ML17300A853

LER 87-020-00:on 870416,Train B Fuel Bldg Essential Ventilation Actuation Signal Received on Balance of Plant ESFAS Due to Communication Malfunction.Caused by Byte Error. Time Delay Relays in Circuitry installed.W/870515 Ltr

ML17300A853
Person / Time
Site:	Palo Verde
Issue date:	05/15/1987
From:	Bradish T, Haynes J ARIZONA PUBLIC SERVICE CO. (FORMERLY ARIZONA NUCLEAR
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
192-00209-JGH-T, 192-209-JGH-T, LER-87-020, LER-87-20, NUDOCS 8705260269
Download: ML17300A853 (11)

Text

REQULATO NFORMATION DISTRIBUTION 'TEI'1 (R IDB>

ACCESSION NBR: 8705260269 DOC. DATE: 87/05/15 NOTARI ZED: NO DOCKET FAC IL: STN-50-528 Pal o Verde Nuc l eav Stations Uni t 1> Av'i zona Pub) i 05000528 AUTH. Nhl 1E AUTHOR AFFILIATION BRA D I SHI T. R. Av'izona Nucleav'o~ cv Prospect <formerly Arizona Public Serv HAYNES, J. G. Arizona Nucleav Paver ProJect (fovmev lg Arizona Public Serv RECIP. NAME RECIPIENT AFFILIATION

SUBJECT:

LER 87-020-00: on 870416> Tv'a in B f uel b l dg essential ventilation actuation signal received on balance of plan+"

engineev ed safety featuv'es ac+uation sos due to communicati on malfunction. Caused bg byte ev'v ov'. W/870515 ltr.

DISTRIBUTION CODE: IE22D COPIES RECEIVED: LTR ENCL BI2E:

TITLE: 50. 73 Licensee Event Repov t (LER) i Inc i ent Rpt> etc.

NOTES: Standardized plant. M. Davis'!RR: iCg. 05000528 RECIPIENT COPIES REC I P I ENT COP I ES ID CODE/NAME LTTR ENCL ID CODE/NAME LTTR ENCL PD5 LA 1 PD5 PD 1 1 LICITRA E 1 DAVIS'

/

INTERNAL: ACRS MICHELBON 1 ACRS MOELLER 2 2 AEOD/DOA 1 1. AEOD/DSP/ROAB 2 2 AEOD/DBP/TPAB 1 1 DEDRO 1 NRR/DEST/ADE 1 0 NRR/DEST/ADS 0 NRR/DEST/CEB 1 1 NRR/DEST/ELB 1 1 NRR/DEBT/ICSB 1 1 NRR/DEST/MEB 1 1 NRR/DEBT/}'1TB 1 1 NRR/DEST/PSB 1 NRR/DEBT/RBB 1 1 NRR/DEBT/SQB 1 1 NRR/DLPG/HFB 1 1 I"RR/DLPG/GAB 1 1 NRR/DOEA/EAB 1 1 NRR/DREP/EPB 1 1 NRR/DREP/R*B 1 1 I'ERR/DREP/RPB 2 2 8:/PAT'S/ I LRB 1 NRR/PMAB/PTSB 1 1 EQ- - "5 02 1 1 RES DEPY QI 1 1 RQN5 FILE 01 1 EXTERNAL: EG8AL OEV/SK)N NUMEEA r 3 HvMFEA Palo Verde Unit 1 0 5 0 0 0, 5 8 7 020 000 2o" 0 TEXT llf/oo/o Iooco 4 /oOoPod. ooo ~ //Oooo/ P//IC Fono 38EA3/ IITI At approximately 1356 MST on April 16, 1987, Palo Verde Unit 1 was in Mode 1 (POWER OPERATION) at 100 percent power when a Train "B" Fuel Building Essential Ventilation Actuation Signal (FBEVAS)(VG) was received on the Balance of Plant Engineered Safety Features Actuation System (BOP ESFAS)(JE). This resulted in the designed cross trips to Train "A" FBEVAS and Train "A" and "B" of the Control Room Essential Filtration Actuation Signal (CREFAS)(VI) on the BOP ESFAS. These represented automatic actuations of Engineered Safety Features (ESF)(JE).

