Vous êtes sur la page 1sur 11

The Effects of Warm Overstressing on Pressure Vessel

Steel Properties
PVRC study shows that overstressing of flaws raises the
fracture resistance of A516 Grade 70 and A533 Grade B but that
subsequent strain aging causes some toughness reduction

ABSTRACT. The beneficial effects of warm response, while 500 and 650° F aging Such proof testing, however, is at
prestressing have been demonstrated in a produced more embrittlement. The stress stress levels high enough to produce
number of laboratory tests and, in princi- relief treatment for both steels improved plastic strain at regions in the vessel
ple, should have considerable value in the Charpy impact properties. where mechanical stress concentra-
enhancing the fracture capability of pres- The fracture toughness data for uni- tions exist, either because they are in-
sure vessels. However, there is some formly prestrained steels showed a mod-
indication that the beneficial effects may erate decreasing trend in Kw for speci- troduced by normal fabrication proce-
be lost under actual vessel operational mens that had been strained and aged. dure or because vessel design incorpo-
conditions such as with prolonged ther- Specimens strained at a fatigue cracked rated such stress concentrations. In
mal exposure. During the warm pre- notch were increased in toughness by the some cases this procedure, designed
stressing (overstressing) process, plastic- strain, but the benefits of this prestress- specifically to produce such strains,
strains occur at regions in the vessel ing were degraded by subsequent aging has been called prestressing or over-
where high stress concentrations exist. treatments. High levels of notch strain stressing. In order to ensure against
When the material is subsequently ex- produced Kic values that were signifi- brittle fracture during the overstress-
posed to elevated operating temperature, cantly higher than the base metal Kic ing cycle, this is often done at tem-
these regions become embrittled by the values. The Km values resulting from
process of strain aging. This embrittle- peratures above ambient and thus is
the high level notch strain followed by
ment may be enough to trigger a catas- an aging treatment were decreased by called "warm prestressing."
trophic failure in the pressure vessel. the aging process. The Kic values result- Although warm prestressing and
Hence, an investigation was undertaken ing from lower notch strain followed by proof testing are similar insofar as
to study the effects of strain aging em- an aging treatment, were not above the executional features are concerned,
brittlement, arising from overstressing level of the base metal and were, in some the basic effect utilized and the end
techniques, on two pressure vessel steels: cases, below it. Finally, both uniform
A533B and A516 grade 70. and notch prestrain studies indicated that result desired are quite different.
stress relief treatments, following strain- In Yukawa"s recent paper, 1 he de-
Charpy impact tests and tension tests. fines warm prestressing as, "a proce-
in conjunction with fracture toughness ing and/or straining and aging treat-
tests, were peformed to determine the ments, tended to decrease the K,t for dure of subjecting the vessel to a pre-
relative degree of material damage both steels rather than improve it. load or prestress under conditions
caused by strain aging. Different levels of where its fracture resistance is inher-
strain and different cycles of straining ently high for the purpose of increas-
and aging were studied. Different aging Introduction ing the fracture resistance under sub-
temperatures were also studied, and a This investigation concerns the sequent conditions where the fracture
stress relief treatment was investigated. strain aging embrittlement of pressure resistance would otherwise be low.
The results of the Charpy impact and vessel steels as a result of overstressing More specifically the procedure usual-
tension test data indicated that strain techniques followed by prolonged ly involves prestressing at some higher
aging embrittlement occurs in A516 thermal exposure. Since one of the temperature to improve the fracture
grade 70 steel and A533B steel but is resistance at a lower temperature."
more severe for the A516 grade 70 than primary requirements of pressure ves-
for the A533B. For both steels, repeated sel operation is the prevention of un- He defines proof testing as, "a
straining and aging treatments were expected catastrophic failure, vessels procedure of subjecting the vessel to a
noted to be more severe than single cycle are carefully designed, fabricated, and stress level higher than the design or
strain age treatments. Different aging nondestructively tested. Also, after the operating conditions for the purpose
temperatures proved that 400° F had vessel has been fabricated it is re- of ascertaining the lower bound of the
lesser effect on the A516 grade 70 steel quired to undergo a proof test. This ultimate fracture capacity. Alterna-
than did 500 and 650° F. For the A533B test usually consists of hydrostatic tively, the purpose may be stated as
steel aging at 400° F produced negligible pressurization to 1.5 times the design ascertaining the absence of a flaw
pressure, as specified by Section VIII larger than a certain calculated size."
L. N. SUCCOP is a former research as- of the American Society of Mechan-
sistant, A. W. PENSE is Associate Pro- In both procedures, pressurization
fessor and R. D. STOUT is Dean of the ical Engineering (ASME) Boiler and is the most feasible method of stress-
Graduate School. Lehigh University, Beth- Pressure Vessel Code. The hydrostatic ing the vessel. Using Fig. 1, Yukawa
lehem, Pa. test is necessary to provide evidence explains the two procedures in greater
Paper presented at the AWS 51st Annual that the vessel will withstand operat- detail. The three fracture strength
Meeting held in Cleveland, Ohio, during ing pressure.
June 8-12. 1970. curves show schematically the influ-

