa)

960

45
0
40 0 m
00
35
m
00
m
30
00
m

2 D 2 2
D
4

Because pl and q vary only slowly with Cvd so, l can increase if Cvd decreases.

800

20

m
50

27
rp m
270
rp m
260
rpm
250
pm
0r
24

640
480

0
23

rp m

0,0
0,0

at = 0,942
fc = 0,92
0,5

r pm

220

320
160

00

m
00
15

m
00
10

500

m
210 rp

3
Qcritical = 2,05 m /s
1,0

Mixture density = 1300 kg/m3

1,5

2,0

2,5

3,0

3
Capacity (m /s)

3,5

0m
4,0

4,5

5,0

PUMP-PIPELINE CHARACTERISTICS

_______________________________________________________________________________________
Page 51 of 66

DREDGING ENGINEERING
Wb 3413
Production [m3/hr]

Pumps and Systems

P-L diagram
I
Production

Capacity

Pipeline length [m]


Section I : Production is determined by other factors than pump or engine
Section II : Operating point at constant torque or constant power line
Section III: Operating pony at constant speed line

PRODUCTION-PIPELINE LENGTH DIAGRAM

_______________________________________________________________________________________
P Page 52 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

17. SERIES OPERATION:


Purpose of serie operation is:

dredging at greater depth.

pumping over greater distance.


From operation point of view there is hardly any difference between pumping with one pump or
with more than one pump. However the pumps should be designed for the same operation area.
For dredgers having more than one pump the first pump is in general a suction pump. (relative
low pressure and a high decisive vacuum)
The pump characteristics of pumps in series can easily be determined by super position of the
manometric pressure and the required power at a given capacity.
So:
N

n =1

n =1

pt = pn ( Q) AND Pt = Pn ( Q)
The total efficiency is defined as:
N

s =

Q pn ( Q)
n =1

P ( Q)
n =1

100%

_______________________________________________________________________________________
Page 53 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

Pressure [kPa]
pump 1=2

pump1

Efficiency [%]

Flow [m3/s]

Power [kW]

Flow [m3/s]

pump 1+2

pump 1

Flow [m3/s]

_______________________________________________________________________________________
Page 54 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

17.1. THE LOCATION OF THE BOOSTER

As long as the incoming pressure at the booster is sufficient positive and out coming pressure is
not too high for the pump and its component, then the location does not matter.

p2

max. allowable
pressure of pump 2
max. Pumping distance
Pressure p1

max pressure p1

p2max

Parallel lines

vacum

min. Pumping distance


Reclamation area
min. Input pressure
area where booster
can be placed

FIGURE: PRESSURELINES ALONG PIPELINE

_______________________________________________________________________________________
P
Page 55 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

18. PARALLEL OPERATION OF PUMPS AND PIPES


Parallel operation is used when a higher capacity is required.
Examples:

Trailing suction hopper dredgers.

PARALLEL SUCTION PIPES WITH CENTRAL DICHARGE SYSTEM

Special purpose vessel Cardium used during the delta works

SUCTION MOUTH CARDIUM

_______________________________________________________________________________________
Page 56 of 66

DREDGING ENGINEERING

Wb 3413
Pumps and Systems
Jet pumps systems.
Jetpump systems on board of trailing suction hopper dredgers have often the possibility
to work in serial and parallel operation.

Drink water works (variable demand)


Due to the variable demand parallel operation is normal in drinkwater supply

Parallel operation with to dredge pumps on one line is some times to be seen on board of trailing
suction hopper dredgers. The two dredgepumps deliver the mixture via one shute or discharge
pipe into the hopper.

18.1. PUMP CHARACTERISTICS OF PARALLEL OPERATION

The combined characteristics can be determined by super position of the capacities at a given
pressure.
This implies that the capacity is expressed as function of the pressure.
N

n =1

n =1

Qt = Qn ( p) AND Pt = Pn ( Q) = f ( Qt )
The total efficiency is:
N

s =

p Qn ( p)
n =1

Pt ( Qt )

100%

_______________________________________________________________________________________
P Page 57 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

Pressure [kPa]

PUMP 1+2
PUMP 1

Efficiency [%]
PUMP 1

PUMP 1+2

Flow [m3/s]
Power [kW]
PUMP 1+2

PUMP 1

Flow [m3/s]
_______________________________________________________________________________________
Page 58 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

_______________________________________________________________________________________
Page 59 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

18.2. PARALLEL PIPELINES

Parallel operation of dredge pumps on one line is only done in the dredging field when
reclamation areas have small fill heights. In that case the main pipeline is devided in two smaller
lines with an equal cross section.
Comfort with parallel operating pumps the pipeline characteristic can be determined by super
position of the capacities at a given pressure.
When different pipeline length are used beware of the critical velocity in the long line! .

_______________________________________________________________________________________
P
Page 60 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

If parallel or series operation is useful depends on pipeline characteristic as shown below.

_______________________________________________________________________________________
P
Page 61 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

19. INFLUENCE OF WEAR ON THE PERFORMANCE OF PUMPS.


Wear is mainly determined, except from the mineral composition of the grains, by the speed of
the mixture.
For pumps it is assumed that the wear is proportional with the third power of the peripheral
speed.
Therefore the peripheral speed is limited to 35- 40 m/s.
The performance of a pump changes as the sizes and shapes differ from the original ones.

19.1. WEAR AT THE SUCTION INLET

Wear at the inlet occurs when the pump is working at a capacity, which differs substantial from
the design capacity.
(shock losses)
The inlet geometry is decisive for the cavitation performance of centrifugal pumps.
So wear at the inlet results mostly in a reduction of the decisive vacuum.

19.2. WEAR AT THE OUTLET.

The manometric pressure is mainly determined by the geometry at the outlet.


Reduction of the impeller diameter due to wear will result in a decrease of the manometric
pressure.
Because the wear is proportional with the third power of the peripheral speed, more wear can be
expected when pumping over long distances.

19.3. WEAR AT THE LINING PLATES

_______________________________________________________________________________________
Page 62 of 66

DREDGING ENGINEERING
Pumps and Systems
Pumphouse

Entrance

Impeller

Wb 3413
Wear at the lining plates does increase the
clearance between the impeller and the wearing
plates, resulting in increase of the fluid
recirculated.
This will induce on its turn a higher wear and a
reduction of the efficiency.
(recirculation requires power)

Pumpshaft

Recirculation between impeller and


pumphouse
19.4. WEAR AT THE CUTWATER

Wear at the cutwater does increase the quantity of recirculation water in the pump casing.
However, compared to water pumps, dredge pumps do have a large cap between the impeller and
the pump casing at the cutwater.
So the influence of wear at the cutwater will decrease the efficiency slightly.

20. BIBIBLIOGRAPHY
4. Wilson, K.C., Addie, G.R. Sellgren, A. and Clift, R. (1997). Slurry transport using
centrifugal pumps, Blackie Academic and Professional.
5. Jonker, J.B. (1995). Turbomachines I, Lecture notes University of Twente, faculty
Mechnical engineering
6. Karrasik, I.J.K., Krutzsch,W.C., Fraser,W.H. and Messina, J.P., Pmp handbook,
McGraw-Hill Book Company, New York, 1976
7. Stepanoff, A.J. (1957). Centrifugal and axial flow pumps, John Wiley & Sons, Inc.
8. Stepanoff, A.J. (1965). Pumps and blowerstwo-phase flow, John Wiley & Sons, Inc.

_______________________________________________________________________________________
Page 63 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

ENCLOSURE A
THE CUBIC EQUATION FOR THE SOLUTION OF THE CONSTANT POWER LINE
GIVEN z + a2 z + a1 z + a0 = 0
3

LET

1
1
1
1
q = a1 a22 ; r = ( a1a2 3a0 ) a23
3
9
6
27

AND

s1 = 3 r + q 3 + r 2 ; s2 = 3 r q 3 + r 2

THEN IS IF:

q 3 + r 2 = c2 > 0;
ONE REAL ROOT AND A PAIR OF COMPLEX CONJUGATE ROOTS.

s1 = 3 r + c ; s2 = 3 r c AND BOTH REAL THEN:

z1 IS REAL AND z2 , z3 ARE COMPLEX

q 3 + r 2 = c2 = 0
ALL REAL ROOTS AND AT LEAST TWO ARE EQUAL.

s1 = s2 = 3 r s1 s2 = 0
q 3 + r 2 = c2 < 0
ALL ROOTS REAL

s1 = 3 r + ci ; s2 = 3 r ci AND ARE COMPLEX


1
+ 2 k
+ 2 k
s1 = ( r 2 + c 2 ) 6 cos
+ i sin
3
3

1
+ 2 k
+ 2 k
+ i sin
s2 = ( r 2 + c 2 ) 6 cos
3
3

SO

1
+ 2 k
s1 + s2 = 2( r 2 + c 2 ) 6 cos
3

1
+ 2 k

s1 s2 = 2( r 2 + c 2 ) 6 i sin
3

_______________________________________________________________________________________
Page 64 of 66

DREDGING ENGINEERING
Wb 3413

Pumps and Systems

c
WITH = arctan
r

RESULTING IN THE REAL ROOTS z1 , z2 , z3


FOR ALL CONDITIONS THE ROOTS z1 , z2 , z3 ARE:

z1 = ( s1 + s2 )

a2
3

1
( s + s ) a32 + i 23 ( s1 s2 )
2 1 2
a
i 3
1
(s s )
z3 = ( s1 + s2 ) 2
2
3
2 1 2
z2 =

AND

z1 + z2 + z3 = a2
z1 z2 + z1 z3 + z2 z3 = a1
z1 z2 z3 = a0

APPLIED TO THE CONSTANT POWER EQUATION z 3 + a 2 z 2 + a1 z + a 0 = 0

A1Q 2 A2 Q 2
P = A0 n D + A1n D Q + A2 nQ n +
n +
n P=0
A0 D 2
A0 D 4
3

WITH

A1Q
A2 Q 2
,
=
,a = P
a
A0 D 2 1 A0 D 4 0

a2 =
SO

1 AQ
A Q2 1 A Q
1 A Q2 A Q
q = 2 4 1 2 , r = 2 4 1 2 3P 1 2
9 A0 D
6 A0 D A0 D
3 A0 D
27 A0 D

AND
3

2
A Q 2 1 A Q 2 1 A Q 2 A Q
1 AQ
q + r = 2 4 1 2 + 2 4 1 2 3P 1 2
9 A0 D 6 A0 D A0 D
27 A0 D
3 A0 D
3

s1 = 3

2
2
A2 Q 2 1 A1Q 2 1 A2 Q 2 A1Q
1 A1Q
1 A1Q
1 A2 Q 2 A1Q

4
2 3P
2
4
2 3P
2 +
4
2 +
6 A0 D A0 D
9 A0 D 6 A0 D A0 D
27 A0 D
27 A0 D
3 A0 D

s2 = 3

2
2
A2 Q 2 1 A1Q 2 1 A2 Q 2 A1Q
1 A1Q
1 A1Q
1 A2 Q 2 A1Q

4
2 3P
2
4
2 3P
2
4
2 +
6 A0 D A0 D
9 A0 D 6 A0 D A0 D
27 A0 D
27 A0 D
3 A0 D

_______________________________________________________________________________________
P
Page 65 of 66

DREDGING ENGINEERING
Wb 3413
SUBSTITUTED IN z1 , z2 , z3 GIVES THE REQUIRED ROOTS.

Pumps and Systems

_______________________________________________________________________________________
P
Page 66 of 66

CECW-EH-D

Department of the Army

EM 1110-2-5025

U.S. Army Corps of Engineers


Engineer Manual
1110-2-5025

Washington, DC 20314-1000

Engineering and Design


DREDGING AND DREDGED
MATERIAL DISPOSAL

Distribution Restriction Statement


Approved for public release; distribution is
unlimited.

25 March 1983

ENGINEER MANUAL

ENGINEERING AND DESIGN

DREDGING AND DREDGED


MATERIAL DISPOSAL

DEPARTMENT OF THE ARMY


CORPS OF ENGINEERS
OFFICE OF THE CHIEF OF ENGINEERS

EM 1110-2-5025
25 March 1983

DAEN-CWE-HD

DEPARTMENT OF THE ARMY


U.S. Army Corps of Engineers
Washington, D.C. 20314

Engineer Manual
No. 1110-2-5025

EM 1110-2-5025

25 March 1983
Engineering and Design
DREDGING AND DREDGED MATERIAL DISPOSAL

1. Purpose. This manual provides an inventory of the dredging equipment


and disposal techniques used in the United States and provides guidance
for activities associated with new work and maintenance projects. This
manual further provides guidance on the evaluation and selection of
equipment and evaluation of disposal alternatives.
2. Applicability. This manual is applicable to all field operating
activities concerned with administering the Corps' dredging program.
3. Discussion. The engineering and design guidance discussed in this
manual is primarily for projects that have been authorized and are in the
preliminary design stages. However, much of the information is equally
applicable to the preliminary engineering and design required during the
authorization phase of dredging projects.
FOR THE COMMANDER:

DAEN-CWE-H
DAEN-CWO-H

DEPARTMENT OF THE ARMY


US Army Corps of Engineers
Washington, D. C. 20314

EM 1110-2-5025

Engineer Manual
No. 1110-2-5025

25 March 1983
Engineering and Design
DREDGING AND DREDGED MATERIAL DISPOSAL
Table of Contents
Subject

CHAPTER 1.

CHAPTER 2.

Paragraph

INTRODUCTION
Purpose and Scope-----------------------Applicability---------------------------Reference-------------------------------Bibliography----------------------------Background------------------------------Considerations Associated With Dredging
and Dredged Material Disposal----------

1-1
1-2
1-3
1-4
1-5

1-1
1-1
1-1
1-2
1-2

1-6

1-4

2-1
2-2
2-3
2-4
2-5
2-6
2-7

21
2-1
2-1
2-2
2-5
2-6
2-6

3-1
3-2
3-3
3-4
3-5
3-6
3-7
3-8
3-9

3-1
3-1
3-3
3-7
3-15
3-18
3-20
3-23
3-26

3-10

3-28

3-11
3-12
3-13
3-14

3-28
3-31
3-33
3-34

DESIGN CONSIDERATIONS
General---------------------------------Preliminary Data Collection-------------Dredging Locations and Quantities-------Physical Properties of Sediments--------Selection of Dredging Equipment---------Disposal Alternatives-------------------Long-Range Studies-----------------------

CHAPTER 3.

Page

DREDGING EQUIPMENT AND TECHNIQUES


Purpose---------------------------------Factors Determining Equipment Selection-Hopper Dredges--------------------------Cutterhead Dredges----------------------Dustpan Dredges-------------------------Sidecasting Dredges---------------------Dipper Dredges--------------------------Bucket Dredges--------------------------Special-Purpose Dredge------------------Summary of Dredge Operating
Characteristics-----------------------Locations of Dredges in the United
States--------------------------------Agitation Dredging Techniques-----------Advances in Dredging Technology---------Environmental Considerations-------------

EM 1110-2-5025
25 Mar 83
Subject
CHAPTER 4.

Paragraph

DISPOSAL ALTERNATIVES
Introduction-----------------------------

Section 1.

Section II.
Section III.

Section IV.

Section V.

Page

Evaluation of Dredged Material Pollution


Potential
Influence of Disposal Conditions on
Environmental Impact------------------Methods of Characterizing Pollution
Potential-----------------------------Sediment Resuspension Due to Dredging
Factors Influencing Dredging Turbidity--Open-Water Disposal
Behavior of Discharges from Various Types
of Dredges----------------------------Dredged Material Dispersion at the
Discharge Site------------------------Environmental Impacts in the Water
Column--------------------------------Environmental Impacts on the Benthos----Overview of Open-Water Disposal---------Confined Dredged Material Disposal
Containment Area Design-----------------Containment Area Operation and
Management----------------------------Productive Uses-------------------------Environmental Considerations------------Habitat Development as a Disposal
Alternative
General Considerations for Habitat
Development---------------------------Marsh Habitat Development---------------Upland Habitat Development--------------Island Habitat Development--------------Aquatic Habitat Development--------------

4-1

4-1

4-2

4-1

4-3

4-2

4-4

4-4

45

4-6

4-6

4-6

4-7
4-8
4-9

4-11
4-12
4-16

4-10

4-17

4-11
4-12
4-13

4-24
4-27
4-28

4-14
4-15
4-16
4-17
4-18

4-29
4-31
435
4-37
4-40

APPENDIX A.

BIBLIOGRAPHY

A-1

APPENDIX B.

CHECKLIST FOR REQUIRED STUDIES

B-1

ii

EM 1110-2-5025
25 Mar 83
CHAPTER 1
INTRODUCTION
1-1. Purpose. This manual provides an inventory of the dredging
equipment and disposal techniques used in the United States and provides
guidance for activities associated with new work and maintenance
projects. This manual also presents engineering and design guidance for
use on both new work and maintenance dredging projects. The guidance is
primarily for projects that have been authorized and are in the
preliminary design stages. However, much of the information is equally
applicable to the preliminary engineering and design required during the
authorization phase of dredging projects. This manual further provides
guidance on the evaluation and selection of equipment and evaluation of
disposal alternatives.
1-2. Applicability. This EM is applicable to all field operating
activities concerned with administering the Corps' dredging program.
1-3. References. The references listed below provide practical guidance
to Corps personnel concerned with dredging and dredged material disposal.
a. ER 1110-2-1300, Government Estimates and Hired Labor Estimates for
Dredging.
b.

ER 1110-2-1404, Deep Draft Navigation Project Design.

c.

EM 1110-2-1906, Laboratory Soils Testing.

d.

EM 1110-2-1907, Soil Sampling.

e. EM 1125-2-312, Manual of Instructions for Hopper Dredge Operations


and Standard Reporting Procedures.
f. WES TR D-77-9, Design and Construction of Retaining Dikes for
Containment of Dredged Material.
g.

WES TR DS-78-1, Aquatic Dredged Material Disposal Impacts.

h. WES TR DS-78-4, Water Quality Impacts of Aquatic Dredge Material


Disposal (Laboratory Investigations).
i. WES TR DS-78-6, Evaluation of Dredged Material Pollution
Potential.
j. WES TR DS-78-10, Guidelines for Designing, Operating, and Managing
Dredged Material Containment Areas.
k. WES TR DS-78-11, Guidelines for Dewatering/Densifying Confined
Dredged Material.
l-l

EM 1110-2-5025
25 Mar 83
l. WES TR DS-78-12, Guidelines for Dredged Material Disposal Area
Reuse Management.
m. WES TR DS-78-13, Prediction and Control of Dredged Material
Dispersion Around Dredging and Open-Water Pipeline Disposal Operations.
n. WES TR DS-78-16, Wetland Habitat Development with Dredged
Material: Engineering and Plant Propagation.
o. WES TR DS-78-17, Upland Habitat Development with Dredged Material:
Engineering and Plant Propagation.
p. WES TR DS-78-18, Development and Management of Avian Habitat on
Dredged Material Islands.
WES TR DS-78-21, Guidance for Land Improvement Using Dredged
q.
Material.
The WES Technical Reports referenced above are available from the
Technical Information Center, U. S. Army Engineer Waterways Experiment
Station, P. O. Box 631, Vicksburg, MS 39180.
1-4. Bibliography. Bibliographic items are indicated throughout the
manual by numbers (item 1, 2, etc.) that correspond to similarly numbered
items in Appendix A. They are available for loan by request to the
Technical Information Center Library, U. S. Army Engineer Waterways
Experiment Station, P. O. Box 631, Vicksburg, MS 39180.
1-5. Background. The Corps of Engineers has been concerned with the
development and maintenance of navigable waterways in the United States
ever since Congressional authorization was received in 1824 to remove
sandbars and snags from major navigable rivers. The Corp's dredging
program involves the planning, design, construction, operation, and
maintenance of waterway projects to meet navigation needs. The Corps'
responsibility includes developing and maintaining the Nation's waterways
and harbors, as well as maintaining a minimum dredging fleet to meet
emergency, national defense, and national interest dredging requirements.
The importance of the Corp's dredging program to the economic growth of
the country is suggested by the fact that the total waterborne commerce
of the United States continued its record-breaking advance during the
1970's. The viability of the economy of the United States is clearly
dependent upon maintenance of the waterways, ports, and harbors for
navigation. The Corp's annual dredging workload is approximately 287
million cu yd of material, including both maintenance and new work. The
Corps accomplishes the majority (70 percent in FY 81) of its annual
dredging workload by contracting privately owned equipment under
competitive bidding procedures; it performs the remaining work using
hired labor to operate Corps-owned dredges (item 5). An overview of the
Corps' dredging program is shown in figure l-l.

1-2

EM 1110-2-5025
25 Mar 83

1-3

EM 1110-2-5025
25 Mar 83
1-6. Considerations Associated with Dredging and Dredged Material Disposal.
Some considerations associated with dredging and dredged material disposal
are as follows:
a. Selection of proper dredge plant for a given project.
b. Determining whether or not there will be dredging of contaminated
material.
c.

Adequate disposal facilities.

d.

Long-term planning for maintenance dredging projects.

e. Characterization of sediments to be dredged to support an engineering design of confined disposal areas.


f. Determining the levels of suspended solids from disposal areas and
dredge operations.
g.

Disposal of contaminated sediments.

h.

Disposal in remote areas.

i.

Control of dredging operation to ensure environmental protection.

j.

Containment area management for maximizing storage capacity.

1-4

EM 1110-2-5025
25 Mar 83
CHAPTER 2
DESIGN CONSIDERATIONS
2-1. General. A dredging and dredged material disposal operation requires
consideration of both short- and long-term management objectives. The primary
short-term objective of a dredging project is to construct or maintain
channels for existing navigation needs but not necessarily to authorized
project dimensions. This should be accomplished using the most technically
satisfactory, environmentally compatible, and economically feasible dredging
and dredged material disposal procedures. Long-term objectives concern the
management and operation of disposal areas to ensure their long-term use.
This chapter outlines the design consideration usually needed to meet the
objectives of a dredging project.
2-2. Preliminary Data Collection. In order to gather the data required for a
dredging and dredged material disposal project, it is necessary to do the
following:
a. Analyze dredging location and quantities to be dredged, considering future needs.

2-3.

b.

Determine the physical and chemical characteristics of the sediments.

c.

Evaluate potential disposal alternatives.

d.

Identify pertinent social, environmental, and institutional factors.

e.

Evaluate dredge plant requirements.


