WELDING RESEARCH

SUPPLEMENT TO THE WELDING JOURNAL, MAY 1987

m
Sponsored by the American Welding Society and the Welding Research Council

Readers are advised that all papers published in the Welding Journal's Research Supplement undergo
Peer Review before publication for: 1) originality of the contribution; 2) technical value to the welding
community; 3) prior publication of the material being reviewed; 4) proper credit to others working in the
same area; and 5) justification of the conclusions based on the results.
The names of the more than 160 individuals serving on the AWS Peer Review Panel are published
periodically. All are experts in their respective technical areas, and all are volunteers in the program.

Weldability Test for Thin Sheet Materials

Modified circular patch test successfully determines hot

cracking sensitivity of thin sheet materials

BY S. A. DAVID AND J. J. WOODHOUSE

ABSTRACT. Metals and alloys in thin
sheet form are difficult to evaluate reli-
ably for weldability (in particular, resis-
tance to hot cracking) by the existing
standard tests. The few tests that have
been developed are time consuming. A
modified circular weldability test for thin
sheet materials has been developed and
applied to successfully qualify hot crack-
ing sensitivity of austenitic stainless steels
and iridium alloys. Cracking in iridium
alloys has been found to be influenced by
the thorium content and by the fusion
zone grain structure.
Introduction
One of the major concerns in welding
is the question of weldability of thin sheet
materials. Considering the large tonnage
of metals and alloys used in the form of
thin sheets, it is essential to determine the
weldability of thin sheet material, espe-
cially as a function of heat-to-heat varia-
tions. The term "weldability" is used here
to mean the capacity of a metal or
5. A. DAVID and I. ). WOODHOUSE are with
the Materials joining Group, Metals and
Ceramic Division, Oak Ridge National Labora-
tory, Oak Ridge, Tenn.
combination of metals to be welded
under fabrication conditions into a specif-
ic, suitably designed structure and to
perform satisfactorily in the intended ser-
vice (Ref. 1). Weldability is a complicated
property, however, and not easily
defined. One of the important considera-
tions of weldability is the ability of the
material to avoid hot cracking during
weld fabrication. Although several tests
(Ref. 2) have been developed to deter-
mine the hot cracking tendency of metals
and alloys, most of these tests are only
applicable to thick sections, that is, great-
er than about 2.5 mm (0.1 in.). Metals and
KEY W O R D S
Thin Sheet Weldability
Weldability Test
Hot Crack Sensitivity
Circular Patch Test
Modified Patch Test
Iridium Alloys
Hot Crack Resistance
Sheet Weldability Test
Iridium Hot Cracking
Austenitic SS Cracking
alloys in thin sheet forms are difficult to
evaluate reliably for weldability (in partic-
ular, resistance to hot cracking) by the
existing standard tests. Cracks can form
within the weld metal during welding, in
the heat-affected zone (HAZ), or occa-
sionally in both places. Currently, there
are three tests, i.e., Houldcroft (Ref. 3),
Lambert (Ref. 4), and circular patch (Refs.
5, 6), available to test the hot cracking
tendency of thin sheet materials.
The Lambert test involves making a
series of fusion passes side-by-side along
the length of a steel strip, well secured in
a seam weld unit. A dye penetrant check
then allows assignment of a qualitative
rating, ranging from no cracks to bad
cracking. Although this test is good for
characterizing the hot cracking sensitivity
of materials qualitatively, the process has
been found to be very time consuming
and operator dependent. In the Hould-
croft test, the specimen design is such
that a gradation of self restraint is
imposed on the specimen by slots of
varying length in the edges of the sheet.
Depending on the cracking sensitivity of
the material, as a weld pass is made on
the specimen a crack may or may not
initiate and propagate. The length of the
crack is used as a measure of hot cracking
susceptibility. Finally, the circular patch

