ML20205D800

Applicant Answers to State of VT Second Set of Interrogatories & Requests for Production of Documents to Vermont Yankee Nuclear Power Corp.* Certificate of Svc Encl. Related Correspondence

ML20205D800
Person / Time
Site:	Vermont Yankee File:NorthStar Vermont Yankee icon.png
Issue date:	10/19/1988
From:	Gad R, Reid D ROPES & GRAY, VERMONT YANKEE NUCLEAR POWER CORP.
To:	VERMONT, STATE OF
References
CON-#488-7341 OLA-2, NUDOCS 8810270210
Download: ML20205D800 (40)

Text

SYl

) $QATED CORMESPONDENCl k l:LHLTLC g (b N- 0 UNITED STATES OF AMERICA NUCLEAR REGUIATORY COMMISSION '88 OCT 24 P3 :17 before the ,.y ,  ;

ATOMIC SAFETY AND LICENSING BOARD  !:'.(

~

)

In the Matter of )

) Docket No. 50-271-OLA-2 VERMONT YANKEE NUCLEAR ) (Tosting Roquirements for POWER CORPORATION ) ECCS and SLC Systems)

)

(Vermont Yankee Nuclear )

Power Station) )

)

APPLICANT'S ANSWERS TO STATE OF VERMONT'S SECOND SET OF INTERROGATORIES AND REQUESTS FOR THE PRODUCTION OF DOCUMENTS TO VERMONT YANKEE NUCMAR POWER CORPORATION Interrocatory No. 1.

Question:

1. Please identify all persons who participated in the preparation of answers to these interrogatories and production requests, and identify the portions of your response to which each person contributed.

Besconse:

Don A. Reid: 3, 4, 8, 9, 11, 13, 14, 15, 16 and 20.

John T. Herront 2, 3, 4, 7, 8, 9, 10, 11, 13, 14, 15, la and 20.

Jay K. Thayer 6, 8, 9, 11 and 17.

Kevin J. Burns: 5, 6, 11, 12, 13, 15, 16 and 19.

1 i

B810270210 PDR 001019ADOCK 05000271 PDR G

l 0

}5

h

' Interrocatory No. 2.

Questi2Dt

2. Do any of the American Society of Mechanical Engineers (ASME) inservice testing provisions from which the Applicant was granted relief, pursuant to 10 CFR 50.55a(g), in 1980 (as being "impractical") relate to

testing of Emergency Core cooling System (ECCS) or Standby Liquid Cooling (SLC) System components?

Resnonset Please note that Vermont Yankee identifies this to be Question 1 of the questions propounded by the Board during the pre-hearing conference in this matter on June 28, 1988, see II. 40 (6/28/88), and now provides its response to the Board's question.

Requests for relief from certain ASME inservice testing provisions were granted in 1980 for ECCS and SLC equipment.

These will be produced for inspection and copying at the offices of Vermont Yankee Nuclear Power Corporation, Brat-tieboro, Vermont, at a time mutually convenient to counsel.

Technical Specification related alternate testing does not form the basis for the relief requested in any of these Cases.

Vermont Yankee points out that the 1980 requests for relief are no longer those under which VYNPS operates.

Requests for relief from ASME Section XI testing requirements currently being implemented for ECCS and SLC components hre set forth in a document distributed to the Board and parties under date of July 28, 1988 (PVY 88-63). The requeste r _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . __ _ _ _ __ . _ - _ _ . _ - - _ _

applicable to the components that are affected by the proposed license amendment are as follows:

Service Water RRB-PI, RRB-V1, RRB-V2 SJ4: RRB-P2, RRB-V20, RRB-V21 Diesel Fuel Oil Transfert RRB-P5, RRB-P6 Diesel Air Start System RRB-V4 Core Sorav RRB-V14, RRB-V15, RRB-V16 HPCIt RRB-V17 EliEt RRB-V22, RRB-V23 RCIC RRB-V24, RRB-V25 Generic Relieft RRB-GP1 RRB-GP2 RRB-GP3 RRB-GV1 RRB-GV2 RRB-GV3 A review of the above Relief Requests was performed and it was determined that Technical Specification related alternate testing is not involved as a basis for the relief requested in any case.

Interroaatory No. 3.

Question:

3. For each component for which tests are proposed to be eliminated, what are the testing intervals under the ASME inservice tosting program?

RRaponset Please note that vermont Yankee identifies this to be Question 2 of the questions propounded by the Board during the pre-hearing conference in this matter on June 28, 1988, see II, 40 (6/28/88), and now provides its response to the Board's question.

For all of the ECCS and SLC equipment for whict slter-nate testing is proposed to be eliminated and for which there is an ASME test requirement, the ASME inservice testing program test interval is quarterly (though other Technical Specification provisions may require more frequent testing).

t r

Please note that the ASME tests do not apply to all of the l t

components for which alternate testing is proposed to be j deleted. These other components will remain subject to existing Technical specification provisions for which no amendment is proposed.

Interroaatory No. 4.  !

Question:

4. Are there any differences in the types of tests carried out under the ASME inservice testing program and the tests which are proposed to be eliminated? Describe any differences.

Etaponse: l Please note that Vermont Yankee identifies this to be  ;

Question 3 of the questions propounded by the Board during

• the pre-hearing conference in this matter on June 28, 1988, see II. 40-41 (6/28/88), and now provides its response to the Board's question.

Tests conducted per the ASME inservice testing program are performed using test methods that are identical to those currently used for Technical Specification related alternate testing, with the following exceptions:

a) During pump testing per the ASME program, a number of quantitativa eaacurements are taken at various pump and driver locations. These measurements may not be required to be taken during alternate testing. (E.g., vibration, flow, pressuro.)

b) Run times for diesel testing are different. Please see our Response to Interrogatory No. 15 of the State of Vermont's First Set of Interrogatories.

Interroaatory No. 5.

Question:

5. What is the historic out-of-service frequency for each valve or component for which testing is proposed to be eliminated?

