ML18087A910

Evaluation of Failure to Trip,Of Reactor Trip Circuit Breakers on 830222 & 25,Salem Nuclear Generating Station Unit 1, Interim Technical Evaluation Rept

ML18087A910
Person / Time
Site:	Salem
Issue date:	04/07/1983
From:	TOMAN G J FRANKLIN INSTITUTE
To:	SHEMANSKI P NRC
Shared Package
ML18087A907	List: ML18087A910
References
CON-NRC-03-81-130, CON-NRC-3-81-130 TAC-49795, NUDOCS 8305170106
Download: ML18087A910 (84)

Text

1. *. / APPENDIX B . . \ INTERIM TECHNICAL EVALUATION REPORT EVALUATION OF FAILURE TO TRfP *OF REACTOR TRIP CIRCUIT BREAKERS ON FEBRUARY 2i AND 25J 1983 SALEM NUCLEAR GENERATING STATION UNIT 1 NRCCOCJCETNO.

50-272 NRCTACNO.

49795 NRC CONTRACT NO. NRC-03.a1-130 .Prepared t:Jy Franklin Research Canter 20th and Raes Streets Pl'llladatphia.

PA 19103 Prepared for Nuclear Regulatory Commission Washington.

O.C. 20555 FRC PROJECT C508 FRC ASSIGNMENT "O" Author: G. J. Toman FnC Group Leader: G. J. Toman Lsad NRC e..,gineer:

l?. Shemansk:f.

April 7, 1983 This report was prepared as an account of wonc sponsored by an agency ot the United States Government Neitner the United States Government nor any agency thereof. or any of their employees.

makes any warranty, expressed or Implied. or assumes any legal liability or responsibility for any third party's use, ortne results of sucn use, ofany lntcrmatlon, ratus, product or precess dlsclcsed In this report, or represent3 that las use by such third party would not Infringe privately owned rtgnts. Prepared ,,/ Reviewed by: * * .

Apprcvee by: _ ,r-8:305170106 830429 I PDR ADOCK 05000272 I s PDR Project: Manager Oepartment Director ( A e.

1) Date: i=/7 I Cate: I I I Research *center A Division of The Franklin Institute The BenjMWI Fr.nlcift PhilL. PL 191 OJ (21,) "'8* I 000. ..

• --*-*-**---*-*--***

... -... *"*: . .

--... :

...

.. ...........................

• * .. FORmORD This Tedlnical Evaluation ,Report was prepared Franklin Research Center under a contract with the cr.s. NUclear Eegulatory commission (Office of Nuclear Reactor Regulation, Division of Operating Reactors) for technical assistance in support of NIC operating reactor licer1sing actions. 'l'he technical evaluation was conducted in ac:::crdance with criteria established by the NRC.

ReseMcll Center iii INTBODtX:TION This Interim Technical Evaluation Report contains a compendium of the reports issued by Franklin Research Center (FRC). to the Regulatory . . Commission (NRC) and updates to April 6, l.983, FRC's findinqs concerning the ' failure of the Salem Unit l Westinghouse OB-50 reactor trip circuit breakers on February 22 and 25, l.983. 'l1le evaluation centered upon the. undervoltage trip attachments of the circuit breakers which were supposed to trip the circuit breakers when deenergized.

This is a report on work in pro9ress1 thus, the conclusions contained herein may change as new information from the proposed Licensee verification testing becomes available and subsequent evaluation is completed.

Table l lists the reports and letters transmitted to Mr. v. Noonan, C!lief, NRC E:ruipment Qualification Branch, prior to April 6, 1983. O!RONCLOGY OF E"lAI.UAT!ON

/ G. Toma."'l of FRC accompanied Messrs. V. Noonan and ::>. She.'Uanski of NEC Equipment Qualification Branc.'l and Mr. J. Beard of NRC Operating Reactors Assessment Branch to Salem Nuclear Generating Station (Salem NGS) on 3, .!.983. The purpose of the trip was to gather information concerning the 5alem anit l events of February 22 and 25, 1983. The information obtained is contained in Appendix A. Appendix A also contains a discussion of initial findings concerning the undervoltage trip attachment (UVTA) that did not operate Failure of the CVTA to operate prevented the reactor* trip circuit breakers (RrCB) from opening automatically.

Figure l contains photographs of the *as-C'eceived*

state of the onit 2 *B* IJVTA supplied to FRC by ?ubJ.ic service Electric and Gas Company (PSE&G). on March 3, 1983. on page 3 of Appendix A, the second paragraph states that Chemical Rubber Company CRC-2-26 lubricant was used on the OVTAs associated with Unit l. on March 18, 1983, during a meeting with the NPC, Westinghouse personnel stated that the Westinghouse technician's report states that only one RrCB was worked _upon at the Salem plant on January l3, 1983 and that the lubricant used was Calfonex 78A. FRC has been unable to find a supplier for this lubricant.

Center A°"""""' al The F..-., i1umaite ..... -l-'* *.

'.*. ... 1 . *: .* .* ; Appendix B contains the review of Revision l of Sa1em NCS Maintenance Procedure M3Q-2. Since tb.e reyiew, Bevisicn 2 has bl!i!p issued. Revision 2 requires ten successful t:ippings of the RrC:S by a new OVTA prier to returning the circuit breaker to use when the OVTA is replaced.

The procedure is titled *Reactor Trip and Bypass ACB Semi-Annual Inspection and Testing*;

the title now states the periodicity of testi119.

Cautions concernill9 handling of the lJVTA and further tests of the RrCB and tJVTA were added in Bevision 2. Appendix C describes the *as-received*

electrical testing of the onit 2 *s* OVTA. * 'l"he device was found to operate at 26% of normal voltage ratb.er than between 30 and 60% of normal voltage as stated in the manufacturer's literature.

'l"he evaluation of the device during these initial-tests indicated tb.at the device should be compared to the other tlVTAs available at t.."le Salem plant_to determine visible variations f:om device to device. On March 10, 1983, G. Teman and R. Cranisky o_f went to Salem NGS for the purpose of comparing the oni t 2 *:s* OVTA to the remaining Oni t l OVTAs. Upon arr::.val at the Salem plant, it was found that onl.y one C'rlA *11as available.

Comparison showed variations in manufacture and reset lever adj:Jstment spring setting between the two de*1ices.

Messrs. Toma."1 and Cranisky also observed operation of a t:ni t 2 RrCB with the 2 *:s* !'NTA. When t:le OV'l'A was attached to the circuit breaker, power was not applied to the CVTA. T!lerefore, when the RrCB was closed, the CVTA reset iever was held back manually to prevent a trip-free opening of t!le circuit breaker. The UVTA :eset lever was released.

'l'he' UVTA latch released and the trip lever :noved up to tbe trip bar, but the Rl'CB did net trip. A second attempt also resulted in.no trip. 'l'WO mere attempts at manual operation cf the tNTA did t:ip the Rl'CB as did.subsequent electrical operations.

Appendix D contains the letters of transmittal for :ni.c:oscopic graphs cf the latch, latch pin, and latch spring from the Onit 2 *B" CVTA shcwingrougbness and wear on the mating surfaces.

Appendix E contains the FlC findings from the beqinn_ing of the evaluation up to 21, 1983. It discusses.two possible failure modes for the OVTA and findinqs from a March 17, 1983 trip to Salem. Franklin Research Center .. :**: *

\ ** . Appendix F is *a trip report from a Marcil Jl, l.983 visit to Salem to eval.uate testinq.perfcrmed by on the t:Init l Rt'C:Bs .. E"INDINGS The following-is a summary of the significant findings of the evaluation to date: l. Manufacturing variations existed on the original.

OVTAs supplied to . the Sa1em plant. 2. Manufacturing variations exist on the four new tJVTAs supplied to Sal.em Otii t l. 3. No lubrication of the OY'I'As appeared to nave been performed from the time of manufacture (appcaxima.tely 1972) until January l.983 when a lubricant otb.er than that reeommended by the manufacturer was used. 4. The Chit* 2 *13* TJV'l'A had its reset lever arm spri119 adjustme."lt dlanged some-time after instal.laticn, possibly to remedy inadvertent openings of the circuit breaker during operation or while attempting to close the circuit breaker prior to operation.

5. Improper adjustment of the OV'l'A reset lever arm can increase the probability of an inadvertent opening of the circuit breaker (trip free) if the arm is. set to overtravel too far. 6. Loosening of the reset lever arm spring reduces the tendency to trip .

but also reduces the force that woul.d *cause the OV'l'A to unl.atcb when deenergized.

1. variations in reactor trip bar forces exist among the Salem Unit l

8. 'ttle mating surfaces of the Unit 2 *B* latch components are rough and appear not to have been honed. Adc:!iti*onal.

wear from operation appears to be significant.

9. Hesitation in lever arm motion when the OV'l'A is slowly deenerqized was noted in the unit 2 *a* UVTA and the device. operated bel.ow the expected 30% of nominal. voltage point. lO *. on Marcil 18, Westinghouse switchgear Division personnel al.so indicated that the CVTA must be replaced some time during life of the plant. Criteria fer determining when to replace the OVTA do not appear to be available. , * -enklln Research Center A Clwiliail alThe Frmnlcln ..___ I' .-......... .

CON:I.USIONS

.. It is believed that in as-manufactured

• new* condition, the l983 UV'l'A will. properly trip a circuit breaker that has a trip bar force requirement within the limit of 31 ounces, and would probably consistently trip a circult breaker with an-as-found trip bar force r.equirement of up to 38 ounces (the force required to trip the unit l "A* RrCB). sowever,*sufficient

'* evidence has not been presented to show that current manufacturing processes for the UVTA, when coupled with maintenance, will eliminate long-term failures that appear to be mechanical, age-related phenomena.

The variations from device to device cause concern. 'ttle fact that honing of latch surfaces is a hand operation indicates.that variations in the surfaces of the latch will remain even though no extreme roughness should be expected.

In addition, the lack of quantitative acceptance criteria for the OV'l'A adds concern that impending failures might be missed during inspection and maintenance.

RECOMMENDATIONS

l. Acceptance criteria be set for parameters affecting correct operation of the O'VTA. 2. Testing :nethodology for acceptance*tests be prepared for factory and Licensee use. 3. uniformity of construction be instituted or suf;icient testing be performed showing that the variations in the devices are of no consequence to reliable operation.

4. Life testing of the UV'l'A be performed to show that the device can successfully operate for the intended lifetime with proper maintenance.

s*. Criteria be developed to determine a replacement interval for the UV'rA such that replacement occurs significantly before end-of-life failure.

Research Center . A OMlian cl"The Franie!in

._,.. *

• I. s.* '!he base.line testing of OVTA output forces and RrCB trip bar trip forces is a f.irst step in proving the adequacy of the OVTAs for continued use. 'lhese test resu.lts sbou.ld be compared with results taken periodically during the life of the tJVTAs. In order to allow preventative action to be ta.ken, trending of the variati'ons in the

• results should be performed to determine if degradation in OVTA and performance is occurring.

Center A Clwiliml *a1 The Fl'Wlldift

"-a /

. Tabl.e l..

• Chronologic:a.l Li.st of Beports and Correspondence from Fg: to the NR: concerning Failures of Reactor Trip Circuit Breakers at Saiem unit l on February 23 and 25, 1983 APoendix Date/Subject A March 4, 1983 Initial Report on the salem Onit 2 Trip Onit B March 9, 1983 Reviews of Proposed Corrective Action by PSE&G and of Maintenance Procedure M3Q-2 c March 9, 1983 Report of Initial Testing of the Salem Onit 2 *B* Ondervoltage Trip Attachment D Marcil 21, 1983 Photographs of the Salem Unit 2 "B" Undervoltage Trip Attachment (Transmittal letter and description of views only) E March 22, 1983 Salem Events of February 22 and* 25, 19831 Failures of

• P.eactor Trip Circuit Breakers1 Final Report of Initial Investigation March 31, 1983 Trip Report: (Transmitted this Interim . JUUu-Fra°nkli:1 Resezsrc:h Center " CMsion c1 The rfWllclin IM1mu Salem Unit l Reactor Circuit Breakers to the NBC for the first time **as part of Report) * .. *, .;

• A. c. * *:., ondervolta9e

':rip Attadmlent

?cp View tlndervoltage Trip Attachment.

Side View B. c. .. -* .. :_ ... *:_ ... .-:. -* .**.*. -"-* '* Oftdervoltage Trip Attadm:ent .en CU:cuit Brealce: .I ondervoltac;e Trip Attacbmen.t Side View Pigu:re l. onit 2 *B* tl'ndervoltage Trip Attachment (A, c, and o a:e pbotcgrapis of the as-received

_aonditicn*.) ______ .. _

--**-------':""'":'"'

..... ---

• ... **.:* .. *: . .: ... *: *** ... *-,_ ..... ** .... . . ****-.. _ *_*._ :_.,_. * . ?'.' ..........

..... .

• • INITIAL REPORT ON THE SALEM UNIT 2 UNDERVOLTAGE TRIP UNIT APPENDIX A ' '. . ... -. . ; *.*.::

Center A Division of The Franklin Institute The Ben,.min Franklin Pa. 19103 448-1000 * * .........

M. *: -*-** *** *" ....*. '."'"'.;' . ***, **.:: ... -;* **. * .. '* . .... . -.--* ...

.... -* . '

• Research Center

• A Division "of The Fnnkiln Ir.stttute March 4, 1983* Vincent Noonan Chief, Equipment Qualification Branch U.S. Nuclear Regulatory Commission Pnillips Building, MS-P-234/RM P-234D 7920 Norfolk Avenue 3echesda, MD 20014

Dear Mr. Noonan:

Attached is the initial an the Salem Unit 2 under-voltage trip unit. ! have detailed same possible causes and i:idicatad the initial tests and evaluations to be per!or:ed.

Also attached is a request for additional equipment from PSE&G to be used ill the investigation.

Should the;re be any questions, please call me a't 215/4.i.8-1257.

I intend to begin the tests on Monday, March 7, 1983. GJT/ih Encl

• 10th&: Rao:e Streets.

Pa. 19103 (215) 448-1000 TWX-710 670 1889 I ** ..... -.. * . *. ,** .. * ... '**; \ * .. . . .-:* .. ***-",,'" . . ... -

st!MMAll AND I'N'ITIAL-FINDINGS Initial inspection of the cnit .2, B Wlder-voltage tl:ip unit indicates a possillility of multiple cont:ibuting causes of failure. Possible contributors are tl) lack of lW::>rication:

(2) wear; (3) 1%1Qre operation than intended by desisn:. (4) from improper lubrication in January, 1983: and (5) nicking of latdl surfaces caused by vibration f:om repeated operation of the breaker.

ibutors appear "tc be cumulative with no one main . . cause. 'n:le initial investiqation does indicate that the failure is aqe related and that a new device perfoca properly.

Many surfaces of the " latdl :nec:hanism ue worn and the additional.

friction tends to prevent proper . operation.

Proper llJ.bdcaticn througilcat the life of* the device may have . prevented the weu that c:an be seen on the sample * . -* . tests and examinations E'IIC proposes to perform will. attempt to the cawae of failure and if possible reproduce it. '!:le followinq summarizes the initial findinqs and lists the pro?lsed tests. OISCCSSIOR OP'

l. !nitial thcuc:hts fro1u information received crier to Marcll 3, 1983 visit to Salem aeneratina station. ?rem infcr:uation received by J;ilcne and by telec=pier, 'fil'f initial bel.ief prior to visitin; site was that the under-'lcltage t:ip device did cot operate because it had "not been exercised very frequently, my be.lief being that dust or dir*.: acCUmul.ation on the device in two months time prevented its operation.