WELDING RESEARCH SUPPLEMENT 1129-s

 WELDABILITY TEST FIXTURE

8-32 CAP SCREW

SOCKET
HEAD 8-32 HEXAGON
SOCKET

Fig. 7 Clad fuel produced in plutonium fuel

form facility at the Savannah River Plant (Ref.
13)

test involves making a full penetration gas

tungsten arc weld bead in a circle on a
test plate which is prepared with or
without a simple circular butt joint.
Iridium alloys (lr-0.3W) doped with
about 60 w t ppm thorium (designated
DOP-26) are currently used as contain-
ment material for radioactive fuel in ther-
moelectric generators that provide sta-
0.50
ble electrical power for a variety of outer
planetary missions (Ref. 7) Fig. 1. Iridium -1.75-
alloys were chosen for this application 2.125
because of their high melting point, good
high-temperature strength, oxidation re- 2.75 -
sistance, and compatibility with the oxide 3.00-
fuel and the insulation materials. Thorium
is added as a grain boundary strengthen-
er, segregating strongly to the grain H 0 L D DOWN PLATE
boundaries and inhibiting intergranular
fracture. Although the susceptibility of
iridium alloys to hot cracking has been -0.024
investigated (Ref. 8), experience suggests
that the DOP-26 iridium alloy hardware
fabricated by approved procedures and
specifications varies in weldability. Defec-
tive welds represent a large investment in
L 0.125
starting materials and in manufacturing
costs. The only accurate measure of the
weldability of a given group of iridium
alloy hardware is the production weld
acceptance rate. The capability to identi-
DIMENSIONS
Z SPECIMEN

VENT H O L E , 0 . 2 5 DIAM

fy batches of DOP-26 iridium alloy mate-
rial with poor weldability early in the clad
manufacturing step could result in an
enormous cost saving.
Recently, a modified circular discrimi-
natory weldability test that is qualitative in
nature was developed at Oak Ridge
National Laboratory (ORNL). The test was
developed to determine the hot cracking
sensitivity of iridium alloy in the form of
thin sheets as a function of heat-to-heat
variation in alloy composition. This paper
describes the test procedure and its appli-
cation. Although the test was developed
particularly for iridium alloy sheets used
for space applications, it may be used for
other alloys, such as stainless steels.
IN INCHES SECTION A - A
Fig. 2 Schematic of the weldability test fixture
Experimental Procedure
The weldability testing was conducted
with a fixture developed at ORNL by S. A.
David and J. J. Woodhouse (Ref. 9). The
test fixture is shown schematically in Fig.
2. The details of the fixture are described
elsewhere (Ref. 9). Major elemental com-
positions of the iridium alloys tested are
given in Table 1. Thorium level was
determined by spark source mass spec-
troscopic analysis. Also, where available,
analysis for thorium as determined by
isotope dilution mass spectrographic
analysis is reported. The latter technique
has been found to have better accuracy.
In addition to the iridium alloys, a series of
austenitic stainless steels, Types 304 and
310, were tested to qualify the testing
procedure. The tests were conducted
with optimized test variables for produc-
tion materials.
The specimen, in the form of a disc 2
in. (50 mm) in diameter and 0.025 in.
(0.63 mm) thick, is inserted into the test
fixture. First, t w o circular autogenous
welds of 0.875-in. (22.3-mm) and 1.375-
in. (35.0-mm) diameter are made by the

130-s | M A Y 1987
Table 1Composition of Alloy DOP-26 Specimens

Composition (wt ppm)

