Text

Ro:On 890915 & 18,reactor Trips Occurred When Safety Channel 1 High Voltage Power Supply Dropped to 90% or Less of Rated Voltage.Caused by Instabilities in 15-volt Dc Power Supply. Trip Relays Replaced & Bistable Adjusted
	ML20248F108
Person / Time
Site:	05000083
Issue date:	09/29/1989
From:	Vernetson W; FLORIDA, UNIV. OF, GAINESVILLE, FL
To:	Ebneter S; NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
Shared Package
ML20248F110	List: ML20248F108; ML20248F111;
References
	NUDOCS 8910060218
	Download: ML20248F108 (5)
	v • d • e

] a 4 e.: . NUCl. EAR ENGINEERING SCIENCES DEPARI' MENT Nuclear Reactor Facility University of Florido I

- us.vemee on,m.cnn .

cucu. mero. .uun,.. .

Ga.nevels,Ho,kao 32611 Phone (900 392-1429.Toten 64330 September 29, 1989 Final 14 Day Report:

Safety Channel 1 High Voltage Trip Nuclear Regulatory Commiss'on Region II 101 Marietta Street, N.W.

Suite 2900-Atlanta, Georgia, 30323 Attention: Stewart D. Ebneter Regional Administrator Re : University of Florida Training Reactor Facility License: R-56, Docket No. 50-83 EVENT SCE'01R10:

On Friday 15 September 1989, the University of Florida Training Reactor (UFTR) was started up at 1435 hours0.0166 days 0.399 hours 0.00237 weeks 5.460175e-4 months ; at 1703 hours0.0197 days 0.473 hours 0.00282 weeks 6.479915e-4 months , after seven (7) minutes of full power operation, with SRO W.G. Vernetson as operator at the controls, P.M.

Whaley as- SRO on-call and the UFTR operating at 100 kW, a reactor full trip occurred (Safety Channel I high voltage power supply dropped to 90% or less of its rated voltage).

The operation in pregress was support.ing neutron activation analysis as well as gamma irradiation of several samples in the shield tank. Following a three minute delay, primary cooling was restored and the high voltage trip signal was l noted to have cleared. At the start of the following working day, on Monday, September 18, 1989, the high voltage trip was again noted to have occurred (over the weekend), indicating that the problem was not related to operating at power and was intermittent. Preliminary investigation revealed the power supply output to be normal at the time of measurement (normal voltage, ne circuit noise) indicating that the problem was intermittent and in the high voltage power supply or in the high voltage trip bistable circuit.

On September 18, 1989, under Maintenance Lag Page 89-51, the Safety Channel 1 voltage trip bistable and the Safety Channel I high voltage power supply were examined, with intermittent instabilities in the power supply documented by a chart recorder. The instabilities were traced to a 15 VDC positive power supply, as the source of the problem.

8910060218 800929 3

{DR ADOCK 0500 %- 2

. i

}

e w %u,m-%,

g .

fi}

< i

(" c .

l.; .i NRC1

. Safety; Channel'1lHV-Trip Page 2

" September 29, 1989s Since this problem followed an extensive maintenance program involving work on Lboth Safety Channel 2 and the control blade drive circuitry, the decision was made to treat this event as a potentially promptly reportable event and make.

notification (although the reactor trip from failure of Safety Channel I high voltage power. supply - positive 15 volt power. supply - was a trip from a known cause, and therefore not necessarily a reportable occurrence per UFTR Technical Specifications Section 6.6.2). This initial notification was carried out to Mr.

Ed McAlpine of Region II with a followup letter on September 18,-1989 (See Attachment I).

Since the event was reported b, ed on a series of problems followed by the SC-1 high voltage power supply trip, a brief summary of recent maintenance history is presented.

7 September 1989 The UFTR tripped from 100 kW operation due to an electrical transient; che trip relays for the primary coolant pump

-(deenergized) and the dilutant fan (deenergized) indi-cated tripped.. Both the primary coolant pump and the dilute fan were' running norme11y through the event.

8 September 1989 The quarterly scram checks (due in September) were performed 4 as part of corrective action prior to restart. Safety Chan- !

nel 2 high voltage trip test circuit was found to be not func-tioning, although removal of high voltage was verified to properly trip the UFTR. The bistable trip point was readjust-ed and the trip circuit tested and shown to perform properly on 12 September 1989.

8 September 1989 During trouble shooting procedures on Safety Channel 2, the auto flux control reference potentiometer was found to be mechanically failed. The potentiometer was replaced with an on-hand spare and the auto flux controller calibration veri-fied on 15 September 1989 with a brief operation at power.

12 September 1989 The auto flux controller calibrat ion check was first attempted on 12 September 1989, but the control blade drive system fail-ed to withdraw the safety control blades. Trouble shooting revealed that the switch logic contacts on the Regulating Blade were operating erratically, causing the multiple blade interlock to engage and inhibit movement of all control blades except the regulating blades It was decided to replace all

. hu 3 !

7

Nb

l 4

' j 1 .- ..

NRC 1 3 Safety. Channel?1 HV-Trip -

Page September 29, 1989 )

control. blade drive. switches with on-hand spares to prevent .

recurrence _of the failure. On 15 September-1989, control 1 blade >3 drive circuit maintenance: was completed with a retest program on potentially affected systems. including measurement of

. control blade drop times and controlled withdrawal and insertion time checks.

