ML20105B102

Containment Leakage Rate Testing, Technical Evaluation Rept

ML20105B102
Person / Time
Site:	Browns Ferry
Issue date:	12/23/1981
From:	Delgaizo T Franklin Research Ctr, Franklin Institute
To:	Huang Y Office of Nuclear Reactor Regulation
Shared Package
ML18029A246	List: ML18029A245 ML20105B102
References
CON-NRC-03-79-118, CON-NRC-3-79-118, TAC 08715, TAC 08716, TAC 08717 NUDOCS 8112280441, TER-C5257-5-6-1, TER-C5257-5-6-176
Download: ML20105B102 (23)

Text

O 4

e 7-TECHN'ICAL EVALUATION REPORT

~

CONTAINMENT LEAKAGE RATE TESTING TENNESSEE VALLEY AUTHORITY BROWNS FERRY NUCLEAR PLANT UNITS 1, 2, AND 3 NRC DOCKET NO. 50-259/260/296 FRC PROJECT cs257 NRC TAC NO. 08715/08716/08717 FRC ASSIGNTAENT 1 NRC CONTRACT NO. NRC-03-79-118 FRC TASKS 5/6/176 j

3 Preparedby Franklin Research Center Author:

T. J. DelCai=o The Parkway at Twentieth Street' Philadelphia, PA 19103 FRC Group Leader:

T. J. DelGairo Prepared for Nuclear Regulatory Commission

~

Washington, D.C. 20555 Lead NRC Engineer: Y.

S. P.uang Dece ber 23, 1981 This report was prepared as an account of v'/ork sponsored by an agency of the United States Governmerit. Neither the United States Government nor any agency thereof, or any of their emplo3 ces, makes any warranty, expressed or implied, or assumes any legal liability or responsib!!ity for any third party's tse, or the results of such use, of any information, apparatus, product or process disclosed in this report, or represents that its use by such third party would not infringe privately owned rights.

9112.2 8 6 % i O

_ Franklin Research Center A Didsion of The Franklin Institute The Benjamin Franun Park.ey. PNta. Pa. 19103 (215) 448 ICOG

(

G t

TER-C5257-5/6/176

~.

l l

CONTENTS 8

Section

' Title Page 1

l 1

BACKGROUh"J.

1 I

2 EVALUATION CRITERIA.

.~.

2 3

TECHNICAL EVALUATION

~

3 3.1 Requests for Exemption from the Requirements of Appendix J.

3 3.1.1 Draining and Venting of Instrument Sensing Lines During Type A Testing 4 _.

I ~

i 3.1.2 Draining and Venting of the Seal Water supply to the Reactor. Recirculation Pumps During 5-Type A Testing.

.(~

3.1. 3

'"ype B Testig cf I'm'-aiment Airlocks.

8 3.1. 4 Reverse Direction Type C Testing of Certain Isolation Valves 11 3.1. 5 Reduced Pressure Type C Testing of M' in Steam a

Isolation Valves 12 Type C Testing with Water in Lieu of Air or 3.1. 6 Nitrogen as a Medium 12 3.1. 7 Type C Testing of Valves in the Reactor Bu'ilding Closed Cooling Water System 14 3.1.8 Type C Testing of Traversing In-Core ' robe

~

System Valves.

15

• i

3. 2 Clarification of the Requirements of Appendix J 15

1. ' '

4 CONCLUSIONS 18 5

REFERENCES 20 i

iii

.M Frenklin Research Ce'nter A Ca*se of N Frs $m wy

l

(

TECHNICAL EVALUATION REPORT L

CONTAINMENT LEAKAGE RATE TESTING TENNESSEE VALLEY AUTHORITY a

BROWNS FERRY NUCLEAR PLANT UNITS L 2, AND 3 i

NRC DOCKET NO. 50-259/260/296 FRC PROJECT CS257 NRC TAC NO.

08715/08716/08717 FRC ASsLGNMENT 1

~

, NRC CONTRACT NO. NRC-03-79-118 FRC tasks 5/6/176

$=

Preparedby Franklin Research Center Author:

T. Jr. DelCaizo The Parkway at Twentieth Street' "

Philadelphia, PA 19103 FRC Group Leader:

T. J. DelGaizo

--(

Preparedfor Nuclear Regulatory Commission Washington, D.C. 20555 Lead NRC Erg.ineer: Y. S. Huang Dece=ber 23, 1981 This report was prepared as an account of work spe:nsored by an l

agency of the United States Government. Neither the 1)nited States Government nor any agency thereof, or any of theirr emplo3ces, l

makes any warranty, expressed or implied, or assurnes any legal l

liability or responsibility for any third party's use, or the results of l

such use, of any information, apparatus, product or process disclosed in this report, or represents that its use by such third party would not infringe privately owned rights.

O Franklin Research Center A Division of The I ranklin Institute The Benjam.n Frenkhn Pa-% sy. PMa-. Fa 19103 (215) 448-1 COG

(

TER-CS257-5/6/176

.{

' d.

1.

