You are on page 1of 9

WELDING RESEARCH

SUPPLEMENT TO THE WEL DI NG JOURNAL, MARCH, 1980


Sponsor ed by t he Amer i can Wel di ng Soci et y and t he Wel di ng Research Counci l l l D r
Effect of Manganese on the
Microstructure and Properties of
Al l -Wel d-Metal Deposits
Going from 0.6 to 1.8% Mn increasingly refines weld
microstructures and promotes acicular ferrite formation,
and optimal impact is attained with approximately 1.5%
Mn although strain aging affects notch toughness and
displaces optimum Mn to a higher concentration
BY G. M. EVANS
SYNOPSIS. The effect of manganese,
in the range 0.6 to 1.8%, on the mi cro-
structure and mechanical properties of
manual metal arc deposits (ISO 2560)
has been investigated. It was f ound
that manganese increasingly refined
the mi crostructure and promot ed the
f ormat i on of acicular ferrite. Both ten-
sile strength and yield strength i n-
creased by approxi matel y 10 N/ mm
2
per 0.1% Mn addi t i on to the deposit.
Charpy V, Schnadt and COD tests
graded "as-deposi t ed" wel d metals in
the same relative order, the opt i mal
impact properties bei ng attained at a
manganese level of approxi matel y
1.5%. Stress relieving was f ound to
have only a marginal effect on impact
properties. Strain aging, on the other
hand, markedly affected notch t ough-
ness and displaced the opt i mum man-
ganese level to a higher concentra-
ti on.
I nt roduct i on
The worki ng program of Subcom-
mission l l -A of the Internati onal Insti-
tute of Wel di ng calls for a j oi nt effort
to study the mi crostructure of wel d
metal. As a first step, four al l -wel d
metal deposits have been di stri buted
to various laboratories wi t h recom-
mendati ons' for the characterization
of the microstructural components.
The present paper details the f i nd-
ings of the Swiss del egati on in col l abo-
ration wi t h the Wel di ng Institute
(Uni ted Ki ngdom). In addi t i on to the
metallographic studies, a test program
was conducted to evaluate the i n-
fluence of manganese on the tensile
and impact properties of the wel d-
ments.
Experimental Procedure
Electrodes
Four experi mental iron powder type
basic electrodes, coded A, B, C and D,
Paper to be presented at the AWS 61st
Annual Meeting in Los Angeles, California,
during April 14-18, 1980.
G. M. EVANS is Chief Metallurgist, Welding
Industries Oerlikon Buehrle Ltd., Zurich,
Switzerland.
were prepared using 4 mm (0.16 in.)
diameter core wi re. The ferro-manga-
nese contents of the coatings were 3,
5, 7 and 9%, respectively, and the ferro-
silicon content was balanced. The
coati ng factor was 1.70 and the elec-
trodes were baked for 1 h at 400C
(752F) to yi el d a di ffusi bl e hydrogen
content of 2.3 ml /100 g deposit,
accordi ng to the ISO procedure.-
Weld Preparation
The wel d preparation empl oyed was
that specified in the Internati onal
Standard for the code of symbols for
manual metal arc electrodes, namely
ISO 2560-1973.
Wel di ng was done in the flat posi -
ti on using the stringer bead techni que.
Direct current (electrode positive) was
empl oyed, the amperage bei ng 170 A,
the voltage 21 V and the heat-i nput
nomi nal l y 1 kj / mm. The interlayer
temperature was 150 C (302F).
Heat Treatment
The wel dment s were tested in bot h
the as-welded and the stress-relieved
WEL DI NG RESEARCH SUPPLEMENT I 67-s
Table 1-VVeld Metal Composition (As-Welded), Wt-%
Electrode C Mn Si S O
A
B
C
D
0.035
0.038
0.049
0.051
0.66
1.00
1.42
1.82
0.30
0.30
0.34
0.34
0.006
0.005
0.005
0.006
0.013
0.014
0.013
0.017
0.007
0.010
0.009
0.009
0.049
0.046
0.041
0.039
(2 h/580C or 2 h at 1076F) condi t i on.
Impact tests (Charpy V notch) were
also conduct ed on strain aged speci-
mens, compressed 10% and aged for Vz
h at 250C (482F).
Mechanical Testing
Two sub-size al l -wel d-metal tensile
specimens (Mi ni trac) were machined
and tested for each type of el ectrode
and condi t i on. Also approxi matel y 35
Charpy V notch specimens were
struck, so as to obtai n the compl ete
transition curve.
Schnadt impact specimens
3
were
prepared f rom as-welded deposits and
were tested under bradycoheracy (B)
and tachycoheracy (K) condi ti ons. In
addi t i on, as-welded plates were COD
tested in ful l thickness (20 mm or 0.79
in.) at the Wel di ng Institute. The wel d
metal was saw notched (0.15 mm or
0.006 in.) transversely to provi de sub-
sidiary-type specimens, as proposed in
DD 19" but wi t hout a fatigue crack.