Dredging Locations and Quantities.

a. Dredging locations and the quantities of material to be dredged are


two of the most important considerations in planning dredging projects. Since
disposal of dredged material is usually the major dredging problem, it is
essential that long-term projections be made for disposal requirements of each
project. Records should be kept of quantities dredged and maintenance
interval(s) to forecast future dredging and disposal requirements.
b. Hydrographic surveys are the principal dredged contract management
tool of the Corps. Hydrographic surveys should be made prior to dredging
to determine existing depths within the project area and after dredging
to determine the depths that were attained as a result of the dredging
operation. Each district should have the capability, either in-house or
by contract, to make accurate, timely, and repeatable hydrographic surveys.
To ensure accuracy, quantity calculations must be made from survey data
gathered in a timely manner using proper equipment and based upon precisely
established horizontal and vertical controls. Direct tide level readings
must be made at the site of the work to eliminate gross errors in quantity
calculations. Quantity measurement methods must be fully consistent

2-1

EM 1110-2-5025
25 Mar 83
between work per formed by contract and work per formed by hired labor.
2-4. Physical Properties of Sediments. In planning any dredging operation
which constitutes a specialized problem in earthmoving or excavation, it is
essential that field measurements and computations be made to determine the
location, characteristics, and quantities of material to be removed. The
characteristics of the dredged material determine dredge plant and, to some
extent, disposal requirements. Refer to Chapter 4 for specific
characterization tests required for evaluation and design of disposal
alternatives for dredged material.
a. Sampling. Sediment samples should be taken of the material above the
depth to which removal will be credited. This should be done concurrent with
the pre-dredge survey. For maintenance dredging of a recurring nature,
samples will be taken before each dredging until the characteristics of the
sediments are well known. For subsequent dredging, a small number of samples
will be taken to identify and changes in sediment characteristics. Normally
the sediment sampling depth will be the authorized project depth plus an
allowable tolerance (usually 2 ft) to compensate for the inherent inaccuracies
of the dredging process. The number of sediment samples taken should be
sufficient to obtain accurate information regarding the characteristics of the
material to be dredged. Samples in soft materials can be obtained by push
tube or grab samplers.
(1) Tube sampling.
(a) A tube sampler is an open-ended tube that is thrust vertically into
the sediment deposit to the depth desired. The sampler is withdrawn from the
deposit with the sample retained within the tube. Differences among tube
samplers relate to tube size, tube wall thickness, type of penetrating nose,
head design including valve, and type of driving force. Tube samplers (also
called harpoon samplers) are available with adjustable weights in the range of
from 17 to 77 lb and with fixed weights in excess of 90 lb. The amount of
weight required depends upon deposit texture and required depth of penetration.
(b) The split barrel sample spoon (also known as split-spoon sampler) is
capable of penetrating hard sediments , provided sufficient force is applied to
the driving rods. The sampler is thrust into the deposit by the hammering
force exerted on rods connected to the head. During retrieval, the sample is
retained within the barrel by a flap. The nose and head are separated from
the barrel in order to transfer the sample to a container. Refer to EM
1110-2-1907 for more information on soil sampling.
(2) Grab sampling. A grab sampler consists of a scoop or bucket
container that bites into the soft sediment deposit and encloses the sample.
Grab samplers are used primarily to sample surface materials, with depth of
penetration being 12 in. or less. Grab samplers are easy and inexpensive to
obtain and may be sufficient to characterize sediment for routine maintenance
dredging. Grab sampling may indicate relatively homogeneous sediment
composition, segregated pockets or coarse- and fine-grained sediment, and/or
mixtures. If segregated pockets are present, samples should be taken at
2-2

EM 1110-2-5025
25 Mar 83
a sufficient number of locations in the channel to adequately define spatial variations in the sediment character and quantities of each material.
(3) New work. Samples taken by conventional boring techniques are
normally required for new work dredging. Samples should be taken from
within the major zones of spatial variation in sediment type or along the
proposed channel center line at constant spacing to define stratification
within the material to be dredged and to obtain representative samples.
Borings are required for new projects and should be advanced below the depth
of anticipated dredging. The relative density of sands can be determined
by driving a split-spoon sampler and recording the number of blows required
to penetrate each foot of sand. Refer to EM 1110-2-1907 for information
on conventional soil sampling methods and standard split-spoon penetration
tests. Information on the soil above and below the authorized new work
depth is needed to properly design the channel slopes. It is essential to
obtain the characteristics of the material to be dredged to preclude determination of unsuitable dredge plant, unrealistic production and cost estimates, etc. Pertinent information regarding sediment samplers is summarized in table 2-1.
b. Laboratory Testing. Laboratory tests are required to provide data
for determining the proper dredge plant, evaluating and designing disposal
alternatives, designing channel slopes and retention dikes, and estimating
long-term storage capacity for confined disposal areas. The tests discussed below are to be used to characterize the material to be dredged so
that a proper dredge plant can be selected. Specific tests for evaluation
and design of disposal alternatives are discussed in Chapter 4. The required laboratory tests are essentially standard tests and generally follow
procedures found in EM 1110-2-1906 . The extent of the testing program is
project-dependent:
fewer tests are required when dealing with a relatively
homogeneous material and/or when data are available from previous tests and
experience, as is frequently the case in maintenance dredging; for new work
projects and unusual maintenance dredging projects where considerable variation in sediment properties is apparent from samples, more extensive laboratory testing programs are required. Laboratory tests should always be
performed on representative sediment samples. Tests required on finegrained sediments (those of which more than half pass through a No. 40
sieve) include natural water content, plasticity analyses (Atterberg
limits), and specific gravity. The coarse-grained sediments (those of
which more than half are retained on a No. 40 sieve) require only grain
size analyses and in situ density determinations. These tests are described below.
(1) Natural water content test. Natural water content refers to the
in situ water content of the sediment. It is used to determine the in situ
void ratio and in situ density of fine-grained sediments. Water content
determinations should be made on representative samples from borings and
grab samples of fine-grained sediment obtained during field investigation.
Fine-grained sediments do not drain rapidly; thus, representative samples
taken from borings and grab samples are considered to represent in situ
water contents. Detailed test procedures for determining the water content
are found in Appendix I of EM 1110-2-1906.
2-3

EM 1110-2-5025
25 Mar 83
Table 2-1.

Summary of Sediment Sampling Equipment

Sampler

Weight

Remarks
2
Samples 144-in. area to a
depth of up to 12 in., depending on sediment texture

Peterson

39-93 lb

Shipek

150 lb

Ekman

9 lb

Suitable only for very soft


sediments

Ponar

45-60 lb

Samples 81-in.2 area to a


depth of less than 12 in.
Ineffective in hard clay

Drag Bucket

Varies

Skims an irregular slice of


sediment surface. Available
in assorted sizes and shapes

Phleger Tube
(gravity
corer)

Variable :
17-77 lb;
fixed in
excess of
90 lb

Shallow core samples may be


obtained by self-weight
penetration and/or pushing
from boat. Depth of penetration dependent on weight
and sediment texture

Conventional
Soil Samplers

Refer to
EM 1110-2-1907

2
Samples 64-in. area to a
depth of approximately 4 in.

Conventional soil samplers


may be employed using bargeor boat-mounted drilling
equipment. Core samples
attainable to full depth of
dredging

2-4

EM 1110-2-5025
25 Mar 83
(2) Plasticity analyses. Plasticity analyses (Atterberg limits) should
be performed on the separated fine-grained fraction (passing the No. 40 sieve)
of sediment samples. A detailed explanation of the tests required to evaluate
the plasticity of sediments is presented in Appendix III of EM 1110-2-1906.
Samples should be classified according to the Unified Soil Classification
System (USCS) (item 12).
(3) Specific gravity test. Values for the specific gravities of solids
in fine-grained sediments are required for determining void ratios and in situ
densities. Procedures for conducting the specific gravity test are given in
Appendix IV of EM 1110-2-1906.
(4) Grain size analyses. Grain size analyses are required only on the
coarse-grained fraction of samples. Grain size analyses should follow the
procedures contained in Appendix V of EM 1110-2-1906.
c. In situ density. In situ density is used to evaluate dredgability to
sediments and aid in equipment selection, to estimate production rates, and to
estimate volume required for storage in confined disposal areas. In situ
density can be estimated from field investigations of sediments or from
laboratory test data using geotechnical engineering formulas. Refer to
Appendix II of EM 1110-2-1906 for guidance in estimating in situ density from
laboratory tests. For sand sediments, relative density has a decisive
influence on the selection of equipment for dredging. The relative density of
sands can be estimated from standard split-spoon penetration tests (para
2-4a). Table 2-2 presents estimates of relative density of sands based on
standard penetration tests. Where no field tests are performed on coarsegrained materials (i.e. sand, gravel, etc., ) the material in its densest state
based on laboratory tests will be considered comparable to its in situ
condition.
Table 2-2.

Relative Density of Sands According to Results


of Standard Penetration Tests

No. of Blows/ft

Relative Density

0-4
4-10
10-30
30-50
Over 50

Very loose
Loose
Medium
Dense
Very dense

2-5. Selection of Dredging Equipment. Most Corps dredging is performed by


private industry under contract, and the specifications should not be
written such that competitive bidding is restricted. However, in certain
situations limitations may be placed on the equipment to be used to minimize the environmental impact of the dredging and disposal operation. In
cases where available upland containment areas are small, the size of the
dredge should be restricted to minimize stress on the containment area
dikes and to provide adequate retention time for sedimentation to minimize
2-5

EM 1110-2-5025
25 Mar 83
excessive suspended solids in the weir effluent. Environmental
protection is adequate justification for carefully controlling the
selection and use of dredging equipment. The dredging of contaminated
sediments requires careful assessment of the dredging operation. The
information presented in Chapters 3 and 4 will provide guidance for
proper equipment selection based on the materials to be dredged, dredging
environment, contamination level of sediments, transport and disposal
requirements, and production requirements.
2-6. Disposal Alternatives. The major considerations in selecting
disposal alternatives are the environmental impact and the economics of
the disposal operation. Much of the recent knowledge concerning dredged
material disposal was gained as a result of the Dredged Material Research
Program (DMRP) conducted by the U.S. Army Engineer Waterways Experiment
Station (WES) and reported in WES Technical Reports. The major
objectives of the DMRP were to provide definitive information on the
environmental impact of dredging and dredged material disposal operations
and to develop new or improved dredged material disposal practices. The
research was conducted on a national basis, excluding no major types of
dredging activity or region or environmental setting. It produced
methods for evaluating the physical, chemical, and biological impacts of
a variety of disposal alternatives in water, on land, or in wetland
areas, as well as tested, viable, cost-effective methods and guidelines
for reducing the impacts of conventional disposal alternatives. Summary
reports produced under this program are listed in para 1-3, and a
detailed discussion of disposal alternatives is presented in Chapter 4.
Two fundamental conclusions were drawn from the results of the DMRP
concerning disposal of dredged material: (1) no single disposal
alternative can be presumed most suitable for a region, a type of dredged
material, or a group of projects before it has been tested, and (2)
environmental considerations make necessary long-range regional planning
for lasting, effective solutions to disposal concerns. There is no
inherent effect or characteristic of a disposal alternative that can rule
it out of consideration from an environmental standpoint before specific
on-site evaluation. This holds true for open-water disposal, confined
upland disposal, habitat development, or any other alternative.
Case-by-Case project evaluations are time-consumig and expensive and may
seriously complicate advanced planning and funding requests.
Nevertheless, from a technical point of view, situations can be
envisioned where tens of millions of dollars may have been or could be
spent for disposal alternatives that contribute to adverse environmental
effects rather than reduce them. Also, easily obtained beneficial
impacts should not be overlooked. No category of disposal alternative is
without environmental risk or offers the soundest environmental
protection or reflects the best management practice; therefore, all
disposal alternatives should be fully investigated during the planning
process and treated on an equal basis until a final decision can be made
based on all available facts. It is hypothesized that all alternatives
could be considered to dispose of even the most highly contaminated
dredged material if a plan could be devised for management that was
adequate and legally acceptable under domestic regulations and
international treaty.
2-7.

Long-Range Studies.

Dredging and disposal activities cannot be


2-6

EM 1110-2-5025
25 Mar 83

designed independently for each of several projects in a given area. While


each project may require different specific solutions, the interrelationships among them must be determined. Thought must also be given to changing particular dredging techniques and disposal alternatives as conditions
change. Long-range regional dredging and disposal management plans not
only offer greater opportunities for environmental protection and effective
use of dredging equipment at reduced project cost, but they also meet with
greater public acceptance once they are agreed upon. Long-range plans must
reflect sound engineering design, consider and minimize any adverse environmental impacts, and be operationally implementable.

2-7

EM 1110-2-5025
25 Mar 83
CHAPTER 3
DREDGING EQUIPMENT AND TECHNIQUES
3-1. Purpose. This chapter includes a description of the dredging equipment and techniques used in dredging activities in the United States and
presents advantages and limitations for each type of dredge. Guidance is
provided for selection of the best dredging equipment and techniques for a
proposed dredging project to aid in planning and design.
3-2.

Factors Determining Equipment Selection.

a. The types of equipment used, by both the Corps and private industry, and the average annual amount of dredging associated with each type
are shown in Figure 3-1. The dredging methods employed by the Corps vary
considerably throughout the United States. Principal types of dredges include hydraulic pipeline types (cutterhead, dustpan, plain suction, and
sidecaster), hopper dredges, and clamshell dredge. The category of "other"
dredges in Figure 3-1 includes dipper, ladder, and special purpose dredges.
However, there are basically only three mechanisms by which dredging is
actually accomplished:
(1) Suction dredging. Removal of loose materials by dustpans,
hoppers, hydraulic pipeline plain suction, and sidecasters, usually for
maintenance dredging projects.
(2) Mechanical dredging. Removal of loose or hard, compacted materials by clamshell, dipper, or ladder dredges, either for maintenance or
new work projects.
(3) A combination of suction and mechanical dredging. Removal. of
loose or hard, compacted materials by cutterheads, either for maintenance
or new work projects.
b. Selection of dredging equipment and method used to perform the
dredging will depend on the following factors:
(1) Physical characteristics of material to be dredged.
(2) Quantities of material to be dredged.
(3) Dredging depth.
(4) Distance to disposal area.
(5) Physical environment of and between the dredging and disposal
areas.
(6) Contamination level of sediments.
(7) Method of disposal.

3-1

EM 1110-2-5025
25 Mar 83

3-2

EM 1110-2-5025
25 Mar 83
(8) Production required.
(9) Type of dredges available.
3-3.

Hopper Dredges.

a. General. Hopper dredges are self-propelled seagoing ships of from


180 to 550 ft in length, with the molded hulls and lines of ocean vessels
(fig. 3-2). They are equipped with propulsion machinery, sediment containers (hoppers), dredge pumps, and other special equipment required to
perform their essential function of removing material from a channel bottom
or ocean bed. Hopper dredges have propulsion power adequate for required
free-running speed and dredging against strong currents and excellent maneuverability for safe and effective work in rough, open seas. Dredged material is raised by dredge pumps through dragarms connected to drags in
contact with the channel bottom and discharged into hoppers built in the
vessel. Hopper dredges are classified according to hopper capacity: largeclass dredges have hopper capacities of 6000 cu yd or greater, medium-class
hopper dredges have hopper capacities of 2000 to 6000 cu yd, and smallclass hopper dredges have hopper capacities of from less than 2000 to
500 cu yd. During dredging operations, hopper dredges travel at a ground
speed of from 2 to 3 mph and can dredge in depths from about 10 to over
80 ft. They are equipped with twin propellers and twin rudders to provide
the required maneuverability. Table 3-1 gives available specifications for
all vessels in the Corps hopper dredge fleet.
b.

Description of Operation.

(1) General. Operation of a seagoing hopper dredge involves greater


effort than that required for an ordinary ocean cargo vessel, because not
only the needs of navigation of a self-propelled vessel but also the needs
associated with its dredging purposes must be satisfied. Dredging is accomplished by progressive traverses over the area to be dredged. Hopper
dredges are equipped with large centrifugal pumps similar to those employed
by other hydraulic dredges. Suction pipes (dragarms) are hinged on each
side of the vessel with the intake (drag) extending downward toward the
stern of the vessel. The drag is moved along the channel bottom as the
vessel moves forward at speeds up to 3 mph. The dredged material is sucked
up the pipe and deposited and stored in the hoppers of the vessel. Once
fully loaded, hopper dredges move to the disposal site to unload before resuming dredging. Unloading is accomplished either by opening doors in the
bottoms of the hoppers and allowing the dredged material to sink to the
open-water disposal site or by pumping the dredged material to upland disposal sites. Because of the limitations on open-water disposal, most
hopper dredges have direct pumpout capability for disposal in upland confined sites. Before there were environmental restrictions, hopper dredges
were operated with the primary objective of obtaining the maximum economic
load; i.e., removing the maximum quantity of material from the channel
prism in the shortest pumping time during a day's operation.
(2) Hopper dredging is accomplished by three methods: (a) pumping
past overflow, (b) agitation dredging, and (c) pumping to overflow. The
3-3

EM 1110-2-5025
25 Mar 83

HOPPERS

Figure 3-2.

Self-propelled seagoing hopper dredge.


3-4

EM 1110-2-5025
25 Mar 83

3-5

EM 1110-2-5025
25 Mar 83
use of these methods is controlled to varying degrees by environmental
legislation and the water quality certification permits required by the
various states in which dredging is being accomplished. The environmental
effects of these methods must be assessed on a project-by-project basis.
If the material being dredged is clean sand, the percentage of solids in
the overflow will be small and economic loading may be achieved by pumping
past overflow. When contaminated sediments are to be dredged and adverse
environmental effects have been identified , pumping past overflow is not
recommended. In such cases, other types of dredges may be more suitable
for removing the contaminated sediments from the channel prism. If hopper
dredges are not allowed to pump past overflow in sediments that have good
settling properties, the cost of dredging increases. The settling properties of silt and clay sediments may be such that only a minimal load increase would be achieved by pumping past overflow. Economic loading, i.e.
the pumping time required for maximum production of the hopper dredge,
should be determined for each project. These determinations, along with
environmental considerations, should be used to establish the operation
procedures for the hopper dredge.
(3) Agitation dredging. Agitation dredging is a process which intentionally discharges overboard large quantities of fine-grained dredged material by pumping past overflow, under the assumption that a major portion
of the sediments passing through the weir overflow will be transported and
permanently deposited outside the channel prism by tidal, river, or littoral
currents. Agitation dredging should be used only when the sediments dredged
have poor settling properties, when there are currents in the surrounding
water to carry the sediments from the channel prism, and when the risk to
environmental resources is low. Favorable conditions may exist at a particular project only at certain times of the day, such as at ebb tides, or
only at such periods when the streamflow is high. To use agitation dredging effectively requires extensive studies of the project conditions and
definitive environmental assessments of the effects. Agitation dredging
should not be performed while operating in slack water or when prevailing
currents permit redeposit of substantial quantities of the dredged material
in the project area or in any other area where future excavation may be
required.
Refer to para 3-12 for more information on this topic.
(4) Refer to ER 1125-2-312 for instructions for hopper dredge
operations.
c. Application. Hopper dredges are used mainly for maintenance dredging in exposed harbors and shipping channels where traffic and operating
conditions rule out the use of stationary dredges. The materials excavated
by hopper dredges cover a wide range of types, but the hopper dredge is most
effective in the removal of material which forms shoals after the initial
dredging is completed. While specifically designed drags are available for
use in raking and breaking up hard materials, hopper dredges are most efficient in excavating loose, unconsolidated materials. At times, hopper
dredges must operate under hazardous conditions caused by fog, rough seas,
and heavy traffic encountered in congested harbors.
d.

Advantages.

Because of the hopper dredges design and method of


3-6

EM 1110-2-5025
25 Mar 83
operation, the self-propelled seagoing hopper dredge has the following advantages over other types of dredges for many types of projects:
(1) It is the only type of dredge that can work effectively, safely,
and economically in rough, open water.
(2) It can move quickly and economically to the dredging project
under its own power.
(3) Its operation does not interfere with or obstruct traffic.
(4) Its method of operation produces usable channel improvement almost
as soon as work begins. A hopper dredge usually traverses the entire length
of the problem shoal, excavating a shallow cut during each passage and increasing channel depth as work progresses.
(5) The hopper dredge may be the most economical type of dredge to
use where disposal areas are not available within economic pumping distances of the hydraulic pipeline dredge.
e. Limitations. The hopper dredge is a seagoing self-propelled vessel
designed for specific dredging projects. The following limitations are associated with this dredge:
(1) Its deep draft precludes use in shallow waters, including barge
channels.
(2) It cannot dredge continuously. The normal operation involves
loading, transporting material to the dump site, unloading, and returning
to the dredging site.
(3) The hopper dredge excavates with less precision than other types
of dredges.
(4) Its economic load is reduced when dredging contaminated sediments
since pumping past overflow is generally prohibited under these conditions
and low-density material must be transported to and pumped into upland disposal areas.
(5) It has difficulty dredging side banks of hardpacked sand.
(6) The hopper dredge cannot dredge effectively around piers and
other structures.
(7) Consolidated clay material cannot be economically dredged with
the hopper dredge.
3-4.

Cutterhead Dredges.

a. General. The hydraulic pipeline cutterhead suction dredge is the


most commonly used dredging vessel and is generally the most efficient and
versatile (fig. 3-3). It performs the major portion of the dredging
3-7

EM 1110-2-5025
25 Mar 83

DISCHARGE LINE

CUTTERHEAD
Figure 3-3.

Hydraulic pipeline cutterhead dredge.

3-8

EM 1110-2-5025
25 Mar 83
workload in the United States. Because it is equipped with a rotating
cutter apparatus surrounding the intake end of the suction pipe, it can efficiently dig and pump all types of alluvial materials and compacted deposits, such as clay and hardpan. This dredge has the capability of pumping dredged material long distances to upland disposal areas. Slurries of
10 to 20 percent solids (by dry weight) are typical, depending upon the
material being dredged, dredging depth, horsepower of dredge pumps, and
pumping distance to disposal area. If no other data are available, a pipeline discharge concentration of 13 percent by dry weight (145 ppt) should
be used for design purposes. Pipeline discharge velocity, under routine
working conditions, ranges from 15-20 ft/sec. Table 3-2 presents theoretical pipeline discharge rates as functions of pipeline discharge velocities for dredges ranging in sizes from 8 to 30 in.
Table 3-2.

Suction Dredge Pipeline Discharge Rates,a


cu ft/sec

Discharge
Velocity
ft/sec

a in.

10
15
20
25

3.5
5.2
7.0
8.7

Discharge Pipe Diameter


18 in.
24 in.
17.7
26.5
35.3
44.2

31.4
47.1
62.8
78.5

30 in.
49.1
73.6
98.1
122.7

Discharge rate = pipeline area x discharge velocity.

Production rate is defined as the number of cubic yards of in situ sediments dredged during a given period and is usually expressed in cu yd/hr.
Production rates of dredges vary according to the factors listed above and
other operational factors that are not necessarily consistent between
dredges of the same size and type. For example, a 16-in. dredge should produce between 240 and 875 cu yd of dredged material per hour, and a 24-in.
dredge should produce between 515 and 1615 cu yd per hour. The range for
typical cutterhead production as a function of dredge size is shown in figure 3-4. This figure illustrates the wide range of production for dredges
of the same size. The designer can refer to figure 3-5, which shows the
relationships among solids output, dredge size, and pipeline length for
various dredging depths, as a preliminary selection guide for the size of
dredge required for a given project. This is only a rough guide, and accurate calculations based not only on the type of material to be dredged
but on the power available and other considerations should be completed before a final engineering recommendation can be made. The designer should
refer to the data available from ENG Form 4267, "Report of Operations-Pipeline, Dipper, or Bucket Dredges," for use in estimating production
rates, effective working time, etc. These data on past dredging projects
are available in the Construction-Operation Divisions of the Districts.
Specifications and dimensions for several cutterhead dredges ranging in
pipe diameter from 6 to 30 in. are presented in table 3-3.
3-9

EM 1110-2-5025
25 Mar 83

DREDGE SIZE. IN.

Figure 3-4.

Typical cutterhead dredge production according to


dredge size.

3-10

EM 1110-2-5025
25 Mar 83

Figure 3-5. Relationships among solids output, dredge size, and


pipeline length for various dredging depths- (WES TR DS-78-10 )

3-11

EM 1110-2-5025
25 Mar 83

3-12

EM 1110-2-5025
25 Mar 83
b. Description of Operation. The cutterhead dredge is generally
equipped with two stern spuds used to hold the dredge in working position
and to advance the dredge into the cut or excavating area. During operation, the cutterhead dredge swings from side to side alternately using the
port and starboard spuds as a pivot, as shown in figure 3-6. Cables attached to anchors on each side of the dredge control lateral movement. Forward movement is achieved by lowering the starboard spud after the port
swing is made and then raising the port spud. The dredge is then swung
back to the starboard side of the cut centerline. The port spud is lowered
and the starboard spud lifted to advance the dredge. The excavated material may be disposed of in open water or in confined disposal areas located
upland or in the water. In the case of open-water disposal, only a floating discharge pipeline, made up of sections of pipe mounted on pontoons and
held in place by anchors, is required. Additional sections of shore pipeline are required when upland disposal is used. In addition, the excavated
materials may be placed in hopper barges for disposal in open water or in
confined areas that are remote from the dredging area. In cutterhead dredging, the pipeline transport distances usually range up to about 3 miles.
For commercial land reclamation or fill operations, transport distances are
generally longer, with pipeline lengths reaching as far as 15 miles, for
which the use of multiple booster pumps is necessary.

6
ANCHOR

ANCHOR
I0 .,

3WN)

Figure 3-6. Operation of a cutterhead dredge (viewed


from above).
3-13

EM 1110-2-5025
25 Mar 83
c. Application. Although the cutterhead dredge was developed to
loosen up densely packed deposits and eventually cut through soft rock, it
can excavate a wide range of materials including clay, silt, sand, and
gravel. The cutterhead, however, is not needed in maintenance dredging of
most materials consisting of clay, silt, and fine sand because in these materials, rotation of the cutterhead produces a turbidity cloud and increases the potential for adverse environmental impacts. Common practice
is to use the cutterhead whether it is needed or not. When the cutterhead
is removed, cutterhead dredges become in effect plain suction dredges. The
cutterhead dredge is suitable for maintaining harbors, canals, and outlet
channels where wave heights are not excessive. A cutterhead dredge designed to operate in calm water will not operate offshore in waves over
2-3 ft in height; the cutterhead will be forced into the sediment by wave
action creating excessive shock loads on the ladder. However, a cutterhead
dredge designed to operate offshore can operate in waves up to about 6 ft.
d. Advantages. The cutterhead dredge is the most widely used dredge
in the United States because of the following advantages:
(1) Cutterhead dredges are used on new work and maintenance projects
and are capable of excavating most types of material and pumping it through
pipelines for long distances to upland disposal sites.
(2) The cutterhead operates on an almost continuous dredging cycle,
resulting in maximum economy and efficiency.
(3) The larger and more powerful machines are able to dredge rocklike
formations such as coral and the softer types of basalt and limestone without blasting.
e.