Sample
a
Identification W Th< > Th" Al Cu Si Zr

RR921-5 3400 52 50 10 10 20 1
RR923-6 3900 48 54 50 10 5 1
RR831-4 3200 44 57 62 5 5 3 <1
S019-3 2800 41 54 10 3 5 <1
S019-4 2800 41 54 10 3 5 <1
S035-1 2400 41 65 5 to 3 10
LR304-7 2600 20 30 3 5 3 0.03
LR325-1 4800 25 56 10 - 10 <1
LR325-2 4800 25 56 10 40 10 <1
V205-2 3100 57 58 43 3 3 1 <1
V208-1 2800 36 59 54 3 3 1 <1
SR063-5 3400 37 53 53 10 10 1 <1
SR065-1 2500 50 55 43 5 20 5 10
SR066-2 3600 34 55 58 3 10 3 5
MERR-42-1 2900 56 42 50 10 5 1
MERR-42-2 2900 56 42 50 10 5 1
X302-5 2700 62 64 60 3 1 3 <0.3
X304-5 2700 50 69 37 10 10 5 <0.3
X305-4(c> 3200 70 73 60 3 5 3 <0.3
X326-5 3300 41 69 53 20 5 3 <0.3
X333-1 2400 63 70 32 3 1 3 <0.3
(a)
Spark Source mass spectrography determination.
(b)
lsotope dilution mass spectrography determination.
Cc)
Possibly contained residual gas in the ingot.

gas tungsten arc (GTA) welding process in
an inert-atmosphere glove box. The
welding variables used are given in Table
2. The t w o specific diameters, which
were established after a number of
experiments, establish varying restraint
levels. Secondly, the disc is removed and
inverted, and the weld procedure is
repeated. The first welding procedure
renders the microstructure (solidified) of
the material susceptible to cracking and
also increases the residual stress within
the disc, which could enhance the mate-
rial's susceptibility to cracking. The
restraint may be varied through the
extent to which the top plate is clamped
down.
The sequence in which the circular
welds are made is very important and
was maintained throughout this investiga-
tion. The 1.375-in. diameter (35.0-mm)
weld was made first, the 0.875-in. (22.3-
mm) weld was made second, the disc
was turned over, and the sequence was
repeated. If cracking was observed in the
first sequence, the second sequence was
not repeated. Also, the cracking was
examined carefully to identify whether it
occurred along the centerline or in the
Table 2Welding Variables
Initial current
Up-slope time
Weld current
Down slope time
Finish current
Welding speed
Arc oscillation
initial weld bead under the arc taper. The
latter has been a critical problem in pro-
duction welding (Ref. 10). Standard
Fig. 3 - Circular welds of different diameters.
metallographic techniques were used for A Type 304 stainless steel disc showing no
optical microscopic examination of the tendency for cracking; B Type 310 stainless
welds. The iridium alloy welds were elec- steel showing tendency to hot crack
trolytically etched in a solution of 400 ml
H 2 0 , 100 ml HCI, and 50 g NaCI in a
stainless steel container.
Results and Discussion
Acceptance Criteria Stainless Steels
If the disc subjected to the above Results of the tests on stainless steels
procedures does not show any evidence are tabulated in Table 3. In the test, Type
of cracking in either of the t w o circular 304 stainless steel sheet showed no
welds, the material is classified as weld- cracking during any of the sequential
able. If the circular weld of smaller diam- welds of various diameters, showing that
eter shows evidence of cracking and the Type 304 is weldable. For Type 310
larger diameter does not, the material stainless steel, the test showed no crack-
may be classified as one that is suscepti- ing during the first sequence of welds.
ble to cracking. If both circular welds However, upon turning the disc over and
show evidence of cracking, the material welding with the procedure described
may be classified as one that is highly earlier, the inner weld showed cracking
susceptible to cracking. during welding. Results also show that
increasing the diameter of the smaller
circular weld to 0.937 in. (23.8 mm) and
repeating the procedure eliminated the
Table 3Stainless Steel Weldability Test
Results
cracking. Figure 3 shows the results of the
dye penetrant check on these welds.
Weldability Test
Work conducted on several other stain-
less steels with this test agrees well with
Side 1 Side 2 the results obtained by using the Sigmajig
1.375- 0.875- 1.375- 0.875- test developed by C. M. Goodwin (Ref.
m. in. in. in. 11).
50 A Material diam. diam. diam. diam.
0.5 s
Type 304 NC NC NC NC
85 to 120 A
Type 310 NC NC NC C Iridium Alloys
2.5 s
Type 310 NC NC(a> NC NC'a>
20 A
30 ipm The weldability of iridium alloys has
NC = no cracking.
6.6 Hz C = cracked. been found to be very sensitive to the
(a,
lnner weld diameter increased to 0.937 in. amount of thorium additions. Although

