ML18153A927
| ML18153A927 | |
| Person / Time | |
|---|---|
| Site: | Surry  |
| Issue date: | 01/21/1994 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML18153A926 | List: |
| References | |
| TAC-M88122, TAC-M888123, NUDOCS 9404250282 | |
| Download: ML18153A927 (4) | |
Text
e e
UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 2066IH>001 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION REVIEW OF RELIEF REQUEST:
DEFERRAL OF CODE REPLACEMENT REQUIREMENTS FOR THE SURRY UNIT l AND UNIT 2 SERVICE WATER VALVES VIRGINIA ELECTRIC AND POWER COMPANY DOCKET NOS.:
50-280 AND 50-281 TAC NOS. M88122 ANO M881223
1.0 BACKGROUND
By letter dated November 2, 1993, Virginia Electric and Power Company (the licensee), requested relief from the replacement requirements of the American Society of Mechanical Engineers (ASHE) Boiler and Pres.sure Vessel Code (the
- Code),Section XI, for a total of 40 service water (SW) system valves i nsta 11 ed at Surry Power Station Uni ts 1 and 2.
The 11 censee.. observed 1 eakage through the valve body material on a number of valves on both units. The valves were discovered to be degraded as a result of dealloying, a form of corrosion. Because the dea 11 oyi ng is not repa i rab 1 e-, the 1 i censee must replace the valves. However, replacements were not i11111ediately obtainable, and not all affected (or potentially affected) valves could be replaced during unit operation. Relief was requested to defer replacements until new valves could be procured and until the next planned outage of sufficient duration (for the respective unit) to replace those valves that cannot be replaced during unit operation.
- '..... \\
The subject valves are nominal pipe size (NPS) 2* and under.
The material is ASHE/ASTM SB-148, alloy UNS No. 95400, a cast aluminum-bronze with 11%
aluminum as the major alloying element. The valves are Jamesbury ball valves with integral cast flanges.
The SW system uses brackish water.
Nominal operating conditions are 45 psig at ambient temperature. The SW system provides cooling water to various safety-related equipment, and is, thus, a Code class 3 safety-related syste11.
2.0 DISCUSSION 2.1 ROOT CAUSE ANALYSIS A number of valves on each unit were observed to be showing signs of slight leakage through the valve body.
The leakage, at worst, was minimal.
If the accumulated moisture was wiped off the valve, it would take some minutes before a wet spot or droplet would form.
/------
9404250282 940121 -~
PDR ADOCK 05000280
- - f p
2 Metallurgical examination of two valves removed from service revealed they were deteriorating as a result of dealloying. Dealloying is a form of corrosion that is manifested by the selective removal of an alloying element from the metal.
For an aluminum-bronze alloy it involves the loss of aluminum and the formation of a spongy porous copper. This deterioration reduces the tensile strength and toughness of the material.
The licensee attributed the dealloying to a combination of factors.
The SW system experiences significant biofouling. This requires periodic disassembly of portions of the SW system for removal of the accumulated biomass.
Because of this persistent accumulation, crevice conditions are created. This can result in crevtce corrosion which occurs when differential chemical compositions exist between the fluid in the crevice and the bulk fluid flowing in the pipe.* Additionally, microbiologically induced corrosion (MIC) is encouraged in the crevice under the biomass.
The staff observes that these conditions are exacerbated by the fact that the SW system employs brackish water.
The chloride content in a crevice is concentrated above that of the bulk medium, as for other aggressive constituents.-
Additionally, alloy microstructure is known to affect the dealloying resistance of aluminum-bronzes.
Its significance in this case was not explored because the licensee decided to specify a different material for the replacement valves.
Upon finding that the aluminum-bronze SW valves were experiencing dealloying, the licensee considered all SW valves of the same material to be at risk and, thus, candidates'for replacement, although half of the valves in the system showed no signs of. leakage.
The total of affected or potentially affected valves for the two units was 40.
Leakage-rate monitoring,-0ver a period of approximately one month did not reveal any noticeable *increase.
Based upon this, and experiences learned from another licensee, VEPCO concluded that the dealloying was progressing at a slow, linear rate. Thus, the licensee concluded that the structural integrity of the valves would not change substantially over a period of months while replacements were sought and/or installed.
2.2 STRUCTURAL EVALUATION To bound the conditions affecting structural integrity, and thus, operability, the two valves subjected to metallurgical examination were selected on the basis of degree of through-wall leakage. One valve was selected as an example of worst leakage, and another as an example of moderate leakage.
Metallurgical sections through the two valves revealed the extent and nature of the dealloying.
The worst valve (based upon leakage) revealed a general degree of dealloying affecting the metal to a depth approaching 2/3 of the wall thickness in places. At a location of the leakage, the dealloying penetrated the wall 1n an area of less than 25% of the valve circumference.
The valve with the lesser degree of leakage revealed less extensive degradation.
3 With these bounds on the severity of the degradation, the structural integrity was assessed for the valve body and separately for the integral cast flanges.
