ML20212F570
| ML20212F570 | |
| Person / Time | |
|---|---|
| Site: | Prairie Island  |
| Issue date: | 12/22/1986 |
| From: | NRC |
| To: | |
| Shared Package | |
| ML20212F113 | List: |
| References | |
| NUDOCS 8701120059 | |
| Download: ML20212F570 (10) | |
Text
.
4 APPENDIX I Evaluation of Westinghouse Report WCAP 10456, "The Effects of Thermal Aging on the Structural Integrity of Cast Stainless Steel Piping for. Westinghouse Nuclear Steam i
^^
Supply Systems" INTRODUCTION The primary coolant piping in some Westinghouse Nuclear Steam Systems (NSSS) contains cast stainless steel base metal and weld m
~ Q The base metal and weld metal are fabricated to produce a duplex s re of delta (6) ferrite in an austenitic matrix.The duplex structure pro-duces a material that has a higher yield strength, improved welda and greater resistance to intergranular stress corrosion cracking than a single phase austenitic material.
However, as early as 1965 (Ref.1),
it was recognized that long time thermal aging at primary l oop water temperatures (550*F-650'F) could significantly affect the Charpy impa toughness of the duplex structured alloys.
Since the Charpy impact test is a measure of a material's resistance to fracture, a loss in Charpy impact toughness could result in reduced structural stability in %e piping system.
.i 8701120059 861222 PDR ADOCK 05000282 p
PDR I
The purpose of Westinghouse Report WCAP-10456 is to evaluate whetljer cast stainless steel base metal and weld metal containing postulated cracks will be sensitive to unstable fracture during the 40 year life of a nuclear power plant.
In order to determine wh'ther a piping system will behave e
in such a fashion, the pipe materials' mechanical properties, design
_. criteria and method of predicting failure must be established.
In this J.
evaluation, the NRC staff asseseas the mechanical properties of thermally aged cast stainless steel pipe materials, which are reported in WCAP-10456.
DISCUSSION 1.
Weld Metal Report WCAP-10456 refers to test results reported in a paper by Slama, et.al. (Ref. 2) to conclude that the weld metal in primary loop piping would not be overly sensitive to aging and that the aged cast pipe base metal material would be structurally limiting.
In the Slama report, eight (8) welds were evaluated.
The tensile properties were only slightly affected by aging.
The Charpy U-notch impact energy in the most highly sensitive weld decreased from 7daJ/cm2 (40 ft-lbs) to near 2 (24 ft-lbs) afte' aging for 10,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> at 400'C (')52*F).
4daJ/cm This change was not considered significant.
The relatively small
. effect of aging on the weld, as compared to cast pipe matenial was reported to be caused by a difference in microstructure and lower levels of ferrite in the weld than in the cast pipe material.
2.
Cast Stainless Steel Pipe Base Material Report WCAP 10456 contains mechanical property test results from a number of heats of aged cast stainless steel material and a metallurgical study, which was performed by Westinghouse, to support a statically based model for predicting the effect of thermal aging on the Charpy impact test properties of cast stain-less steel. As a result of these tests and the proposed model, Westinghouse concluded that the fracture toughness test results from one heat of material tested represents end-of-life conditions for the 10 plants surveyed. The 10 plants surveyed are identified as Plants A through J.
a.
Mechanical Property Test Results Reported in WCAP-10456 i
Mechanical property test results on aged and unaged cast stainless steel materials were reported in papers by Landerman and Bamford (Ref. 3), Banford, Landerman and Diaz (Ref. 4), and Slama et. al (Ref. 2). These papers werd l
I
i K
discussed in WCAP-10456.
In addition, Westinghouse performed confirnatory Charpy V notch and J-integral tests on aged cast stainless steel material, which was tested and evaluated by Slama's group.
(1) The fatigue crack growth rate of aged or unaged, material in air and pressurized water reactor environments were equivalent.
(2) Tensile properties were essentially unaffected except for a slight increase in tensile strength and a decrease in ductility.
(3) J-integral test results indicate that the J and tearing 1C modulus, T, are affected by aging.
i b.
Mechanism Study in WCAP-10456 The tests and literature. survey conducted by Westinghouse indicate that the proposed mechanism of aging occurs in the range of operating temperatures for pressurized water reactorsandthedatafromacceleratedagingstudiescanl, be used to predict the behavior at operating temperatures.
Cast Stainless Steel Pipe Test c.
The materials. data discussed in section 2 of this evalua ettained from small specimens.
As a consequence, the J-R results are limited to relatively short crack extensions.
To investigate the behavior of cast stainless steel in ac,tual piping geometry
, j, Westinghouse performed two experiments:
one test was performed on thermally aged cast stainless steel and the other test was performed on cast stainless steel that was not thermally aged.
Each pipe tested contained a through-wall circumferential crack to the extent specified in WCAP-10456.
The pipe sections were i
closed at the ends, pressurized to nominal PWR operating pressure and then had bending loads applied.
1
?
The results of the tests were very similar, in that both pipes displayed extensive ductility, and stable crack extension.
