Supplementary Technical Document for Senior Welding Engineer Certification

I. Determination of Necessary Preheat Temperature

July 2009

THE JAPAN WELDING ENGINEERING SOCIETY

1. Welding heat input

Heat supplied per unit weld length by a heat source is given as:
EI ( J / cm)  60  ( I  E / v) (1)
or
EI ( kJ / mm )  0.006( I  E / v ) (2)
where, I is welding current (A), E is welding voltage (V), and v is welding speed
(cm/min). EI is called “energy input” in AWS (American Welding Society) and is same as
AE (arc energy) used in EN (European Standard).
The energy supplied into a weld is HI (heat input):
HI (kJ / mm)    EI (3)
where,  is heat efficiency of a heat source.
 for SAW  1.0
 for SMAW  0.8
 for GMAW  0.8
 for TIG  0.6

2. HAZ hardness

The maximum HAZ hardness, Hmax governs the occurrence of cold cracking and
sulfide stress corrosion cracking. So, the limitation of Hmax is often specified in welding
fabrication in such that Hmax should be less than 350Hv for avoiding cold cracking, or
be less than 248Hv for avoiding sulfide stress corrosion cracking.
Hmax is determined by the welding cooling rate and chemical composition of a steel.
The cooling rate in welding is generally represented by the cooling rate at 540oC,
R540(oC/s) or the cooling time between 800 and 500oC, t8/5(s). This is because the phase
transformation on cooling in a mild steel starts at about 800oC and finishes at about
500oC, There is the following relationship between R540(oC/s) and t8/5(s) :

R540 ( o C / s )  300 o C / t 8 / 5 ( s ) (4)

t8/5(s) changes depending upon the welding heat input, plate thickness, preheat and
inter-pass temperature and ambient temperature. In order to obtain t8/5(s),
nomographs1) and charts2) are prepared. However, this can be online-calculated very
easily at the website of the Japan Welding Engineering Society. Its URL is:
http://www-it.jwes.or.jp/weld_simulator/index.jsp
Fig.1 shows how Hmax changes as t8/5 changes. As t8/5 decreases (the cooling rate

2
increases), HAZ hardness increases and the HAZ microstructure becomes hardened
martensite. As t8/5 increases (the cooling rate decreases), HAZ hardness gradually
decreases and the volume of martensite in HAZ decreases. The HAZ microstructure
consists of 100% martensite at the cooling times shorter than the point A, while it does
of 0% martensite at the cooling times longer than the point B. The hardness of 100%
martensite is determined solely by the carbon content. Haradenability of a steel
represents how easily the martensite structure can be obtained in a heat-treated steel
or steel HAZ. As hardenability increases, the Hmax- t8/5 curve shifts to the right hand
side; i.e., the martensite microstructure can be obtained even at the longer t8/5(slower
cooling rate).

Fig.1 The effect of steel composition on Hmax- t8/5 relation

Based on the principle of the hardness behavior above-mentioned, the following

Hmax prediction formula is porposed3):
HM  HB HM  HB
H MAX   arctan( x) (5)
2 2.2
log( t 8 / 5 / t M )
x 4 2
log( t B / t M )

where, t8/5 is given by the welding conditions (heat input, plate thickness, preheat

3
temperature), and HM, HB, tM, tB are given by the steel chemical composition as follows:
HM : 100% martensite hardness:
H M  884C ( 1  0 .3C 2 )  294 (6)
tM: the critical (longest) cooling time when HAZ becomes 100% martensite (point A):
t M  exp( 10 .6 CE I  4 .8 ) (7)

Si Mn Cu Ni Mo Cr (1  0.16 Cr )
CE I  C       (8)
24 6 15 12 4 8
CEI is the carbon equivalent representing HAZ hardenability and its applicable range is
C  0.3% .
HB : 0% martensite hardness:
H B  145  130 tanh( 2 .65 CE II  0 .74 ) (9)

Si Mn Cu Ni Cr Mo V Nb (10)
CE II  C        
24 5 10 18 5 2 .5 5 3

tB : the critical (shortest) cooling time when HAZ becomes 0% martensite (point B):
t B  exp( 6 .2CE III  0 .74 ) (11)

