JMEPEG (2008) 17:30–36
DOI: 10.1007/s11665-007-9132-1
ASM International
1059-9495/$19.00
Effect of Auxiliary Preheating of the Filler Wire on Quality
of Gas Metal Arc Stainless Steel Claddings
Amandeep S. Shahi and Sunil Pandey
(Submitted June 8, 2006; in revised form November 9, 2006)
Weld cladding is a process for producing surfaces with good corrosion resistant properties by means of
depositing/laying of stainless steels on low-carbon steel components with an objective of achieving maximum economy and enhanced life. The aim of the work presented here was to investigate the effect of
auxiliary preheating of the solid filler wire in mechanized gas metal arc welding (GMAW) process (by using
a specially designed torch to preheat the filler wire independently, before its emergence from the torch) on
the quality of the as-welded single layer stainless steel overlays. External preheating of the filler wire
resulted in greater contribution of arc energy by resistive heating due to which significant drop in the main
welding current values and hence low dilution levels were observed. Metallurgical aspects of the as welded
overlays such as chemistry, ferrite content, and modes of solidification were studied to evaluate their
suitability for service and it was found that claddings obtained through the preheating arrangement,
besides higher ferrite content, possessed higher content of chromium, nickel, and molybdenum and lower
content of carbon as compared to conventional GMAW claddings, thereby giving overlays with superior
mechanical and corrosion resistance properties. The findings of this study not only establish the technical
superiority of the new process, but also, owing to its productivity-enhanced features, justify its use for lowcost surfacing applications.
Keywords
austenitic stainless steel, ferrite, preheated filler wire,
UGMAW process, weld cladding
1. Introduction
Increasing productivity of any welding process while
maintaining or even improving the weld quality has been the
task of researchers in the field of development of welding
processes. Previous predictive studies on gas metal arc welding
(GMAW) process have had various purposes. Researchers have
attempted to model GMAW process in different metal transfer
modes and tried to optimize it using different techniques (Ref
1-3) apart from accounting for wire melting rate in this process
(Ref 4-6).
1.1 Cladding
The term weld cladding usually denotes the application of a
relatively thick layer (approximately 3 mm or 1/8th in.) of weld
metal for the purpose of providing a corrosion-resistant surface
(Ref 7). In modern industry, increasing use is being made of
clad materials as a means of achieving the optimum balance of
strength, special surface properties, and economy. Some of the
typical base metal components that are weld-cladded include
the internal surfaces of carbon and low-alloy steel pressure
Amandeep S. Shahi, Department of Mechanical Engineering, Sant
Longowal Institute of Engineering & Technology, Longowal, Punjab
148106, India; and Sunil Pandey, Department of Mechanical
Engineering, Indian Institute of Technology Delhi, New Delhi
110016, India. Contact e-mail: ashahisliet@yahoo.co.in.
30—Volume 17(1) February 2008
vessels, paper digestors, urea reactors, tube sheets, and nuclear
reactor containment vessels. Among the various welding
processes employed, GMAW process has become a costeffective choice for cladding smaller- and medium-sized areas
due to its superior quality, all position capability, and ease of
mechanization. The characteristics and typical uses of various
weld-surfacing processes are mentioned in Table 1.
1.2 Dilution
It is defined as the ratio of the cross section of weld metal
below the original surface to the total area the weld bead
measured on the cross section of the weld deposit (Ref 8).
Various combinations of procedural parameters like primary
parameters viz. welding current, voltage, welding speed, and
secondary parameters like polarity, electrode size, wire stickout,
welding position/inclination, arc shielding, electrode oscillation, welding technique, additional filler metal etc., which affect
dilution, can be incorporated into a procedure (Ref 9). Various
processes like SAW, GTAW, PAW, GMAW, ESW, FCAW, Strip
cladding, Explosive welding (Ref 10-13), etc., have been used
for cladding operation with an aim of minimizing dilution to as
low value as possible without sacrificing the joint integrity. This
requires a thorough understanding and proper control over a
number of variables, which affect dilution. Use of hot filler
additions (Ref 14) in various conventional processes like TIG,
Laser, Plasma arc, etc., have been reported which affect dilution
to a significant extent.