Response

Please note that Vermont Yankee identifies this to be Question 4 of the questions propounded by the Board during the pre-hearing conference in this matter on June 28, 1988, see Tr. 41 (6/28/88), and now provides its response to the Board's question.

Components may be out of service for two main reasons repair (unscheduled) o routine preventative maintenance (scheduled). Routine preventative maintenance is generally performed when the reactor is shutdown and the component in question is not required to be operable, a situation to which the existing requirements for alternate testing do not apply.

Vermont Yankee therefore interprets this interrogatory to be limited to the out-of-service frequency on those occasions where components are under unscheduled repair.

Repairs are required whenever a component failure occurs. Component failures are almost always detected at the test demand, since demands due to an actual accident signal are very rare. Thus, we can estimate the out-of-service frequency for these components from their failure rates as observed at tests.

Failure rates for each component can be calculated from plant data for the number of failures and the total number of test demands and/or operating hours. This type of data was compiled in the Report for the Core Spray System pumps and valves, and for the diesel generators. Associated failure rates (out-of-service frequencies due to failures) are also provided in the Report in Tables 5.7 and C-3.

Note that there are two types of units used to quantify failure rates: failures per demand and failures per hour.

Since valves operate by either opening or closing, valve failure rates are given per demand. Pump and diesel failure to start rates are also given per demand. However, since pumps and diesel generators run after they are **.arted, failure rates are also given for failure per hour of run time.

Detailed plant-specific failure data for other com-ponents was not compiled in the Report. However, failure rates for similar components are expected to be approximately the same. By "similar" we mean most motor-operated valves are similar, and most standby pumps are similar. An exten-sive industry-wide data base has been developed for perform-ing probabilistic studies of nuclear plant systems. This generic data bmse includes values for motor-operated valves, standby pumps, and diesel generators. This data is presented l

1

}' in Tables 5.5 and C-1 of the Report. Note that the generic i data combines similar components (i.e., all motor-operated i valves) from different systems into a single group.

l Comparison of the Vermont Yankee plant-specific data to l the generic data shows that the Vermont Yankee plant-specific failure rates for the Core Spray System components and diesel i

j generators are lower than the generic values. However, since we have not performed detailed, plant-specific analyses for J other systems, the generic values give a reasonable estimate a

! of failure rates for components in other systems.

Interroaatorv No. 6.

j ouestiont

6. What is the projected service life, in both time and

! number of occasions used, for each valve or component for which testing is proposed to be eliminated?

{

I

Response

Please note that Vermont Yankee identifies this to be Question 5 of the questions propounded by the Board during

]

the pre-hearing conference in this matter on June 28, 1988, l see II, 41 (6/28/88), and now provides its response to the Board's question.

Please see our response to Interrogatory No. 8 of the h

l State of Vermont's First Set of Interrogatories, Attachment A Note 3. As set forth therein, components such as pumps I

l and valves in Emergency Core Cooling Systems are designed to l

function for the life of the plant. Typically, the plant I

! life is assumed to be the duration of the operating license 1

l l

l

• I or 40 years for Vermont Yankee. Conservatism in design can l I j lead to a significantly longer service life. l The service life of a component (such as a pump) does l l not necessarily mean that the pump will continue to run for {

t

! that period of time without the routine replacement or  ;

1 l refurbishment of constituent parts (such as a bearing). The l

\

1 frequency of such repair or refurbishment depends on 6Leh factors as how often a component is demanded and the type and i frequency of maintenance performed.  :

Interroaatory No. 7.

l Questiont  ;

. l

7. Are any of the valves for which testing is proposed to I be eliminated "check valves", within the meaning of I&E l 4

Bulletin 83-03 (March 10, 1983)?  ;

l i Responset

Please note that Vermont Yankee identifies this to be ,

j Question 6 of the questions propounded by the Board during

) i i the pre-hearing conferenca in this matter on June 28, 1988, l

) see Ir. 41 (6/28/88), and now provides its response to the

! (

l

. Board's question. [

l f Yes. Two check valves were identified in response to {

l I&E Bulletin 83-03: Valves v70-43A and V70-43B. These valves r' are currently tested during normal emergency diesel generator I

l surveillance as well as during any alternate testing as- .

1 i sociated with the diesel generators. In addition, the internal integrity of these valves is verified at least once [

t every five years. [

l l

f L

f

Interrocatory No. 8.

QuestiED:

8. On the list of active components and devices necessary to perform the safety function of the systems (Attach-ment A of "Applicant's Answers to State of Vermont's First Set of Interrogatories and Requests for the production of Documents to Vermont Yankee Nuclear Power  :

Corporation," - hereinafter, "Applicant's Answers"), .

identify which items are instrumentation and control l items.

Repuonse i

This interrogatory is vague, because it does not supply a definition of what the interrogating party intended to include within the set of "instrumentation and control items" for the purpose of this question.  ;

, The components listed in Attachment A to our response to i Interrogatory No. 8 of the State of Vermont's First Set are i i

all nochanical components and electrical components. Those components are generally not considered to be I&C equipment, though they may be controlled by I&C equipment. Vermont  ;

I Yankea therefore believes that, if it correctly interprets l l

this interrogatory, the answer is "None."  !

J l Interrocatory No. 9.

Question l

9. Are there any instrumentation and control items (sen- ,

sors, actuation devices, relays, logic devices, etc.)

not listed on Attachment A of "Applicant's Answers" which are part of the systems for which testing is '

proposed to be eliminated, and which I a. Are required to perform the safety function of the

! system? Identify these instrumentation and control items in the same format provided in Attachment A  ;

I of "Applicant's Answers,"

l t

l

b. Could fail and cause inoperability of the system?

Identify these instrumentation and control items in the same format provided in Attachment A of "Applicant's Answers."

c. Are all of the instrumentation and control items identified in paragraphs a. and b. above tested under the testing of alternate systems requiremer.ts which are proposed to be deleted? If there are instrumentation and control items which are not tested under current requirements, identify each item and describe why it is not prosently tested.

d. Since Applicant's proposal is to amend the design bases of the systems to rely on the ASME XI testing program in lieu of testing the alternate system in the event that one redundant train is out-of-service, are any of the instrumentation and control items from paragraphs a. and b. above not included in the Applicant's Inservice Testing Program? If so, identify which items are not included. For these items, explain why Applicant considers it acceptable to eliminate their testing from the design bases.