All safety equipment operated as designed with the exception of essential cooling water pump "A" (BJ,P). The pump was started manually by control room operators (utility-licensed). The actuations were annunciated (ANN) and responded to by utility-licensed operators in the control room. All actuated equipment was restored to normal operating status and the BOP ESFAS was reset at approximately 1457. The duration of the event was approximately 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />.

A review of the Radiation Monitoring System (RI'IS)(IL) alarm (ALM) typer data revealed that the FBEVAS trip signal did not originate from the low range Fuel Building (ND) Vent Exhaust Monitor (RU-145)(IL). A gas grab sample taken in the vicinity of RU-145 verified that there were no unusually high radiation levels present which would have caused the actuation. The Shift Supervisor (utility-licensed) informed the RMS system engineer (contractor non-licensed) that he witnessed the remote indication controller (RIC) for RU-145 cycle from approximately E-7 microcuries per cubic centimeter (cc) to F 2 microcuries /cc and back to E-7 microcuries /cc in approximately 1 second. This occurred 2 or 3 times approximately 4 minutes after the ESF actuation. The HIGH, ALERT and RATE alarm lights also cycled on and off during this evolution. In the event of a valid alarm signal, these lights are designed to lock in.

A spike from lE-7 microcuries /cc to lE-2 microcuries /cc is an acceptable result for the monitor algorithm to process. However, the algorithm does not support a return from 1E-2 microcuries /cc to lE-7 microcuries /cc within 1 second.

Testing conducted under an authorized work document involved simulating a high count rate at the detector (DET) by utilizing the installed check source. These tests verified that all alarms locked in and that it took in excess of 1 minute for the count rate to decay to 1E-7 microcuries /cc.

A review conducted by the Computer'roup of the communications from the micro-processor (CPU) to the RIC determined that the cause of the FBEVAS was a parity error on a byte of data causing the counter to reset and establish a new set of sequencing data for the RIC display. This is the first time this situation has occurred.

As corrective action, a modification is being developed to install time delay relays (RLY) into the circuitry to the BOP ESFAS input to eliminate the possibility of ESF actuations due to 1 or 2 second spikes.

hlAC IMAM 3OEA 19 83i

I NRC Form 3SSA

/943/ U.S. NUCLEAR REOULATORY COMMISSION LICENSEE EVENT REPORT ILERI TEXT CONTINUATION APPROVED OMS NO 3I50 0/04 EXPIRES: 8/31/88 FACILITY NAME Ill DOCKET NUMSER lll LER NUMbER IS) PAOE LSI YEAR SEQUENTIAL rl8 V IS IO N NVM SR NVM SR Palo Verde Unit 1 TEXT /// moro o/roco rl /o0/rwo//. Iroo a5 //oorro/ P//I C Fomr 38/rrA 3/ I I Tl 05000.5287 0 2 0 0 0 3 OF 0 4 A root cause of failure investigation was also conducted to determine the cause of essential cooling water pump "A" not automatically starting. This investigation revealed that a breaker alignment pin on essential cooling water pump "A" start

, breaker 1EPBAS03M (BKR) was binding on the breaker frame. It cannot be determined why the pin was misaligned. The breaker pin was realigned and the breaker was cycled to ensure that it was operating properly. Additionally, the override relay (RLY) that permits manual operation of the pump was serviced and cleaned.

Following corrective maintenance essential cooling water pump "A" was successfully started by the automatic start circuitry by simulating an ESF actuation.

Similar events involving an ESF actuation and essential cooling water pump "A" not automatically starting were reported in Unit 1 (Docket No. STN 50-528)

LER 86-014-01 and LER 86-037-01. The events have not occurred on Unit 1 Train "B" or either train in Units 2 and 3. This indicates that this occurrence is isolated to Unit 1 Train "A". In LER 87-014-01, likely cause of essential cooling water pump "A" not automatically starting was it was determined that the most dirty contacts in the override relay. A subsequent event reported in LER 037-01 could not identify the cause of the pump not starting. In both events, the breaker was inspected and no problems were identified. Therefore, it is believed that the alignment pin discovered in this event is not related to the previous events.