1.3 Auxiliary Preheating Arrangement in GMAW Process
(Universal Gas Metal Arc Welding Process)
This process makes use of a specially designed torch as
shown in Fig. 1. It employs two contact tips and a secondary
Journal of Materials Engineering and Performance
Table 1 Characteristics and typical uses of various weld-surfacing processes
Approximate minimum
deposit thickness, mm
Deposition
rate, kg/h
Dilution of
single layer, %
Oxy-acetylene (OA)
Powder weld (PW)
Manual metal arc (MMA)
Tungsten inert gas (TIG)
Plasma transferred arc (PTA)
Gas metal inert gas (GMAW)
0.5
0.1
3
1.5
2
2
1
0.2-1
1-4
2
10
3–6
15
ÆÆÆ
15-30
5-10
2-10
15-30
Flux-cored arc (FCAW)
2
3–6
15-30
Submerged arc wire (SA)
Submerged arc strip (SA)
Electroslag (strip) (ESW)
3
4
4
10-30
10-40
15-35
15-30
10-25
5-20
Process
Typical uses
Small areas deposits on light sections
Small areas deposits on light sections
Multilayer on heavier sections
High quality low dilution work
High quality low dilution work
Faster than MMA, no stub end loss,
position work possible
Similar to GMAW, mainly for iron-base alloys
for high abrasion resistance
Heavy section work, high-quality deposits
Corrosion resistant cladding of large areas
High-quality deposits at higher deposit rates
than SAW, Limited alloy range
2. Experimental Work
2.1 Base Material and Filler Used
The popularly used structural steel, which was cut down to
suitable sizes of 200 · 150 · 12 mm plates each, was used as
the substrate material for the present investigation and the solid
filler wire used was 316L (extra low-carbon grade) of 1.14 mm
diameter, which because of higher molybdenum content has a
higher corrosion and creep resistance, thus making it a suitable
choice for chemical, pulp handling, photographic, and food
equipment. The chemical composition of the base and the filler
metal is given in Table 2.
2.2 Trial Runs
Fig. 1 Schematics of GMAW process with preheating arrangement
(Ref 15)
power source to preheat the filler wire prior to its emergence
from the welding torch, thereby providing an additional and
independent power source. In this arrangement, the major
role of welding current is dissipation of sufficient heat to
support the arc, to melt the surface of the base plate, and to
fuse the hot incoming wire. The main difference between
conventional GMAW and this arrangement, in terms of
heating, is that the preheated wire further experiences I2R
heating after it leaves the lower contact tip. This allows
breaking of the fixed relationship between welding current,
wire stickout, and the deposition rate, which often limits
conventional GMAW process. The use of the independent
secondary power source enables the heat content of the filler
wire to be independently controlled, thus providing the ability
to weld at a desired deposition rate while reducing the
welding current,the wire stickout, the arc force, and the heat
input (Ref 16).
Journal of Materials Engineering and Performance
Trial runs were conducted for establishing the working
range of the input parameters viz., wire feed rate, open circuit
voltage, welding speed, electrode stickout, and preheat current
to the filler wire. Weld quality was considered to be acceptable
when the input parametric combination resulted in beads which
were free from various visual defects like macrocracking, nonuniform ripples on the bead, excessive convexity and spatter,
surface porosity, geometrical inconsistency, etc. Welding was
done in the mechanized mode using the model Power Wave355 from Lincoln Electric Co., with constant voltage system,
which facilitated the variation of wire feed rate and voltage in
steps of 0.05 m/min and 0.1 V, respectively. Owing to the high
resistivity of the filler wire it could withstand a maximum
preheat current of 110 A only, which was provided using a
transformer (Table 3). Other secondary process parameters
used for the final beads were:
Torch angle = 90
Shielding gas used = industrially pure Argon
Shielding gas flow rate = 20 L/min
Electrode polarity = Reverse
Cladding position = Flat
2.3 Quantitative Comparisons of GMAW and Preheated Filler
GMAW Process as Regards Chemical Composition in
Single Layer Cladding
This included weld overlaying in the mechanized mode, of
austenitic stainless steel 316L filler wire of 1.14 mm diameter
Volume 17(1) February 2008—31
Table 2 Chemical composition of the base and filler wire (wt.% age) with Fe as balance
Material
Base metal
Filler wire
C
Mn
Si
Cr