Resoonset Vermont Yankee objects to this interrogatory, on the ground that it relates entirely to equipment for which the proposed amendment proposes no change in testing require-monts, and is therefore beyond the scope of the admitted contention. Vermont Yankee notes that the subject of testing for logic systems was recently addressed in connection with License Amendment No. 106.

Without waiving this objection, but rather relying thereon, Vermont Yankee supplies the following information:

(a) -

(d). The identification requested does not exist in any readily retrievable form and providing the identifica-tion requested would require the performance of original research which has been estimated to require 300-500 man i

hours of effort at a cost of $20,000 to $30,000. The specific information could be extracted by the State of Vermont from analysis of the following documentations (i) -

surveillance procedures, (ii) Control Wiring Diagrams and (iii) Process and Instrumentation Diagrams. If inspection of these documents is requested, please contact counsel for i

Vermont Yankee. J I

Interroaatory No. 10.

Questions

10. Are there any items listed in Attachment A to "Ap-plicant's Answers," which are tested at frequencies .

greater than once every three months? If so, identify I each item.

Etaponse:

All of the items listed on Attachment A to "Applicant's f

Answers" are tested more frequently than once every three months, except the il Fan for the west cooling tower, which >

f does not have a specific surveillance interval requirement, I but is normally not tested more frequently than every three r

months. 6 Interroaatory No. 11.

Question:

1

11. Section 5.3.1.1.2 ("Use of Time-Related Models") of i' NUREG/CR-2300, "PRA Procedures Guide," January 1983, contains the following statements:

"Standby components are usually subjected to periodic testing, which occurs, for example, once a month or perhaps once a year. The time between tests is the length of time the component is  !

exposed to failure without detection, and hence the  !

term "fault-exposure time." ... The fault-exposure  ;

l i

r t

l

time is usually determined from plant procedures, but some caution should be used when examining a system for test intervals. As an example, consider

[a system thatD is tested in various pieces; that ds, the logic :,s tested once a month, as are che spray pumps. The sensors are calibrated once a year and tested once a year through the logic.

However, the entire system is never tested end to end. This results in a specific contact never being tested during the life of the plant."

a. Are there any portions of the systems for which testing of the alternate system is proposed to be eliminated which contain a combination (sensor - >

logic - > Device), as indicated in Section 5.3.1.1.2 of NUREG/CR-2300, which is not tested during the life of the plant? Identify each case,

b. If there is a portion of a system which is not tested (either never, or over a long period) is it not possible that the time-related unavailability, Q t , could be large enough to obscure the effects identified in Figure 5-1 of "The Report?" If this were the case, what would be the value of eliminat-ing testing of the alternate system when the overall effect on system (un) availability would be negligible?

c. For each portion of the Core Spray System iden-tified in paragraph a. above, identify how the fault exposure time was determined for the analysis in "The Report."

d. Describe how Applicant would perform a Probabilis-tic Risk Assessment (PRA) on the portions of systems in paragraph a. which are not tested. If there are any differences between your method and that described in Section 5.3.1.1.2 of NUREG/CR-2300, please explain the differences.

c. In Section 9.5 of "The Report," it is stated, "Because of the similarity in types and numbers of components that must actively function during a true demand, conclusions based on quantitative results for the Core Spray System are used in the evaluation of testing requirements ... for other pumping systems." This appears to be inconsistent with your response to Interrogatory No. 28, "First Set," where it is stated that a portion of the SLC system is not tested. How would the system unavailability curve for the SLC system dif fer from that of the core Spray System, considering the high time-related failure unavailability, Qt, of the "squib" valves? Provide qualitative and quantita-tive responses.

f. In response to Interrogatory No. 28, "First Set,"

it is stated that the "the squib valves...could not be, and should not have been 'taken into account' by the Report." But does not section 5.3.1.1.2 of NUREG/CR-2300 identify how the squib valves 'could' and 'should' have been taken into account in the Report?

• l Responset "1

a. We are not aware of any portions of the subject systems that are not tested during the life of the 1

, plant, and we believe that there exists no such portion. However, in order to establish this fact definitively, the effort described in our response I to Interrogatory No. 9 would have to be performed.

Such an analysis was performed, however, in connection with License Amendment No. 106, and it

]

l demonstrated that there was no logic system within i

the scope of that amendment, or associated sensors, i

not tested at least once during the life of the plant.

b. Figure 5-1 of the Report considers the portions of the core Spray System that are tested as part of alternate testing. Since these are the components

) involved in the proposed change, the analysis was limited to these components. If other devices are postulated to dominate the total time-dependent j availability, such that alternate testing has a I

I  ;

i j

]

I negligible impact on overall system availability, i then the elimination of alternate testing would j have a negligible impact on the health and safety of the public. If this were the case, then the proposed amendment would satisfy the applicable criteria for approval. ,

c. No portions of the core Spray System were iden-tified in Part a, above.

d. No portions of the subject systems were identified in Part a, above.

a. The title of the Report ist "Impact of Alternate Testing on component and System Availability." The system components considered in the Report are ,

those involved in alternate testing. As stated in response to Interrogatory No. 28, "First Sett" j "detonation of the squib valves is not (and could not be) within the scope of alternate testing." i Thus, a "system unavailability" curve that includes l

the squib valves is not meaningful to this amend- h ment request.

l

f. NUREG/CR-2300 is a general guide for performing i Probabilistic Risk Assessment (PRA) analyses. As such, it provides guidance for analyzing systems in the context of a PRA. For a PRA, actuation logic '

l and sensing devices are normally considered in 1 f

developing a system fault tree (failure model). l i

i I

t

Although it utilizes probabilistic techniques, the Report is not a PRA.