2. Initial. eval.uaticn of information received at Salem site on March 3. When I arrived at the Salem site, I met vit:h Vince Neenan, Paal. Shemanski, and J. T. Beard, al.l of the HBC. we were given a walk-around inspecticn of the plant by an oper.ator named Ray McCarthy.

We inspected the onit 2 Reac1:cr T:ip breaker area and visaal.ly examined the circuit breakers which were laying on the fleer near their cabi_cles.

'l'he unde.c-Yoltage trip devices were remcved frcm eac::i of these circuit breakers.

'l'!le shunt t:ip coils were in place. 'l.'he cucuit breaker trip bar moved freely. We then went to the cnit l Reactor Trip circuit breaker area. 'Ihese breakers had been removed and brougilt to the electrical maintenance area. It was noted* that an inverter wnidl was approximately 4 ft f:cm the circuit-.

Franklin.

Rese!lrch Center A°""-al"ni.F..---... .*: . . . ;.*.:--*----

• . .. *. -* :. !.. *, .:*:_ I breaker ccmpart:nent was blowing hot ai: into the ream, and into the circuit .. ( :-": l:lreaker compartments, and that the 9eneral area was well above 80 *. We then were taken to the elect:ical maintenance area, where we interviewed Mr. Keteum, an electrical maintenance supervisor, who discussed the circumstances of the removal of the circuit breakers that were involved with the incident on cnit l. We were also introduced.

to Hr. Leo :Roland, ancthe= electrical supervisor, who bad aJ.so worked on the circuit breakers*

in .question in AUgust of l.982. 'l:he information received was that the circuit breakers and their under-.,oltage trip device had been operated frequently and had operated within a few days prior to the incident.

'!'bis dispelled my i."l.itial concern that the .devices llad net been exercised often eneugh. I requested that Salem management provide one of the under-voltage t:ip devices, and a shunt trip coil, for testing at Franklin Researd:l Center. did so. It appears that new under-voltage trip coils are not readily available.*

'l!'!is*wiJ.l be investigated at a later date. P-ESUL'l'S Ol!' INITIAL EX>>!INA'!'!ON; SOME l?OSSI:3LE Since receipt of the under-voltage device, whidl is the cnit 2 a trip under-'loltage device, I have noted rougi:ness in the operation the trip late. '!'here is some drac;gi."l.g of the mechanism, and portions of t."le latch mechanism have obvious signs of wear. POssibla c:ont:ibuting factors to the failure to operate are a lack of lubrication, wear, jarring of the under-voltage device f:cm the <;ircuit breaker operation and more frequent operation of the under-voltage trip device than was intended during design. My belief is that under most' industrial applications, under-'loltage device would be used very infrequentl.y, and probably would onl.y be operated durinc; . . test sequences at yearly or longer intervals.

'!herefoz:e in industrial applications, the device would operate onl.y a few times, perhaps 20 or 30 during its lifetime, and would not be a normal tripping mechanism for the breaker. However, in its use at Salem, it is the prime trippinc; device fer the circuit breaker, and as such is called upon to operate on the order of 50 per year. 'ttlat would mean, at its current age, in l983 th!!re ....ould be possibly 400 to 500 trip operations on this device. curing t."le investigation, it was noted that the shunt trip coil has been operated since' August, l982 once every 7 days rather than at longer intervals.

.'ttlis means that t."le circuit breaker is now tripped and closed . .

F:'1.lnk!in Research Center Ao;.._, GI,,_ F.---***.: **"*l.-:**:

.* * .:

...

'. !. every 7 days, causing_ :he jarring of .the entire lDl!ehanism of the circuit " breaker and its attached relays and coils dr.ie to the 1lOntal. oceraticn of the . -breaker. This may er .may not be significant in that t:ie re.lay would stay energized during these trippings and its latch mechanism would be jarred somewhat by operation of the breaker. coul.d possibly add to the friction which is building up in tne latch medlani.sm from nor:nal .operation by caL1sing the latc:::i medlanism to just slightly nick the surface that it rides on thereby tending to prevent operation.

PUrther investigation will t.ry to determi.:ie whether this indeed a ccnt.ributcr tc t:le mis-operation.

It woul.d appear from initial inspection of the device that wear and roughness of matin9 surfaces in the t::ip latdl are ccnuibuting causes. Proper J.u.bric:aticn may have prevented the current situation o:r c:cul.d reduce this roughness to the point where proper opera t.ion c:oul.d oc::w:. F:c will at:empt tc de-cl!l:mine whet.her the .OC-2-26 lu.bricating, cleaner spray added to t:le problem by ei tiler causing corrosion or removing al..l :'S!Sidual lubrication

£rem initial. c:cnstruction.

It appears cat from t'le time of initial const:uction of the under-'Toltage trip units, up until J'anuaey of lS83, no lubrication whatsoever had been performedr and then in January of lS83, lubrication was performed by the maintenance personnel in conjunction wit."l a Westinghouse tedlnician.

i\t this tUE!,

lubricant cleaner was sprayed on a..ll four trip devices associated with onit l.. 'lhi.s lW:lricant is procured by FRC for testing purposes.

LISTING OF :::XPECTED INVESTIGATION

l. The firs*i: test will be to perform various de-energizations and energizations of the under-voltage trip unit and mcnitcr the device under various conditions.

2. 'nle second test will be to disassemble the latdl mechanism to observe the surfaces of the various parts of the latch, and i=botcgrapb these surfaces t.'lrough a m.ic:oscope to determine various leveJ.s of wear in these surfaces.

3. 'l11e tbi.=d test is to determine the effects of Cl!C-2-26 spray on the various types of metals used_ in this device. An attempt will be made to use metals ot.'ler than these in t."le actual device. If possible, we will deter:nine the c:::iemical consistency of this spray, hopefully through the manufacturer.

4. 'nle fourth test is only possible if the spare circuit breaker can be oi;)t.ained from Salem. '!he under-voltage trip device and. the shunt ccil. . -". .. ' *.-;':**

' *-.. )* **-;_ . .. *. **.,_*:. ,-* * ...

• would be upon tbe breaker, and the breaJcer woW.d be _cpe:a'ted repeatedly to determine effect

• imder-volta9e t:ip unit. It is surmised that while the devic:e is ener;i=ed and the breaker t:ips and closes a number of times, aaditional fricticri of the t:ip latdi '112.Y cc:cur from the vibration. test vol:ld primarily deter.tine whet!:ier such .additional friction occurs from operation.

'l'O prove t!lat the sample under-'1'oltage trip Wlit is identical tc the Onit l devices, a visual inspection of all existing Salem Onit l and 2 wider-9oltage t:ip wiits should be perfcr:Ded

• . can take place at Salem. Ne disassembly is needed. The devices can be mcunted on the 0 s or lcose. sbould be dcne as* soon as possible.

'!.'Uesday March a, l.9 83 is rect1mmended.

?urther tests will be deteained upon the basis of these initial tests. All. tests will be ncn-dest:uctive tests sudl that the devic**wil.l.

be* aeie tQ be 1.2sed fer further testinq and possible return. to the 11tility.

AD'CI'l'IONAL f!EQUESTS FOR SALm STATION !9.N1\GF.MENT

l. For use in investigaticc.

of d:le incident, Fli: requests that t!le spare DB-50 circuit breaJcer be made available fer ase at !'BC for a period of 2 weeks. Delivery is requested as seen as practical.

SOpefuJ.ly no later t!lan March 9, l9 83. 2. n:lC woulc like to v.isually inspect all of the remaininq widervoltaqe t:ip devices frcm sa.lem l and 2r includinq these at t:le NB: Headquarters.

It would be beneficial to have t!lese ali in one place to speed the review. '.there would be no ciisassembly and the devices o;,ulci be on the circuit breakers fer this. inspection.

It is hoped. that this could be done on * 'l'llesday Marcil 8, 1983.

F'ranklin Reearc:h Center A ei...-, al The F,_ _,.. . ;-*' .. --. >. -._-*.:_. . ; ... * ... .-.-*-.*-..

'".:': .. ... -.. ,

..

• -:_ .. -***::.-: ....... .::. ... I I

-* ll REVIEWS OF PROPOSED CORRECTIVE ACTIONS BY PSE&G ... AND OF MAINTENANCE PROCEDURE M3Q-2 APPENDIX B --. ** **-..... --* -... _., *.: ..... . *.-*.: ' "ftnklin Center A Division of The Frankiin Institute The Senjarrun Franldin Parnay, Ptlit... P11. 19103 (215) 448-1000 . -. .. :*-*

-*-: . -.... . . . . --.. : *. . . . .. . .****: .

Research Center A Division 9i The Frankli."'l Instmtte Mr. Vincent Noonan Chief, Equipment QUalification Branch Il.S. Nuclear Regulatory Commission Phillips Building, MS-P-234/R..'!

P-234D 7920 Nerf elk Avenue Bethesda, 20014

Dear 2'4..r. Noouan:

9, 1983 Attached are the reviews of the proposed corrective action by Public.Service

& Gas Co. of Ne..-Jersey as detailed in the letter from Mr. Richard A. Underlitz to Mr. D. G. Eisenhut dated 1, 1983, and of Maintenance Procedure M3Q-2. These documents

..-ere previously mitted to you on March 9, 1983, at approxil:lately 11: 00 am, by telecopy.

GJT/:!.b F.ncl.' cc: M. Carrington P. Shemanski Very truly yours,

• I

• Gary µ. ;oman Sectib Leader, Qualification Analysis and Planning Section, Nuclear Engineering Dept. 20th & Race Streets. Philadelphia.

Pa. 19103 (21SJ TWX-ilO 670 1889 . * .. ,_*, . ,,-.: .. . .. : . . **.* .. * .... -..

. ', RmI:::W -OF LE'!TER FROM MR. RICB'.ARD A. O?mERLI'rZ, VICE PRESIDENT NOCLEAR, PSE&G CO. TO MR. D. G. EISENHUT, DIRECTOR, DIVISION OF' LICENSING, NRC, DATED l, l983.

Subject:

Reactor Trip Breaker .Failure, NO. l tlnit, salem Generating Station, Docket No. 50-272. 'I'he following are resul.ts "of the review of the referenced letter. This review is restricted to concerning the "shunt trip* and voltage trip* attadlments of the Reactor Trip Breakers.

Page l of t:h_e letter states that the sole reason for the failure_ of tbe undervoltage trip attadlment is laclc cf proper lubrication.

FBC continues.

ta evaluate this possibility and makes no finaJ. determination at tnis time. nc is also evaJ.uating other possible related conditions such as, .improper cation o:f the* device, dust and debris in the moving coiJ. section of the device, b incling of the late pi.n, and residuaJ.

magnetism

'in the ccre. ::'P.C' s findings will. be forwarded ta the NEC on or. before April l, l98J.

• has no comments on Page 2 of the letter. . I On Page 3 , in Item 2, a new maintenance procedure MJQ-2 is described.

However, no periodicity for maintenance and no other method of invoking tnis p:ocedure are stated. (Nc1:e: tne M3Q-2 procedure al.so does not contain this information.)

Since laclc of mairJtenanc:e is indicated by PSE&G co_. and Westinghouse as the probable cause, the* periodicity for maintenance must be addressed.

On i'age 4, Item 7 describes new testing cf the Reactor Trip Breaker. One test requires timing of the breaker operation when tripped by the undervoltage trip attadlment.

PRC recommends that this test be performed 3 times and that t:!le average time be compared to previous tests (as test data accumulates)

• '!:his wouJ.d provide a means of determining variations in .the trip attachment and circuit breaker response.

Inc:easin9 time lag woul.d be indicative of an impending problem. On Paqe 4, Item 10, FEC agrees that mcving tne undervolt:age trip ments from one circuit breaker to another is inadvisable and aqrees with the Licensee's proposed action. On the Licensee does not define whether the shunt trip is safety related or not. Since tnis device proved to be hiqhl.y important when tbe undervoltage trip attachment did not work, it wouJ.d very important ta -consider the shunt trip as safety-related.

In addition, the shunt trip is in series with the circuit breaker auxiliary switch tbereby requiring the liary switch to be safety related. Prepared by G. J. Toman 3/8/83.

Researdt Center :Aa..m.a-n.

F.....,_ *:. **.:-... -... :*-."*. '. *. . _ .... ,._. :::."' :::-*:-:*

• . .. .., . , * .. .-.. ..... ........ . . '

' ! *.._* ... REVIEW' Ol!' PSE&G CO. PRcx::EJJU'R!:

M3Q-2, ..

• TITLE:

TRIP ANO B'l'P11SS ACS INSPECTION ANO REV. l. 3/4/83. 'ttle copy of this procedure received by FBC contained an additional one page document laeelleq .*salem NUclear Generating Station, Reactor Switchgear, Operational Verification Program.*

':.nc has the followiiiq comment on the Operational Verification Program. Item 2 states that requires 10, 40, then 50 trips of the circuit breaker depending upon the number O*f failures Of the unde.rvoltaqe trip attachment

• . ?iQ3-2 does not contain sudi a requirement.

Allowinq any failures durinq testing is wholly inappropriate for the unde.rvoltaqe t:ip unit and maintenance procedure M;23-2 should not be modified*

to allow the undervo.ltage trip attadiment to fail, no matter how many successful operations follow. Failure to operate once during a sequence of trippings of the attachment indicates severe prob.lems in the mechanism and places the reliability of its function in doubt. Item 2 of the document al.so states that the lO, 40, then SO trip test system is statistically sound for the application.

'!his is correct for the application.

AS the test method would allow 2 failures in 100 tests. '!his would be a 0 .02 probability of failure. The.re are two reactor trip circuit breakers in series. ASswning independence of failures (i.e., no common mode mechanism) the probaeilit-/

of s:i:nu.ltaneous fa:l.lure of the t'WO. circuit breakers would be a .02 x a .02 or o .0004. i:ttiis dicates that a failure of both reactor protection systems would occur once every 2500 trips. For Sa.lem tlr1it l, whic:n is said to have so. trips per year, a fa.llu.re to. trip would occur at least once per SO years. 'ttle allowance of any failures to t:ip dw:inq testing is totally unac::eptable.

of M3Q=2 l. NO statement of the period between maintenances is made. NO method of voking this procedure is given. 2. Page .lO of 23, Item 9.7.2.2 indicates that lubrication could de some good, but does not indicate bow to determine when to lubricate the undervoltaqe trip attacment.

Fm: has found that the onit 2. t:ip attachment trip lever hesitates when depressed.

It is suqgested that maintenance perscnne.l slcw.ly depress and release the t:ip lever to determine if it binds du.ring . operation.

If it does hesitate or bind, and lubrication does not remove this bindinq, replacement sbou.ld be performed.

Also, the p.rocedw:e sbou.ld indicate.

a required interval between lubrication.

'Ibe Manufacturer's

'rechnical Balletin, NSO *Data Letter 74-2, suggests starting with a 6 month interval between

':ttie second parag.ra¢ of Item 9.7.2.2 indicates the portions of tbe voltage trip attacment to be lubricated; however, no mention is made of the latch to latch spring (the copper alloy flat sprinq) surface, the bearinq points of the latch spring pin, and the bea.rinq points of the reset lever a.rm.* All of these, especially the latch to latch spring face, are friction sources that c:ou.ld prevent correct operation and should be considered for lu.b.rieation.

-.l_ .... -.--:*-. -.* .----.. \

3. Section 9 .8 d_isc:usses timing of. the* circuit breaker wl:\en t:ipped by the undervcltage trip att.aennent. suqgests that th.ree timing tests be performed and tne average ti:e be c:cmpared tc previous tests as successive tests are perfoced.

'!!lis woulci allow degradation in performance tc be determined.

Note: If the trip times are known to be consistently near tile trip limit \ of 4 c:yc:les, trending of tile trip times would not be usefu.l. However; if a new breaker starts witil lower trip times such as 2 cyc.les, trending could be indicative of cieg:radation.