BACKGROUND I

on August 4,1975 [1], the NRC requested the Tennessee Valley Authority i

(TvA) to review the ecsntainment leakage testing programs at Browns Ferry

.I Nuclear Plant.and to provide a plan for achiev"ing full com:pliance with 10CFR50, I

Appendix J, including appropriate design modifications, changes to technical specifications, or requests for exemption from the requirements. pursuant to

~

~

10CFR50.12, where necessary.

~

o WA responded to the NRC's request in a series of four letters dated Emptember 9, 1975 [2), October 10, 1975 [3), January 15, 1976 [4], and

' February 26, 1976 15]. As result of this corresh>ondence, 27A requested certain exemptions from the requirements of ' Appendix J.

The NRC responded in a letter dated December 27, 1976 [63 requesting clarification of certair$

f areas.

,:e I

,7 Cn July 8, 1977 [7], WA provided thelsdeditional information and also i(,

e j

requested another. exemption which had not been previously indicated. on,May 27,'1980 [8], WA provided still further information in response to another NRC request for information dated April 21, 1980 [9].

~

The purpose of this report is to provide technical evaluatiens of out-

. standing submittals regarding the implementation of the requirements -of 10CFR50, Appendix J, at Browns Ferry Nuclear Plant Units 1, 2, and 3.

Conse-

~

quently, technical evaluations of the requests for exemption and one item of clarification of the requirements, as submitted in the above correspondence, r -.are provided.

f I

.g g.

t l

1

-6 I

A, -

hMl Franklin Research Center 3

A Cheme siTm FrsWn bu wne

l C

q I

. 6 I

k-CS257-5/6/176

~

2.

EVALUATION CRITERIA l

I Code of. Federal Regulations, Title 10, Part 50 (10CFR5G), Appendix J, i

I Containment Leakage Testing, wat specified by the NRC as containing the

\\'

criteria for the evaluations.

Where applied to the followiz:g evaluations, the a

criteria are either referenced or briefly stated, where nee:essary, in support of the determinations or conclusions. Furthermore, in.re. cognition of plant-specific conditions that could lead to requests for -ramption not ' ~

explicitly covered by the regulations, the NRC directed that the technical i

review constantly ecphasize the intent of A'ppendix J, that stential containment' atmospheric leakage paths,be identified, monitc:ced, and maintained 6

~

~

below established limits.

4-

..a e r

.* e,1.-

.-%w

• kd*

n..=.@

e e+

em e.

e E

4 9

e l

t.

r

.I o

i l

l l

4A

• he u

y~

r

(

E R-C5257-5/6/176 3.- TIX'HNICAL EVALUATION 3.1 REQUESTS EUR EXEMPTION FROM THE REQUIREMENTS OF APPENDIX J Reference 5 superseded the partial responses to the NRC's generic letter,

. hich had been submitted by TVA in References 2, 3, and 4.

In Reference 5, w

TVA requested exemptions from the requirements of Appendix.J in eight categories.

7he. current status of these requests for exemption, as modified f

by subsequent correspondence, is provided below:

Exemption Recuest Stat 21s

1. Venting of the containment Exemption withdrawn by Reference 7.

inerting system during System to be modified to perrait Type A testing.

venting.

2. Vesting of the conainment Exemption withdrawn by Reference 7.

,l air dilution system during

le.Systar; to be nodified to permit f

Type A testing.

venting.

.(

3. Draining and venting of Te=porary exemption requested while instrument sensing lines excess flow check valves are,,

during ?/pe A testing.

installed.

~

4. Drairiing and venting of Te=porary exe=ption requested while the seal water supply of the test connections are installed.

I reactor recirculation pumps during Type A testing.

5. Type B testing of containment Permanent exemption requested from airlocks.

the frequency of the Pa (peak

~

calculated accident pressure} test.

6. Reverse direction Type C Permanent exemption requested for l

testing cf certz.in isolation three valves.

valves.

7. Reduced pressure Type C Permanent es emption requested to testing of main steam test at 25 psig rather than Pa.

isolation valves.

8. Type C testing with water Exemption request for certain in lieu of air or nitrogen valves withdrawn in Reference 8.

.as a medium.

These valves are now tested with air.

d.

l ELUU Franklin Research Cer.:er

[_

1 A %.s w r - %.

~

i

}

l

~

l TEE-C5257-5/6/176 l

~

i I

Since the exemption requests for Items 1 and 2 have been co=pletely.

withdrawn, these items were not evaluated. Technical evaluat. ions of the remaining six items are pres'ented below, a3cng with the additional exemptions requested in Reference 7.

l 3.1.1 Draining and venting of Instrument Sensing Lines Durina Tvoe A Testing In Reference 5, TVA requested an exemption from Sectiob III.A.1. (d) for instrument sensing lines which are not drained and vent during the rype A test.'

IVA stated:

"In order to maintain adequate indication and control for afe cperation of the unit, instrument sensing lines are not drained and vented' inboard and outboard of the excess flow checic valve.

The instrument system, outboard of primary containment, is a r-'med, qualified system.

