McMaster University

DigitalCommons@McMaster
Open Access Dissertations and Theses Open Dissertations and Theses
9-1-1979
Cold-Cracking Control in Low-Alloy Steel Welds
Vivek Pavaskar
This Thesis is brought to you for free and open access by the Open Dissertations and Theses at DigitalCommons@McMaster. It has been accepted for
inclusion in Open Access Dissertations and Theses by an authorized administrator of DigitalCommons@McMaster. For more information, please
contact scom@mcmaster.ca.
Recommended Citation
Pavaskar, Vivek, "Cold-Cracking Control in Low-Alloy Steel Welds" (1979). Open Access Dissertations and Theses. Paper 2753.
http://digitalcommons.mcmaster.ca/opendissertations/2753
McMaster
Handlton; Ontario

r ..
"
S.eptembcr 1979
, .
(
.-

-,
'.
-'
' ..
,,'
"
.'
0"
.-
'- ""
.-
....
,
...
..
/
-. . ..
,
'.
"
,
.\

.. ....
."-
"\ .
~
'.
_,L
r"
I
I
)
a
. .
. ,
#
..
c
IN
'.
"
LOW-ALLOY S'1'E!:L WELDS
-
. .
" .
'.
'.>

, .
..
, ,

, .
" }.
'.
,
,"
. ' .
- .
MSUR Of' DCD+Rnax: (197'9)
~
(MetaUurcJY aDd ICateria1.a SC1eftCe)
. -
AT11'HOR:
SUPERVISOR:
Vivek Pavaskar. B. '1'ech. ( I ~ InSt!tute of
'recbnology. BcGlbay)
Dr. J _ S. JCirkaldy
-.
-,
.'
,
4 \.
"
. t
, ~

ii
....
"
0
.
tIIf
.
(
~
\
/'
/

.:
,.,.
-',
,
...
" .
, '.
....
..
..
..
,.
..
-\
, 's. . I
BAZ microst:ructure" anc1'1-lyckogen level
, \ -
t&J. faet="<; 5!"'. cold s""ceptihlUty RAZ.
Implant v ious hYdroqen leve'1.s over A range of steel. CCIIIpO-
I, , ':. . ...
and he 1n s ..shows how the
,
level influence the cx::itic4l -stress for cold-cracking.
,
_._.--_.. -
, ,
. 'Based on implant test 4 cOrrelation fOrlUula prC'dieting. the
. ,
, stress necessary for cold cracking for given HAZ ,
transfet ,
'"
marten'site in the HAZ -4nd hydrogen -revel, is proposed. Elnpioying this
. ......
. ,
prediction'of 'in the
. .
correlation. together

'HAZ hardness. on
..
.\
tion-cooling relations anc
"
tions, an which can p!cdict the stress necessary for

cracking given implant cooling rate and level,
.p
is constructed.

....
niis, method of predicting the-critical stress'necessary for cold
cracki.ng is an over tht: ex.isting regression formulas fo'l;

. ,-
estindting cold-,racking susceptibility.
, . .
This formula has been to predict cold-crack-
ing susceptibility data as obtained such as rigid
and to to cold-crack-
ing. Development and use of <l:"l inplant machine ""ith-an automatic
and loading facility reported.
I'
n
f
.
iii
)

.\' , .
,..
" .
1.
;-.-'
., "
f
'.
:
'\...
. '
"
- '
- -
-,
.-
-
.
.........
"
...
. ...,
..
...... . ......
'1'be.. abthoc.1a indebted his aaperviaor, Dr. J. S. J:1%'ka1dy,
,
'.
. .
.
I
for de&lt 'for hia cont.iJluoua
- ... .. 1#
throughout the course of this 'nlanks. are a1.so'due to
"many\ of gradUoite students in the De-partment of Meta11urgy
for their advice and assistance. X wou1d al.so
like to thank the technical staff of the Department of Metallurw and
the machine shop for their assjstan<:e in de-
o "
veloping the Equipment, Thanks are also due to Mrs.
the excellent typing 'of this thesis.
/'"
, , .
suppOrt of the Natural Sciences and Engineering
Research Couneil of Canada is gratefully acknowledged .
./
, .
,
,
-.._-
..
. -.

r
i
1
1
"
iv
'.
G ,/
'.
-.
. ;
PAGE"
OIAP1'ER 1 .IN'l'RODOCTION 1

CHAP'1"ER 2 ASSESSDlC THE OF THE HEAT ,
. .
AFP'EX:"tEO ZONE (HAZ) IN S'l"EEL
'-.
2.1 Weldability Formulas for !,-ssessi{\g
Susceptibility
/
4
2.2 Implant Testin; for
: Susceptibi lit.y
. 15
J
CHAPTER 3 INFLUENCE OF HAZ .;.:::> HYOROGEtl LEVEL ON
, ,
3.1

Test Data
..
19
3.2 Influence of , Harcness on a
CR
20
3. 3 (' f i ed' Corre ticnt"Forr..u Ill, . : or a
CR
ROle in o.
. .
23
25
3.4.1 Concentration and

, .
. '
Str*:;ss I::tensity in tl'\t: of Haximum
i 36
3.4.2 , . 38
CHAPTER 0 . BAS!S 0; tNPG1, PLATE
CR .

