05000334/LER-2008-001
| Docket Number | |
| Event date: | |
|---|---|
| Report date: | |
| Reporting criterion: | 10 CFR 50.73(a)(2)(i)(B), Prohibited by Technical Specifications 10 CFR 50.73(a)(2)(v), Loss of Safety Function |
| 3342008001R00 - NRC Website | |
There were no structures, components, or systems that were inoperable at the start of the event that contributed to the event (other than the control room envelope boundaries described below). Energy Industry Identification System (EIIS) codes are identified in the text as [XX].
DESCRIPTION OF EVENT
Beaver Valley Power Station (BVPS) has a common air envelope boundary which encompasses both Unit 1 and Unit 2 control rooms. A separate Control Room Emergency Ventilation System (CREVS) is provided on each side of the joint Unit 1/Unit 2 Control Room Envelope (CRE) which will isolate both Units' normal CRE ventilation penetrations and will draw outside air into the CRE through filters to pressurize the CRE following actuation.
On November 10, 2008, the results of the CRE tracer gas in-leakage testing for the BVPS Unit 1 and Unit 2 combined control room [VI] were reported, which was performed per Technical Specification Surveillance Requirement 3.7.10.4. The measured CRE in-leakage value for the Unit 1 CREVS Pressurization Mode was 0 cubic foot per minute (cfm). The measured CRE in-leakage value for the Unit 2 CREVS Pressurization Mode was also 0 cfm.
These values are less than the design basis dose calculation assumption of a maximum of 30 cfm unfiltered in-leakage during a design basis accident (DBA) following CREVS initiation.
CRE during normal plant operation was measured to be 1257 cfm plus/minus 101 cfm, (the initial preliminary value). This exceeded the design basis dose calculation assumption of a maximum of 500 cfm (nominally 300 cfm intake from the Unit 1 ventilation system and 200 cfm from the Unit 2 ventilation system) used in the dose calculations for all DBAs described in both Units' Updated Final Safety Analyses Reports (UFSARs). The normal air intake into the CRE at each Unit is controlled by a manual throttle damper, which is isolated by two upstream in-series safety-related dampers when CREVS is initiated. The CREVS filtered air intake is through a separate pathway. [Note: control room charcoal/HEPA air filters are located only on each Unit's CREVS intake flow path; there are no charcoal/HEPA filters on either of the Units' control room recirculation flow paths.] Each of the two Units' intake throttle dampers were found to be partially degraded and part of the excessive normal intake flow was determined to be due to a corrosion hole in the BVPS Unit 1 normally-closed control room purge intake damper blade, which is in parallel to the normal intake throttle damper in the normal air intake duct path. This purge intake damper is normally closed and is designed only to be opened at the conclusion of a DBA when the control room would need to be purged with a very large outside air flow rate.
DESCRIPTION OF EVENT (Continued) The subject CRE tracer gas in-leakage testing also identified that the inleakage during the control room recirculation ventilation mode (outside air is isolated, but CREVS fans have not yet been manually started) was 292 +/- 23ofrp, which could be slightly above the assumed design value of 300 cfm if the full tolerance of +23 cfm is assumed.
The limiting BVPS Unit 1/Unit 2 calculated control room dose which credits CREVS occurs during a DBA Loss of Coolant Accident (LOCA). Although the normal unfiltered intake flow value may only be utilized for a very short time in the limiting LOCA dose calculation (i.e., the first 77 seconds - time for full CRE isolation as initiated by a Containment Isolation- Phase B signal), the contribution of normal unfiltered intake flow to the calculated control room operator post-DBA dose is not insignificant for all DBA dose calculations.
There was no need to credit CREVS actuation for some DBAs described in the Units' UFSARs since assuming continuous normal control room intake for 30 days post-DBA yielded acceptable calculated control room doses. The effect of an excessive normal intake flow rate is more pronounced for those DBAs which do not credit CREVS initiation in their control room design basis dose calculation than for the DBAs which do credit CREVS initiation. For example, the current BVPS Unit 1 Steam Generator Tube Rupture (SGTR) DBA dose calculated for the control room operator does not credit any CRE isolation or CREVS operation and produced a calculated control room dose which was smaller than the limiting LOCA control room dose which does credit CREVS operation. However, re calculating control room dose using the recently measured excessive normal operation air intake value would cause the BVPS Unit 1 SGTR DBA control room dose to exceed the value currently listed in the UFSAR for SGTR and also above the current limiting LOCA calculated control room dose. Thus, the excessive normal operation air intake value creates an inadequate CRE boundary since it, by itself, invalidates the bounding control room dose calculation by resulting in a calculated control room dose which exceeds the current BVPS Unit 1 and Unit 2 bounding control room dose. An inadequate CRE boundary represents a condition prohibited by plant Technical Specification 3.7.10 for CREVS.