Results
Chemical Composition
Typical chemical analyses of the
deposits are given in Table 1. The
systematic increase in the amount of
ferro-manganese in the coati ng re-
sulted in four distinct wel d metals
cont ai ni ng, nomi nal l y, 0.65, 1.0, 1.4
and 1.8% Mn.
The wel d silicon content was rela-
tively constant, but the carbon and
phosphorus contents increased pro-
gressively over the range. Wel d metal
oxygen level, on the other hand,
decreased, thus substantiating the
deoxi dati on potenti al of manganese.
Essentially the same results as given in
Table 1 were obtai ned on repeated
analysis for the stress-relieved sam-
ples.
Metallographic Examination
General. A transverse section of one
of the mul t i -run deposits is shown in
Fig. 1, a total of nine layers bei ng
required to fi l l the gap. Three beads
were deposi ted per layer, and the
macroscopic effect was of repeated
sequences of as-deposited and super-
cri ti cal l y heat-affected wel d metal
zones.
The wi dt hs of the col umnar, coarse
grained and fine grained regions were
measured in the vertical mi d-pl ane
posi ti on and the dupl i cate results,
obtai ned by exami ni ng as-welded and
stress-relieved specimens, are de-
picted in Fig. 2. The percentages of the
Fig. 7Cross section of multi-run deposit
5 -
LU
U_
O
LLI
o
<
LL
CC
ID
CO
Q .
O
s
o
LL
10
15
LLI
o
< 2 0
l -
cn
a
columnar
coarse , "I fi ne
grained Jgrai ned
Plate
surface
E
E
o
.c
u
>
o.
CO
.c
O
- STRESS- RELI EVED
Fig. 2 Zone distribution along the vertical centerline position
Table 2Zone Percentages in the Equivalent ISO-V Notch Position (AW = As-Welded,
SR = Stress-Relieved)
Zone
Col umnar
Coarse grai ned
Fine grai ned
AW
18
35
47
SR
32
24
42
AW
23
34
43
SR
19
35
46
AW
22
34
44
SR
12
37
51
AW
11
34
55
SR
20
37
45
Average
20
34
46
68- sl MARCH 1980
>: :;K> 1m
J H
Fig. 3Photomicrographs of top heads (co-
lumnar). X200 (reduced 35% on reproduc-
tion)
US
"HP
;- V"'
im.^MmM^iMmiM
Fig. 4Photomicrographs of coarse grained
regions. X200 (reduced 35% on reproduc-
tion)
m
..
&
w t ' J S V " ' " * ' '' V' '' i
F/g. 5Photomicrographs of fine grained
regions. X.315 (reduced 35% on reproduc-
tion)
zones in the central 10 mm (0.39 in.)
region of the deposits, i.e., the Charpy
V-not ch l ocati on, are given in Ta-
ble 2.
The wi dt h of the col umnar regions
varied from layer to layer, and penetra-
ti on was such that, in some instances,
t wo critically heated regions were
adjacent to one another. A difference
due to manganese coul d not be ascer-
tai ned, the values for dupl i cate speci-
mens scattering to an equi val ent
extent. The average value for the sum
of the coarse and fine grained regions
at the notch l ocati on was f ound to be
80%. A slight vertical di spl acement
woul d affect the relative proporti ons
of the zones, since the l ower runs
tended to contai n wi der col umnar
bands.
The col umnar grains broadened as
the wel d progressed duri ng deposi -
t i on, due to the epitaxial growt h
effect. As an approxi mati on, however,
it can be presumed that the sequence
is repeti ti ve t hroughout and that the
central t op bead and the adjacent
heat-affected wel d metal serve to
characterize the bulk of the deposit.
Typical microstructures of the four
manganese-containing wel d metals
are shown in Figs. 3, 4, and 5, for the
col umnar, coarse grained and fine
grained regions, respectively.
Columnar Region. The t op central
bead of each specimen was examined
at X200 and quanti tati ve metal l ograph-
ic measurements were made as de-
scribed in Doc. II-A-389-76
1
, using a
WEL DI NG RESEARCH SUPPLEMENT I 69-s
Fine grained
100
90
Fig. 6Diagram of top bead and adjacent areas
Swift poi nt counter. The area traversed
measured 2.5 X 2.0 mm- (Fig. 6), and
500 points were recorded by each of
t wo investigators.
Three major mi crostructural compo-
nents (Fig. 3B) were i dent i f i ed, name-
ly:
1. Pro-eutectoid ferrite (light etch-
ing).
2. Intermedi ate lamellar products,
mai nl y ferrite side plates resembling
upper bai ni te (light etchi ng).