Limitations.

The limitations on cutterhead dredges are as follows:

(1) The cutterhead dredges available in the United States have


limited capability for working in open-water areas without endangering personnel and equipment. The dredging ladder on which the cutterhead and suction pipe are mounted is rigidly attached to the dredge; this causes operational problems in areas with high waves.
(2) The conventional cutterhead dredges are not self-propelled. They
require the mobilization of large towboats in order to move between dredging locations.
(3) The cutterhead dredge has problems removing medium and coarse
sand in maintaining open channels in rivers with rapid currents. It is difficult to hold the dredge in position when working upstream against the
river currents since the working spud often slips due to scouring effects.
When the dredge works downstream, the material that is loosened by the cutterhead is not pulled into the suction intake of the cutterhead. This
causes a sandroll, or berm, of sandy material to form ahead of the dredge.
(4) The pipeline from the cutterhead dredge can cause navigation
problems in small, busy waterways and harbors.
3-14

EM 1110-2-5025
25 Mar 83
3-5.

Dustpan Dredge.

a. General. The dustpan dredge is a hydraulic suction dredge that


uses a widely flared dredging head along which are mounted pressure water
jets (fig. 3-7). The jets loosen and agitate the sediments which are then
captured in the dustpan head as the dredge itself is winched forward into
the excavation. This type of dredge was developed by the Corps of Engineers to maintain navigation channels in uncontrolled rivers with bedloads
consisting primarily of sand and gravel. The first dustpan dredge was developed to maintain navigation on the Mississippi River during low river
stages. A dredge was needed that could operate in shallow water and be
large enough to excavate the navigation channel in a reasonably short time.
The dustpan dredge operates with a low-head, high-capacity centrifugal pump
since the material has to be raised only a few feet above the water surface
and pumped a short distance. The dredged material is normally discharged
into open water adjacent to the navigation channel through a pipeline
usually only 800 to 1000 ft long.
b. Description of Operation. The dustpan dredge maintains navigation
channels by making a series of parallel cuts through the shoal areas until
the authorized widths and depths are achieved. Typical operation procedures for the dustpan dredge are as follows:
(1) The dredge moves to a point about 500 ft upstream of the upper
limit of the dredging area and the hauling anchors are set. Two anchors
are used, as shown in Figure 3-8. The hauling winch cables attached to the
anchors are crossed to provide better maneuverability and control of the
vessel while operating in the channel prism.
(2) The dredge is then moved downstream to the desired location. The
suction head is lowered to the required depth, dredge pump and water jet
pumps are turned on, and the dredging commences. The dredge is moved forward by the hauling cables. The rate of movement depends on the materials
being dredged, depth of dredging, currents, and wind. In shallow cuts, the
advance may be as rapid as 800 ft/hr.
(3) When the upstream end of the cut is reached, the suction head is
raised and the dredge is moved back downstream to make a parallel cut. This
operation is repeated until the desired dredging widths and depths are
achieved.
(4) The suction head may have to be lowered or raised if obstacles
such as boulders, logs, or tree stumps are encountered. Experience with
dustpan dredges indicates that the best results are obtained when the
height of the cut face does not exceed 6 ft in depth.
(5) The dr edge is moved outside the channel to let waterborne traffic
pass through the area simply by raising the suction head and slacking off
on one of the hauling winch cables. The propelling engines can be used to
assist in maneuvering the dredge clear of the channel. The vessel is held
in position by lowering the suction head or by lowering a spud.

3-15

EM 1110-2-5025
25 Mar 83

Figure 3-7.

Dustpan dredge.

3-16

EM 1110-2-5025
25 Mar 83

Figure 3-8. Operation of dustpan dredge


(viewed from above).
c. Application. The pipeline system and the rigid ladder used with
the dustpan dredge make it effective only in rivers or sheltered waters; it
cannot be used in estuaries or bays where significant wave action occurs.
Because it has no cutterhead to loosen hard, compact materials, the dustpan
dredge is mostly suited for high-volume, loose-material dredging. Dustpan
dredges are used to maintain the navigation channel of the uncontrolled
open reaches of the Mississippi, Missouri, and Ohio Rivers. Dustpan dredging is principally a low-stage season operation. River channels are surveyed before the end of the high-stage season to determine the location and
depths at the river crossings and sandbar formations, and dustpan dredging
operations are planned accordingly. The existing fleet of Corps dustpan
dredges is described briefly in table 3-4.
Table 3-4.

Corps Dustpan Dredges.

Name

District Location

Discharge Diameter, in.

Age, years

Mitchell
Burgess
Ockerson
Potter
Jadwin

Kansas City
Memphis
Memphis
St. Louis
Vicksburg

34
32
32
32
32

47
47
49
49
47

These dredges are high-volume dredges capable of excavating a navigation


channel through river sediment in a short time. During FY 71, the dredge
Jadwin excavated over 6,200,000 cu yd, with an average production rate of
approximately 3600 cu yd/hr. Detailed operations data for all the dustpan
3-17

EM 1110-2-5025
25 Mar 83
dredges are reported on ENG Form 4267, "Report of Operations--Pipeline,
Dipper, or Bucket Dredges." Refer to table 3-3 for specifications for a
typical dustpan dredge.
d. Advantages. The dustpan dredge is self-propelled, which enables it
to move rapidly over long distances to work at locations where emergencies
occur. The attendant plant and pipeline are designed for quick assembly so
that work can be started a few hours after arrival at the work site. The
dustpan dredge can move rapidly out of the channel to allow traffic to pass
and can resume work immediately. The high production rate and design of
the dustpan dredge make it possible to rapidly remove sandbar formations
and deposits from river crossings so that navigation channels can be maintained with a minimum of interruption to waterborne traffic.
e. Limitations. The dustpan dredge was designed for a specific purpose, and for this reason there are certain limitations to its use in other
dredging environments. It can dredge only loose materials such as sands
and gravels and only in rivers or sheltered waters where little wave action
may be expected. The dustpan dredge is not particularly well suited for
transporting dredged material long distances to upland disposal sites; pumping distances are limited to about 1000 ft without the use of booster pumps.
3-6.

Sidecasting Dredges.

a. General. The sidecasting type of dredge (fig. 3-9) is a shallowdraft seagoing vessel, especially designed to remove material from the bar
channels of small coastal inlets. The hull design is similar to that of a
hopper dredge; however, sidecasting dredges do not usually have hopper bins.
Instead of collecting the material in hoppers onboard the vessel, the sidecasting dredge pumps the dredged material directly overboard through an elevated discharge boom; thus, its shallow draft is unchanged as it constructs
or maintains a channel. The discharge pipeline is suspended over the side
of the hull by structural means and may be supported by either a crane or a
truss-and-counterweight design. The dredging operations are controlled by
steering the vessel on predetermined ranges through the project alignment.
The vessel is self-sustaining and can perform work in remote locations with
a minimum of delay and service requirements. The projects to which the
sidecasters are assigned for the most part are at unstabilized, small inlets which serve the fishing and small-boat industries. Dangerous and unpredictable conditions prevail in these shallow inlets making it difficult for
conventional plant to operate except under rare ideal circumstances.
b. Description of Operation. The sidecasting dredge picks up the bottom material through two dragarms and pumps it through a discharge pipe supported by a discharge boom. During the dredging process, the vessel travels
along the entire length of the shoaled area casting material away from and
beyond the channel prism. Dredged material may be carried away from the
channel section by littoral and tidal currents. The construction of a
deepened section through the inlet usually results in some natural scouring
and deepening of the channel section, since currents moving through the
prism tend to concentrate the scouring action in a smaller active zone. A
typical sequence of events in a sidecasting operation is as follows:
3-18

EM 1110-2-5025
25 Mar 83

Figure 3-9.

Sidecasting dredge.

3-19

EM 1110-2-5025
35 Mar 83
(1) The dredge moves to the work site.
(2) The dragarms are lowered to the desired depth.
(3) The pumps are started to take the material from the channel bottom and pump it through the discharge boom as the dredge moves along a designated line in the channel prism.
(4) If adequate depths are not available across the bar during low
tide levels, dredging must be started during higher tide levels. Under
these conditions, the cuts are confined to a narrow channel width to
quickly attain the flotation depth necessary for dredging to be continued
during the low tidal periods.
(5) The dredge continues to move back and forth across the bar until
the channel dimensions are restored.
(6) The discharge can be placed on either side of the dredge by rotating the discharge boom from one side of the hull to the other.
c. Application. The Corps of Engineers developed the shallow-draft
sidecasting dredge for use in places too shallow for hopper dredges and too
rough for pipeline dredges. The types of materials that can be excavated
with the sidecasting dredge are the same as for the hopper
dredges
__
(para 3-3c).
d. Advantages. The sidecasting type of dredge, being self-propelled,
can rapidly move from one project location to another on short notice and
can immediately go to work once at the site. Therefore, a sidecasting
dredge can maintain a number of projects located great distances from each
other along the coastline.
e. Limitations. The sidecasting dredge needs flotation depths before
it can begin to work because it dredges while moving over the shoaled area.
Occasionally, a sidecaster will need to alter its schedule to work during
higher tide levels periods only, due to insufficient depths in the shoaled
area. Most areas on the seacoast experience a tidal fluctuation sufficient
to allow even the shallowest shoaled inlets to be reconstructed by a sidecasting type of dredge. A shallow-draft sidecasting dredge cannot move
large volumes of material compared to a hopper dredge, and some of the material removed can return to the channel prism due to the effects of tidal
and littoral currents. The sidecasting dredge has only open-water disposal
capability; therefore, it cannot be used for dredging contaminated
sediments.
3-7.

Dipper Dredges.

a. General. The dipper dredge is basically a barge-mounted power


shovel. It is equipped with a power-driven ladder structure and operated
from a barge-type hull. A schematic drawing and photograph of the dipper
dredge are shown in figure 3-10. A bucket is firmly attached to the ladder
structure and is forcibly thrust into the material to be removed. To
3-20

EM

Figure 3-10.

3-21

Dipper dredge.

1110-2-5025
25 Mar 83

EM 1110-2-5025
25 Mar 83
increase digging power, the dredge barge is moored on powered spuds that
transfer the weight of the forward section of the dredge to the bottom.
Dipper dredges normally have a bucket capacity of 8 to 12 cu yd and a working depth of up to 50 ft. There is a great variability in production rates,
but 30 to 60 cycles per hour is routinely achieved.
b. Description of Operation. The dipper type of dredge is not selfpropelled but can move itself during the dredging process by manipulation
of the spuds and the dipper arm. A typical sequence of operation is as
follows :
(1) The dipper dredge, scow barges, and attendant plant are moved to
the work site.
(2) The dredge is moved to the point where work is to start; part of
the weight is placed on the forward spuds to provide stability.
(3) A scow barge is brought alongside and moored into place by
winches and cables on the dipper dredge.
(4) The dredge begins digging and placing the material into the
moored barge.
(5) When all the material within reach of the bucket is removed, the
dredge is moved forward by lifting the forward spuds and maneuvering with
the bucket and stern spud.
(6) The loaded barges are towed to the disposal area and emptied by
bottom dumping if an open-water disposal area is used, or they are unloaded
by mechanical or hydraulic equipment if diked disposal is required.
(7) These procedures are repeated until the dredging operation is
completed.
c. Application. The best use of the dipper dredge is for excavating
hard, compacted materials, rock, or other solid materials after blasting.
Although it can be used to remove most bottom sediments, the violent action
of this type of equipment may cause considerable sediment disturbance and
resuspension during maintenance digging of fine-grained material. In addition, a significant loss of the fine-grained material will occur from the
bucket during the hoisting process. The dipper dredge is most effective
around bridges, docks, wharves, pipelines, piers, or breakwater structures
because it does not require much area to maneuver; there is little danger
of damaging the structures since the dredging process can be controlled accurately. No provision is made for dredged material containment or transport, so the dipper dredge must work alongside the disposal area or be accompanied by disposal barges during the dredging operation.
d. Advantages. The dipper dredge is a rugged machine that can remove
bottom materials consisting of clay, hardpacked sand, glacial till, stone,
or blasted rock material. The power that can be applied directly to the
cutting edge of the bucket makes this type of dredge ideal for the removal
3-22

EM

1110-2-5025
25 Mar 83

of hard and compact materials. It can also be used for removing old piers,
breakwaters, foundations, pilings, roots, stumps, and other obstructions.
The dredge requires less room to maneuver in the work area than most other
types of dredges; the excavation is precisely controlled so that there is
little danger of removing material from the foundation of docks and piers
when dredging is required near these structures. Dipper dredges are frequently used when disposal areas are beyond the pumping distance of pipeline dredges, due to the fact that scow barges can transport material over
long distances to the disposal area sites. The dipper type of dredge can
be used effectively in refloating a grounded vessel. Because it can operate with little area for maneuvering, it can dig a shoal out from under
and around a grounded vessel. The dipper dredge type of operation limits
the volume of excess water in the barges as they are loaded. Dipperdredged material can be placed in the shallow waters of eroding beaches to
assist in beach nourishment.
e. Limitations.
It is difficult to retain soft, semisuspended finegrained materials in the buckets of dipper dredges.
Scow-type barges are
required to move the material to a disposal area, and the production is
relatively low when compared to the production of cutterhead and dustpan
dredges. The dipper dredge is not recommended for use in dredging contaminated sediments.
3-8.

Bucket

Dredges.

a. General. The bucket type of dredge is so named because it utilizes


a bucket to excavate the material to be dredged (fig. 3-11). Different
types of buckets can fulfill various types of dredging requirements. The
buckets used include the clamshell, orangepeel, and dragline types and can
be quickly changed to suit the operational requirements. The vessel can be
positioned and moved within a limited area using only anchors; however, in
most cases anchors and spuds are used to position and move bucket dredges.
The material excavated is placed in scows or hopper barges that are towed
to the disposal areas. Bucket dredges range in capacity from 1 to 12 cu yd.
The crane is mounted on a flat-bottomed barge, on fixed-shore installations,
or on a crawler mount. Twenty to thirty cycles per hour is typical, but
large variations exist in production rates because of the variability in
depths and materials being excavated. The effective working depth is
limited to about 100 ft.
b. Description of Operation. The bucket type of dredge is not selfpropelled but can move itself over a limited area during the dredging process by the manipulation of spuds and anchors. A typical sequence of operation is as follows:
(1) The bucket dredge, scows or hopper barges, and attendant plant
are moved to the work site by a tug.
(2) The dredge is positioned at the location where work is to start
and the anchors and spuds lowered into place.

3-23

EM 1110-2-5025
25 Mar 83

Figure 3-11.

3-24

Bucket dredge.

EM 1110-2-5025
25 Mar 83
(3) A scow or hopper barge is brought alongside and secured to the
bucket dredge hull.
(4) The dredge begins the digging operation by dropping the bucket in
an open position from a point above the sediment. The bucket falls through
the water and penetrates into the bottom material. The sides or jaws of
the bucket are then closed through the use of wire cables operated from the
crane. As the sides of the bucket close, material is sheared from the bottom and contained in the bucket compartment. The bucket is raised above
the water surface and swung to a point over the hopper barge. The material
is then released into the hopper barge by opening the sides of the bucket.
(5) As material is removed from the bottom of the waterway to the desired depth at a given location, the dredge is moved to the next nearby location by using anchors. If the next dredging area is a significant distance away, the bucket dredge must be moved by a tug.
(6) The loaded barges are towed to the disposal area by a tug and
emptied by bottom dumping if an open water disposal area is used. If a
diked disposal area is used, the material must be unloaded using mechnical
or hydraulic equipment.
(7) These procedures are repeated until the dredging operation is
completed.
c. Application. Bucket dredges may be used to excavate most types of
materials except for the most cohesive consolidated sediments and solid
rock. Bucket dredges usually excavate a heaped bucket of material, but during hoisting turbulence washes away part of the load. Once the bucket
clears the water surface, additional losses may occur through rapid drainage of entrapped water and slumping of the material heaped above the rim.
Loss of material is also influenced by the fit and condition of the bucket,
the hoisting speed, and the properties of the sediment. Even under ideal
conditions, substantial losses of loose and fine sediments will usually
occur. Because of this, special buckets must be used if the bucket dredge
is to be considered for use in dredging contaminated sediments. To minimize
the turbidity generated by a clamshell operation, watertight buckets have
been developed (fig. 3-12). The edges seal when the bucket is closed and
the top is covered to minimize loss of dredged material. Available sizes
range from 2.6 to 26 cu yd. These buckets are best adapted for maintenance
dredging of fine-grained material. A direct comparison of 1.3 cu-yd typical clamshell and watertight clamshell operations indicates that watertight
buckets generate 30 to 70 percent less turbidity in the water column than
typical buckets. This reduction is probably due primarily to the fact that
leakage of dredged material from watertight buckets is reduced by approximately 35 percent. The bucket dredge is effective while working near
bridges, docks, wharves, pipelines, piers, or breakwater structures because
it does not require much area to maneuver; there is little danger of damaging the structures because the dredging process can be controlled
accurately.

3-25

EM 1110-2-5025
25 Mar 83

Figure 3-12.

Open and closed positions of the watertight bucket.

d. Advantages. The bucket dredge has the same advantages cited for
the dipper dredge, except that its capabilities in blasted rock and compact
materials are somewhat less. The density of material excavated is about
the same as the inplace density of the bottom material. Therefore, the
volume of excess water is minimal, which increases the efficiency of operation in the transportation of material from the dredging area to the disposal area.
e. Limitations. The limitations of the bucket type of dredge are the
same as those described for the dipper dredge (para 3-7e).
3-9.

Special-Purpose

Dredge

a. General. The Corps of Engineers Dredge CURRITUCK (fig. 3-13), assigned to the Wilmington District, is an example of a special-purpose type
of dredge. Designed to work the same projects as sidecasting dredges, the
CURRITUCK has the additional ability to completely remove material from the
It is a selfinlet complex and transport it to downdrift eroded beaches.
propelled split hull type of vessel, equipped with a self-leveling deckhouse located at the stern, where all controls and machinery are housed.
The vessel is hinged above the main deck so that the hull can open from bow
to stern by means of hydraulic cylinders located in compartments forward
and aft of the hopper section. The CURRITUCK has one hopper with a capacity
of 315 cu yd. The hopper section is clearly visible to the operators in
the pilot house, making production monitoring an easy task.
b. Description of Operation. The CURRITUCK operates in much the same
way as a hopper dredge. The operator steers the vessel through the shoal
3-26

EM 1110-2-5025
25 Mar 83

Figure 3-13.

Corps special-purpose dredge.


3-27

EM 1110-2-5025
25 Mar 83

areas of the channel. The dredge pumps, located in the compartments on


each side of the hull, pump material through trailing dragarms into the
hopper section. When an economic load is obtained, the dragarms are lifted
from the bottom of the waterway and the dredge proceeds to the disposal
area. A major difference between the operation of the CURRITUCK and that
of a conventional hopper dredge is in the method of disposal; the CURRITUCK
is designed to transport and deposit the dredged material close to the surf
zone area.
c. Application. The CURRITUCK provides a sand-bypassing capability
in addition to improving the condition of navigation channels. The
CURRITUCK excavates material from navigation channels, transports it to
downdrift eroded beaches, and releases it where it is needed to provide
beach nourishment, rather than wasting it offshore. After the material has
been deposited in the near-shore coastal areas, the dredge backs away and
returns to the navigation channel.
d. Advantages. The CURRITUCK is an effective dredging tool for use
in shallow-draft inlets. All of the dredged material is placed in the littoral zone. The CURRITUCK can also be used to supplement sidecasting
dredges and to transport dredged materials from inlet channels to the nearshore areas of eroded beaches.
e. Limitations. The production rate of the CURRITUCK is limited by
its small hopper capacity. Therefore, it is not effective on major navigation channels. In addition, when the flotation depths are minimal it is
necessary to use a sidecasting dredge to provide access into the project.
3-10. Summary of Dredge Operating Characteristics. The important operating characteristics of each dredge presented in the preceding sections are
summarized in table 3-5. In some cases, a wide range of values is given to
account for the various sizes of plants within each class. In other instances, the information provides a qualitative judgement (high, low,
average) of each dredge types performance in a given area. Table 3-5
should be helpful in making quick assessments of the suitability of a given
dredge type in a known physical setting.
3-11. Locations of Dredges in the United States. Figure 3-14 shows the
distribution of dredging capability for the Corps and industry in the United
States by region. Congress has determined (Public Law 95-269) that the
Corps will operate a dredging fleet adequate to meet emergency and national
defense requirements at home and abroad. This fleet will be maintained to
technologically modern and efficient standards and will be kept in a fully
operational status. The status of the United States dredging fleet as determined in the Corps of Engineers National Dredging Study is comprehensively summarized in a paper of the same title (item 6). A detailed inventory of all dredges in the United States is published annually in World
Dredging and Marine Construction (item 10). The designer can consult this
source for information on the specific types of dredges avail-able in the
proposed project area.
3-28

EM 1110-2-5025
25 Mar 83

3-29

EM 1110-2-5025
25 Mar 83

3-30

EM 1110-2-5025
25 Mar 83
3-12.

Agitation Dredging Techniques.

a. General. Agitation dredging is the process of removing bottom


material from a selected area by using equipment to raise it in the water
column and allowing currents to carry it from the project area.
In the
most detailed study available on agitation dredging techniques, Richardson
(item 7) evaluated past agitation dredging projects and presented guidelines and recommendations for using agitation dredging. Two distinct
phases are involved in agitation dredging: (1) suspension of bottom sediments by some type of equipment, and (2) transport of the suspended material by currents. The main purpose of the equipment is to raise bottom
material in the water column. Natural currents are usually involved in
transporting the material from the dredging site, although the natural
currents may be augmented with currents generated by the agitation equipment. Agitation dredging is accomplished by methods such as hopper dredge
agitation, prop-wash, vertical mixers or air bubblers, rakes or drag
beams, and water jets. Based on the work done by Richardson (item 7), only
hopper dredge, prop-wash, and rake or beam dragging agitation justify more
detailed discussion in this EM.
b. Objectives. The main objective of agitation dredging is the removal of bottom material from a selected area. If the material is
suspended but redeposits shortly in the same area, only agitation (not
agitation dredging) has been accomplished. The decision to use agitation
dredging should be based primarily on the following factors:
(1) Technical feasibility. The equipment to generate the required
level of agitation must be available, and the agitated material must be
carried away from the project area by currents.
(2) Economic feasibility. Agitation dredging must be determined the
most cost-effective method for achieving the desired results; it should not
affect the costs of other dredging projects downstream by increasing
dredging volumes.
(3) Environmental feasibility.
unacceptable environmental impacts.
c.

Agitation dredging should not cause

Hopper Dredge Agitation.

(1) General. Refer to para 3-3a for a general description of hopper


dredges. In agitation dredging, hopper capacity is of secondary importance
compared with pumping rates, mobility, and overflow provisions.
(2) Description of operation. The general operation of a hopper
dredge is discussed in para 3-3b. In hopper dredge agitation, the conventional dredge-haul-dump operating mode is modified by increasing the
dredging mode and reducing the haul-dump mode. It has been reported that
hopper dredge agitation can allow a project to be maintained with a dredge
which is relatively small compared with the size dredge required for a
conventional dredge-haul-dump operation. Hopper dredge agitation is of
3-31

EM 1110-2-5025
25 Mar 83

two types: (a) intentional agitation produced by hopper overflow; and


(b) auxiliary agitation caused by dragheads and propeller wash. Since the
latter is present in all hopper dredge operations and since it is difficult
to quantify separately from hopper overflow, both types are measured together when reporting hopper dredge agitation effectiveness.
(3) Application. Agitation hopper dredging can perform the same
maintenance functions as conventional hopper dredging if the following conditions are satisfied: (a) sediments are fine-grained and loosely consolidated, (b) currents are adequate to remove the agitated sediments from the
project area, and (c) no unacceptable environmental impact results from
the agitation dredging.
(4) Advantages. Because currents, not equipment, transport most of
the sediment from the project area during agitation hopper dredging, the
following advantages are realized: (a) hopper dredge agitation costs can
be several times less per cubic yard than hopper dredge hauling costs and
(b) smaller hopper dredges can be used to maintain certain projects.
(5) Limitations. Hopper dredge agitation should be applied only to
specific dredging sites and not used as a general method to maintain large
areas. The following limitations must be noted when considering this
dredging technique for use at a site: (a) hopper dredge agitation cannot
be used in environmentally sensitive areas where unacceptable environmental
impacts may occur and (b) sediments and current conditions must be suitable
for agitation dredging.
d.