However, excess flow check valves are functionally tested once each operating cycle to ensure 'their integrity. A permanent exemption from W *iv J venting-and'drai=ing recuirements is requested for this item."

In Ref'erence 7',

TVA further stated:

~

"In regard to the propcsed exemptien for instrument sensing lines (item 3 of the February 26 letter), we continue to recuest a permanent

}-

exemption from paragraph III.A.l(d) with regard to venting and draining these lines.during the type A test.

A debiled evaluation which demonstrates that Browns Ferry units 1, 2, and 3 are in accordance with the provisions of Regulatory Guide 1.11 for instrument lines that are part of a protection system and connected to the primary reactor coolant system is discussed in our response to AIM question 5.15 dated March 25, 1971. An additional 17 instrument lines, which were not addressed in this response because they are not part of this protection system, do not meet the requiremen'a of Regulatory Guide 1.11.

These instrument lines are ir:entified in the following '"r M a.2.1.2.

"~ecive flow chet;k valves will be installed in these liner, and temporary exe=ption to the require =en*a of Regulatory Guide 1.11 is requested until N 4 7 i n e M l a tion. "

Evaluation q

Regulatory Guide 1.11, Instrument Lines Penetrating Prir "*y Keactor Containment, provides a basis for ireplementing General Design criteria 55 and i

56 regarding the isolatien of instrument lines penetrating the containment.

4 O

LOJ Fren'rlin Research Center A MY t/ rht frBWPt Mh#f

I i

p c

s

(

TER-C5257-5/6/176 This guide permits the use of self-actuated excess flow check valves to' isolate these lines as a compromise between the competing functions of maintaining the operability of the instrument and preventing contairment outleakage. However, the tjuide does not consider the closed system outside containment sufficiently rupture-proof to rely on it to prevent containment leakage.

'ihe guide states:

"The probebility of such a rupture [a component in an instrurent line outside containment) is considered to be sufficient 1r

~

high that the calculated offsite exposures that might result from such a single failure during normal operation should be substantially below the guidelines of 10CFRlOO." 'Iherefore, leakage testing in accordt.nce with Appendix J should be conducted.

Section III. A.l. (d) of Appendix J does not require the venting and draining duik;g.. Type A testing of systems that are,;.equired to maintain the.

plant in a safe condition during the test. Accord //g5to th'e T7A submittal,

.(

the instrument lines are in this category. - At the same time, however, Section

~

I l

III. A.l. (d) requires Type C testing cf centainment isolation valves which are not drained and vented during the Type A test.

Consequently, no exe:.ption 1

from the requirements of Appendix J is required provided that a Type C leakage test,is performed together with the periodic functionality. test of the excess I

flow check valves or at other convenient intervals as required by Appendix J.

j Results from these local leakage rate tests should be used to back-correct the Q'pe A test results to ensure that the integrated containment leakage rate is I

within specification.

3.1.2 N2 4 9 f nc and ventino of the Saal Water Sucoly to the Reactor Recirculation Pu=Ds During Tvoe A Testing In Reference 5, 'IVA requested' a permanent exemption from the require-l tents of Appendix J to exclude the seal water supply to the reactor recircu-lation pumps from the draining and venting requirements of Appendix J.

TVA stated that this seal water supply could not be vented downstream of the l

3 inboard isolation valves.

In Reference 7, T.'A stated ihat this requent applied to the 3/4-inch supply line frc= the control rod drive (CRD) hydraulic control and that test connections would be installed to perform the recuired 9

y*

A

l Tr2-C5257-5/6/176 draining and venting.

In Reference 7, WA requested that a tc=porary exemption be granted in lieu of the original request for permanent exemption while the modification is being installed.

Evaluation -

~

WA is committed to modifying these lines to permit compliance with the requirements of Appendix J in future Type A tests.

Sinc.e the line involved is a 3/4-inch line which cannot be vented without the modification, FRC finds a temporary exemption to be acceptable.

f.

~

3.1.3

?_ge B Testinc of Containment Airlocks In Reference 5, WA requested an exemption from the requirements of Appendix J to permit continued testing of contniment airlocks at the~ Browns -

. :&Pru Ferry plant at 2.5 psig within a short time after each cpening and at no.decs _

than Pa once-each operating cycle.

In Reference 7, WA stated that it would modify its technical specifica-

.tions to require the reduced pressure test within 3 days of the first of each

~ heries of openings whenever containment integrity is required, but reiterated

~

its request for a permanent exemption from Appendix J to require the Pa test once each operating cycle rather than every 6 months.

WA's basis for this request is that, to conduct the Pa test, the unit

~

must be in cold shutdown and the containment deinerted for entry into the centainment to install the holding devices needed to keep the inner door from unseating.

r,a-ance of -ecive radiation exposure to maintenanch,,and test perscnnel during the installation of the holding devi!$5s and test performance, these employees cannot remain inside containment.

In order to provide an ex.it cnd entrance during this test, containment integrity is broken by providing cccess through the CRD hatch.