Ni
Mo
Cu
S
P
0.295
0.019
ÆÆÆ
1.61
0.18
0.37
0.25
19.12
ÆÆÆ
12.47
0.50
2.83
ÆÆÆ
0.10
0.018
0.014
0.027
0.019
Table 3 Different welding conditions used with recorded responses
Sr. No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Wire feed rate, m/min
Open circuit voltage, V
Welding speed, cm/min
Electrode stick-out, mm
Process
Dilution, %
10
10
6
6
7
7
7
7
7
7
7
7
7
7
7
7
8
8
34
34
34
34
40
40
28
28
34
34
34
34
34
34
34
34
34
34
30
30
30
30
30
30
30
30
40
40
20
20
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
42
42
18
18
30
30
GMAW
UGMAW
GMAW
UGMAW
GMAW
UGMAW
GMAW
UGMAW
GMAW
UGMAW
GMAW
UGMAW
GMAW
UGMAW
GMAW
UGMAW
GMAW
UGMAW
33.33
24.18
27.20
12.45
32.54
22.48
23.22
15.12
22.90
13.18
22.25
13.67
23.45
13.34
33.65
24.72
23.14
13.90
Fig. 2 Specimen cutting plan
on 12 mm thick low-carbon steel (IS: 2062 Grade 1 which is
used as general structural steel) with the objective of producing
a high-alloy fully austenitic surface in one weld layer. Figure 2
and 3, respectively, show the specimen cutting plan and the
cross sections of the weld bead profiles.
Figure 3 shows the cross sections of the bead profiles
obtained using GMAW and Preheated filler-GMAW process.
Welding parameters used were those which would give the
optimum dilution conditions (Ref 17):
Wire feed rate = 6 m/min, Open circuit voltage = 30 V,
Welding Speed = 20 cm/min, Electrode stickout = 30 mm,
Preheat current = 110 A (preheating resulted in 36 A of drop
in the main welding current).
Table 3 shows the variation of dilution with respect to
different input parametric combinations in GMAW and
UGMAW process.
Table 4 represents the relative comparisons of various weld
bead geometry parameters.
32—Volume 17(1) February 2008
After laying down single overlays, the chemical composition, at a distance of 2 mm from the top of the weld bead was
checked and is mentioned in Table 5.
Table 6 shows the input parametric combinations for GMAW
and UGMAW process, yielding the same level of dilution i.e.
33%. This comparison shows the capability of UGMAW
process in giving higher deposition rate than GMAW process
(which is one of the main objectives of cladding operation).
2.4 Effect of Buttering Layer
In order to compensate for increased dilution (Table 6),
generally, use is made of the buttering layer, which is generally
high-chrome filler like 309L filler. First layer of solid filler
309L was laid which was followed by the second layer of 316L
layer with an inter-pass temperature of 150 C using GMAW
process with other welding conditions remaining constant as
used above (Table 7).
Journal of Materials Engineering and Performance
3. Microstructural Studies
low dilution capability). Photomicrographs in Fig 4(a) shows
characteristic primary solidification structures as they appear in
different zones of a weld bead of austenitic stainless steel
overlay with normal cooling in air. The solidification structure
was found to be mainly cellular or cellular-dendritic. Narrow
zones of planar growth were, however, found along the fusion
line in claddings surfaced with UGMAW process. Furthermore,
no equiaxed grains were found in the weld metal.
Figure 4(b) shows the cellular and cellular-dendritic structure of fully austenitic phase solidified in 316L stainless steel
overlay surfaced with preheated filler-GMAW process.
Standard metallurgical procedures like sectioning, grinding,
polishing, and etching (etchants used were 2% Nital for the
base metal and 10 g oxalic acid in 100 mL of distilled water for
the weld metal 316L) were employed to prepare the samples
taken for this study (those using UGMAW process owing to its
4. Ferrite Studies
Since weld microstructure is greatly influenced by chemical
composition, a number of empirical relationships and constitutional diagrams like SchaefflerÕs diagram, Delong diagram,
WRC 1992 diagram (Ref 18-20), and even the latest prediction
models that account for cooling rate effects (Ref 21-23) have
been developed to predict microstructures based on actual or
approximated composition.
Various constitutional diagrams and empirical relationships
were used in order to predict the ferrite content in the clad metal
because the importance of this study lies in the fact that in order
to avoid hot cracking or microfissuring in austenitic stainless
steels a minimum of 4% ferrite is necessary. The following
formulas were used using different constitution diagrams for
predicting the ferrite content of weld metals.