The squib valves are not subject to alternate testing, nor could they be, since for testing l

purposes it is not possible to "fire" the squib valves while the plant is in operation. Since the purpose of the Report was to evaluate the impact of alternate testing on component and system availa-bility, and since the squib valves are not subject to alternate testing, those valves should not have been and consequently were not taken into account in the Report. i Interrocatory No. 12.

Question:

12. The Applicant's analysis described in the report entitled "Impact of Alternate Testing on Component and System Availability (hereinafter, "The Report")," does not explain how the "Time-Related Fractions," in Tables ,

5.7 and 5.8, are used in the time-related failure rate equations on pages 6, 7 and 27.

a. Provide an explanation.

b. Industry time-related failure ratt,s are provided for pumps and valves, respective),y, in NUREG/CR-1205, Revision 1, Data Summaries of Licensee Events Reports of Pumps at U.S. Comar,rcial Nuclear Power Plants, January 1982, and NT' REG /CR-1363, Revision 1, Data Summaries of licensee Event Reports of Valves at U.S. Commercial Nuclear Power Plants, October 1982. Why were these values not used instead of time-related fractions?

Does the use of time-related fractions result in c.

the expected increase over time in Qt, un-availability due to potential time-related failure.

Explain specifically how this is so.

I

< 1 t

l i

Responset

a. Given the followingt

1. a functional failure rate, F (total l i

failures per demand), l

2. a time-related fraction, f, I

a failure rate per demand (Qd) and a standby time-related failure rate (lambda) are calculated as follows:

Od = ( 1 - f ) F i

lambda = f

• r/730 The valttes of F and f used in the base {

I case analysis are given in Table 5.8 of the  ;

Report. The number 730 is the number of hours f I

in a month, which is the current surveillance

! interval. i i

Qd and lambda appear in the equations on Pages 6, 7 and 27 of the report.  !

b. NUREG/CA-1205 and NUREG/CR-1363 previde  !

[

failure information for pumps and valves.  :

i These reports also calculate failure rates ,

i based on this failure information. Two types j i

of failure rates are presented, and are  :

r t

calculated as follows  ;

i Demand Hour Rate = Total Number of Failures Total Number of Hours j Demand Rate = Tctal Number of Failures  !

Total Number of Demands  !

f; t

._ -I

1 T M se failure rates do not contain information regarding the mechanism (i.e.,

demand-related or time-related) of the f allen . They are simply two different rates, 1

each based on the total number of failures, l

l but expressed in different units. These l

failure rates are not approptiate for use in the equations on Page 27 of the report, since what is required for these equations ist railuru Rate Per Demard = )Raber_of_Demnd-RelarM Failums -

'Ibtal ) Amber of Dumards Stardby railure Rate - M =*=r of T 6 aalm e M rallures

' Ictal Nur@er of Hours l

Note that the numerators nf the NUREG definitions are not the same as the numerators l

l of the Report definitions. (If one summed the numerators of the Report definitions, we obtain the total number of failures, while if we summed the numerators of the MUREG defini-tions, we obtain a number equal to two times the total ntrAber of f ailures.) Thus, the WUREG failure rates are not appropriate for l use in the report.

c. As discussed in Part a above, the time-related fraction is used to calculate a time-related failure rate, lambda. Given this value of lambda, the equations on Pages 6, 7, and 27 l

i L

show that the unavailability due to time-related failures increases with increasing time.

Interroaatory_No. 13.

Questiont

13. For the outage events described in Attachment 5 to "Applicant's Answerst"

a. Which items are on the list because they were out of service for planned preventative maintenance?

b. Which items are on the list because they were out of service due to non-operabiliYy/non-functioning in a test?

c. Which items are on the list because they were out of service due to failure while running (a running pump or fan fails, a modulating valve fails)?

l 7

d. On Attachment B of "Applicant's Answers," which are l the six (5) items identified in tasponse to l Interrogatory No. 14c, "First 66t," as occurrenr<4 1 in which testing of the alternete systek resul'oo j in failure which caused power reduction? Plea.,

1 identify the causes for each failure.

j ResponAnt

a. please see our Answers to State of Vermont's First Set of Interrogatories, Interrogatories 14(a) and i

i (b).

b. "The Report" cospiles a list of itsas that failed a surveillance test for Core Spray and Diesel Generators. No similar list for the remaining l

) systems has been compiled, and to prepare such would require original research which is estimated to require approximately 120 manhours of research j and effort and a cost approximately $7,500.

i I -

18-O 1

- - - . , , - - - , - - , . , , , . . - _ _ _ , _ ,----.._._,_.,n.-__--,,..,__.n-n n -n.,,

f

c. In order to categorize the unavailabilities raported in Attachment B in the fashion requeqted, we would have to perform the ant. lysis described in cur response to sub-part (b), above,

d. The cause of each failure and associated event can be found in the following Licensee Event Reports:

LER 83-20 LER 83-27 LER 83-32 LER 84-22 LER 88-19 LER 88-20 LER 87-68 These are within the set of documents that Vermont Yankee has previously undertaken to produce for inspection and copying at its offices in Brattleboro, Vermont, in response to the Inter-venor's prior set of interrogatories and document production requests.

Interrocatory No. 14.

Question:

14. Of the 175 times in 5 years that testing of alternate systems has been required (Attachment B of "Applicant's Answers"):

a. How many times did the alternate system fail?

b. How many times did the alternate system fail and repair of the alternate system was accomplished before power was reduced?

c. How cany times did the alternate system fail and repair af the alternate system was not accomplished before power had to be reduced? r

d. How many times did the alternate system fail and repair of the original system was accomplished before power was reduced?

l

Response

a. This informatio:9 is contained in Vermont Yankee's Answer to State of Vermont's First Set of Inter-rogatories, Interrogatory 14 (c) .

b. As set forth in Vermont Yankse's Answer to State of Vermont's First Set of Interrogatories, Inter-rogatory 26(b):

"Such is not the practice. Initiation of an

'LCO shutdown' is required at VYNPS whenever redundant trains are inoperable, and this requirement will not be changed by the proposed amendment."

c. This interrogatory requires that both the original and the alternate system have failed. As set forth in our previous response to Interrogatories Nos.