This LER also contains a Special Report which is being submitted pursuant to Technical Specification (T.S.) 3.3.3.9 ACTION 42b and T.S. 6.9.2 to report an event in which a high range noble gas activity monitor (RU-146) was inoperable for greater than 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. The 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> limit for operability was exceeded at approximately 1356 MST on April 19, 1987. Pursuant to T.S. 3.3.3.9 ACTION 42a the Preplanned Alternate Sampling Program was initiated to monitor the plant ventilation system.

Monitors RU-145 and RU-146 monitor the fuel building ventilation exhaust for release of activity due to a fuel handling accident. Monitor RU-145 performs the safety function of isolating the normal ventilation system initiating a FBEVAS signal on a HIGH-HIGH activity alarm. Monitors RU-145 and RU-146 work as a pair with RU-,145 being the low range monitor and RU-146 being the high range monitor. Normal configuration consist of RU-145 operating and RU-146 in standby.

When RU-145 reaches it's maximum range, RU-146 starts and RU-145 goes to standby.

Since RU-145 and RU-146 work in tandem, both monitors must be declared inoperable if the other malfunctions. The cause of the RU-145 malfunct'ion, as discussed above, has been attributed to a parity error on a byte of data. This caused the counter to reset and establish a new set of sequencing data for the RIC display and would account for the cyclic behavior described by the Shift Supervisor.

As described above, a modification is being developed to install time delay relays into the circuitry to the BOP ESFAS input to eliminate the possibility of ESF actuations due to spurious spikes. RU-145 and RU-146 are expected to be returned to service by June 15, 1987.

N/Ic I 0/IM ssoo I9 83r

NRC form 3SSA (9 83) U.S. NUCLEAR REQULATOR Y COMMISSIQN LICENSEE EVENT REPORT {LER) TEXT CONTINUATION APPROVEO OMS NO 3)50-0)0s EXPIRES: 3/3) /SS FACILITY NAME (11 DOCKET NUMSER (El LER NUMSER (S) ~ AQE (3)

SEQUENTIAL YEAR gg HUM IA < >'I A(VISION

>>UMSSA Palo Verde Unit TEXT ///mo>o soon is FOI>IO>F. O>>

1 OR>O>'>no/I ///IC %%dm> 3(E(A') ((3) 52 887 020 0 04 oF0 Operator (utility-licensed) actions during this event were proper and in accordance with approved procedures. No inoperable systems, structures or components contributed to the event. No unusual characteristics of the work location contributed to this event.

Since an actual emergency condition did not exist at the time of the event, as verified by the negative results obtained from the gas grab sample taken in the vicinity of RU-145, there was no threat to the safe operation of the plant or the health and safety of the public.

Although events involving spurious ESF actuations have been reported, none of these events were attributed to the same root cause.

AAC lOAM SSSO IS 83>

l l

lf

Arizona Nuclear Power Project P.o. 8OX 52034 ~ PHOENIX. ARIZONA 85072-2034 192-00209-JGH/TRB/MJC May 15, 1987 U.S. Nuclear Regulatory Commission Document Control Desk Washington, D.CD 20555

Subject:

Palo Verde Nuclear Generating Station (PVNGS)

Unit 1 Docket No. STN 50-528 Licensee Event Report 87-020-00 File: 87-020-404

Dear Sirs:

Attached please find Licensee Event Report (LER) No. 87-020-00 prepared and submitted pursuant to 10 CFR 50.73. In accordance with 10CFR 50.73(d), we are herewith forwarding a copy of the LER to the Regional Administrator of the Region V Office.

If you have any questions, please contact T. R. Bradish, Compliance Supervisor at (602) 932-5300 Ext. 6936.

Very tr y yours, J. G. Haynes Vice President Nuclear Production JGH/MJC/cld Attachment CC: 0. M. DeMichele (all w/a)

E. E. Van Brunt, Jr.

J. B. Martin R. P. Zimmerman R. C. Sorensen E. A. Licitra A. C. Gehr INFO Records Center

ll li I

'h

$