ML20214L564

Responds to 840320 Questions Re 840319 Generator Hydrogen Explosion & Bearing Event Might Have on Facility Pending Requests to Continue Operation W/One Fewer Reactor Operator on Shift.Fire Brigade Responded as Required

ML20214L564
Person / Time
Site:	Rancho Seco
Issue date:	05/03/1984
From:	Bishop T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V)
To:	Thompson H Office of Nuclear Reactor Regulation
References
TAC-52561, TAC-54487, NUDOCS 8705300081
Download: ML20214L564 (5)

Text

,

- =

i . -

gfy

1. wg lf . , y h,,;

. umnoswas M%-4 c;

NUCLEAR REGULATORY COMMISSION asoum v 9tg%

1. # , 1450 MARIA LANE, SUITE 210 1NALNUT CREEK CALIFOMNIA 94996

1. a4 .-4 r)

, . ,oD '

MAY 3 1984 MEMORANDUM FOR: Hugh L. Thompson, Jr. Director .

Diiision of Human Factors Safety, NRR -

FROM: T. W. Bishop, Director Division of Reactor Safety & Projects, RV

SUBJECT:

-RANCHO SECO - ADEQUACY OF STAFFING Your memorandum to Mr. John B. Martin dated March 20, 1984 addressed the March 19 generator hydrogen explosion and subsequent events and questioned what bearing, if any, these might have on SMUD's pending request to continue operation with one fewer Ro on shift than required by current regulations.

Answers to your questions are as follows, numbered as presented in your memorandum:

1. The precursors which led to the hydrogen leakage and ignition stemmed from loss of the 2E1 480 volt bus as a result of a ground fault in a gland exhaust fan. The hydrogen side seal oil pump was lost (powered from 2EI) as a result. Although seal oil system design is such that continued operation should be permissible in this condition, with the air side seal oil pumps maintaining the hydrogen shaft seal, pressure in the air side seal oil system was subsequently lost for a brief period (seconds or minutes), allowing hydrogen to escape into the generator / exciter housing.

a. The operators became aware of the loss of the hydrogen side seal oil pump while following the procedure for the loss of the 2El bus. Alarms associated with the hydrogen seal system did not operate properly, but non-licensed operators were aware of and dealing with difficulties in maintaining oil levels in the hydrogen side seal oil receiving tank and drain regulator. The cause of the loss of oil pressure in the outboard (air side) portion of the shaft seal has not been clearly determined.

The operating staff was not aware of actual hydrogen leakage until a small explosion was reported, followed almost immediately by the larger explosion and fire.

b. As mentioned in a., the staff did not know about hydrogen leakage until the explosions and fire. Principal attention was then given to tripping the plant and dealing with the fire. A non-licensed operator was shortly thereafter dispatched to blow down remaining hydrogen in the generator (still at 64 psig, compared with 73 psig prior to the explosion) through an elevated release point above the plant.

0705300081 DR 840503 ADOCK 05000312 PDR c . . .

2 MAY 3 1984

c. It is our conclusion that there was sufficient staff to deal with

a. and b. and the other events which occurred. However, it is noted that additional personnel were on shift at the time of the incident. The' Plant Superintendent.(not licensed) was on site as part of the backshift management _surveillauce program.* In addition i to the assigned STA, an STA with an RO license was on jhift for training for 'an SRO license. Also, an additional RO candidate (trainee) was on shift for training purposes. It is felt that 1

- _ , . m.- ~ avant analysis, supplemental actions e and comunanications with 4

outside agencies could have been slower had these persons not been available. However, we do not feel their absence would have 4

sigr..ficantly affected actions by licensed personnel in ensuring safe operation of the plant.

2. With regard to the actuation of the CO system: i 2

s. It was automatically initiated.

b. The fire brigade responded as required by APS20, with the Senior Control Room Operator (SRO license) in charge. Two hoses were laid out on the turbine deck. Because of the possibility of further hydrogen explosions and because the installed CO, system appeared 1

to be dealing with the fire, the fire team kept Its distance.

Their principal concern was to take over if the fire was not-extinguished by the installed sytem or if an oil fire should result. t

3. With regard to partial or complete loss of NNI:

a. Control room personnel did immediately recognize the loss of NNI.

This was recognized by annunciators and by abnormal indications.

b. The operators believed they had lost all NNI, and followed the appropriate procedure. Subsequent investigation showed that only the "X" NNI had been lost, but following the procedure for loss of all NNI appears to have been a conservative action. This procedure was terminated after the seventh step of " Operator Response", since-NNI had been regained.

c. Supervisory personnel were not required to manipulate controls during the event. The shif t supervisor was the F.nergency Coordinator. The Plant Superintendent was on site at the time of i the occurrence as part of the backshift management surveillance program established by the licensee. He looked out on the turbine deck, after the first small explosion was reported, and informed the control room of the fire. He is not, and has not previously been licensed on Rancho Secc, and acted only in an advisory role. .