Cooling Rate
4.2.1 He<lt-.Tran:ifer wit'h 30"r.",=,H. Fl'Ow
42
46
o

4,_2..1.1 Applic<ltio:'\ 0: 3:) HC<1t F.lm"
\..
..
4.2
r
I
"/
..

\
v
I
J_
-.
.,

.
..
.. '
.0
51
PACZ,';
-
. ',"
... ...... .' .... : .
.'-'-,' ,. -'-
'.;"
2D, Heat Flow
2.SD Heat Flow .
4.2.2 Heat 'transfer With
Heat
( .
4.2.4 Representation of Heat Flow
.... 0
!
,
I
. /
! "
-, .
'.'\ J

,
I
;

/ /
Conditions as a Function of Plate 'l'hiclcness 52
."
4.2.5 Procedure for Preaicting Cooling Rate 53
4.2'.6 Discussion
4.3 Prediction of Heat Affected Zone Microstructure
55
4.3,i , Klcroconstituent -.Composition - COoling
,
Rate 57
4.3.2-Hardness - Composition -'Cooling Rate
Reiations
t
59
;
(--4.3.3 Prediction of HAZ , Martensite and Hardness
, w
59
4.3.4 Discussion 63
CHAPTER 5 DETERMINATION OF PRACTICAL WELDING PROCEDURES ON THE
BASIS OF IMPLANT TEST 82
5.1 General Remarks 82
5.2 Between Implant Test and Otht!T kstraint
Tests 82
'5.3 Determination of Prehea.t to Avoid Cracking 84
5.3.1 Estimation of Reuction Stress Developed in
the Welc 87
.'
5.3.2 of on Resicual Hydrogen
Level 88
5.'.3 Prediction of Prehedt Level to Crucking 91
vi
t
7 CONCLUSIONS
\
!!e!..
92
110
110
110
111
113
113
115
115
120
126
126
-
133
..
.:... ..
, .
';'.:.'
- -'..

6.2.3 The Pressure Systeo for the Implant Test
6.2.1 AutO!Datic Deposition of Electrodes
6.2.2 '!'he Il:lpl4nt Loadinc;' tem
6.3.2 Implant 7est'Details.
6.3.1 Welding
6.4 Prelic1nary ?ests
6.5 DisCI..-ssion
6.3 Standardization of the- Iop1a."1': Testing Procedure
6.1 General Remarks
of th"e Testinc; Machine
, "
._. ':--
.... -: ..

.... ,..
,-,
',.' ,"
.:. .".'
.'
..,p.. :,.
;;"' .. ':"
APPEt'DIX I 1-3
APPENDIX II 137
III 139


144
o
vii
-. l
-
.
LIST 'OPTJU3LES
TABLE
i
PACE
,
Steel weldments
1
2
3
5
6
7
8
9
10
Formulas to Assess Cold-craddng of HAZ in
, ,

Precautions for Various carbon

Equivalents
Index Proposed by Bradstreet
Re1.a.t.ion Between carbon Equivalent and weldability .
.
Low Hydrogen Electrodes
Procedure,for Prevention",of Cold-Cracking
Implant Test' Datd \
Comparison Between Our's and Christensen's Correlation
Influence of Hydrogen Lever on,oCR
Calculated Stress Intensity and HeR ,at
Comparison Between Reported dnd Calculated Cooling
.
I
6-7'::'
18
10
10
.11
i
24
26

" '31
39
J
Times' Between 800C - 500C .. 56
11 Prediction of OCR Basis of Heat Input,
Thickness, Plate and Hydrogen
"
12 Prediction of HAZ and OCR for the Ito-Bcssyo
Data 73-81
13-16 Prchaat Levels Predicted for Various-Steel Compositions,
Plate Thickness, Heat Hydrogen Level With

17
Fillet Weld Configuration
Welding Hcat Input Details
viii
102-109
116
-
. --
"
..
18
.19
t
)
- . .
(bserved
. tiJDeS - _sooC
t
0Mtmic:al CoD:posi.tiCXlS of ...the Steels used for
_ l)apl.ant: Testing
..!..
PACE
--
..
123
.129
...
. ,
. ...
. -
I
,
J
.., .
ix
..
'.
,
..
'.
,
"
:
"
..
PACE
1
2
3
5
6
7
8
9
10
11
12

13
14
Bai.ley' s (14) Nomog%4JD of safe Welding Procedures
for C-Mn Steels
Ito-Bessyo (1'6) System for Determining' safe Weldi.ng
Procedures
!m;)lanting Steel . Specimen by Welding'
SChematic Diagram Showing Implant Te9t
Typical Stress VS Fracture Time Curve fo'r an
"
Implant Test
.
. (23)
of 0ck With HAZ Hardness
, Weld Cracking and HAZ Hardness Plot
C1S
)
'COrrelation Between OCR and Both , Martensite and
Hardness as Parameters
Correlation Between OCR and HAZ INDEX
COmparison Between COrrelated and Observed With
, ,as
Comparison Between Correlated and Observed With
CR
, Martensite and Hardness, as Parameters
Influence of Hydrogen Level on OCR
Variation of Hydrogen Index Wi th HydrC?3en Level
,Stress Intensity and Hydrogen Concentration Combina- ...
12
14
16
16
17
21
22
27
29
30
32
33
'\
tions for Fracture
(4)
as Suggested, Beachem
...
15 Influence of Notch Radius on the Location of the
Hy,drogen Assisted
C35)
Crack1ng ,-
..
x
....
35
37

