ML20211N588

Safety Evaluation Re First Level Undervoltage Protection Testing.Testing Unnecessary

ML20211N588
Person / Time
Site:	Yankee Rowe
Issue date:	02/19/1987
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20211N572	List: ML20211N567 ML20211N588
References
NUDOCS 8703020137
Download: ML20211N588 (7)

Text

'

. nnan UNITED STATES

[ NUCLEAR REGULATORY COMMISSION y yE WASHINGTON, D. C. 20555

,f SAFETY EVALUATION REPORT BY OFFICE OF NUCLEAR REACTOR REGULATION FIRST LEVEL UNDERVOLTAGE PROTECTIOh TESTING YANKEE NUCLEAR POWER STATION YANKEE ATOMIC ELECTRIC COMPANY DOCKET NO.50-029 INTRODUCTION By letter dated June 30, 1986, (Reference 1), as supplemented October 22, 1986 (Reference 2) and December 4, 1986, (R.eference 3), Yankee Atomic Electric Company (YAEC), licensee for the Yankee Nuclear Power Station, submitted a basis for not performing monthly tests of the first level (loss-of-voltage) undervoltage protection and for not including the present refueling interval testing in the Technical Specifications (TS). Presented below is the staff's evaluation of this proposal.

BACKGROUND On June 3, 1977 (Reference 4), the NRC sent a letter to all licensees which established multi-plant Action Item B-23, Potential Equipment Failures Associated with Degraded Grid Voltage. As a result of this issue, YAEC installed on each emergency bus a degraded voltage relay which alarms in the control room. This is in addition to the loss-of-voltage relay which automatically starts the diesel generator for the affected bus on loss of bus voltage. In an SE dated January 4, 1985, (Reference 5), the NRC accepted this design, but requested Technical Specifications (TS) for monthly testing of both the first level (loss-of-voltage) and second level (degraded voltage). On November 30, 1985, (Reference 6), TS on the second level protection monthly testing were implemented. By letter dated June 30, 1986 as supplemented October 22, 1986 and December 4, 1986, YAEC provided its basis why monthly testing of the first level protection is not required and why inclusion of the existing refueling interval testing into the TS is not required.

i EVALUATION Attachment 1 is a one line diagram showing the emergency bus arrangement at l Yankee. This diagram was included in'a Licensee Event Report dated

! July 18, 1986 (Reference 7). The plant's Auxiliary Electrical System features three electrical divisions. The two outer' divisions are permanently connected to two 115 kV incoming lines, and the center division to the main generator.

The center bus is without power immediately following a generator trip. Power is restored by manually closing a tie breaker to either of the two outside 2400 volt buses, unless the trip is caused by a fault in the generator in which case the center bus is automatically tied to 2400 volt Bus 3.

l 8703020137 8702Q,.,,.,

DR ADOCn vDuu g e

The first level undervoltage protection on the emergency buses consists of the following: )

One loss-of-voltage inverse time relay on each 480-volt Class 1E bus.

These relays (induction disc type) are set to actuate in 1.8 seconds on a complete loss of power, or 3.0 seconds at 277 volts (58%), or 7.0 seconds at 370 volts (77%) with a tap setting of 105 volts and time dial of 2. This trip setting corresponds to 399 volts or 83.25% of nominal.

Actuation of this relay under accident condition will energize an auxiliary relay which initiates the following actions:

a. Actuates a lock-out relay which isolates the 480-volt Class 1E bus from the offsite source and starts the diesel generator.

b. Trips the high pressure safety injection (HPSI) pump, while maintaining the remaining loss-of-coolant accident (LOCA) loads

[ low pressure safety injection (LPSI) pump and the motor control center (MCC)] on the bus.

c. Once the diesel generator attains satisfactory voltage and frequency, its output breaker will close. This will deenergize the auxiliary relay which will remove the HPSI trip.

d. Ten seconds after the bus is reenergized via diesel generator, the HPSI pump is loaded onto the bus. The load shedding is retained since no other loads are sequenced on following the HPSI pump start.

The NRC staff's safety evaluation of January 4, 1985, found the first level undervoltage protection design to be acceptable. However, the evaluation noted that Technical Specifications which require monthly functional testing of the relays should be submitted.

, In their letter dated June 30, 1986, the licensee provided their justification l for the current 18 month testing interval for the first level (loss-of-voltage) undervoltage protection relays. They state that their justification js based on the demonstrated high reliability of the first level relay system, the unique and highly reliable Plant Auxiliary System arrangement, and the sim-l plicity of the existing design. Further, they state that the modifications l required to permit on-line testing of the first level undervoltage protection are complex, would complicate the existing circuitry and could reduce circuit reliability.

l l

i- ,

l The licensee states that the Westinghouse Type CV-7 undervoltage relays were '

installed in 1972 and have been tested every 18. months. The available main-tenance history since 1975 indicates that the tripping time in 13 out of 15 i relays tested have not deviated more that +0.06 seconds over 18 months of operation. In two relays, a deviation of +0.16 and +0.15 seconds, respectively, 4

'was found. Also, since 1972, there has not been a single case in which the i undervoltage relays have been challenged and failed to actuate. Further, the undervoltage relay on the 480 volt Emergency Bus 2 is automatically challenged on any reactor scram due to the generator trip.

