ML20196J862

Discusses Three Plants,Evaluated for Consideration as Abnormal Occurrences & Other Event of Interest.Oconee & Quad Cities Plants Are Being Considered as AOs & Big Rock Point Being Considered as Other Event Interest

ML20196J862
Person / Time
Site:	Oconee, Quad Cities, Big Rock Point  File:Consumers Energy icon.png
Issue date:	12/04/1998
From:	Roe J NRC (Affiliation Not Assigned)
To:	Congel F NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD)
References
NUDOCS 9812110049
Download: ML20196J862 (5)

Text

,

[l {

ps sg i

UNITED STATES l

. 2 NUCLEAR REGULATORY COMMISSION j I E ;f WASHINGTON, D.C. 30666-4001 l

• December 4, 1998 l MEMORANDUM TO: Frank J. Congel, Director Incident Response Division, AEOD FROM: f Jack W. Roe, Acting Director b

y [b Division of Reactor Program ManagemAnt, NRR

SUBJECT:

ABNORMAL OCCURRENCE INPUT l

l The Events Assessment and Generic Communications Branch of NRR has evaluated two i

events for consideration as Abnormal Occurrences (AOs) and one event for consideration as an "Other Event of interest"(OEI). As a result of our evaluation, we are recommending two events as OEls.

The first event considered for classification as an AO involves the.OconeeNuclear Station. The licensee found inconsistencies between the piping, instrument, and owner's manual drawings related to the Borated Water Storage Tank (BWST)levelinstrumentation. During a loss of coolant accident, these inconsistencies could delay the manual swapover of the suction for the emergency core cooling system (ECCS) pumps from the BWST to the reactor building emergency sump (RBES), causing pump cavitation. However, based on the staff's evaluation, we have concluded that operator action would ensure that the system would perform its safety function and that at no time was the health and safety of the public at risk. Therefore, we removed this event from further consideration as an AO.

The second event considered for classification as an AO involves Quad _ Cities. Results of the licensee's initial Individual Plant Examination of Extemal Events (IPEEE) analysis calculated a conditional core damage frequency (CDF) of SE-3 per reactor year for fire due to a high reliance on plant shutdown from outside the control room (even for fires that do not affect the control room), a high number of manual actions needed to reach safe shutdown, and reliance on  !

equipment in the unit not affected by the fire. These conditions existed from the mid 1980s until the end of 1997, when the licensee shut down both units.

We believe that this condition represented a major deficiency in design having potentially 0 significant safety implications, required immediate remedial action and, as such, merits i consideration as an abnormal occurrence. However, the safety significance of this condition is e c/ <

uncertain because the licensee is developing new information that indicates the licensee may 9 l have been able to safely shut down the facilities in the event of a fire. Therefore, we are proposing the condition as an OEl.

The other event considered for classification as an OEl involves Big Rock Point. A boroscopic inspection revealed that the discharge pipe of the liquid poison system (LPS) was completely severed, rendering the system incapable of injecting boron into the RCS during an accident. It was determined that this corHition may have existed as eart as 1979.

Contact:

W. Burton, NRR 415-2853 794rg/cxrd>q Ed '

,., o R -m {~-' r n ~7 9,T3; 9812110049 981204 5 ADOCK 0500 BD uLM C0JO QUb d ,

gDR 1

1

s F. J. Congel Dece:ter 4,1998 Big Rock Point is currently in decommissioning and as such, the LPS is not required to be operable. However, during the time that the unit was operational, this system would not have been able to perform its safety function if called upon.

This event was discussed in the congressional hearing held on July 30,1998. Due to the public interest associated with this event, we have determined that it warrants classification as an "Other Event of Interest."

A synopsis for each event is provided below.

Our considerations utilized the final abnormal occurrence criteria approved by the Commission for publication in the Federal Reaister and contained in Management Directive 8.1, " Abnormal Occurrence Reporting Procedure" and the guidance provided in Staff Requirements Memorandum dated September 4,1998, in response to SECY-98-175, " Proposed Guidelines for Appendix C, 'Other Dvents of interest,' to the Abnormal Occurrence Report to Congress."