4. Enclosure l of M3Q-2 was taken from the Westinghouse Low Vo.ltage Meta.l Enc.losed switchgear Manual. 'this diagram shows attachments suc::b as tile overcurrent trip device tilat are not used in tile reactor trip cirC'.iit breakers and does not show tile sb.unt trip or Wldervoltaqe_

trip attachments.

FBC suggests tilat an diagram be included in the procedure.

5. In accordance with 3 above, PEC suggests that Enclosure 7 be modified for tilree timings of t!:ie circ:iit breake: racer t:ian one. Prepared by G. J. 'l'Oman 3/8/83

• Research A Diw-cl The FTWllllln

.,_... -2_ ..

.. ****:-:.*'":.*-*

• _.-*-. .:-_. **

.;**

• ,-; ,*;: ----REPORT ON INITIAL TESTING OF THE SALEM UNIT 2 "B" UNDE;RVOLTAGE TRIP ATIACHMENT APPENDIX C . *. '.: -*:.

Center A Division of The Franklin Institute

• ""1e Benjamin Franldin Parkway. Phil&. Pa. 19103 (21!5) 446*1000 -*--,_ **;**.-.*

...... ,,_ i I .1

Research Center A Division of Tne F:anklin Inst:tt;tte Mr. Vincent Noonan ... March 9, 1983 Chief, Equipment Qualification Branch U.S. Nuclear Regulatory Commission Phil.lips Building, MS-P-234/RM P-234D 7920 Norfolk Avenue Bethesda, MD 20014

Dear M:

r. Noonan: The document describes initial tests cf the* Salem 2 B U'nde-"'"Voltage Trip attac!lment.

Please note the finding on Page 2, !te.m 3. !he as-found condition of the device iS very disturbing.

When energized and de-energi:ed the device does not operate within the specified voltage range and hesitation vas noted in the latch to lat:i:h pin operation.

GJT/ih -eo-,.

cc: M. Carrington P. Shemanski

,ours, // _,,,,. Toman Section Leader, Qualification Analysis and Planning Section, Nuclear Engineering Dept. 20th & Race Streets. Philadelphia.

Pa. 19103 (.215) 448-1000 T\V'X 0 ilO 670 1889 *. . '

.. ..

--*Y .* .. *.*;: REPOµ OF INITIAI..

OF Im:

2 "B II ut"DD.VtlI..TAGE TRI?

3/9/83 1. Described .a. The UnderVoltage (UV) Trip Attachment operates at 60 to 30% of nor::lal voltage per the instruction manual. b. Nor.:al vo_ltage on this device is 48 Vdc. Note: The manual is describing an t.mdervbltage trip attachment an ac coil designed to operate on line voltage. The device supplied for Salem has been modified by replacing the ac coil a 48 Vdc coil. 2. As-Found Tri:ming Point ThLSalem 2 "B" lJV trip attachment consistently tripped at 12.6::0.4 Volts* de *'-hen voltage was gradually reduced from 48 Vdc. This is 26% of nor.:ml. operai:ing voltage. It was noted that the reset ar.11 moved pari:ially wardS unlatching a few "1."0lts before unlatching occurred indicating binding of the la:ch to latch pin. J. The for spring tension for the reset lever spring was found on the Salem 2 "E" UV Trip attachment to be approximately 1'1 tur.is below the surface of the locki:lg nut. The adjustment "las brought to the condition on Fig. 17 .oi the 'Westinghouse Ma.'lua.l., "!nstructions for Types DB-50, DBF-16 and DBL-50 Air Cj.rcuit Breakers," which is approxi:lately 6 tun:is beyond the surface of the locking nut:. (See attached drawing.)

The 1JV trip attacbmec.t then operated at 25.8 Vdc, ....ill.ch is 54: of nor.nal voltage.

tension w--as reduced to the lowest limit: (1 co turns below the as-f ound conditicn) the device operated a 6.3 Vdc or 13% of voltage. This indicates that the setting of the reset ar.:n spring tension is to corTect ooeration and that the as-found oosition of the adjustment is definitely below the desired level.

Fnsnklin Research Center . AC:.-Cln..I"_,.,.__

Page 1 of 2. *.*. -;.*.-* -.-.

• ' .... The =Qvmg core is nor=ally held the s-.. ticr.a:-;

to hold the rod and c:ol".sequen:iy

he reset lever. in the reset pos1con. 'w\'ben :he coil voltage is P-<11.u:ed s-.;::ic:ie!ltiy, the :'!!set lever spmg a-*e:comes

he ar::ac:on o! the ar.d :o:.tes the ::i!Set !eve: cioc:t*wue

.* .;,s the reset le\*er rotates. it c:.a::-:es wit."! it :he !a.u:::. pin re!at:ve to lar.:h u.r.::il :."le la::.'l is re!eased.

\\'"ben the the :rip s-;::"--:g

rip !f!ver to ::O:p the breaker. The lau:h is by eos.s ba:.

adj:..LStable !.:*er a.s

.. Reset Lever Spring Adjustment Screw Page 2 ..

• of 2.

....... . .. * .* .. * "PHOTOGRAPHS OF THE SALEM UNIT 2 "B" UNDERVOLTAGE TRIP ATTACHMENT APPENDIX D * -".J*"'!

-**-.* .,.., ,. --r---.. -* -.. :' . . ;. Research Center A Division of The Franklin Institute The Benjal'nin Framdin Phil-.. Pa. 19103 (21') 448-1000 . . : :=*. -..... * -: .. . ' . ***.* * . ._,, ... --.* . .:.:>: ' . ** ....

' _,: .* U Franklin Research Center *. A Division of The Frankl.in Insc:ute Mr. Vinc:en: Noonan Chief, Equipuienc Qua.lific:aciou Branch U.S. Nuc:l"ear Regulacory CQ:missicn Phil.lips Building, !".S-P-234/R.'!

P-234'0 7920 Norfolk Avenue Bechesd.a, MD 20014 IO 3E BY ADDRESSZZ ONLY

Dear Noonan:

" ?'.arch 21,, 1983 E:u:.loseci herewich a.re the pbot:ographs of the Salem Cnit 2 "3" Cndencltage Trip Ac:ach:ment shcndng the rough surfaces of ehe la.tcb., latch pin, and lace:!! sprl.:ig.

enclosed is a descti.:?tio::

of the and a of che trip attac=:euc

'lli:h che photograph nucbers indic:aced.

GJT/ih Encl. Very truly yours, " '7 ' .b.-i _,-/ _,/'--* )

Gary J/. Toman Leader, Qualif ic:acicn Analysis and Planning Section, Nuclear Engineering Dept. cc: P. Shemanski ) M. carrtngton) wichou; accachmencs

.:0th&,;

Streets.

Pa. 19103 (%15)

TWX-110 670 1889 :*"':'*--...

... ,.: *.*.:"*.';*:.**:':**.

... : ... *. . *.* * .. : *,:.: . -:

' ... -**

• I * *

, .. *". .:* ... SALEM ONI'I' 2 *s**

Tm:l' ClttT A P!::otoa:achs of Late.a,. Latch :?in. !.ateh Serine, and Latch Retention Pin

• Picture Nul:lCer l 2 3 4 5 6 7 8 9 Oescri:tion of Picture Side view cf latch showing-rcugh;.'ess of area . Front view cf latch shewing latdl nctcll area Front view cf latd:l with slight angle to right shewing face and slight mushrcoming cf latch edge* tat=i lcckinq up intc the latch surface showing and mushrooming cf edge Lat=l spring showing wear and abrasion f:o= rllbbing on latch sur!ace (between arrcv) Inside of l.atch loop showing surface that rides on latch spring retention pin Latdl sprinc;J :etenticn pin showing poine of contact of latch inside .locp edge (betveen arrows) tatc!l pin showing surface and edge wear f:om latch 'I'Op view of late: pin showing wear ?ictures taken March 18, 1983 Description prepared March,21, 1983 Franklin Researdt Center 4 * * . *"'_..-*:_-

__ *, -.:

.' ,*: .... .. *. SALEM EVENTS OF FEBRUARY 22 AND. 25-, 1983; FAILURES OF REACTOR TRIP CIRCUIT BREAKERS; FINAL REPORT .-OF INVESTIGATI9N APPENDIX E *, "ftnklin Center A Division of The Franklin Institute 1'1te Ben,amin Franklin Parttway, Phil&. Pa. 19103 <2151448* """:"'""*" ____ .,, ........ . I *.: ... " . ' .. -. -* .... *. -*' . . * * .. , :;<:'"'* -. *. -" .. ** ' -

! ** * *.*

• . ****-* * . ' '

..*

Research Center -A Division of The Franklin Institute

• • _, .

• Mr. Vincent Neenan Chief, Equipment Qualific:aticn Brandi o.s. Nucl.ear Rec;u.l.atory CQmm;ssicn Phillips Buildinq, MS-P-234/lU!

P-234I> 7920 Norfcll Avenue Se t:hesda, MI> :ZOOM Ma;ch. 22, -l.983

Subject:

Salem EVenu c! February 2.2 and 2..5, l.983:

of React.or Trip Circuit Breakers As yo12 requested,.

encl.os1!d is the final. report of t!le Init:ial Investigation of Failure of Sal.em trnit l Reactor Trip Circuit to Trip on February 22 and 25, l.983e GJT/jg Enclosure cc: P. Sheman&K i M. Carrinqton Yo s trul.u ti,,'/

QUaliiicatico Analysis and Planninq Section, aiclear Enqineer inq Cept. 20th & Race: Strec:ts, Philadelphia, 19103 (2lS) 448*1000 TWX 0 710 670 1889 .. *::::* . *-* ..

.Evaluation Initial Investigation of Failure of salem onit l Reactor Trio Circuit Breakers to T:i'O on February 22 and-25, l.983 PrepaJ:ed by Franklin Researdl Center Much 21, l9 83 .-_.' -; . *. '* Investigation of the failw:e o:f the Salem Onit l reactor trip c:ircuit breakers to open when the undervoltage t:ip attad:lments (OVT) were de-enerqi.zed by the solid state protection system on February 22 and 25, l983* inclwded review of.the operatinq, maintenance, and surveillance tesdnq hist=ry for the :CB-50 c:ircuit breakers used at the salem plant. .ini:.ia.l i:ivestigation centered upon the OVT attacbment:

subsequent efforts included the interac1:icn of the tJVT attachment with the c:irc-.:it b:ea.Ker.

'!:be t:ip lever of tbe UV'? attaement must lift t:!le c:ircuit breaker t: ip bar for openinq of the circuit breaker to occur. '?o date, two possible failure mcdes llave been deter:llined fer the Salem 1Jnit l TJVT attadlments. first was observed by the Licensee and by N1C personnel the day of and the day a£ter the 25, 1383 event.* 'l'his failure mode apparently occurs when latcll-to-latdl pin binding prev.ents unl.atdlinq of the TJV'? attachment, there.by preventing the trii? lever f:om lllOV'inq wl:len the device is de-enerqi:ed.

Shortiy after the February 2S event, all but one of the failed devices were luoricated and no furtller failures tc operate c:culd be repeated. :emaininq failed tJVT attachment was subsequently damaged and was not available far inspection.

'l:he second possible failw:e made was reccgnized from inspection of the ."JVT attachment prcvided ta Franklin Beseard1 Center (l'SC) by the Licensee.

The latch. sprins*on this device exerts enough force on the latch to :educe the output force from the trip lever as t:be friction increases between tne latch spring and latdl with aqe and lack of lubrication. reduced fcrce could be significant if the farce required ta lift the circuit breaker t:ip bar is higher expected.

Ori March µ, l983, Westinghouse Switchgear representatives stated that the expected force required to lift the :: u:.JU Franklin Resun:h Center 4 0.-. al 'nl&.F---... . ' -l-

"'. '* *. . .-*: ***.--circ-.Ji t breaker trip bar at time cf manufae-:ure would have a maxi::lu.m

• cf ll .. ounces and a normal range cf 20 to 28 ounces. March l7, l98:3, ?R: personnel measured 28-to 30-:iunce lift force requirements on five cf six Salem reactor t:ip circuit made available for inspection by the

• Licensee.

were the four Onit 2 circuit brea.kers and the onit l *a* bypass circuit breaker. 'I'he sixth circuit breaker, the present onit l *A* trip, requil:ed 38 ounces cf *lifting force for operation, indicating*

that reduced output force trom a tJVT attachment coupled with a high trip bar lift requirement is a possible condition.

Durin9 the investigation, vuiations in construction were noted amcng the original OVT attac.'lments supplied to the SaJ.em plant. device given to FIC had a tight latch spring. 'I'he remaininq device that was made available to PB: for :inSpection had a mudl looser latch spring tnat exerted no force on the latdi excep1: during actual latching operations.

No reset lever spring ac:ijusaent*

lode washer was found on the device provided to ?R:, whereas the remaining Salem device had the. lode washer. Discussions with NFC and Licensee personnel indicated tnat similar variations were noted in the other OV'l' attadlmeuts that were no long-er available for inspection by PBC. '?be latc.!l surfaces c..f t::ie oriqi.naJ.

tJVT attachments were found to be in the as-stamped scate. P.oughness was noted on the late-to-latch pin face and on the latdl-to-latdl spri:iq face. On* the device provided to FBC:, this roughness on the latdi spring side of the latch had caused a 9rcove that is al.so rou9b to be worn into the spring. Irretgularities on pin sw:face of the late were noted on the E'lC device and the device tested by the Licensee.

During-testing cf the F!!C device, hesitation during unJ.atching was observed when voltage_ was graduall.y reduced to the coil of the device, further indicatinq friction*

in the iatdi-to-latch pin surface. Photographs of the latdl, latch pin, and latdi spring surfaces-taken on March l8, l983 shew the il:regular nature of tbe mating surfaces.

• Subsequent to the manufacture of the Salem tJVT attac.lulents, Westinghouse c.!langed the procedure for the latch to include hand honing of the late!: surfaces that mate with other components.  : *. ,. ....... _ '. ... *' --.:****.

i. I, -* .,.* ..... on Maren 17, 1983, fie personnel.

examined the new UV'? attachments supplied for Salem onit l. 'ttl.ese devices were found t.o*have "the latch-spring surface honed. Other surfaces could net be because the devices mcunt.ed on the circuit breakers.

Variations in spring force were noted, and one 1JVT attachment h3d spring forces similar to the original.

device supplied to FlC for evaluation.

on Maren 18_, 1983, Westingllouse.SWitcbgear Division personnel indicated that quantitative acceptance c:ite:ia have not been set for the CVT attadlments.

No output force reqQirement bas been set and no quantitative requirement for medlanic:al unlatdlinq capability exists. In addition, no such requirements lla.ve been set for fieJ.d testing UV'? attachment ope:ation and circuit breaker trip bar force. Conclusions bel.ieves that'in the as-manufactured

• new* condition, the 1983 WT attadlment will prope::ly trip a ci:cuit breaker that bas a eip bar force requirement that. is with.in the design limit of 31 ounces, and would probably consistently trip a circuit breaker with as-found bar force of up to 38 ounces. Eowever, sufficient evidence has not been presented to shew current manufacturing processes for the UV'? attad:lment when coupled with maintenance will eliminate lcnq-ter.n failu:es that appear to be mechanical, age-related pbemcnena.

'?be variations from device to device cause concern. inie fact that boning is a hand operaticn*indicates that variations in the surfaces of the la.tdi will. remain even though no extreme rou¢ness should be expected.

In the lack acceptance criteria adds concern that impending failures might.be missed during inspection and maintenance.

on Marcb 18, 1583,, Westinghouse Division also indicated that the CVT attac:bmen't imst be replaced some time during the life of the plant. Criteria fer determining when to replace the CVT attachment do not appear to be available.