These analyses indicated that the integral cast flanges retained a margin of safety of 4.27. This significant load carrying margin is due to their heavy se~t ion thickness, !=Ons~qu~nt 1 <>wer JoadJ ng ~ 1 ~~er P.erc~n~age,.~f,.de alloyed th 1 ckness and absence** (>'f any through...:the;..th 1 ckness' cleafloyfhg:' *"Conseqlient 1 y, the valve body became the potentiaJly limiting factor.,
... 1--1 i_.
- *,_rt _ _..,.,
,.... ir-
- .-*.* 'l}i:
,\\C YLJ~1___..r.*:1,-~f~'
Dea 11 oyi ng adve.rs.~ly af,fect 0s thE! __ mate,ria 1 proper,t.i es.
AJ ~hough tens i1 e strength is affe'cted, the *most slgni ficant degracfation~ is 1 'in thi materi'a:1 toughness.
Because of this, it was.apparent th~t thE! effect of transient or impact loads would be the dete*rmtning* 'factor itfrasses's"fhtt structural integrity. For this evaluation, the licensee performed a limit load analysis to examine the material's resi'stance-to a ductile -o~erl'oad and a fracture mechanics evaluation to assess the material's perf.onn~nce in the presence of a postulated crack-1 i ke defect.
The detall"s aiid c*al'cu=rat i ans of these ana 1 yses were contained i~ the licensee's technical report No. CE-0070, Rev. 0, submitted a~~- itipplement to the relief request.
The limit* load analysi~~issumed a completely d~illojed valve~body;* Strengt~
data for dealloyed aluminum-bronze was tised1 ;~~the calculatiori. 'The *results*
of the analysts* indicate that**there*.. 1s*at*,1east a 1margin* of safety of 2.17,<
against a net section plasticity. Thus~:~'dUct11~~rupture (6~ijrloadjswill not occur for maximum load situations. * --
. '.,r.-0....
The dealloy~d valves* do not show*any lineaf fl~ws:or indicatio~i of crack-like behavior. But, recognizing the low toughness* property of'*the dealloyed' materi a 1, a through-wal 1 crack was postll1 atedfor *the ana*lysis ~
- Crack length was assumed to be 25" of the valve circumference.*, The postulated' cracked section was subjected to the enveloping (worst' case) loaas~ *including* seismic.
- \\.
- ~-.;-:?.
~.
i:*.
For the purpose of_ this evaluatfon*'a"frcl°c:ture toughness of 65' ks1 (in)* 112 was used for'the remaining unaffected material. This allowable was based on tests conducted elsewhere on the same material type, and previously reviewed by the NRC staff.
No credit was t.aken for-t~e-: reshtance provided**by the fully dealloyed material. The assum~d thkiness -of the *tJnaffected:~inaterial was 1/3 of the original wall thickness~ Thii"assumption was iri*accord'aru:-e with the observed maximum depth of'dealloying at locations other than the leak path.
The result~ of this evaluation show that the applied stress intensity is 14.96 ksi (in) 11, representing a margin of safety of 4.35 against brittle fracture.
- 2. 3 OTHER ACTIONS The licensee committed to replacing the valves in an expeditious manner.
To date, all 20 valves have been replaced in Surry, Unit 2, and 8 of the 20 valves have been replaced in Surry, Unit 1.
The schedule for this replacement would be governed by the availability of repl a.cement components and the ability to replace a given valve during unit operation. All affected valves would be replaced by the conclusion of the next refueling outage for Surry, Unit 1, which is scheduled to commence on January 21, 1994. Additionally, a
4 weekly* surveillance will be conducted to monitor those valves presently exhibiting no indication of the dealloying phenomena and to detect changes in the amount of leakage at each affected valve.
The piping and components are visible and, are observed on, a daily,,basi.s, during routine. housekeeping and, :
operator walkdowns.
3. o CONCLUSION The staff finds that the licensee's determination of the. root cause and the.
structural integrity evaluations are reasonable and conserva.tiveJy bounding.
Consequently, the staff finds.that operabi 1 ity has-. been adequate. ly verified.
Additionally, the. staff finds that the 1 icensee' s actions with respect to expeditious replacement and periodi.c monitoring of-.the remaining valves are appropriate.
Pursuant to 10 CFR 50.55a(a){3){ii), relief. is gran_te~ uniil.the end of the next scheduled refueling outage. for each unit at Surry' Power Station for the deferral o_f replacement of SW. system aluminum"."bronze valves.. At that time, any remaining alumfoum~bronze SW system valve,s m~st be replac*ed.
Principal.
Contact:
- 's~ Hornseth'.,"'.
Date: January 21, 1994
- ~ I '. ' ; :'
I ;"7
- : ;.: (~
':° \\ i
' ~-~
- ~. :..
- 1J!~!~_:;:;_ 1,i'~L-:*~ ~
- "i I l
~ : : '* : *
~I ;
. ***=-*
... *. ~
.-.ir.