Although he could have assumed the role of Emergency Coordinator (since an Unusual Event had beca declared), this function was also  ;

allowed to remain with the Shif t Supervisor. The Manager of Neelear Operations olau arrived at the site while the event was in progress. He consulted with those dealing with the event, but the Shift Supervisor remained in control.

n , -,-- r--rr ,. r r, ~+n-m _--~~,ew-np,,p.e.,-n -, ,,,,- w r ,_g,,,,-,.--n-,,,-, ,,,a-w,---,ema,-w ,, e n ,v--m-..,w,m--nc-.-,-.w,-,,w y a,-- p~--w ww-e-m w.

3 MAY 3 1984

4. With regard to simulator training:

a. Operators interviewed foll> wing the event stated that the events experienced on Macch 19 are not modeled on the Lynchburg simulator. They stated that they had been covered in licensed training at Rancho Seco, huvever, particularly actiona- related to loss of NNI.

b. In interviews with the licensed personnel who were on shift at the time of the event, they stated that they felt prepared to cope with the events which occurred on March 19. They also stated that they did not at any time have concern or questions about the actual status of the plant - that they recognized the basic problems confronting them and were taking actions to resolve them. Our investigation of the events does not indicate otherwise.

5. With regard to the STA:

a. The STA was in the control room when the event occurred.

b. The STA (licensed RO) assisted in the analysis of the event, and in advising the Shift Supervisor. He did not, however, become directly involved in the manipulation of controls. He also made required Emergency Plan notifications to state and local officials. A second STA (licensed RO) was also on shift working on his SRO license. He was not significantly involved in the event other than communicating with the NRC on the ENS.

6. With regard to Staffing Levels:

a. We feel that the number of licensed operators available on shift were sufficient to deal with this event. Although other personnel were on shift at the time of the event, the principal control room actions taken to deal with the events and maintain the plant in a controlled condition were taken by the regular shif t personnel.

As noted above there are usually two extra persons in training available on each shaft. The principal exceptions noted to this policy have been during classroom training periods on day shift, and when extensive simulator training is being conducted at the B&W Simulator in Lynchburg, Virginia. The presence of these trainees on shift could help A urten response times, and aid in the evaluation of any event that could occur. An stated in paragraph 1.c.,

however, we do not feel the absence of these additional persons on shift would significantly affect safe operation of the plant.

b. As stated in a., it appears that the normal shift complement could have dealt properly with this event.

l

4 MAY 3 1984

7. Our investigation has not identified any problems with regard to the operators' handling of the events which occurred, particularly those actions taken by licensed operators. The licensee's investigation noted that when manually initiating SFAS as required by the LOSS OF NNI procedure, only trains 1A and IB were actuated. Channels 2A and 2B, which initiate LPI and auxiliary feedwater, were not initialed. (The casualty procedure was being read aloud as it was perfonsed, but the step which initiates 2A/2B was either not read completely or not' heard.)

However, a prevf ous step called for tripping of the main feed pumps, and subsequent operation of the auxfliary feedwater systeu had been verified.

Low Pressure Injection also would not have injected under the existing plant conditions. This one oversight does not appear to have presented a problem, and it is possible that the operators would have subsequently discovered it if the procedure had not been terminated shortly thereafter j because of regaining NNI.

1 Our overall conclusion from evaluation of the events and the operators' actions in dealing with them is that shift personnel performed in a professional manner. We see no factors from our investigation of this event which would reflect adversely on granting the requested exemption.

l . N.

T. W. Bishop, Director Division of Reactor Safety and Projects, Region V

^

i h

e

.E- '%

.' ) ,

L 2

F

>* N to e

as W u-a 59 P"

4 .e '.

^

. tsp .

WJ@  %,1 c o., p 1

g s

_". n Y W' 'A' F4

• iA 4

l i

1 l

l l

b 8

9

• D ( 3

s NnC room aos U.S. NUCLE AR HEGUL ATOR Y COMMISSION REQUESTER'S MAILIJG ADDRESS

s C #. n.

, = <57 IYSO MMJA LAAJE fur 12 #so CeTv '

FACSIMILE TRANSMITTAL ISTATE REQUEST ggg OA.E RETUH o g

,ogN AL TO 2,353 5 7/g} w E- n ~- -_

v MESSAGE TO N AME ANO ONOAN6F ATION P ACSiMILE PMory E N UM BE R VERIP4E ATION PMdass MWuse#

did CtTT 171<Au.A /

n R rA 4

~3 ST ATE 492- 8IIO AUTOM ATIC 492- 731/

M uman OP P AGE. ( hw'ans I v.u. ...vv 6i=.. awe,ea w a.:

ht* 0 11d _$Q, _ _

MESSAGE FROM

_I MAm 2 [TELEPNCheNO. FACstMILE PHONE NUMBER V ERIP8C ATION T EL.EPHONE M UM BE R 4(#3 Mics.sPEEo( ... -.=i n.o w s P E L u i ,. . . . )

, rL' bNS Vbr2#

• 3$DY Yb 5' SWO$ Vb 5 *" .5 V J6kDeNG AUTOM ATIC lM AIL sTOP AUTOM ATIC ', , a g ,E. v .E.

Asszo.O ~ .. '

V .A'Aftps/x N. Am PRECEDENCE l ovenmacNT l l Pour Houns l lTwoMounY l loNEHoum l l iMM aos Att ss scia L IN sTRUCTIONS I

l l

I TIMt/DAT E (Stamo)

RaCEevEO Yn AMsMaiTED l

_ - - _ _ .