FllC

• recommends the fcllcwing actions:

• UUU.i Franklin Research Center

• .. *,*.!*-*.

• ,.** ..... '7'"""; .. ... ..... "':"** .

-.. . ' ,:': ..

l. Acceptanee criteria be set fQr parameters a.ffect.inq correct of the WT attac:fuaent.

.. 2. 3. 4. 5. Testing methodology for ac:::eptanc:e tests be prepared for and Licensee use. oniformity of const:uction be instituted or sufficient testing be perfor::ied showing that the variations in the devices are of no consequence to reliable operation.

of the OVT attachment be performed to show that the device can successfully opera1;e for the. intended lifetime with proper maintenance.

Criteria ?e developed to deter.nine a replacement interval for.the tiV'l! attachment such that replacement occurs significantly befare the fail;ire.

Data and information prQvided to date indicate that the lonq-tea reliability of the CV'!' attachment h.as not l:>een proven to be adequate.

'!he reliability of the UV'1' attachment appears to be significantly.below that of t.:le OB-50 circuit brealcer to w"hid:l it is

?re pared by G. J. i:cman Franklin Center A_°'-ol The F--, ' ____ , ___ . . -.-.*

.*. :..

I \ TRIP 'REPORT: APPENDIX F .. SALEM UNITl REACTOR CIRCUIT BREAKERS . .

Research Center . A Division of The F'ranldin Institute The Ben;uruft Franldin Parit-Ny, Phila.. Pa. 19103 (21'1 "8°1000 . ** ......

.......

-*-:---**-* ..*.. ., ***. **.:**

.. **

• Tri:> Reoort SaJ.em Nuclea= Generating Station (NGS) !-!arch 3l, l983

Subject:

Salem Unit 1 Reactor Tri'C Circuit Breakers of Tri 'C: At lO am c:in 31, l983, Mr. v. Noonan of NH:, EQB mm, requested that Mr. G. Toman of PR: proceed tc sa.lem NGS to evaluate testing of the reactor trip circuit breaker. (R:Cla) trip bar force requirement and the undervoltage trip attadlment (OVTA) output fcz:ces perfocned at the Salem "Initia.l repcrts of the testinq received by Mr. NoCa.an frcm the NR:: Begion l Resident for sa.lem*indicated t!lat the devices were not operating within the force requirement ranges dictated by the me staff. Mr. Noonan asked Mr. 'l'Oman to represent NRR on this trip tc Salem *and to request the following:

l. Who performed the tests? (Company affi.liation and names of

2. What ':est methodology was 11sed? (F!C toeval.uate adequacy of method) 3. I.icensee is requested t:l prcvide pictures o! (non-lllandatcey)

• 4. I.icensee is reqo..:ested to pr_ovic:Ie a copy of procedures used for testing. 5. I.icens:ee is requested to provide a copy of the data taken on al.l. four Sa.lem unit l

• RrCl!s and tlV'rAs. o*. Did Public Service Elect:ic and Gas Company (l?SE.&:G)

Qua.lity Assurance personne.l witness the testing? 7. What is the proposed corrective action? 8. NIC will. app:ove corrective action prior to Al.l of the discussions with PSE&G by Mr-. Teman were to be in the presence of an NBC :aesident Inspec:tcr*

from Be9icn I. Recort: IJpon ar:riva.l at Salem NGS, Mr. Teman met with Mr. w. Lazarus, me Resident Inspector:, Beqicn I. With Mr. Lazarus, Mr. Teman met with the foll.owing persQnnel in the Salem NGS General. Manager's office:

Research Cehter A Chiooatlol'l'he F....., .._.. -l-* *

..

....

-. -* ** . -

--.

• .* -

. :* ... -:. * .. Mr. R. :Oderi t: * -Vice President, Mr. B. :Midura -General Manager, Mr. J. Boet;er -General. Ma.naqer Mr. Gore -(titl.e not kuown) N 1 clear sa..l.em NGS SUpport Mr. v. Pavincidl

-Principal.

Engineer Mr. Ga.llagh e.r -Manager , Ma in tenance .. Mr. Toman relayed Mr. Noonan*s requests to this PSE&G indicated that the testing was completed on March 25, 1983 and that the test equipment and personnel.

llad returned to the PSE&G.' s Maplewood Energy tab. :a:owever, Mr. Pavincic:h, "who had witnessed the tests, described them through use of color pilotcgra;tis.

PSE&G indicated that the tests were performed to obtain baseline data on the RrCBs and OVTAs and to ful.fill.

me requirements.

concerning OVTA output forces and circuit breaker t:ip bu force requirements.

Mr. l?avincic:!l described the force measurements, which used a load cell in ccmpression or tension depending on the test requirement and a Visicorder gr apilical.

reccrder.

l?SE&G provided a copy of the procedure, Maintenance Procedure

':-94 Bevision O wilic:b was performed on Marcil 24 and 25, 1983. '!he procedure contains all of the data taken during the tests and eac:i step is wi tb a PSE&G QUality Assurance*

Division stamp. Mr. Toman asked if Westinghouse was party tc preparation or of the tests. PSE&G answered tbat the tests were prepared and performed by PSE&G personnel.

Eoweve.r, Westinghouse, Ea.st Pi personnel were expected on site at Salem NGS" during the afternoon of Karell 31, 1983 to review and ccmment on the results of the .testinc;.

With. regard tc tile i;:Qotcgrapbs of the test setups, Mr. :ecet':}e.r stated the pbotcgra;hs being shown were the onl.y copies at the Salem plant and t:b.ac they would be needed for the review by Westinghouse.

Mr. Teman asked if Xe.rox copies could be supplied and new prints sent at a later time. Mr. Eoetger agreed to this. The main tests of interest we.re t::ip ba= static lift force tests, trip bar impul.se lift tests, and OVTA impul.se *output tests. 'l'he requirement that the static trip bar lift force be 31 ounces or less was met on.ly upon Bypass Breaker *B*. Eowever, all. of the RrCBs* tripped upon il!ipulse farces of' one pound or slightly less, and all of the OVTAs exerted .impluse forces of 3 .325 pounds or mere. For the static pull test on the trip ba=, the lead cell. was set up iri tension with the pull. point as cl.ose as possible tc the point at which the TJVTA trip leve.r would hit the trip bar. For the trip bar impulse lifting force, the load* cel.l was used in c:anpression.

'the trip bar was ca.reful.ly given a manual. impu.lse with the lead c:el.l in the area of the OVTA's trip lever. Seve.ral.

attempts were made so that a range of impulses brac:Jceting the trip/no-trip point could be recorded.

Research Center Aa-otTheF_._ -*-*. ----... :.*.:**: __ . *_/_ ...

.. _ ..

.. * .

-.... : -. . . .,,.., ... -.* : ..

.* .. ::-::*

.. *. * .. : . . *

• .*. '-; .., .. , : ---* . Note: 'l:be impa.lse t=at was manua.lly applied 4urinq t:be test. or may not approximate the impulse output of the undervolt.age trip attachment.

No c:cmpari.son data. are ava.i1a.ble to show that the accele:aticr:is, strike velocities, and pul.se widths are similar between the manually applied impul.se and that applied by the tJVTA. For the lJV'rA impul.se output tests, the load cell was _placed 0.090 inches above the trip bar so that the force imparted to the trip bar. by the OVTA could be determined indirectl.y.

• Note: 'l:be method of testin9 the output force of the tlV'rA used in these tests may include additiona.l forces generated

'by the trip bar du.ring un.la.td1ing of the circuit breaker trip mechanism.

These forces may add to or subt:act from the CVTA output force and could be t:be explanation for the wide ranqe of readings f:om 3.325 to 6.50 lb. ?R: agrees witJ:i that, due to the variations between the CVTAs ami the RrOS, teseinc; t:be CVTM 011 the :s:ros is mere useful than testinc; the separate.ly.

PSE&G llas attempted to take as much*base.line data as ?OSSible du:inq these tests fer ccmpa:iaan with fu.tw:a test data. A c:cpy of PSIU.G c:cmpletad t..est pi:ccedure is attached.

PSE&G stated that:. comparison of required i=?*lse t:ip bar force tc 'CV'l'A impulse force output sbcws adequate margin and indicates mat no corrective action is required.

BOwevei:, FSE:t&G wishes ta b.ave Westinc;hcuse review and concur with the test resu.lts.

l?i!C makes no judgment concerning the va.lue of the data (see notes, above)

• With re9ard to personneJ.

performing the. tests, M:. Pavincich witnessed all.*tests, Mr. B. Ricks supervised the load c:el.l tests, and Mr. acmancsky

• supervised the timinq tests of the zrc:ss. G'?/jg Research Center A FtWlda U.... . *.

.,**. Prepared by G.. ':Oman April l, 1983 '.":* .. *' ...

I . I *' " ...... '* PROO:OlJR!:

A-ll l P:t<X:!:ltiltt r l .-_T ...... -.. 94 ..... __ _ '*.... 't.l!'!.Z:

S?ECAL Ac=OIS:TI'(N OF Tm: #l CNJ:'!'

• J

' P-T!V. NO. AND BYPASS EP.E;.XE?.S

!'!US :?liCc=:t)tl'?.E IS ALL COCCMENTAT!C!1 SMI.I. BE IN ACCOROINa WI":H .ADMINIS'I'ItATIVE (AP) ll. .

?RoaDOR!:

HAS Bn:N AND TO BE IN COMPmra WITS Si'ECll'ICA"nONS.

0 I ,qf I DATE .. -I 1

• N/A I "'*-

I -... I .

N/4 *-r-ND!: U:n:::. Ill I I , '* \ APl?'RCV::tl

' N/A .CAT.:: AI :U::v OAT.:: N/A II'\ 'i'Jrk*,.;.

SQ:.Z OA'!'::

v MAIN!'. MGa. s.o.R.c:. MTG. Ntl!'.l3!3 OAn:* i *\._ \ I I MG2. -Ol?S. D APPROVZ:O *-" A-ll. . -*------*.;:

• • _ :**.: ... *.* . . Enclosure l Pa9e l of l . .-

.. . I Rev. 22 * *

-5.0 -6.0 7.0 *. .. SAIZM STATION OC'ARTME:rr MANt1AI.

PRoa:I:lOP.E *, Speeial Oata Acquis.iticn of the il Onit Reactor Trip and Bypass Breakers.

The of this l?:::oc:edu=e is tQ a1low fer certain data acC!Uisition on the #l Onit Reactor Trip and Bypass Breakers.

Data is to be collected by the i'SZ&G Research Corpcraticn under the direction cf the i'SE&G Nuclear E.""lgineerini; Department.

Th.is Procedure is applicable to Salem Onit U only. 4 .l AQi:li.st..:ative Proc:edure (AP) lS -'l'agg'inq R!2les. 4 .2 Procedure 9 -Ccnt:::cl of Station* Maintenance.

4 .3 E:l.eceic I:lst:-..cticn

au.Uetin 32-lOO-U, Lew Voltaqe Metal.-E:nclosw:e Swit:hqear.

ENC!.OSUM:.S S.l Enclosure l -Breaker Positicn and Serial Nu:llb!rs.

5.2 Enclosure 2 -Breaker Position "As Found" Conr!iticn.

5.3 EnclQsure 3 -Breaker Position "As Left" c.:mdition.

As del.ineated in this Prccedure.

l?P.ECAOTIONS ANO ::t.DllT.AT:tONS 7 .l When Maintenance Procedures reqi:i:e wcrkinq arcand electrical equipment,

• safety precautions am.wc:k*habits specified in the Electric Production Oepartment Safety Manuiil*shall bl observed :by all personnel involved in the acccmplishment of those Prccedu;:es.

7.2 When slinqs muSt:

to acccmplish Maintenance l?rccedures, they shall

• be used in a.cco::dance with the requirements of Maintenance M2Q. 7.3 When Procedures requi:e wcrkinc; en stainless steel. components of the primary system, the following items shal.l not be used: 7.3.l Carbon steel wi:e :brushes and scrapers.

Pac;e l of ll Rev. O *.* :*:.*** ..... !:

• . *r-. --

.. ........ -... :* .*. *. .. . .

c <.: _;:* . ' *. i .4* . ' 7.J.2 C!llcrinated solvents, e.q.

7 .3.3 Onapprcved

ask.inc; tape, ot.'ler fo:: use on protective

_cl.otlli:lq.

.. In addi ticn to the no:::ial precautions taken when workinq around eleem..:ica.l as speei:fied in the :E:lec-==ic Pro:iuction Depart::ient Safety Mam:xal, care ::nust be taken to stand clea.:: of t.1le various mechan,ical mechanisms associated wit.11 these breakers.

i. 5 :Se.fore any work perfor.ned, make c.ertain that all control ci..""l:Uits are open and that the breaker is removed fJ:cm the metal-clad unit. 7.6 Prior to applyinq solvents to electrical components, ved.£y ccmpcnents a.re il:cune to the solvent.*

on w-2 switches, no solvent cleaninq is qui:ed or authorized.

Clean wit.'l clot.'l only. S. 0 (l) (2) --... (3) (4) ':'-94 8 .l ca.1;.._atic:i data:

"'M::N'r ZQm?MENT CLDRATION CLDRA."'ION SERDL N'CMSZ2 DAT!: OW-DAT!: , I PL-?t/&

rt-. :i°nz: [\

ffi*oo;; S:t-+=1 s L:t. Is?. : .J'/.y 5"2.4? 4.iz "Cl 8.2 dl\ \l 01" ::::itar the ";s.s Found" seria.l r.i*::nte.rs all fou: (4) Reactcr Trip and Rea.c=r Trip Bypass %ea.'lters on E:iclosm:e No. 2 before rem.ovi:q*

any breaker ::rCZL cabinet. c::::ipa.re serial tc t.'lose listed in :::nc:losw:e No. l. Notify yow: supe..-,isc:

of an-:r disc:epancies.

8 .3 Notify the Senicr I & C SUpe...-viscr tha.t all four (4) Ileactcr Trip and ::.eactcr

"::ip Sypass :Sreake.:s'will l::le removed frc:::n their c:ul::lic:les am t.1lat I & c testinq will be required "after completion of t.'le Procedure.

Notification is due tc the I & C Oepart:nent's

eqiiirement tc perfrJ(fJ;;;:=

Trip Master 1'esp=se 1"_.,e B '3 Senior I & C Supervise:_

Siqna:ture

'Date Physical eis-:ance

!:c:n f:cnt of t::ip latch face tc t::ip bar. Reactor T:rip :Sreake.r INSPECTION HOL:> POINT " ... " ....

II Paqe 2 of ll . I J I ! :. -. __ . T-94 * , ........ . Reactcr T:'ip B:eaker "il" INSPEC'l'lON BOLO ?OTh"'l' R"eac:-...or

ay::..ass Breake= "A" . \ INSP:C:t:nON ROI.:) POINT Reactc:r B'!{IlasS B:reake: "B" StlPEa'VISO:vwr:NESS INSPECTION BOLD POIN'1'

• 2.5\o ,,

• 9 .2 Clearance

-:rip lave: a::C: trip ba:

• Reactcr '!:ip S:reake: "A"

• I 2. StJPE:RVISOR,/WrrNESS INSPEC!.J.ON BOIJ:) POINT IJ ::\eacto:r T:ir B:reake: '"S"

• O/S'

!NSPEC'nON HOLD POINT Reactc:r Bypass Breake: "A"

• INSPEaION BOLD l'Oim' .. .... 1" ,, Reactcr Bypass. :Breaker "?5 11* "":" e 1 '1

• O & C> *.*.

mSPEC'I'ION BOI.Il POM Page 3 cf ll .