354-s I AUGUST 1970

ence of the two important factors of residual compressive stresses there information. It should also be noted
flaw size and temperature. For the when the structure is unloaded. that proof testing, if done at higher
kinds of steels used in pressure ves- 3. "Blunting" the crack tip which temperatures, does at the same time
sels, the general trends shown in Fig. reduces its sharpness or severity. constitute a warm prestress. Yukawa
1 have been established by a large As far as proof testing is concerned points out that the effectiveness of
number of experimental studies. Yukawa explains that this procedure both procedures applies only to the
The theory of warm prestressing involves subjecting the vessel to stress extent that the sense and pattern of
assumes the existence of a "large" level, o-o (Fig. 1) at some designated stressing reproduces those occurring in
flaw. As shown in Fig. 1, the fracture temperature such as Tu T2 or T 3 . If operational service, i.e., it is very diffi-
strength associated with this large flaw the test is carried out at temperature cult by pressure alone to attain
is relatively low at some specific low Tl and the vessel successfully with- stresses that match in magnitude and
temperature, 7,. The objective of stands a stress level o-o, the absence of sense all of the operational stresses.
warm prestressing is to produce an a "large" size flaw can be inferred. It Both Yukawa and Nichols have
increase in this low fracture strength does, however, leave open the possible considered some of the limitations of
by subjecting the structure to stress &., existence of "medium" and "small" the procedures arising from unknown
at temperature T2 or higher. At the size flaws. If the proof testing to stress factors. These factors are:
same time care must be taken to o-2 is done at temperatures T2 or T 3 , 1. The strain aging sensitivity of
perform the warm prestressing at the absence of even a large flaw can- current pressure vessel steels.
conditions that ensure no possibility of not be inferred. Also with this latter 2. The effect of prolonged thermal
fracture during the prestressing. The situation no information could be ob- and radiation exposure on the retained
basic premise of this procedure is that tained about the ability to withstand effectiveness of warm prestressing.
after an excursion to stress o-., at T., stress o-2 at temperature 7',. Another 3. The response to prestressing of
or higher, the structure containing the consideration of proof testing is that radiation embrittled steels. The latter
large size flaw can ideally withstand the proof testing conditions, similar to two factors obviously apply to the use
any applied stress up to o-2 at temper- the warm prestressing conditions, of vessels for nuclear reactor service,
ature T,. Even if this ideal amount of necessarily involve a compromise be- but the possibility of strain aging will
enhancement is not obtained, the pre- tween being sufficiently discriminating apply to all vessels that are strained
mise is that some substantial increase for flaws and avoiding actual vessel and subsequently placed in elevated
in ultimate fracture capability is ac- failure. temperature service.
complished. In evaluating warm prestressing and As indicated above, the plastic
Nichols- points out that these ben- proof testing procedures, Yukawa strain produced during the overstress-
efits of warm prestressing arise in at concludes that beneficial effects for ing combined with subsequent ele-
least three ways, all of which are a warm prestressing have been demon- vated temperature exposure can lead
consequence of local plastic deforma- strated in a number of laboratory tests to strain aging embrittlement, particu-
tion: and, in principle, should have consid- larly at the stress concentration loca-
1. Work hardening the material at erable value in enhancing the fracture tions in a vessel. This embrittlement is
the tip of the crack, which leads to an capability of pressure vessels. On the potentially dangerous since such loca-
increase in yield strength. other hand, proof testing is basically tions will always be at higher than
2. Producing local yield at the an inspection technique that provides working stress and could conceivably
crack tip which introduces favorable only marginally small increments of trigger a failure at pressures below the

'SMALL' A516 Grade 7 0

T2 a - Tested as Norma I ized

b -Tested after1.23kStrain
Fig. 1—Schematic representation of proof testing and warm
prestressing c -Tested after1.2f35(>Strain
Followed byICOOhr at
2 0 — 1.23V-


0 3 4
Fig. 2 (right)—The response of A516 Grade 70 to strain aging Strain-%


discontinuous yielding behavior re-
Table 1—Chemical Analysis of the Steels, %
turns and the stress strain curve fol-
Steel Grade C Mn P S Si Ni Cr Mo Cu Al lows a curve such as (c) in Fig. 2.
The yield point is now higher than the
A533 B .18 1.25 .024 .025 .24 .52 .13 .51 .29 .024
flow stress at the end of prestraining.
A516 70 .23 .97 .010 .020 .25 .09 .05 .02 .25 .025
This increase in yield or flow stress on
unloading and aging is the most uni-
versal indication of strain aging. There
proof stress level, thus eliminating the four categories: may also be an increase in ultimate
beneficial effects of prestressing. tensile strength and a decrease in
(a) Elements which interact weakly
Basically, strain aging is an embrit- elongation and reduction of area, but
or not at all with nitrogen and
tling phenomenon that takes place af- these do not always take place.
carbon—Cu, Ni, Mn, P.
ter or during plastic strain. Baird:: The effect of strain aging on the
(b) Nitride formers—Al, Si, B.
reports that the strain aging process tensile and impact properties of steels
(c) Carbide formers—Mo.
consists of two parts: has been extensively studied by
(d) Nitride and carbide formers—
1. The formation of solute "atmo- P V R C and others."' The PVRC in-
Cr, V, Nb, Ti.
spheres" around dislocations produced vestigations have shown that both car-
by straining. With respect to the alloying ele- bon and alloy pressure vessel steels
2. The formation of rod-like precip- ments listed above, it becomes appar- have a tendency for strain aging in the
itate particles centered on the disloca- ent that killed steels containing alumi- 500-700° F range if given a tensile
tion core. num and/or those containing strong prestrain of 5 % . For the carbon
At the present time, strain aging is carbide (and nitride) formers such as steels, the maximum aging response
considered to be due primarily to the Ti, V, Nb and to a lesser extent Cr, occurred at 500° F while for alloy
migration of carbon and nitrogen will be less susceptible to strain aging steels it was closer to 700° F. The
atoms to dislocations and locking phenomena, while semikilled steels aging was characterized by an in-
them. Other elements are also impor- may be more so. Previous PVRC in- crease in yield and, to a lesser extent,
tant and Baird:i summarizes all ele- vestigations4 of strain aging have tensile strength of the steel and a
ments affecting strain aging into two shown this to be generally true. Ex- decrease in tensile ductility. The Char-
groups: perimentally, the general effect of py impact transition temperature in-
1. The solutes which can lock dislo- strain aging on the tensile properties creased substantially as a result of
cations and which can diffuse suffi- of A516 grade 70 is illustrated in Fig. aging. The effect of strain aging on
ciently quickly to produce strain aging 2. If the specimen is strained to a the fracture toughness of carbon and
(CandN). point well into or beyond the lower low alloy steels that were warm pre-
yield extension, unloaded and then stressed, then aged, has been studied
2. The elements which themselves
immediately retested, the stress-strain by Harrison and Fearnehough"1 and by
do not produce strain aging but affect
curve follows the same curve. Howev- Hawthorne and Loss." These results
the process by altering the concentra-
er, if the specimen is unloaded after show that although high levels of ten-
tion or mobility of the solute atoms
plastic strain and then allowed to age sile prestrain raised the toughness of
that produce strain aging. These ele- the specimens, a 25% loss of the
ments can themselves be divided into at room temperature or above, the

w ^ y

• * t L^

L ^ -f^h-^r

£3B> tf=?