Prop-Wash Agitation.

(1) General. Prop-wash agitation dredging is performed by vessels


especially designed or modified to direct propeller-generated currents into
the bottom shoal material. The agitated material is suspended in the water
column and carried away by a combination of natural currents and prop-wash
currents. Unintentional prop-wash agitation dredging often occurs while
vessels move through waterways. This type of sediment resuspension is uncontrolled and is often considered undesirable.
(2) Description of operation. The prop-wash vessel performs best
when work begins at the upstream side of a shoal and proceeds downstream
with the prop-wash-generated current directed downstream. The vessel is
anchored in position, and prop-wash-generated currents are directed into
the shoal material for several minutes. The vessel is then repositioned
and the process in repeated.
(3) Application. Prop-wash agitation dredging has been successfully
used in coastal harbors, river mouths, river channels, and estuaries. It
is a method intended for use in loose sands and in maintenance dredged
material consisting of uncompacted clay and silt. Cementing, cohesion, or
compaction of the bottom sediment can make prop-wash agitation dredging
difficult to perform. Waves may cause anchoring problems with the agitation vessel. Optimum water depths for prop-wash agitation dredging in sand
3-32

EM 1110-2-5025
25 Mar 83
are between two and three times the agitation vessel's draft. Based on
studies by Richardson (item 7), the average performance of vessels specially designed for prop-wash agitation range from 200 to 300 cu yd/hr in
sand and are a little higher for fine-grained material.
(4) Advantages. The major advantages of prop-wash agitation dredging
are related to economics. In some areas, prop-wash agitation dredging has
been found to cost 40 to 90 percent less per cubic yard dredged than conventional dredging methods.
(5) Limitations. The limitations on prop-wash agitation dredging are
as follows: (a) prop-wash agitation seems best suited for areas with little or no wave action, (b) prop-wash agitation should be applied in water
depths less than four times the agitation vessel's draft, and (c) the sediments must be loose sands, silt, or clay.
e. Rakes and Drag Beams. Rakes, drag beams, and similar devices work
by being pulled over the bottom (usually by a vessel), mechanically loosening the bottom material, and raising it into the water column to be carried
away by natural currents. Since rakes and drag beams do not produce currents of their own and since they do not resuspend material as much as
loosen it, these devices must be used in conjunction with currents strong
enough to transport the loosened material away from the shoaling site; in
addition, the vessel towing one of these devices may provide some resuspension and transport by its propwash. A wide range of dredging rates have
been reported for agitation dredging by rakes and beams. Little value
would be obtained by reporting these rates because they are highly dependent upon site conditions; however, it has been reported that the cost
of agitation dredging by rakes and beams can be less than 10 percent of the
cost for conventional dredging. Data show a definite correlation between
dragging speed and dredging rate. The advantages and limitations for rake
and drag beams are similar to those reported for other agitation dredging
techniques.
f. Environmental Considerations. The environmental considerations
discussed in Chapter 4 also apply to all agitation dredging techniques.
The properties of sediments affect the fate of contaminants, and the shortand long-term physical and chemical conditions of the sediments at the agitation dredging site influence the environmental consequences of contaminants. These factors should be considered in evaluating the environmental
risk of a proposed agitation dredging technique.
3-13. Advances in Dredging Technology. Advanced dredging technologies are
generally directed toward one or more of the following areas of improvement:
greater depth capability; greater precision, accuracy, and control over the
dredging process; higher production efficiency; and decreased environmental
harm. Following are brief descriptions of the major recent innovations in
production dredging:
a.

Ladder-mounted submerged pumps for higher production.

3-33

EM 1110-2-5025
25 Mar 83
b. Improved designs of dredging heads to minimize material
resuspension.
c. Use of spud barges aft of the dredge to extend hull length and increase dredge swing. This will increase production efficiency of cutterhead dredges.
d. Longer ladders, connected further aft on the dredge hull to increase depth and permit greater control.
e. Tandem pump systems for greater production efficiency and
reliability.
f. Better hull designs equipped with liquid stabilizing systems
(motion compensators) to allow use in heavier seas.
g. Improved production instrumentation to monitor flow rates, densities, cumulative production, etc.
h. Improved navigation, positioning, and bottom profiling instrumentation. The state of the art includes advanced laser, electronic, and
acoustical systems.
i. Closed-bucket modifications to reduce loss of fines and liquid
from bucket dredges.
j.

Depth and swing indicators for mechanical dredges.

k. Use of silt curtains during dredging and open-water disposal to


restrict turbidity plumes and, in the case of contaminated materials, limit
the added dispersion due to dredging.
3-14. Environmental Considerations. The adverse environmental effects normally associated with dredging operations are increases in turbidity, resuspension of contaminated sediments, and decreases in dissolved oxygen. Selection and operation of the type of dredge plant as well as the type of
sediment being dredged affect the degree of adverse impacts during dredging.
Investigations which have been conducted by WES under the DMRP have studied
the environmental effects caused by dredging and disposal operations. The
results of these studies have been published as WES Technical Reports.
Guidance on the environmental aspects of dredging and disposal is presented
in Chapter 4.

3-34

EM 1110-2-5025
25 Mar 83
CHAPTER 4
DISPOSAL ALTERNATIVES
4-1.

Introduction.

a. While selection of proper dredging equipment and techniques is essential for economic dredging, the selection of a disposal alternative is
of equal or greater importance in determining viability of the project,
especially from the environmental standpoint. There are three major disposal alternatives available:
(1) Open-water disposal.
(2) Confined disposal.
(3) Habitat development.
Each of the major disposal alternatives involves its own set of unique considerations, and selection of a disposal alternative should be made based
on both economic and environmental considerations.
b. This chapter describes considerations in evaluation of disposal
alternatives , primarily from an environmental standpoint. Sections on
evaluation of pollution potential and sediment resuspension due to dredging
apply to all disposal alternatives, while separate sections describe considerations of each of the three major disposal alternatives.
Section I.
4-2.

Evaluation of Dredged Material Pollution Potential

Influence of Disposal Conditions on Environmental Impact.

a. As discussed in WES TR DS-78-6, the properties of a dredged sediment affect the fate of contaminants, and the short- and long-term physical
and chemical environment of the dredged material at the disposal site influences the environmental consequences of contaminants. These factors
should be considered in evaluating the environmental risk of a proposed
disposal method for contaminated sediment. The processes involved with release or immobilization of most sediment-associated contaminants are regulated to a large extent by the physical-chemical environment and the related bacteriological activity associated with the dredged material at the
disposal site. Important physical-chemical parameters include pH,
oxidation-reduction conditions, and salinity. Where the physical-chemical
environment of a contaminated sediment is altered by disposal, chemical and
biological processes important in determining environmental consequences of
potentially toxic materials may be affected.
b. The major sediment properties that will influence the reaction of
dredged material with contaminants are the amount and type of clay; organic
matter content; amount and type of cations and anions associated with the
sediment; the amount of potentially reactive iron and manganese; and the
oxidation-reduction, pH, and salinity conditions of the sediment. Although
each of these sediment properties is important, much concerning the release
4-1

EM 1110-2-5025
25 Mar 83
of contaminants from sediments can be inferred from the clay and organic
matter content, initial and final pH, and oxidation-reduction conditions.
Much of the dredged material removed during harbor and channel maintenance
dredging is high in organic matter and clay and is both biologically and
chemically active. It is usually devoid of oxygen and may contain appreciable sulfide. These sediment conditions favor effective retention of
many contaminants, provided the dredged materials are not subject to mixing,
resuspension, and transport. Sandy sediments low in organic matter content
are much less effective in retaining metal and organic contaminants. These
materials tend not to accumulate contaminants unless a contamination source
is nearby. Should contamination of these sediments occur, potentially
toxic substances may be readily released upon mixing in a water column, or
by leaching and possibly plant uptake under intertidal or upland disposal
conditions.
c. Many contaminated sediments are reducing and near neutral in pH,
initially.
Disposal into quiescent waters will generally maintain these
conditions and favor contaminant retention. Certain sediments (noncalcareous and containing appreciable reactive iron and particularly reduced
sulfur compounds) may become moderately to strongly acid upon gradual
drainage and subsequent oxidation as may occur under upland disposal conditions. This altered disposal environment greatly increases the potential
for releasing potentially toxic metals. In addition to the effects of pH
changes, the release of most potentially toxic metals is influenced to some
extent by oxidation-reduction conditions, and certain of the metals can be
strongly affected by oxidation-reduction conditions. Thus, contaminated
sandy, low organic-matter-content sediments pose the greatest potential for
release of contaminants under all conditions of disposal. Sediments which
tend to become strongly acid upon drainage and long-term oxidation also
pose a high environmental risk under some disposal conditions. The implications of the influence of disposal conditions on contaminant mobility are
discussed below.
4-3.

Methods of Characterizing Pollution Potential.

a. Bioassay. Bioassay tests are used to determine the effects of a


contaminant(s) on biological organisms of concern. They involve exposure
of the test organisms to dredged material (or some fraction such as the
elutriate) for a specified period of time, followed by determination of the
response of the organisms. The most common response of interest is death.
Often the tissues of organisms exposed to dredged material are analyzed
chemically to determine whether they have incorporated, or bioaccumulated,
any contaminants from the dredged material. Bioassays provide a direct indication of the overall biological effects of dredged material. They reflect the cumulative influence of all contaminants present, including any
possible interactions of contaminants. Thus, they provide an integrated
measurement of potential biological effects of a dredged material discharge.
For precisely these reasons, however, a bioassay cannot be used to identify
the causative agent(s) of impact in a dredged material. This is of interest, but is seldom of importance, since usually the dredged material cannot be treated to remove the adverse components even if they could be identified. Dredged material bioassay techniques for aquatic animals have been
4-2

EM 1110-2-5025
25 Mar 83
implemented in the ocean-dumping regulatory program for several years
(item 1) and are easily adapted for use in fresh water. Dredged material
bioassays for wetland and terrestrial plants have also been developed
(item 2) and are coming into ever-wider use.
b. Water Column Chemistry. Chemical constituents contained in or associated with sediments are unequally distributed among different chemical
forms depending on the physical-chemical conditions in the sediments and
the overlying water. When contaminants introduced into the water column
become fixed into the underlying sediments, they rarely if ever become part
of the geological mineral structure of the sediment. Instead, these contaminants remain dissolved in the sediment interstitial water, or pore
water, become absorbed or adsorbed to the sediment ion exchange portion as
ionized constituents, form organic complexes, and/or become involved in complex sediment oxidation-reduction reactions and precipitations. The fraction of a chemical constituent that is potentially available for release to
the water column when sediments are disturbed is approximated by the interstitial water concentrations and the loosely bound (easily exchangeable)
fraction in the sediment. The elutriate test is a simplified simulation of
the dredging and disposal process wherein predetermined amounts of dredging
site water and sediment are mixed together to approximate a dredged material
slurry. The elutriate is analyzed for major dissolved chemical constituents
deemed critical for the proposed dredging and disposal site after taking
into account known sources of discharges in the area and known characteristics of the dredging and disposal site. Results of the analysis of the
elutriate approximate the dissolved constituent concentration for a proposed dredged material disposal operation at the moment of discharge. These
concentrations can be compared to water quality standards and mixing zone
considerations to evaluate the potential environmental impact of the proposed discharge activity in the discharge area.
c. Total or Bulk Sediment Chemistry. The results of these analyses
provide some indication of the general chemical similarity between the
sediments to be dredged and the sediments at the proposed disposal site.
The total composition of sediments, when compared with natural background
levels at the site, will also, to some extent, reflect the inputs to the
waterway from which they were taken and may sometimes be used to identify
and locate point source discharges. Since chemical constituents are partitioned among various sediment fractions, each with its own mobility and biological availability, a total sediment analysis is not a useful index of
the degree to which dredged material disposal will affect water quality or
aquatic organisms. Total sediment analysis results are further limited because they cannot be compared to any established water quality criteria in
order to assess the potential environmental impact of discharge operations.
This is because the water quality criteria are based on water-soluble
chemical species, while chemical constituents associated with dredged material suspensions are generally in particulate/solid-phase forms or mineralogical forms that have markedly lower toxicities, mobilities, and chemical
reactivities than the solution-phase constituents. Consequently, little
information about the biological effects of solid-phase and mineral constituents that make up the largest fraction of dredged material can be gained
from total or bulk sediment analysis.
4-3

EM 1110-2-5025
25 Mar 83
Section II.
4-4.

Sediment Resuspension Due to Dredging

Factors Influencing Dredging Turbidity.

a. Occurrence and Extent. The nature, degree, and extent of sediment


suspension around a dredging or disposal operation are controlled by many
factors, as discussed in WES TR DS-78-13. Chief among these are: the particle size distribution, solids concentration, and composition of the
dredged material; the dredge type and size, discharge/cutter configuration,
discharge rate, and solids concentration of the slurry; operational procedures used; and finally the characteristics of the hydraulic regime in the
vicinity of the operation, including water composition, temperature and hydrodynamic forces (i.e., waves, currents, etc.) causing vertical and horizontal mixing. The relative importance of the different factors may vary
significantly from site to site.
b. Hopper Dredge. Resuspension of fine-grained maintenance dredged
material during hopper dredging operations is caused by the dragheads as
they are pulled through the sediment, turbulence generated by the vessel
and its prop wash, and overflow of turbid water during hopper filling operations. During the filling operation, dredged material slurry is often
pumped into the hoppers after they have been filled with slurry in order to
maximize the amount of solid material in the hopper. The lower density,
turbid water at the surface of the filled hoppers overflows and is usually
discharged through ports located near the waterline of the dredge. In the
vicinity of hopper dredges during maintenance operations, a near-bottom
turbidity plume of resuspended bottom material may extend 2300 to 2400 ft
downcurrent from the dredge. In the immediate vicinity of the dredge, a
well-defined, upper plume is generated by the overflow process. Approximately 1000 ft behind the dredge the two plumes merge into a single plume
(fig. 4-1). Suspended solid concentrations above ambient may be as high as

DREDGE

Figure 4-1. Hypothetical suspended solids plume downstream of a hopper dredge operation with overflow in
San Francisco Bay (all distances in feet)*
4-4

EM 1110-2-5025
25 Mar 83
several tens of parts per thousand (grams per litre) near the discharge port
and as high as a few parts per thousand near the draghead. Turbidity
levels in the near-surface plume appear to decrease exponentially with increasing distance from the dredge due to settling and dispersion, quickly
reaching concentrations less than 1 ppt. However, plume concentrations may
exceed background levels even at distances in excess of 4000 ft.
c. Bucket or Clamshell Dredge. The turbidity generated by a typical
clamshell operation can be traced to sediment resuspension occurring when
the bucket impacts on and is pulled off the bottom, turbid water spills
out of the bucket or leaks through openings between the jaws, and material
is inadvertently spilled during the barge loading operation. There is a
great deal of variability in the amount of material resuspended by clamshell dredges due to variations in bucket size, operating conditions, sediment types, and hydrodynamic conditions at the dredging site. Based on
limited measurements, it appears that, depending on current velocities, the
turbidity plume downstream of a typical clamshell operation may extend approximately 1000 ft at the surface and 1600 ft near the bottom. Maximum
concentrations of suspended solids in the surface plume should be less
than 0.5 ppt in the immediate vicinity of the operation and decrease
rapidly with distance from the operation due to settling and dilution of
the material. Average water-column concentrations should generally be less
than 0.1 ppt. The near-bottom plume will probably have a higher solids
concentration, indicating that resuspension of bottom material near the
clamshell impact point is probably the primary source of turbidity in the
lower water column. The visible near-surface plume will probably dissipate
rapidly within an hour or two after the operation ceases.
d. Cutterhead or Hydraulic Pipeline Dredge. Most of the turbidity
generated by a cutterhead dredging operation is usually found in the vicinity of the cutter. The levels of turbidity are directly related to the
type and quantity of material cut, but not picked up, by the suction. The
ability of the dredge's suction to pick up bottom material determines the
amount of cut material that remains on the bottom or suspended in the water
column. In addition to the dredging equipment used and its mode of operation, turbidity may be caused by sloughing of material from the sides of
vertical cuts; inefficient operational techniques; and the prop wash from
the tenders (tugboats) used to move pipeline, anchors, etc., in the shallow
water areas outside the channel. Based on limited field data collected
under low current conditions, elevated levels of suspended material appear
to be localized in the immediate vicinity of the cutter as the dredge
swings back and forth across the dredging site. Within 10 ft of the cutter,
suspended solids concentrations are highly variable but may be as high as
a few tens of parts per thousand; these concentrations decrease exponentially from the cutter to the water surface. Near-bottom suspended solids
concentrations may be elevated to levels of a few tenths of a part per
thousand at distances of less than 1000 ft from the cutter.

4-5

EM 1110-2-5025
25 Mar 83
Section III.
4-5.

Open-Water Disposal

Behavior of Discharges from Various Types of Dredges.

a. Hopper Dredge. The characterisitics and operation of hopper


dredges are discussed in para 3-3 of this manual. When the hoppers have
been filled as described, the dragarms are raised and the hopper dredge proceeds to the disposal site. At the disposal site, hopper doors in the bottom of the ship's hull are opened and the entire hopper contents are emptied
in a matter of seconds; the dredge then returns to the dredging site to reload. This procedure produces a series of discrete discharges at intervals
of perhaps one to several hours. Upon release from the hopper dredge at
the disposal site, the dredged material falls through the water column as a
well-defined jet of high-density fluid which may contain blocks of solid
material. Ambient water is entrained during descent. After it hits bottom,
some of the dredged material comes to rest. Some material enters the horizontally spreading bottom surge formed by the impact and is carried away
from the impact point until the turbulence of the surge is sufficiently reduced to permit its deposition.
b. Bucket or Clamshell Dredge. Bucket dredges remove the sediment
being dredged at nearly its in
- situ- density and place it in barges or scows
for transportation to the disposal area, as described in para 3-8. Although several barges may be used so that the dredging is essentially continuous, disposal occurs as a series of discrete discharges. The dredged
material may be a slurry similar to that in a hopper dredge, but often sediments dredged by clamshell remain in fairly large consolidated clumps and
reach the bottom in this form. Whatever its form, the dredged material descends rapidly through the water column to the bottom, and only a small
amount of the material remains suspended.
c. Cutterhead or Hydraulic Pipeline Dredge. The operation of a cutterhead dredge, described in para 3-4, produces a slurry of sediment and water
discharged at the disposal site in a continuous stream. As the dredge progresses up the channel, the pipeline is moved periodically to keep abreast
of the dredge. The discharged dredged material slurry is generally dispersed in three modes. Any coarse material, such as gravel, clay balls, or
coarse sand, will immediately settle to the bottom of the disposal area and
usually accumulates directly beneath the discharge point. The vast majority of the fine-grained material in the slurry also descends rapidly to the
bottom in a well-defined jet of high-density fluid, where it forms a lowgradient circular or elliptical fluid mud mound. Approximately 1 to 3 percent of the discharged material is stripped away from the outside of the
slurry jet as it descends through the water column and remains suspended as
a turbidity plume.
4-6.

Dredged Material Dispersion at the Discharge Site.

a. Water-Column Turbidity. The levels of suspended solids in the


water column around a discharge operation generally range from a few hundredths to a few tenths of a part per thousand. Concentrations are highest
near the discharge point and rapidly decrease with increasing distance
4-6

EM 1110-2-5025
25 Mar 83
downstream from the discharge point and laterally away from the plume center line due to settling and horizontal dispersion of the suspended solids.
Concentrations also decrease rapidly between each discrete hopper or barge
discharge and after a pipeline is shut down or moved to a new location.
Under tidal conditions, the plume will be subject to the tidal dynamics of
the particular bay, estuary, or river mouth in which the dredging activity
takes place. Many of the Corps projects have been studied in physical hydraulic models, and estimates of plume excursion can be made from their
model reports. Rough estimates can be made from numerical models. Mathematical model result can be materially improved when calibrated by physical
and/or prototype data; except under very simple conditions, all models have
to be verified with prototype or prototype-derived data. In rivers where
the flow is unidirectional, the plume length is controlled by the strength
of the current and the settling properties of the suspended material. In
both estuarine and riverine environments the natural levels of turbulence
and the fluctuations in the rate of slurry discharge will usually cause the
idealized teardrop-shaped plume to be distorted by gyres or eddylike patterns, as in figure 4-2.
b. Fluid Mud. A small percentage of the fine-grained dredged material slurry discharged during open-water disposal is dispersed in the water
column as a turbidity plume; however, the vast majority rapidly descends to
the bottom of the disposal area where it accumulates under the discharge
point in the form of a low-gradient fluid mud mound overlying the existing
bottom sediment, as shown in figure 4-3. If the discharge point of a hydraulic pipeline dredge is moved as the dredge advances, a series of mounds
will develop. The majority of the mounded material is usually high-density
(nonflowing) fluid mud that is covered by a surface layer of low-density
(flowing or nonflowing) fluid mud. Under quiescent conditions, more than
98 percent of the sediment in the mudflow remains in the fluid mud layer at
concentrations greater than 10 ppt, while the remaining 2 percent may be
resuspended by mixing with the overlying water at the fluid mud surface.
Fluid mud will tend to flow downhill as long as the bottom slope is approximately 1 percent or greater. A study of hopper dredge disposal at
Carquinez Strait, San Francisco Bay, showed concentrations of dredged material in the water column were generally less than 0.2 ppt above background
and persisted for only a few tens of minutes. However, 3 to 8 ft above the
bottom, concentrations reached 20 ppt in a fluid mud layer. Similar occurrences of low suspended sediment concentrations in the water column with
concentrations on the order of several tens of parts per thousand just
above the bottom, as in figure 4-4, have been discussed for pipeline dredge
discharges in WES TR DS-78-13. These conditions persist for the duration
of the disposal operation at the site and for varying times thereafter as
the material consolidates to typical sediment density.
c. Mounding. If bottom slopes are not great enough to maintain mudflows, the fluid mud will stop and begin to consolidate. When suspended
sediment concentrations exceed 200 ppt the fluid mud can no longer flow
freely but will accumulate around the discharge point in a low-gradient
(e.g., 1:500) fluid mud mound. At the water column/fluid mud interface,
the solids concentration increases very abruptly from perhaps a few tenths
of a part per thousand in the water to approximately 200 ppt in the fluid
4-7

EM 1110-2-5025
25 Mar 83

Figure 4-2. Middepth (3.0 ft) turbidity plume generated by


a 28-in. pipeline disposal operation in the Atchafalaya Bay.
Current flow is generally toward the northeast.

4-8

EM 1110-2-5025
25 Mar 83

NO CURRENT

C.

PREDOMINANT
CURRENT
VERTICAL
Figure 4-3.

DISCHARGE

D
HORIZONTAL

DISCHARGE

Effect of discharge angle and predominant current direction


on the shape of a fluid mud mound.

4-9

EM 1110-2-5025
25 Mar 83

Figure 4-4. Relationship between suspended solids concentration


along the plume center line and distance downcurrent from several
open-water pipeline disposal operations measured at the indicated
water depths,

4-10

EM 1110-2-5025
25 Mar 83

mud. The solids concentration within the fluid mud increases above 200 ppt
at a slower rate with depth until it reaches normal sediment densities.
Deeper layers of fluid mud reach their final degree of consolidation more
rapidly than thinner ones. Depending on the thickness of the fluid mud and
its sedimentation/consolidation characteristics, complete consolidation of
a fluid mud mound may require from one to several years. In those situations where material dredged by bucket or clamshell is of slurry consistency, the above description is generally applicable. More commonly, however, muddy sediments dredged by a clamshell remain in large clumps and
descend to the bottom in this form. These may break apart somewhat on impact; but such material tends to accumulate in irregular mounds under the
discharge vessel, rather than move outward from the discharge point. Whatever the dredging method, sandy sediments tend to mound directly beneath
the discharge pipe or vessel.
d. Special Circumstances. Knowledge of the behavior of discharged
dredged material allows control of the dispersion of the material at the
disposal site. When minimum dispersal is desired, the dredged material can
be discharged into old underwater borrow pits, sand or gravel excavation
sites, etc. Such deposits may be further isolated from the overlying water
column by covering with a layer of uncontaminated sediment. It is also
possible to place such a covering, or "cap," over dredged material discharged onto a flat bottom.
4-7.