18 September 1989 Following the, trip on 15 September, 1989 and under Maintenance

. Log Page 89-51, it was . determined that the positive 15 volt DC power supply.to'the high-voltage power supply was operating erratically. Safety Channel,2 high voltage power supply was monitored overnight to assure'prope'r operation of Safety Channel 2, and the_ Safety Channel 1 high voltage power supply' replaced on 19 September 1989. . During an operational retest of the reactor, the 7extended range light did not extinguish at the required indication, and an unscheduled shutdown was performed per UFTR operating procedures.

Under. Maintenance Log Page 89-52, the' extended range bistable was-found to be activating at about 2-1/2 times higher than

. required and was corrected. Because of the physical location of the extended range bistable with respect to the bistable _ relays for Safety Channel I high voltage power supply. trip, it was thoughtLthat work on the high voltage bistable might have caused the extended range light bistable trip point to change. . Two steps were accomplished to assure that this was a unique occurrence; first, the trip point for the other bistable that is not-routinely checked for the trip point value (source interlock at 2 cps) was checked, with satisfactory results, and second, the extended range bistable trip' point was checked and noted on the Daily Preoperational Checkouts (to be accomplished daily until September 30,-1989). No difference in set point has been noted.

After a successful operation with a power run to check calibration of the auto flux controller, the UFTR was returned to service. Approximately 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months later, the reactor operation supporting neutron activation analysis and a gamma irradiation was terminated by the trip on the Safety Channel I high voltage power supply, the subject of this report. Although this series of failures occurred during a short time span, no evidence of a common root cause is available.

' *- l Sf$

UR

.l .

l l

.NRC' i Safety Channel 1 HV-Trip Page 4 September 29,'1989 {

21 September 1989 Between 1100 and 1300, with.the reactor secured, the coolant pump and the stack dilute scram relays' tripped. 'Although this event could have been an electrical switching transient, no one

, had been present to observe the occurrence and therefore this hypothesis could not be verified.. Because of previous reactor trips, it was decided to replace the field and trip relays for both scram channels. Following retesting, the maintenance log page was left open to provide observation of the reactor protection system behavjor until it was closed out on 25 September 1989 with no recurrence of either trip indicated.

29 September 1989 During preoperational checkout, the wide range drawer was noted to be downscale. Under Maintenance Log Page 89-55, the discriminator circuit of the wide range drawer was found failed.

A f ailed transistor and a resistor were replaced with on-hand spares with the system available on this date for performing checks for return to service.

This scenario through September '18 was communicated to Mr. Ed McAlpine (Region II NRC) on September 18, 1989. prior to having finally resolved the source of the problem later in the day. As indicated above, it was promptly reported primarily because of the extensive maintenance program as indicated ~in the attached one page confirmation of the conversation (See Attachment I). As promised in Attachment I, this event was reviewed by the UFTR Reactor Safety Review Subcommittee at its regular meeting on 19 September 1989. Since September 19, 1989,'several additional events have occurred as outlined above to include the tripped scram relays on 21 September 1989 and the failed discriminate circuit on September 29, 1989. The latter simply involved two electronic components at or near failure which have now been replaced.

EVALUATION / CORRECTIVE ACTION The general evaluation of this event is that the various failures are unrelated l except possibly though the electrical transient that caused the initial trip on September 7 and could have accelerated failure of certain other components, though some failures such as the flux controller are clearly mechanical and unrelated.

Corrective action over the past ten days has involved extensive maintenance with replacement of various trip relays and adjustment of the extended range bistable when the extended range was found to be malfunction during the startup on September 18, 1989. In this case (following procedures) an unscheduled shutdown was performed and -

the bistable aligned, and the extended range and the source interlock bistables verified to be properly functioning. The check of these 2 bistables (not responsible for reactor trip) is being added to the regular quarterly surveillance to check reactor trips.

l

NRC Safety Channel 1 HV-Trip Page 5 September 29, 1989 Based upon the above analysis, this event is concluded to constitute a reactor trip from a known cause. There was no compromise to reactor safety in this event nor was there a potential for abnormal radiation exposure. All reactor safety and control systems responded properly and in a conservative manner. Therefore, the event is not considered promptly reportable; nevertheless, the prompt notification process has been utilized because this event followed performance of extensive maintenance including corrections made to the high voltage trip setting on Safety Channel 2 as part of the Q-1 quarterly scram checks. The subsequent maintenance to replace several trip relays was primarily preventive in nature while the maintenance to repair the discriminator circuit in the wide range drawer is part of normal repairs dueto aging circuitry.

CONSEQUENCES The Reactor Safety Review Subcommittee (RSRS) Executive Committee met on September 19, 1989 and agreed with the trip evaluation and that it was not promptly reportable.

The committee agreed with restart for normal operations upon successful completion of repairs.

Reactor Staff and administration including Director W.G. Vernetson present on September 15, 1989 agreed there was no compromise to reactor safety in this event, nor was there danger of personnel receiving excessive radiation doses. All reactor safety and control systems responded properly for all failures identified in this rcport.

If further information is needed, please advise.

Sincerely, b < C' -

William G. Vernetson Director of Nuclear Facilities WGV/p LD L Notary Public Notary Public. State Of Flo% At Large My Commission Expir" var. 30,1991 cc: P.M. Whaley ""*"'""**"'"'*'"='

Reactor Safety Review Subcommittee Document Control Desk Attachment l

l

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - _ _ _ _ _ _ _ _ _ _ _ _ - _ _ - _ _ -

ML20248F108

Text

Navigation menu

Search