The staff finds the above data does not by itself disclose whether the licensee's undervoltage protection relays are more or less reliable than average.

IEEE Standard 500-1984 (Reference 8) indicates protective relay failures per million i cycles of operation as follows:

Low Hich Fails to Open U T5 Tts Fails to Close 0.85 5.95 The licensee's relatively few cycles of operation (as compared to a million) does not-indicate that the relays are better or worse than average. However, the staff notes that the average probability of failure upon demand is very

. low. The following discussion centers on this observation.

The licensee states that two to six hours will be required to perform the monthly testing for all three CV-7 relays (Reference 3). This equates to

approximately one to two hours for each relay. During the testing, the relay i would be unavailable to initiate an action, i.e., equivalent to a failure to

! actuate upon demand.

(conservative directionAssuming each relay)is for this analysis , unavailable for one hour

failuretoactuateupondemandof1389X10,ghisequatestoaprobabilityof (1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> divided by 720 hours0.00833 days <br />0.2 hours <br />0.00119 weeks <br />2.7396e-4 months <br /> per month). However, the ave i

upondemandis5.95X10-gagefailurerateofprotectiverelaystoactuate (highside). Assuming then that the licensee's relays are equivalent to the average relay, monthly on-line testing of the relays would increase their probability'of failure to actuate upon demand by i more than two orders of magnitude. Thus, the staff agrees that on-line

! monthly ~ testing of the relays is not necessary or desirable. However, the staff will require that the Technical Specifications be amended to require l, testing on a refueling cycle basis at least once per 18 months. Presently,

the testing of the under-voltage (loss-of-voltage) relays is not included in l l the Technical Specifications, only in the operating procedures (OP-5762).

1 i

i l

i

, . _ _ . . _ _ . _ _ . . _ , . _ _ _ _ _ . . . . . , _ _ _ _ . _ . . . . . _ _ , . . . , , . _ - _ ~ , _ , , _ . . - , . , . _ . . , _ . . , _ , , . _ . . . .

, ~4-CONCLUSIONS Our review of the licensee's first level (loss-of-voltage) undervoltage pro-tection system, and other relevant features of the licensee's system, indicates that on-line monthly testing (as compared to testing on a refueling basis) of the first level undervoltage protection system is more likely to decrease over-all reliability than increase it. Therefore, we find that it is not necessary or even desirable to test the first level protection system on a monthly basis.

However, the staff will require the licensee to amend their Technical Speci-fications to provide for testing of the first level undervoltage protection system on a refueling basis at least once per 18 months consistent with the Westinghouse Standard Technical Specifications (Rev. 4). The licensee is required to submit such Technical Specifications for staff review and approval.

i i

d

REFERENCES:

1. Yankee Atomic Electric Company letter, G. Papanic, Jr. to E. McKenna, NRC, dated June 30, 1986.

2. Yankee Atomic Electric Company letter, G. Papanic, Jr. to E. McKenna, NRC dated October 22, 1986.

3. Yankee Atomic Electric Company letter, G. Papanic, Jr. to E. McKenna, NRC, dated December 4, 1986.

4. NRC letter, A. Schwencer to R. H. Groce, Yankee Atomic Electric Company, dated June 3, 1977.

5. NRC letter, J. Zwolinski to J. Kay, Yankee Atomic Electric Company letter, dated January 4,1985.

6. NRC letter, G. Lear to G. Papanic, Jr. Yankee Atomic Electric Company letter, dated November 30, 1985.

7. Licensee Event Report, LER 86-07, Yankee Nuclear Powre Station, July 18, 1986.

8. IEEE Standard 500-1984, Reliability Data for Nuclear Powr Generating Stations.

, - _ - - ~ , y . _ _ . , _ _ . . .c._ . . _ _ . . . _ _ . . , , , .- . , _ . _ , _ _ _ . , _ _ _ _ _ , _ . . ,. .. _ , ..- . _ - _ .

..' LER 86 97 y.

A i i AcHMENT .1 YANK'EE NUC. LEAR FDUJER STATIChi M AlN ONE LJNE Dil62RM ,

i:

Cf,267 Il[KV IISKY HAERINAN h

7/d VLA.A> MAlM d3Sil5 ry'm XFMR ZS/lE 3SST /S/8 2SST I VR I

/SST rym m l VR l 0 o o

/ 9 e 9 i 9 Z400V EDS 3 242/ BUSf 2400v EDS 2 -

AN T P oSST l

TP ASST W 555T og og o

,4yg 454a SAS

@V 9)LAS BUS to-3 Q 480V BJS4-l 9) Q AAS i 480V EUS 5 6 6 6 BTIA BTZA ST?A o o o o G O 9/ E'EUS I y/ E EDS 2 9/ EBJS .2

~b 6 6 6 6 6 6 c c

- o/ o/ o/ o/ O/ O/ o/ o/ o/ l i

~ ~+, ; pgg ,

1987 istribution Copies:

81 ' .'~;Ft M NRC PDR-Local PDR PAD #1 r/f.

-PAD #1'p/f

-TNovak, Actg. DD NThompson, DHFT OGC-Bethesda.

.EJordan BGrimes JPartlow Glear PShuttleworth EMcKenna AToalston FRosa ACRS (10)

LFMB