ML20153F351

From kanterella
Jump to navigation Jump to search
Safety Evaluation Accepting Util Latest Design Mods to MSIV Actuation Control Sys,Per IEEE 279 & GDC 21
ML20153F351
Person / Time
Site: Nine Mile Point, 05000000
Issue date: 02/07/1987
From:
Office of Nuclear Reactor Regulation
To:
Shared Package
ML20150F672 List:
References
FOIA-88-356 NUDOCS 8809070234
Download: ML20153F351 (7)


Text

,\, UNITED STATES E E e NUCLEAR REGULATORY COMMISSION

+{'*

{. #

, WASHING TON, D. C. 20555  ?

%, .....f $.. j .1 g, - y.

q$y.ltg .4 .

SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION q_ D..g RELATED TO THE MSIV ACTUATION LOGIC DESIGN Tf r.'

AT NINE MILE POINT, UNIT 2 3'<"

t DOCKET NO. 50-410 'Y f.l[

[<[ .

1.

[ Q:.

Uf. Ep

1.0 INTRODUCTION

J .N PI w k.dp# g.

3

( The main steam isolation valve (MSIV) actuator control system for Nine f;g Mile Point, Unit 2 (NMP-2) was rrodified recently to compensate for GQ.

deficiencies found within the mechanical portion of the original MS!V f ,]

actuator design. The original actuator design consisted of a mechanical 9 -[

latching mechanism to hold the valve open and two normally energized

"*p';*]

g

{ solenoid operated spring plungers to release the mechanism and pemit spring closure of the valve. Each solenoid was powered from a separate, h independent Uninterruptable Power Supply ((!PS) system, and it took tile deenergization of both solenoid spring plungers to initiate closure of the associated MSIV (i.e., disengage the mechanical latch). Testing of the 5 original design revealed that in some cases the force produced by the h solenoid operated spring plungers was inadequate to trip the trechanical 3 latch.

e Subsequently, the licensee decided to eliminate the use of the mechanical iatch trip system described above and to utilize the hydraulic system, originally intended only to open the MSIVs, to maintain the MSIVs in the g open positic'1. This re-design proposes using the solenoid operated valves m

(SOVs) which are part of the original hydraulic system. These S0Vs will g be normally energited to keep the solenoid valves closed, thereby maintaining the hydre,'alic pressure to keep the MSIVs open. Decnergiration of any one of the SOVs will cause rapid closure (release of hydraulic F pressure) of the associated MSIV. In an attempt to enhance MSIV

[ availability and to be consissent with the original MSIV design, the r-8809070234 880e10 L PDR FOIA KUDLICKGB-356 PDR ,/fp

licensee proposed to provide power to each SOV from either of two independer.t UPS systems through automt. tic transfer circuitry in order to assure that the loss of one power supply system would not deenergize (open) either SOV. After review of the modifications performed on the MSIVs, the staff informed the licensee by letter dated Jenuary 14, 1987 that the autcratic transfer logic was not in full compliance with the requirements of 10 CFR 50, Appendix A, General Design Criterion (GDC) 21 and IEEE Standard 279 [10 CFR 50.55 a(h)).

By letter dated January 15, 1987, the licensee comitted to remove the automatic transfer system and informed the staff of a revised method of supplying power to the electrical SOVs associated with the MS!V actuttion system. Tho staff's evaluation of the latest MSIV design i modifications propnsed for the MS!V actuator control system is provided s below. {

2,0 EVALVA110E The modified NMP-? MS!V actuator logic control circuitry will utilize relaylogic(coil-to-contact)operationtoassurethatactuationofa single emergency trip sensor (i.e., one-out-of-two in trip system A or B) will not cause inadvertent closure of the MSIVs. This is consistent with the original design basis of NMP-2 whereby the logic is set up as a one-out-of-two taken twice logic (i.e., one-out-of-two in trip system A l and B are required to close the MSIVs). To reiterate, the automatic h transfer between the redundant Class IE reactor protection system (RPS)

UPS systems has been rcr.oved. Thus, a loss of a single RPS UPS system will deenergize its associated 50V and result in the closure of the associated MSIVs which is consistent with the fail-safe design feature.

Also, the relay logic design alle, s th., i kensee to perforr, their required Technical Specification monthly channel functional test surveillance requiremetics and provides coil-to-contact isolation between

(

the redundant logic channels. The staff questioned the use of these relays as cualified isolation devices. In response (letters dated January 15 and February 4, 1987), the licensee described the relays and comitted to perform maximum credible fault tests to qualify the relays as acceptable isolation devices. The device has been identified as a

, ITE/Telemechanique (Gould-Type J10) relay. The relay contains a metal barrier that separates the coil section and its associated wiring from the contact section of the relay and its wiring. The licensee has confirmed that the subject relays are seismically and environmentally cualified for their safety-related application in accordance with IEEE 344-197G and 10 CFR 50.49 respectively.

in the February 4,1987 letter, the licensee provided an outline of the test procedure used to qualify the Gould J10 relays and provided a sumary of the test results. The staff audited the test procedure used to demonstrate the qualification of the subject relays as acceptable isolation devices. The test procedure identified the maxirum credible fault voltagu and current to be 119 Vac and 15 amps respectively. After review of the analysis (supplied by letter dated January 15, 1987) used to detemire these fault values, the staff considers 110 Vac/15 amps to be acceptable fault test values fe qualification of the relays. The nereistance eritpris penvided 1H the telt ifrutedute wat found (b be acceptable (i.e., upon application of the r:aximum credible fault, the isolation between the coil and contact will be maintained and no spurious signals will be transmitted across the barrier). The test sunmary (provided in the February 4,1987 letter) shows that the relay perfomed in accord 0nce with the established acceptance criteria stated above (i.e., no spurious signals were observed when the mhximu:n credible faults were applied in the transverse modes). Based on the results of the tests and the actual relay configuration (metal barrier between the coil and contact portions of the device), the staf# concludes that the Gould J10 relays as installed in the MSIV actuator control system are acceptable isolation devices.