J. *.-: .. *. . ' .: ; *. . ' .....

Rev. 0 ---.

.;;_:*":

• • : .. **,. *.* . . ..

( .. *--... *; ... ..; , .. 9.3 Cistance :rem Cente= o! Trip Bar (Breaker Mechanism) am St=iJce Point of Or..dervcl:aqe

'!:ip Leve:. :Rea.etc:

Trip Breaker "A" S_.

,, .

!NSPEC':'ION HOU> POn."'l' Reactor Trip Breaker "'S"

-'r:.., .. .:=. %,\ ' e .-: ...-..:, !HP-o-=.>"""--* is-: .

l t SUPERV;SOlVWITNESS INSP!:C.,J.O?r POI?."T Reactcr ByPass Breaker *A" INSPECTION HOU>

,, R_._

0 1/ Rea.C-:cr Bypass Breaker __ ....;:9.w.*

..::';":..;:C..::;.::Z.:..5".._

______ '"!"'"' ____ _ S'CPEltVISOlVWI"r.mSS

(-*

!CI.I) l'OIN'l' \, ........

• 9.4 Measure ::orce Reqa.:ired to beqi:i to Cc::press

!.atch Sprinq (Flat Brcn.:a Sprin;) Reactor '!:ip Breaker "A"'* 100 o .::s+O\C.

.. INSPEC.l:l.CN HOU> POINT .

Reac:tcr 'i_ .:.ri :Sreaker *3 11 .2. 00 SOPEltVISOR,/W!TNESS BOU> POINT I

\ . .<--> .J, :Rea.etc:=

Bypass Breaker "A" SUPERVISOR/WITNESS Uoc>a\e

/ .u. ... :,, "'W*.;,

I I** I?1SPECTICN HOU> PODrr T-94 0 . IL-\

= 6 . *t ""'-::! __ .c,o Page 4 of ll . **. ::.: . .; :: ... . *****.-.** . '-* .

3 '"'"n&.:1'

' .Rea-""tcr lirr19aes l!reaker *s* .li-=

..

.... ON BOIJj PO:Dn 9.5 Record Conditicn of Latch Surfaces or Not) Reactcr 'I'rii;> Breaker "A" ,?\ ... }) -

efk.q_ J.

• * \ ,,.-,rJL..l

.. . . ... . .

Breaker ":s*

c.fP'gAS,. . 4 . p 0\,')\..-cQ.

-+.

e \:.........

• IN?PECTION HOLI:I PO:Dn POINT R.eaC""..cr Bypass Breaker "A" HOI.:I POINT *** .:.*c* * *, -c* .. eak ,._,. :.ea -c ... =. '!3*-S ... r e:

zNSPEcriCN BOLO POINT fg; , BEMAPJCS:

.. I. i . I . i " .. ME:AStmEMENTS

-SHONT *ntDS A1'TAC3MENT 9.6 Physical Cist.a.nce

frcm Shunt Trip I.ever Pin to 'I'rip*aa::

• * -: *-..... **-** .-94 Reactor Trip Breaker "A" *

,. SW)

I c-._ *.* (IE . ..

BOLI:I POXN'l' Page S of 11 ... **, .. .. .. **:

-.... :-:--* *-

...* : . .*. .. :*. . '* :-. '* .... *. *,.,'. :;. <, ' .*.* _,_

• 0 . ' .'. *.

(---'. . ' ... *l *('* \.. '. *--::.' ... 9 ... . , T-94

'::ip

• 3t1 DISPEC'::ON BOLO ilonr?

CyPass :Sreake= "An n:sl'EC'l'ION BOUl PO!N'l' ReaC-..or Bypass Breaker n'Sn S'CPERV:CSOR/WITNESS HO!.tl POINT *

• g i;-"'i i;-..,

rt' £<'t, -,, -=--Clearance f=om S.T. Lever to T::ip Bar

• II Raac""..or T::ip 3: ea.leer *A* ,\\2.0 *stniERT..SCR/WIT?:ZSS r.lSPE-r-:?C.': ?ODn' R.eactQr T:ip Breaker "1!"

• 0 s:.b.

I:?SPECTION

!OI.l:) ?OI?:T Reactor Bypass Breaker nA" .o-;a St!PERV'ISOR/WI'l'NESS INSPEC.!..!.ON HOI.:J POI:rr "' . Reactcr Bypass Breaker . n3* 0 St!PERVISOlVWZ'r.tESS INSPECTION HOLD POn:T Paqe 6 of ll ..

..-;;; ::-: ;-. ..

• S:.'i-1 .: ** .. ! e .,, ;J"..; L -. -* . J ',-(Imt)"

....

A .1 l i-r W"" '<::t::P > * , "--Rev. O i* .* . .:,,* .

• T .. 94 s.:. Lever.-:rnS?Ec:r.:CN BOLO POINT \ -7." '1 ., '2. loq 'e> Reactor T:'ip B:reaker *B 11 HOLO POINT .. ' '1 'i 7"2 .. (#Cl) \:> R.eactcr

Sreaker "A" nGPEC!lON HOW POINT y 7 e, INSP!.:c:lON BQU) POIN'!' _, ... z.., 'I 1'Z Reac:tcr T:'ip Breaker "B" Reactcr Bypass B:reake: ...... _. **.*.* ,. :1.:F*.:*

_-.. , *-----*--*---* --.. _..,......, .........

'-...... .. . . *.*.-*--.... ..

.. .-

-. t'... . -\. I '-. T-94 Reactc: T.:ip :al:eake.r

""A" lb 1 -:..c.i..:;

\'-=

H,-. S" ::

• 1

.o \'., .Z.,lo'Z."S"'l.., 1 2.,,S' \'o, "2,,(9z.<;Jb INS"i?Et...

:.l.ON

S:CI.:l ::io:nrr 1'-\\ ....

\\oi .. *'"' . *"? t Aea.ctcr Trip Breaker F'u.t\ .:..Ce \\a 'Z..,\2."$"\\,.

\.?i15'1t..

7...eee\\,;, , r.;{' °': \,9/'5'\b.

I .Cif?I: Reactcr Bypass :Sreaker "A" P .... l "Z. * \ z. c;' \ 'o. 2 .2. i;' \\, '? . \ 2. <)\\. 'Z. "Z.S"

'2..4a ReaC:-..o:

3ypass :Sreaker ":S" S 1.;.:-EXV!SOIVWn'NESS .LON Hoi:c POINT . . Aea.C:-:cr T:ip :Srea.ker "3"

R.eactcr Bypass*Sreake.r "A" POINT R.eac-tcr Bypass Breaker ":S" ., . *** .. * .. I '" I 1 ' I PN.'\. O'( s ':, \I ?co lb. *".'75" \'a 'C$'C * 'S"c:i '..la * .,-o

• Q.e.\e.c:..

I ic;' J \::)o; . nzc;

• i5'o =s l\Q.\Eo.H K' '-l t.e.S e Paqe a of ll

\ .. C!m?Y .. .. :* _J

_, 9.l2 Fo:ce by Undervoltaqe Trip Lever on the T:ip :aa:.

Reactor Trip Breaker "'B" INSPECTION BOLD POXNT J t:; ": .;z. 'i' o"St

9. l.3 =orce Generated by Trip Lever on the Trip Bar. -* Reac;:tor Trip Breaker "A" Trip Breaker "'B* . INSPECTION HOLD. POIN'r_ ' . .. SCPJ:!RVISC!V'WX'!'NESS INSPEC'nCN HOU> POIN'l' Reactor Bypass .Breaker . ..... .......

• , *..._. ___ ,, ... *-94

• oc; Pa.qe 9 of l1 . .. .. '.-.. \<" .. .. .*, . . :**:**.* . *. -.,_;_.-.*

.* * . .. .. ':

I . , \

j .

Rev *. O ** ..... * ..

{ \ ... **-,

... ) 9.14 T-94

!NS?EC'l'ION*SOI.D PO:tNT Reac::for T:ip B:eaker "B" INSPECTION HOLD PO:tNT Reactor Bypass Breaker "An INS?ECTION HO:m POINT ' ?.eaeto:r Bypass Breaker INSPEC'!l:ON

s:o:m POINT INS?EC"nON HC!:r.D POINT* Reactor Bypass Breaker StJPERVISOR/WI'l'NESS INS?ECTZON ROU> POINT ::?aqe lO of ll 7'*/p ri"mti!!

Rev. 0 ., -c: ... * . -

... *.

• 9.2 ?.ei:stall all four (4) tip am Bypass in their respective.

as indic:ated en Enclosure P.ecord on Enclosure 13 * .. INSPEL.llCN BOU> POnr? lO.l Notify I & c Senior Supervise=

that Maintenance*

has been ':. cc:i.pleted

• en all four (4) Reactor T:ip Breakers and Reactor 'r:ip Bypass Breakers on i Unit. I & c Oepa.rt::nent c:an proceed with their t:i:ne response testinc;.

I Seni.cr I ; C Supe.rvi.scr Data Time 11..0 ll.l c:cmpleted: -----ll. 2 Remarks : ll. 3 'l'est perfor.:ted by

--.. ...,--=*=.i:k;::;:..---------.

• Siq:ia-=:e ll.4 Reviewed by: . ----------------------------

bo..-T-4' . 4.\\\ I r...

...

Sc:.l4 ** . \.\) t>"" °J.

!:ic.fo\

I

..... 4u. '. rP ' Gc:-e T-94 Maintenance Supe:visc:

QUa.lity Assurance

.. ,"\ \ *. iiolc

'-2 \ IJ.S.-C. "

'?tt, 0"" -?o PA ' 6 iC.. . -i-o U.l> -e c.J.:e. ?J..,oN aa . c,) 1 lu* I P-:Z... Pac;e ll of ll Oate Cate Oate .**-.-** .

"": .... 1'.-::** ...**. * -. . . . . . . . . . . . ... ' *: .. " "':

': . (SW) (IBP) -ReV. 0 /.: . ' .,., '. *. '* *,* . . .. . r'

.. .. -----------. MA!N'nNANCE T-94 ENCLOSURE l .. BREAKER SERIAL AND POSITIONS SERIAL NUMEER: l24Y7268B 224Y7268B 324Yi268'B 424Y7269B A TRIP 124Y7269B B TRIP 224Y7269B 124Y7269B 224Y7269B 324Y7269B 424Y7269B A BY-PASS 324Y7269B Enclosure l Paqe l of l B BY-PASS 424Y7269B

.. .* Rev. 0

.. ' .

• .... :1 . _,..* *. *-UNIT il I Reviewed By: --------

PROCZDOzr::

'l'-94 " ENCLOSURE 3 EREAKZR AS U::'T CONDITION . \ A TRIP _____ A __ B_Y_-_P_AS

__ s ____ _...I B TRIP .Maintenance Supervisor Quality Assurance Enclcsu:e l Paqe l c:f l B BY-:?A.5S i:iev. a .. ..

.APPENDIX C Wes: i ngl'Hl lS$! E=:nit Co;porati.on Wat!r Reacicr . --\ Mr. H. Denton, Oire<:tor . . Office of CDc:ni s$ion Phi11ips Building 7920 1"l} 20014 Hr. Oen-ton: NS-EPR-Z737 22., l983 Tht! ?Urpose of this let.ter to provide you ..,*th tne 1 on on>-

iru1es'tiga'tion into the malfunctions of the Sale= . P1 ant rea4;tor tl"'i p Our gati on, a 1 on9 wf tll analyses for the

?1 an-:, d=onstrat!!s that the pl ants

• wi"tr. tM s iequi pment c:.:ara continue to operate wr1 undlJ@ ri to pub11 c llea1th safety.

• On Z0-21, 1983, Westinghousa a detai1edll fnsp.:ctian of the unaervo1uge trip at'tad'lment t21at was provided tg Westi ng;,ouse and was r9presented by ?S£&G && tne UV attachment.

that malfunctioned on Reac:t:)r Trip Sreaker S at Sa1e11 Unit 1 on 25, 1983. P!1ot.ograjjt?s

!nd an aud1 o tape recording of the P'arch 20-21 1nspe-etion are. available at ltestinghcusa fer your review. A. detailed ..-Mt-ten 1nspect1on is being-de...e1oped fl"m ttds eva1uat1on.

io tn;$ is the only such detai1ad inS"?eCtion condtN:ted ta date of a IJY attachment rei>rtS1tnt.ed 1s OM of the tllo that u1fllnctioftt!d on 25, 19a3, at Saleti 1.

• In this 1n$pection, Westinghau:e a list of i::ostuhted malfunction sce:narios for this devic*e (s.ee Tabla 1) *. Thi' 1nspei;'t1on w.as conducted 1n order to whicn af th41se potentia1 scanarios 11igh't have produced the mal of the Sale11 Unit , devii:e.s..

me fo11Clring 1s a of the key* findings in tbis examination.

l. /4s l"'eceived fro= PSUG. the UV device wou1 d not 1 at.ch. was also 1 hist.cr;r of bt"eaker c1osure prob1r.zrs on Reactor Trip a ;s repo!'"t8d in the !ilRC TasK Force Report. MUR.EG 09i7. .a.s a the test circu;t bre1ter on this device was installed not clo$ed e1ectrical1y or manually.

Th1s ;n.a,bi1ity to -..

• * .. latch tile. UV deviee was traced to a bent and deformed -phosphor--

• bronze leaf spring which ..ould not iaaint41n the against the traveling lat.en mechaniSl'I. defoM11ation observed en the leaf -sprl ng d not l'lave !:>een caused by nQr:iN1 operation or wear of this deYice. Had this UV device been 1rista11ed on Salen lkJit l ReactrJr 'rrip Breaker Bin t.ie condition.as rei:eived by Westinghouse.

ft would have been d1fficult to close the circuit breaker. It should notad that. had tM s part.i,ullr'"1eaf s-prl ng been vely defor:ied against the trave11 ng 1 at.ch, the IJV could tfave prevented frt.im unlati:hing thus preventing the from opening. A Wes ti oghouse repn!sentative sent to the Sal em site en February 27. 1983, noted. that a 1eif spr1ng ...as deformed on at 1east one !IV device shown to him at the by ?SE!G perscnne.1.

Tni s cievi ce was described U> hi a by PSE!S as one of the !JV attachments that ri;a 1 functi onad .on February 25, 1983, at Sal em I.hit l. 2. The axmi nation at 'liesti nghouse of the UV disclosed a 11fssing *

• 1oct washer on the drop.out voltaie SCr!':i *Chani sm. The l!Cfj ustment scTew was e.xcessi ve ly turned*i n *, a condition wM c:h reduces un1atd1ing fon:e available llhen tile UY deVi(;e is ., .... *ene1"9i zed. In eendition.

a visual !xam1naticn by 'Jesti nghouse revealed that the device -.as 1 ubr1 cated. PSE!G r..as advised that a l!JDrii:;int adde<i to this 6evice after tne event of February 25, 1983.

ngnou:se is cur:-ently uia1yzing this iubMcant in order to 1ts t,ypa. 4. 'Near on the 1 atc:h and latch 1 nurlace was not excessi 'fe 4nd there -.as no evi aence of burrs.

e.:ccessi ve frl cti or.a 1 cannot be ruled out as a malfunction scenario since p.ost i!"tCident ha.0011ng

{manualiy exercisi!i9 the device and lubrication) prier to by *iestinghouse cou1d have raasted a friction force !:la l f\mct7 en s:anar1 o. No v1s1ble found of corrosion or brcicen or is:is:sing pal"'t.s, other tne prev1ous1y

• ntfoneci lock washer. There ...ere no obvious $.i;r.s of 1mproper nanufactura..

A check of each p.a:-t against dilll!nsioos and toleranc:as is being made.

tests that tne device -was ca.pa.ble of 1ts electrical*

function.