Fig. 3 Cross sections of weld bead profiles, as obtained with
GMAW process (a) 182 A arc and preheated filler-GMAW (b)
146 A arc-observe significant decrease in the penetration in (b) and
peaky bead appearance
1. Schaeffler Chromium Equivalent = (%Cr + %Mo + 1.5%
Si + 0.5%Nb) Schaeffler Nickel Equivalent = (%Ni +
30%C + 0.5%Mn)
2. Delong Chromium Equivalent = (%Cr + %Mo + 1.5%
Si + 0.5% Nb) Delong Nickel Equivalent = (%Ni +
30%C + 30%N + 0.5%Mn)
3. WRC-1992 Chromium Equivalent = (%Cr + %Mo +
0.7%Nb) WRC-1992 Nickel Equivalent = (%Ni + 35%C +
20%N + 0.25%Cu)
4. Hammer and Svensson Chromium Equivalent = (%Cr +
1.37%Mo + 1.5%Si + 2%Nb + 3%Ti) Hammer and
Svensson Nickel Equivalent = (%Ni + 22%C + 1.31%Mn
+ 14.2%N + %Cu)
Table 4 Relative differences of weld bead geometry
parameters in GMAW and UGMAW (preheatied filler
GMAW) process
Parameter
GMAW
process
UGMAW
process
Relative
difference
4.24
9.0
3.36
20.32
4.96
7.5
1.24
11.35
14.51% increase
16.67% decrease
63% decrease
44.14% decrease
Height, mm
Width, mm
Penetration, mm
Dilution, %
Table 5 Chemical composition of single layer claddings
using GMAW and UGMAW processes
Process
Table 7 Chemical composition of GMAW claddings
using buttering layer of solid filler 309L
C
Cr
Ni
Mo
0.050
0.040
13.92
16.86
9.02
10.80
1.76
2.14
C
GMAW
UGMAW
0.050
Cr
Ni
Mo
17.97
11.02
1.55
Table 6 Comparison of GMAW and UGMAW process in terms of dilution
Process
GMAW
UGMAW
Wire
feed rate,
m/min
Open
circuit voltage,
V
Welding
speed, cm/min
Nozzle-to-plate
distance, mm
Welding
current, A
Heat
input/weld
length, kJ/mm
Dilution,
%
Deposition
rate, kg/h
4
8
28
36
28
42
20
16
162
214
0.826
0.902
33
33
3.242
4.230
Journal of Materials Engineering and Performance
Volume 17(1) February 2008—33
Fig. 4 (a) and (b) Microscopic view of the weldment (100·)
Table 8 Chromium and nickel equivalents (in percentage) of claddings
Schaeffler-Cr Schaeffler-Ni Delong-Cr
WRC-1992-Cr WRC-1992-Ni
equiv.
equiv.
equiv.
Delong-Ni equiv
equiv.
equiv.
Process
GMAW
Preheated filler-GMAW
16.175
19.495
11.34
12.84
16.175
19.495
12.343
13.825
15.75
19.07
12.397
13.827
Hammer & Hammer &
Svensson- SvenssonCr equiv.
Ni equiv.
16.826
20.186
13.5
15.126
Table 9 Comparison of predicted ferrite content (in percent) of the welds and predicted modes of solidification
Process
Schaeffler
Delong
WRC-1992
Hammer & Svensson
GMAW
UGMAW
Nil (A+M)
1 (A+F)
Nil (Below A+M line)
2.75 (A+F)
Not applicable due to low Cr content
1.2 (AF)
1.24 (A/AF)
1.33 (A/AF)
Table 8 shows the tabulated values of various equivalents
using the formulas as mentioned above.
The predicted solidification modes are represented in the
brackets as mentioned in Table 9 whose notation is as given
below:
Fig. 5 Multilayer stainless steel overlays
A + M indicates austenitic and martensitic mode
A + F is austenitic and ferritic mode
AF is austenitic-ferritic mode
A/AF is austenitic and austenitic-ferritic mode
5. Corrosion Test
5.1 Specimen Preparation
Fig. 6 Stainless steel specimen for corrosion testing
Predicted ferrite from Seferian equation = 3(Crequivalent 0.93Niequivalent -6.7), where Crequiv and Niequiv are defined by
Schaeffler and = 2.56 both for GMAW and UGMAW process.
