Skip to main content

Full text of "NASA Technical Reports Server (NTRS) 19920006682: HP9-4-.30 weld properties and microstructure"

See other formats


1991 


N92-15 


NASA/ASEE SUMMER FACULTY FELLOWSHIP PROGRAM 


MARSHALL SPACE FLIGHT CENTER 
THE UNIVERSITY OF ALABAMA 


HP9-4-.30 WELD PROPERTIES AND MICROSTRUCTURE 


Prepared By: 


George W . Watt , Ph . D . 


Academic Rank: 


Assistant Professor 


Institution: 


Utah State University 
Department of Industrial Technology 


NAS A/MS FC: 

Laboratory: 

Division: 

Branch: 


Materials and Processes 
Metallic Materials 
Metallurgy Research 


MSFC Colleague: 


Tina W. Malone 


Contract No. : 


NGT-0 1-008-021 

The University of Alabama 


XX 


*Q ( 




HP9-4-.30, ultra high-strength steel, the case material for 
the Advanced Solid Rocket Motor (ASRM) , must exhibit acceptable 
strength, ductility, toughness, and stress corrosion cracking (SCC) 
resistance after welding and a local post weld heat treatment 
(PWHT) . Testing, to date, shows that the base metal (BM) 
properties are more than adequate for the anticipated launch loads. 
Tensile tests of test specimens taken transverse to the weld show 
that the weld metal overmatches the BM even in the PWHT condition. 
However, there is still some question about the toughness and SCC 
resistance of the weld metal in the as welded and post weld heat 
treated condition. 

To help clarify the as welded and post weld heat treated 
mechanical behavior of the alloy, subsize tensile specimens from 
the BM, the fusion zone (FZ) with and without PWHT, and the heat 
affected zone (HAZ) with and without PWHT were tested to failure 
and the fracture surfaces subsequently examined with a scanning 
electron microscope (SEM) . Table I shows the test results (the 
average of 5-6 test specimens) and, for comparison, average 
results from a large number of full scale BM tensile tests 
accomplished at MSFC. The full size specimen tensile test results 
and the subsize specimen tensile results for the BM are comparable 
indicating at most a small specimen size effect for the subsize 
tensile results. Without PWHT the FZ and HAZ materials have a 
reduced 0.2% offset yield strength (YS) , a very high ultimate 
tensile strength (UTS) , and significant loss of ductility as 
measured by either %E1 or %RA. The reduced YS is probably due to 
the presence of appreciable amounts of relatively soft retained 
austenite that begins to plastically deform at a lower stress than 
the higher strength martensite matrix surrounding it. The 
influence of retained austenite on the fracture of maraging steels 

TABLE I 


Subsize (0.16" dia.) 


NASA 


Prooertv 

full size 

BM FZ.NoPWHT 

FZ . PWHT 

HAZ . NoPWHT 

HAZ . PWHT 

YS(ksi) 

206 

211 

181 

221 

188 

215 

UTS(ksi) 

227 

237 

271 

250 

293 

251 

%E1 

22.8* 

14.3 

7.5 

14.6 

8.2 

16.2 

(1/2" GL) 







%RA 

54.3 

54.3 

35.5 

43.6 

45.2 

57.2 

* The gage length (GL) for 

the NASA 

full size specimens 

was 2". 


XX- 1 


has been reported in the literature. 1 It was suggested that the 
highly strained pools of austenite contain carbide or other 
particles that are sites for the creation of voids which grow, 
coalesce, and result in failure by microvoid coalescence. The data 
in Table I, also, clearly show that the PWHT used (950 F for 2 
hours ) results in return of the FZ and HAZ to almost base metal YS 
and UTS values. The %E1 and %RA of the HAZ also indicate recovery 
of most of the BM ductility, but the %RA value for the post weld 
heat treated FZ are considerably below the BM values indicating the 
ductility of the FZ remains low. These results would suggest that 
a good PWHT should return the HAZ to desired strength and 
toughness, but probably not recover the FZ toughness. 

SEM examination of the fracture surfaces of representative 
samples from the subsize tensile tests (BM, FZ with and without 
PWHT, and HAZ with and without PWHT) shows the following results 
and trends. All fracture surfaces exhibit a microvoid coalescence 
failure mode. In general, there appears to be a bi-modal 
distribution of dimple or void sizes on the fracture surface with 
small dimples being less than 1 micron in diameter and the large 
dimples being from 1-5 microns in diameter. Another clearly 
discernible characteristic was that the FZ and HAZ without PWHT 
specimens had a much higher fraction of the small dimples on the 
fracture surface than the BM. After PWHT, the HAZ showed an 
increase in the fraction of larger dimples while the FZ seemed to 
show a decrease. This reduction in FZ dimple size is probably due 
to sample to sample variation and should not be interpreted as 
caused by the PWHT. Preliminary estimates of the average dimple 
size on the fracture surface (D 0 ) , which depends on the dimple size 
distribution, and the strain to fracture estimated from %RA (e f ) 
are given in Table II. These data show that as D 0 increases, at 
least within the range of sizes considered here, e f increases. 
Garrison, et al, 2 have related void or dimple spacing to crack tip 
opening displacement. As dimple spacing increases dimple diameter 
will increase in direct proportion and, since, crack tip opening 
displacement is proportional to strain to fracture the data in 
Table II should follow a relationship similar to that determined by 
Garrison. He showed, at low void spacmgs, the relationship 
between void spacing and crack tip opening displacement should be 
linear; experimental data appeared to confirm this result. Thus, 