i

. 4 f Upon further review of the HSIV logic design modifications, the staff l 3 questioned the use of redundant fault fnterrupting devices for isolation between the Class IE and non-Class 1E circuits. The latest modifications E show that the non-Class 1E MS!Y partial stroke test solenoid valve will

$ be powered from the RPS Class 1E bus through redundant 3 amp fuses. The

( licensee has stated that the fuses are procured and installed as Class 1E

~

,- devices and that both are properly coordinated with the upstream

[ protective devices. In their February 4, 1987 letter, the licensee E confirmed that a continuous 3 ampere load will not degrade thz ability of E , the Class IE sy: tem to perfom its required protective function (i.e., . -

S the Class IE system is designed for and rated to support the 3 ampere currentdraw). Further, the licensee has verified that the replacement i of fuses is covered by the hMP-2 Standing Order No.14. "Operations h Departnent Fuse Control Program," which will enstre, through quality g assurance review, verification that a correct size fuse will be -

installed. The staff finds this acceptable.

The staff pursued the issue related to periodic verification of operation _

of the redundant emergency trip SOVs associated with each MSiv. The licensee previded information in their January 15 and Febmary 4,1987

[ letters to avdress this issue. The correct performance of tbs NSIV 4

) emergancy trip actuatien solenoid valves can be verified during MSIV full .

I closure testing. The MSIVs will be full stroke tested at each cold '

p shutdown in accordance with Section XI of the ASMF Boiler and Pressure Yessel Code (Part IWV 3412). This commitment is described in the NMP-?

preservice inspection plan pump cnd valve program and is based upon Technical Specification Section 4.0.5 which requires compliance with ASME g _

Section XI. The licensee has further co rtitted to revise Operating _

L Surveillance Procedure (OSP) NP-MSS-CS-001 to require positive confirma-a tion of the functioning of each MSIV actuator solenoid during each cold

=

shutdown full stroke test. Reed switch position indication is provided in h -

the control room for the SOVs whereby the operator can verify proper solenoid valve operation. The OSP identified above should be revised

i i 4

)

prior to the next full stroke test required by ASME Section XI. The licensee has also verified that the Electrical Protection Assembly (EPA) f setpoints are coordinated with the load requirements of the SOVs and that calculations were perforred to ensure that both the high and low voltage i EPA trip setpoints maintain the operating voltage within allowable values.

l

3.0 CONCLUSION

1

]'

Based on the above evaluation, the staff concludes that the latest design nodifications made to the MSIV actuation control system comply with the 1

applicable regulatory ecceptance criteria including the requirenents of 1 IEEE 270 and GDC 21 and are, therefore, acceptable.

l 1

.I 1

I i

i i

i l

1 l

l i I s

4 I

1

i ENCLOSURE 2 E!CSB/ DBL SALP INPUT FLANT: Nine Mile Point, Unit 2 LICENSEE: Niagarc Mohawk Pcwcr Corporation DOCKET NO: 50-410 LICENSEE STATUS: OR SER SUBJECT- Safety Evaluation Report INPUT PERFORMANCE PARAMETER $:

Management Involvement in 4 suring Quality Approach to Resolution of Technical Issues From a Safety Standpoint Response to NRC Initiatives (4 Staffing (IncludingManagement)

Reporting and Analysis of Reportable Events Training and Qualification Effectheness ,

I Any Other SALP Functional Area  !

t  ;

PERFORMANCE NAARAtiVE LEERIPTION OF CATEGORY / {

PAPAMETER RATING AOPL!tANi&lfENSEE'S PERFORMANCE 1 The licensee did not appear to adequately 3 l understa/.d staff policies ano did not make decisions based on Ldequate managerent ievolvement. An appropriate level of manacement was not present and significantly involved at the various review reett.ngs held with i the licensee. '

2 The licensee's submittals showed that there 3 was not an adeouate understanding as to the infomation necessary to resolve various issues. The approach to resolve the issues appeared to be viable but the infomation

2-PERFORMANCE NARRATIVE DESCRIPTION OF CATEGORY /

, PARAMETER APPLICANT / LICENSEE'S PERFORMANCE RATING provided to resolve the issues was lacking significantly in thoroughness and depth and net minimum requirerrents. Much effort was expended by the staff to provide guidance to the applicant necessary to resolve many of the issues. Repeated requests and clarifications of requirements had to be made to obtain necessary information. The lack of sufficient information has caused much delay in the resolution of the issues.

3 Thelicenseerespondedpoorly(lackof 3 thorouanness) to concerns raised by the staff.

In particular, where design changes were made ,

that required staff re-review, the applicant  !

was reluctant to provide sufficient details to allow the staff to complete its review, The staff has spent considerable effort to obtain an acceptable resolution of the issue (i.e., the NRC staff had to generate specific guidance on inferr.ation needed to resolve the issues).

OVERALL APPLICANT / LICENSEE PERFORMANCE RATING 3 I

l l