ML20067B039

Responds to NRC 901228 Ltr Re Deviation Noted in Insp Rept 50-336/90-22.Corrective actions:10CFR50.59 Evaluation Performed to Support Procedure Change W/No Resulting Unreviewed Safety Questions

ML20067B039
Person / Time
Site:	Millstone
Issue date:	02/01/1991
From:	Mroczka E NORTHEAST NUCLEAR ENERGY CO., NORTHEAST UTILITIES
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
A09246, A9246, NUDOCS 9102080103
Download: ML20067B039 (7)

Text

-

q 1

. NORTHEAST UTILITIES o noro Ome.. . somen street. Berun, connecucm 9 UNNbE$E[$2 e..*' " *"""" P.O BOX 270 H ARTFORD, CONNECTICUT 061410270 k U J 75 .UC3Ci[$1 (203) 665-5000 February 1, 1991 Docket No. 50-336 6922M -

Re: 10CFR2.201 U.S. Nuclear Regulatory Commission i

Attention: Document Control Desk Washington, DC 20555 Gentlemen:

Millstone Nuclear Power Station, Unit No. 2 Response to Notice of Deviation NRC Reaion 1 inmection No. 50-336/90 By letter dated December 28, 1990,II) the :NRC transmitted its Inspection Report No. 50-336/90-22 and associated Notice of Deviation. The deviation involves failure to test reactor protection channels and failure to operate the loose-parts monitor in accordance with Final Safety Analysis Report -(FSAR) commitments. The Staff requested that Northeast--Nuclear Energy Company (NNECO) respond to the Notice of Deviation within. 30 days of the date of the notice. In a telephone conversation with Region I personnel on. January-28, 1991, an extension was requested and subsequently granted. This extension was-requested in order to ensure a quality response. NNEC0 hereby submits its response to the Notice of Deviation as Attachment 1.-

NNECO trusts that the information provided herein ' fully addresses the NRC Staff's concerns regarding these issues. Please contact us if you have- any questions.

Very truly yours, NORTHEAST NUCl. EAR ENERGY-COMPANY I

E. J. 'Flp6cika f& f O'

Senior'Vice President Attachment cc: See'next page (1) C. W. Hehl letter to E. J. Mroczka, "NRC Region I Combined Ins)ection Nos. 50-245/90-20, 50-336/90-22, and 50-423/90-20," dated Decem]er 28, 1990.

i m2 w na t r70030 9.102080103 910201

$R ADOCK0500g6 p%)1

(,)

[l 1 m

1

. l 1

U.S. Nuclear Regulatory Commission A09246/Page 2 february 1, 1991 j cc: T. T. Martin, P.egion I Administrator G. S. Vissing, NRC Project Manager, Millstone Unit No. 2 W. J. Raymond, Senior Resident . Inspector, Millstone Unit Nos.1, 2, and 3  :

STATE OF CONNECTICUT)

) ss. Berlin COUNTY Of HARTFORD )

Then personally appeared before me, E. J. Mroczka, who being duly sworn, did state that he is Senior Vice President of Northeast Nuclear Energy Company, _ a Licensee herein, that he is authorized to execute and file the foregoing information in the name and on behalf of the Licensee herein, and that the statements contained in said information are true and correct to the best of his knowledge and belief, akke ))/Nr JJotaryPutytf flyComm!:dnMp!rps Pach31,1993 l

l l

1

. 1 1

1 I

Docket No. 50-336 A09246 l

l Attachment No. 1 Millstone Nuclear Power Station, Unit No. 2 NRC Region I Inspection No. 50-336/90-22 Response to Notice of Deviation l

l February 1991 l l

'I D

S U.S. Nuclear Regulatory Commission A09246/ Attachment 1/Page 1 February 1, 1991 ,

1 o

Millstone Nucl6ar Power Station, Unit No.-2 l NRC Region I' Inspection No. 50-336/90-22 ,

Resoonse-to-Notice of Deviation  !

l A. Staff Statement of-Deviation (first part).

