ML19259D219

Responds to NRC 790821 Ltr Re Check Valve Leak Testing.Eccs Check Valves Performing Essential Pressure Isolation Function Will Be Classified as ASME Section Xi.Info to Be Included in FSAR Amend 62

ML19259D219
Person / Time
Site:	Sequoyah
Issue date:	10/12/1979
From:	Mills L TENNESSEE VALLEY AUTHORITY
To:	Rubenstein L Office of Nuclear Reactor Regulation
References
NUDOCS 7910170389
Download: ML19259D219 (8)

Text

+ .

TENNESSEE VALLEY AUTHCRITY CH ATTANOOGA. TENNESSEE 37401 400 Chestnut Street Tower II October 12, 1979 Director of Nuclear Reactor Regulation Attention: Mr. L. S. Rubenstein, Acting Chief Light Water Reactors Branch No. 4 Division of Proj ect Management U.S. Nuclear Regulatcry Commission Washington, DC 20555

Dear Mr. Rubenstein:

In the Matter of the Application of ) Docket Nos. 50-327 Tennessee Valley Authority ) 50-328 This is in response to your August 21, 1979, letter to H. G. Parris concerning check valve leak testing for Sequoyah Nuclear Plant.

Enclosed is TVA's response to questions 212.115 which states the NRC positions on classification of isolation check valves and leak criteria.

This response will be incorporated into the Sequoyah Nuclear Plant Final Safety Analysin Report by Amendment 62.

Very truly yours, TENNESSEE VALLEY AUTHORITY

l. $

L. M. Mills, Manager Nuclear Regulation and Safety Enclosure j* * ,-

7910170 389t

RESPONSE T0 1 TEM 212.115 0F AUGUST 21, 1979, LETTER FROM L. S. RUBENSTEIN TO H. G. PARRIS 6.65 In their response to scaff question 212.74, TVA detailed their (212.115) proposed plan to leak test check valves which isolate high and low pressure systems. The staff finds several features of the response to be unacceptable. The applicant has not committed to the staff requirements that the indicated pressure isolation check valves be classified ASME IWV-2000 category AC. Due to the essential pressure isolation function which these valves perform, the staff will require that the applicant place them under the ASME inservice testing category whis- does require leak testing. Additionally, the applicant has inc sated that his acceptance criteria for check valves being leak tested, is 9 gpm. The staff finds this leak rate to be excessive, and will require that the check valves indicated in the response to question 2.2174 meet a leak criteria of 1 gpm.

Response

The ECCS check valves that perform as essential p,ressure isotation function will be classified ASME Section XI IWV-2000 category AC, and the criteria for leak rate test acceptance will be one gpm, as requested.

The TVA response to question 212.74 was for Watts Bar, not Sequoyah.

Attached is a reprint of the Watts Bar questic. c with the response amended to apply to Sequoyah and to comply with NRC's position on valve categoriza-tion and leak rate limit. Testing under Section XI of the ASME Code will be as described.

j i 3 ,-

-- J

! 9 R {U , AlL

-ter:F=rr g

i2.'74 Cuestion E l 7. 7 Y hvom. b d a. + 6 s S e F 5 /! Ps KS. 2i 2)

Check valves in the discharge side of the high head 7

safety injection, low head safety injection, PHF, e charging, and heron injection systems perfere an

\ isolation function in that they protect lcw pressure systems fro.. full reactor pressure. The staff will require that these check valves be classified ASME KEV-2000 categcry AC, with the leak testing for this class of valve being performed to ccde specifications. Each

(~

check valve in the systems identified above must te leak A ,

tested; it is not satisfactory to just pull a suction en the outer cost check valve. This only verifies that one of ti series check valves is seated. The necessary frey-ency of testing vill be that specified in the ASME Code, e:: cept in cases where only one or two check valves in series separate high tc lcw pressure systems. In these cases, leak testing will be performed at each refueling af ter the valves have been ,e:<ercised.

valvereferredtointheacovedis[cussion. Identify Verify the ASME that19V-2000 Se you will meet the required leak testing schedule, and that you have the necessary tast lines to leak test each valve. Provide che leak detecticn criteria that bill be

(.

used.

Responsa . A m e a[Q[ o et a v y v I /h Series check valves which separate high pressure reactor ccolant systen (PCS) piping frcm louer tressure piping and compenents associated with the safety injecticn (SIS), residual heat removal (BURS) , and upper head injecticn (C31) systems will be pericdically tested to assure each valve's closed position during normal pot.er opet .cion. (2crcn injection line check valves, which do not form a high/lcw pressure interface, will also te

( tested.) The ma::imum cime hetueen cicsure tests will act exceed the interval between refueling cutages. The perranently installed tesc systers will normally be used for closure verification.