34—Volume 17(1) February 2008
In order to evaluate the suitability of the claddings for nitric
acid environment, three layers of 316L were overlaid on the
low-carbon substrate using preheated filler-GMAW process as
shown in Fig. 5.
Thereafter corrosion test specimen was prepared in accordance with ASTM Practice A-262 for corrosion testing (Ref
24). The stainless steel overlay was machined out so as to make
it free from the base material. Then it was machined and ground
to the suitable size as shown in Fig. 6.
5.2 Performing Nitric Acid Test (HUEY Test)
The test solution used was 65 ± 0.2 wt.% nitric acid. The
solution was prepared by adding distilled water to concentrated
Journal of Materials Engineering and Performance
of carbon besides having relatively higher ferrite content as
compared to conventional GMAW claddings.
7. Conclusions
From the study undertaken, as above, the following
conclusions can be drawn:
Fig. 7 Corrosion testing apparatus
nitric acid (reagent i.e., HNO3, sp. gr. 1.42) at the rate of
108 mL of distilled water per liter of concentrated nitric acid.
As shown in Fig. 7 the stainless steel specimen was put in
the boiling nitric acid for 24 h and the weight loss was
determined.
Corrosion rate which is generally reported in in. /month or
mils /year was calculated as follows:
Inches per month ¼ ð287 wÞ=ðA d tÞ;
where t = time of exposure, h; A = Total surface area, cm2;
w = weight loss, g; and d = density of the sample, g/cm3.
Observed data was t = 24 h, A = 42.86 cm2, W = 0.08 g,
and d = 7.99 g/cm3.
2.793 · 10-3 in. per month or 33.51 mils per year (using
conversion factor of inches per month · 12,000 = mils per
year).
1. Dilution achieved in preheated filler-GMAW cladding is
significantly lower as compared to GMAW cladding
because preheating of the filler wire reduces base metal
penetration, apart from relatively smaller variations in
other bead geometry parameters, due to significant drop
in the main welding current.
2. Owing to lesser arc force, finger-like penetration was absent in preheated filler-GMAW process and the weld beads
obtained were peaky as compared to GMAW weld beads.
3. Preheated filler-GMAW claddings possessed higher contents of chromium, nickel, and molybdenum than GMAW
claddings indicating the productivity-enhanced feature of
the new process, i.e., by way of cutting costs due to lesser amount of clad metal build-up required for achieving
fully austenitic composition.
4. New process is capable of substituting buttering layer to
a significant extent thus resulting in considerable savings
of high-chrome filler 309L.
5. For the same value of dilution higher deposition rate
(more than 30%) is given by UGMAW process than
GMAW process.
6. Higher ferrite content was present in preheated fillerGMAW claddings as compared to GMAW which shows
its capability to give claddings having lesser tendency to
hot-cracking.
References
6. Results and Discussion
Auxiliary preheating of the filler wire reduces dilution
significantly which is due to the fact that for any given set of
welding conditions the heat content of the filler wire is partially
controlled by the preheating current (I2R heating) whereas
remaining energy required for melting the wire is provided by
the main welding current. Since reduction in arc force and the
heat transmitted to the weld pool are directly related to welding
current, any decrease in welding current will result in decreased
dilution. Hence the reason for obtaining significant reductions
in the penetration and consequently dilution values due to the
auxiliary preheated filler wire. Although the single layer
claddings obtained both by the GMAW and preheated fillerGMAW process did not meet the fully austenitic composition,
but new arrangement, besides capable of giving claddings with
superior mechanical and corrosion resistance properties, certainly has an upper edge over its conventional counterpart
GMAW in meeting the needs of low-cost surfacing applications. This is evident from the preheated filler-GMAW
claddings which possess higher content of expensive materials
namely chromium, nickel, and molybdenum and lower content
Journal of Materials Engineering and Performance
1. P.J. Modensi, C.M.D. Starling, and R.I. Reis, Wire Melting Rate
Phenomena in Gas Metal Arc Welding, Sci. Technol. Weld. Joining,
2005, 10, p 120–124
2. G. Huismann and H. Hoffmeister, Sensing Metal Inert Gas Process by
Measuring Wire Feed Rate and Current, Sci. Technol. Weld. Joining,
1999, 4, p 352–356
3. D. Kim, S. Rhees, and H. Park, Modelling and Optimization of GMA
Welding Process by Genetic Algorithm and Response Surface Methodology, Int. J. Prod. Res., 2002, 40, p 1699–1611
4. E. Halmøy, Wire Melting Rate, Droplet Temperature and Effective
Anode Potential, Proc. Conf. Arc Physics and Weld Pool Behaviour,
May 1979 (London, UK), Paper 29, The Welding Institute, 1979,
p 49–54
5. J.H. Waszink and G.J.P.M. Van Den Heuvel, Heat Generation and
Heat Flow in the Filler Metal in GMA Welding, Weld. J., 1982, 61,
p 269s–280s
6. M. Suban and J. Tusek, Dependence of Melting Rate in MIG/MAG
Welding on the Type of Shielding Gas Used, J. Mater. Process.