Fig. 3—Speci.nen orientations in A516 Grade 70 piate Pig, 4—Specimen orientations in A533 Grade B plate

356-s AUGUST 1970

Table 2—Tensile Test Data

A516 grade 70 s t e e l - A533B steel (heavy section)

0.2% Ultimate Reduc- 0.2% Ultimate Reduc-
offset tensile Elon- tion in offset tensile Elon- tion in
yield, strength, gation, area, yield, strength, gation. Area,
ksi ksi % % Base properties ksi ksi % %
48.4 76.6 34.0 66.6 72.5 90.7 24.5 57.1
51.5 78.1 31.0 64.5 Strained 1.25% at 200° F 75.1 90.6 24.9 56.4
56.9 80.7 33.5 64.8 Strained 1.25% at 200° F-aged at 79.2 89.9 19.4 52.4
650° F for 1000 hrs 1 cycle
67.5 84.3 27.0 61.0 Strained 1.25% at 200° F-aged at 650° F 90.5 96.9 17.0 53.7
for 1000 hr 3 cycle
55.6 79.9 31.7 67.5 Strained 1.25% at 200° F hen aged 79.8 89.9 19.5 55.2
at 500° F for 1000 hr 1 cycle
85.6 91.2 24.2 61.3 Strained 1.25% at 200° F then aged 92.4 97.2 18.5 59.3
at 500° F for 1000 hr 3 cycle
58.4 80.0 31.3 65.6 Strained 1.25% at 200° F then aged 77.9 86.2 24.2 58.3
at 400° Ffor 1000 hr 1 cycle

Table 3—Charpy V-Notch Impact Test Data

-A533B steel (heavy section)-
. Transition temperature (°F) a t — •. Shelf —Transition temperature (°F) at , Shelf
50% energy, 50% energy,
15 ft-lb 20 ft-lb 15 mils shear ft-lb Condition 15 ft-lb 30 ft-lb 15 mils shear ft-lb
#1 - 8 3 -68 -85 +30 83
#2 - 7 4 -57 -89 +33 81.5 Base properties
#3 - 8 4 -63 -90 +35 77 -14 + 12 -20 +60 46
-60 -41 -68 +46 82.5 Strained 1.25% at 200° F -11 + 64 0 +64 43
-45 -33 -46 +56 65 Strained 1.25% at 200° F then aged +20 + 87 + 6 +68
at 650° F for 1000 hr 1 cycle 40
+ 2 +20 - 6 +56 72 Strained 1.25% at 200° F then aged +40 +116 +30 +77
at 650° F for 1000 hr 3 cycles 36
-47 -32 -45 +43 76 Strained 1.25% at 200° F then aged +28 + 75 - 4 +58
at 500° Ffor 1000 hr 1 cycle 40
+16 +37 +20 +69 62 Strained 1.25% at 200° F then aged +20 +100 +22 +76
at 500° Ffor 1000 hr 3 cycles 48
-50 -31 -46 +48 72 Strained 1.25% at 200° F then aged -14 + 52 -40 +56
at 400° Ffor 1000 hr 1 cycle 52
-43 -28 -51 +56 85 Strained 1.25% at 200° F then aged -38 + 25 -51 +48
at 650° F for 1000 hr 1 cycle stress re- 63
lief at 1100° F for 8 hr furnace cool
50° F/hr
-33 -18 -36 +60 85 Strained 1.25% at 200° F then aged -36 + 33 -36 +62 61
at 650° F for 1000 hr 3 cycles stress re-
lief at 1100° F for 8 hr furnace cool
50° F/hr

prestressing effect resulted from a 500 reliability is particularly critical, and Vessel Steels Subcommittee, Materials
hr age at 550° F. When tensile and flaw growth is known to be likely Division, PVRC, undertook an experi-
then compressive prestrain is applied during operation, periodic proof test- mental program at Lehigh University
the toughness is not improved, and ing of the vessel has sometimes been to examine this response in several
subsequent aging at 482° F for V 2 hr part of the operating plan. Thus re- pressure vessel steels. The objectives
reduces the toughness below the origi- peated cycles of plastic straining and of the program were to determine
nal plate level. service are anticipated for such ves- what tensile prestrain followed by rep-
As may be seen from the above sels. resentative elevated temperature (ag-
discussion, the use of warm prestress- The factors that counterbalance the ing) cycles would do to the material
ing has many advantages. It will ac- advantages are metallurgical ones— itself, and also what influence pre-
complish not only a proof test but will the response of the material to radia- strain and aging would have on the
also serve as a means of brittle frac- tion damage and to aging phenomena fracture behavior of the same materi-
ture control. Indeed, in cases where induced by the prestraining. If either al with a controlled flaw present dur-
thermal stress relieving of a comoleted are severe, the potential advantages of ing the prestrain. As a corollary study,
vessel is not economical or practical, prestraining will be either negated or the influence of thermal stress relief
it has been suggested2 that mechani- perhaps overshadowed by the poten- on strain aged material, with and
cal overstressing should be used in tial danger of reduced fracture resis- without flaws, was also examined.
place of thermal stress relief. Even in tance.
cases where overstressing does not Since the metallurgical response of
replace thermal-stress relief, tradition- Method and Procedure
pressure vessel steels to warm pre-
al proof testing at levels high enough straining in the presence of flaws has The two steels investigated in this
to produce plastic yielding will result not been given sufficient attention to project were A533B and A516 grade
in the same effect in terms of subse- clarify the extent to which material 70. Table 1 shows the chemical com-
quent aging behavior. When a pres- degradation occurs under conditions positions of these materials.
sure vessel is placed in service where of prestraining and aging, the Pressure The A533B steel was sectioned into


mens is listed in Table 4.
Table 4—Summary of Experimental Program
Uniform Prestrain Study