CAUSE OF EVENT
The cause of this event was less than adequate (LTA) CRE boundary definition stated in site procedures which resulted in the failure to ensure adequate maintenance and surveillance activities on appropriate CRE components.
The three different modes of the control room ventilation operation (i.e., Normal, Recirculation, and Pressurization Modes) and their effect on the CRE boundary (as addressed by Technical Specification 3.7.10) was not described in site procedures. The CAUSE OF EVENT (Continued) current site procedures are inadequate since they mainly describe only the CRE building structure components. As a result of this LTA CRE boundary definition, the function of the event related dampers to establish a CRE in-leakage boundary in the Normal Mode of operation had not been clearly recognized as a critical aspect of CREVS operability. In the Pressurization Mode of operation (i.e., post-DBA), two upstream series isolation dampers are closed to form the ventilation boundary - not the event related (normal intake throttle) dampers. The common site understanding of the definition of the CRE boundary was typically associated with the Pressurization Mode of operation. Therefore, the need to establish periodic preventative maintenance tasks and/or flow, measurements to maintain normal intake flow rates within limits for the event-related dampers was not recognized until after the dampers degraded to the point that the in-leakage flow rates in the normal mode of operation exceeded the maximum value of 500 cfm.
ANALYSIS OF EVENT
The BVPS Unit 1 intake purge damper (1VS-D-40-1G) was degraded by a hole created by corrosion (which was subsequently determined to have been potentially accelerated by ice-removal material spread for winter walkway clearance). Corrosion is a slow process which does not occur in a short time frame. Hence, this was firm evidence that the abnormally high CRE intake flow, exacerbated by the 1VS-D-40-1G pathway hole, existed prior to the recent tracer gas testing. This flow rate through the closed purge damper would have been in addition to any flow via the normal intake pathways, and would have exceeded the licensing basis analyses limit for the total normal intake flow rate assumed in the dose calculations of record (less than 500 cfm total intake limit). This would have caused a higher-than-bounding CRE dose and hence, an insufficient CRE boundary.
Technical Specification 3.7.10 Action B applies for an inoperable CRE boundary and contains three Required Actions with Completion Times ranging from 'Immediately' to '90 days'. The inadequate CRE boundary would have been a condition prohibited by plant Technical Specifications for greater than the time frames allowed by Technical Specification Action B completion times. Therefore, this is reportable pursuant to 10 CFR 50.73(a)(2)(i)(B).
The potential higher than design inleakage during the control room recirculation ventilation mode (292 cfm +/- 23 cfm) was not caused by the 1VS-D-40-1G corrosion hole, and thus, it was not a firm pre-existing condition. Therefore, this was not reportable pursuant to 50.73(a)(2)(i)(B).
An evaluation was performed to assess the impact of the changes in the control room ventilation intake/inleakage flows measured by the tracer tests on the control room ANALYSIS OF EVENT (Continued) operator dose following a design basis accident. The evaluation determined that the regulatory limit of 5 REM TEDE would not be exceeded following any DBA. Therefore, no significant impact on the control room operators to perform their duties following postulated design basis accidents are expected, and this event is considered to have very low safety significance.
There was no loss of safety function for the CRE pursuant to 10CFR 50.73(a)(2)(v) since the largest increase in calculated control room dose resulting from the measured increase in normal intake air flow would not have exceeded the General Design Criteria 19 regulatory limit of 5 REM TEDE, even though the dose increase would have been more than the bounding value currently listed in the Unit's UFSAR.
CORRECTIVE ACTIONS
1. Immediate actions included closing both Units' CREVS isolation dampers to reduce the normal intake of unfiltered air to a value less than assumed in the DBA dose calculations.
2. The CRE dampers were repaired and the subsequent normal in-leakage rate was satisfactory.
3. A site document will be developed to clearly define the CRE boundary to also include boundaries used during normal and recirculation modes of control room ventilation modes of operations, and not just the post-CREVS initiation pressurization mode of.
operation.
4. Periodic preventative maintenance and periodic intake flow measurements will be incorporated for the normal CRE intake dampers.
Completion of the above and other corrective actions are being tracked through the BVPS corrective action program.
PREVIOUS SIMILAR EVENTS
A review found no prior BVPS Unit No. 1 or BVPS Unit No. 2 Licensee Event Report within the last three years for an event involving the control room boundary/envelope.
CR 08-49260