Inonerability of Emeroency Core Cool!ng System and Building Sorav Pumos at Oconee The BWST is normally used as a source of borated water for filling the refueling cavity and fuel transfer canal during refueling operations. Its emergency function is to provide the initial source of borated water during a loss-of-coolant-accident (LOCA) for the ECCS, composed of the high pressure injection (HPI) and low pressure injection (LPI) systems, and the building spray (BS) system.

During a LOCA, flow is initiated in the HPI and LPI systems from the BWST to the reactor vessel and in the BS system from the BWST to the building spray headers. When the BWST approaches minimum level as part of the transition into long-term recirculation cooling, operators manually realign the ECCS pump suctions from the BWST to the RBES. Wide-range reactor building water level instrumentation is used to indicate the RBES level following a LOCA.

In February 1998, during a Self-Initiated Technical Audit conducted by the licensee to assess the operational readiness and functionality of the HPI and LPI systems, the licensee identified two problems impacting the ECCS and BS system. The first problem involved BWST level transmitters that did not have a specified zero elevation to ensure that the transmitter zero corresponded to the BWST zero reference. The associated calibration procedure did not provide guidance to ensure that the calibration source was positioned properly to correlate the transmitter zero to the tank zero. As a result of these errors, the indicated BWST levels at each unit indicated approximately 12 inches (18 inches in the worst case for the Unit 3 instrument) i higher than the actual level. This level error could result in a delay in transferring the ECCS l and BS pump suctions, causing vortexing at the pump suctions, pump cavitation, and subsequent pump failure.

l

F. J. Congel Deceter 4,1998 l l

The second problem involved level in the RBES. The Emergency Operating Procedure (EOP) presumed reactor building water level would indicate greater than 4 feet when the BWST level decreased to 6 feet or less when realigning ECCS and BS pump suctions to the RBES.

While addressing the BWST level instrumentation problem, a licensee engineer discovered that, due to unaccounted for instrument uncertainties, the RBES water levelinstrument might not indicate greater than 4 feet when BWST level reached 6 feet. This created a conflict between the BWST/ reactor building (RB) levels specified in the EOP for swapover to the RBES and the BWST/RB levels indicated in the control room. As a result, during certain desip basis accident scenarios, the level indication errors would have resulted in the failure to satisfy EOP requirements for the combination of indicated levels for the BWST and RBES and would have delayed swapover initiation, resulting in vortexing in the BWST and air binding of the ECCS and BS pumps. The BWST level instruments were recalibrated and interim guidance was provided to the operators until the EOPs could be changed.

These deficient.ies did not result in an actual challenge or failure of a safety system.

Consequently, at no time was the health and safety of the public at risk. Further, should a LOCA have occurred with these deficiencies in place, operators would have been alerted to the abnormal condition by erratic pump behavior and could have taken steps to transfer the pump suctions to the RBES. However, during certain small break LOCA scenarios, HPl pumps with less margin for net positive suction head at the pump suction could be damaged. For such an occurrence, operators would be required to depressurize the plant to within the discharge pressure of the LPI pumps. Therefore, the staff concludes that long-term core cooling could be l achieved with the identified deficiencies in place. Staff analysis calculated a CDF for this event '

in the low E-5/ year range, resulting in the categorization of this as a precursor event for the AEOD Accident Sequence Precursor program. ,

We have determined that this condition did not result in the inability of the system to perform its safety function if called upon, and therefore does not warrant classification as an AO.

l Qgad Cities Fire Procram Deficiencies Commonwealth Edison, the licensee for Quad Cities, by letter dated February 17,1997, provided the NRC with the results of an IPEEE analysis. The conditional CDF due to fire reported in the analysis is SE-03 per reactor year. Preliminary results of a new fire risk analysis indicates the initial CDF may have been too high by about a factor of 100. Major contributors to this risk are: the high reliance on post-fire safe shutdown methodologies that require plant shutdown from outside the main control room (even for fires that do not directly affect the control room), the significant number of manual operator actions required to perform the post-fire safe shutdown, and the reliance on equipment and systems from the non-fire affected unit to shut down the reactor in the fire affected unit. In addition, fires in certain plant areas require shutdown of both units from outside the control room. The licensee implemented an interim

. altemate shutdown method and took other compensatory actions in March 1997, to reduce the risk from a fire.