In addition, WA's experience has shown that...- w.

the inner door acts as a variable orifice when tests at Pa are conducted, and high leakage rates into containment are observed through the inner door seals.

Since there is no detectable leakage through the inner door seals during pressurizatien from the proper direction (i.e., during tihe Type A test), WA O -- rn - - - - -. - - - -. - - - - - - -.

('

(

\\

(

TER-C5257-5/6/176 concludes that tests conducted at Pa are ultraconservative and of little value l

in determining overall leakage from containment via the airlock asse=bly.

Experience has also shown that reduced pressure tests, without holding clamps, will not only demonstrate that the door seals are intact, but will also d:monstrate that the entire airlock ' assembly is intact.

WA reiterated its belief that the 6-month test at Pa adds an, unnecessary burden leading to increased manpower requirements, personnel exposure, and

~

less of revenue and only results in unrealistic leakage rates.

TVA continued to request a permanent exemption to paragraph.III.D.2 of Appendix J.

TVA

~

inoicated that it would develop procedures to determine a method of mserva-tively extrapolating the leakage obtained during reduced pressure tests to the leakage' rate that would be experienced under accident conditions.

To verify

.-that there is no deterieratien of the airlock' assembly, WA also proposed tob-%-

c conduct reduced pressure tests within 6 months of the first of each series of

.(

openings whenever contain=ent integrity is required.

T B eluaticn l

Sections III.B.2 and III.D.2 of Appendix J require that containment nirlocks be tested at Pa at 6-month intervals and af ter each opening in the interibt between 6-=cnth tests.

Taese requirements were impos'ed because airlocks represent pctentially large leakage paths which are more subject to human error than other containment penetrations.

Type 3 penetrations (other than airlocks) require testing in accordance with Appendix J at intervals not to exceed 2 years.

~ '

Apper. dix J was published in 1973. A compilation of airlock events from Licensee Event Reports submitted since 1969 shows th.at airlock testing in recordance with Appendix J has been ' effective in the prompt identification of airlock leakage, but that rigid adherence to the af ter-each-epening require-crnt may not be necessary.

Since 1969, there have been approximately 70 reported airlock leakage trsts in which measured leakage exceeded allowable limits. Of these events, 251 were the result of leakage other than from imp' roper seating of airlock O W r,..-

. s r~

I

.l C

C

.r I.

{

TERkS257-5/5/176 door seals. These failures were generally caused by leakage past door operatin'g mechanism handwheel packing, door operating cylinder shaf t seals, I

equalizer valves, or test lines.

These penetrations resemble other Type B or C contain: dent penetrations except that they may be operated ucre frequently.,

Since airlocks are tested at a pressure of Pa every 6 months, these pene-trations' are tested, at a minimum, four times more frequently than typical Type B or.C penetrations.

The 6-month test is, therefore, considered to be both justified And adequate for the prompt identification of this leakage. ~

Improper seating of the airlock door seals, however, is not only the most frequent cause of airlock failures (the remaining 75, but also represents a potentially large leakage path. While testing at a pressure of Pa after each . opening will identify seal leakage, it can also be identified by alternative methods such as pressuriring between double-gasketed door sah (for airlocks I designed with this. type of seal) or pressuriring the airlock to pressures - other than Pa. Furthermore, experience gained in testing airlocks since the l. issuance of Appendix J indicates that the ::se of one of these alternative methods may be preferable to the full-pressure test _of the entire airlock. [' Reactor plants designed prior to the issuance of Appendix J of ten do not have the capability to test airlocks at Pa without' the installation of strong-backs or the performance of mechanical adjustments to the operating mechanisms r of the inner doors. The reason for this is that the inner doors are designed f to seat with accident pressure on the containment side of the door, and there-the operating ' echanisms were not designed to withstand accident pres-i

fore, m

l sure in the cpposite direction. When the airlock is pressurized for a local airlock test' (i.e., pressurized between the docrs), pressure is exerted on the l airlock side of the inner door, causing the door to unseat and preventing the I performance of a meaningful test. ' ':he strongback or mechanical adjustments prevent the unseating of the inner door, allowing the test to proceed. "te installation of strongbacks or performance of mechanical adjustments is time consuming (of ten taking several hours), may result in additional radiation exposure to operating personnel, and may also cause degradation of the operating mechanism of the inner door,'with consequential loss of reliability O M Fr%'in Reservh Frnter