" Final Safety Analysis Report (FSAR)-section7.2.2-states that! the I requirements of IEEE. Standard 1338 1971: will be met"for - the < testing of- ,

reactor protection system :(RPS) channels.. IEEE : Standard 338-1971 i requires -that tests be conducted by:- inserting -aMsimulated i signal '"as close to the sensor as practicable." Licensee procedure SP:240lG' was . i developed -to meet the above commitments, andiprovides the monthly func- '!

tional test of RPS channels. ,

" Contrary to the above, 'as 'of -November 15,1990, .the monthly functional tests -for the RPS channels for reactor coolant 4 system flow, reactor coolant pump speed and' the zero power mode bypass--. interlock were not-performed by inserting a simulated signal .as close to the primarycsensor -

as practicable."

-Response The Millstone Unit No. '2. FSAR. Section 7.2.2L . states 1that-: testing of 3'

reactor protection system (RPS) channels _" meets. the general requirements of 'IE:t ?'.8 1971. " We do not consider- this to be ,a commitment to ~ a.

literal interpretation of: the- IEEE guidance. The FSAR specifically addresses the.RPS testing-methodology _ currently-being used.

The Millstone Unit No. 2 Technical Specification Section 1.10 states:

l "A channel functional :est shall be the injection of a simulated signal into the channel as. close 'to the sensor- asi practicablec to verify Lthe -

OPERABILITY including the alarm'and/or trip functions."'

In thei case of the - RCS ' flowi and: RCP speed tests, Lthectest signal is '

injected at the input -to the- RPS trip ' module. In some <other functional-l test procedures, the test signal- .is introduced at the ; input to; .the E SPEC 200 cabinet in lieu -of - the transmitter input.: Iny these other

. procedures, 'the test signal- is injected at the location that most effec-l tively tests multiple functional units. .In<the case of the.RCP speed and l the RCS flow: instrument. loops, the RPS trip. units are the only functional un4ts in the loop that require functional testing.

NNEC0's . understanding, is that the functional test -is to determine that the functional aspects of a given channel- are -demonstrated to' operate in

, a'"go/no go": manner of assessment. Functional tests are not required to -

! assess - calibration accuracy. Channel checks . provide the assurance that j: the overall instrument channel is providing an operable signal from i

. _ _ _ _ _ _ _ . _ _ _ _ -m _* -- -- --wa

- . - - . . . - - . . - . . . .- - .- .. . - - -~

U.S. Nuclear Regulatory Commission

, A09246/ Attachment 1/Page 2 i February 1, 1991 sensor to point of display. Calibration activities provide.the necessary comparison of the system measurements. to'a . traceable standard to address the issues of instrument accuracy and instrument drift.

FSAR Section 7.2.4- addresses - the = testing methodology. The fourth _ para-graph identifies how' the channel check and functional test are performed ~

and that proper overlap exists. This section supports the existing basis for why the functional test method and ~ point of simulated. signal. injec-tion are appropriate.

The first paragraph identifies IEEE Standard 338-1971 as a document that provides auidance for testing protection systems. Section 2.1 identifies that testing scope may be accomplished by several methods. One includes:

introducing and varying, as appropriate, a substitute input to the sensor of the same nature of the measured variable. Another isby cross-checking between outputs of _ channels that bear a- known relationship to each other. In addition, Section 5.3.2(3) addresses testing.when access tc the sensor is not available, as is the case with these channels, and the use of simulated signals. Section 5.3.5 identifies:that if a-portion -

of a channel-is not included in artest, its operability may be verified by comparing readings between channels which-bear a known relationship to -

one another. = NNEC0 believes that through: the combined use of. these methods, the test overlap is established as stated in the FSAR and-is in- -

full compliance with the intent of-IEEE-338.

NNEC0 has reviewed the desirability of performing functional testing with the test signal being introduced into.the channel at the earliest posi- ,

tion of the control room loop. . Since, -in the present configuration of the bistables,- operability is' adequately verified, it _is NNEC0's. position that the intent of IEEE 338-1971 is satisfied.