Aged 1000 hr
For the uniform strain part of the
at 400° F study the blanks, ranging in size from
•Vs x 1V 2 x 18 in. to l ' / 4 x l 3 / 4 x
Uni- Strained Aged 1000 hr Strained 1.25% at 200° F, aged 1000 27 in., were uniformly strained 1.2-
form 1.25% at at 500° F hr at 500° F plus strained 1.25% at 1.5% at 200° F in a standard univer-
strain 200° F 200° F, aged 1000 hr at 500° F sal testing machine. An environmental
tests testing chamber, using forced air, was
Aged 1000 hr Strained 1.25% at 200° F, aged 1000 used to heat the specimens. The defor-
at 650° F' hr at 650° F plus strained 1.25% at mation during the straining was moni-
200° F, aged 1000 h rat 650° F>
tored by a dial gage extensometer at-
Base tached to the outside of the chamber.
con- Low notch Aged 400 hr
dition strain at at 650° F» Following the straining, some of the
200° F a b blanks were immediately machined
Aged 1000 hr Low notch strain at 200° F, aged 1000 into Charpy and tensile specimens,
at 650° F" hr at 650° F plus low notch strain at while others were aged for 1000 hr at
200° F, aged 1000 hr at 650° F" 650, 500, and 400° F. After the single
Notch strain-age cycle, some of the bars
strain Medium Aged 1000 hr aged at 650° F were machined into
tests notch at 650° F
strain at
standard Charpy, tensile, and slow
200° Fa'« bend fracture toughness specimens.
The others were then given two more
High notch Aged 1000 hr strain-age cycles before they were
strain at at 650° F made into the same three types of
200° Fa' A specimens. For materials aged at 500
and 400° F, only Charpy and tensile
• Data obtained both before and after thermal stress relief, 8-12 hr at 1100° F. specimens were investigated. The 500
Prestrain K of 38 ksi V i n . for A516 grade 70, 40 ksi in. for A533B—5% plastic zone size. and 650° F aging temperatures alone
' Prestrain K of 63 ksi V i n . for A533B—15% plastic zone size. were investigated for the three-cycle-
'' Prestrain K of 88 ksi V i n . for A516 grade 70, 77 ksi X i n . for A533B—25% plastic zone size strain-age condition.
3 The Charpy and tension specimens
/ 4 - l 1 / 2 in. thick slices from the 7V 2 3) in order to save material. For the were tested on a standard impact test-
in. thick plate as shown in Fig. 3. A516 grade 70 steel, the specimens ing machine and a standard 10,000 lb
The slices were then given a heat were cut parallel to the rolling direc- capacity universal testing machine, re-
treatment to match the thermal his- tion (shown in Fig. 4) from center, spectively. Tension data for A516
tory of the center of the quenched, surface and quarter thickness positions grade 70 steel over a range of temper-
tempered and stress relieved plate 6 of the 2 in. thick plate. Mechanical atures from —250 to +200° F were
in. thick. 7 tests, as shown in Table 3 indicated also obtained.
The A516 grade 70 steel (2 in that the A516 grade 70 plate was rela-
tively mechanically uniform through The slow bend fracture toughness
thick) needed no additional heat specimens for the uniform strain part
treatment and was tested in the nor- its thickness. Hence, it was decided
that specimens cut from any of the of the study, seen in Fig. 5, ranged in
malized condition. size from 0.60 x 1.20 x 10 in. to 0.8 x
Standard Charpy impact tests and three positions would give satisfactory
evaluation of the mechanical proper- 1.50 x 10 in. After being cut from the
tensile tests were then performed on blanks and machined they were then
both steels to determine the base ties of a 2 in. thick plate.
precracked at 286-300 cycles/sec.
mechanical properties. These results After the base mechanical proper- A three point bend loading arrange-
are shown in Tables 2 and 3. For the ties of the steels were determined, ment was used to grow fatigue cracks
first material, A533B, the 1-1V 4 in. uniform and notch prestrain specimen that varied from 0.45 to .055 times
slices were of such a shape that the blanks were machined from the plates the depth of the specimen, W. The
specimens had to be cut transverse to as shown in Figs. 3 and 4. The strain- chevron notch configuration, shown in
the rolling direction (as shown in Fig. ing and testing schedule for the speci- Fig. 5 was used in order to form a
fatigue crack which did not deviate
from the notch plane and yet extend-
< Q = ^ 3^ f ed substantially beyond the notch root
BW /o (w) throughout most of the specimen
thickness. This substantial extension
was necessary to avoid undue influ-
ence of the notch on the crack stress
field. The chevron notch also expe-
dited nucleation of the fatigue crack. 8
In all the specimens, the cracks were
grown in two steps. The first step grew
the crack beyond the notch root at a
SECTION fast growth rate. The second step grew
THROUGH the crack at a slow rate so that the
THICKNESS- B NOTCH final .050 in. was grown in 50,000
DEPTH - W cycles or more.
SPAN L E N G T H - L Following precracking, all the slow
Fig. 5—Slow bend specimen bend specimens were loaded in a three

358-s AUGUST 1970

point bend fixture, at approximately a 5% of the remaining uncracked sec-
60-120 ksi Vin. per min loading rate tion of the specimen, resulted. The
— Fig. 6. A double cantilever clip-in second level of tensile prestrain was
displacement type gage was inserted higher, designed to simulate the be-
into attachable knife edges already on havior of a larger flaw in a vessel.
the specimen. The displacement out- This produced a plastic zone size of
put from the clip gage, along with the about 15% and was used for A533
load output from the load cell, was grade B only. The third level, rep-
fed into an x-y recorder. Different test resenting a high stress intensity such
temperatures from -250 to -150° F as might occur at the tip of a large
were obtained with liquid nitrogen deep flaw, produced a plastic zone size
applied through perforated copper of about 25% of the remaining sec-
coils surrounding the specimen in a tion.
cooling box insulated with 2 in. thick For the notch prestrain part of the
styrofoam. The temperature was mon- study, the slow bend specimens ranged
itored by means of a thermocouple in size from 1.0 x 2.0 x 18.0 in. for
attached to the center thickness of the A533B to 2.0 x 4.0 x 18.0 in. for
specimen. The specimen was held at A516 Grade 70. After being cut from
temperature for 15 min before test- the plates and machined as shown in
ing, and a minimum of 2 specimens Figs. 4 and 5, they were precracked,
were tested for each condition. Fig. 6—Slow bend test apparatus: 1— as described before, on a fatigue test-
load cell; 2—specimen; 3—clip gage; ing machine. All the specimens, in the
Notch Prestrain Study 4—knife edges; 5—bend fixture; 6—X-Y configuration shown in Fig. 8 were
In developing a test program to recorder heated to 200° F, slowly loaded in
study warm prestressing, it soon be- tension to a stress level based on the
came apparent that uniform straining specimen size, and held for five min-
of a small flaw free specimen could bend and the compact tension fracture utes. As described above, prestress
not duplicate the mechanical effects toughness specimens. The basic levels of 38 and 88 ksi Vin. were
that exist in the region at the tip of a procedure was to machine and fatigue studied for A516 grade 70, while 40,
sharp crack in a thick walled pressure crack these specimens as in the uni- 63 and 77 ksi Vin. were used for
vessel. The stress state conditions rep- form strain study, and then to treat A533 B. The prestrain K level was
resented by this situation may, indeed, them as a section from the membrane calculated on the basis of the preload
never be duplicated satisfactorily in of a vessel given a prestressing treat- and the crack length measured on the
any laboratory test. However, a simu- ment. This was accomplished by giv- surface of the specimen after fatigue
lation of this condition could be at- ing a tensile prestrain to the specimen cracking. The formula used was that
tempted by studying the behavior of a and then applying the aging and/or recommended for single-edged cracked
specimen which contained a sharp stress relief treatments which were to plates in tension in ASTM STP 410: 9
natural flaw. The influence of pre- be studied. The tensile prestrain
strain and aging on this specimen, in created a plastic zone at the tip of the
fatigue crack which simulated, at least pall2
conjunction with more traditional K (applied) Y-tfL- (i)
tests, could then be used to clarify the to some degree, that associated with a
flaw in a vessel. BW '
relative role of those effects, such as
response to aging and work harden- Three levels of tensile prestrain where K = applied stress intensity fac-
ing, which are primarily metallurgical were employed in the investigation. tor ksi V i n . ; P = preload (lbs.);
in nature and those, such as crack The first, representing a condition of a a = crack length after precracking (in.);
blunting and residual stress state, small flaw in the wall of a thick vessel, B = thickness (in.); W = width (in);
which are primarily mechanical in was applied at a prestrain only great Y = calibration factor for specimen
origin. enough to produce a relatively low geometry.
The test specimens used to dupli- stress intensity at the fatigue crack The corresponding plastic zone sizes
cate such a condition were the slow tip. A plastic zone of small size, about (ry) were calculated from:

.25W DIA KQ =
BW/2 ® © T o
T H I C K N E S S IS 2"


y "2TT\py
2 r v =DIA.of P L A S T I C
w-1.25 ^

Fig. 7—Compact tension specimen Fig. 8—Prestraining technique for the specimens


of Kic on testing were invalid by
Table 5--Low Temperature Tension Test Data (- -250° F to +200° F) specimen size criteria. All load-
70 steel- , — A533Bs1 displacement records were linear or
Reduc- Reduc- nearly so and all specimens failed
0.2% Ultimate tion 0.2% Ultimate tion before a 5% secant modulus line was
offset tensile Elon- in Temper- offset tensile Elon- in reached. 8 Indeed, low testing temper-
yield, strength, gation, area, ature, yield, strength, gation, area, atures were employed to ensure that
ksi ksi % % °F ksi ksi % % this would be so.
96.6 118.0 35.3 53.6 -250° F 114.1 124.5 21.0 56.7 In another sense, none of the notch
68.0 98.7 39.0 55.7 -150° F 91.6 111.4 20.3 60.7 prestrained specimens could be con-
56.9 88.5 37.0 61.8 - 50° F — — — — sidered as "valid." This is because,
54.0 87.1 36.2 61.2 0° F — — — — after careful precracking in fatigue,
48.4 76.6 35.0 66.6 + 70° F 71.4 91.4 18.0 65.5
36.2 68.7 +200° F 67.6 86.6 17.0 67.5 all of the specimens except the base
47.5 75.0
condition ones were deliberately given
* From Mager and Thomas. 1 treatments that were designed to sim-
ulate strained flaws in vessels. As a
the broken ends of the slow bend result, each was given a strain that
ry (2) fracture toughness specimens. would introduce a substantial but con-
2JT \<ry)
Only specimens tested in the base trolled plastic zone at the crack tip.
Following the prestress, some of the condition (without prestrain or aging) Because of this large plastic zone,
specimens were plane strain fracture were used for specimen blanks and the none of these tests could be consid-
toughness tested as described before portions near the fracture surfaces ered as normal fracture toughness
and the others were aged for 400 and were not utilized. These specimens tests.
1000 hr at 650° F. After the notch were 2.0 x 1.0 x 2.087 in. for A533B
prestrain-1000 hr age cycle some of and 4.0 x 2.0 x 4.175 in. for A516
the specimens were tested while the grade 70. The specimens were pre- Results and Discussion
remaining ones were given two more strained as shown in Fig. 7 and were Charpy V-Notch Impact Test Data
notch prestrain-age cycles of 1000 hr tested in tension. Data on the condi- The results of the Charpy V-notch
aging each before they were finally tion of stress relief after prestrain and impact tests are presented in Table 3.
tested. The prestrain K level used dur- after aging was obtained by fabricat- It should always be kept in mind when
ing the second and third cycle was the ing specimens, giving them the prior viewing these data that the A533B has
same as applied previously. As before, treatment to be studied, and then been given a heavy section heat treat-
a minimum of two specimens were stress relieving them at 1100° F for 8 ment simulating a 6 in. thick plate,
tested for each condition. or 12 hr. The stress relief was fol- while the A516 grade 70 steel is 2 in
After all the specimens had been lowed by a furnace cool (approximate- thick.
machined and after a large number ly 50° F / h r ) . Both Charpy impact The base properties of the A533B
had been tested, it was decided that tests and fracture toughness tests were steel indicate that this material has
some additional conditions, particular- performed on these specimens. The generally less impact toughness than
ly stress relief treatments, should be A533B specimens were stress relieved the A516 grade 70 steel and a low
investigated. Since most of the test 12 hr, the A516 grade 70, 8 hr. tipper shelf energy. However, this re-
material available had now been used Some special comments need to be sult is expected from the heavy section
for the slow bend tests, in order to made with regard to the validity of heat treatment given to the A533B
undertake these tests it was necessary the fracture toughness test data. In plates and the fact that they represent
to modify the specimen design in or- one sense, almost all of the fracture transverse not longitudinal properties.
der to conserve material. Therefore, toughness test data may be regarded The plate used in this program was
additional fracture toughness speci- as valid Kic information. Only speci- not made to A533B specifications but
mens were fabricated of the compact mens which received very high pre- was quenched and tempered nickel
tension type, shown in Fig. 7, from strain values and thus had high values modified A302 grade B steel. It prob-

A 516 GRADE 70 A 533 GRADE B

TESTED AT -250°F TESTED AT - 2 0 0 ° F
55 .' 50 •


45 O O
o [E 40 O
o "
40 ' R
o :
35 8 1 T'
y 30
30 O o o
• 25
• 20
20 '

A 3 C Bs» CSR A 3 C CSR A B
1000 HR. 1000 HR. RELIEVED RELIEVED 1000 HR. 1000 HR. RELIEVED 1000 HR.
3CYCLES 1100'F 12HR. 1100eF 12 HP. 3 CYCLES 1100°F12HR.