Mn Cu Ni Cr Mo (12)
CE III  C     
3.6 20 9 5 4

The HAZ maximum hardness, Hmax given by the above equations can be
online-calculated very easily at the website of the Japan Welding Engineering Society.
Its URL is:
http://www-it.jwes.or.jp/weld_simulator/index.jsp

3. Carbon equivalent

Weldability of a steel represents, in a narrow sense, how HAZ hardens. HAZ

hardness is directly related with susceptibility to hydrogen induced cold cracking.
Therefore, good weldability means that HAZ is less hardened and thereby more
resistant to the occurrence of cold cracking. As shown in Fig.1, HAZ hardness is
determined by the carbon content and HAZ hardenability. Cold cracking becomes a
problem generally when the heat input is between 0.8 and 2.0kJ/mm (t8/5 is between 4
and 11s). In this condition, HAZ hardness significantly changes as hardenability
changes for a steel containing rather high content of carbon. However, for a steel with
lesser amount of carbon, HAZ hardness is less affected by hardenability. Rather, it is
more influenced by the carbon content itself.
Table 1 shows well known carbon equivalents The carbon equivalent of group A is
similar to Eq. 8. Therefore, this type of carbon equivalent represents HAZ hardenability

4
and is considered to be an index of weldability of a steel containing rather high content
of carbon. CEIIW in this group is widely used and carbon equivalent often means CEIIW.
Pcm of group C, in which the effect of carbon is significant, is thereby considered to be
an index of weldability of a steel with reduced carbon. CET of group B is in between
group A and B4).

Table 1 Carbon equivalents mainly used

Group Carbon equivalent formula Applicable range Specification

Ａ Mn Cu  Ni Cr  Mo V AWS D1.1
CE IIW  C   
6 15 5
C  0.08% BS 5135
Si Mn Ni Cr Mo V
CEWES C      
24 6 40 5 4 14 EN 1011-2-2001

旧 WES 3001

Ｂ Mn Cu Ni Cr Mo
CET  C     
10 20 40 20 10 0.08  C  0.12% EN 1011-2-2001

Ｃ WES 3009
Si Mn Cu Ni Cr Mo V
Pcm  C         5B
30 20 20 60 20 15 10 C  0.12% AWS D1.1

Ｄ ASTM A 1005/A-00
 Si Mn Cu Ni Cr  Mo  Nb  V 
CE N  C  f (C )       
 24 6 15 20 5  C  0.3% ASME

B16.49-2000

f ( C )  0 .7 5  0 .2 5  ta n h 2 0 ( C  0 .1 2 ) 

Table 2 Coefficient of CEN carbon equivalent

C(%) f(C) C(%) f(C) C(%) f(C) C(%) f(C)

0.02 0.51 0.08 0.58 0.14 0.85 0.20 0.98

0.03 0.51 0.09 0.62 0.15 0.88 0.21 0.99

0.04 0.52 0.10 0.66 0.16 0.92 0.22 0.99

0.05 0.53 0.11 0.70 0.17 0.94 0.23 以上 1.00

0.06 0.54 0.12 0.75 0.18 0.96

0.07 0.56 0.13 0.80 0.19 0.97

As shown by the change of HAZ hardness in Fig.1, HAZ hardness is determined

by an interactive effect of the carbon content and HAZ hardenability. Therefore, there
must be an applicable range depending on the carbon content for each carbon
equivalent in the groups of A, B and C in Table 1. CEN carbon equivalent of group D

5
considers the interactive effect of C and hardenability5). It approaches to CEIIW as C
increases, while it approaches to Pcm as C decreases. It is considered to be an index of
weldability of a wide range of steel.

4. Determination of Necessary minimum preheat temperature

1) Hardness control method of American Welding Society (AWS) D1.1, ANNEX I

( C  0.11% )
This method avoids cold cracking by reducing HAZ hardness less than the critical
level through reducing the cooling rate or increasing cooling time. As shown in Fig.1,
the reduction of HAZ hardness by increasing t8/5 is effective for a steel with a rather
high content of C. The AWS D1.1, ANNEX I specifies that this method is applicable for a
steel with carbon higher than 0.11%.