14(c) and 2C/'), this has occurred six times. In each case, the init4ation of an LCO shutdown

- occurred immediately.

d. As set forth in Vermont Yankee's Answer to State of Vermont's First Set of Interrogatories, Inter-rogatory 26(b):

"Such is not the practice. Initiation of an

'LCO shutdown' is required at VYNPS whenever redundant trains are inoperable, and this requirement will not be changed by the proposed amendment."

Interrocatory No. 15.

Questign:

15. Table 5.6 and Appendix B of "the Report" contains a description of Observed Fai3 4 ms for Vermont Yankee core Spray System. The data used .or the analysis are based on these failures. Yet Attachment B to "Applicant's l

l l

l Answers" lists Core Spray failures which are not identified in Table 5.6 or Appendix B.

a. Provide a description in the Appendix B format for each of the following:

1) Core Spray 7B out-of-service, 8-26-83

2) 'B' Core Spray out-of-service, 10-12-83

3) 'A' Core Spray out-of-service, 2-9-84

4) 'A' C/S, 'B' C/P out-of-service, 2-20-85

5) 'B' C/S Pump out-of-service, 12-8-86

6) 'B' C/S Pump out-of-service, 12-19-86

7) 'B' C/S Pump out-of-service, 1-20-87

8) 'B' C/S Pump out-of-service, 4-2-87

9) C/S SA out-of-service, 12-11-87

10) 'A' C/S out-of-service, 2-12-88

b. Why were the above out-of-service events omitted from "the Report?"

c. Does consideration of the events listed in para-graph a. above alter the data used in Tables 5.7 and 5.8 of "the Report?" If yes, please explain how the data is altered.

d. Does consideration of the events listed in para-graph a. above alter the results of Figure 5-1 of "the Report?" If yes, please explain how the results are altered.

e. How many other failures end out-of-service events have occurred in the Core Spray System over the life of the plant which should be included in "the Report?"

f. Why are the following failures listed in Table 5.6 and Appendix B of "the Report" not identified in Attachment B of "Applicant's Answers?" For each failure listed, indicate why it did not result in testing of the alternate system in accordance with the Technical Specification?

1) C/S V-llA failure, 9-5-87

2) C/S V-5A failure, 12-8-87

3) C/S V-26A failure, 2-3-88

g. Are there instances of Diesel Generator failure identified on Attachment B of "Applicant's Answers" which should have been included in Report, Appen-dix C, Vermont Yankee Diesel Generator Failures?

If so, do these omissions affect your data, results and/or conclusions?

l i

a

Response

a. The descriptions follow:

1. Auaust 26. 1983. V-7-B This valve is normally open. After closing for a surveillance test, the breaker tripped while attempting to reopen the valve. The breaker trip was attributed to the attempt to l reopen the valve before it was fully closed and while the motor was still in motion. The breaker was reset, and the valve was tested successfully. This is not considered a l functional failure since the failure was attributed to the manner in which the test was performed. This could be considered a test-related failure, since this valve is normally in its safety position (open), and since the t

failure was attributed to the manner in which the test was performed. However, this failure was not included in the report because Valves 7-A and 7-B are no longer tested as part of alternate testing. ,

2. October 12. 1983. V-7-B r l This failure was of the same nature as the August 26, 1983 failure of V-7-B. The reasonc

! for not including it in the report are the  !

same.

i

3. February 9, 1984 The Core Spray System was declared inoperable due to a blown fuse while performing Core Spay System logic testing. This was not a failure of a component involved in alternate testing, hence it was Tst included in the report.

4. February 20, 1985. V-5 The Core Spray System, V-5, was declared inoperable to replace a wire from a terminal strip to a flow switch. The wire, although functional, was replaced with a wire that was qualified for a harsh environment. Since there was no observed failure, this event was not included in the report.

5. December 8. 1986. Pumo B Core Spray Pump B was made inoperable for testing of the motor. The motor was run uncoupled from the pump to take vibration readings. Since there was no observed failure, this event was not included in the report.

6. December 19. 1986. Pumo B This event was of the same nature as the December 8, 1986 event. The reasons for not including it in the report are the same.

7. Jgj1uary 20, 1987, Pumo B Core Spray Pump B was made inoperable by uncoupling the motor from the pump. The motor was balanced, and Pump B was returned to service. Since this was preventative main-tenance and did not involve a failure, the event was not included in the report.

8. ADril 2, 1987. Pumn B This event involved a differential pressure indicating switch, and Pump B was made inoperable for the repair. No failure of Pump B occurred, and alternate testing is not performed on the differential pressure indicating switch, hence this event was not included in the report.

9. December 11, 1987. V-5-A This event is discussed on Page B-3, Appendix B, of the report, under the date December 8, 1987. Thus, this failure was included in the report.

10. Februhrv 12. 1988. V-26A This event is discussed on Page B-3, Appendix B, of the report, under the date February 3, 1988. Thus, this failure was included in the report.

b. The response to Part a, above, states why each of the events Numbers 1 through 8 is not included in the report. The response to Part a, above, also states where the events, Numbers 9 and 10, are included in the report.

c. As stated in response to Part a, above, all of the subject events are either not applicable to tha report or are already included in the report.

Thus, consideration of these event does not alter the data in Tables 5.7 and 5.8 of the report.

d. For the reasons discussed in response to Part c, above, consideration of the subject events does not alter the result of Figure 5-1 of the report.

c. A review of Vermont Yankee Core Spray System failure data was performed as part of the study documented in the report. All failures that were found during the study, and that were judged to be applicable to the proposed amendment, are docu-mented in the report.

Further review of Vermont Yankee Core Spray System failure data was performed to respond to Part a, above. No new failures that are applicable to the report were found.

f. 1. Attachment B to Vermont Yankee's answers to the First Set of Interrogatories was derived from the Control Room Equipment Status logs,

which indicate equipment that is out of service when it is required to be operable.