5. Artificial of the UY device reset IU'"1 was requ1red to ;;easure t?le tr1p lever by the UY device. The tf/st breaker "tr1 pped normally by the U.V at*tK.tnent on severa I atte!%:pts liiUl nor.!a1 tr1p Dar 1oad o( 1.15 pounds and )lf'1tn an 1riCM!ased 1oad" to Z.3 pounds.* A furt.her fn<:rsase in trip bar 1o.a:d to 3.3 p<>unds resulted in eri"a't1c breaker tripping by the UY dev1 i:e. The mu1&W:B e.xpect.ea fort:e re qui rail to actuau the tr1 .P bar i!037Q

---

* on a normal 08-50 is 31 ounces. lience, it wA$ concluded tha:t the UV at.tac:l"lment was capa..ble of providing sufficient fon::e*

• with to trip a properly "Cainta1ned pB-50 circuit Sased on. the above infonaation, this appears to rule out the postulated scenarios l throug!l 6 of Table 1. The effects of dirt/contamination (mecnanis::a Sl could net be evaluated due to the lubricant applieij to device af't2r the events en FebMJar.t 25, 1983. Eviaence of of t:ie drop out vo1tige adjustment scr!w (£lechanis:i

9) was present and may a contribut:Jr, but lta.S not, in itestinghou.se engineer-ing judgerAent, reason f.or malfunction.

Based en the resul'ts of this

'tl1e C1U11funct1on Qf the UY attacr=ent was not cttributable to design or :nanufactJJr1ni;.

The ?QStu1at.ed lll41funct1on scenarios which cannot be eliminated or whtch were present in the examination fall intc four categor1es:

l) frictional ano=alies, 2}

3) bent or .defoMDed

• • parts, and/or 4) misadjust=ent.

These fgur categories considered to fall under orre bN>ad cate9ory of mainter.a.nce r"1!lated causes. Sa 1 e!ll Restart He1' uv trip attac.ments-nave 1ns'talled at 5.aleii Unit 1. Thesa devices wer-e built by We$tingh0use and as manufac":JJre;j to Westinghouse specif1c.atfons, 1ncluding 1 ubr1 ca t1 on. 1hey llll!M! *subjec4"2d Z5 conseetrti ve opera';i ons without naifurction at the Switchgear 01v1s1on and. after **r.sta11ation a"t Sclle11 lh1t 1, were t.o at 1east. 10 add1t'!on.,1 peraticns also without ulfunct.1on1ng.

_ $1estinghouse nas a ccmpleta of Tecnn1ca1 Bu11et1ns and Data to A

Su11e"tin clarifyfo; the c1rcu1t.

• and und.enoltage trip at.tacht:'ient.

1ubr1cants an¢ lW>rlca"tion p.o1nts w111 be 1siued to a11 West1ngnouse plants, 1nclud1ng Salem, by. Jilaren 24, 19-a3. 'lifesti ngtiouse na.s evaluated tne in th!t bul 1et.i n and has-d2tsm1 ned tt\.1t 1 t does not ccrrosion af 'the undervo l'tage atuc:tsaent maur141 s g1 ven the equ1 prsent oper!t.i ng env1rQnments and 1uDr1cat1on s-pec1t1cat1ons fer tne Westinghouse-suppl iea OB-SO reacur 'tMP sw1ti:hgear.

• West1nghouse analys1 s of the S41em events, to 1ou en March 14, 1983 and 1neluded as Attad'Dlen"t 1,

th&t the pub1fc be41tn safetY ..oul d not have been affected had the gf the Februar.r 25,* l 983 OCCUM"ed at .full J)C.er *. 4dditiona11y, if one of the t.o no"1al1y l'Min ffftiwcur pusnps 1ost 4t fµl1 power, if t:le operat.Dr ae't1on tg unual1y trip the did not oc;eur for five fl1 nut.es, tne resul Unt prima17 s1ste11. pnssure* wcu1 d have been be low the pressur1z.er relief valve set.+10irrt.

Furtnermre.

for the 1e:s 1ikal1 case of less of 111 main pi::z:ps, opentcr 1c:tion at 30

• l.cul <1 nave resulted in pMear.r systera pressuM! r-eachi ns only the relief valve OperiJt.Or action at f1Te a'inut:s, upon loss of t:ioth feedlfat.er pr.a:ps, and with a reslisti' l'IOd:ratgr cc-efficient, would have resulted 1.n a primary sys't-""'m prt!ssure which the AS1'E Code *

• 4-037Q .... ....,._ *-.--. -----"------*-

• ** * .. *.

• SerYice l..!Ye1 C 1i-:a1t of 3200 psi a,. the accep'tanc!

criteria for t.'\i s event. This sequen;e ...cul d have requ; red the ope!"at.GT" to ignere SC?ven 11e,Jor tMp demand alan;s. Even if the reactor-.. trip b-reai.ers

-..ere not able to be tripp&d mnuilly*fraa t!'le control roes.

• there are other diverse 11eans ava1-1dlle fora the Salem to a react.or

&s prescrieed itr the'tl"" ATWS s:inx2dure. the'°'fOM!, r:cnc:1Ude that even if the hfghly unlfkely event -ere ta at full power, there would De no effect on pub11 c he&l th and safety. 6f"'n the results of this analysis, other nieasures by *the HRC, wM1e l"!Ot.

prov1de additio.nal iaargi-n and shculd further ccnfi denc! that Salem Un1t 1 can be resqrted safely. For exup1e, manual reactor trip from the control room after an tr1p demand .exerc'f ses both the undervoltage and shunt tr!P ..... attaci::ents of the DB-50 ntactar trip switchgear.

fe$t1ngncuse has con<luct.ed recent tests on a shurtt trl p att1chllent and has detenai ned the dev1ce a force at least JOm greate,. than fon

e necessary tr1 p the In response t;) HRC IE Circular 81-12. liest1nghouse 1 s pre?ari ng a Techn1ca1 Bc.111 e't1 n giving ons for independent . testing of the and shunt trip for manu41 l"!ac-:or t:"f p. .1li1 s Tedini ca 1 Bul1eti n wi 11 be 1 s:sued to all West1 nsncuse p1 anu, 1 nclud1 ng by Maren ZS, 19S3. IJY Tr1 P ..!.tt.ac:h=ent Desi an: A't tM C.O:sn'f ss1 crers' 11eeti ng of March 15. l 963 t.4ie NRC Staff e:i:pressed uncerta.1n-ey over the-UY "trip at.t.acha!nt design 11fe'til:ll!

and the foherent nzal"'1i n the tM p force generated by the undervo 1 tage trip 11tuchvaent and t.ie forr:t requ11"9U to 11 ft tbe bre.ake tl"'f p !><<r. 'iiie bel1eva uncerta1nt1es have eeen n!solved by furt.M-r infonsat'fon

.. Tests bY PS£J6, F'rank11n Cen't!!T", and nghouse f ndieau a nanna l tr1p farce marg1 n of loo-zoo perc2nt. .A1 sc, in 19i2 an uncteMQ l tage attad=ent, lil0d1 f1 ed as a l t of 3 re po rt!d UY a1funr:-t1ons at Rcb'f nson Unit z. was suc:cessfully usud for more than BOOO. oper&ti ons without 1a l func-t'i on. "t>>d1 fi ed undervo 1 U.ge trip 1ttaetcents sant as replacements to all operating phnts with DB-50 reactor tT1 p sw1tchgear

• at that t'fm. A1l subsequent Wl!st1 ngl\ause of under'tOliage t"r1p* attact=ents h4s i n=orporated the =odi fi cat1 orts Mele in A ntvin of t.ERs on l1Y trip attachment lt4lfunc'tions si nee Uie 197?

'fnd1cate.

that appn>xii:aatel,y t'ifO-thirds of t:ie ralfuncticns appear to be. so'fntenanca The overal 1 for pe,.. demand en Wes't1 ngnouse OB-50 ci r"Ctrlt breeke:-s has appli.ed 1n PRA es and has no't snown an undue-to tQ'bl r1 st. In view of the the c;un-ent design of. tl'\e OB-SO reactor trip switchgear a pT'Qper for n*uclear power plant protect1on properly fnstal1ed, tested. and 11a1nta1ned

• 4 4037Q *-

..

• Not'w'ithstandi"9 the foregoing, Tr.I-to you Qf March 1, 198.3 COl2nitt.ed to a tnerough evaluation and pro9rar1 of the UV trip attacf'slent to be by June z. 1993. Test and 'JH'"09r"aJI definition ar"'Q Q.x;per:ted "to-be e=-plet.ed t:>y Mare?\ 25, 19a3. FurthQn::wn"'t.

we intend to do an ; n-depth analysis of CB-SO r"1!act.cr trl p swi tchg1iar ma1funct1ons-'w'hich have ec::urred on nuclear p1ants to provide* additional .data t.c further its r-eHability.

This reviQW *

• wi 11 be based on hi sU>ri ca 1 LER data uld an1 O.fini t1 v& data wn-t ch lfe ean obtain frOfl ou.,.

phrrt rehtive to wtal m.Dbel9 of reac;t.or trip and l"Aa 1 functi ens reeordad..

We wi,, advi $8 !fOU of the completion data of review. P1ease contact me if you "'°u1d Hke tc di-'Z:uss this -zat't!!Y6 further'.

tMil.Y yours, E. P'. Rahe,

>bJc:leir Safety Attactnent($)

s . :t

• ** T"b1 e l Postu1 ated >ii.a 1 tuneti on Scenario'

1. Corrosf on: 2. Missing Parts 3. Brotsn Parts 4.

Failure a. OUt of Parts

• b.

• JiCi l;r Insufficient irip Force Frict1ona1 Area

a. wear b. Burrs9 H.!Jllj-U?S

c. bubt1caticA

s. Di i-t/Cantam nati en 9. Mi $adju:st:aent 1 O. Bent or Def orsed Parts 4037Q 6 ,-.. \

.. , I -* ... ATTACHMENT.

I of the Salem P1ant fQr Postulatad .Feedn:ater Malfunct1on llrit.hout Automatic Reactor Trip 'WESiINGHOUSE ElECTRIC CORPORATIOH M. ?. OsbcrM .Trans1t:nt*

hla7ysis E.afety

• • /J; t\IJ

?.ami.ger

'iransient Ana.lysis safety Department

' l .. .. -. t I t "" l c I I: t: "1C' t .,. i 'l" "" #-c " ! ' . ! ' . i , i . ( l l. . ' i .

• • --*.--* : .*-..... ---** --

-*-* *--. *"'.

• * * --------------SCOPE In light of.the recent of th!

trip breakers automat1ca11y at the-Salem p1ant, the of th\i is to predict the consequences of a failure to fer 1imiting plant transients whi1e the plant is at full power. The trensients analtzed.

specifically for*the Salem plant, a paJ-tial loss of generator main feedwater (1ow*due*

to the trfp .of 4 sfngle main feedwater.

pump and also a *complete 1 oss Qf main flow due to the loss .of beth main feedwatar pumps.

less event is that in the Salem plant FSAR. As $t4ted previously, tna purpose of this study is to predict of the to these and. as such, the plant are assumed to function_

norma17y.

with the so1e exception l>eing the c.Qnman mode fai1unt of the r*eactor to *automatical1y function as

!!'.7.'fti!'T*ienced art February 22 and 25, 1983. It snoul d be noted that t.'\e sl)uri ou.s steaw generator level trip generated on 2/ZS/83 wes as a ot nonnal expected control di fficu1t1es expericncad at -1ow n1i1 . -power ievels. also should be noted th4t the loss of a fetdwater pump on 2/22/83 was* due to a nonna 1 manueveri ng of 4n a1ectriea-l bu:s whf1 . configuring the p1ant 1n a Both of events are net normally expected at ful1 power and thu$ one should eonsjder mere credib1e events such a! a feedw5ter heater tian the more i'fmiting and mui:n less freqyent feedwat&r ma1functions.

The study a*thirty seeond.oparator time for 5 manuai reactor trip fo11owing the automatic protection system demand a simulation of the actual response of th!

ZS, 1383 event. The study al so ciers a more c:onserva.ti ve cperi!to r respDnse of five minutes in order to detenni'ne the sensitivity of the plant ta action. O::SClUPilON OF

£FFtCTS . -

studies (We.AP 8330 We$tinghouse Anticipated Withal.rt Ana11sh) of failur-e to tr1p events previously s.ubmitt;-d t.o the URC have identified the 11m1ting full power events to dffect1ng st2am generator main flow. The in main feedwater f1ow affects the overall heat capabf1tt,r the steam* generators and, as a result of the m1$1\atch bet'llreen prfmary Sida heat generation and the secondary side heat a heqtup of the primary system coolant..

If the reac:t.ar is prt=ptly > the *. auxiliary feedwater system heat r'moval to remove decay heat. ft>wever ff. feedwater fl ow to s"team generators is or te:nninated without trip the $econdary tem 11 be unab 1 e _to remove a 11 of the neat tna t is in the core. This heat buildup in the primary system is i t\mcticn of the amcunt of the feedwater and 1s 1nd1cated by risin9 coolant SY$tem temperature and by inCf"l!4Sfrt9 pressuriter water level due to t.ie insut"'ge of the r!actor 1eve1 in !team generators drop5 as the remaining 1n the steam generators 1s boiled off due to inadequate spply

-When the generator water-level falls to the point **he?'!!'

the-stem tube buncf1 e i $

pr-1ma ry to se-conau*y

'heat .*. transfer is reduced, reactor-cool ant system and temJ:>eT'ature z 3999Q

• .**--* ....... -* ...*.

.* =----:s--=*

=-=-___:__--_:__::_-'---_:_:____:__:___;.:***.:_-

-=*-=-=*-*--=--.!..:'

=====--_:_:_

• .

• tncr-ease at a* greater rate *. This 9reat.er rat! of 1nd * #

1hc'l'"!ase fs maintained the pres:surizer-fil1$ comp1iete1y weter-fs dischar9ed through the pressurizer re11ef Rear;ti vi ty due to the high primary system redu'=es cora power. As a result the system presur'! begins to and a steam space is a9ain formed fn the pressurizer.

• iht limiting criteria for the postu1ated transient3 that reactor coolant maintained

$Uff1c1ently below the to the AS>£ Code Service i.eve1 C (£.mergenc)".)

1im1tG. For the coolant system, tne is 3200

• COHTR(X. R0Cfi4 lNOjCATIONS ANO MJT!GATING ACTIONS . . Although reactor is from tripping the cor.:nor. hOure c"f the reactor trip b'!"eakers, then: are

<:i)ntrGT room indications and a1anns wh1ch.4ra dur1ni the tl'"ansient_

would servi the that the event has tixan p1aee.

ind1ca'tions in addition to procedures, ,mfch tb.e ver1f1cation of a suc:essfu1 reactor trip all wouTci the mitigation of the consequences of the-transient..

F'o1" a of narmal fe&dwaU.r event .. i" add1t1on to nol"'!l'ia1 precess 'gntro1 a1arms (pump trip, p'l"'essure, level and alanns for both primary $e<<:Ondary systems}. -audible alarms ,.,,uld tle generated:

l.

flow mismatch and low 1evet (each generator}

2. Overtemoerature Ce l ta-T t.urbi ne

3.

Oelta-T trip demand 4.

Del ta* T turb'f ne

5. Overpower-Del'ta-T reactor trip demand* S.. High pressure nusc:tor6 trip demand 7 .. High pressurizer 1eve1 re actor trip demand 8. Steam generator l ow-1 ow 1 eve l r-eactor trip demand 9.

safety injection (in eo1nc1dence with nigh flow) 10. Lo-w r!actor coolant. loop flow reactor trip demand Tables l and 2 show the time sequences for these alarm&-. ' .As part of the procedures the opera ts to se following any reac-toi-trip demand, tne operator h required to first the successful accomplistlnent of the tr'ip by Ob$ervin9 rod position indicatorst rod bottom flux, or reac'ttlr trip brgaker position ind1cat1ons.