Fig. 2 Critical cooling rate to obtain critical HAZ hardness

Fig.2 shows the critical cooling rate, R540(oC/s) necessary to satisfy the critical
hardness of 350Hv and 400Hv for a given CEIIW. For fillet welding, the cooling rate is

6
governed by the web and flange thickness. AWS D1.1 prepared a number of figures
which show the relation of R540(oC/s) to the web and flange thickness, and the energy
input. Fig. 3 is one example for the 12mm web thickness. The energy input (EI) to avoid
cold cracking in this method is given, first by obtaining the critical R540(oC/s) against the
critical hardness (350Hv or 400Hv) in Fig. 2 and secondly by finding EI corresponding
to the web and flange thicknesses and R540(oC/s) in Fig.3.

Fig.3 Minimum energy input for critical cooling rate (flange thickness of 12mm)

EI of the vertical axis of Fig.3 is for SAW. In the case of welding processes other than
SAW, the minimum necessary energy input is given by multiplying the EI obtained for
SAW (Fig.3) by the following factor..
Welding process Multiplication factor
SAW 1.0
SMAW 1.50
GMAW, FCAW 1.25

7
2) Hydrogen control method of American Welding Society (AWS) D1.1, ANNEX I
( C  0.11% )
This method avoids cold cracking by reducing diffusible hydrogen for a steel whose
HAZ hardness changes little by decreasing R540 or increasing t8/5. The AWS D1.1,
ANNEX I specifies that this method is applicable for a carbon reduced steel ( C  0.11% ).
The susceptibility index, SI is given by:
SI  12  Pcm  log10 H GC (13)
where, Pcm is carbon equivalent of group C in Table 1 and HGC is weld metal diffusible
hydrogen content by a gas chromatography (JIS Z3118, ISO 3690). The necessary
minimum preheat temperature is given by Table 3 according to the SI value, degree of
restraint and plate thickness.
It should be noted that if this method is applied to a steel with higher content of
carbon, too conservative (high) preheat temperatures are given6).

Table 3 Necessary minimum preheat temperature by hydrogen control method

Susceptibility index, SI=12Pcm + log10 HGC

Restraint Thickness 3.0 3.1 – 3.6 – 4.1 – 4.6 – 5.1 – 5.6 –

Level (mm) 3.5 4.0 4.5 5.0 5.5 7.0

< 10 < 20 < 20 < 20 < 20 60 140 150

10 - 20 < 20 < 20 20 60 100 140 150

Low 20 - 38 < 20 < 20 20 80 110 140 150

38 - 75 20 20 40 95 120 140 150

> 75 20 20 40 95 120 140 150

< 10 < 20 < 20 < 20 < 20 70 140 150

10 - 20 < 20 < 20 20 80 115 145 160

Medium 20 - 38 20 20 75 110 140 150 160

38 - 75 20 80 110 130 150 150 160

> 75 95 120 140 150 160 160 160

< 10 < 20 < 20 20 40 110 150 160

10 - 20 < 20 20 65 105 140 160 160

High 20 - 38 20 85 115 140 150 160 160

38 - 75 115 130 150 150 160 160 160

> 75 115 130 150 150 160 160 160

3) Preheat specification in Table 3.2 in AWS D1.1

8
 AWS D1.1 alternatively gives the necessary minimum preheat temperature as
shown in Table 4.
Table 4 Necessary minimum preheat temperature
(Extracted from AWS D1.1 Table 3.2)

Steel Welding process Thickness,t(mm) Necessary

ASTM Corresponded Preheat (oC)
JIS
A36 SMAW with 3  t  19 None
A53 Gr.B electrode other than 19  t  38 66
A131 Gr.A,B SM400A,B,C low hydrogen type 38  t  63.5 107
A139 Gr.B 63.5  t 150
A381 Gr.Y35
Same as above SM400A,B,C SMAW with low 3  t  19 None
and TS500MPa SM490A,B,C hydrogen electrode 19  t  38 10
class of steel SM490YA,YB SAW 38  t  63.5 66
SM520B,C GMAW 63.5  t 107

The necessary minimum preheat temperatures for ASTM steels other than those
mentioned in Table 4 should be referred to AWS D1.1 Table 3.2.