Fig. 9—Influence of uniform straining, aging and stress relief Fig. 10—Influence of uniform straining, aging and stress relief
on the fracture toughness of A516 Grade 70 steel on the fracture toughness of A533 Grade B steel

360-s i A U G U S T 1970
A 516 GRADE 70 A 533 GRADE B
STRAINED AT 200°F - AGED AT 6 5 0 ° F TESTED AT -150°F
90 o
100' O
80- o o
_ 80 o
_ 60
* 70 o
o o
£ 60 o
9 * 40- o
50- o o 30-
40- 8 20-
(5-/.ZONE) 400HR. 1000 HR 3 CYCLES (25-/.ZONE 1000 HR. (SV.ZONE) 1000 HR. 3CYCLES (BV.ZONE) 1000 HR (25-/.ZONE 1000HR
1000 HR, 1000 HR.

Fig. 11—Influence of notch straining and aging on the frac- Fig. 12—Influence of notch straining and aging on the frac-
ture toughness of A516 Grade 70 steel ture toughness of A533 Grade B steel

A516 GRADE 70
110 TESTED AT -150 F
O 110
• 100 '
90 O
90 O •

RO • , 80 ° o
o • •
Z 70 ^70
in O .
* 60 • •
60 • . o •
0 t o * 50 o • o •
* 50 o o R
40 o o
40 8 o • • '
30 •

30 •
20 "
' . : : -

1100 *F 1100 °F HR30CLES 1100°F 1100" F

Fig. 13—The influence of notch strain, aging, and stress Fig. 14—The influence of notch strain, aging, and stress relief
relief on the fracture toughness of A516 Grade 70 on the fracture toughness of A533 Grade B

ably does not wholly represent the temperatures increased further and served for the A533B steel. For the
melting or manufacturing practice for the shelf energies decreased. For the A516 grade 70 steel, the 15 ft-lb, 20
current A533 B heats. A516 grade 70 steel the 15 ft-lb, 20 ft-lb and 15 mil transitions increased
The as-strained condition for the ft-lb, and 15 mil transition tempera- approximately 50° F over the single
A516 grade 70 steel showed an in- tures increased approximately 35° F cycle strain age level. The 50% frac-
crease of approximately 20° F for above the base property level. The ture appearance transition remained
both the 15 ft-lb and 20 ft-lb transi- 50% fracture appearance transition the same and the shelf energy slightly
tion temperatures as well as for the 15 temperature increased 20° F above its increased to 72 ft-lb. For the A533B
mil (lateral expansion) transition tem- base level and the shelf energy was steel the 15 ft-lb, 30 ft-lb, and 15 mil
perature. The 50% fracture appear- noted to fall 10-15% below its origi- transitions increased about 25° F over
ance transition temperature increased nal level. For the A533B steel the 15 the single cycle strain age level. The
slightly and the shelf energy also in- ft-lb and 15 mil transition tempera- 50% fracture appearance transition
creased slightly compared to the base tures increased approximately 30° F increased slightly to 77° F, while the
properties. The A533B material above their base levels while the 30 shelf energy decreased to 36 ft-lb.
showed a similar response to this con- ft-lb transition was observed to in- The behavior exhibited by both steels
dition, except to a lesser degree. It crease 75° F above its base level indicates that repeated straining and
may be seen that the 30 ft-lb transi- (recall that 52° F of this increase was aging is more severe than a single
tion temperature increased 52° F. noted in the as-strained condition). cycle treatment.
This response is regarded as large; The 50% fracture appearance transi- As far as aging temperature is con-
however, it may also be noted that the tion for this steel increased only 8° F cerned, 400° F produced lesser aging
shelf energy decreased slightly which above its original level while the shelf in the A516 grade 70 steel and none
shifts the upper end of the data curve energy decreased about 12%. at all in the A533B. For the 500 and
to lower values and thus contributes Upon straining and aging both ma- 650° F temperatures, little difference
to a larger change in the higher ener- terials 3 cycles at 650° F, substantial in respective embrittlement was
gy transition temperatures. increases in transition temperatures noted.
When both steels were strained and were noted for the A516 grade 70 The results of the stress relief treat-
aged 1 cycle at 650° F, the transition steel while lesser increases were ob- ment showed general improvement in