The failure of Core Spray Valve 11A on September 5, 1987, occurred during a refueling outage and therefore the component was not required to be operable. Consequently, this failure did not trigger any requirement of alternate testing, and therefore it did not meet the criceria for inclusion in Attachment B. However, it was a failure, and therefore it was included in the Report.

2. Core Spray Valve SA failure on December 8, 1987 is the same event as interrogatory 15.a.9 of the State's second set of interrogatories.

The date discrepancy is the difference between the date a Maintenance Request was initiated and the date the valve was declared in-operable.

3. Core Spray Valve 26A failure on February 3, 1988 is the same event as interrogatory 15.a.10 of the State's second set of inter-rogatories. The date discrepancy is the difference between the date a Maintenance Request was initiated and the date the valve was declared inoperable.

- g. An "Index of System Status" was provided in Attachment B to Vermont Yankee's priors answers to inter-rogatories. This list comprises instances where one of the subject systems was declared inocerable. Appandix c of the report documents diesel general failure rate data. The terms "declared inoperable" and "failure rate data" are not synonymous. A system may be declared inoperable for reasons other than a failure.

A detailed evaluation of all instances of diesel generator out-of-service events has not been performed, ad performing such an analysis would require the obtaining of additional information and the expenditure of substantial amounts of engineer man-hours and associated dollars.

However, it can be assumed with a high degree of confidence that, were such an additional study performed, it would not affect the conclusions drawn in the Report. The reasons for this are as follows:

1. The data collection and review process undertaken during the preparation of the Report identified a number of diesel generator out-of-service events  ;

that were not applicable to the Report, 1.g., that did not involve a failure.

2. As noted on page C-4 of the Report, several "slow start" instances were detected during that study, which were included in the Report data as "fail-ures," even though it is unlikely in any given case that the "slow start" would have resulted in the diesel generator actually failing to perform its safety function. This results in an added degree of conservatism in the Repcrt.

3. The diesel generator mean failure rates used in the base case analysis are provided on page C-6 of the Report. The mean failure rates actually used are higher than the Vermont Yankee-specific failure rates shown on page C-5 of the Report. This results in an added degree of conservatism in the Report.

4. Figure 7-2 of the Report shows the sensitivity of the results to the d'esel generator failure to start rate. This figure shows that the basic conclusion is insensitive to the diesel failure rate. That is to say, daily testing produces higher unavailability than monthly testing, even if the diesel failure to start rate is increased by more than a factor of ten.

Interrocatory No. 16.

Question:

16. Concerning maintenance of the Core Spray systems l

a. Describe the maintenance required for Valves V-11, V-12, V-S and V-26. What is the maintenance frequency of these valves for the past 5 years?

How often, or in what percentage of maintenance activities, does maintenance result in declaring the system inoperable and thus requiring testing of the alternate system? How often, or in what percentage of maintenance activities, is the

1 1

maintenance able to be scheduled during plant outrage.

b. Describe the maintenance required for a Core Spray Pump. What is the maintenance frequency of this pump for the past 5 years? How often, or in what percentage of maintenance activities, does main-tenance result in declaring the system inoperable and thus requiring testing of the alternate system?

How often, or in what percentage of maintenance activities, is the maintenance able to be scheduled during plant outrage.

c. For an overall system unavailability analysis, what is the value which should be used for number of outages due to maintenance (per tims period)?

Resoonse:

a. The only routine maintenance that is scheduled for Core Spray Valves 11A/B, 12A/B, SA/B and 26A/B is an inspection of the Limitorque Operator in accordance with OP 5220 approximately once every third refueling outage. Routine preventive maintenance activities are normally scheduled during refueling outages when the system is not required to be operable. Unscheduled maintenance could be performed during power operation based on consideration of the factors listed in our response to previous interrogatories 16(a) and (b).

Based on a review of Attachment B from our response to the first set of interrogatories, Alternate Testing was required approximately five (5) times due to the inoperability of Core Spray Valves 11A/B, 12A/B, SA/B or 26A/B.

- OP 5220 and the maintenance records associated with the subject valves will be produced for inspection and copying at the offices of Vermont Yankee Nuclear Power Corporation, Brattleboro, Vermont, at a date and time mutually convenient to counsel in this proceeding.

b. The routine preventive maintenance activities that are scheduled for core Spray Pumps are:

- Annual vibration readings (performed during a normal surveillance test).

- Change of the motor bearing oil, lubrication, and visual inspection each refuel outage.

Inspection of the motor every third refueling outage.

Visual inspection of support brackets every fifth refueling outage.

The routine preventive maintenance activities are normally scheduled during refueling outages when the system is not required to be operable.

Unscheduled maintenance could be performed during power operation based on consideration of the factors listed in our response to previous inter-rogatories 16a and b.

Based on a review of Attachment B from our response to the first set of interrogatories, Alternate Testing was required approximately five (5) times due tc the inoperability of Core Spray Pump P46-1A/B.

The maintenance records associated with these pumps will be produced for inspection and copying at the offices of Vermont Yankee Nuclear Power Corporation, Brattleboro, Vermont, at a date and time mutually convenient to counsel in this proceeding.

Y

c. The value for "number of outages due to maintenance per time period" appropriate for use in an un- i availability analysis depends on the nature of the analysis being performed and on the model being used. For the analysis performed in the Report, which addresses the effect of the proposed amend-ment upon system unavailability, the approach is stated on page 5 of the Report:

"We assume that unavailability due to scheduled maintenance is independent of the testing policy. This is a conserva-tive assumption, since more frequent testing may actually lead to increased ,

component wear, which in turn would i require more maintenance."

Since alternate testing was assumed not to affect this value, no value was required in the model documented in the Report.

Interrocatory No. 17. ,

I ouestion:

17. Were "the Report" and "Applicant's Answers" prepared i under an applicable Quality Assurance program or under applicable procedures? Provide the procedures. If Applicant believes this request to be burdensome, then summarize the requirements of the program or procedures.