C C TER-CS257-5/6/176 cf the airlock. In addition, when conditions require frequent openings over a short period of time, testing, at Pa af ter each opening becomes both impractical (tests of ten take from 8 hours to several days) and accelerates the rate of exposure of personnel and the degradation of mechanical equipment. ' For these reasons, FRC concludes that the intent of Appendix J is satis-fied, and the undesirable effects of testing af ter each opening are reduced, , if a satisfactory. test of the airlock door seals is performed within 3 days of each opening or every 3 days during periods of frequent openings, whenever - containment integrity is required. The test of the airlock door seals may be performed by pressurizing the space between the double-gasketed seals (if so Quipped) or by press'urizing the entire airlock to a pressure less Wm Pa that does not require the installation of strongbacks or performance cf other ~ mechanical adjustments. If the reduced pressure airlock test is to be employed, the results of the leakage test must'be conservatively extrapolated to equivalent Pa test results. - WA maintains that iirlock testing at Pa every.:S months adds an ? unnecessary burden leading to increased ~ manpower requirements, personnel expcsure, and loss of revenue to cbtain unrealistic leakage rate results. FRC .does not believe that WA characterizes the airlock test results at Pa as unrealistic when the holding devices are in place but rather when the holding ' devices are not used. With the holding devices in place, tiie entire airlock assembly is tested at Pa (simulating accident conditions), except for the gasket of the inner door which is conservatively tested since the ac',-M gasket seating force (Pa pressure within the containment) is much larger than the seating force =aintained by the holding devices 1. s,isting the internal airlock test pressure. If the holding devices are not capable of sealing the inner door during the Pa test, this is a problem which has not been identified ~ by other BWR licensees. Consequently, FRC finds that a Pa air test every 6 months is essential to ensure the integrity of the entire airlock assdly and. %at WA's request to test airlocks at Pa once each operating cycle is unacceptable. FRC concurs with WA, however, that when 6 ;nonths has passed since the last cuccessful airlock test at Pa and there have been no airlock entries in i O -g- $.2 Frenkb Rescerth Cer:.er A In.nen ed TM Frarman huomm

C c ~ ( N -C5257-5/6/176 n-the interim, titere is inadequate justification for shutting down the unit, and deinert'ing 'the containment to perform an airlock test at Pa. In fact, when the airlock has not been opened since the last successful Pa test, the reasons for testing airlocks more frequently than other Type B penetrations (e.g., more frequent use, more prone to human ~ error) are inapplicable. Consequently, FRC finds t' at'the interval between 6-month tests may be extendied up to one year, h provided that there have been no operations of the airlock _ since the last successful Pa test and provided that a Pa test is performed following the hiext airlock entry. In conclusion, FRC finds TVA's proposal to test contaizzment airlocks once each operating cycle at a pressure of Pa and every 6 months at a pressure of 2.5 psig to be unacceptable. FRC finds that containment airlocks must be tested at 6-month intervals at a pressure of Pa _in accordance with Appendix J, except that this testing interval may.be extended if there have been no. . airlock operations since the last sucessful Pa L ir test. and if a Pa test ris.- a perfor=ed following the next airlock cpening. FRC finds that testing in,, accordance with these requirements is acceptable in meeting the intent and -- objective of Appendix J. FRC further finds that TVA's proposed exemption from Appendix J to test airlocks at 2.5 psig rather than at Pa af ter each opening is acceptable if the test is accomplished within 72 hours of each cpening or every 72 hours during periods of frequent openings, and if the test results are conservatively extrapolated to the'Pa test results. For the Licensee's information, the following correlation has been found to be conservative in ex'trapolating the mass flow rate at pressure Pt (i:t) to the mass flow rate,at. pressure Pa (ma) I (assuming constant temperature; Pat = atmospheric pressure): ha (Pa + Pat) - (Pat) 2 (Pat) ~ ' ist (Pt + Pat) The above airlock requirements should be in effect whenever containment integrity is required. M *e*,.

( (_ s TER-C5257-5/6/176 3.1.4 Reverse Direction Type C Testing of Certain Isolation Valves In Reference 5, TVA identified 27 valves for which test pressure could not be applied in the direction that the valves would be required to perform their safety function. TVA stated that these valves could be tested, by local " pressurization in the opposite direction and, with few exceptions, that they were also subjected to Type A differential test pressure. In Reference 7, TVA indicated that the response of Reference 5 did not consider valve type, only pressure direction. A further ' review revealed that only 3 of the 27 valves originally identified were not water sealed to prevent the escape of containment air, were pressurized for testing from the opposite direction, and had leakage rates that could not be conservatively established when pressurized from the opposite directi.on.- These valves are single-wedge gate valves which, because of plant design, cannot be pressurized in the directioninwhichtheyperformtheirsaf6ty'dfunctiop. TVA requested 'a permanent exemption from the' requirements of Appendi7g J to permit continued' -( reverse direction testing of these three valves.,,w.,, Evaluation - i Section 'III.C.1 of Appendix J permits Type C testing of containment isolation valves in a direction other than that in which the valves perform f their safety function, if it can be shown that the resultIs of testing are l equivalent to or more conservative than results of testing in the direction of I the s'afety function. For valves which are relied upon to perform a containment \\ l isolation function ~ during post-accident cor.ditions, there does not appear to be a satisfactory justification for an exemption from thesa requirements when

...s1 a shewing of equivalent or more conservative results is not possible. "