3. Aci cul ar ferrite, consisting of a
fi ne structure of i nterl ocki ng ferrite
plates (dark.etchi ng).
The results obtai ned on poi nt count-
ing are pl ot t ed in Fig. 7. It can be seen
that the amount of acicular ferrite
increased markedly, at the expense of
pro-eutectoi d ferri te, as the manga-
nese cont ent increased. Also, a clear
trend existed for the i ntermedi ate
lamellar component to decrease wi t h
increasing manganese.
Carbon replicas of the t op beads
were exami ned at the Wel di ng Insti-
tute, in a transmission electron mi cro-
scope (TEM), and a linear intercept
met hod was appl i ed at a magni fi cati on
of X2500. The results are given in Table
3, the values for the high manganese
welds bei ng i ndi cati ve of the acicular
Table 3Average Linear Intercept in Top
Beads of As-Welded and Stress-Relieved
Specimens
Average linear intercept, jtm
Electrode
A
B
C
D
As -depos
3.30
2.87
1.72
1.05
ted St
r
ess-relieved
3.96
2.60
1.70
1.59
ferrite lath size.
Examination of the replicas showed
that there was a gradual transition
between acicular ferrite and pro-
eutectoi d ferrite. Also, the di sti ncti on
normal l y made between the t wo
microstructures in the opti cal mi cro-
scope was purely arbitrary at high
magni fi cati on.
Small and wi del y dispersed areas of
retained austenite were observed on
the replicas. The amount of austenite
increased wi t h increasing manganese
but onl y in the case of wel d D was
sufficient austenite present (1%) to be
detected by X-ray di ffracti on. Al l the
replicas from stress-relieved wel ds
coul d be readily di sti ngui shed by the
presence of grain boundary carbides
~AS WELDED
O-STRESS-fl ELIEVED
ACICULAR
FERRITE (3)
PRO-EUTECTOID
FERRITE (1)
0-5 K) 1-5
MANGANESE IN WELD, % .
Fig. 7Effect of manganese on microstruc-
ture of top bead
whi ch were formed by the temperi ng
out of the retained austenite.
Any martensite whi ch mi ght have
formed wi t hi n the retained austenite
was di ffi cul t to detect because of the
fi ne scale of the structure. Such i ndi ca-
tions of martensite, as were f ound,
were of considerably smaller areas
than those reported by Garland and
Ki rkwood
5
as occurri ng in submerged
arc wel ds. Furthermore, it was not
possible to i denti fy the areas as either
lath or t wi nned martensite or t o assess
them quanti tati vel y.
Coarse Grained Region. Phot omi -
crographs of the reheated wel d metal
taken di rectl y bel ow the central t op
bead are shown in Fig. 4. Wi t h increas-
ing manganese the structure became
increasingly more dark etchi ng, and
the pro-eutectoi d ferrite del i neati ng
the prior austenite grain boundaries
whi ch became finer and hence tended
to accentuate the coarse grained
nature of the zone. The fusion bound-
ary in the case of the lowest manga-
nese wel d (A) was di ffi cul t to locate
mi crostructural l y but the segregation
bands (ripples) coul d readily be seen
by varying the focus.
7
E
Q 5
O
O 1
As
1
'
deposited region.
Reheated region.
-
/
y y
X ^$ ^-' ,
y-"
W.I.
A .
J -
/
C,
..-
A
y-$-
--'
0 200 400 600 800 1000
NUMBER OF GRAIN BOUNDARIES INTERCEPTED .
Fig. 8Grain boundaries intercepted on
traversing as-deposited and reheated re-
gions
Table 4Linear Intercept Results From Fine Grained Region (H = horizontal,
V = vertical)
Intercept
per mm
155
146
Ratio
1.06
I nt er cept / mm
150
a
s
c
Interval / (fi)
6.7
P'
/2
, mm '-'
12.22
B
C
D
H
V
H
V
H
V
173
171
199
208
237
253
1.01
0.96
0.94
172
203
245
5.8
4.9
4.1
13.13
14.28
15.62
70-s I MARCH 1980
The mi cr ost r uct ur e wi t h i n t he pr o-
eut ec t oi d f er r i t e envel opes appear ed
t o be opt i cal l y i dent i cal t o t he aci cul ar
f er r i t e oc c ur r i ng i n t he as- deposi t ed
we l d met al . The areas s ur r ounded by
t he pr o- eut ec t oi d f er r i t e di f f er ed i n
si ze and t he boundar y of t he coar se
gr ai ned z one was di f f i cul t t o l ocat e,
si nce t he mi cr ost r uct ur e t e n d e d t o be
dependent o n t he under l y i ng sol i di f i -
cat i on and t r ans f or mat i on pat t er n.