Those checP valvas adjacent to the BCS (intcard check

( valves) vill be veri.ied closed by the fclicuing procedure identified as Test Type La in Table C212.74- 1:

1. Pressurize the downstream side of the check valve by incre.asing the BCS pressure acove 1800 pounds, k

),,

w -<

.+BUP 2. Isolate all water sources upstream of the check valve to be testel,

3. Open the volume upstrei..n of the check valve to the installed test line, and c

4. Evaluate flow conditiens in :he test line for an indication of va.1.ve closure. .If the flow-rate-isy 9,0_g gm..o r- -le ss ,-the -te st-is acce p table .

The .second check valves f rom the FCS (outboard.

I check values) can be tegted for closure by j , observing their ability to maintain an established

)1F) dif tarential pressure, by leak test, cr by any

( other e qually acceptable alternative.- ,

, / b Closure testing by leak measurement appears most satisfactory and may be dcne by the iclicwing methods:

!('[ 1.

Test Type L2 (for safety injection pump cold leg injectica line valves63-551, 553, 555, and 557) .

Back pressure is provided by an SIS accumulatcr.

Isolate the volume upstream of the valve except fcr connection to a leak test line. Measure leakage.

2. -Test Type L3 (for SIS accumularcr injection and RHR

pump cold and hot leg injection line valves63-622, 623, 624, 625, 632, 633, 634, 635, 640, and 643).

Use the safety injection pump, through the appropriate cold cr het leg injectica line, tc apply. back pressure to the valve. Measure leakage as for Test Type L2-

3. Test Tyt.c L 4 (for bcron injecticn line and safety injectica pump hot leg injection line valves63-581, 543, 545, 547, and 549) . Use the safety inf =ction purcp, through the appropriate test line connection, to apply back pressure to the valve.

Measure back leakage by isclating the upstream volume, venting a high point, and collecting ~5!

leakage in a calibrated vessel.

4. Test Type Ls (for UHI injection line valves87-562 and 563). Use a charging pump through the appropriate test line to apply back pressure to the valve. Measure leakage as for Test Type L2 -j Acceptance for any single check valves test leakage will depend on durenstrating its capability to fully prctect its connected, low pressure system from an overpressure transient in the rare event that the valve's similarly 39 tested, redundant counterpart experiences gross leak

(

,, _ 3- ---

i

_s

e .

. d BW-39 j-

=

tightness failure. Also, acceptance will aunure that the normal, prirary systerr charging capetility is not challenged by such a failure and the plant can proceed with an cederly shutdcwn. Saall check valve leakages are ieportant for normal cperatico becauce of 1) requirements to restrict FCS coilcakage as discussed in C Fogulatory Guide 1.45, 2) waste dispccal system capacity and operating costs, and 3) piping systen contamination in norrally atagnare lines. Thece saall leakages will te addrecsed by normal plant operating censideraticns.

and-cont-inuously-conitored a s-descrited-in-cur-response

.to-FSAi-Cues t ica-212. 7 5.- be will atte=pt to ccrrect any (e srall leaks at the earliest cpportunity to avoid forced plant outageu because of FCS cutleabge. For the purpose cf protecting lower pressure systems in the event of a check valve failure, the leak rate tercugh any single valve can be significantly greater than that allowed generally for thc check valve system. be need to take advantage of this fact becaure the single check valve integrity test is dcnc during the final stages of plant heatup when returnirg to Ec.cr operation after a ref ueling cu tage. Finding an unacceptably leaking check valve at this ti e will then require returning tc a cold shutdcwn conditon and possibly draining the FCS (in the case of an intoard check valve) to gain eccess to the -

N

( valve internals, just when the plant is otherwise fully operaticnal end ready to prcduce powcr.

economic it;crtance of checsing a reascnacle leak rate Hence, the criteria for this particular test beccJes ap;arent. We will use 4-gpia-ee-the-test-criteria Thiwvalue, as ,

__ o e~ _\ daw hSkC11hlCW. Y U!f '

W n$1oc.) -.ccWN c k&C . ,8c k*?* 's - - - - - - - -

11) Leak rate creater han one but less than nine nam - 13

'ino leaking caecx valve util ue utsasscnolcd, 19 inspected, lapped and rotested as accessary to reduce the leakage below 1 gpa. This will be dono 20 at the first opportunity but not latcr than the i next ref ueling autage, f,

[y2

} }' 12) Leak rate 9 ypa r greater - The plant vill be brougnt tu cu N sautcown and necessary caintenance 23 l jk ( fl will be {crforoed to assure check valve leakage 24 less than one gpn before the plant returns to power i operation. -~:. .. ~ _.- __ _ 2$

c,,. __ ,, - -' ' %c-t r e r-ebe-et- a .' _ - o. ,

( W.