..-: 9 g In the case of valves FCV l-55, FCV 71-2, and TCV 73-2, TVA has stated that they are not water sealed to prevent the escape of contain=ent air. Further, they are single wedge gate valves which are not capable of meeting the equivalent or more conservative test of Section III. Col. The 'f act that-plant design does not currently permit testing in the direction of accident pressure is not, in itself, sufficient justification for an exemption from the l ^ t p.~ ~ 7;.. c s i *, ). U, g.t,,, g.1i a,. ,'4 Edhd Frankfin,"Research Center I, d f.n i. A t> on e he Fre =M hsevte

( L l TE5257-5/6/176 (~ : requirements and intent of Appendix J. Consequently, valves FCV l-55, FCV 71-2, and FCV 73-2, should be tested in the direction of their safety ~ function. An exemption in this case is not acceptable. Reduced Pressure Tvoe C Testino of Main Steam Isolation valves' (MSIVs)_ 3.1. 5 In Reference 5, TVA requeste'd exemption from the requirements of Appendix J to permit continued testing of MSIVs r.t 25 psig rather dum Pa. TVA stated that the design of the Browns Ferry MSIVs makes it impossible to conduct, leak rate tests at Pa since Pa must be applied in the wrong direction. t Evaluation Section III.C.2 of Appendix J requires that containment isolation valves be local leak rate tested at,a pressure of Pa. The design of the main steam system in cost operating EWR plants necessitates leak testis;1pf:.the MSIVs by ' k..e pressurizing between the valves. The MSIVs are angled in tNhaih steam lines to' afford-better sealing in the direction of accident 1sakage..Because of ~ this design, a test pressure of Pa acting on the inbo[r"d disc tends to lift. the disc ef f the seat, thus resulting in excessive leakage into the reactor ~ vessel. The NRC staff considered this matter when the original test pressure of 25 psig was established for these MSIVs, at 'the design -stage of the plant. Since testing of the MSIVs at the reduced ~ pressure between the valv.es th'e gives rise to a greater leakage than Pa applied upstream of the valves, testing procedure results in a conservative determination of the leakage rate through the valves. Consequently, FEC finds that TVA's proposal to test MSIVs at a pressure 'of 25 psig by pressoriring between the valves, is ar. acceptable exempticn 'to the requirements of Sectica I.II.C.2. 3.1. 6 Tvoe C Testino with Water in Lieu of Air or' Nitrocen as a Medium In Reference 5, TVA requested permanent exemption from the requirements - - of Appendix J to permit testing of approximately 59 isolation valves with water in lieu of air or nitrogen as a medium, stating that the syste=s. involved could not be drained to permit the pneumatic tests. TVA also stated g IM Frank,w rr. arch Center Rese

w. w.=

i ( ( i, TER-C5257-5/6/176 ( that water represented tiie medium which would be present in cas'e of leakage from these valves during accident conditions. In Reference 7, however, TVA stated that *.he request of Reference 5 had censidered that Appendix J inferred that all valves be tested with air. Since the submittal of Reference 5, TVA determined that the real intent of' Appendix J was to prevent containment air. leakage t'o the outside atzmosphere. In support of this contention, TVA referred to paragraphs II.D, II.N, III.C.2, and III.C.3 of Appendix J. Based upon this determination, TVA stated that the ~ requests for exemptions for the 59 valves of Reference 5 were no longer ~ t i necessary. TVA proceeded to identify the isolation val.es at the Erowns Ferry f plant ir. four categories, describing the confor=ance with _ Appendix J a'ssociated with each as follows: _Catecory Conformance with 3ccera h J '~ 1. valves to be' tested with wat.cr This category has been deleted as a medium with results con- ... by Reference 8. These valves are verted to equivalent air now tested with air. leakage. 2. Valves water realed by the Appendix J does not require air suppression chamber. testing of these valves because atmospheric leakage is prevented by a water seal from the ~ suppression pool throughout the post-accident pariod. 3. Valves in safety systems Appendix J does not require air which are water sealed, testing of these valves because atmospheric. leakage is prevented by a closed loop, sei ic Class I system which is water sealed throughout the post-accident 1 period. These safety systems are designed so that no loss of water seal occurs despite a possible 2 single active failure. 4. Vr.lves which are air tested. Tested in accordance with Sections II.D, II.N, and III.C of Appendix J, except for the 'exe: ption for MSIVs. k E Frank:in Research Ce :t-r % e, r-o. =

L ( -TER-C5257-5/6/176 Evaluation ~~ FPC conctirs With TVA's interpretation of the requirenents of 'Appendi2It J as stated above. No further evaluation ~ of this item is undertaken since exemptions from the requirements of Appendix J are no longer requested. 3.1. 7 Tvbe C Testino of Valves in the Reactor Buildino Closed Coolina hte'r (RBCCW) System In Reference 7, TVA reinstated its request of Refererx::e 3 which had been inadvertently omitted from Reference 5. In Reference 3, TVA had requested an exe=ption from the Type C testing requirements of Appendix J for the isolation val.ves of the RBCCW system. TVA'sbasisforthisrequestisthatthe} valves are in a closed-loop system external to the reactor process system, operate at a pressure greater than 75 psig (greater than design peak accident pressure), operate centinuously, and were not designed to permit routine testing.of the valves.