4) BS5135-1984 (EN 1011-2 -2001 A) method

This method is similar to the AWS hardness control method in which cold cracking is
avoided by increasing t8/5 to reduce HAZ hardness. Therefore, CEIIW is used in this
method. Fig.4 shows a figure giving the necessary preheat temperature against the
thickness and arc energy for the combinations of carbon equivalent and weld metal
hydrogen content; a) CEIIW :0.38 and HGC higher than 15ml/100g; b) CEIIW :0.40 and
HGC between 10 and 15ml/100g; c) CEIIW :0.41 and HGC between 5 and 10ml/100g; d)
CEIIW :0.46 and HGC less than 5ml/100g. This method prepares 13 figures like Fig.5. The
necessary preheat temperature is determined by selecting the figure corresponding to
the given combination of CEIIW and HGC(JIS Z3118, ISO 3690).
The arc energy, which is same as the energy input, in Fig. 4 is for SMAW. For the
other processes with different heat efficiencies, the arc energy values calculated from
Eq. (2) should be divided by the following factors to give the values to be used in Fig. 4.
SAW : 0.8

9
 MAG (solid wire) : 1.0
MIG : 1.0
TIG : 1.2

The combined thickness, CT in Fig.4 is given as follows:

CT for butt joint = plate thickness x 2
CT for fillet joint = (flange thickness x 2 + web thickness)/2

Fig. 4 Necessary preheat temperature based on BS5135

5) Pc method (WES 3009)

The crack susceptibility index, Pc is given by Pcm, the weld metal diffusible

10
 hydrogen content, HGL (ml/100g) and plate thickness, d(mm) as follows:
H GL d
Pc  Pcm   (14)
60 600
where, Pcm is a carbon equivalent of C group shown in Table 1, HGL is the weld metal
diffusible hydrogen measured by the glycerin method ( JIS Z3113). The glycerin method
is poor in measuring accuracy and JIS Z3113 was abolished.
The minimum preheat temperature necessary for preventing root cracking in
y-groove restraint cold cracking testing with the energy input of 1.7kJ/mm is given as
follows:
Tph (oC) = 1440 Pc - 396 (15)
The y-groove restraint testing is conducted under a very sever condition of the high
restraint, short bead, sever root notch and single pass. Therefore, the preheat
temperature 75oC lower than that given by the y-groove restraint testing is generally
adopted in welding of a TS490MPa grade of high strength steel. For steels with rather
high carbon contents, this method gives too conservative (high) preheat temperature in
the same way as the AWS hydrogen control method dose6).

6) Method by CET4) (EN 1011-2-2001-B)

The minimum necessary preheat temperature is given by the following equation:
Tph( o C )  700CET  160  tanh(d / 35)  62 H GC0.35  (32  53CET ) HI  330 (16)
where, CET is a carbon equivalent of the group B in Table 1, d(mm) is the plate
thickness, HGC(ml/100g) is the weld metal hydrogen content (JIS Z3118, ISO 3690), and
HI(kJ/mm) is the heat input calculated by Eq.3.
This method gives appropriate preheat temperatures for steels with carbon
contents between 0.08 and 0.12% but does too conservative preheat temperatures for
both the rather high carbon steels and reduced carbon steels6).

7) Method by CEN3)
This method is based on the muster curves of the minimum preheat temperatures
for y-groove restraint testing as a function of the group D of carbon equivalent, CEN and
the plate thickness (mm) shown in Fig.5. These master curves are for the standard
condition of 5ml/100g of weld metal diffusible hydrogen content, HGC (JIS Z3118, ISO
3690) and of 1.7kJ/mm of energy input, EI. Fig.6 shows the values of CEN to be
corrected by the deviation of HGC from the standard,  CEN (HGC ), and Fig.7 shows the
values of CEN to be corrected by the deviation of EI from the standard,  CEN (EI ).
The energy input in Fig.7 is for SMAW (=0.8). For the other process with a different

11
heat efficiency, , the arc energy values calculated from Eq. (2) should be multiplied by
the following factors to give the values to be used in Fig. 7.

Welding process Multiplying factor

SAW (=1.0) 1.25
MAG (=0.8) 1.0
MIG (=0.8) 1.0
TIG (=0.6) 0.75

Fig. 5 Muster curves for necessary preheat by CEN method

12
 Fig. 6 Correction of CEN according to HGC deviation

Fig. 7 Correction of CEN according to EI deviation

CEN corrected by the deviation of HGC and EI from the standard is given as:
CEN(corrected)= CEN + ΔCEN (HGC) + ΔCEN (EI) (17)
where, CEN is the carbon equivalent value calculated from the steel composition. By
introducing CEN(corrected) into the horizontal axis of the master curves of Fig. 5, the
necessary preheat temperature for y-groove testing, Tph(y-restraint) is given.