WELDING RESEARCH SUPPLEMENT 1131-S

Table 4Iridium Alloy Weldability Test Results

Weldability Test
Side I Side 2
Sample
Identification 1.375- in. diam 0.875- in. diam. 1.375- in. diam. 0.875- in. diam.

RR921-5 NC NC NC NC
RR923-6 NC NC NC C
RR931-4 NC NC NC C
S019-3 NC NC C NW
S019-4 NC NC c NW
S035-1 NC NC c NW
LR304-7 NC NC c NW
LR325-1 NC NC c NW
LR325-2 NC NC c NW
V205-2 NC C NW NW
V208-1 NC c NW NW
SR063-5 NC NC NC NC
SR065-1 NC c NW NW
SR066-2 NC c NW NW
MERR-42-1 C c NW NW
MERR-42-2 c NW NW NW
X302-5 NC c NW NW
X304-5 c c NW NW
X305-4 C(a) C (- NW NW
X326-5 C C NW NW
X333-1 c C NW NW

NC = no cracking.
C = cracked.
N W = no w e l d .
<8)
C = cracked and excessive amount of w e l d porosity

alloy DOP-26 with 60 wt ppm thorium
can be welded with the restraints
imposed during a melt run test (autoge-
nous bead), alloys containing 100 wt ppm
thorium or greater have been found to
be very sensitive to hot cracking during
GTA welding (Ref. 12). At this level of
thorium, the alloy is likely to become very
sensitive to minor changes in welding
variables as they affect hot cracking.
The weldability test results on the iridi-
um alloy discs are listed in Table 4. Given
the fact that iridium alloys are prone to
cracking during welding, the test was
able to rank successfully the various heats
of material as to their weldability. Of the
various heats tested, the RR batch had
the best weldability characteristics, fol-
lowed by the S, LR, V, SRO, MERR, and X
batches. The MERR and X batches were
rated the worst. The observed cracking
in some of these batches is shown in Fig.
4. Most of the observed cracks were
centerline cracks. Sometimes both radial
cracking and underbead cracking in the
HAZ of the single-pass weld bead under
the arc taper section of the weld were
observed. The cracking tendency during
the test has been observed to be related
only to the thorium level. For all the
heats, the lower the thorium level, the
better the weldability. Reported thorium
levels of > 5 0 ppm (with the error
involved in Th analysis) could very well
put the possible thorium level well above
the critical level of 100 ppm, at which the
alloy is very sensitive to cracking. The
level of other residuals did not correlate
well with the cracking tendency of the
alloy.
Fig. 4 Circular welds of different diameters in
two heats of alloy DOP-26 showing varying
tendency to hot crack
Weld Cracking and Microstructure
Microstructural features of the circular
welds that resisted cracking are shown in
Fig. 5. The fusion zone grain structure
appears relatively fine grained. However,
the microstructural features of the welds
that cracked revealed a coarse-grained
fusion zone Fig. 6. Grain structure of
the fusion zone plays an important role in
hot cracking behavior of alloys. In iridium
alloys, fine-grained fusion zone micro-
structures have been found to resist hot
cracking (Refs. 12, 13).