Environmental Impacts in the Water Column.

a. Contaminants. Although the vast majority of heavy metals, nutrients, and petroleum and chlorinated hydrocarbons are usually associated
with the fine-grained and organic components of the sediment (see WES
TR DS-78-4), there is no biologically significant release of these chemical
consituents from typical dredged material to the water column during or
after dredging or disposal operations. Levels of manganese, iron, ammonium
nitrogen, orthophosphate, and reactive silica in the water column may be
increased somewhat for a matter of minutes over background conditions during open-water disposal operations; however, there are no persistent welldefined plumes of dissolved metals or nutrients at levels significantly
greater than background concentrations.
b. Turbidity. There are now ample research results indicating that
the traditional fears of water-quality degradation resulting from the resuspension of dredged material during dredging and disposal operations are
for the most part unfounded. The possible impact of depressed levels of
dissolved oxygen has also been of some concern, due to the very high oxygen
demand associated with fine-grained dredged material slurry. However, even
at open-water pipeline disposal operations where the dissolved oxygen decrease should theoretically be greatest, near-surface dissolved oxygen
levels of 8 to 9 ppm will be depressed during the operation by only 2 to
3 ppm at distances of 75 to 150 ft from the discharge point. The degree of
oxygen depletion generally increases with depth and increasing concentration of total suspended solids; near-bottom levels may be less than 2 ppm.
However, dissolved oxygen levels usually increase with increasing distance
4-11

EM 1110-2-5025
25 Mar 83
from the discharge point, due to dilution and settling of the suspended
material.
(1) It has been demonstrated that elevated suspended solids concentrations are generally confined to the immediate vicinity of the dredge or discharge point and dissipate rapidly at the completion of the operation. If
turbidity is used as a basis for evaluating the environmental impact of a
dredging or disposal operation, it is essential that the predicted turbidity
levels are evaluated in light of background conditions. Average turbidity
levels, as well as the occasional relatively high levels that are often associated with naturally occurring storms, high wave conditions, and floods,
should be considered.
(2) Other activities of man may also be responsible for generating as
much or more turbidity than dredging and disposal operations. For example,
each year shrimp trawlers in Corpus Christi Bay, Texas, suspend 16 to 131
times the amount of sediment that is dredged annually from the main ship
channel. In addition, suspended solids levels of 0.1 to 0.5 ppt generated
behind the trawlers are comparable to those levels measured in the turbidity plumes around open-water pipeline disposal operations. Resuspension
of bottom sediment in the wake of large ships, tugboats, and tows can also
be considerable. In fact, where bottom clearance is 3 ft or less, there
may be scour to a depth of 3 ft if the sediment is easily resuspended.
4-8.

Environmental Impacts on the Benthos.

a. Physical. Whereas the impact associated with water-column turbidity around dredging and disposal operations is for the most part insignificant, the dispersal of fluid mud dredged material appears to have a
relatively significant short-term impact on the benthic organisms within
open-water disposal areas. Open-water pipeline disposal of fine-grained
dredged material slurry may result in a substantial reduction in the average abundance of organisms and a decrease in the community diversity in the
area covered by fluid mud. Despite this immediate impact, recovery of the
community apparently begins soon after the disposal operation ceases.
(1) Disposal operations will blanket established bottom communities
at the site with dredged material which may or may not resemble bottom sediments at the disposal site. Recolonization of animals on the new substrate
and the vertical migration of benthic organisms in newly deposited sediments
can be important recovery mechanisms. The first organisms to recolonize
dredged material usually are not the same as those which had originally occupied the site; they consist of opportunistic species whose environmental
requirements are flexible enough to allow them to occupy the disturbed
areas. Trends toward reestablishment of the original community are often
noted within several months of disturbance, and complete recovery approached
within a year or two. The general recolonization pattern is often dependent
upon the nature of the adjacent undisturbed community, which provides a
pool of replacement organisms capable of recolonizing the site by adult migration or larval recruitment.

4-12

EM 1110-2-5025
25 Mar 83
(2) Organisms have various capabilities for moving upward through
newly deposited sediments, such as dredged material, to reoccupy positions
relative to the sediment-water interface similar to those maintained prior
to burial by the disposal activity. Vertical migration ability is greatest
in dredged material similar to that in which the animals normally occur and
is minimal in sediments of dissimilar particle-size distribution. Bottomdwelling organisms having morphological and physiological adaptations for
crawling through sediments are able to migrate vertically through several
inches of overlying sediment. However, physiological status and environmental variables are of great importance to vertical migration ability.
Organisms of similar life-style and morphology react similarly when covered
with an overburden. For example, most surface-dwelling forms are generally
killed if trapped under dredged material overburdens, while subsurface
dwellers migrate to varying degrees. Laboratory studies suggest vertical
migration may very well occur at disposal sites, although field evidence is
not available. Literature review (WES TR DS-78-1) indicates the vertical
migration phenomenon is highly variable among species.
(3) Dredging and disposal operations have immediate localized effects
on the bottom life. The recovery of the affected sites occurs over periods
of weeks, months, or years, depending on the type of environment and the
biology of the animals and plants affected. The more naturally variable
the physical environment, especially in relation to shifting substrate due
to waves or currents, the less effect dredging and disposal will have.
Animals and plants common to such areas of unstable sediments are adapted
to physically stressful conditions and have life cycles which allow them to
withstand the stresses imposed by dredging and disposal. Exotic sediments
(those in or on which the species in question does not normally live) are
likely to have more severe effects when organisms are buried than sediments
similar to those of the disposal site. Generally, physical impacts are
minimized when sand is placed on a sandy bottom and are maximized when mud
is deposited over a sand bottom. When disposed sediments are dissimilar to
bottom sediments at the sites, recolonization of the dredged material will
probably be slow and carried out by organisms whose life habits are adapted
to the new sediment. The new community may be different from that originally occurring at the site.
(4) Dredged material discharged at disposal sites which have a naturally unstable or shifting substrate due to wave or current action is
rather quickly dispersed and does not cover the area to substantial depths.
This natural dispersion, which usually occurs most rapidly and effectively
during the stormy winter season, can be assisted by conducting the disposal
operation so as to maximize the spread of dredged material, producing the
thinnest possible overburden. The thinner the layer of overburden, the
easier it is for mobile organisms to survive burial by vertical migration
through dredged material. The desirability of minimizing physical impacts
by dispersion can be overridden by other considerations, however. For example, dredged material shown by biological or chemical testing to have a
potential for adverse environmental impacts might best be placed in an area
of retention, rather than dispersion. This would maximize habitat disruption in a restricted area, but would confine potentially more important
chemical impacts to tha same small area.
4-13

EM 1110-2-5025
25 Mar 83
(5) Since larval recruitment and migration of adults are primary
mechanisms of recolonization, recovery from physical impacts will generally
be most rapid if disposal operations are completed shortly before the seasonal increase in biological activity and larval abundance in the area.
The possibility of impacts can also be reduced by locating disposal sites
in the least sensitive or critical habitats. This can sometimes be done on
a seasonal basis. Known fish migratory routes and spawning beds should be
avoided just before and during use, but might be acceptable for disposal
during other periods of the year. However, care must be taken to ensure
that the area returns to an acceptable condition before the next intensive
use by the fish. Clam or oyster beds, municipal or industrial water intakes, highly productive backwater areas, etc., should be avoided in selecting disposal sites.
(6) All the above factors should be evaluated in selecting a disposal
site, method, and season in order to minimize the habitat disruption of disposal operations. All require evaluations on a case-by-case basis by persons familiar with the ecological principles involved, as well as the characteristics of the proposed disposal operations and the local environment.
b.

Contaminants.

(1) Dredging and disposal do not introduce new contaminants to the


aquatic environment, but simply redistribute the sediments which are the
natural depository of contaminants introduced from other sources. The potential for accumulation of a metal in the tissues of an organism (bioaccumulation) may be affected by several factors such as duration of exposure,
salinity, water hardness, exposure concentration, temperature, the chemical
form of the metal, and the particular organism under study. The relative
importance of these factors varies from metal to metal, but there is a
trend toward greater uptake at lower salinities. Elevated concentrations
of heavy metals in tissues of benthic invertebrates are not always indicative of high levels of metals in the ambient medium or associated sediments.
Although a few instances of uptake of possible ecological significance have
been shown, the diversity of results among species, different metals, types
of exposure, and salinity regimes strongly argues that bulk analysis of
sediments for metal content cannot be used as a reliable index of metal
availability and potential ecological impact of dredged material, but only
as an indicator of total metal context. Bioaccumulation of most metals
from sediments is generally minor. Levels often vary from one sample period to another and are quantitatively marginal, usually being less than
one order of magnitude greater than levels in the control organisms, even
after one month of exposure. Animals in undisturbed environments may
naturally have high and fluctuating metal levels. Therefore, in order to
evaluate bioaccumulation, comparisons should be made between control and
experimental organisms at the same point in time.
(2) Organochlorine compounds such as DDT, dieldrin, and polychlorinated biphenyls (PCB's) are environmental contaminants of worldwide significance which are manmade and, therefore, do not exist naturally in the
earth's crust. Organochlorine compounds are generally not soluble in surface waters at concentrations higher than approximately 20 ppb, and most of
4-14

EM 1110-2-5025
25 Mar 83
the amount present in waterways is associated with either biological organisms or suspended solids. Organochlorine compounds are released from sediment until some equilibrium concentration is achieved between the aqueous
and the solid phases and then readsorbed by other suspended solids or biological organisms in the water column. The concentration of organochlorines
in the water column is reduced to background levels within a matter of hours
as the organochlorine compounds not taken up by aquatic organisms eventually
settle with the particulate matter and become incorporated into the bottom
deposits in aquatic ecosystems. Most of these compounds are stable and may
accumulate to relatively high concentrations in the sediments. The manufacture and/or disposal of most of these compounds is now severely limited;
however, sediments that have already been contaminated with organochlorine
compounds will probably continue to have elevated levels of these compounds
for several decades. The low concentrations of chlorinated hydrocarbons in
sediment interstitial water indicate that during dredging operations, the
release of the interstitial water and contaminants to the surrounding environment would not create environmental problems. Bioaccumulation of
chlorinated hydrocarbons from deposited sediments does occur. However, the
sediments greatly reduce the bioavailability of these contaminants, and tissue concentrations may range from less than one to several times the sediment concentration. Unreasonable degradation of the aquatic environment
due to the routine maintenance dredging and disposal of sediment contaminated with chlorinated hydrocarbons has never been demonstrated.
(3) The term oil and grease is used collectively to describe all
components of sediments of natural and contaminant origin which are primarily fat soluble. There is a broad variety of possible oil and grease
components in sediment, the analytical quantification of which is dependent
on the type of solvent and method used to extract these residues. Trace
contaminants, such as PCBs and chlorinated hydrocarbons, often occur in
the oil and grease. Large amounts of contaminant oil and grease find their
way into the sediments of the Nations waterways either by spillage or as
chronic inputs in municipal and industrial effluents, particularly near
urban areas with major waste outfalls. The literature suggests long-term
retention of oil and grease residues in sediments, with minor biodegradation occurring. Where oily residues of known toxicity became associated
with sediments, these sediments retained toxic properties over periods of
years, affecting local biota. Spilled oils are known to readily become adsorbed to naturally occurring suspended particulates, and oil residues in
municipal and industrial effluents are commonly found adsorbed to particles.
These particulates are deposited in sediments and are subject to suspension
during disposal. Even so, there is only slight desorption, and the amount
of oil released during the elutriate test is less than 0.01 percent of the
sediment-associated hydrocarbons under worst-case conditions. Selected
estuarine and freshwater organisms exposed for periods up to 30 days to
dredged material that is contaminated with thousands of parts per million
of oil and grease experience minor mortality. Uptake of hydrocarbons from
heavily contaminated sediments appears minor when compared with the hydrocarbon content of the test sediments.
(4) Ammonia is one of the potentially toxic materials known to be released from sediments during disposal; it is routinely found in evaluations
4-15

EM 1110-2-5025
25 Mar 83
of sediments using the elutriate test and in the water near a disposal area
where concentrations rapidly return to baseline levels. Similar temporary
increases in ammonia at marine, estuarine, and freshwater disposal sites
have been documented in several DMRP field studies, but concentrations and
durations are usually well below levels causing concern.
(5) The potential environmental impact of contaminants associated
with sediments must be evaluated in light of chemical and biological data
describing the availability of contaminants to organisms. Information must
then be gained as to the effects of specific substances on organism survival
and function. Many contaminants are not readily released from sediment attachment and are thus less toxic than contaminants in the free or soluble
state on which most toxicity data are based.
(6) There are now cogent reasons for rejecting many of the conceptualized impacts of disposed dredged material based on classical bulk analysis
determinations.
It is invalid to use total sediment concentration to estimate contaminant levels in organisms since only a variable and undetermined
amount of sediment-associated contaminant is biologically available. Although a few instances of toxicity and bioaccumulation of possible ecological consequence have been seen, the fact that the degree of effect depends
on species, contaminants, salinity, sediment type, etc., argues strongly
that bulk analysis does not provide a reliable index of contaminant availability and potential ecological impact of dredged material.
4-9.

Overview of Open-Water Disposal.

a. Prediction of physical effects of dredging and disposal is fairly


straightforward. Physical effects include removal of organisms at dredging
sites and burial of organisms at disposal sites. Physical effects are restricted to the immediate areas of dredging or disposal. Recolonization of
sites occurs in periods of months to 1-2 years in case studies. Disturbed
sites may be recolonized by opportunistic species which are not normally
the dominant species occurring at the site.
b. Many organisms are very resistant to the effects of sediment suspensions in the water; aside from natural systems requiring clear water,
such as coral reefs and some aquatic plant beds, dredging or disposalinduced turbidity is not of major ecological concern. The formation of
fluid muds due to disposal is not fully understood and is of probable environmental concern in some situations.
c. Release of sediment-associated heavy metals and chlorinated hydrocarbons to the water column by dredging and disposal has been found to be
the exception, rather than the rule. Metals are rarely bioaccumulated from
sediments and then only to low levels. Chlorinated hydrocarbons may be bioaccumulated from sediments, but only very highly contaminated sediments
might result in tissue concentrations of potential concern. There is little
or no correlation between bulk analysis of sediments for contaminants and
their environmental impact.

4-16

EM 1110-2-5025
25 Mar 83
d. Oil and grease residues, like heavy metals, are tightly bound to
sediment particles, and there appears to be minimal uptake of the residues
into organism tissues. Of the thousands of chemicals constituting the oil
and grease fraction, very few can be considered significant threats to
aquatic life when associated with dredged material.
e. Many laboratory studies describe worst-case experimental conditions where relatively short-term exposures to high concentrations of sediments and contaminants are investigated. Although limited in scope, experimental results showing the lack of effects under these worst-case conditions
support the conclusion that the indirect long-term and sublethal effects of
dredging and disposal will be minimal. An integrated, whole-sediment bioassay using sensitive test organisms should be used to determine potential
sediment impacts at a particular site. Appropriate chemical testing and
biological evaluation of the dredged material can be used to resolve any
site-specific problems which may occur.
Section IV.
4-10.

Confined Dredged Material Disposal

Containment Area Design.


a.

Concepts of Containment Area Operation.

(1) Diked containment areas are used to retain dredged material


solids while allowing the carrier water to be released from the containment
area. The two objectives of a containment area are: (a) to provide adequate storage capacity to meet dredging requirements and (b) to attain the
highest possible efficiency in retaining solids during the dredging operation in order to meet effluent suspended solids requirements. These considerations are interrelated and depend upon effective design, operation,
and management of the containment area. Major considerations in design of
containment areas are discussed below. Detailed design guidance may be
found in WES TR DS-78-10.
(2) The major components of a dredged material containment area are
shown schematically in figure 4-5. A tract of land is surrounded by dikes
to form a confined surface area into which dredged channel sediments are
pumped hydraulically. In some dredging operations, especially in the case
of new work dredging, sand, clay balls, and/or gravel may be present. Thi
coarse material rapidly falls out of suspension and forms a mound near the
dredge inlet pipe. The fine-grained material (silt and clay) continues to
flow through the containment area where most of the solids settle out of
suspension and thereby occupy a given storage volume. The fine-grained
dredged material is usually rather homogeneous and is easily characterized
The clarified water is discharged from the containment area over a weir.
This effluent can be characterized by its suspended solids concentration
and rate of outflow. Effluent flow rate is approximately equal to influent
flow rate for continuously operating disposal areas. To promote effective
sedimentation, ponded water is maintained in the area; the depth of water
is controlled by the elevation of the weir crest. The thickness of the
dredged material layer increases with time until the dredging operation is
completed.
Minimum freeboard requirements and mounding of coarse-grained
4-17

EM 1110-2-5025
25 Mar 83

Figure 4-5.

Schematic diagram of a dredged material containment area,

material result in a ponded surface area smaller than the total surface area
enclosed by the dikes. In most cases, confined disposal areas must be utilized over a period of many years, storing material dredged periodically
over the design life. Long-term storage capacity for these sites is influenced by consolidation of dredged material and foundation soils, dewatering
of material, and effective management of the disposal area.
b. Evaluation of Dredging Activities. Effective planning and design
of containment areas first requires a thorough evaluation of the dredging
program. The location, volumes, frequencies, and types of material to be
dredged must be estimated. The number, types, and sizes of dredges normally
employed to do the work should also be considered. This information is important for defining project objectives and provides a basis for containment
area design.
4-18

EM 1110-2-5025
25 Mar 83
c.

Field Investigations.

(1) Samples of the channel sediments to be dredged are required for


adequate characterization of the material and for use in sedimentation and
consolidation testing. The level of effort required for channel sediment
sampling depends upon the project. In the case of routine maintenance work,
data from prior samplings and experience with similar material may be available to reduce the scope of field investigations. Since maintenance sediments are in an essentially unconsolidated state, grab samples are normally
satisfactory for sediment characterization purposes and are easy and inexpensive to obtain. For unusual maintenance projects or new work, more extensive field investigations will be required.
(2) Field investigations must also be performed at the containment
area site to define foundation conditions and to obtain samples for laboratory testing if estimates of long-term storage capacity are required. The
extent of required field investigations is dependent upon project size and
upon foundation conditions at the site. It is particularly important to
define foundation conditions, including depth, thickness, extent, and composition of foundation strata, and to obtain undisturbed samples of compressible foundation soils and any previously placed dredged material. If possible, the field investigations required for estimating long-term storage
capacity should be planned and accomplished along with those required for
the engineering design of the retaining dikes.
d.

Laboratory Testing.

(1) Laboratory tests are required primarily to provide data for sediment characterization, containment area design, retention dike design, and
long-term storage capacity estimates. The laboratory tests and procedures
required are essentially standard tests and generally follow accepted procedures. The required magnitude of the laboratory testing program depends
upon the project. Fewer tests are usually required when dealing with a
relatively homogeneous material and/or when data are available from previous tests and experience, as is frequently the case in maintenance work.
For unusual maintenance projects where considerable variation in sediment
properties is apparent from samples, or for new work projects, more extensive laboratory testing programs are required. Refer to WES TR DS-78-10
for details on testing procedures.
(2) Sedimentation tests, performed in 8-in.-diameter ported columns
as shown in figure 4-6, are necessary to provide design data for retention
of suspended solids (item 4). These tests are designed to define the flocculent or zone-settling behavior of a particular sediment and to provide
information concerning the volumes occupied by newly placed layers of
dredged material. Sedimentation of freshwater sediments at slurry concentrations of less than 100 ppt can generally be characterized by flocculent
settling properties. As slurry concentrations are increased, the sedimentation process may be characterized by zone-settling properties. Salinity
greater than 3 ppt enhances the flocculation of dredged material particles;
therefore, the settling properties of saltwater dredged material can usually
be characterized by zone-settling tests. The flocculent settling test
4-19

EM 1110-2-5025
25 Mar 83

Figure 4-6.

Schematic of apparatus for settling tests.

consists of measuring the concentration of suspended solids at various


depths and time intervals by withdrawing samples from the settling column
ports. The zone-settling test consists of placing a slurry in a settling
column and timing the fall of the liquid-solids interface.
(3) Determination of containment area long-term storage capacity requires estimates of settlement due to self-weight consolidation of newly
placed dredged material and due to consolidation of compressible foundation
soils. Consolidation test results, including time-consolidation data, must
therefore be obtained. Consolidation tests for foundation soils should be
4-20

EM 1110-2-5025
25 Mar 83
performed as described in EM 1110-2-1906 with no modifications. The consolidation testing procedure for sediment samples generally follows that
for the fixed ring test for conventional soils, but minor modifications in
sample preparation and loading are required (WES TR DS-78-10).
e.

Design for Retention of Suspended Solids.

(1) Sedimentation, as applied to dredged material disposal activities,


refers to those operations in which the dredged material slurry is separated into more clarified water and a more concentrated slurry. Laboratory
sedimentation tests must provide data for designing the containment area to
meet effluent suspended solids criteria and to provide adequate storage
capacity for the dredged solids. These tests are based on the gravity
separation of solid particles from the transporting water.
(2) The sedimentation process can be categorized according to three
basic classifications:
(a) Discrete settling. The particle maintains its individuality and
does not change in size, shape, or density during the settling process.
(b) Flocculent settling. Particles agglomerate during the settling
period with a change in physical properties and settling rate.
(c) Zone settling. The flocculent suspension forms a lattice structure and settles as a mass, exhibiting a distinct interface during the settling process.
(3) The important factors governing the sedimentation of dredged material solids are the initial concentration of the slurry and the flocculating properties of the solid particles. Montgomery (item 4) demonstrated by
experiments that, because of the high influent solids concentration and the
tendency of dredged material fine-grained particles to flocculate, either
flocculate or zone-settling behavior governs sedimentation in containment
areas. Discrete settling describes the sedimentation of sand particles and
fine-grained sediments at concentrations much lower than those found in
dredged material containment areas. Test results using the 8-in.-diameter
settling column are used to design the containment area for solids retention based on principles of flocculent or zone settling. Detailed design
procedures found in WES TR DS-78-10 will determine surface area, containment area volume, ponding depth, and freeboard requirements. The designs
must consider the hydraulic efficiency of the containment, based on shape
and topography, and the proper sizing of outlet structures.
f.

Evaluation of Long-Term Storage Capacity.

(1) If the containment area is intended for one-time use, as is the


case in some new work projects, estimates of long-term storage capacity are
not required. However, containment areas intended for use in recurring
maintenance work must be sized for long-term storage capacity over the service life of the facility. Storage capacity is defined as the total volume
available to hold additional dredged material and is equal to the total
4-21

EM 1110-2-5025
25 Mar 83
unoccupied volume minus the volume associated with ponding and freeboard
requirements.
(2) The following factors must be considered in estimating long-term
containment area storage capacity:
(a) After dredged material is placed within a containment area, it
undergoes sedimentation and self-weight consolidation resulting in gains in
storage capacity.
(b) The placement of dredged material imposes a loading on the containment area foundation, and additional settlement may result due to consolidation of compressible foundation soils.
(c) Since the consolidation process is slow, especially in the case
of fine-grained materials, it is likely that total settlement will not have
taken place before space in the containment area is required for additional
placement of dredged material. For this reason, the time-consolidation relationship is an important consideration.
(d) Settlement of the containing dikes significantly affects the
available storage capacity.
(3) Estimation of gains in long-term capacity can be made using results of laboratory consolidation tests and application of fundamental principles of consolidation modified to consider the self-weight consolidation
behavior of newly placed dredged material. Detailed procedures for estimating long-term storage capacity are found in WES TR DS-78-10.
g. Weir Design. The purpose of the weir structure is to regulate release of ponded water from the containment area. Proper weir design and
operation can control resuspension and withdrawal of settled solids. This
is possible only if the containment areas have been properly designed to
provide sufficient area and volume for sedimentation. Weirs are designed
to provide selective withdrawal of the clarified upper layer of ponded
water.
In order to maintain acceptable effluent quality, the upper water
layers containing low levels of suspended solids should be ponded at depths
greater than the depth of the withdrawal zone; i.e., the area through which
fluid is removed for discharge over the weir. The size of the withdrawal
zone as determined by the weir loaction and configuration affects the
velocity of flow toward the weir. Detailed considerations in weir design
and design nomographs for determining required weir crest lengths are
found in WES TR DS-78-10. Weirs should be structurally designed to withstand anticipated loadings at maximum ponding elevations, with consideration given to uplift forces and potential piping beneath or around the wier.
Outlet pipes for the weir structure must be designed to carry flows in excess of the flow rate for the largest dredge size expected to provide for
emergency release of ponded waters.
h.