There was no observed unstable crack extension i
or fast fracture.
4 4
l I
__. _ _ -.. _... _,,. ~ _ _ _. _. _.. -.... _ _,... _,, _.
. :~
The results of the Westinghouse pipe experiments indicate that cast stainisss steel, both aged and unaged, can withstand crack extensions well beyond the range of the J-R results with small specimens.
However, if crack extension is predicted in an actual application of thermally aged cast stainless steel
~
in a piping system, the staff finds that it is prudent to limit the applied J to 3000 in-lbs/in2 or less unless further studies and/or experiments demonstrate that higher values are tolerable.
Loss of initial toughness from thermal aging of cast stainless steels at normal nuclear facility operating temperatures occurs slowly over the course of many years; therefore, continuing study of the aging phenomenon may lead to a relaxation of this position.
Conversely, in the unlikely event that the total loss of toughness and the rate of toughness loss are greater than those projected in this evaluation, the staff will take appropriate action to limit the values to that which can be justified by experimental data.
Because the aging is a slow process, the staff finds there would be sufficient time for the staff to recognize the problem and to rectify the situation.
However, the staff finds this to be a highly unlikely situations because the staff has accepted only the lower bounds of data that were gathered among 10 plants encompassing the range of materials in use.
d.
Effects of Thermal Aging on Vestinghouse-Supplied Centrifugally Cast Reactor Coolant Piping Reported in WCAP-10456 The reactor coolant cast stainless steel piping materials in the plants identified in WCAP-10456 as A through J, were produced to Specification SA-351, Class CF8A as outlined in ASME Code Section II, Part A, and also to Westinghouse Equipment Specification G-678864, as revised.
For these materials, Westinghouse has calculated the predicted end-of-life Charpy U-notch properties, based on their proposed model.
The two standard deviation end of-life lower limit value for all the plants surveyed was greater than the Charpy U-notch properties of the aged reference materials, which Westinghouse indicates represents end-of-life properties for all the plants.
As a result, Westinghouse con-cluded that the amount of embrittlement in the aged reference material exceed the amount projected at end-of-life for all cast stainless steel pipe materials in Plants A through J.
CONCLUSIONS l
3 On the basis of its review of the information and data contained ib' Westinghouse
~
Report WCAP-10456, the staff concludes that:
1.
Weld metal that is used in cast stainless steel piping system'is initially less fracture resistant than the cast stainless steel base metal.
However, the weld metal is less susceptible to thermal aging than the cast stainless steel base metal.
Hence, at end-of-life the cast stainless steel base metal is anticipated to be the least fracture
~ ~
resistant material.
2.
The Westinghouse proposed model may be used to predict the relative amount of embrittlement on a heat of cast stainless steel material.
The two standard deviation lower confidence limit for this model will provide a useful engineering estimate of the predicted end-of-life Charpy impact properties for cast stainless steel base metal.
3.
Since there is considerable scatter in J-integral test data for the heats of material tested, lower bound values for J and T 1c should be used as engineering estimates for the fracture resistance of the aged reference material.
The staff believes these values should also provide a lower bound for the fracture resistance of aged and unaged weld metal.
If crack extension is predicted in an actual application of cast stainless steel in a piping system, the staff concludes that the applied J should be limited to 3000 in-lbs/in2 orlessunlesIfurther i
studies and tests demonstrate that higher values are tolerabie.
The Westinghouse pipe tests demonstrate that this may be possible.
l
=
~
f
.g.
[
4.
Since the predicted end of-life Charpy impact values for the materials in Plants A through J are greater than the value measured for the aged reference material, the lower bound fracture properties for aged reference material may be used to determine the fracture resistance for the cast stainless steel material in Plants A through J.
I t
i
REFERENCES (1)
F. H. Beck, E. A. Schoefer, J. W. Flowers, M. E. Fontana, "New Cast High Strength Alloy Grades by Structural Control," ASTM STP 369 (1965)
(2)
G. Slama, P. Petrequir., S. H. Masson, T. R. Mager, "Effect of Aging on Mechanical Properties of,,ustenitic Stainless Steel Casting and Welds,"
presented at SMIRT 7 Post Conference Seminar 6 - Assuring Structural Integrity of Steel Reactor Pressure Boundary Components, August 29/30, 1983, Monterey, Ca.
(3)
E. I. Landerman and W. H. Bamford, " Fracture Toughness and Fatigue Characteristics of Centrifuga11y Cast Type 316 Stainless Steel After Simulated Thermal Service Conditions." Fresented at the Winter Annual Meeting of the ASME, San Francisco, Ca., December 1978 (MPC-8 ASME)
(4)
W. H. Bamford, E. I. Landerman and E. Diaz, " Thermal Aging of Cast Stainless Steel and Its Impact on Piping Integrity," ASME Pressure Vessel and Piping Conference, Portland, Oregon, June 1983, American Society of Mechanical Engineers, N.Y., Circumferential Cracks in Pressure and Pipin.g, Vol.II, 1984, PP. 137-172.
5
- .