Were these procedures followed in the preparation, review and approval of these documents?

Resoonse:

Please see Vermont Yankee's Answers to State of Ver-mont's First Set of Interrogatories, Interrogatory No. 19(b),

which is equally applicable to those answers.

Interroaatory No. 18.

Ouestion:

r J 18. In order to meet a Technical Specification Limiting Condition for Operation (LCO) to be in COLD SHUTDOWN in 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, when must the decision be made to begin reducing power. When must actual power reduction begin?

Resoonse:

Vermont Yankee has historically interpreted the Limiting Condition for Operation to require that the plant immediately begin to reduce power once it has been determined that the

! LCO shutdown condition exits.

Interrocatory No. 19.

Question:

19. In response to Interrogatory No. 27b, "First Set,"

Applicant declined to provide an graphical representa-tion showing the sensitivity of results to the assump-tion of trying to repair a failed component while both trains were out-of-service, rather than proceeding immediately to safe shutdown. This evaluation has been performed, with results shown in Attachment A. This evaluation was performed using data and equations from "the Report" except that time-related failure rates were used from the NUREG's identified in Interrogatory No.

12b above. Results are plotted as in Table 5.1 of "the Report," and are shown for two cases:

1- Remaining at power when the second train fails, and attempting repair (the case evaluated by the Applicant), and 2- Proceeding immediately to safe shutdown upon failure of the second train (in this case the repair unavailability is inactive because the unit is shutdown).

a. Since Applicant's response the Interrogatory No. 26b, "First Set," states it is not the practice to attempt repair when both redundant trains are inoperable, why is the immediate shutdown plot (Item 2 above) not the proper result of the svaluation instead of Figure 5.1 of "the Report?"

b. The results in Attachment A correctly show the benefit from daily testing of the alternate system.

When the repair portion is removed, it is seen that elimination of daily testing and resorting to 30-day ASME XI tests roughly doubles the un-availability of the second train, assuming the first is already out of service. Considering this change, why is the correct risk avoidance decision for public health and safety not rather to retain the presently required daily testing of alternate systems and instead proceed directly to safe shutdown without repair, as stated in Applicant response to Interrogatory No. 27b, "First Set?"

Response

a. Applicant's response to Interrogatory 26(b), "First Set," stated:

"Such (repair in lieu of shutdown) is not the practice. Initiation of an 'LCo shutdown' is required at VYNPS whenever redundant trains are inoperable, and this requirements will not be changed by the proposed amendment."

The "immediate shutdown" curve provided in Attach-ment A to this Interrogatory is not the "proper result" of the evaluation because changes in the shutdown Limiting condition for operation (Lco) time period are not the subject of the proposed amendment.

i

b. Applicant's response to Interrogatory No. 27b, "First Set," stated:

"As discussed in Part a (response to Inter-rogatory No. 27a, First Set) above, the analysis in the report already considered the Technical 3pecifications for shutdown given that two ECCS subsystems are declared in-operable. S_nce no change in the duration of the Technical Specification shutdown LCo is being sought by this amendment, further analysis discussing different shutdown LCO durations is not meaningful."

This above statement explains why Attachment A to the second set of interrogatories is not ap-plicable.

Interrocatory No. 20.

Ouestion:

20. In the application for elimination of testing of alternate systems from the Vermont Yankee Technical Specifications (VY-TS), dated December 7, 1987, Ap-plicant states, the elimination of testing of alternate systems "is consistent with the testing requirements contained in BWR Standard Technical Specifications (BWR-STS)." However, the testing requirements in the BWR-STS, or lack of them, are predicated on meeting other requirements of the BWR-STS, as well as system and component design to the state-of-the-art at the time of issue of the BWR-STS (Revision 3, December 1980).

a. In BWR-STS Section 3/4.7.1.1, daily testing of the alternate system is not required for the residual heat removal service water system. However, in 3.7.1.1.c the system is only allowed to remain inoperable without this testing for 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> before shutdown is required. Vermont Yankee-TS Section 3.5.C.3 allows the system to be inoperable for a full 7 days before shutdown is required. If testing is proposed to be eliminated in accordance with the BWR-STS, why should not Vermont Yankee adopt the more stringent LCO of the BWR-STS?

b. In BWR-STS Section 3/4.7.1.2, daily testing of the alternate system is not required for the plant service water system. However, in 3.7.1.2.a.3 the system is only al. owed to remain inoperable without this testing for 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> before shutdown is required. Vermont 'ankee-TS Section 3.5.D.2 allows the system to be in parable for a full 15 days before shutdown is lequired. If testing is proposed to be elimitated in accordance with the BWR-STS, why should not Vermont Yankee adopt the more stringent LCO of the BWR-STS?

c. In BWR-STS Section 3/4.8.2, daily testing of the alternate system is not required for the 480 V Uninturruptible Power Systems. However, in 3.8.2.1.a the system is only allowed to remain inoperable without this testing for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> before shutdown is required. Vermont Yankee-TS Section 3.10.B.4, which refers to Section 3.5.A.4, allows the system to be inoperable for a full 7 days before shutdown is required. If testing is proposed to be eliminated in accordance with the BWR-STS, why should not Vermont Yankee adopt the more stringent LCO of the BWR-STS?

d. In BWR-STS Section 3/4.8.1, for the emergency diesel generators, one diesel generator is only allowed to remain inoperable for 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> before shutdown is required. Vermont Yankee-TS Section 3.10.B.1, which refers to Section 3.5.H.1, allows the system to be inoperable for a full 7 days before shutdown is required.

Further, the BWR-STS requires testing of the alternate diesel generator in Section 4.8.1.1.2.a.4 within one hour and at least once per 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> thereafter. In the BWR-STS both the LCO and the testing requirements are more stringent than Vermont Yankee-TS. Why should not Vermont Yankee adopt these BWR-STS requirements? Vermont Yankee-TS Section 3.5.C.3 allows the system to be in-operable for a full 7 days before shutdown is required. If testing is proposed to be eliminated in accordance with the BWR-STS, why should not Vermont Yankee adopt the more stringent LCO of the BWR-fe rS?