toughness over the aged condition. aging at all in the A533B material. at 200°F. As a result, the plastic zone
For the A516 grade 70 steel the single The tensile data shown in Table 5 resulting from the prestrain is the
cycle strain age treatment followed by are results that were obtained for use same as or greater than that at fail-
the stress relief treatment showed no in validating fracture toughness tests. ure.
improvement (no decrease) in the The A533B data were obtained from Beyond 1000 hr of aging, repeated
material transition temperatures, but another investigation 10 and are repre- straining and aging did not produce a
the shelf energy increased back to its sentative of this steel heat treated to continued loss in toughness. Within
base level. However, for the repeated the yield and tensile strength used in the limitations imposed by test scatter
straining and aging condition followed this investigation. it appears that the final condition of
by the stress relief treatment, the re- aging was to below the original tough-
Fracture Toughness Test Data
sults indicate a substantial improve- ness level of the plate. These results
ment in the transition temperatures. The results of the fracture tough- confirm the uniform strain test data,
For the A533B steel similar behavior ness tests are shown in Figs. 9-14. It but also demonstrate that the loss in
was noted, indeed, the stress relief should always be kept in mind when toughness in the notch prestrained
treatment improved the properties to viewing these data that both the uni- condition is less than in the uniform
the degree that they were better than form and notch prestraining was per- tests.
the base condition. For both steels formed at 200° F while the fracture The results of tests on material
these results indicate that stress reliev- testing was conducted at or below- tested in the condition of higher levels
ing results in improved toughness over 150°F. of notch prestrain (63, 77 and 88 ksi
the aged condition. It should also be Vin.) followed by aging are also
Uniform Prestrain and Aging Results
noted that after stress relieving treat- shown on Figs. 11 and 12. It is clear
ments were given to the 1-cycle and The uniform prestrain K,c results that the higher levels of warm pre-
3-cycle strain aging conditions, the for the A533B and A516 grade 70 stressing did produce improvement in
transition temperatures of both steels steels are shown graphically in Figs. 9 fracture toughness, but that this
were approximately the same despite and 10. The steels showed trends for a toughness improvement is to a large
the original difference in base proper- decrease in K„ for increasing extent lost through subsequent aging.
ties. The A516 grade 70 steel, howev- amounts of straining and aging. For The aging loss left a residual improve-
er, still maintained a higher shelf en- the A533B the single cycle strained ment in toughness over the base con-
ergy. and aged condition reduced the aver- dition of from 20 to 6 5 % , as com-
age toughness approximately 10%. pared to almost 100% increase pro-
Tensile Test Data For the A516 grade 70 steel, the same duced by the warm prestressing alone.
A summary of the tensile test data treatment lowered the average KIC These experimental data confirm the
for the base conditions and the differ- value about 2 0 % . For the triple cycle results of Hawthorne and Loss, 6 and
ent straining and aging cycles is straining and aging condition, the further suggest that strain aging can
presented in Table 3. In addition, A533B indicated a further drop to a negate much of the benefit of even
tensile data over a range of tempera- value 16% below the original plate,
high levels of prestrain.
tures for both steels are presented in while the A516 grade 70 steel showed
a slight increase in toughness. This Higher levels of prestrain were not
Table 5. employed in this study and higher
For both steels, the yield strength, increase is small and the magnitude of
changes in toughness between the levels are probably not practical in
tensile strength, elongation, and re- actual practice. At the level 88 ksi
1-cycle and 3-cycle treatments may
duction of area data are typical of the Vin. slow crack growth occurred in
have been the result of test scatter. It
changes associated with strain aging.3 A516 grade 70 during prestraining.
should be noted for both steels that
As explained in the Introduction, Thus the A516 Grade 70 " H " speci-
the repeated straining and aging treat-
strain aging increases the yield ment did not affect Kic as significantly mens reported in Fig. 11 have already
strength while the tensile strength, as it did Charpy impact and tensile reached the limit on experimental
elongation, and reduction of area are results. The Kic test results indicate prestrain values without risk of failure
less affected. For the A516 grade 70 that the A533B steel is not as suscep- during prestraining.
steel the increase in yield strength tible to strain aging as the A516 grade
from the base condition to the strain 70 steel, which is consistent with the Thermal Stress Relief Results
age 3-cycle condition was approx- impact test data. The influence of thermal stress re-
imately 7 5 % . For the A533B steel, lief on both the uniformly strained
the same conditions showed about a and notch prestrained material pro-
Notch Prestrain and Aging Results
28% increase. This result is in agree- duced results that were both informa-
ment with the Charpy impact test data The notch prestrain and aging re- tive and puzzling. These results are
and once again suggests that the sults for both steels are found in Figs. seen on Figs. 9, 10, 13 and 14 where a
A533B steel is less susceptible to 1 1-14. For both steels, the base con- variety of strained and aged condi-
strain aging. Based on yield point dition tests show extensive scatter, and tions are plotted along with the
elevation, it appears that the repeated thus interpretation of the aging results equivalent material given a thermal
straining and aging condition pro- is less certain, particularly at the low- stress relief of from 8 to 12 hr at
duced the most embrittlement. This er prestrain levels. However, the influ- 1100° F. It may be seen by such a
result is also in agreement with the ence of the lower level prestrain comparison that, in almost every case,
impact data. (38-40 ksi Vin.) did appear to pro- thermal stress relief was not helpful in
The effect of different aging tem- duce a slight improvement in fracture improving toughness of either notch or
peratures was consistent with the im- toughness in the specimens which was uniformly prestrained steel and served
pact data. The 500 and 650° F aging subsequently lost by aging. Since the to decrease toughness when applied
temperatures produced no significant plastic zone produced by the prestrain directly to notched specimens after
variation in respective strain aging was small, the toughness im- prestrain and prior to aging. The frac-
behavior. The 400° F aging tempera- provement was limited. It should be ture surfaces of the highly prestrained
ture produced lesser aging in the remembered that, although the pre- material before and after thermal
A516 grade 70 steel than did the strain loads were less than the ob- stress relief were suggestive of this
served Kn-, the prestrain was applied toughness decrease as may be seen in
other temperatures, and hardly any

362-s i A U G U S T 1970

Spec f P
no. Q1 Q2
1 25.5 3 0 3
2 18.6 28.1

Fig. 15—A516 Grade 70 strained 88 ksi

Vin. at 200° F and tested at -150° F.
A—no stress relief B—stress relieved.
S—denotes slow crack growth during DISPLACEMENT
straining Fig. 16—Load vs. deflection curve and fracture surface for A533 Grade B
strained, aged and stress relieved

Figure 15. ployed were difficult to control, the K crack ran from a region of low tough-
Since the influence of prestrain is to level during fatigue cracking of these ness into a region of higher toughness
produce a large plastic zone at the specimens (and only these four speci- and was arrested. This behavior would
crack tip and thus induce residual mens) was higher than recommended. suggest that the conditions in the
compressive stress as well as blunt the It is possible that the fatigue cracking region at the crack tip were made
crack, it should not be surprising that procedure after stress relief raised the more severe by the stress relief treat-
stress relief decreases the effectiveness Kn level above that of B s u by in- ment.
of the prestrain by eliminating residu- ducing plastic strains at the crack tip.
al compressive stress. Moreover, as However, it may also be suggested Conclusions
long as some crack blunting remains, that the presence of the fatigue crack The results of this experimental in-
the toughness should still be higher itself, with the plastic zone that is vestigation may be summarized as fol-
than the base condition value. always associated with it, is a prior lows:
These are the results obtained from condition for this toughness loss. For 1. Charpy impact and tension test
these experiments. They suggest, at the A533B specimens, where fatigue data indicate that strain aging embrit-
least for the high levels of notch cracking was more carefully con- tlement occurs in A516 grade 70 steel
prestrain, that both crack blunting and trolled, the loss between C and C s u is and A533B steel, but is more severe
residual compressive stress do play a equally great. for the A516 grade 70 than for the
role in toughness improvement from Another interesting feature of this A533B.
warm prestressing. However, thermal unexpected result was the atypical 2. Charpy impact and tension test
stress relief was not helpful in improv- load-displacement records and frac- data indicate that strain aging embrit-
ing toughness after strain aging and, ture surfaces of the uniform prestrain tlement is more severe for the re-
in some cases, was distinctly harmful. specimens exhibiting these low Kic peated straining and aging condition
This is entirely contrary to the influ- numbers after stress relief. All four than for the single cycle strain age
ence of thermal stress relief on the specimens (2 of A533B and 2 of condition.
impact test specimens where stress A516 grade 70) showed load vs. dis- 3. The study of different aging tem-
relief was shown to improve tough- placement plots and fractured sur- peratures showed that 500 and 650° F
ness. faces such as the example in Fig. 16. produced almost the same response
The stress relief treatments when These results indicate that the crack while 400° F produced slightly lesser
applied to the uniformly prestrained experienced several cycles of popping response in the A516 grade 70 steel
A516 grade 70 steel produced differ- through a locally embrittled region and negligible response in the A533B
ing results, as may be seen by compar- and then being arrested. steel.
ing specimens B and B SIt and C and An effort was made to correlate the 4. Stress relief treatments following
C g K on Figs. 9 and 10. In this in- most distinct load points (p,,j, P q 2 , straining and aging treatments im-
p p witn tne m o s t
stance, specimen B S R had much lower q3' q4) distinct proved the Charpy impact toughness
toughness than specimen C S R , al- crack front advancements on the frac- of both steels.
though the strained and aged tough- tured surfaces (a^ a2, a3, a 4 ) . It was 5. The fracture toughness data for
nesses were nearly equivalent. The difficult to discern what were and uniformly prestrain steels showed a
difference in these two sets of speci- were not crack front advancements on moderate decreasing trend in Kw for
mens was the sequence of specimen the fractured surfaces; hence, only the specimens that had been strained and
preparation. Specimens B SIt were fa- first and second most distinct load aged.
tigue cracked prior to stress relief points were correlated with the first 6. The notch strain investigation
while specimens C SK were fatigue and second most distinct crack fronts showed that the benefits of warm
cracked after stress relief. respectively. The results of this corre- prestressing were degraded by subse-
Since these uniform strain speci- lation, as shown on Fig. 16, indicate quent aging treatment.
mens were small (0.60 in. thick for that the Kn2 values were higher than 7. High levels of notch strain at
A516 grade 70) and the loads em- the Kql values. This suggests that the 200° F followed by aging at 650° F


