05000206/LER-1981-018, Forwards LER 81-018/01T-0.Detailed Event Analysis Submitted

Forwards LER 81-018/01T-0.Detailed Event Analysis Submitted

ML20010B044
Person / Time
Site:	San Onofre
Issue date:	07/31/1981
From:	Haynes J SOUTHERN CALIFORNIA EDISON CO.
To:	Engelken R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV)
Shared Package
ML20010B045	List: 05000206/LER-1981-018, Forwards LER 81-018/01T-0.Detailed Event Analysis Submitted 05000206/LER-1981-018-01, /01T-0:on 810717,hydrogen-air Mixture Ignited in North Radwaste Gas Decay Tank Resulting in Manway Damage & Gaseous Release.Caused by check-valve Back Leakage of Volatile Gas Mix at Dinitrogen Instrument Air Interface
References
TAC-65149, NUDOCS 8108130250
Download: ML20010B044 (3)

LER-1981-018, Forwards LER 81-018/01T-0.Detailed Event Analysis Submitted

Event date:
Report date:
2061981018R00 - NRC Website

text

_-

4 h5 Southern California Edison Company q

P, O. BOX 8 00 22 44 WALNUT GROVE AVENUE f gg -

ROSE M E AD. CALIFORNI A 91770 i

0 agou v re..,~o~r 4.c.wavurs July 31, 1981 Wosc/$8 C '"""""'

8

- ~... o, ~ u c u.. o,.....o ~,

T1 U. S. Nuclear Regulatory Commission Office of Inspection and Enforcement Region V 1990 North California Boulevard Suite 202, Walnut Creek Plaza A

Walnut Craek, California 94596 l

Attention: Mr. R. H. Engelken, Director e

7 412198k TL DOCKET No. 50-206 A

SAN ONOFRE - UNIT 1

\\(g WY

Dear Sir:

By letter dated July 20, 1981 we provided prompt notification of a reportable occurrence involving the Radioactive Waste Gas System. This letter constitutes the two week follow-up report.

Subm1ttal is in accordance with the reporting requirements stipulated in Section 5.6.3 of Appendix B to our Provisional Operating License DPR-13.

At 2125 on July 17, 1980 a venting process was underway for the' north radwaste gas decay tank C-6A (located in the Reactor Auxiliary Building) through the cryogenic waste gas treatment system. The system had been placed in service in accordance with approved procedures.

It is noted that just prior to processing the north decay tank the set th decay tank had been processed without incident. As processing began on the north decay tank difficulty was experienced in adjusting the flow rate through the system.

Investigation of the problem determined that a reducing valve which controls dcwnstream gas pressure into the gas treatment system was cycling. While attempting to alleviate this problem an operator noted " popping" noises, apparently originating in the piping downstream of the reducing valve.

Sometime thereafter an ignition of the gas mixture in tank C-6A occurred.

Fire alanns were observed in the control room, and operating personnel within the gas treatment system area were notified. These personnel reported a loud noise, apparently from the gas decay tank area. Operating personnel entered the Reactor Auxiliary Building and reported the presence of smoke. Local pressure indication on tank C-6A was observed to read 0-psig. The operating-personnel exited the area and requested a radinlogical survey. While the survey was in progress, operating personnel re-entered the area in Scott-Air-Packs and inspected the north waste gas decay tank.

At this time the area was observed cleared of smoke and no fire was prdsent.

The tank manway cover bolts were found loose and minor damage around the tank manway noted.

(=)108130250 810731 DR ADOCK 05000 3g

4 U. S. Nuclear Regulatory Commission LER 81-018 Page # 2 Post investigation of this event found that damage to the tank is apparently limited to the manway area and the manway bolting.

It appears that the source of the ignition was the oxygen recombiner located in the cryogenic waste gas treatment system. At oxygen concentrations greater than 3% the chemical reaction in the recombiner can generate temperatures above the ignition temperature for an oxygen-hydrogen combination. The manufacturer provides an oxygen analyzer with this system to detect oxygen levels exceeding this value. However, at the time of the event this meter was inoperable and its use was not required by the operating procedure. Local ignition of the hydrogen-oxygen atmosphere evidently created pressure spikes that caused the control valve, CV-M2, to cycle resulting in the popping noise observed by the operators. During one of the local ignitions the flame front propagated upstream of CV-M2 and ignited the gas mixture in the tank.

An analysis of the plant stack monitoring system subsequent to the event indicated that the activity release due to the man-way cover failure (8.8 curies) occurred during an 18 minute period.

Utilizing an extrapolated peak for the most limiting isotope involved (Xe-133), the dilution "clume through the stack and meteorological dilution factors, the calculated plant boundary activity concentration was determined to be a factor of ten below the maximJm permissible Concentration delineated in Appendi.' B, Part 20 of CFR Title 10. Additionally, onsite personnel exposure associated with this event was well within limits.

Subsequent investigation into the cause of high oxygen concentrations within the waste gas system revealed high oxygen concentrations in the station's Nitrogen System. This had been discovered the day prior to the event and had led to sampling of the flash tank gas space (completed just prior to the ignition). This sample showed the presence of oxygen in the tank.

It has been determined that instrument air (maintained 10 to 20 psig above the Nitrogen System pressure) was ente.ing the Nitrogen System at locations where nitrogen serves as a backup to instrument air. The backup function is for the following safety related valves: Pressurizer relief and pressurizer relief block valves, isolation valve on the nitrogen supply to the pressurizer, Auxiliary Feedwater System control valves and the auxiliary feedwater turbine driven pump, steam control valve.

The primary source of leakage was confined tn those valves associated with the auxiliary feedwater system (insignificant leakage was found at the other locations).

The check valves isolating the two systems had not been installed in accordance with manufacturer's recommendations during the recent TMI backfit modifications.

As a result of this installation the valves failed to seat properly providing a direct and rather large communication path between the two systems.

S

U. S. Nuclear Regulatory Commissicr.

LER 81-018 Page # 3 The following action will be taken in regard to this incident:

(1 )

The Instrument Air System and the Station's Nitrogen System will be physically separated from each other and bottled gas used in lieu of _

the station Nitrogen System.

t (2)

The procedure used to operate the waste gas treatment system will be revised to prohibit system use if oxygen concentrations above 3% are present (manufacturer's upper limit on oxygen concentration). Sampling for oxygen prior to rel(ise will be required.

(3)

Tank C-6A shall be examined with liquid penetrant, repaired and pressure tested in accordance with the ASME Boiler and Pressure Yessel Code,Section VIII.

(4)

The pressure boundary components in the immediaw vicinity of tank C-6A shall b uamined for indications of damage and action taken accordingly.

(5)

In the cryogenic waste gas treatment system, the oxygen recombiner and system filter will be replaced, and the oxygen analyzer repaired.

Items (1) and (2) will be complete prior to return to power.

Items (3), (4) and (5) are currently in progress.

It is our intent to complete these activities prior to return to power subject to material lead times and the results of inspections t:nderway.

Should you have any questions on the above, please call me.

Sincerel,

fynt/$-

J. G. Haynes Manager of' Nuclear Operations JTR:dh:900

Enclosures:

Licensee Event Report 80-018 cc:

U. S. Nuclear Regulatory Commission Office of Inspection and Enforcement U. S. Nuclear Regulatory Commission Office of Management Infonnation & Program Control Nuclear Safety Analysis Center L. F. Miller (USNRC Resident Inspector) a