Technol., 2001, 119, p 15–192
7. ‘‘Welding, Brazing and Soldering,’’ ASM Metals Handbook, 1983,
Vol. 6, p 816–820
8. J.F. Lancaster, The Metallurgy of Welding. 4th ed., Allen and Unwin,
London, 1987
9. U.D. Mallaya and H.S. Srinivas, Effect of Magnetic Steering of the Arc
on Clad Quality in Submerged Arc Strip Cladding, Am. Weld. J., 1993,
72, p 289s–293s
Volume 17(1) February 2008—35
10. Y.K. Oh, J.H. Devletian, and S.J. Chen, Low-Dilution Electroslag
Cladding for Shipbuilding, Am. Weld. J., 1990, 69, p 37–40
11. N. Murugan and R.S. Parmar, Stainless Steel Cladding Deposited
by Automatic Gas Metal Arc Welding, Am. Weld. J., 1997, 76,
p 391s–403s
12. N. Murugan and T. Kannan, Effect of Flux Cored Arc Welding Process
Parameters on Duplex Stainless Steel Clad Quality, J. Mater. Process.
Technol., 2006, 176, p 230–239
13. K. Ramazan and A. Mustafa, An Investigation on the Explosive
Cladding of 316L Stainless Steel-Din-P355 GH Steel, J. Mater.
Process. Technol., 2004, 152, p 91–96
14. I.D. Harris, TIG Hot Wire Offers High Quality, High Deposition, Met.
Construct., 1986, p 445–449
15. A.S. Shahi and S. Pandey, AWS Professional Program, Fabtech
International and AWS Welding Show, Chicago, USA, November 2005
16. I. Stol, Advanced Gas Metal Arc Welding Process, First International
Conference on Advanced Welding Systems, Organised by The Welding
Institute, London, 1985
36—Volume 17(1) February 2008
17. A.S. Shahi and S. Pandey, Welding Current Prediction in GMAW and
UGMAW Process Using Response Surface Methodology, Sci. Technol.
Weld. Joining, 2006, 11(3), p 341–346
18. G.R. Speich, Metals Handbook, 8th Ed., Vol. 8, ASM, 1973, p 424
19. J.A. Brooks and J.C. Loppold, Selection of Wrought Austenitic
Stainless Steels, ASM Handbook, Vol. 8, p 456–460
20. G.M. Goodwin, The Effects of heat Input and Weld Process on Hot
Cracking in Stainless Steels, Am. Weld. J., 1998, 77, p 43
21. D.J. Kotecki, Ferrite Determination in Stainless Steel Welds-Advances
Since 1974, Weld. J., 1997, 76, p 24s
22. J.M. Vitek, S.A. David, and C.R. Hinman, Improved Ferrite Number
Prediction Model that Accounts for Cooling Rate Effects-Part 1: Model
Development, Am. Weld. J., 2003, 82, p 10s–17s
23. J.M. Vitek, S.A. David, and C.R. Hinman, Improved Ferrite Number
Prediction Model that accounts for Cooling Rates Effects-Part 2: Model
Results, Am. Weld. J., 2003, 82, p 43s–50s
24. ‘‘Standard Practices for Detecting Susceptibility to Intergranular Attack
in Austenitic Stainless Steel,’’ ASTM A 262-91, 1992, p 8–10
Journal of Materials Engineering and Performance