produced Kn values that were signifi- ects Task Group and the Pressure Vessel S t e e l s . " WELDING JOURNAL. 38(4).
Research Suppl. 182-s (1959).
cantly higher than the base plate Kic Vessel Steels Subcommittee of the 5. H a r r i s o n , T. C . and F e a r n e h o u g h . G.
values. The Ki< values resulting from Materials Division, the authors have D.. " T h e Influence of S t r a i n s due to Proof
Loading on Welding Embrittlement."
the high notch strain followed by an been given much valuable advice and Metal Construction and British Welding
aging treatment were still above the guidance in the course of the work Journal, Vol. 1. No. 10, (Oct. 1969). p . 476.
Kn- of the unstrained plate. The low and they appreciate very much this 6. H a w t h o r n e . J. R., and Loss. F . J..
"Influence of O p e r a t i n g E n v i r o n m e n t and
level of notch strain followed by aging assistance. Mr. Noel Huettich of Le- Procedures on Material B e h a v i o r : T h e Ef-
produced toughness values below the high University was particularly help- fects of Coupling Nuclear Radiation with
State and Cyclic Service Stresses and of
Kic level of the plate. ful in the experimental portions of the Periodic Proof T e s t i n g . " NRL 6620, U. S.
8. Both uniform and notched pre- work, and the authors wish to ac- Naval Research L a b . . A u g u s t 1967.
strain fracture toughness studies at knowledge his effort in behalf of the 7. S t r u n c k . S. S.. Pense, A. W.. and
Stout. R. D.. " T h e P r o p e r t i e s and Micro-
— 150° F indicated that thermal stress program. s t r u c t u r e of S p r a y Quenched Thick-Section
relief treatments, either after prestrain- S t e e l s . " Welding Research Council Bul-
References letin No. 120 ( F e b . 1967), p . 1.
ing or following aging, did not in- 8. Recommended Practice for Plane-
crease the toughness of the material 1. Y u k a w a , S., " E v a l u a t i o n of Periodic Strain Fracture Toughness Testing of High
tested. In some cases, the fracture Proof T e s t i n g and W a r m P r e s t r e s s i n g P r o - Strength Metallic Materials Using a Fa-
cedures l o r Nuclear R e a c t o r Vessels," tigue Cracked Bend Specimen. A report
toughness of the plate was decreased Heavy Section Steel Technology Program p r e p a r e d by Subcommittee I on F r a c t u r e
by this treatment. Technical Report No. 1, J u l y 1969. Oak Testing of H i g h S t r e n g t h M a t e r i a l s of
Ridge National L a b o r a t o r y H S S T P - T R - 1 . ASTM C o m m i t t e e E-24 on F r a c t u r e Test-
2. Nichols. R. W., " T h e Use of Over- ing of Metals. ASTM, 1967.
Acknowledgment stressing Techniques to Reduce the Risk 9. Brown. W. F.. and Srawley. J . E..
The authors wish to acknowledge of S u b s e q u e n t Brittle F r a c t u r e . " British " P l a n e Strain Crack T o u g h n e s s T e s t i n g of
Welding Journal, Vol. 15. Jan.-Feb., 1968, High S t r e n g t h Metallic M a t e r i a l s , " ASTM.
the technical guidance and financial pp. 21-42 and 75-84. STP410, 1966.
support given to this program by the 3. Baird. J . D.. " S t r a i n Aging of Steel 10. Mager, T. R„ and T h o m a s . F . O..
—A Critical Review." Iron and Steel, " E v a l u a t i o n by L i n e a r Elastic F r a c t u r e
Materials Division of the Pressure Mav-September. 1963. pp. 186-192. 326-334. Mechanics of Radiation D a m a g e to P r e s -
Vessel Research Committee, an agen- 368-374. 400-405, and 450-457. s u r e Vessel S t e e l s . " Heavy Section Steel
cy of the Welding Research Council. 4. Rubin. A. I.. Gross. J . H.. and Stout. Technology Program Technical R e p o r t No.
R. D.. "Effect of H e a t T r e a t m e n t and 5. Oak R i d g e N a t i o n a l L a b o r a t o r y , Octo-
Through the University Research Proj- Fabrication on Heavy Section P r e s s u r e ber 1969.

L^alt for j aperd .

Two symposia sponsored by the Wrought High-Nickel Alloys Committee

of the Welding Research Council will be held at the 52nd Annual AWS
Spring Meeting in San Francisco, California, during April 26-30, 1971.

Symposium # 1 will deal with: "Nickel-Base Alloy Weldments

for Elevated Temperature Service."
Those interested in presenting papers at this symposium
should write to: G. S. Hoppin, III
Bldg. 500 (M79)
Aircraft Engine Group
General Electric Company
Cincinnati, Ohio 45215

Symposium # 2 will deal with: "Fissuring of High Alloy

Authors interested in presenting papers at this symposium
should contact: Dr. D. A. Canonico
Oak Ridge National Laboratory
Post Office Box X
Oak Ridge, Tennessee 37830

500 to 1000 Word Abstracts Required by September 1, 1970

Completed Papers Required by January 15, 1971

364-s AUGUST 1970