The following actions are available to the operatcr in the main control. if an unsuccessful l"'t'actor trip occurs: 1. Martu3l reactor trip (with automatic turbine tTiP.) z. Manua1 turt)ine trip J. Manual turb f ne runbac:k (

n. ) 4.

safety injection 5.. flianual contro1 rod insertion

• 39990 J .. i. i i

.... , .. Outside the obvious benefit trip. th-e tur'.>ine

• trip or runbact action is the most im-portant, trip be obtained to tenninate tbe steam flow f.rom steam to preserve steam generator inventory.

Steam pre$$U"r"e

!nd hen'e primary temperaturt wf 1i be by means of dump controi system, steam generator relief safety Other medns out3ide .the main control rocm arc avai1ab1e:

1.

manual trip of any reactor trip

2. Local manual trip of "the rod cont?"'Ol system sets 3. Local manua1 trip of the turbine TRANSlENT SlMUlATIOH performed to both a and 10$$ of miin feedwater.

These analyses are based upon previous mo4el$ consistent with previous to the HRC by Westinghou'e on ATiiS {MS-TK\-Z182y T. M

• to Or. S. Hanauer, 12/30/79!'

but .11so tc lŽ)del the Sa1em Plant. ihe were for both analyses:

1.. In1t.ial nomai ful1 power cperetion at t:eginning of c;:)re Hfi!.

to the current cond1t1on of the S<<1em rtant and is a1sc the 11m1t1n9 eor.dition since the i;;ogerator teizroeratuF;!

coefficient is at leut vaiue. A value of -a'ix:"m/*F.

\ihicn*;s valid for 95i of core Hfe, was

z.

the pressuriier relief and valve$

tn funtt-:cn.

There are two and three safety

..

heaters and spray also function autcmaticany

.. J. The automatic turbine runbac:k on either OJ" Oe 1 ta-T si*gm1 l 5 f s operable *.

• The runtJ4c);

setpo i n t i s below ttre setpcint.

The turbine runback operates on 1 30 seccnd cycle. Turbtne 1 oad is first r-educed s: 1 n LS sacancu. * !f .Ji. the end of the 30 seconds the runbacx signal $'t11l exists,* the lozid further reduced another si and so on.

• The lQad reduction nas i effect on the transient and helps r-educe peak primar:r system pressure.

4. The rod c:ontrol system is assumed to be in the manual mode consist.ant with aetua l praet1c:e.

Autcmati c action a f the rod system ft'OU}d cause rod in$ert1on when primar.'

f ncreases and would be less conservat1ve.

5. The* steam dump control system is avai1ab1e.

The of the dump is of nominal steam f1ow at full power. a. Auxiliary feedwater f1ow (1760 gpm) begin*s at io Se$;Q.ndS f'oHolrfng of t.ne low-low 1eve1 signa1. Tnis response time is based upon actual data f'f"Om the Salem

7. Operator action is assumed to initiate a

!lUlr..ual t?"ip.

1s 1nitidted v1s the reaetor trip breaker

• ........

.... *-.. . . --* . -.... . "i i ' " . ; * : ' i 1 ; t i t . ,. t i .. l t * . L !

.. * * ' 8

• For the complete loss of feedwater tran$ient, the main feedlitater put!!ps are aslumed to coastdcwn to zero f1ow in five For the *loss C!f i $ingle pump. one PLlnP is assumed* to caastdcnm

'to zero flow five howevert the remaining pump has C.!pa.eity of of" nominal full power feedwater flow. Therefore,.

the pump {the Salem Plant pumps) will its flow tc 70: fiow. re$pOnse time for the second pump is.20 seeonds. * * . 9 *. Notni nal and protection system setpoints were assumed. TP.AHS!£NT RESULTS. 1. Loss of. a Ma 1n Pump The sequence of events for both a 30 sacond and 300 se-cond delay of isanua1 trip shown in Table 1. The transient prirnar.t pressu:-e i:alcu1 ations are shewn 1 n F'1 gure L The low-lcw*i."team level $etpoint is at 99 seconds; auxi1iary 1s autom.atica11Y-initiated.

Ten aux11iary fesdwater begin$ to be delivere<1 to tht! steam generatcrs4 30 For the where there is only 30 $f!COnd delay, there ire no subsequent reactor trl p si gna1 s There is no large heatta? of the reac:Ulr c01llant becau!.:e the stH:m generator tube bund1e doe$ _ not uncover. Thus there is a1ways adeQuate secondary s:ida beat removai. The peak ef ZZS6 psia which occurs 4t JO seccnd$> 1 s only slightly above. t."le presiure at which the sprays are actuated'. . **

• For this trans1Qnt.

the c:oa1ant system not challenged.

• *

• S f.fi nute De l ay . For the c:ase whel"'e operator act1on 1 s delayed 300 seet>nd$ ( 5 minutes), the reactor int system umperature in,r-:eases,, t't:!&ching the Overpalll'er Oelta-T setpoint.

for turbine runt>ack at 190 seconds. This signal is maintained and thus turbine power c;ontinves to every 30 seconds until the turbine .load is at At this

• point, the sUi-n of the ma1n feedwater flow h-ourone plus tt'ie feedwater flow is equal to the turi)ine steam flO'tl. Therefore.

steam generator level does not cont1nue decraasfag and The operator 1n1tiatad and*turbine tr;p at J99 seconds occurs atur the steam and f eedwater fl ow have :natcned.

The J)eak primar:r system of 2330 ps'f a at 267 seconds bi!fore the steam and feed fl ow are matched. TM s pressure f s below the relief valve setpaint (2350 psia). The pressurizer sprays, ;gmt,ined with the effeet of reduced turbine load preYent. 51 gni fie ant overpressuri

a ti on.

rea,tor coo 1 ant pr-es sure stays below $ervi Level C 1 i;;f t$ of 3200 psi a. . -s 3999Q * * ' *-*. *--------=-=--=-===-=-'=*--=-**o.______:"---'-':=......:-=**=*

-=**-"-"'*-==-*..::.:**

• o_*

Z. Loss of A11 Main Feedwater The sequenca cf events foia this.

are presented tn iab le* 2. The tran!1ent pressure ca1cu1ations are depicted 1n

2. The low-iow steam generator setpoint is reac:heD at 3.3 10 seconds *iater. aux1liury feedwater is to the steem generators. . -* JO 'Seeori'd .Del av An turbine runba.cl *due to an Overpewer Oelta-T is initiated at 43 seconds and turbine. 1oad f s reduced ihe Open and maintain pressure at*the setpoiAt value {23501 unti 1 tne oper-ator tr1 the plant 11t 63 seconds. Steam dump f s initiated ana reduces tt>e tem;=:eratvre.

to tn.? no load value of 547*F. thii transient the system is well below 3200 psia. 5 Minute As in the previous case.* the heatti-p of the l)rimar-1 hnt c turt>i nH. fa tefj by an Overpawer De 1 t.a--T si -gna L The turbine 1cad is reduced twice in until the lead is of M=i nA l load. Steam pressure to drop due tn ooi 1 off of .. water in the n9 a low ste= pressure . al.am. A.t tM s time primary pressu'"i!

iUrts to there is on insurge into the bo"th high

• 1eYel and pressure*trfp alarms to be actuated.

The stsam genel'"atgr tube bundle begins uncover, causi"9 a larger of increase in primary and ihe.

fill$

tne peak pressur'1:!

is 3491 ps1a. Huclear power has dtM:reased tn1s po1nt to a.bout 30'i of nominal due to feedback *. As the of **ter t:trough the relief ind safety valves the pr;mary pressure starts to and th-e and relief va1 aoout3&

afteia 'dle time of peak The opes-ator trips the reactor at 333 seconds. COHO.US IOMS The here that for 1oss of =a1n puap, at si.,; ma.jar i!lan1s 1n to generated to alert to the fact that occun!d..

Furthermora.

@ven for eYent witn a five dela,,v 1 n "trip automatit; turbine MJnbac.k reduces f1ow to mtch the capabfHtjr of the auxiliary F'os-this tta threat of in that the relief yalve

• setpoint*

ts *not !ven reached. 6 39990 '* -----,.. \ \ ; * ! * ' ' l t r t L ! f I l i i l . {

.. Como1 ete of Main For comp1ete 1oss of operator action the: action time taken at the pl ant on the Februaey 25, 1983 ts sufficient to prevent ove1'1)ressuri:ation of the reactor system. Peak primary system pressure results only tn relief valve actuation without the actuation*

of pressurizer sa.fJty valves. FurtheJ"m1)re.

there 3 major .a-1an:s$

which are actuated fn ti>* the steam generated 1ow-1aw level a1ann to alert the operatDr tD take action. -. * , As discussed earlier. 1t. ts a reduction tn

-heat t:rans capabi 11 ty wM ch cau&es the primary system h:-.atup and pres sure i nere ase.. A 'f;urbi ne trip reduces the arnount of 11 ow and the rate at which the leve-1' in the steam generator drops. 1f turb1ne is before is i significan't las' of stem 911nerator the tubes will not uncover and the pnmary syste= wi11 net overpr'!ssuri ze. Based upon the rt!Ults discus-sad in the prtYious --s*ct1on. operator action to trip tha turbine at or-befar-e one to one and a half minuus following the 1ow.low level trip and a1anA .ould prevent overpressuri

ut1 on of the reictor coo hnt syst&D btro-ttd 3200 psi a. lt be noted that the core nucle1r (a reac'tiv1 ty coeffic'f eirt of -8 rx:m/*r) used are not of the actual design for-the Salem P1ant. Previous AntS nave shown the peak pressure ta be stron9 func:tion of the and *there & 100 psi ever:t l pt;:t fn the t:1)eff1ehnt.

The Sa 1 em core is desi tt> opar-ate 5uc:h by time the plant reached full 1t would have a coefficient of -10.s or 2.S less *tjnm th@ c:ceff1c:1ent fn the study .. This ..

would be reduced even further Z pcs1*F per rionth o1 oparation 3)._ ibe 10.5 in a peat pressure for the l"lm1t1 ng of f1 ve minute Ol)t!l"_a1:0r .action of 3241 ?Si a (a 250 psi a reduc:ti on -from 3491 psi a} 1 s 'ff1 thin the .

• caiculat1ona1 band of the ASJ.£ Stress LeYe1 C 11m1t.

ttie case 1 n Figure 2 would not exceed thtt ai teri a ..

• Su=nary eonc:1u$1on,

$tudy has desi:ionstrated the ability of the Silem jUelear p1ant to withstand effects of postulated

=al functions without reactor triJ) at full ;>a.el"' w1 an long* delay for operator action. The re!ults snow reS?onse

• wbicn is within uncertainties Qf the AS1'£ Stre-ss Level C Hta1t-s. These r11sults*are further &ffect*d by the low 9robabllity of these

()j;Cur1 ng 1t full power in addition to the 1nc:nastngly beneficial nuclear characteristics of tJte plant over eore Hfe. 7 .

... i f l 1 I ' l \ t t.-t :: ' "" "i: " j .. i ! .. I i l ; ..

• v i i '1! 1' *f t :. i* I .. t "' ? .. ,.. l 'I I I ., l .. * ,.. .. i 1 s. .. i i ' i t ! t i i . * ' ._ i ! : -. ' * . ! I I .... .* --........ ___ *_* -_--_-_. _*

.. _. -_. -'---'

"""'r."'-'--'-'.

-"-::--::___,.

__ ___:__,__;*

_

-:;-* * ....

..

• TABLE 1 Se_quence of Events Less of One FeedWater-Pump Event -Loss of one

{4lan"D) pi.mip nme1 -0 de 1 i Vf!rs max 1 mum f 1 20 Low-low SG level setpoint {alarm);

feedwater signa1 (a1arm) 99 Auxiliar,y feedwater begins 109 OperatQr tr1ps reacior and turbine i29 O? AT setpcint turbine load reduced *-iurt:ine load reduced S: --OP T" trip setpoi nt .(al arm) ---Turbine 1oad reduced --Pe.ax Pressure ();i:urs .. --Turtli 7oad reduced 5% ---Turbine 1oad reduced High pressurizer level setpoint {alarm} Operator trips N!actor and turbine (1} (2) 30 de 1 tI.'I beftire manua 1 tr1 D . . . 30.0 secand delay before manual trip a 3999Q ..... 3 alarms

• prior-to tr1p TiQ!2 -JJ 20 99 109 ---190 233 zzo zso .. Z67 psia} 2SO 310 31-1 399 6 pr'iar to trip ,.. E I* I* l I re i. i };: "' t t* i i ., I ! * .. :I l . i l i l!; i :; t i f * "' i 1 .. . ; i l j ! **, .

• __ ..

.... *--* . *

.* --:.** __ .. __ *::a---. *--*-

• .... TABl.E 2 S!quence of Events Complete l.oss of lifain Fe*edwater Event Loss of main feedwate,.

pumps (alann} low-1ow*SG 1eve1 setpafnt (alar:=);

auxiliary .iignal generation

°" b T runDacx se tpc;i nt (al arm) turbine load reduced 51 T trip setpoint (alaJ"'ll)

Auxiliary fe!'dwater begins Pressurizer reH!!t' valves. open 0 33 34 43 43 55 * *O?eJ"atoi-trips. reactor/turbine oJ Tur'Ji ne *load reduct!d 51 --Hf gh pressurizer 1eve1 trip setpaint (a1ania} Lew staam pn!ssure SI {aianD) High pressu,.izer pressure setpaint (alarm) *** SG tubes 0egi n to uncover;

• new drap$ pressurizer safet¥ va 1 ves open Pressurizer fi11 s Peak pre-ssurt Pressurizer safety.valves close Pressuri4er relief valves close Low RC f1ow (alarm) -Operator tr1 ps reactor/turbine (1) 30 second delay before raanua1 trip C 2} 300 se-:cnd de 1 ay bef o-re manual tr. p g *-*----* --4 alarms prior-to trip .. l3 43 55 54 .as .SS 88 92. .95 ll4 (3491 psia) l4t . 155* Ji! 3:33 1 alarms prior to trl-p

-*'

• I * .5 -"'"" a ! i c -10 QI -... -0 '-1:1 -15 := -'° l-a: ;a; :;,I --20 c -19 !D "'= i *25 .. :ao . 0 FIGURE.

TEM?ERATUR£ COEfF1CI2iT CURING CYCL! S --AT HP., ARO, EOUILISRIUM XENON CONO!i!ONS

'It ii i 4 6 s Cyc1e Surnuc

J_ -10 1 lZ * ..,.CAP 1024,, "The N1J<

lear-o'f Unit One Power Pl.int Cych: S" . *--... . -*-* ---.. . '.'""'*:"'--_

.. :--u

" 6 * * -.. -.... ... -JZOO FIGURE Z

.. t.css of A.11 zoa = rr ..

--:----*-... . .*. --* . -

.... :-*.:-.