Thi s l at t er p h e n o me n o n was par t i cu-
l arl y not i ceabl e at t he per i pher y of t he
t op bead wher e t he heat - af f ect ed
zone we l d met al had t r ansf or med
back i nt o c o l u mn a r t ype grai ns. Al so,
i nt er act i on oc c ur r ed b e t we e n super -
i mposed heat - af f ect ed zones, a t ypi cal
occur r ence bei ng t he c o n t i n u a t i o n of
a f i ne gr ai ned r egi on i nt o a coar se
gr ai ned r egi on at t he poi nt of i nt er cep-
t i on wi t h a new f usi on boundar y . The
depos i t i on sequence, howev er , was
such t hat ov er l appi ng of heat - af f ect ed
zones di d not occur at t he Ch V- n o t c h
l ocat i on.
The scanni ng el ect r on mi c r os c ope
( SEM) was used at t he We l d i n g I nst i -
t ut e t o st udy t he as- deposi t ed and
r eheat ed r egi ons of t he we l d me n t s . A
l i near i nt er cept me t h o d was appl i ed
and t he resul t s obt ai ned f or as- wel ded
speci mens are gi ven i n Fi g. 8, t he
change i n i nt er cept bei ng mo n o t o n i c
wi t h i ncr easi ng manganese. The f usi on
boundar i es wer e cl ear l y vi si bl e, and
t her e was al so a sudden change i n
l i near i nt er cept wh e n t he boundar i es
wer e cr ossed. The boundar i es be-
t ween t he i nt er cr i t i cal l y and f ul l y
r eheat ed r egi ons, howev er , c o u l d not
be l ocat ed by di r ect obs er v at i on, nor
wer e t hey det ec t ed by t he i nt er cept
measur ement s (Fi g. 8) , si nce l i t t l e or
no di s c ont i nui t y of sl ope occur r ed
wi t h i n t he r eheat ed r egi ons.
Fine Grained Region. The f i ne
gr ai ned r egi ons (Fi g. 6) wer e p h o t o -
gr aphed at X630, and l i near i nt er cept s
of gr ai n boundar i es wer e made as
descr i bed i n Doc . I I - A- 389- 76. The
resul t s obt ai ned f or t he ver t i cal
( t hr ough- t hi ckness) and t he h o r i z o n -
t al di r ect i ons are gi ven i n Tabl e 4 and
show a f ai r degr ee of equi axi al i t y.
The r eci pr ocal of t he squar e r oot of
t he mean gr ai n i nt er val is pl ot t ed,
agai nst we l d met al manganese c on-
t ent , i n Fi g. 9. A st r ai ght - l i ne r el at i on-
shi p was o b t a i n e d , manganese agai n
bei ng f o u n d t o have a mo n o t o n i c
i nf l uence. Of par t i cul ar i nt er est is t hat
t he pr esent gr ai n si ze measur ement s
can vi r t ual l y be super i mposed on
t hose r epor t ed by Tul i ani
6
f or r eheat ed
r uns of submer ged arc we l d met al .
Mechani cal Properties
Tensile Results. The t ensi l e test dat a
obt ai ned are gi ven i n Tabl e 5 f or bot h
t he as- wel ded and st r ess- r el i eved c o n -
5 0-5
10 1-5
MANGANESE IN WELD, %
2 0
Fig. 9Effect of manganese on the mean linear grain intercept (fine grained
region)
Table 5-Tensi l e Test Results
1
"
1
Condi t i on
As-wel ded
Stress-relieved
El ectrode
A
B
C
D
A
B
C
D
YS
392
413
468
514
370
402
436
479
N/ mm
1
UTS
466
498
551
588
456
490
529
576
/c
El
31.9
31.2
29.4
28.0
35.2
31.0
31.6
27.4
RA
80.6
80.6
78.7
76.8
80.6
80.6
78.8
76.9
11
YSyield st r engt h; UTSul ti mate t ensi l e st rengt h; El el ongat i on; RAreduct i on i n area.
E
z
c/)
600
550
SOO
450
400
350
1
A
^W y
- &
/
y
P-
y
y
y
y
1 1 1
B C J ^ UXS. -
y&
y^ s
^y s
y^s' y ' Y.S.
^y*'
y^ y
s y
. y
' y
y^ y
/ y '
s * * AS WELDED
O STRESS-RELIEVED
1 1 1
0 - 5 1 0 1-5 2 0
M A N G A N E S E I N W E L D , %.
Fig. 10 Effect of manganese on the tensile properties of multi-run deposits
W E L D I N G RESEARCH S U P P L E ME N T I 71- s
2 5 0
2 0 0
^150
a
a
UJ
5
100
a
UJ
DO
a
o
S2 5
CO
<
A
B
C
D-
I I I I
Char py
- ' ' / /
/ / /
1
i:
If \
/ / / /
/ ', /
/ ' / /
-yy
i i i i
i
- V
fy
i
i
.-
i
- 80 - 60 - 40 - 20 0 20
T E S T T E MP E R A T U R E , C .