Chemical Clarification for Reduction of Effluent Suspended Solids.

4-22

EM 1110-2-5025
25 Mar 83
(1) When dredged material slurry is disposed in a well-designed, wellmanaged containment area, the vast majority of the solids will settle out
of suspension and be retained within the settling basin. However, gravity
sedimentation alone will not remove all suspended solids. Any fine-grained
material suspended in the ponded water above the settled solids will be
discharged in the effluent water. In addition, the levels of chemical constituents in the effluent water are directly related to the amount of suspended fine-grained material; therefore, retention of fine-grained solids
in the containment area results in a maximum degree of retention of potentially toxic chemical constituents. Effluent standards may require removal
of suspended solids over and above that attained by gravity sedimentation.
(2) In the absence of a fully engineered treatment system, several
expedient measures can be employed to enhance retention of the suspended
solids within a containment area of a given size before effluent discharge.
They include: intermittent pumping, increasing the depth of ponded water,
increasing the effective length of the weir, temporarily discontinuing operations, or decreasing the size of the dredge.
(3) Flocculation. One method specifically for reducing the levels of
fine-grained (clay-sized) suspended solids levels in the effluent involves
treating the containment area effluent or the dredged material slurry with
chemical flocculants to encourage the formation of flocs (i.e., particle
agglomerates) that settle more rapidly than individual particles. This agglomeration or coagulation process is accomplished by an alteration of the
electrochemical properties of the clay particles and the bridging of particles and small flocs by long polymer chains. Because of the large number
of manufacturers of polyelectrolytes and the types available, preliminary
screening of flocculants is necessary. Evaluation and determination of the
optimum dose of several nontoxic polymers may be accomplished using jartesting procedures. These procedures will indicate the most cost-effective
polymer and the optimum dosage of the polymer solution for treating the suspended solids levels, as well as the optimum mixing intensities and durations for both rapid- and slow-mixing stages. Optimum detention times and
surface overflow rates for clarifying the flocced suspensions and a general
indication of the volume of flocced material that must be stored or rehandled can be determined from settling tests. Schroeder (item 8) presents
design guidance for the use of chemical clarification methods.
i. Dike Design. Dikes for retaining or confining dredged material
are normally earthen embankments similar to flood protection levees. Dike
locations are usually determined by land available-for disposal areas;
therefore, dikes sometimes must be constructed in areas of poor foundation
quality and from materials of poor construction quality. In past years, retaining dikes for dredged material have been designed and constructed with
less effort and expense than other engineered structures. The potential for
dike failures and the environmental and economic damage which can result
dictate that retaining dikes be properly designed and constructed using the
principles of geotechnical engineering. Foundation investigations and laboratory soils tests and analyses must be conducted to design dikes to the desired degree of safety against failures. Procedures used in dike design
4-23

EM 1110-2-5025
25 Mar 83
generally parallel those required for design of flood protection levees or
earth-filled dams. WBS TR D-77-9 contains detailed guidelines for the design and construction of retaining dikes.
4-11.

Containment Area Operation and Management.

a. Containment Area Operation. A major consideration in proper containment area operation is providing the ponding necessary for sedimentation and retention of suspended solids. Adequate ponding depth during the
dredging operation is maintained by controlling the weir crest elevation,
usually by placing boards within the weir structure. Before dredging commences, the weir should be boarded to the highest possible elevation that
dike stability considerations will allow. This practice will ensure maximum possible efficiency of the containment area. The maximum elevation
must allow for adequate ponding depth above the highest expected level of
accumulated settled solids and yet remain below the required freeboard. If
the basin is undersized or if inefficient settling is occurring in the
basin, it is necessary to increase detention time and reduce approach velocity to achieve efficient settling and to avoid resuspension, respectively.
Detention time can be increased by raising the weir crest to its highest
elevation to increase the ponding depth; or it may be increased by operating the dredge intermittently to maintain a maximum allowable static head
or depth of flow over the weir, based on the effluent quality achieved at
various weir crest elevations. Once the dredging operation is completed,
the ponded water must be removed to promote drying and consolidation of
dredged material. Refer to WES TR DS-78-10.
b.

Containment Area Management.

(1) Periodic site inspections. The importance of periodic site inspections and continuous site management following the dredging operation
cannot be overemphasized. Once the dredging operation has been completed
and the ponded water has been decanted, site management efforts should be
concentrated on maximizing the containment storage capacity gained from
continued drying and consolidation of dredged material and foundation soils.
To ensure that precipitation does not pond water, the weir crest elevation
must be kept at levels allowing efficient release of runoff water. This
will require periodic lowering of the weir crest elevation as the dredged
material surface settles.
(2) Thin-lift placement. Gains in long-term storage capacity of containment areas through natural drying processes can also be increased by
placing the dredged material in thin lifts. Thin-lift placement greatly
increases potential capacity through active dewatering and disposal area
reuse management programs. Thin-lift placement can be achieved by obtaining sufficient land area to ensure adequate storage capacity without the
need for thick lifts. It requires careful long-range planning to ensure
that the large land area is used effectively for dredged material dewatering, rather than simply being a containment area whose service life is
longer than that of a smaller area. Dividing a large containment area into
several compartments can facilitate management; each compartment can be
managed separately so that some compartments are being filled while the
4-24

EM 1110-2-5025
25 Mar 83
dredged material in others is being dewatered. One possible management
scheme for large compartmentalized containments is shown conceptually in
figure 4-7. For this operation, thin lifts of dredged material are placed
into each compartment in the following sequence: filling, settling and surface drainage, dewatering, and dike raising (using dewatered dredged
material).
c.

Dewatering and Densification.

(1) The removal of excess water from dredged material through active
site management may add considerably to containment area storage volume,
especially in the case of fine-grained dredged material. The most successful dewatering techniques involve efforts to accelerate natural drying and
desiccation of dredged material through increased surface drainage. Dewatering efforts may be implemented in conjunction with other periodic inspection and management activities of the containment.

Figure 4-7. Conceptual illustration of sequential dewatering operations


possible if disposal site is large enough to contain material from
several periodic dredging operations.
4-25

EM 1110-2-5025
25 Mar 83
(2) Dredged material is usually placed in confined disposal areas in
a slurry state. Although a significant amount of water runs off through
the overflow weirs of the disposal area, the confined fine-grained dredged
material usually sediments/consolidates to only a semifluid consistency
that still contains large amounts of water. Not only does the high water
content greatly reduce available future disposal volume, but it also makes
the dredged material unsuitable or undesirable for any commercial or productive use.
(3) Three major reasons exist for dewatering fine-grained dredged material placed in confined disposal areas:
(a) Promotion of shrinkage and consolidation to increase volume in
the existing disposal site for additional dredged material.
(b) Reclamation of the dredged material into more stable soil form
for removal and use in dike raising or other engineered construction, or
for other productive uses, again increasing volume in the existing disposal
site.
(c) Creation of stable, fast land at the disposal site itself, at a
known final elevation and with predictable geotechnical properties.
(4) Allowing evaporative forces to dry fine-grained material into a
crust while gradually lowering the internal water table is the least expensive and most widely applicable dewatering method. Good surface drainage, rapidly removing precipitation and preventing ponding of surface
water, accelerates evaporative drying. Shrinkage forces developed during
drying return the material to more stable form; lowering the internal water
table results in further consolidation.
(5) Trenching. The most efficient method for promoting good surface
drainage is to construct drainage trenches in the disposal area. Because
several types of equipment have been found effective for progressive trenching to improve disposal area surface drainage, no unique set of trenching
equipment and procedures exists. The proper equipment for any dewatering
program will depend upon the following factors: size of the disposal area,
whether or not desiccation crust currently exists (and, if so, of what
thickness), time available for dewatering operations, type of site access,
condition of existing perimeter dikes, time available between disposal
cycles, and availability of and rental and operating cost for various types
of trenching equipment.
(6) Underdrainage. Underdrainage is another dewatering method which
may be used either individually or in conjunction with improved surface
drainage. In this procedure, collector pipes are placed in either a naturally occurring or artificially placed pervious layer before dredged material disposal. Upon disposal, free water in the dredged material migrates
into the pervious underdrainage layer and is removed via the collector pipe
system. Although technically feasible, underdrainage may not be costeffective in many disposal situations. Detailed discussions of dredged
material dewatering are found in WES TR DS-78-11.
4-26

EM 1110-2-5025
25 Mar 83
d. Disposal Area Reuse. Removal of coarse-grained material and dewatered fine-grained material from containment areas through proper management techniques will further add to capacity and may be implemented in conjunction with dike maintenance or raising. Removal of fine-grained dredged
material is a logical followup to successful dewatering management activities and can allow partial or total reuse of the disposal area. A reusable
disposal area can be regarded as a dredged material transfer station, where
dredged material is collected, processed if necessary, and removed for productive use or inland disposal. The advantages provided by a reusable disposal area (one from which all or a large portion of dredged material is
removed) and not by a conventional area are:
(1) Elimination or reduction of land acquisition requirements, except
for inland disposal.
(2) Justification for increased costs for high-quality disposal area
design and construction.
(3) Long-term availability of disposal areas near dredging sites.
(4) Availability of dredged material for use as landfill or construction material.
Detailed guidance on disposal area reuse is found in WES TR DS-78-12.
4-12.

Productive Uses.

a. When planning a reusable disposal area, major consideration should


be given as to how the dredged material solids will be used. If off-site
productive uses could be found for all the solids being dredged, the site
would theoretically have an infinite service life. The fact that dewatered
dredged material is a soil, may be analyzed as a soil, and can be used as a
soil encourages the productive use of dredged material as a natural resource. The following should be evaluated as potential off-site productive
uses for dredged material:
(1) Landfill and construction material.
(2) Surface mine reclamation.
(3) Sanitary landfill cover material.
(4) Agricultural land enhancement.
Compatibility of dredged material with the use in question and feasibility
of transport must be considered in evaluating off-site productive use.
Detailed guidance is found in WES TR DS-78-21.
b. Containment areas that have been filled also have potential productive use as industrial, recreational, or waterway-related sites. Filled
containment areas have been commonly used for commercial/industrial sites,
and most ports have such facilities built on former dredged material
4-27

EM 1110-2-5025
25 Mar 83
disposal sites. Recreational use of containment areas is popular because
it requires minimum planning and lower cost as compared to industrial/
commercial uses. In addition, the nature of recreation sites with much
open space and light construction is especially suited to the weak foundation conditions associated with fine-grained dredged material. Dredged
material sites may be used for purposes closely related to the maintenance,
preservation, and expanded use of waterways and the surrounding lands, such
as shore protection, beach nourishment, breakwaters, river control, etc.
Such uses of dredged material sites are influenced by the method and sequence of the dredging operation as well as the layout of the disposal area.
Waterway-related use normally involves the creation of landforms and thus
permits opportunities for imaginative multiple-use site development. These
landforms commonly result in a secondary recreational use.
4-13.

Environmental

Considerations.

a. Upland disposal of contaminated sediments must be planned to contain potentially toxic materials to control or minimize potential environmental impacts. There are four possible mechanisms for transport of contaminants from upland disposal sites:
(1) Release of contaminants in the effluent during disposal
operations.
(2) Leaching into groundwater.
(3) Surface runoff of contaminants in either dissolved or suspended
particulate form following disposal.
(4) Plant uptake directly from sediments, followed by indirect animal
uptake from feeding on vegetation.
b. The physiochemical conditions of the dredged material at an upland
disposal site may be altered substantially by the drainage of excess water.
Marked changes in the chemical mobility and biological availability of some
contaminants may result. In many cases, contaminant levels exceed applicable surface water quality criteria if mixing and dilution with large
volumes of receiving water is limited. Almost all of the contaminants in
initial dewatering effluents (with the possible exception of ammonia and
manganese) are associated with suspended particulates; increasing suspended
solids removal will be effective in reducing these levels.
c. Disposal sites should not be selected where subsurface drainage
could result in contaminant levels exceeding applicable criteria for drinking water supplies or adjacent surface waters. Management practices to reduce leaching losses may be beneficial in some cases. Coarse-textured materials will tend to drain freely with little impediment, with time. Some
fine-textured dredged material tends to form its own liner as particles
settle with percolation drainage water, but it may require considerable
time for self-sealing to develop; thus, an artificial liner may be useful
for some upland sites. Because of the gradual self-sealing nature of many
4-28

EM 1110-2-5025
25 Mar 83
fine-textured dredged materials, temporary liners subject to gradual deterioration with time may be adequate in many cases.
d. Plant populations may be managed to minimize uptake and environmental cycling of metals from contaminated sediments applied upland. Such
a technique may be more effective where plant populations are intensively
managed, as in an agricultural operation, since different species and even
subspecies differ greatly in their ability to take up and translocate toxic
materials. It may be possible to grow crops in which metals tend to accumulate in the plant tissue which is not harvested. Where contaminated
dredged material is used to amend agricultural soil or improve other unproductive soils, liming can be an economical and effective method for reducing the bioavailability of many toxic metals.
e. Covering contaminated dredged material with clean soil or clean
dredged sediments is a potential management practice that applies to all
three of the major disposal alternatives. Where contaminated dredged material is to be used for habitat development, agricultural soils amendment,
land reclamation, or as fill for engineering purposes, covering with clean
material can be an effective method for isolating contaminants from biological populations growing in or living on the disposal site. The depth
of clean material should be sufficient to isolate contaminants from plant
roots and burrowing animals. Care should also be exercised to ensure that
leaching from contaminated sediments into adjacent groundwater does not
take place.
Section V.
4-14.

Habitat Development as a Disposal Alternative

General Considerations for Habitat Development.

a. Habitat development refers to the establishment of relatively


permanent and biologically productive plant and animal habitats. The use
of dredged material as a substrate for habitat development offers a disposal
technique that is, in many situations, a feasible alternative to more conventional open-water, wetland, or upland disposal options. Refer to Smith
(item 8) for more detailed information.
b. Four general habitats are suitable for establishment on dredged
material : marsh, upland, island, and aquatic. Within any habitat, several
distinct biological communities may occur (fig. 4-8). The determination of
the feasibility of habitat development will center on the nature of the surrounding biological communities, the nature of the dredged material, and
the site selection, engineering design, cost of alternatives, environmental
impacts, and public approval. If habitat development is the selected alternative, a decision regarding the type or types of habitats to be developed must be made. This decision will be largely judgmental, but in
general, site peculiarities will not present more than one or two logical
options.
c. The selection of habitat development as a disposal alternative will
be competitive with other disposal options when the following conditions
exist:
4-29

EM 1110-2-5025
25 Mar 83

Figure 4-8.

Hypothetical site illustrating the diversity of habitat


types that may be developed at a disposal site,

(1) Public/agency opinion strongly opposes other alternatives.


(2) Recognized habitat needs exist.
(3) Enhancement measures on existing disposal sites are identified.
(4) Feasibility has been demonstrated locally.
(5) Stability of dredged material deposits is desired.
(6) Habitat development is economically feasible.
d. Disposal alternatives are often severely limited and constrained
by public opinion and/or agency regulations. Constraints on open-water
disposal and disposal on wetlands, or the unavailability of upland disposal
sites, may leave habitat development as the most attractive alternative.
e. Habitat development may have strong public appeal when the need
for restoration or mitigation or the need for additional habitat has been
demonstrated. This is particularly true in areas where similar habitat of
considerable value or public concern has been lost through natural processes or construction activities.
f. Habitat development may be used as an enhancement measure to improve the acceptance of a disposal technique. For example, seagrass may be
planted on submerged dredged material, or wildlife food plants established
on upland confined disposal sites. This alternative has considerable
4-30

EM 1110-2-5025
25 Mar 83
potential as a low-cost mitigation procedure and may be used to offset environmental impacts incurred in disposal.
g. The concept of habitat development is more apt to be viewed as feasible if it has been successfully demonstrated locally. Even the existence
of a pilot-scale project in a given locale will offset the uncertainties
often present in the public perception of an experimental or unproven
technique.
h. The vegetation cover provided by most habitat alternatives will
often stabilize dredged material and prevent its return to the waterway.
In many instances this aspect will reduce the amount of future maintenance
dredging necessary at a given site and result in a positive environmental
and economic impact.
i. The economic feasibility of habitat development should be considered in the context of long-term benefits. Biologically productive
habitats have varied but unquestionable value (e.g., sport and commercial
fisheries) and are relatively permanent features. Consequently, habitat
development may be considered a disposal option with long-term economic
benefits that can be applied against any additional costs incurred in its
implementation. Most other disposal options lack this benefit.
j. Habitat development may be most economically competitive in situations where it is possible to take advantage of natural conditions or where
minor modifications to existing methods would produce desirable biological
communities. For example, the existence of a low-energy, shallow-water
site adjacent to an area to be dredged may provide an ideal marsh development site and require almost no expenditure beyond that associated with
open-water disposal.
4-15.

Marsh Habitat Development.

a. Marshes are considered to be any community of grasses and/or herbs


which experiences periodic or permanent inundation. Typically, these are
intertidal fresh, brackish, or salt marshes or relatively permanently inundated freshwater marshes. Marshes are often recognized as extremely valuable natural systems and are accorded importance in food and detrital production, fish and wildlife cover, nutrient cycling, erosion control, floodwater retention, groundwater recharge, and aesthetic value. Marsh values
are highly site specific and must be interpreted in terms of such variables
as plant species composition, wildlife use, location, and size, which in
turn influence their impact upon a given ecosystem.
b. Marsh creation has been the most studied of the habitat development
alternatives, and accurate techniques have been developed to estimate costs
and to design, construct, and maintain these systems. Over 100 marshes have
been established on dredged material; examples are shown in figures 4-9 and
4-10. Refer to WES TR DS-78-16 for specific information on wetland habitat
development. The advantages most frequently identified with marsh development are: considerable public appeal, creation of desirable biological

4-31

EM 1110-2-5025
25 Mar 83

a.

b.

An aerial view of the 420-sq-ft freshwater marsh developed on


fine-textured dredged material confined by a sand dike.

Within 6 months of dredged material placement, a lush growth of


wetland plants had been established by natural colonization.

Figure 4-9.

Windmill Point marsh development site, James River, Virginia.


4-32

EM 1110-2-5025
25 Mar 83

a. A salt marsh was established on poorly consolidated fine-textured


dredged material confined behind an earthen dike on this dredged
material island.

b.

Vigorous growth was obtained from sprigged smooth cordgrass and


salt-meadow cordgrass.
Figure 4-10.

Apalachicola Bay marsh development site.


Apalachicola Bay, Florida .
4-33

EM 1110-2-5025
25 Mar 83
communities, considerable potential for enhancement or mitigation, and the
fact that it is frequently a low-cost option.
c. Marsh development is a disposal alternative that can generate
strong public appeal and has the potential for gaining wide acceptance when
other techniques cannot. The habitat created has biological values that are
readily identified and are accepted by many in the academic, governmental,
and private sectors. However, application requires an understanding of
local needs and perceptions and of the effective limits of the value of
these ecosystems.
d. The potential of this alternative to replace or improve marsh habitats lost through dredged material disposal or other activities is frequently overlooked. Techniques are sufficiently advanced to design and construct productive systems with a high degree of confidence. Additionally,
these habitats can often be developed with very little increase in cost
above normal project operation, a fact attested to by hundreds of marshes
that have been inadvertently established on dredged material.
e. The following problems are most likely to be encountered in the
implementation of this alternative: unavailability of appropriate sites,
loss of other habitats, release of contaminants, and loss of the site for
subsequent disposal.
f. The most difficult aspect of marsh development is the location of
suitable sites. Low-energy, shallow-water sites are most attractive; however, cost factors will become significant if long transport distances are
necessary to reach those sites. Protective structures may be required if
low-energy sites cannot be located, which can add considerably to project
cost.
g. Marsh development frequently means the replacement of one desirable habitat with another, and this will likely be the source of most opposition to this alternative. There are few reliable methods of comparing
the various losses and gains associated with this habitat conversion; consequently, relative impact may best be determined on the basis of the professional opinion of local authorities.
h. The potential for plants to take up contaminants and then release
them into the ecosystem through consumption by animals or decomposition of
plant material should be recognized when contaminated sediments are used
for habitat development. Although this process has not been shown to occur
often, techniques are available to determine the probability of uptake.
i. Development of a marsh at a given site can prevent the subsequent
use of that area as a disposal site. In many instances, any further development on that site would be prevented by State and Federal regulations.
Exceptions may occur in areas of severe erosion or where the initial disposal created a low marsh and subsequent disposal would create a higher
marsh.
j.

There are types of wetland habitat development other than marshes,


4-34

EM 1110-2-5025
25 Mar 83
such as bottomland hardwoods in freshwater areas.
WES TR DS-78-16.
4-16.

These are addressed in

Upland Habitat Development.

a. Upland habitats encompass a variety of terrestrial communities


ranging from bare soil to dense forest. In its broadest interpretation,
habitat occurs on all but the most disturbed upland disposal sites. For
example, a gravelly and bare freshwater disposal area may provide nest
sites for killdeer; weedy growth may provide cover for raccoons or a food
source for seed-eating birds; and water collection in desiccation cracks
may provide breeding habitat for mosquitoes. Man-made habitats will develop regardless of their management; however, the application of sound
management techniques will greatly improve the quality of those habitats
and the speed with which they are populated.
b. Upland habitat development has potential at hundreds of disposal
sites throughout the United States. Its implementation is largely a matter
of the application of well-established agricultural and wildlife management
techniques. Examples of successful sites are shown in figures 4-11 and
4-12. Refer to WES TR DS-78-17 for more detailed information on upland
habitat development. Upland habitat development as a disposal option has
several distinct advantages, including: adaptability, improved public acceptance, creation of biologically desirable habitats, elimination of problem areas, low-cost enhancement or mitigation, and compatibility with subsequent disposal.
c. Upland habitat development may be used as an enhancement or mitigation measure at new or existing disposal sites. Regardless of the condition or location of a disposal area, considerable potential exists to convert it into a more productive habitat. For example, small sites in
densely populated areas may be keyed to small animals adapted to urban life,
such as seed-eating birds and squirrels. Large tracts may be managed for a
variety of wildlife, including waterfowl, game mammals, and rare or endangered species.
d. The knowledge that a site will ultimately be developed into a useful area, be it a residential area, park, or wildlife habitat, improves
public acceptance. Many idle and undeveloped disposal areas that are now
sources of local irritation or neglect would directly benefit from upland
habitat development, and such development may well result in more ready acceptance of future disposal projects.
e. In general, upland habitat development will add little to the cost
of disposal operations. Standard procedures may involve liming, fertilizing, seeding, and mowing. A typical level of effort is similar to that applied for erosion control at most construction sites and considerably less
than that required for levee maintenance.
f. Unless the target habitat is a long-term goal such as a forest,
upland habitat development will generally be compatible with subsequent
disposal operations. In most situations, a desirable vegetative cover can
4-35

EM 1110-2-5025
25 Mar 83

Figure 4-11. Barley was planted on this sandy dredged material island
in the Columbia River, Oregon, greatly improving its value to wildlife,

Figure 4-12. Sandy and silty dredged material were combined at Nott
Island, Connecticut, to produce a pasture for wild geese.
4-36

EM 1110-2-5025
25 Mar 83
be produced in one growing season. Subsequent disposal would simply require recovery of the lost habitat. Indeed, the maintenance of a particular vegetation stage may require periodic disposal to retard or set
back plant succession.
g. The primary disadvantage of this alternative is related to public
acceptance. The development of a biologically productive area at a given
site may discourage subsequent disposal or modification of land use at that
site. This problem can be avoided by the clear identification or establishment of future plans before habitat development, or by the establishment
and maintenance of biological communities recognized as being most productive in the earlier stages of succession. In the latter case, subsequent disposal may be a necessary management tool.
h. Some habitat types will require management. For example, if highproductivity annual plants are selected for establishment (i.e., corn or
barley as prime wildlife foods), then yearly planting will be necessary.
If the intent is to maintain a grassland or open-field habitat, planting
may be required only initially, but it may be necessary to mow the area
every 1 to 5 years to retard colonizing woody vegetation. In most cases,
it will be possible to establish very low-maintenance habitats, but if the
intent is to establish and perpetuate a given habitat type, long-term management may be essential and expensive.
4-17.