The5r . c MW Ny

/phosetrn hould not result in undue forced outages ,ead,, w

-i Wie welb uity ,)r1itany is limits cnly 5% required or 1r.ntoofeccure the plant' saf rpety, ove res Lufuh'otection su re pr relief capaci'q :ct the lo.

pressure systems which would cone unccafcrtably alese to exceeding precsure ecundary design cafety rargins if subjected to full FCS pressure, 2) te is 15f. or less of relief capacity for low prescure sycters whicn have a high enough design pressure to preclude their gecs=

( f ailure when exposed to f ull FCG presst,ro , and 3? the ir 4 - - " :s low enough tc have negligible effect en the norreal charging systen and no affect on a nor:ral shutdoun capctility, and 4) it is within the permanently inctalled leak tect ne&curcrent capatility. Further, although a pctential leak rate cf this rugnitude (rcsulting tren groIs failure of a rcGundant check

( valve) is r% ' desiratic, it is nct unsafe and wculd te detceted carl. w tan a sgall leak, reducinq the tir.e of plant operation ' Ocu t the Lcnefic cf double check valve protection.

(

s-*

Dt.A. -la - *3

3. ^ j

Eelow is a table of piping systems and safety valve data which shows ccnservati sm in available relief rates fcr a '

9 gpm leak.

Safety Valve Volumetric Felief 9 gpm leak rate ,

Piping Setroint ( r.sio) Caracity (ccm) as a % cf Cat.acitv f BHB Pump 600 820 1.1%

Discharge SIS 700 235 3.8%

Accumu-lator SI Pump 1750 60 15%

Cischarge UHI Ac- 1800 70 13%

cumulator Baron 2735 N.A. G.A.

Injection

f. s Kl&tN 0X ! _ .- .M - TA

.--y ..-.v. ^ bt ek5 e'u---fffJJ.asa ban." ih.kr h3 AA. tMbkS-,

1. J- u' lH W C- O .f <;d iM4if<  !* 5 IIED-

/'?] h W [.-

g- . _ .

-We-ob ject--to-ass igning - ASME Section--XI, Category AC to any-of--these check valves. A Category AC would . ; . e * .r; 0 g.

incorrectly.. impose. seat. leakage testing requirements .g when it is the valve pasition in which we are primarily interected. We are u';ing a leak test en the inbcard check valves simply hecause this is a very positive as well as convenient reans of determining valve closure.

We believe that any cther method of determining cicsure is also acceptable (such as cbserving differential pressure ratentica capability as on the cuthcard check valves), withcut a determination of individual valve leak rates as required fcr Category AC valves.

The cNeck valve testing thus far described is intended to limit the probability of a double check valve failure, and undesirable events related theretc, tc -=

tragnitudes indicated in references 1 and 2 in the '

neightcrhcod of 5.5 x 10-S per year.

1.

P00RORGN1 EPRI NP-252, Prcject 767-1, September 1976, "EWF ,

Sensitivit y to Alterations in the Interfacing 1 Systems LCCA," (page 37, item c).

2. Nuclear Technology, Vol. 37, Jan 78, "Prchabilistic Analysis of the Interfacing System Loss-ct-Ccclant Accident and Implications cn Design Decisions," pp 5-12 (see pages 11, item 3) . ,,, , - ,

(

TABLE C212.74-1 CHECK VALVE TABLE Piping TVA Valve No. Location Test Type Sec. XI Catecory Boron Injec- 63-586 1 Ia (X) C tion 587 1 It (X) C

( 588 589 1

1 La (X)

La (X)

C C

63-581 2 L. (X) C SIP /BHFP/ SIS63-560 1 It AC Accumulators 561 1 La 4C

( 562 1 12 4C

( 563 1 It AC SIS Accunu- 63-622 2 L3 /JC lators 623 2 La dC 624 2 13 dC 625 2 13 dC SIP (CL)63-551 2 La dC 553 2 La dc 555 2 La /ic 557 2 La </C RERP (CL)63-632 2 *L 3 //C

{ 633 2 La sic

( -

634 2 13 dC 635 2 13 4C SIP /BHFP (HL)63-641 1 Lt AC 63-644 1 La AC SIP. (HL)63-558 1 I t AC 559 1 It AC 543 2 L. ,1C 545 2 L. 1C 547 2 L. ,

1C 549 2 L. dC

( R ER P (HL)63-640 643 2

2 La La AC AC UHI 87-558 1 La tic i _. ;

Sheet 1 Jtey.ised. by-Amendmen L 39-

TAELE Q212.74-1 (Continued)

CHECX VALVE TABLE (c

Piping TVA Valve No. Eccation Test Type Sec. XI Catecorv 559 1 L1 A,C 560 1 13 /ic -

561 1 La AC 1 ,

562 2 L3 AC 563 2 L3 AC Lccation: 1 = valve closest to the E.S (inboard check valve) e 2 = second valve inside containcent (cutboard check valve) 3 = third valve inside containment Test Type: L = leak test X = not a high/ low pressure boundary interface

, ,. s ., . -

) '

i _s

• 5-0

(

Sheet 2 Fev.ined by~ Amendmenti23.