13
 Fig. 8 Correction of necessary preheat according to weld metal yield strength

As mentioned above, Tph(y-restraint) is not necessarily adopted as the preheat

temperature in actual welding practices. Fig. 8 shows the values of the necessary
preheat temperature to be corrected in welding practices, Tph (oC) from Tph(y-restraint)
depending on the weld metal yield strength. Finally, the necessary minimum preheat
temperature in actual welding practices, Tph (oC) is given as follows:
Tph (oC) = Tph(y-restraint) + Tph (18)
The concept of the correction of the necessary preheat temperature of Fig. 8 is
based on the following welding practice experiences: 1) the necessary preheating
temperature can be reduced by 75oC from Tph(y-restraint) in TS490MPa or YS360MPa
steels; 2) cold cracking such as toe cracking, under-bead cracking, and weld metal
cracking is more likely to occur as the welding residual stresses increase, i.e., the weld
metal yield strength increases.
This method uses many figures and thus, a calculation error may arise. It is
recommended to use online calculation provided by the Japan Welding Engineering
Society whose URL is http://www-it.jwes.or.jp/weld_simulator/index.jsp.

8) Specification by the Japan Road Association

According to the Specification of Highway Bridges, the Japan Road Association, the
necessary preheat temperature is given as shown in Table 5:

14
9) Specification by the High Pressure Institute of Japan
According to the standard of the use of high strength steels, the High Pressure
Institute of Japan, the necessary minimum preheat temperature is given as shown in
Table 6:

10) Preheat temperature standard for heat resistant steels

The necessary preheat temperature ranges for heat resistant steels are given as
shown in Table 77):

Table 5 Standard condition for necessary preheat temperature

Preheat temperature (oC)

Steel Welding process Plate thickness (mm)

25 or less 40 or less 50 or less 100 or less

SMAW with electrode other None 50 - -

than low hydrogen type

SM400
SMAW with electrode of None None 50 50

low hydrogen type

SAW None None None None

GMAW

SMAW with electrode of None None 50 50

SMA400W low hydrogen type

SAW None None None None

GMAW

SM490 SMAW with electrode of None 50 50 80

SM490Y low hydrogen type

SAW None None 50 50

GMAW

SM520 SMAW with electrode of None 80 80 100

SM570 low hydrogen type

SAW None 50 50 80

GMAW

SMA490W SMAW with electrode of None 80 80 100

SMA570W low hydrogen type

15
 SAW None 50 50 80

GMAW

Table 6 Necessary minimum preheat temperature for high strength steels

Thickness Necessary minimum preheat (oC)

t(mm) TS590MPa class TS780MPa class
t  19 none 100
19  t  25 60 125
12  t  32 75
32  t  38 100 150
38  t  50 125
50  t  76 -

Table 7 Necessary preheat temperature range for heat resistant steels

Steel 0.5Mo 1Cr-0.5Mo 2.25Cr-1Mo 5Cr-0.5Mo Enh.2.25CR-1Mo

0.5Cr-0.5Mo 1.25Cr-0.5Mo 3Cr-1Mo 9Cr-1Mo 2.25Cr-1Mo-V
0.75Cr-0.5Mo 3Cr-1Mo-V
P-number 3-1, 3-2 4-1 5-1 5-2 5C-1
Preheat 80 - 200 120-300 150-350 200-350 200-350
temp., (oC)

References
1) M. Inagaki, report of National Institute for Metal Science, 3-1 (1960), p.24
2) N. Yurioka and S. Ohkita, “Welding of Iron and Steel”, Sanpo Publishing Inc. (1998)
3) N. Yurioka et al., “Prediction of HAZ hardness of transformable steels”, Metal
Construction, 19 (1984), No.4, p217R
4) D. Uweer, H. Hohne: IIW Doc. IX-1931-91, Determination of suitable minimum
preheating temperature for the cold-crack-free welding of steels, (1991)
5) N. Yurioka et. al., “Determination of necessary preheating temperature in steel
welding”, Welding Journal, 62 (1983), No.6, p147s
6) N. Yruioka, “Comparison of preheat predictive methods”, Welding in the World, 48
(2004), No.1/2, p.21
7) Advanced Welding and Joining Technologies, First edition, The Japan Welding
Engineering Society (2009)

16