Fig. 11Charpy V-notch impact results (as-welded)
3 0
2 0
E
I-
a
10
40
^
3 0
>
a
tr
UJ
UJ 2 0
Q
UJ
m
cr
8 10
CO
<
0
I
A
B
C
" D
i
I I I I I
Sc hnadt , K
0
y_
,yy
/.' j
/ ' n
/' /
/ 1
i i
i 1
y
/in
1 i
* ' '
/My
-y
i i i i i
-
_
- 1 5
10
E
a
- 5
-80 -60 -40 -20 0 20
T E S T T E MP E R A T U R E , C .
Fig. 13Schnadt impact test results K (as-welded)
1-2-
1 0
E
E
^ S
a
o
o
<
o
on
o
0-8
0 - 6 -
0-2
I I
A
_ B
c
D
| COD
-
-
..- y
. I I
I
clip
f
>'/ /
'/ y
I
gauge
I
l i mi t .
I
-
-
-
- 2 0 0
-150 -100
TEST TEMPERATURE,
Fig. 12-COD test results (as-welded)
di t i ons. Yi el d st r engt h and ul t i mat e
t ensi l e st r engt h are pl ot t ed i n Fi g. 10
and are seen t o i ncr ease l i near l y wi t h
i ncr easi ng manganese.
For t he as- wel ded c o n d i t i o n , t he
resul t s ( i n N/ mm
2
) are descr i bed as
f ol l ows wher e YS is yi el d st r engt h and
UTS is ul t i mat e t ensi l e st r engt h:
YS =314 + 108 Mn (1)
UTS =394 + 108 Mn (2).
For t he st r ess- r el i eved c o n d i t i o n , t he
equi val ent equat i ons wer e cal cul at ed
t o be:
3
C.
5 0
- 80 - 60 - 40 - 20 0
T E S T T E MP E R A T U R E
20
Fig. 14Schnadt impact test results
C
(as-welded)
YS = 3 1 1 + 89 Mn (3)
UTS = 3 9 0 + 98 Mn (4).
For t he speci f i c we l d i n g c ondi t i ons
e mp l o y e d , it was f o u n d t hat an
i ncr ease of 0. 1% manganese i n t he
deposi t i ncr eased t he t ensi l e par ame-
ters by appr ox i mat el y 10 N/ mm- .
Stress r el i evi ng of t he syst em ( C- Mn )
i nduc ed t he t ensi l e par amet er s t o
decr ease, t he d r o p bei ng dependent
on t he manganese l evel .
Toughness Results. The Char py V
t r ansi t i on cur ves f or as- wel ded depos -
i ts are gi ven i n Fi g. 11. The C O D test
resul t s obt ai ned f or saw not c hed spec-
i mens are pl ot t ed in Fi g. 12 and t he
Schnadt t est resul t s are gi ven i n Figs.
13 and 14 f or t he K and B c ondi t i ons ,
r espect i vel y.
The dat a are r epl ot t ed consecut i vel y
i n Figs. 15 t o 18, as a f u n c t i o n of we l d
met al manganese and it is seen t hat
t he f our di f f er ent test pr ocedur es
exhi bi t ed t he same gener al t r ends.
I ncr easi ng manganese l ower ed t he
upper shel f and di spl aced t he t r ansi -
t i on cur ves t o l ower t emper at ur es unt i l
an o p t i mu m c o n d i t i o n had been
at t ai ned at a manganese c ont ent of
appr oxi mat el y 1.5%. Ther eaf t er , i n-
cr easi ng manganese bec ame del et er i -
7 2 - s l M A R C H 1980
2 5 0
2 0 0
^ 1 5 0 -
O
a c
UJ
5
1 0 0
a
UJ
DQ
CC
o
W 5 0
<
Cha r py- V
0-5 1 0 1-5 2 0
MANGANESE I N WEL D, %.
Fig. 15Effect of manganese (Charpy V-notch)
0-5 1 0 1-5
MANGANESE I N WELD
%
Fig. 17Effect of manganese (Schnadt K)
ous, except at very l ow temperatures
where the l ower shelf was raised.
The Charpy V-notch impact curves
for stress-relieved deposits are pl otted
in Fig. 19 and the data are reconsidered
in Fig. 20, as a f unct i on of manganese.
Comparison wi t h the as-wel ded con-
di t i on (Fig. 11) indicates onl y a slight
di spl acement, the heat treatment hav-
ing had a beneficial effect at l ow man-
ganese and a detri mental effect at high
manganese contents. The extent of the
temperature di spl acement, at the 100 J
level, is given in Table 6.
The Charpy V curves obtai ned on
testing strain aged impact specimens
are shown in Fig. 21 and the equi val ent
results are pl ot t ed against manganese
3 0
2 0
E
I-
a
10
0-5 10 T5
MANGANESE IN WELD %
2 0
Fig. 16Effect oi manganese (COD test, 20 x 26 mm, i.e.,
0.79 x 1.02 in.)