• m,.

Evaluation .- { TVA's basis for this exemption request is that t ie RBCCW system is a

, closed system inside containment which continuously operates at pressures

~ greater than containment accident pressure. However, in order for a closed system to qualify as a barrier to leakage which prevents the isolation Walves I from being relied upon to isolate the containm'ent, the closed system must meet i i certain criteria relative to its post-accident integrity. The criteria for a closed system to qualify es a containment isolation barrier are given in Standard Review Plan (SRP) 6.2.4. Among other things, l the system must be protected against missiles and pipe whip, must be designed Mer.- l to seismic Category I, and -a meet safety-c:. ass piping requirements. Unless _ all the requirements of SRP 6.2.4 are met, the system can be postulated to l in which case the isolation valves will be relied upon to perform a

rupture, I

containE5nt isolation function and therefore =ust be Type C tested. The closed system described by TVA in Reference 3 does not meet all the l l criteria of SRP 6.2.4. Consequently, valves 70-47 and 70-506 must be Type C 1 .,4 $ Frank!in Research Center %.in r,.,o.,w.m i

( ( . 4 O' 6 '::'ER-C5257-5/6/,176 tested since they will be relied upon to prevent the escape of containment air to the outside atmosphere in case of a post-accident rupture of the RBCCW system. 3.1. 8 Tvoe C Testing of Traversing In-Core Probe (TIP) Svstem Valves In Reference 7, TVA requested a permanent exemption frcun Type C testing requirerents for the TIP ball valves (penetrations 35A thrc>ogh 35E) and a temporary exemption for the nitrogen supply to the TIP indexers (penetration 35F).intil facilities can be installed to permit testing. TVA's basis for 7I this request is that any addition of valves in penetrations: 35A through 35E to - permit testing could cause interference with or failure of -the TIP probe as it passes through these valves.. These penetrations are now strbjected to the dif-ferential pressure of the Type A test, and in TVA's opiniorm, the cost and reduction in syster reliability would not be offset by the d=provement in con-tainment integriti. Evaluation .(7

.l.

R Although the TIP penetratiens are small lines, because: of the.m=nber of lines involved, tire potential for a rakage of containment a~ sphere can be substantial and does not justify permanent exemptions.

Fur-the rmor e, another BWR licensee has successfully tested these valves without imstalling additional valves in the lines by disconnecting the TIP tub >es at fittings just inside the drywell. This technique is now in effect at sev eral BWR units. ~ Consequently, FRC fincs that TVA's proposal to permanently exempt these lines from Type C testing is unacceptable and that these' valves ubculd be tested in accordance with Appendix J. E ftemporary exemption, while r:odifying the nitrogen supply to the TIP indexers, is acceptaole.

3. 2 CLARIFICATION OF TEE REQUI' DLENTS OF APPENDIX J R

In Reference 5, TVA stated: "For clarification of the requirements of Appendix J, we _ wish to advise that, in summing Type C test results to determine whether Type B and C test results satisfy the O UU!! FraN& Research Cemer A r-m r,.%.

.l TER-C5257-5/6/176 Appendix J limit (total less than 0.60 La), we sum path leakage-that is, the greater leak rate of either of the pair of valves in a leak path is assumed to be the path ~ leak rate." Evaluation-The purpose of Type C testing of containment isolation valves in .accordance'with' Appendix J is two-fold. First, it provides for leak check of ' containment isolation barriers which are operated frequently and thereby more ~ subject to degradation of integrity than the passive ba'rriers (i.e., Type _C tests are required at each shutdown for refueling and are not to exceed 2 years, whereas Type A testing, occurs approximately every 40 months).

Second, Type C testing verifies the leak-tightness of each isolation valve, whereas Type A testing verifies the leak-tightness of. penetration's, generally with.

' two shut isolation valves in series. Consequently., the Type C test provides for the case in which a single active failure:.caurcs one of the two isolation r. valves in series to remain open. " i t-rq By summing pat'h leakage, TVA achieves both of the ' desired results of' the Type C testing program. Each valve is tested at the required frequency,a'nd the testing accounts for possible single active f ailure, where the valve with the lesser 1eakage of the two in series is conservatively assumed to have. I ~ 2 remained open. Consequently, FRC finds that TVA's procedure of su= ming path i leakage, where the path 1_akage is assumed to be the creater leak rate of the two valves tested, is in accordance with tne requirements of Appendix J. One concern with TVA's procedure for sum =ing path leakage should be noted. Namely, when the initial su==ation of the greater leakage rate of each pair of valves awaaAc O. 6 I.a, such that r'epairs must be accomplished, the, Licensee must be can.ful to e *bt post-repair sum =ations continue to include only the greater leakage, rate from each pair of valves. FRC's concern is that once repairs have been performed on one of the tiwo valves of a particular penetration, the other valve may now be the one with the greater'- leakage rate. In this case, the path leakage rate is now the leakage rate of the valve that was not repaired rather than the new leakage rate of the l. recently repaired valve. As long as the Licensee ensures that path leakage , \\ -16.- gg dd Franklin Rescerrh Center j A Dnesson s/ The franktn instr.Aa

.,- s TER--C5257-5/6/,176 (' rate is always the leakage rate of the greater of the two leakage rates for a particular path, there is no problem with the TVA procedure. FRC interprets TVA's statement to mean that, regardless of subsequent repairs, the -total path leakage which 1 ust not exceed 0.6 La will always include the greater of the two leakage rates for a particular penetration., FRC's concern is expressed 'only to provide complete clarity. l ~ -(

m.. -

i l - U00] Frakrjin Researrh dex.er a % a w r. w a.