The proposed modification to the test would require the--channel to be. out -

of service for additional time. .The- potentialyfor error in . performing the more complex test would be increased._-As additional components-would be' included, . increases in the acceptance criteria- tolerancei would be required. This would result in a less ~ precise assessment .of the func-tional trip . unit operability. In order for such;a- test-~ method to improve -

or equal the existing ~ testing process, calibration-type measurements and.

-comparisons -would have to- be performed during the functional test. ~ This ,

is clearly not required by either the functional test definition or the- '

IEEE standard.

- NNEC0 has reviewed the need to functionally test the additional! compo-nonts- in the instrument -loops and has concluded ' that no additional

' demonstration of operability would' be -achieved nor additional functionalL ~q attributes would be demonstrated. by inserting the signals at a point- 1 earlier in the loop.

l 1

We U.S. Nuclear Regulatory Commission f

- A09246/ Attachment 1/Page-3  !

February 1, 1991 I

-In summary, NNECO does not believe the currentimetho'd o'f' performing the- l functional RPS test deviates from statements in the - FSAR 'or MillstoneL 1 Unit No. 2 Technical _ Specifications. The point of: insertion -is consis-5 tent with: respect to the . FSAR, the referenced IEEE- guidance, and- the  !

Technical Specifications, i B. Staff Statement of Deviation'(second part)-

a "FSAR-Section 7.5.7.4 states that during normal 1 loose parts monitoring 1 system operation,. both loop magnetic: recorders are ..in the record mode--

making an audio- record-- of the output from -e'ach of the eight sensors ' on;

- i the primary coolant -system. Licensee procedure OP-23878 Revision: 2 was developed to meet the above commitment during normal system operations.

" Contrary -to the above,. as of November 8,1990, at 2:00 p.m.,: the, operat-ing procedure OP-23878 was not being implemented in thatfmagnetic-record--

ers were-- not in service- providi.ng an audio record of output from Lthe loose parts monitoring sensors."

]buon for Deviation NNEC0 -acknowledges - that 0P-23878 has not- been fully ~ implemented due to:

certain compcnents.of the system being considered out of service.

The Millstone Unit - No 2 loose-parts- monitor (LPM) system- has '.had 1a

. history of poor reliaoility. After the removal of the thermal shield in 1983, efforts were undertaken to improve its performance. OP-2387B' was. -

revised to include current guidance for use.by operations.- These efforts-proved unsuccessful. In 1988,.the status of the. system:was: reviewed and .

l the decision to pursue; total _ replacement of the system was -made'since

+

design of- the system was Joutdated _ and inadequate, :and :neither - of the-8-track style recorders function or'are' maintainable. Also, spare parts for the existing equipment are notLavailable and in-containment- repairs-

. to the -existing system ' are not; considered appropriater due to the high 4 radiation exposure required Land.the limited reliability 'of the existing

design.  ;

Cprrective Actions j A

OP-2387B has been changed to address this deviation. The changes to i

- OP-23878 included removal ~ of any reference to tape or recorder manipula-- j tion and .the requirement to listen to the audio output of all channels.

r Experience has shown that only specially trained individuals can discern -!

any' difference in the' audio output and very little meaningful information 1

can be gathered through this effort. j

A 10CFR50.59 evaluation was performed to support the procedure change, with no resulting unreviewed safety questions, k.

l l

U.S. Nuclear Regulatory Commission A09246/ Attachment 1/Page 4 February 1, 1991 Actions to Prevent Recurrence A record of the operable channel output will be taken on a monthly basis using existing I&C test equipment. The records will be reviewed and -

trended by NNEC0 until the LPM system vpgrade is completed at the end of the 1992 refuel outage, pate of Full Compliance The deviation has been resolved in the short term by the procedure change. A project assignment has been authorized to remove the old ,

system, from RCS sensor to control room electronics, and install- one of improved design, reliability, and maintainability. This project is currently planned for the 1992 refueling outage. ,