4 0 -
- 3 0 ( -
>
CD
or
LU
UJ 2 0
a
UJ
CD
ce
O
CD
<
1
A
-
1
B
1
1
C
Schnadt
i
Bo
1
D
^-40C
~~-50
" " ^ ^ - 6 0
" " ~ ^ - 7 0
. - 80
i
- 15
10
E
I
a
1 0 -
0 5 1 0 1-5 2 0
MANGANESE IN WE L D , %.
Fig. 18Effect of manganese (Schnadt B)
content in Fig. 22. Agi ng displaced the
curves to higher temperatures, the
shift at the 100 J level bei ng reported in
Table 7.
The lateral shift to higher tempera-
tures di ffered accordi ng to manganese
content, attai ni ng a maxi mum (C) and
then decreasing at the highest concen-
trati on investigated. The overall effect
was for el ectrode D t o become the
best of the series and for the opt i mum
to be displaced relative to that exhi b-
ited for as-wel ded and stress-relieved
deposits.
Discussion
The metallographic studies of the
four different manganese-containing
deposits revealed marked differences
in mi crostructure. In as-deposited
wel d metal , as exempl i fi ed by the t op
central bead, increasing amounts of
manganese progressively increased the
amount of acicular ferri te, at the
expense of pro-eutectoi d ferrite and of
intermediate lamellar component . Fur-
thermore, the acicular ferri te, per se,
became progressively more refined
(Table 3). The reheated regions were
WEL DI NG RESEARCH SUPPLEMENT I 73-s
250
2 0 0
^1501-
O
LU
5
1 C
Q
UJ
CD
CC
O
en 50
CO
<
A -
B -
c -
D "
-
I i i
Char py- V
1
1''/
/ /-
/ /
ft
/ / /
/ , ' /
/ / / /
/ , /
/ / /
V7 /
/ /
/ y
i
STRESS-
1 1 1
1
RELIEVED
,
250
- 3 0
2 0
E
Q.
10
- 80 - 60 - 40 - 20 0 20
TEST TEMPERATURE, C.
Fig. 19Charpy V-notch impact results (stress relieved)
Charpy-V
D
RELIEVED
3 0
2 0
E
1-
a
10
1 0 1 5 2 0
MANGANESE IN WE L D, %.
Fig. 20Effect of manganese (Ch V, stress relieved)
similarly affected, the coarse grained
and the fi ne grained zones also
becomi ng increasingly finer. The over-
all i nfl uence of manganese on mi cro-
structure thus appeared to be benefi -
cial t hroughout , the measured param-
eters changing monot oni cal l y.
The specific wel di ng condi ti ons em-
pl oyed were such as to i nduce the
larger port i on of the central part of the
ISO 2560 deposits to recrystallize. For
example, at the Charpy V-notch loca-
t i on, it was f ound, on the average, that
onl y 20% of the structure remained in
the col umnar f orm. The amount of
recrystallization is considered to be an
i mportant factor i nfl uenci ng mechan-
ical properties
751
and must therefore be
borne in mi nd when attempti ng to
evaluate the i nfl uence of al l oyi ng ele-
ments.
Tensile tests results confi rm that
manganese increases the yi el d
strength and tensile strength of i ron-
manganese alloys.
11
For the range of
manganese contents investigated, sol-
id sol uti on hardeni ng and grain refine-
ment led to a linear i nfl uence, an
increase of 0.1% Mn increasing the
tensile parameters by 10 N/ mm
2
The
latter value compares favorably, but
perhaps i nadvertentl y, to that quot ed
by Brain and Smith
1
" for mi l d steel CO;,
wel d metal and by Tul i ani
6
for sub-
merged arc wel d metal.
The tensile properties decreased
after stress rel i evi ng, t he drop bei ng
greater in the case of yi el d strength
and high manganese levels. Carbide
preci pi tati on occurred at grain bound-
aries duri ng the heat treatment, but
evi dentl y no secondary hardening oc-
curred in the plain C-Mn wel d metal
svstem over the ti me i nvol ved.
Table 6 -
Electroc
A
B
C
D
Ef f ect of St
Temp.
e AW
- 2 7
- 44
- 53
- 43
ress Relief
C at 100 |
SR
- 32
- 44
- 50
- 3 6
(at 100 J)""
Di spl ace-
ment,
C" "
- 5
0
+ 3
+ 7
'"AWas-welded; SRstress-relieved.