. i F. J. Congel Decater 4,1998 The NRC staff visited the site, gained additionalinsights related to plant fire va'nerabilities and made recommendations regarding needed short term actions. The licensee subsequently expressed concem that fire induced circuit failures could impact the implementation of the post-fire safe shutdown methodology and declared all post-fire shutdown paths for both units l inoperable. The licensee determined that a postulated fire could adversely impact cabling of l

both units that is required for shutdown and that the post-fire safe shutdown procedures were inconsistent with the safe shutdown analysis. As a result, the licensee shut down Unit 2 in late September 1997 and Unit 1 in December 1997.

These deficiencies did not result in an actual challenge or failure of a safety system. However, the potential for endangering public health ard rafety did exist as evidenced by the total time the condition existed, i.e., from initial fire protection efforts in the mid-1980s until 1997. I Therefore, this condition was believed to warrant consideration as an abnormal occurrence, based upon criterion D, a major deficiency m design having significant safety implication requiring immediate remedial action. However, since the safety concem was identified, the licensee has been evaluating procedures, equipment vulnerabilities, and training, and has made ,

improvements to their fire protection capability. As a result of this continuallearning and l improvement process, it is difficult to establish the capability to cope with a fire should one have l occurred prior to the time when the program shortcomings were identified. Therefore, the I actual safety significance of the shortcomings is uncertain. If it were known that the licensee '

could not cope with a fire, the condition would have been classified as an abnormal occurrence.

However, information currently available from the licensee indicates that they could have safely shut down the facilities during the period of concern. Considering the potential safety j significance, public interest and increased NRC staff attention to the fire program, the staff recommends that this condition be reported as an "Other Event of interest."

l Loss of Liauld Poison System at Bla Rock Point l Big Rock Point was permanently shut down on August 29,1997. The last fuel bundle was  !

removed from the reactor vessel on September 20,1997. On March 27.,1998, an unsuccessful attempt was made to discharge the contents of the LPS. On April 24,1998, a boroscopic inspection revealed that the discharge pipe of the LPS tank was completely severed approximately 6 inches above the water line.

The purpose of the LPS is to inject boron into the reactor vessel to shut down the reactor in the event of a failure of the normal reactor control rod system. The LPS tank is filled with a concentrated solution of sodium pentaborate to accomplish the shut down. The severed pipe rendered the system inoperable. The licensee's root cause analysis concluded that the probable root cause of the failure was inadequate curing of the phenolic coating on the discharge pipe at the time of manufacture in 1961. After the phenolic coating failed, the carbon steel discharge pipe was exposed and subject to corrosion. Based on metallurgical analysis performed by the licensee, the licensee estimates that the failure of the carbon steel pipe occurred between 1979 and 1984 due to corrosion. Therefore, the LPS was rendered inoperable during the last 14 years of reactor operation.

• i F.J.Congel m 4, 1998 3 .. I The increase in CDF associated with this event was small (4 percent). Currently the unit is undergoing decommissioning and LPS is not required to be operable. During operation, the LPS served as a safety-related backup to other ECCS systems. As such, the ECCS systems  !

were available to inject water into the reactor if a LOCA had occurred. Therefore, the failure of the LPS did not endanger the public health and safety.

Because this event was discussed during the congressional hearing held on July 30,1998, we i have determined that it warrants classification as an "Other Event of Interest"in accordance with current Commission guidance. l DISTRIBUTION Central File H. Karagiannis PUBLIC W. Burton ,

PECB R/F J. Carter J. Roe J. Zwolinski H. Berkow D. LaBarge S. Weiss P. Harris i

DOCUMENT NAME: GnWFB\AO2 To receive a copy of this document, indicate in the box C= Copy w/o attachment / enclosure E= Copy with attachment / enclosure N = No copy l OFFICE PECB PECB C:PECB D:DSSA (A)DfDh hI NAME WBurton:jkd* RDennig* JStolz* GHolahan* JRoe b l DATE 10/05/98- 10/05/98 10/05/98 11/20/98 [l/ /98 OFFICIAL RECORD COPY l

l 1

i

' l I

i

,