( ( l e TER-C5257-5/6/176 ~ " -4. CONCLUSIONS ~ Technical evaluations of the outstanding submittals. by TVA, relative to the implementation of the requirements of 10CFR50, Appendir J, at Browns Ferry gaclear Plant Units 1, 2, and 3, have been conducted. These evaluations include TVA's req ests for exemption from the requirements of Appendix J as ' well as one item of clarification of the requirements. The conclusions of these evaluations are presented below: Instrument lines penetrating the reactor containment do not require o draining and venting during Type A testing provided the lines are Type C tested as required by Section III. A. l. (d). A temporary exemption from the requirements of Appendix J to exclude o draining and venting of the seal water supply to the reactor recirculation pu=ps during Type A testing is acceptable sM'le a' modification is being installed to permit future draining and. venting. This modification,df _not already accomplished, should be completed as soon as possible. 9g g ee.. e TVA's exemption request to test containment airlocks once eac'h o operating cycle at a pressure of Pa is unacceptable.,Airlocks must be ' { tested at 6-month intervals at Pa as required by ' Appendix J, except l that this testing interval may be extended up to one year if there have been no airlock cperations since the last successful Pa air, test and if a Pa test is performed following the next ai,rlock opening. TVA's exemption request to test airlocks at 2.5 psig rather than at Pa o af ter each opening is cceeptable if the test is acco=plished within 72 hours of each opening or every 72 hours during periods of frequent openings and if the test results are conservatively extrapolated to the Pa test results. A TVA's exemption request to test main steam line drain valve FCV l-55 o in the direction opposite to that of its safety function is unacceptable. This valve must be tested in the direction of its {.' safety function. ~'"JA's exemption request to test steam supply valves to the RCIC and EPCI turbines, FCV 71.2 and 73-2, in a, direction These opposite to that cf their safety functions is unacceptable. valves should also be tested in the direction of their safety function. TVA's exemption request to test main steam isolation valves at 25 psig o by pressurizing between the valves is acceptable because the procedure results in a conservative determination of the leakage rate through -the valves. t i ..g M Frankhn Rese:n.h Cemer A Cm.cn af N FreeM, Wow

( Q s TER-CS257-5/6/176 TVA withdrew exemption requests to test certain isolation valves with ( o is in water in lieu of air or nitrogen because TVA's proposed testing [ co=pliance with the requirements of Appendix J. lves TVA's request to exempt reactor building closed cooling water o 70-47 and 70-506 from Type C testing is not acceptabl*e because these valves can be relied upon te> perform a containment isolation function. These valves should be Type C tested. TVA's exemption request to exclude the traversing in-core probe ball o valves fro = Type C tasting is unacceptable. These valves must be tested in accordance with Appendix J. A te=porary' exemption while modifying the nitrogen supply to the trip indexers is acceptable; however, the modifications, if not yet accomplished, should be ~ ccepleted as soon as possible. 7 o, TVA's interpretation cf the requirements of su ing the leakage from Type C tests to determine acceptability satisfies the requirements of The Licensee's proposed method of 'snmirs the path Appendix J. leakage from Type C tests is acceptable provided that the path leak' age will be rechecked after repairs. x m 5 l ~ i E .. ~ . en O h lb Fr:nk'in Rese:rch Center A ce.i n. r,e %.

l', (. U o. c' TER-C5257-5/6/176 5. REFERENCES ~ T u. ~ 1. NRC Generic Letter Implementation of 10CFR50, Appendix J l August 4, 1975 2 J. E. Gilleland (WA) Letter to B. C. Rusche (NRR) September 9, 1975 3 H. G. Parris' (TVA) Letter to B. C. Rusche (NRR) ~ October 10; 1975 C' 4. J. E. Gilleland (WA) Letter *to B. C. Rusch'e (NRR) ~ January 15, 1976 5. J. E. Gilleland (TVA) Letter to B. C..Busche (NRR) February 26,.1976 6. NRC Letter to G. Williams, Jr. ../.3 *"' December 27, 1976

---

7.

J. E. Gillelar.d (WA) Letter to A. Schwencer (ORB 1) July 8, 1977 l 8. L. M. Mills (TVA) Letter to T. A. Ippolito (ORB 3) l'ay 27,1980 9. NRC Letter to H. G. Parris (WA) April 21, 1980 A, d[3 Franklin Research Center A Com on of The Froen inssw}}