TABLE II 


Material 

D° fmicron) 

et 

BM 

1.7 

0.78 

HAZ no PWHT 

1.3 

0.42 

FZ no PWHT ( 1 ) 

1.1 

0.44 

FZ no PWHT (2) 

1.2 

0.47 

HAZ PWHT 

1.4 

0.70 

FZ PWHT 

1.0 

.49 


XX-2 

C'f 


for the dimple sizes in this work, it is expected that the 
relationship between ef and D 0 should be linear with a zero 
intercept. The data do indicate a linear least squares fit of 
ef=0.4D o . It appears from these data that the ductility of the HAZ 
is reqained to a larqe extent by the PWHT while that of the FZ is 
only slightly improved. Consequently, the weak link, as far as 
toughness and stress corrosion cracking of the post weld heat 
treated weldment is concerned, would seem to be the fusion zone. 

Limited tensile SCC tests using the ASTM 3 . 5% NaCl Alternate 
Immersion and the 5% Salt Spray Tests were performed at MSFC and 
indicate a possible problem in the weld metal. In the as welded 
condition most of the failures occurred in three weeks or less at 
stresses equal to or greater than 75% YS with the cracking tending 
to initiate in the HAZ. After PWHT there appeared to be an 

improvement, but failures still occurred at close to three weeks in 
some cases. However, after the PWHT the failures seemed to 
initiate more in the FZ. Several specimens were sectioned 
perpendicular to the fracture surface, polished, and etched in 
order to determine the location of the fracture initiation. These 
specimens were also used for microhardness measurements to 
determine if there was a relationship between hardness and the 
location of the crack initiation site. Gouch 3 has concluded that, 
for high strength steels, welding greatly increases susceptibility 
to SCC with the hardest regions of the weld zone being the most 
susceptible with microstructure having a secondary effect. He also 
found that PWHT improves SCC resistance since it results in a 
reduction in hardness in the FZ and HAZ. 

Comparison of the microhardness data, taken in this instance 
from a limited number of the HP9-4-.30 weldments tested in SCC, 
with the apparent crack initiation point indicates that in the as 
welded condition the crack does appear to initiate in one of the 
harder regions of the FZ or HAZ. However, the PWHT seems to shift 
the failure location into the FZ even though there are still 
harder regions in the HAZ. Apparently, in the post weld heat 
treated condition the microstructure and/or microsegregation become 
more dominant than the hardness in controlling the SCC. Of course, 
the effect of the surface condition may also play a major role, and 
it may be that in the PWHT condition the exposed surface of the FZ 
is more susceptible to pitting so it is easier to initiate the 
cracks. Once the pits (or other surface defect) has been initiated 
the crack probably grows by a hydrogen embrittlement mechanism. 
Tromans 4 studied the stress corrosion cracking of HY-180 steel at 
various corrosion electrochemical potentials and found in all cases 
that the crack propagation was consistent with hydrogen 
embrittlement . 

It has been suggested that a higher temperature PWHT and 
perhaps a different temper during BM processing could improve the 
SCC resistance of the weldments. It is possible to achieve some 


improvement, but if the FZ susceptibility problem is associated 
with the microstructure/microsegregation then that cannot be 
changed by a PWHT. 


REFERENCES 


1. Kenyon, N., "Effect of Austenite on the Toughness of Maraging 
Steel Welds," Welding Journal, May 68, p. 193-a. 

2. Garrison, W.M. Jr, Raghavan, K.S., and Maloney,. J.L., "Fracture 
Toughness: A Discussion of the Influence of Particle kSpacing at 
Constant Particle Volume Fraction," submitted to Metall. Trans. A. 

3. Gouch, T.G., "Stress Corrosion Cracking of Welded Joints in 
High Strength Steels,” Welding Journal, July 74, p. 287-s. 

4. Tromans, D., "Stress Corrosion Cracking of HY-180 Steel in 
Aqueous 3.5 Pet NaCl,” Metall. Trans. A, Vol 12A, p. 1445. 


XX-4