Island Habitat Development.

a. Dredged material islands range in size from an acre to several


hundred acres. Island habitats are terrestrial communities completely surrounded by water or wetlands and are distinguished by their isolation and
their limited food and cover. Because they are isolated and relatively
predator-free they have particular value as nesting and roosting sites for
numerous species of sea and wading birds; e.g., gulls, terns, egrets,
herons, and pelicans. The importance of dredged material islands to nesting species tends to decrease as the size increases because larger islands
are more likely to support resident predators. However, isolation is more
important than size; and thus large isolated islands may be very attractive
to nesting birds. Dredged material island habitats are pictured in figures 4-13 and 4-14. Refer to WES TR DS-78-18 for specific information regarding island habitat development.
b. Dredged material islands are found in low- to medium-energy sites
throughout the United States. Typically, these are sandy islands located
next to navigation channels and are characteristic of the Intracoastal
Waterway. In recent years, many active dredged material islands have been
diked to improve the containment characteristics of the sites.
c. The importance of dredged material islands as nesting habitats for
sea and wading birds cannot be overemphasized. In some states (e.g. North
Carolina and Texas) most nesting of these colonial species occurs on manmade islands.
d.

Island habitat development has the following advantages: it


4-37

EM 1110-2-5025
25 Mar 83

4-38

EM 1110-2-5025
25 Mar 83

Figure 4-14. Mixed-species colony of royal and Sandwich terns located


on a dredged material island in Pamlico Sound, North Carolina. The
colony comprised 2988 royal tern nests and 897 Sandwich tern nests,
employs traditional disposal techniques, it permits reuse of existing disposal areas, it provides critical nesting habitats, and its management is
conducive to subsequent disposal.
e. Island habitat development utilizes a traditional disposal technique : the confined or unconfined disposal of dredged material in marsh or
shallow water or on existing islands. Consequently, unconventional operational problems seldom occur in its implementation.
f. In many coastal areas, the careful selection of island locales and
placement will encourage use by colonial nesting birds. Properly applied,
island habitat development is an important wildlife management tool: it
can replace habitats lost to other resource priorities, provide new habitats
where nesting and roosting sites are limiting factors, or rejuvenate existing disposal islands.
g. Planned disposal on existing dredged material islands is often conducive to their management for wildlife. Nesting is almost always keyed to
a specific vegetation successional stage, and periodic disposal may be used
to retard succession or set it back to a more desirable state. As a practical matter, disposal on existing islands has largely replaced new island
development because of opposition to the loss of open-water and bottom
4-39

EM 1110-2-5025
25 Mar 83
habitats.
Consequently, habitat development on dredged material islands
will frequently be keyed to the disposal on and management of existing
islands.
h. Island habitat development has the following disadvantages: it
may interrupt hydrologic processes, it may destroy open-water or marsh
habitats, and it requires careful placement of material and selection of
the disposal season to prevent disruption of active nesting.
i. Alteration of the water-energy regime by the placement of barriers
such as islands deserves particular attention because it can change the
temperature, salinity, circulation patterns, and sedimentation dynamics of
the affected body of water. Large-scale projects or projects in particularly sensitive areas may warrant the development of physical, chemical,
and biological models of the aquatic system before project implementation.
j. Dredged material islands, by the nature of their location, may reduce the presence of wetlands and/or open-water and their associated benthic habitats. This impact will be minimized by careful site selection or
disposal on existing sites. Containment behind dikes will lessen the lateral spread of material but will probably adversely affect the value of
the island to birds.
k. Disposal on any dredged material island should be immediately preceeded by a visit to determine if the site is an active nesting colony.
The use of dredged material islands by birds will occur with or without
management. When colonies are present, scheduling of subsequent disposal
operations and placement of material should be planned to minimize disruption of the disposal operations as well as of the nesting colonies involved.
Destruction of the nests of all colonial waterbirds is a criminal offense
punishable by fine and/or imprisonment.
4-18.

Aquatic Habitat Development.

a. Aquatic habitat development refers to the establishment of biological communities on dredged material at or below mean tide. Potential
developments include such communities as tidal flats, seagrass meadows,
oyster beds, and clam flats. The bottoms of many water bodies could be
altered using dredged material; in many cases this would simultaneously improve the characteristics of the site for selected species and permit the
disposal of significant quantities of material. Planned aquatic habitat
development is a relatively new and rapidly moving field; however, with the
exception of many unintentional occurrences and several small-scale demonstration projects, this alternative is largely untested. There are no
general texts or manuals currently available; however, potential users may
obtain updated information by contacting the Environmental Laboratory at
the U. S. Army Engineer Waterways Experiment Station.
b. The major advantages of aquatic development are that it produces
habitats that have high biological production and potential for wide application and can effectively complement other habitats.

4-40

EM 1110-2-5025
25 Mar 83
c. Aquatic habitats may be highly productive biological units. Seagrass beds are recognized as exceptionally valuable habitat features, providing both food and cover for many fish and shellfish. Oyster beds and
clam flats have high recreational and commercial importance. Dredged material disposal projects affecting aquatic communities often incur strong
criticism, and in these instances reestablishment of similar communities
may be feasible as a mitigation or enhancement technique. In many instances it will be possible to establish aquatic habitats as part of marsh
habitat development.

4-41

EM 1110-2-5025
25 Mar 83

APPENDIX A
BIBLIOGRAPHY
1.

Environmental Protection Agency/Corps of Engineers Technical Committee


on Criteria for Dredged and Fill Material. 1978 (Apr). "Ecological
Evaluation of Proposed Discharge of Dredged Material into Ocean Waters;
Implementation Manual for Section 103 of Public Law 92-532 (Marine
Protection, Research, and Sanctuaries Act of 1972)," U. S. Army Engineer Waterways Experiment Station, CE, Vicksburg, MS.

2.

Folsom, B. L., Jr., and Lee, C. R. 1981. "Zinc and Cadmium Uptake by
the Freshwater Marsh Plant Cyperus esculentus Grown in Contaminated
Sediments Under Reduced (Flooded) and Oxidized (Upland) Disposal Conditions," Journal of Plant Nutrition, Vol. 3, pp 233-244.

3.

Herbich, J. B. 1975.
ing Company, Texas.

4.

Montgomery, R. L. 1978 (Dec). "Methodology for Design of FineGrained Dredged Material Containment Areas for Solids Retention,"
Technical Report D-78-56, U. S. Army Engineer Waterways Experiment
Station, CE, Vicksburg, MS.

5.

Montgomery, R. L., Horstmann, H. L., Jr., Sanderson, W. H., and


McKnight, A. L. 1981 (Oct). "Problem Identification and Assessment
to Determine the Need and Feasibility of an Update of the National
Dredging Study," U. S. Army Engineer Waterways Experiment Station, CE,
Vicksburg, MS.

6.

Murden, W. R., and Goodier, J. L. 1976. "The National Dredging


Study," Proceedings of WODCON VII, World Dredging Conference, San
Pedro, CA.

7.

Richardson, T. W. 1983. "Agitation Dredging: Lessons and Guidelines


from Past Projects," Technical Report in Press, U. S. Army Engineer
Waterways Experiment Station, CE, Vicksburg, MS.

8.

Schroeder, P. R. 1982. "Chemical Clarification Methods for Confined


Dredged Material," Technical Report in Press, U. S. Army Engineer
Waterways Experiment Station, CE, Vicksburg, MS.

9.

Smith, H. K. 1978 (Dec). "An Introduction to Habitat Development on


Dredged Material," Technical Report DS-78-19, U. S. Army Engineer
Waterways Experiment Station, CE, Vicksburg, MS.

Coastal and Deep Ocean Dredging, Gulf Publish-

10.

Symcon Publishing Company. World Dredging and Marine Construction,


San Pedro, CA.

11.

Turner, T. M. 1977 (Feb). "The Bucket Wheel Hydraulic Dredge," World


Dredging and Marine Construction, Vol 13, No. 3, pp 23-27.

A-1

EM 1110-2-5025
25 Mar 83

12.

U. S. Army Engineer Waterways Experiment Station. 1963 (Mar). "The


Unified Soil Classification System," Technical Memorandum No. 3-357,
Vicksburg, MS.

A-2

EM 1110-2-5025
25 Mar 83

APPENDIX B
CHECKLIST FOR REQUIRED STUDIES
The development of a dredging project involves the study and evaluation
of many factors to assure that dredging and disposal is carried out in an
efficient, economical, and environmentally compatible manner. The following
are some of the factors that should be considered in the planning and design
phase:
a. Analysis of dredging locations and quantities.
b. Dredging environment; i.e., depths, waves, currents, distance to
potential disposal area, etc.
c. Evaluation of physical, chemical, and biological characteristics of
sediments to be dredged.
d.

Identification of social, environmental, and institutional factors.

e.

Evaluation of dredge plant requirements.

f.

Evaluation of potential disposal alternatives.

g.

Hydrographic surveys of proposed project.

h.

Field investigations of sediments to be dredged.

i. Performance of required laboratory tests; i.e., chemical characterization, sedimentation, engineering properties, bioassay, bioaccumulation, etc.
j.

Evaluation of in situ density of sediments to be dredged.

k. Evaluation of long-term dredging and disposal requirements for


project.
l. Coordination of project plans with engineering, constructionoperation, and planning elements of District.
m.

Evaluation of potential productive uses of dredged material.

n. Coordination of project plans with other agencies, public, and


private groups.
o.
impact.

Evaluation of proposed project to determine potential environmental

B-1
*U.S. GOVERNMENT PRINTING OFFICE: 1993 - 718-983/61137

The Basics of Optimally Loading a TSHD


Introduction
Within the wide range of dredging vessels the trailing suction
hopper dredger has an almost unique position. In contrast to
cutter suction dredgers. profile dredgers, bucket dredgers and
so forth the actual dredging activities of this type of vessel are
not continuous. The loading, sailing, discharge to shore or
dumping takes longer or not, according to location and the
type of vessel.
The operating cycle of a trailing suction hopper dredger can be divided into the following
phases:
Loading of the hopper.
Transport of the soil from dredging location to dumping or shore discharge location.
Dumping of the load or discharging it to shore.
The return of the empty vessel to the dredging location.
We can only speak of optimum use of the trailing suction hopper dredger when every single
phase has been optimised and when the different phases are well adjusted to each other. In this
article we first briefly discuss some aspects which are of importance for the optimisation of the
separate phases of the dredging cycle. It appears that it is the loading phase that offers the best
opportunities for improvement. That is why we go deeper into the method to determine loading
and the correct moment to stop dredging. The question is what are the effects of using modern,
more extensive than usual measuring and signal-processing techniques in the right way.
Transport
During both vessel transport phases, empty and loaded, savings may be made by sailing with
reduced engine output. This is most likely when sailing on shallow waters. That the reduction of
fuel costs has to be balanced against the costs of the slightly prolonged cycle time does not
need to be emphasised.
Dumping
The time necessary for dumping can be minimised by the most effective fluidisation of the load,
pumping water on the load with the dredgepump and by means of water jets. This, of course,
largely depends on the type of soil and in most cases the effects on the cycle time are marginal.
Shore discharge
The most effective extraction method is the one where the load is discharged with the highest
possible density and speed. Attentive valve operation together with the correct use of water jets
and hopper discharge duct water supply is of major importance. Most of the time, however,
unloading time is determined by the type of soil, the delivery distance, pump output and the
optimal flow-concentration combination that can be maintained.
Separate dumping and shore discharge
When one has to discharge to shore a third possibility can lead to reduction of the cubic meter
price in certain circumstances. In general shore discharge with a trailing suction hopper dredger
is a costly operation. However, by dumping the soil via the bottom doors and then pumping it to

shore by means of a separate small stationary suction dredger a considerable reduction of


costs can often be reached.
Trailing suction
The effectiveness of the trailing suction hopper dredger is largely determined by the draghead
used, the availability of jet water at the head, the flowrate, the propeller pull available for drag,
the adjustment of the draghead and how the flowrate and trailing speed are handled. By playing
sensibly with these variables output can be optimised. A considerable amount of time can be
saved particularly by searching for the right trailing speed in each situation.
Loading
When loading, a distinction should be made between loading the hopper with fine silt, clay or
with granular material. When loading silt the overflow is limited to a few minutes at the most.
When it concerns polluted slurry there will be no overflow at all. With the loading of clay and
granular material there is virtually always overflow. The answer to the question when the
overflow should be stopped is defined by factors such as: type of soil and granulation, mixture
concentration, hopper geometry, flow rate and the actual liquid/solid loading condition during
overflowing. During loading overflow losses can occur which, depending on the situation, can
vary between 0% and 100% of the actual dredged material flow.
Measurement of the dredging parameters
A complex of factors determines the effectiveness of the dredging process and loading. To have
detailed insight here expensive instrumentation is employed. We might mention the
measurement of the mixture velocity and concentration, from which the dredged production is
calculated, and the measurement of pump vacuum and discharge pressure and hopper loading.
For insight into the effectiveness of the loading of the hopper a load indicator is necessary.
Before discussing this further the loading phase will be first explained.
A closer look at the loading phase
During the loading of the hopper it goes through the following phases:
First the mixture is pumped into the hopper until it is completely full and starts to overflow.
As long as the hopper does not overflow no overflow loss will take place and a maximum
material collection will be achieved.
Overflow at a rate (m3/s) equal to the flow rate. The water in which the solid material had
settled, flows over the overflow(s) in their highest position or over the side of the hopper until
the ship has reached the maximum admissible draught and with that its maximum loading
capacity. The overflow losses are still relatively low.
Overflow at a rate (tons/s) equal to the pump production. The vessel is kept at a constant
draught (constant load). With a pumped-in concentration which is higher than the overflow
concentration the out-going volume stream must be larger than the incoming one. Because
of this the liquid loading level will drop gradually. At the same time the solid loading level will
rise because of settling. This leads to a lower liquid section as a result of which the mixture
transport from inlet pipe to overflow speeds up and becomes more and more turbulent with,
as a result, decreasing settling and even erosion. This explains the increasing overflow
losses until the pump production directly leaves the hopper. The loading efficiency has then
decreased to 0% (or sometimes even less).
End of loading and overflow of the remaining water. Far before the point that 100% of
overflow loss is reached the crew will have stopped drawing material in. There is still
residual water on top of the load which can be discharged after some time by overflowing so

that the vessel can sail to the dumping site or shore connection without unnecessary load
and at reduced draught.
The load indicator as aid for optimal loading
When loading a trailing suction hopper dredger one mostly uses a recording load indicator. The
current type is based on measurement of the sinking of the vessel hull and shows the progress
in the time of the displacement. Fig. 1 schematically demonstrates such a recording. From this
displacement curve one can derive the most cost effective point at which to stop loading. For
this purpose one usually chooses by feel the point on the displacement curve whereby it does
not increase further. As a further help one sometimes also uses the "tangent method" that will
be discussed later. Because of its limitations this method is only useful when dredging silt and
with trailing dredgers with a so-called "high specific gravity hopper" (high load capacity - hopper
volume ratio).

Fig.1
The load indicator based on vessel hull displacement has some clear limitations. Together with
some general critical aspects these are:
The load indicator shows the displacement of the vessel hull and thus the weight of the total
hopper contents that can be derived from it consisting of dredged material mixed with
water. However, from the displacement curve the 'paying' load cannot be derived.
On reaching the maximum hull displacement and with that the maximum weight of the
hopper contents, that part of the load that pays - the solid or dry material - is not maximised.
Besides the settled (solid) material, the load consists of a mixture with a considerable
amount of water which can still be replaced by an amount of paying solid material of
approximately the same weight.
That is why with most trailing dredgers it is impossible to derive the optimal final point of
loading of granular material from the hull displacement curve. More information is needed.
Sometimes the trim of the vessel hull is insufficiently taken into account which results in a
less specific indication of the displacement. This effect can also be the result of non-optimal
positions of the sensors or flexing of the vessel hull.
To accurately define displacement pressure measurement is used. This can be done by
means of pressure meters in the bottom of the ship or indirectly and mostly less accurately
by means of bubble points.

The optimal load indicator


The limiting aspects mentioned above can be met by an optimal load indicator:
By measuring the displacement of the vessel by using pressure meters in the bottom. These
should be installed in such a way that under every possible sea condition a reliable trim can
be derived. Possibly the displacement can be defined on the basis of four instead of two
sensors so that both listing and flexing can always be taken into account.
By programming the load indicator with extensive ship's tables for water displacement as a
function of draught and trim which tables list all possible draught and trim conditions.
By measuring the volume of the mixture inside the hopper, taking into account the trim and
listing conditions.
By continually determining the actual volume or weight of the paying load from water
displacement and hopper volume.
By including in the graphic presentation an aid for determining the optimal point to stop
loading.
By equipping the load indicator with an indication of mutual inconsistent trim positions that
are defined on the basis of draught and hopper level. By this means erroneous sensors can
be monitored.
By a provision for the automatic control of overflow during loading at maximum draught.
The actual hopper contents is derived from the level of the fluid load in the hopper. To measure
this ultra-sound measuring techniques are used as a rule. The actual weight in the hopper is
derived from the difference between the actual displacement and the displacement of the empty
vessel. From the hopper volume and the weight of the hopper contents the volume and weight
of the paying load can be derived. The paying volume or weight is determined by an adjustable
reference density (for instance in situ, water-saturated sand, dry material). For signal
processing, calculations and presentation a micro computer system is used. The same
functionality can, of course, also be included in a modern integrated and computerised dredging
control system. The definition of the paying load takes place on the basis of the following
equations:

In which:
Volpl = Volume of paying load
Sgpl = Specific Gravity of paying load
Wgtpl = Weight of paying load
Volh = Volume hopper
Wgth = Weight of hopper contents
SGw = Specific Gravity outboard water

The progress of the weight of the paying load is graphically shown in Fig. 2, together with the
displacement curve. From the curve for the paying load the optimal loading point can be
derived. In the next paragraph the background and working method will be explained.

Fig. 2

Determination of optimal duration for hopper loading


(by means of the tangent line method to the paying load curve) Contrary to the displacement
curve (weight hopper contents), from this "paying load curve" the optimal point to stop loading
can be defined. The point where the volume or weight of the paying load per cycle-time unit is at
its maximum is regarded as the optimal loading point. That is why a maximum output cost ratio
occurs at the optimal loading point. Graphically this point can be defined as follows: As starting
point of the new, we take the time of the previous cycle when dredging stopped. When following
the new loading curve every point of that curve can be linked to the starting point of the cycle at
the 0-level of the paying load. The tangent of the angle this line makes with regard to the time
axis then corresponds to the relation in that point between the paying load and the duration of
the cycle up to that moment. At the point where this connection line hits the loading curve, the
angle, and also the ratio between output and costs, will be at a maximum. By increasing
overflow losses the growth of the paying load per time unit decreases. So from that moment on
the paying load curve runs below the tangent line. That is why every next point will show a lower
output per unit time. Stopping dredging is thus imperative. The course of affairs as outlined here
is shown graphically by means of Fig. 3.

Fig. 3

Optimal loading by using optimal equipment


From the foregoing it can be deduced that optimal loading is only possible when using
equipment that is optimally suited for this task. This implies that the equipment should at least
be able to present on a monitor the time progress of the paying load in the hopper, plus an
indication of the loading optimum. An example of such a presentation is again shown in both
figures below. Fig. 4.1 shows the cycle with a short sailing distance, while Fig. 4.2 shows the
cycle with a long sailing distance. In these diagrams for the definition of the loading optimum the
tangent line method has been applied both on the displacement curve (dashed line) and on the
paying load curve (unbroken line). The differences have been made visible

Fig. 4.1

Fig. 4.2

Financial aspects and conclusions


Both in the case of short and long sailing times application of the tangent line method to the
paying load curve leads to considerably more favourable ratios between paying load and cycle
time. Quantitative comparative research has shown that, especially with short cycle times,
loading is ended too early on the basis of the usual loading measurement. Within a cycle time of
120 minutes and a loading time of 60 minutes it appears that when loading moderately fine
sand, differences of 10 minutes and more can occur. This means that with the improved method
loading can go on up to a point with a higher cycle efficiency. The accompanying output
improvement is in the order of 5 to 10%. In the considerations above it is assumed that in the

case of loading measurement which is only based on displacement, the tangent line method is
always used. In practice, however, it is mostly different.
Because the displacement curve does not give any further information after passing the point of
maximum displacement, while one still has the feeling that loading can go on, most of the time
loading will go on for too long. This probably causes an even worse loss of efficiency.
These things show that the improved working method described based on the definition of the
paying load, will always lead to an improved loading efficiency. In addition to the financial one
there is yet another aspect of the improved method for loading measurement which should not
be left unmentioned. With clients as well as with controlling authorities there is an increasing
effort to measure the actual dredged soil and relocated dry material as a basis for payment.
This, together with the determination of the optimal loading time, can contribute to an important
increase in the efficiency of the trailing suction hopper dredger.
(Source : Ports and Dredging 137, IHC Holland 1991)

Dredge monitoring system (DMS)


The DMS is an integrated monitoring system available both for refits and newly built
dredgers. The DMS monitors the complete dredging equipment and is customised to
meet customer requirements. A typical DMS constellation consists of:
Suction pipe position indicator
Draft and load monitoring
Production monitoring
Process monitoring for:
Status of dredge valves, jet valves and flushing valves
Status of the dredge- and jet pumps
Status of the bottom doors/valves
Status of the hydraulic system
Alarm monitoring and data logging
Typically the implementation of the DMS does not influence the existing control system.
It is designed to act as independent system, collecting all necessary data for its
calculation and representation tasks. Depending on the size and complexity of the
dredge, an appropriate design for a highly integrated Dredge Monitoring System is
developed. Fibre optic and industrial interfaces using modern networks techniques are
used to connect decentralised process interfaces with centralised controlling units and
human
interfaces.
Needless to say that, if required, applicable control functionality can be integrated into
our system to fulfil tasks such as automatic light mixture overboard or dredge pump
speed control.
IN a nutshell, the DMS is a modern monitoring system for dredge, displaying the
complete dredging application in a clear manner on several displays.

Page 1 of 6

Dredge control and monitoring system (DCMS)


These days automation takes on an ever increasing role in performing the various tasks
in the dredging world. In the past each task was realised by a separate controller to fulfil
a certain job, which lead to a large number of different units, each with its own
operational philosophy. Today all tasks are integrated into one system with a clearly
defined human interface. Dredging is a business where real time is real money. Every
unproductive hour means lost profits for operators. So VOSTA LMG provides smart
solutions for safe, highly efficient, precise and easy to handle dredge operations. VOSTA
LMG's formula for these systems are: Safety, Efficiency, Precision & User Acceptance.
Safety
The DCMS supplies essential information about how the dredge functions and the
environmental influences which can affect this process. It shows all incoming information
and simultaneously controls all relevant functions. Good visibility of equipment (Mimic
Pages), instruments and controls guarantees a safe and smooth dredging process.
Efficiency
Real-time production monitoring and automated operation of dredge equipment support
the
operator
in
maintaining
the
highest
possible
production.
Precision
Real-time monitoring and tracking of primary dredge tools ensure continuous and precise
control
of
dredging.
User
Acceptance
ETL (easy-to-learn) and ETU (easy-to-use) principles drive development of the user
interface, which is always done in close cooperation with operators.
All this results in a highly integrated Dredge Control and Monitoring System with
decentralised process interfaces, centralised controlling units and human interfaces.
The DCMS is available for new vessels as well as for refits. It combines all required
monitoring and control functions of the complete dredging installation by means of
different, task-oriented mimic diagrams and will be designed to the customer needs.
Depending on the type of automation, the visualisation system is either a stand-alone
system or has a clientserver structure. For maximum reliability, redundant systems are
available
as
well.
The mimics show the status of the equipment in real time. The mode is shown by means
of different colours. All relevant process data are shown in bar graphs or numeric values.
The operation is done mainly by a trackball or function keys. The trackball is again used
to click the appropriate symbols, which change in colour and start to flash to confirm
their function.

Page 2 of 6

Suction pipe position indication


Nowadays there is fierce competition on the world dredging market to win tenders for
dredging contracts. In order to be competitive, a dredge must perform with a high level
of efficiency. Failure to meet project targets can lead to difficulties with the client
however, overdredging reduces profits. For efficient dredging, VOSTA LMG offers
various side suction pipe measurement systems.
The suction pipe positioning indicator (SPPI) calculates and displays the position of the
suction pipe with reference to the ship's side and the water line. The position
calculation is performed using the information supplied by angle sensors mounted on
the pipe or the uncoiled rope length of the winch with the respective transverse angle of
the rope. Typically two views of the side suction pipe are displayed on the screen.
These are the side elevation view and the plan view. The side elevation view provides a
representation of the pipe giving an indication of the draghead depth and pipe vertical
angles. The plan view gives an indication of the position of the pipe in relation to the
ship side and of the pipe horizontal angles.
The mimic also shows bar graphs and values for the dredging depth below the water
line and below the keel, the position of the draghead relative to the ship's side and the
various horizontal and vertical angles of the respective pipes. Additionally the system
warns the operator in case of a dangerous pipe position. Interfaces to various survey
systems are also available, as well as the possibility of internal or external data
logging.