... t

-.. . .: :ne

!"-

• • --.. *I s. *

• I f ! ., ,. .....

e =.:' .:: r': *--",.. 1 *** I -.. . *""* .. .... fl" -':. fr 'i: *---f ... :; . ---* t ........ *I -.. .. i ;

.. <<; .. It< :.. " .... *f -,::i...w ..... *-*-=-= .. . ... --et#V v j .,.

t £ 1:: : t 11! t j s ! l .,. ' i l i " . 300

e i a:1 !e ;Cr'if . !'9. c :; i .t i t i .,. r ., .. i-.. ___,. f -..__ -' ;i ;;:-c \ i ! -\ j i ! ' * :: £ l

• t .. t ' .. .::.:. ' I .:<. '. ,. >> -I .. ;;, . :-:: ; .v ... -":!le HC* * **.**:*--

• In the Matter of PUBLIC SERVICE ELECTRIC AND GAS COMPANY (Salem Nuclear Generating Station, Units 1 & .2) UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION ) ) ) ) ) ) ) Docket Nos. 50-272 and 50-311 License Nos. *DPR-70 and DRP-75 ORDER MODIFYING LICENSE EFFECTIVE IMMEDIATELY I. Public Service Electric and Gas Company (the licensee) holds License Nos. DPR-70 and DPR-75 which authorize operation of Units 1 and 2 of the Salem Nuclear Gen-erating Station. The facilities are Westinghouse pressurized water reactors (PWRs) located at the licensee's site at Hancock's Bridge, Salem County, New Jersey. II. On February 25, 1983, an event occurred at Unit 1 of the Salem Nuclear ing Station when the control rods failed to insert when the reactor-trip circuit *breakers failed to _automatically open following receipt of a valid trip signal from the Reactor Protection System (RPS). The manual trip system was used to shut down the reactor. Subsequently, it was concluded by the licensee that the failure to trip was caused by a malfunction of the undervoltage (UV) trip attachments in both reactor-trip circuit breakers.

Evaluation the event of February 25, 1983 revealed that a similar failure had occurred on February 22,

• I

• 1983, at Salem 1. There had also been a previous event Salem 2 involving a failure of one reactor trip circuit breaker to trip on January 6, 1983. The malfunction of the unde_rvoltage device on February 25 was determined by the licensee to have resulted from lack of proper lubrication on the latch of the UV trip attachment.

It appears that no preventative maintenance had been ducted on the Salem 1 DB-50 circuit breakers until January 1983. Additionally, the recommendati-0ns of a Westinghouse 1974 Technical Bulletin

• cetter (NSD DATA LETTER 74-2) were not implemented during the January maintenance since the personnel who performed the maintenance were not aware of the bulletin commendations.

The specific details of the event and the licensee 1 s response are contained in the staff 1 s Safety Evaluation Report, which is hereby porated by reference.

The NRC review of the event revealed a number of significant deficiencies relating to management supervision and control of the procedures governing the classification of the reactor trip breakers as safety-related, management vision of maintenance techniques, and management attention to the safety cations of system malfunctions.

This review is set in NUREG-0977, 11 NRC Fact-Finding Task Force Report on the ATWS Events at Salem Nuclear Generating Station. Unit 1, on February 22 and 25, 1983, 11 March 1983, which is hereby incorporated by reference.

If there were a potentially severe transient, from a worst case set of initial conditions, and the reactor shutdown system did not function, an extremely severe accident could occur in the absence of timely operator action. Therefore, *.

• the aforementioned failures are technically very significant.

When coupled with the cause of the challenge to the reactor protection system, i.e., a feedwater system transient, and the frequency of past feedwater system transients 1 at the Salem facilities, the event raises serious safety questions regarding the continued safe operation of the Salem facility . . III. Analysis of the event described in Section II reveals operational and manage-ment issues at the Salem facilities which are discussed in the staff's Safety Evaluation Report. By letters dated March 14, April 4, April 7, and April 8, 1983, the licensee submitted its Corrective Action Program related to the reactor trip breaker failures.

This program includes short-term remedial actions to be completed prior to startup of the unit and a number of longer term items which have not been completed but for which the licensee has established completion dates. These items are summarized in the attached Table, 11 PSE&G Corrective Action Report, Short and Long Term Items.11 (Attachment

1) The various remedial actions involve equipment, operational and management issues. The equipment issues involve (1) safety classification of breakers, 1 0f primary concern the NRC is the ATWS (anticipated transient scram) event initiated by a loss of feedwater transient.

In 1981 and 1982, Salem 1 experienced about 11 and 5 feedwater transients, respectively, while Salem 2 experienced about 14 in 1981 and about 11 in 19B2 .. This results in average of about 10 transients per unit year of operation for each unit.

• * (2) identification of cause of failure, (3) verification and (4) tenance and surveillance procedures.

The operational i.ssues involve (1) ing procedure for reactor trips and anticipated transients without scram (ATWS), (2) operator training, and (3) operator response.

The management issues involve (1) master equipment 1 i st, (2) procedures, (3) work-order procedures, (4) post-trip review, (5) timeliness of event notification, (6) updating supplied information, (7) involvement of QA personnel with other station ments, (8) post-maintenance operability testing and (9) overall management capability and performance.

The NRC staff has reviewed the corrective action program as discussed in the staff's Safety Evaluation Report, and determined that the licensee's actions, as modified below, are necessary to assure tinued safe operation of the facility.

In addition, I have determined that because of the history of the Salem facility described above, the need for further long-term actions to increase the reliability of the mitigation features of the Salem facility must also be addressed.

Accordingly, I have determined that the public health, safety, and interest require that the actions set forth below be required by an immediately effective order. IV. In view of the foregoing, pursuant to Sections 103 and 161(i) of .the Atomic Energy Act of 1954, as amended, and the Commission's regulations in 10 CFR -Parts 2 and 50, IT IS HEREBY ORDERED EFFECTIVE IMMEDIATELY THAT:

1. The licensee shall implement and/or rr.aintain the items specified in Attachment 1 to this Order, as more fully described and in the manner described in the licensee 1 s submittals dated March 14, April 4, and April 7, and 8, 1983, no later than the dates specified in Attachment 1, with the following modifications: (a) The Nuclear Oversight Committee shall, in addition ta the actions described in the licensee 1 s.April 7 and 8, 1983 submittals, provide copies of its reports to the Executive-Vice President and to the NRC Regional Administrator at the same time they are submitted to the Vice President-Nuclear. (b) Within 60 days of receipt of the appraisal performed by ment Analysis Company (MAC), the licensee shall provide t.o the Director of the Office of Nuclear Reactor Regulation, an analysis of each recommendation made _by MAC, the action to be taken in response to each, and a schedule for implementation. (c) Prior to return to power, engage Beta Corporation, an outside consultant, to review the PSE&G investigation of the events of February 22 and 25, 1983, including the corrective action program. The consultant will make*an independent evaluation of the action plan and advise on the adequacy of the program to insure that Salem Unit 1 can be safely returned to power.

• *

• 2.

---

*------Within 60 days of the date of this Order, the licensee shall submit to the Director, Office of Nuclear Reactor Regulation a detailed report describing either how and on what schedule the following actions will be accomplished, or why any of these actions should not be taken. (a) Implementing at the Salem facility (Units 1 and 2) the following feature of the proposal by the Industry Group on ATWS submitted on April 23, 1982, on Docket PRM-50-29 (page 10 of Appendix C): provision of automatic initiation of turbine trip and auxiliary feedwater independent of the reactor protection system; (b) Providing at the Salem facility diversity in activating (tripping) the reactor scram breakers, for example, by incorporating the breaker shunt trip function into the automatic trip circuits of the reactor protection system; Any schedule submitted shall be subject to approval by the Director and shall be implemented following such approval.

The Director may modify the approved schedule in writing for good cause. v. The Licensee may request a hearing on this Order. Any request fo_r a hearing shall be submitted to the Director, Office of Nuclear Reactor Regulation, U.S. Nuclear Regulatory Commission, Washington, D."C. 20555. A copy of the request-or answer shall also be sent to the Executive Legal Director at the

• same address. ANY REQUEST FOR A HEARING SHALL NOT STAY THt IMMEDIATE EFFECTIVENESS OF THIS ORDER. If a hearing is held on this Order, the Commission will issue an order designat-ing the*time and place of hearing. If a hearing is held, the issue to be con-sidered at such a hearing shall be whether this Order should be sustained.

FOR THE NUCLEAR EGULATORY COMMISSION:

Harold R. Denton, Director Office of Nuclear Reactor Regulation Dated at Bethesda, Maryland this day of April 1983 .

ITEM ATTACHMENT 1 PSE&G CORRECTIVE ACTION REPORT SHORT AND LONG TERM ITEMS .. A. EQUIPMENT ISSUES A.l Determination of Safety Classification of Breakers A.2 Identification of Cause of Failure a. Short-Term Actions 1. . Confirm that new U/V trip attachments on Salem Units 1 & 2 incorporate all design changes made to these devices. 2. Measure and confirm the force required.

to trip the breakers using the breaker trip bar and that the breakers trip with an output force from the U/V trip lever of <31 ounces. b. Long-Term Actions 1. Submit a test program to determine the. life cycle & replacement interval for the UTAs & to verify the adequacy of the new* maintenance

& surveillance programs used COMPLETION Comp Tete* Complete Complete on the reactor trip circuit breakers May 1983 2. Establish a procedure for periodically Complete measuring the force required to trip the breakers.

A.3 Verification Testing Program a. Short-Term Actions 1. Manufacturer will electrically test U/V trip attachment on Test CB 25 times. 2. After installation, U/V trip attachment tested 10 times. Complete Complete ITEM 3. 4. After installation in appropriate breaker compartment, Response Time Test. Test for independent operation of UV and shunt within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of restart* from each shutdown.

.. b. Long-Term Actions

• 1. Provide detailed test program A.4 Maintenance and Surveillance Procedures

a. Short-Term Actions* 1. 2. 3. 4. 5. 6. 7. Resolve breaker cabinet/switchgear room cleaning deficiency.

Revise maintenance procedure and/or other documents to require all replacement U/V attachments to have been successfully tested 25 consecutive cycles. Provide acceptance criteria in Maintenance.Procedure M3Q-2 for ten cycle test that allows NO failures for acceptance.

There will"ti'e a 30-minute interval between each test. Modify Maintenance .procedure M3Q-2, Section 9.8 to include three timing tests and an average time computed for comparison to previous tests.* Revise Maintenance Procedure M3Q-2, Section 9.7 & other appropriate procedures

• to require that a sealant be applied to the head of the locking screw on the U/V attachment.

Specify in Maintenance Procedure M3Q-2, Enclosure 7 (Ref. Section 9.7) U/V coil dropout voltage acceptance tolerance

& actions to be taken if out of specification.*

Lubricate the circuit breakers and UV trip attachments in accordance with W Technical Manual.* *See attached Table 1. COMPLETION Complete Complete May, 1983 Complete Complete Complete Complete Complete Complete Complete

• ITEM .. 8. Perform a trip force measurement on the trip bar of each breaker.*

9. Modify maintenance procedure to test bypass breakers every 6 months.* 10. Provide functional test procedures for UV trip, shunt trip, and manual scram.* 11. Perform monthly reactor trip breaker timing test.* b. Long-Term Actions 1. Propose Technical Specifications B. OPERATOR PROCEDURES, TRAINING, AND RESPONSE B.l Operating Procedures for ATWS a. . Short-Term Actions 1. Identify indications in control room providing positive of reactor trip demand. 2. Review the basis for ATWS procedure steps and order of priority, revise procedures, as necessary, and train operators.

b. Long-Term Actions 1. Provide detail description of procedures to ens*ure operability of SSPS status

• COMPLETION Complete Complete Complete*

Complete Within 30 days of Unit S/U Complete Complete panel indicators.

Complete 2. Provide schedule for the upgrade program for Emergency Operating Procedures

• April 15, 1983 8.2 Operator Training a. Short-Term Action 1. Conduct training for operators on revised procedures . 2. Conduct practical exercise in Control Room of revised procedures.

• See attached Table 1. Complete April 12, 1983

• *

• ITEM 3. 4. 5. 6. ------Conduct walkthrough on alarms & RPS indicators.

.. Conduct training for auxilary operators Evaluate trainees*

performance against established objectives Review training material and ensure it is current and properly referenced B.3 Operator Response a. Short-Term Action 1. Caution operators in use of J handle control. b. Long-Term Actions 1. 2. Replace Reactor Trip Switch Modifications to clarify First Out Annunicator Alarms C. MANAGEMENT CAPABILITY AND PERFORMANCE C.l Master Equipment List (MEL) a. Short-Term Actions b. 1. 2. Verify MEL is complete & accurate.with respect to ECCS, including actuation systems, RPS, Aux. Feedwater_and containment isolation systems. Instruct appropriate personnel in purpose & use of MEL. -

Long-Term Actions 1. Verify completeness

& accuracy of MEL for remaining Q list systems -and issue as a controlled document . COMPLETION April 12, 1983 April 12, 1983 Complete Complete Complete Next outage of sufficient duration May 1, 1983 Complete Complete May 1983

• ITEM .. C.2 Procurement Procedures

a. Short-Term Actions 1. PSE&G Sampling review of past ment documents

b. Long-Term Actions 1. Evaluate & modify procurement procedures to ensure appropriate classification of items/services important to safety. C.3 Work Order Procedures

a. Short-Term Actions 1. 2. 3. QA Department review all related work orders prior to starting work. Implement a program & training to ensure that work orders are properly classified.

Review work orders written since issuance of the MEL for proper classification

& evaluate safety consequences of those found improperly classified.

C.4 Post-Trip Review a. Short-Term Actions 1. Develop and implement AD-16 C.5 Timeliness of Event Notification

a. Short-Term Actions 1. 2. Assign dedicated communicator to each shift. Review importance of reporting ments with supervisors COMPLETION Complete Ju1y* 19s3 Complete Complete Complete Complete Complete Complete

• ITEM C.6 Updating Vendor-Supplied Information

a. Short-Term Actions 1. Update existing documentation on safety equipment and ensure that vendor mentation is under a controlled system. 2. a) Audit Station files for manuals existence, revision level, and date. b) Audit Nuclear Engineering files for manuals existence, revision level, and date. c) Compare Station & Nuclear ing; Audit and use lastest manual revision d) -Contact vendors to confirm that manuals are technically current Request updated copies identified as more recent) Review Westinghouse Technical Bulletins and Data Letters b. Long-Term Actions a) Audit Station files for manuals existence, revision level, & date b) Audit Nuclear Engineering files for manuals existence, revision level, and date. c) Compare Station & Nuclear ing; and use lastest manual revision d) -Contact vendors to confirm that manuals are technically current Request updated copies (*where identified as more recent) e) Revise Station procedures where appropriate COMPLETION Complete Complete Complete Complete Complete June 1983 July 1983 Aug. 1983 Dec. 1983 July 1983

• • ITEM

• f) Index & control new/revised manuals received g) Develop procedures for controlling vendor manuals C.7 Involvement of Quality Assurance Personnel With Other Departments

a. Short-Term Actions 1. 2. Retain outside consultant to assess QA program Modify QA organization policy to more fully integrate with.overall nuclear activities

b.

Actions 1. Provide additional detail training on processing work orders to emphasize QA test/retest requirements C.8 Post Maintenance Operability Testing a. Long-Term Actions 1. 2. 3. 4. Review and revise AP-9 and other station procedures Complete review of vendor and Engineering recommendations and incorporate necessary changes into departmental documents Incorporate items identified into Inspection

_Order System Complete Managed Maintenance Program C.9 Overall Management Capability

& Performance

1. 2. . Complete staffing of Nuclear Assurance

& Regulation Department Independent assessment of QA Operations COMPLETION May 1983 May 1983 Complete Complete September 1983 July 1983 Janua_ry 1984 Aug. 1983 Jan. 1984 Jan. 1984 July 1, 1983 ITEM 3. Implement training program for first-level supervisors

4. Develop training program for senior supervisory level 5. Develop program for periodic or regular training for supervisory managment personnel

6. Develop Technical Training Program for non-Station personnel

7. MAC management diagnostic

-final report COMPLETION Sept. 1983 Oct. 1983 Spring 1984 -spri hg 1.984 May 30, 1983 ..