-
..
.' ... .. :.
;PIGDRE
.
".
,'lr. -
:1
...). .
.'
."
,.
.-
",
PAGE
-
..
.'
,
, .
..
.'
...
..
18
19
20
21
22
23-25
CUWbiua' of St:ress."Intensity and
".' .-
.CoDceDt:raj:ion critical at (Sea
'Block PridictiDg (Sea on the of ..
. - ........
."
Wel.di:D9 Beat anput, Plate Thickness," Plate C'omposi-
.....
tieD.. and Hydrogen Level

Beat Cone.it:i.ons in Welding

vari&t5:-on. .with
:- . .
of B
3D
with
of Coo1i:lg Rate 'with Plate ickness ,
Tr
f .' d Cool' . Cha (42).
ans ODDa:'t:ion an Rate rt
J ":-
" Martensite and Cooling Rate for Three c-Mn
SteelS'
40
49
54
58
61-62
..
. ,
26 . Cqmpa.rison Between predicted and Observed " Martensite
.' (for.Christense:\"s :,)ata(22,23'
.'.
27 Between Predicted A."\d ObserVed HAZ
(fOr Christensen's Data(22,23
..
'64
65
....
28
29
30
". "
31
Cpmparison .Beo.:ee,n 'Predicted A."\d Observed RAZ Hardness
(f
. (lS} .
or . Data)
...
Between'Predicted and Observed OCR (for
. Data)
, Plot of " weld Cracki:\g Predicted 0CR'!for
(15) Data)
. (15t
Plot of " Weld Ito-Bessyo Cracking
..
Parameter P
c
66
68
69
'70
, .
.j
.' .
xi
A
..
. .

.' -
..
,- .,
..
" -
a..
.

.'-
.-
. .-

"
'.
P:tGORE
32 CoIIIpari.saD Between Lower Critical. .
.
Obtained frcm Implant '!'est and from RIC and TBr:::
'.
Tests
'.
83
-
....
.33
'34
350
Critericn or A'V'Oiding Cold-Cracks in Welded
Structures
Procedur; fOr Rec:ommend.ing Preheat.to A'V'Oid Cracking
.
Intensity of Restraint vs Plate Thickness for Welded
85
86
- (43)
Construction
1.
-.
89

36
37-44
45
46
47
Bead .Model for .. Diffusion Analysis
Be.bleen Predicted Preheats and ReCOlllJDended
(B.S. Standard) Preheats
System for AutOmatic Deposi ti6n of MSMAW Electrodes
Implant l.oading System
.-
System for
89
94-107
112
114
114
.. <
48 General View of the Implant Complete
Wi th tic ana Loading Facility 117
I
49
so
51.
S2
53

- .
Details of the weld;ing Arrangement
Detailed View of the Loading System
Weld Bead Sections for Three Heat Inputs
Typical Thermal Cycle Profile in Welding
Implant Specimen Details
.Base Used for Implant Test
118
119
121
122
124
125

55 Section of Implant Specimen Showing the Location

of the Notch 127

56
.
. Typical Variation of Hardness Across the
xii'

128
58
57
FIGtIBE
Diagam for St2e1 No.:
50.
.
.
'.

0
'.
.
.' ....
PAGE
1
-130
.
2
.,-
131
'0
..
"-.-
110': .....
o
.'
'.
.,
.
.'
,.
--
..
..

0
0
-

. .
xiii
,
CHAPTER 1
....
Hydrogen induced cold-crac:king has one of the .major problems'
of this' type can take many fonlS _ although
they have some qene.r'a1 characteristics and are influenced by CO""IDOn
basic factors.
, As the name implies, these cracks form at low temperatures
generally below and often exhibit a delay phenomenon. Even after
the weld has cooled to room temperature, there may be a further lapse of
time ranging from a' few minutes to severa1 hours.
not tolerated in a.structure, and since they are
'0.
..
often' difficult to detect and expensive to repair, it becomes essential
the fabricator to take precautions during welding to prevent their
formation,
J
Even though the general causes of hydrogen cracking and the means
of preventing these cracks are known" recent industrial surveys have
shown that cold-cracking is widely encountered, indicating that
the ability to satisfactorily predict cold-cracking susceptibility has.
not been developed.
Hence, development of a t'el;'able, quick and economic method for
predicting hydrogen craclcing in a real welded.. joint is desirable to 00-
termine when conventional welding techniques can be used wi thout the
likelihood of failure from hydrogen cracking. In addition, needless
expense of preheating and low-hydrogen techniques cou1d be avoided; re-
1