ResDonse:

Responding to this question is rendered difficult because the question appears to be based upon the apparent promise that there is, or can be derived, a direct comparison between LCO durations and the presence or absence of a 4

requirement for alternate testing. It also appears to be founded upon the premise that LCO durations can be directly compared for plants with dissimilar systems. Vermont Yankee does not accept either premise to be valid. Nonetheless, Vermont Yankee supplies the following information:

The basis for proposing the license amendment that is the subject of these proceedings was the enhancement to l safety that would result from the elimination of testing that is both unnecessary and itself a contributor to the un-availability of safety systems. Even assuming that the basis 1 for LCO periods in the VYNPS Technical Specifications included some relicnce upon the alternate testing, which is believed historically not to be the case, this incremental

^

increase in safety will be achieved without regard to any change in LCO periods.

Unlike the generic Standard BWR Technical Specifica-tions, which are used by the NRC Staff as guidelines in reviewing plant-specific technical specifications, the VYNPS

Technical Specifications take credit for specific plant features and enhancements that in some cases are not ap-I Examples include the Vernon l plicable to all BWRs.

Hydroelectric Station Tie Line, the Alternate Cooling Tower l

Sub-system, and the use of Dedicated Diesel Generators.

These types of differences between the generic and the plant-specific situations mean that direct comparisons between the l

I l

1 i

l

generic Technical Specifications and plant-specific Technical Specifications can be misleading.

For instance, in its comparison of section 3.7.1.1.C of the Standard Technical Specifications to section 3.5.C.3 of the VYNPS Technical Specifications, this Interrogatory fails to take note or apparent account of the fact that were one RHR/SW Pump in each sub-system to be declared inoperable, the VYNPS Technical Specifications require the plant to be in cold shutdown within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, while the Standard Technical specifications do not require the facility to be in cold shutdown until after 7 days.

Similarly, section 3/4.7.1.2.a.3 of the Standard Technical Specifications, relating to the Plant Service Water System, allows plants to continue operating for 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> under conditions of failure, where the cognate provision of the VYNPS Technical Specifications appears to allow for continued operation for 15 days. However, VYNPS has avail-

)

able to it the alternate cooling tower sub-system (a sub-l system not available in some other BWR cases and for which no credit is taken in the Standard Technical Specifications),

which difference may be sufficient to justify the different i

i LCO periods. Moreover, the 15-day LCO period only applies if this alternate cooling tower sub-system is available if that sub-system is not available (a situation more comparable l

to the Standard Technical Specifications), VYNPS is required to be in cold shutdown within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. This data suggests I

1 l

j

that, when rendered comparable, the VYNPS Technical Specifi-cations are more restrictive than the Standard Technical Specifications.

Similar analyses might be made of the cituations involving diesel generators and 480-volt power supplies: the Standard Technical Specifications take no credit for a power supply such as the Vernon Hydroelectric Station Tie Line, and differences such as these are sufficient to justify the VYNPS LCO periods. Given the differences in plant designs, features and resources, it suffices, we believe, to conclude that direct comparisons of the sort invited by this inter-rogatory do not produce either valid comparisons or ureful information.

i I

4 l

l 1

l i

t t

I i

_ _ _ _ _ . - _ - - . - . _ - - _ - _ _ _ _ _ . . . _ - . _ . _ m_., . - , , . , . . . _ _ _ _ _ - , - _ . . - _ _ . - -,

1.0. 1-9. 88 03 20PM = VERMONT YANMDC POWDR PC2 B

signatures Donald A. Reid, being first duly sworn, states that the foregoing answers are true, except insofar as they are based on information that is available to Vermont Yankee but not within his personal knowledge, as to which he, based on such information, believes than to be true, this f day of October, 1988.

hb Donard A. Reid Than personal d Donald A. Reid, before and per-

'd d;#% first duly a o , made oath sonally known to true, this day or

'that the forego ' statament october, 1988.

M*/ '.'l h N"U6 /

> < in @ck Notary Publio M  :

d (p. V +$'Consissionexpirast l

g,((( fc La ta ehAsetta t.19 gl- ^-

MOUN1jp ' C

,3gm

(~. K. Gad 11I .

}

_m R / '

Ropes & Gray / reat 225 Franklin St l Boston, Massachusetts 02110 (617) 423-61004 l

I l

1 AMr01 $3 era TOIET 00, 6T Do M 'd _ . .

O

'f m I I { l VYNs135 (OLA-2)

ASLB - Reg. Mail RKGCOSRG.VY CERTIFICATE OF SERVICE 10 DCT 24 P3 :17 crs 4 ,

I, R. K. Gad III hereby certify that op0CSis;1 Mudj.

October 19, 1988, I made service of the within document in accordance with the rules of the Commission by mailing a copy thereof postage prepaid to the following:

Charles Bechhoefer, Esquire, Samuel H. Press, Esquire Chairman Vermont Department of Administrative Judge Public Service Atomic Safety and Licensing 120 State Street Board Panel Montpelier, VT 05602 U.S. Nuclear Regulatory Commission Washington, DC 20555 Mr. Glenn O. Bright George B. Dean, Esquire Administrative Judge Assistant Attorney General Atomic Safety and Licensing Department of the Attorney Board Panel General U.S. Nuclear Regulatory One Ashburton Place Commission Boston, MA 02108 Washington, DC 20555 Mr. James H. Carpenter Ann P. Hodgdon, Esquire Administrative Judge Office of the General Counsel Atomic Safety and Licensing U.S. Nuclear Regulatory Board Panel Commission U.S. Nuclear Regulatory Washington, DC 20555 Commission Washington, DC 20555 Adjudicatory File Atomic Safety and Licensing Board Panel Docket (2 copies)

U.S. Nuclear Regulatory Commission Washington, DC 20555 mm -

('

\

R. K. Gad III L 'c_ E

<