Fig. 5 Fusion zone microstructure of circular weld in alloy DOP-26 without hot cracking. A Top surface; B Transverse section

132-s I MAY 1987

 2
a.
O
_i
UJ
>
Ul
a
x
o
tr
<
UJ
in
Ul
tr
Z
UJ
E
Fusion zone microstructure of circular weld in alloy DOP-26 exhibited tendency to hot crack. A Top surface; BTransverse section a.
O
Hot cracking susceptibility of an alloy to the fusion line and may or may not lr5Th at about 2000C (3632F). In a
depends on its composition, its segrega- extend into the weld metal. Sometimes recent weld cracking problem during
tion behavior, and the microstructural cracking may initiate within the fusion welding of general purpose heat source
characteristics, in particular, grain struc- zone, usually along the centerline, capsules, all the above factors contribut-
O
ture within the fusion zone. During weld- depending on the level of thorium segre- ing to cracking were identified. Occasion- tr
ing of the iridium alloys, the fusion zone gation and fusion-zone grain structure. ally, a significant amount of porosity was <
UJ
grain structure and segregation charac- Low-melting eutectic patches of possible observed in the weld. This was the case V)
Ul
teristics were found to be very critical in IrsTh-lr phases have been observed on in certain heats of the iridium alloys that
determining the hot cracking susceptibili- the fractured surface of the weld metal. were severely prone to cracking. Figure 9 tr
ty of the alloy. With the thermal and Figure 8 shows the presence of a liquid is a scanning electron photomicrograph H
solidification stresses acting transverse to film and the eutectic. The observed of the crack surface exhibiting pores. The
the welding direction, a fusion zone with eutectic patches relate well to the iridi- origin of porosity in this case is attributed
a coarse grain structure is very prone to um-thorium phase diagram (Ref. 15) that to the quality of the iridium ingot and a
O
cracking. During the weldability test, indicates a eutectic between iridium and associated gas content. _i
cracking in the iridium alloys predomi- UJ
>
nantly followed the centerline and was
intergranular in nature. However, in the
production welds, the significant problem X
was the formation of small underbead
o
WELDING cc
cracks that formed in the single-pass weld DIRECTION <
UI
bead under the arc taper (Ref. 10). OUTSIDE SURFACE <n
Cracks were intergranular and did not UJ
DOUBLE PASS ec
extend through the capsule wall. This SINGLE PASS
phenomenon is described schematically
in Fig. 7A, and a transverse section of the CRACK
weld showing such cracking is shown in
INNER SURFACE a.
Fig. 7B. O
_i
Earlier examinations of the microstruc- UJ
tures and crack path had indicated that >
UJ
the crack had initiated in the HAZ as a
a
liquation crack. This type of crack initia- ->
tion depends on the thorium level in the x
alloy. The HAZ cracking is possibly o
tr
caused by melting in the highly thorium <
segregated base metal grain boundaries, UJ
to a distance determined by the thermal tn
gradient normal to the weld and the la
tensile stresses that develop in the HAZ as 1 ff
the heat source passes by. Thorium in this
S
alloy has been added as a grain boundary a.
strengthener, as noted earlier. Also, seg- O
-J
regation of thorium to the grain bound- ui
aries has been observed (Ref. 14). If the >
-
thorium concentration along the grain / ,1 ' 2 .
boundaries of the base metal and in the
last liquid to solidify is high enough, it o
could lead to the formation of low- oc
<
melting constituents (in the thorium-iridi- Ul
um system, a eutectic). Once the crack is tn
Fig. 7 Typical underbead weld cracks in alloy DOP-26. A Schematic of longitudinal section
initiated in the HAZ, it could grow normal showing crack location; B Transverse weld microstructure with the crack