Page 3 of 6

Draught and load calculator


The Draft and Load Calculator is used for continuous, quantitative determination of the
dredged material. The system provides the operator with a quick and complete overview
of the status of the ship (draught, trim, heel and hopper level), the operational data
(displacement, load, solid) and the operation mode. The measured and calculated data
are shown on a dedicated screen in a clear manner, displaying all relevant data.
With this system the operator can easily decide when the target load has been reached
and is informed about his daily efficiency at all times. The reports generated by the
system provide a comprehensive documentation of each job. The determination of the
load condition of the ship is performed using the values measured by various sensors.
The draught is measured by a selection of pressure sensors and the hopper level is
measured by different ultrasonic sensors. Trim is calculated from the values provided by
the draft sensors. The ship's displacement and hopper volume are obtained from the
tables stored in the loading computer using the draught, hopper level and trim values.
The dredged material without water content (dry) is calculated using the displacement
and volume values obtained from these tables and the manual inputs of the seawater, in
situ
and
dry
density.
The measured and calculated data are shown on a dedicated screen in a clear manner,
displaying all relevant data.
In addition reports can printed per trip, consisting of the most important data values and
the loading/unloading diagram. The main trip data is also stored for later reporting and
evaluation by third party systems using office tools such as Excel or Access. VOSTA LMG
offers a range of DLC varying in accuracy and accordingly in price.

Page 4 of 6

Draught measuring system


The micro computer controlled draught measuring system measures and monitors the
ship's
draught's
contents
continuously.
All data, parameter, measuring values etc. arc indicated on an LCD display located in the
controlling unit. An integrated keyboard makes the selection of the individual draught
points and the system's operation very easy. The controlling unit can be mounted in a
switchboard.
Following measuring values can be indicated:
PP draught forward

(Draught at perpendicular)

PP draught aft in cm

(Draught at perpendicular)

Draught forward in cm (Draught at draught mark)


Draught aft in cm

(Draught at draught mark)

Trim in cm
Trim in degrees
List in cm
List in degrees
Seawater density

(manual input)

The draught measuring system presents alternatively on demand the true draught at the
perpendiculars or the draught at the draught mark. The true draught indication is
corrected automatically by measuring the ship's trim and list through an inclinometer.
Seawater density
automatically.

can

be

changed

manually,

all

indications

will

be

corrected

All date and values can be transmitted to computer stations and/or to the ship's
monitoring system via a serial connection RS 485.

Page 5 of 6

Production monitor (PM)


Production monitor is available for all dredging applications. Its purpose is to assist in
increasing production. Using the information from the PM data display allows the
operator make the right decision at the right time. Knowing the actual density, velocity
and
production
leads
to
efficient
operational
decisions.
The production meter comprises the density measurement, the velocity measurement,
the calculator incl. display and optionally an analogue instrumentation panel.
Based on the absorption of nuclear radiation, the density measurement is performed
using either Caesium or Cobalt as the radioactive source. The decision which source
should be used depends on the size of the mixture pipe and the required measuring
range. Whereever possible Caesium 137 is used due to it's longer half life. Today the
modern scintillation counters allow using Caesium with its low radiation for large nominal
diameters mixture pipes, in situations where Cobalt was necessary in the past.
The velocity measurement units (electro magnetic flow rate) are available for pipes of
varying diameters, with or without liner wear detection.
The Production Monitor calculates and displays the actual production. On a touch TFT
display the actual velocity, density and the calculated wet and dry production are
presented in a bar graph, as numeric values and in a time-dependent curve.
For adjustment all necessary settings can be entered directly via a dialog box on the
touch screen.

Page 6 of 6

Dredging and Reclamation: Trends and Future


(by H. van Muijen 1) (IHC MTI Holland)
Dredging is an industry in constant transformation. It has changed tremendously in the last decades. The
traditional dredging activity like construction and maintaining ports and harbors, desilting of drainage and
irrigation channels, keeping reservoirs at depths and removing sediments from waterways is still of
importance. However it is surpassed by other applications of dredging technology. Reclamation is an
important example in this respect, where large amounts of sand are dredged, transported over large
distances and used to make new land for industrial-, housing-, airport- and other infrastructural purposes.
Beach nourishment and other protective measurements should be mentioned here as well. Other
applications like offshore winning of oil & gas, environmental clean-up, mining of interesting minerals like
gold, diamonds, titanium and tin as well as aggregates from inland and offshore deposits are of importance.
For all of these dredging applications equipment is required and with changing circumstances and
constraints, the design and use altered tremendously. Larger dredging depths and increasing capacities
require an impressive scale-up of equipment and tools. Operation under more difficult circumstances and
with other soil characteristics necessitate development of new dredging components and systems like
submerged pumps, active dragheads, special cutter heads and wear resistant materials. Higher accuracy
and selectivity require a better control which leads to more automation and development of enhanced
control and monitoring means. Design and construction of this kind of modern dredging technology also
requires a better understanding of the dredging processes. This requires more focused research on
dredging fundamentals and development of innovative tools, systems and components.
Based on the developments in dredging applications of the last decade, this paper will highlight expected
trends for future use and summarize the required dredging technology development to enable such use.

Introduction
For centuries, water has been mans friend and opponent. We try to transport goods and
people over water and try to survive during storms and floods. Waterways and harbours
are required to fulfil our transport needs. Dikes and other structures are built to protect
us against flooding. New land is conquered from the sea and beaches are restored.
Global economic development requires new energy sources of oil and gas, which are
exploited from offshore deposits more and more. A lot of minerals like gold, diamond, tin
and industrial minerals are often found in wet environments. The same counts for
aggregates for concrete manufacturing and other building material applications.
Contaminated wet soil has to be cleaned in order to reduce a negative impact on the
environment.
To realize all this, dredging equipment is of vital importance. It is a requirement to
gain the experience and knowledge to build this equipment.
Recent changes and expansions in dredging assignments led to the development and
innovation of equipment to comply with the increased demands for dredging tools. It is
expected that this growing trend will continue the coming years which requires ongoing
innovation and development of dredging equipment.
Trailing suction hopper dredge design
Land reclamation is one of the main areas of interest for the dredging industry. On many
locations in the world expansion of industrial areas is desired for e.g. airports, container
terminals or industrial plants and residential areas. Although this has been a dredging
activity that goes back for a long time, the scale up we see nowadays is tremendous.
Hongkong, Singapore, Palm Island I, II + III and The World are some examples of
extensive land reclamation projects of recent times. Many millions of m3 of sand have to
be transported to connect islands and construct areas; existing waterways are deepened
to give access to large ships.

1)

General Manager MTI Holland BV (www.mtiholland.com


PO Box 8, 2960 AA, Kinderdijk, Netherlands
p: +31-78-6910335; f: +31-78-6910331; e.mail: h.vanmuijen@mtiholland.com

Page 1 of 11

Environmental, geological and political considerations may result in long sailing


distances and the need for dredging in deeper waters. These constraints ask for large
Trailing Suction Hopper Dredgers (TSHDs) with high dredging capacity and efficiency.
Especially when sailing distances tend to increase, sand can be dredged at lower cost
with a large capacity TSHD compared to a smaller unit. Until 1992 TSHDs had a
capacity less than 10,000 m3.

Figure 1: Palm Island in Dubai and project in HongKong (new airport Chec Lapkok)

After 1992 a change of this trend occurred. In 1994 the first so called Jumbo Hopper
Dredger, the Pearl River (17,000 m3) was build. Followed in 1998 by the Volvox
Terranova (20,000 m3), in 2001 by the Ham318 (23,700 m3) and in 2000 the Vasco da
Gama (33,000 m3). The last one is also referred to as the first Mega Hopper Dredger.
These are examples from a range of large TSHDs from which a substantial part was
built by IHC Holland. Recently the Fairway has been extended and can carry 35,000 m3
now. By this extension it became the second Mega Hopper, soon to be followed by
more.

Figure 2: Jumbo TSHD HAM 318 (compared to 2 Boeing 747) and the Pearl River and Nile River working
together on a reclamation project

Besides the application of TSHDs for reclamation projects, other use of this type of
dredgers is of importance as well, but not in similar size ranges. Sand suppletion for
beach nourishment and for deposition close to the shoreline require shallow draft
hoppers. As an example the Waterway and Coastway, built in 2000 and 2002 are
able to sail close to shore. The loaden draught of both ships is only 6.6mtr.

Page 2 of 11

For specific maintenance dredging tasks hopper designs will be fitted to suit this
purpose, allowing a maximum volume of soil to be dredged at relatively low specific
densities. Similar hopper design specialties can be mentioned like those for offshore
aggregate dredgers and offshore oil recovery.
In general one can state that a trend in the design for TSHDs can be noticed towards
more specialization. The better the special design is adapted to a specific application,
the more cost effective a dredge can be exploited.
Trenching and glory holes
With increasing demand of oil and gas a lot of new offshore deposits have to be
developed. Besides a proper preparation of the actual exploration and exploitation sites,
these energy sources have to be transported and landed ashore. A proper base for
drilling rigs, protective placement of valve arrangements on the seabottom, pipe and
cable laying and large landing facilities and port structures are required at an ever
increasing scale. The trend towards deeper oil and gas winning projects also requires
deeper dredging possibilities.
Trenching is preparing the oceanfloor to bed in a pipeline. There is a variety of
techniques to trench a pipeline depending on the type of soil and an important one of
them is by means of dredging techniques.
After the trench is formed and the pipeline is installed it is sometimes necessary to cover
the pipeline to protect it against anchors, fishing nets, etc. There are several covering
methods like rock dumping, natural ocean currents and backfill by the suction pipe of a
hopper dredger. In the latter case, the sand in the hopper will be transported to the
trench by use of pumping system and the suction pipe.
Many large TSHDs nowadays have the possibility of deep dredging for trenching or the
creation of so called glory holes for de protection of oil and gas winning structures at the
oceanfloor.

Figure 3: oil and gas winning structure located on the ocean floor

A number of Jumbo Dredgers have been equipped with a deep dredging installation that
is capable of dredging down to depths of 130 m. To reach this depth and maintain
sufficient inlet pressure for the dredge pump for proper functioning, high efficiency
submerged pumps have been fitted in the large size suction pipes with diameters up to
1400 mm. Installations with larger dredging depths are technically feasible. The large
sized submerged dredging pumps require highly powered drive systems up to 6500 kW.

Page 3 of 11

Finite Element Methods


Dredging projects are performed under more and more demanding circumstances.
Operating at larger dredging depths and increased capacities, more constraining deposit
characteristics and site conditions, all within high efficiency requirements. This puts
stress on the design and material selection.
Finite Element Methods (FE methods) are essential during the design process of a
dredging vessel. The professional use of these calculations has led to strong and solid
constructions which are relatively light weighted. Because of this light weight they are
able to carry a relatively large load.
Due to the very nature of dredging operations in open sea conditions and the continuous
process of loading respectively discharging of soil, the ships structure of a TSHD
experiences a variety of cyclic loads in a strong corrosive environment. Design features
of a TSHD include a relative small ratio of hopper to ship length, implying a
concentration of payload in the midship region, resulting in high hull girder bending
moments and shear forces. Furthermore, the presence of heavy dredging equipment, in
combination with the hopper structural arrangement including bottom door openings,
enabling the discharge of soil, inevitably requires the structural design of the ship with
numerous discontinuities. To optimize the stress-weight ratio of the ship structure and to
maximize the strength and fatigue life of structural details finite element (FE) calculations
are performed.
Large Cutter Suction Dredgers (CSD) experience large fluctuating loads in exposed
working conditions induced by waves and structural vibrations when cutting rock for
example. The probability of crack initiation is relatively high at locations in the ship
structure forming the (flexible) link between seabed and ship, ladder trunnions and spud
carrier. FE calculations are performed to reduce stress concentrations to a minimum and
to maximize fatigue life. FE calculations are further performed to optimize the strength
and stiffness of cutter ladder and spud. To prevent resonance or excessive deformations
due to vibrations, FE-models are being setup of the cutter ladder and ship to judge the
vibrational behavior.
High sailing efficiency
The ships sailing resistance depends on the waterflow around the hull and the waves
made by the ship during sailing. By designing the hull correctly, its sailing resistance can
be reduced significantly and its overall energy economy will be improved. The design of
the bows is of special importance. On many large ships, a so called "bulb" (i.e. artificial
nose) is mounted on the hull at the waterline. The bulb changes the flow round the hull in
such a way the resistance is reduced because of the reduction of induced waves.

Figure 4: practical results of improved hull shapes by CFD calculation; left the Lange Wapper (without bulb);
right the Charlemagne with bulb.

The waves induced by the ship are mainly caused by the shape of the stem. Picture 4
shows a ship with a conventional stem and one with a bulb. It can be seen that the

Page 4 of 11

induced waves of the stem with a carefully designed bulb are significantly smaller
compared to the original stem.
The experience on design of ship hulls gained with CFD calculations resulted in a
special shape of the stern. The stern is designed for an ideal approach of the flow to the
propellers resulting in less vibration and a higher efficiency of the propellers.
Computational Fluid Dynamics (CFD)
CFD methods make it possible to calculate the impact of changes to the design of the
ship. Like already mentioned, the stem of the ship has a large impact on the sailing
efficiency. Also the interaction between the hull and the propellers can be calculated by
means of CFD. Vortices induced by the hull have a certain effect on the efficiency and
may create vibrations. CFD methods give information which supports an optimized
design of the ship, where the design of the stern result in an optimal flow to the
propellers.
For a dredging vessel information on the effect of wind, waves and currents is essential
to predict the manoeuverability during differing weather-conditions. The
manoeuverability can be predicted by use of certain coefficients which are calculated by
CFD methods.
Heavy duty cutters operating in swell
Construction and extension of ports like that of Quatar and other ones around the world
require hard digging capabilities. This also applies for a lot of offshore oil and gas
applications. Heavy duty cutter suction dredgers like the JFJ de Nul and the d Artagnan
are able to extend large cutting forces. Often operations have to be performed in swell.

Figure 5: Self sailing heavy duty cutter suction dredger JFJ de Nul and flexible spud carriage installation of
dArtagnan

To allow high cutting forces under these conditions earlier designs of larger cutter
suction dredgers used sturdy spud constructions. The tendency to install even higher
cutting forces up to 6000 kW on the cutter head of the above mentioned cutter dredgers,
made it difficult to come up with strong enough designs of the spud systems. This led to
the design and use of flexible spud installations that allow high cutting forces in sea
conditions with swell up to 1.5 2 m.
Pumps
IHC developed high efficiency dredge pumps. With the same installed power more
production can be achieved compared to the so called standard pumps.
To ensure pump performance and predict wear resistance, lab research supported by
CFD analyses is performed at the research department (MTI Holland) of IHC Holland.

Page 5 of 11

MTI Holland has a circuit where pumps can be tested. The pipe diameter in this circuit is
300mm which gives a reliable representation of the full scale pumps.
To increase production, the modern TSHD is equipped with jetpumps. They typically
have three main functions:
1. Fluidizing bottom material and reduction of cutting forces of modern dragheads
2. Fluidizing dredged material in the hopper during discharge in order to reduce the
discharge time.
3. Clearing the hopper after discharge.
Versatile, limited size high efficiency dredge pumps are more and more used as jet
pumps in order to prevent high wear rates caused by a sometimes high content of sand
particles entrained in the jet water.

Figure 6: IHC range of high efficiency dredging pumps; from left to right medium pressure, cutter special
and high pressure pump

To optimize efficiency of the dredging process, the pump speed, during the dredging
process, has to be optimally adjusted to dredging parameters such as vacuum, mixture
velocity and density. The Variblock (developed by IHC) is a continuously variable speed
transmission gearbox that offers outgoing speed variation with minimal loss of efficiency
at constant input power and speed. Various output speed options are possible, for
instance, an increase or decrease in output speed of 15% with an overall gearbox
efficiency of 94%. The Variblock makes it possible to avoid adjusting the impeller
diameter for different pipeline lengths in case of stationary hydraulic dredgers like cutter
suction dredgers. The investment is significantly lower than for an electric drive for the
pump. All hydraulic components are in stock around the world and immediately available
on demand.
Besides a high efficiency demand, pumps for cutter suction dredgers also require a large
ball clearance. This is especially inevitable for the larger heavy duty cutter dredgers that
are used in harder materials. Harder material is dredged in larger particle size ranges
and to avoid blockage a larger clearance is of essence. IHCs Cutter Special Pump
provides such a large ball clearance at high pump efficiencies and suction capabilities.
Dragheads
The draghead of a Trailing Suction Hopper Dredger has a major effect on the
performance of the dredger, so its design, quality and versatility are essential.
The increase in size of the TSHDs is also to be found in the development of dragheads.
One of the first dragheads was suited for suction pipes with a diameter up to 300mm.
The latest developments show dragheads for a pipe with a diameter of 1400mm.
While the production of excavated soil is governed by draghead width, thickness of the
layer removed and trailing speed, other factors can also play a role such as the required

Page 6 of 11

trailing force. Draghead design is always considered in close relation to all the other key
elements of the dredging process.

Figure 7: Modern active draghead

Many types of dragheads have been developed through the years. Originally, dragheads
were fully based on the principle of erosion: a water flow was created between the
movable visor of the draghead and the bottom. In the so called Dutch (IHC-) head this
water flow entered the draghead mainly on backside of the visor, in the California
draghead mainly at the sides of the visor. By the water flow over the bottom, creating a
pressure difference between bottom and draghead, material (sand) was loosened and
taken up by the draghead. Nowadays modern dragheads are equipped with jet water
supply over the total width of the draghead in the fixed part and with cutting knives or
cutting teeth in the movable visor. With the combined effort of the two, the required
pressure difference over the draghead is reduced while the production is significantly
increased. Through adjustable water inlet flaps on the backside of the visor, sufficient
additional water can be supplied. In some cases the density of the mixture dredged and
the amount of jet water supplied, are so high that only a minor amount of additional
water is required. Sometimes the visor is kept in position relative to the fixed part of the
draghead by means of hydraulic cylinders, so that it is possible to counteract an upward
movement resulting from the cutting knives (figure 7). This can be done to a pre adjusted
value of the force.
In very fine compact sands the penetration of the knives or teeth can, even with the
conventional supply of jet water, be insufficient. This results in low production. To
overcome this, studies and model investigations have led to the development of the so
called IHC Wild Dragon Head. Investigations showed already promising results which

Page 7 of 11

were proven by full size tests in the entrance channel of Shanghai. An additional
advantage is a further reduction of the specific energy: the amount of kW required per
m3 of dredged material.
The Wild Dragon drag head allows TSHD operation in material conditions earlier only
possible with cutter suction dredgers. Especially when dredging in busy ports and
entrance channels, the manoeuvrability and navigational flexibility of hopper dredgers is
an advantage as compared to stationary dredgers.

Figure 8: IHC Wild Dragon draghead

Loading Process
An improvement in the efficiency and the profitability of a TSHD can also be achieved by
improving the settling process in the hopper. The loading time of a TSHD is particularly
sensitive to the settlement when operating in fine sands. An efficient settling process
leads to a shorter loading time and therefore to a shorter dredging cycle, which in turn
will lead to a more profitable operation. Moreover, and this may even be more important,
the amount of sand contained in the hopper increases. Considering the investments
required today, a more efficient operation leads to a substantial improvement of a
vessel's profitability.
In the interest of gaining new insights, IHC initiated an extensive hopper loading
research program in cooperation with Dredging International (DI). To improve the
knowledge and to be able to compare several design alternatives on a realistic scale,
MTI has built a large-scale test rig, which was based on the TSHD Antigoon.
The settling of the soil not only depends on the hopper design but also on the inlet flow
into the hopper.
Besides the design of the hopper, the settling of the soil is among others dependent of
the inlet flow in the hopper. Appropriate chosen velocity distribution of the inlet flow for
example will improve the settling process significantly. The observations and evaluations
of the tests with the large-scale test rig have led to new insights in general as well as in
specific situations. Applying modifications based on the acquired insights made it
possible to develop new design tools and to improve the hopper design in order to
improve the hopper settling efficiency and thus the net production of the TSHD.

Page 8 of 11

Discharge Process
Besides the loading process of a hopper dredger the discharge process also defines the
overall economics to a large extend. Besides dumping material through the bottom
doors, two other methods of discharge are available, i.e. rainbowing or using floating
pipelines. They both use the inboard pumping system to empty the slurrified hopper load
and pump it ashore. Especially the rainbowing method offers a high efficient operation
with a relatively short pumping distance. It will be clear however that the discharge
method is determined by local conditions.

Figure 9: cutter suction discharge pipe with high density slurry mixture

Emptying the hopper in these cases is a very controlled process. By using a balanced
design of self emptying channels, high pressure jet system, inboard pumps, discharge
system and operational control automation, very high specific densities can be gained
during the discharge process. Normal practice allows densities up to 1.6 and even 1.7.
(t/m3).With a natural average deposit density of 2 t/m3 this is very near to the maximum
physical possibilities and required adaptation of the generally accepted theoretical
phenomena for hydraulic transportation.
A similar approach accounts for other discharge pump operations. The use of
submerged pumps in stationary dredgers also allows high density mixture pumping.
However, it will be clear that this is only possible when the material actually entered the
pipe. A proper balance between hydraulic pumping system and cutting head operation is
a prerequisite.
Automation and control.
Operating the dredger accurately at a set position or track is often a necessity, in
particular when trenching, dumping or pumping ashore. Not only the dredging efficiency
of a TSHD may be improved, but in some cases a Dynamic Positioning and Dynamic
Tracking (DP/DT) System is even a requirement to prequalify the dredging equipment for
an offshore or ports maintenance dredging job.
Integrated control systems take care of many parameters during the dredging process
by the use of advanced measurements and automation systems. For instance: a pump
controller maintains a mixture velocity just above the critical speed, so particles do not
settle while the flow resistance is minimal. This gives the best fuel and wear efficiency.
Automatic control of the mixture velocity improves hopper loading, which ensures better
settling of the finest particles. Thereby overflow losses, fuel consumption and wear are
reduced. It can also be used in other types of dredgers.
Relevant data are comprehensively presented on video screen(s). Presentation of data
and the control of all functions, such as: loading process, automatic dredging, discharge

Page 9 of 11

process, power management, reporting, communication with survey equipment and


geographic positioning, take place at one or more MMI's (Man Machine Interface). For
optimum operator environment, the Operator Control Center (OCC) is introduced as an
operator-friendly workstation.

Figure 10: Integrated bridge

Wear and tear


Dredging is hard labour for equipment and tools. The scale up of equipment led to other
type of material use, necessitated to keep the required strength within allowable design
weights. Besides this requirement, other material quantities are aiming at a reduction of
wear and tear. In order to reduce costs more wear resistant materials were developed to
be used in more wear resistant designs.
This applies for all components in the sand bearing systems and already led to an
impressive reduction of wear costs.
Wear resistance is not the only cost effective measure. Prediction of wear processes
allows defining the right moment of exchange. This can lead to proper wear
management with a minimum of life cycle costs. IHC developed a wear management
package that predicts the wear of pumps and pipe lines. This will allow dredging
management to select the best timing for exchange.
Environment
Our environment gets a lot of attention and a sustainable use of it is very important. This
counts especially for water environments as biological balance is very sensitive to
changes. The environmental impact of dredging operations is a topic that gets a lot of
attention. As an example port management is confronted every day with the effects of

Page 10 of 11

environmental impact and control, which necessitates dredging contractors to comply


with strict environmental rules and regulations. This also affects the design and
operation of dredging tools, which is of importance to leading dredge builders like IHC
Holland Merwede as well.
More and more attention is focused on low impact designs, aiming at lower turbidity,
exhaust and noice levels, reduction of vibrations and power use, less wear and more
efficient operation. This will not only lead to a more sustainable use of the dredging
equipment, but also to a better economic operation.

Figure 11: MTI Holland pump test facility

Conclusion
It is evident that innovations and developments in the present dredging industry have not
come to an end. In the years to come a lot more development will continue to play a major
role in providing dredging solutions for a wide area of applications.
Important considerations in this respect are an improved sustainable use with less
environmental impact, more efficient operations in more striking and demanding
circumstances as well as for more difficult materials.

Page 11 of 11

You might also like