""F = (9/5)C + 32
The toughness data obtai ned using
the Charpy V-not ch, Schnadt (K
(
, and
B) and COD test revealed the same
general t rend in all cases. Thus, it can
be concl uded that the universal Char-
py V test can be confi dentl y appl i ed
for routi ne classification of electrodes
accordi ng to ISO 2560. In practical
appl i cati ons, however, when consider-
ing properties in the ful l thickness of
the j oi nt , t he COD test is a requisite
for eval uati ng fitness for purpose and
determi ni ng critical defect sizes. A
possible correl ati on existing between
Charpy V-not ch and COD perhaps
onl y applies when the extent of strain
aging is l ow. Furthermore, a rel ati on-
ship coul d possibly be onl y expected
when the electrodes compared are, as
in the present instance, of the same
slag-base type.
In accord wi t h Nakayama et al." it
was f ound that increasing manganese
decreased the upper shelf of the tran-
sition curves. The increasing yield
strength resulted in a greater tendency
to mi crovoi d coalescence duri ng the
ducti l e mode of fracture. In contrast,
manganese had a beneficial i nfl uence
on the l ower shelf due to the effect on
cleavage resistance.
9
In the transition region of the
impact curves for as-wel ded deposits,
manganese had an opt i mum i nfl uence
at 1.5% manganese, despite the pro-
gressive i mprovement in mi crostruc-
ture t hroughout . The pattern of behav-
ior is thus dependent on the compet -
ing actions of the el ement to:
1. Increase the yi el d strength.
2. Increase the acicular ferrite vol -
ume fracti on and to refine the grain
size in the reheated region.
Stress rel i evi ng of the deposits had
vi rtual l y no i nfl uence on the Charpy
V-notch test results, and peak proper-
ties were also exhi bi ted at the 1.5% Mn
level. Furthermore, it appears that the
decrease in toughness expected as a
result of carbide preci pi tati on was
compensated for by an opposi ng
mechani sm, e.g., a softeni ng of the
ferrite.
Strain aging of the four experi mental
wel d metals i nduced a considerable
degree of embri t t l ement . In C-Mn
deposits it is generally accepted that
the major solute causing the decrease
in resistance to cleavage fracture is
ni trogen. Wi t hi n the range of scatter,
no trend in ni trogen content existed
over the range of manganese contents
studi ed, the values bei ng between 69
and 96 ppm. Manganese is reported
1
'
2
to di mi ni sh the aging tendency of
steel, and opi ni on varies as to whether
the element shoul d be j udged on its
own merits or whether the combi ned
effect of manganese and carbon is of
greater significance. The relative dis-
placement on strain aging (Table 7)
was inconsistent and the reason
remains eni gmati c, other than that
grain refinement alone becomes the
cont rol l i ng factor. The observed trend
74-sl MARCH 1980
i !OU
200

1
^150
a
UJ
z
m
100
Q
UJ
CD
QC
O
v> 50
CO
<
0
I
A
B
c -
_ D
-
STRAIN
I
I
~
AGED
I
I I I I
Charpy-V
/
yS7
/ / / /
/ / /
i'
f
/
nil -
/ ii/
' y
i i i i
- 3 0
- 80 - 60 - 40 - 20 0 20
TEST TEMPERATURE ,C .
Fig. 21Charpy V-notch impact results (strain aged)
- 2 0
E
f
a
j *
- 1 0
2 5 0
200
Charpy-V
30
20
E
I-
Q.
10
10 T5
MANGANESE IN WELD ,5
Fig. 22Effect of manganese (Ch V, strain aged).
was such t hat deposi t D ex hi bi t ed t he
best i mpact pr oper t i es, t he o p t i mu m
bei ng di spl aced away f r o m t he pr e-
vi ous l evel of 1.5% Mn .
The pr esent wo r k is consi der ed as an
i ni t i al st ep f or t he ul t i mat e under -
st andi ng of t he r ol e of mi cr ost r uct ur e
i n mul t i - r un manual met al arc depos-
its. Event ual l y, it is i nt ended t o add
al l oyi ng el ement s, e.g., Mo , Ni and Cr,
t o t he f our di f f er ent manganese l evel s
and eval uat e t he changes i n st r uct ur e
and pr oper t i es. Fi rst l y, howev er , it is
f el t t hat f ur t her wo r k s houl d be c o n -
duc t ed on t he C- Mn syst em so as t o
appr eci at e t he part pl ayed by car bi de
di s t r i but i on and mo r p h o l o g y . To f aci l i -
t at e t hi s, it is i nt ended t o st udy t he
f our we l d me n t s i n t he nor mal i z ed and
n o r ma l i z e d - a n d - t e mp e r e d c ondi t i ons .
Conc l us i ons
For I SO 2560 we l d me n t s deposi t ed
at 1 k j / mm wi t h basi c el ect r odes of a
speci f i c sl ag-base t ype, t he f o l l o wi n g
c onc l us i ons appl i ed:
1. I ncr easi ng manganese, i n t he
r ange 0.6 t o 1.8%, i ncr eased t he
a mo u n t of aci cul ar f er r i t e i n as- depos-
i t ed we l d met al and decr eased t he
a mo u n t of pr o- eut ec t oi d f er r i t e and
i nt er medi at e c o mp o n e n t .