WELDINC RESEARCH SUPPLEMENT 1133-s

Fig. 8Scanning electron micrographs showing eutectic patches on the surface of the fractured weld in alloy DOP-26
% . 9-Scanning electron micrographs showing porosity on the surface of the fractured weld in alloy DOP-26
Conclusions
A simple weldability test t o d e t e r m i n e
the hot cracking sensitivity of thin sheets
o f materials has b e e n d e v e l o p e d . T h e
test has b e e n successfully a p p l i e d t o char-
acterize the h o t cracking sensitivity of
thin sheets of iridium alloys (heat-to-heat
variations) a n d stainless steels.
T h e o b s e r v e d cracking in iridium alloys
has b e e n a t t r i b u t e d t o variations in t h o r i -
u m c o n t e n t leading t o H A Z liquation
cracking and solidification cracking. T h e
presence of l o w - m e l t i n g constituents, in
particular, eutectic, has b e e n identified
o n the f r a c t u r e d surface.
A ckno wledgments
T h e a u t h o r wishes t o a c k n o w l e d g e W .
R. Kanne o f t h e Savannah River Plant f o r
p r o v i d i n g s o m e of t h e p h o t o m i c r o -
graphs, G. M . G o o d w i n a n d J. F. King f o r
r e v i e w i n g the manuscript, a n d K. W .
G a r d n e r f o r t y p i n g . Finally, a c k n o w l e d g -
m e n t is d u e t o R. H. C o o p e r a n d M . M .
M a r t i n as p r o g r a m managers f o r their
s u p p o r t and e n c o u r a g e m e n t t o c o n d u c t
this w o r k . T h e research w a s s p o n s o r e d
134-s I MAY 1987
"^JR*^
b y t h e Division of Special Applications,
O f f i c e of D e f e n s e Energy Projects and
Special Applications, O f f i c e o f Nuclear
Energy, U.S. D e p a r t m e n t of Energy,
under contract DE-AC05-840R21400
w i t h M a r t i n M a r i e t t a Energy Systems,
Inc.
References
1. Welding Handbook. 1976. Vol. 1, 7th
Edition, p. 137, American Welding Society,
Miami, Fla.
2. Vagi, J. )., Meister, R. P., and Randall, M.
D. August 1968. Weldment evaluation meth-
ods. DMIC 244.
3. Houldcroft, P. T. 1955. A simple cracking
test for use with argon-arc welding. British
Welding lournal 2:471-475.
4. Brooks,). A., and Lambert, F.). 1978. The
effects of phosphorus, sulfur, and ferrite con-
tent on weld cracking of Type 309 stainless
steel. Welding journal 56(5): 139-s to 143-s.
5. Rundell, G. R., and Nehrenberg, A. E.
1966. Weld metal cracking of Invar in circular
patch tests. Welding lournal 45(4): 156-s to
160-s.
6. Borland, |. C , and Rogerson, |. H. 1962.
Examination of the patch test for assessing hot
cracking tendencies of weld metal. British
** m+ S *
Welding lournal 9(8):494-499.
7. Liu, C. T., and Inouye, H. October 1977.
Development and Characterization of an
Improved lr-0.3 W Alloy for Space Radioisoto-
pic Heat Sources. ORNL-5290, Oak Ridge
National Laboratory.
8. David, S. A., and Liu, C. T. March 1980.
Weldability and hot cracking in thorium doped
iridium alloys. Metals Technology 7':102-106.
9. David, S. A., and Woodhouse, |. |. U.S.
patent number 4,499,758.
10. Kanne, W . R., |r. 1983. Welding iridium
heat source capsules for space missions. Weld-
ing lournal 62(8): 17-22.
11. Goodwin, G. M. Sigmajig Weldability
Test, Private communication.
12. David, S. A., and Liu, C. T. 1982. High
power laser and arc welding of thorium doped
iridium alloys. Welding lournal 61(5):157-s to
163-s.
13. Scarbrough, ). D., and Burgan, C. E.
1984. Reducing hot-short cracking in iridium
GTA welds using four-pole oscillation. Welding
lournal 6(6):54-56.
14. White, C. L. 1979. The effect of trace
element additions on the grain boundary com-
position of lr-0.3 W alloys. Met. Trans.
10A:683-692.
15. Thomson, ). R. 1964. Alloys of thorium
with certain transition metals. /. Less Common
Metals, Vol. 6, pp. 3-10.