2. I ncr easi ng manganese r ef i ned t he
aci cul ar f er r i t e i n t he as- deposi t ed
we l d met al .
3. I ncr easi ng manganese r ef i ned t he
coarse gr ai ned r egi on of t he heat -
af f ect ed we l d met al .
4. I ncr easi ng manganese r educed
t he gr ai n si ze of t he equi axed f i ne
gr ai ned z one of t he heat - af f ect ed
we l d met al .
Table 7-Ef f ect of Strain
Electrode
A
B
C
D
Temp.
AW
- 2 7
- 4 4
- 53
- 43
Agi ng
C at 100 J
SR
+ 5
- 5
- 12
- 19
(at 100 ))<*'
Di spl ace-
ment
C><"
+ 32
+ 39
+ 41
+ 24
181
AWas-welded; SRstress-relieved.
""F = (9/5)C + 32
5. The yi el d and t ensi l e st r engt hs of
t he deposi t s i ncr eased by appr ox i -
mat el y 10 N / mm
2
per 0. 1% i ncr ease of
manganese.
6. Char py V- n o t c h , Schnadt (K,, and
B
0
) and C O D tests gr aded t he test
wel ds i n t he same r el at i ve or der .
7. The o p t i mu m i mpact pr oper t i es
of as- wel ded and st r ess- r el i eved de-
posi t s wer e at t ai ned at 1.5% Mn , d u e
t o t he c o mp e t i t i v e i nf l uence of yi el d
st r engt h and mi cr ost r uct ur e.
8. St rai n agi ng embr i t t l ed t he de-
posi t s and c hanged t he r el at i ve or der
such t hat o p t i mu m i mpac t pr oper t i es
wer e achi eved at a hi gher manganese
c ont ent .
Ac/cnow/ec/gments
The aut hor wi shes t o express hi s
t hanks t o t he staff of t he We l d i n g
I nst i t ut e f or c o n d u c t i n g par t of t he
met al l ogr aphi c wor k under cont r act .
In par t i cul ar , t hanks are ext ended t o
Dr. R. E. Do l b y f or super vi si ng t he
st udi es.
References
1. Davey, T. G., and Wi dgery, D. )., " A
Techni que for the Characterisation of Wel d
Metal Mi crost ruct ures, " IIW Doc. II-A-389-
76.
2. " Det er mi nat i on of Hydrogen in Wel d
Met al , " ISO 3690.
3. Schnadt, H. M., and Li enhard, E. W
"Experi ment al Investi gati on of the Sharp
Not ch Behaviour of 60 Steels at Di fferent
Temperatures and Strain Rates," IIW Doc.
IX-348-63.
4. "Bri ti sh Standard Draft for Devel op-
ment on Met hods f or COD Test i ng, " DD
19:1972.
5. Garl and, |. G., and Ki r kwood, P. R.,
"The Not ch Toughness of Submerged Arc
Wel d Metal in Mi cr o- Al l oyed Structural
Steels," IIW Doc. IX-892-74.
6. Tul i ani , S. S., "The Role of Manganese
in Mi l d Steel Submerged Arc Wel d Met al , "
Internat. Conf. on Wel di ng Research,
CEGB, Mar chwood, Engl and, Sept. 1972.
7. Baach, H., " Dependence of Mechan-
ical Properties on the Number of Layers
duri ng Submerged Arc Wel di ng, OERLI-
KON Schwei ssmi ttei l ungen No. 54 (1965),
pp. 16-24.
8. Stout, R. D., McLaughi n, P. F and
Strunk, S. S., "Heat Treatment Effects of
Multi-Pass Wel ds, " Welding Journal, 48(4),
Apri l 1969, Research SuppE, pp. 155-s to
160-s.
9. Rees, W. P., Hopki ns, B. E and Ti pl er,
H. R., "Tensi l e and Impact Properties of Iron
and Some Iron Al l oys of High Puri ty," J.I.S.L,
169 (1951), pp. 157-168.
10. Brain, A. G., and Smi t h, A. A.,
Mechani cal Properties of CO., Wel d Met al ,
Brit. Weld. /., 9 (1962), pp. 669-677.
11. Nakayama, O., Nakana, H., and Tashi-
ko, H., "The Effects of Al l oyi ng E lements on
Stress Corrosi on and Mechani cal Properties
of Wel d Metal s in Arc Wel di ng wi t h Cov-
ered El ectrodes," The Welding World, 20
(1968), p. 509.
12. Szczepanski, M The Brittleness of
Steel, L Wi l ey and Sons, New York, 1963,
p. 104.
WELDI NG RESEARCH SUPPLEMENT I 75-s

You might also like