IR 05000261/1989018
| ML14176A772 | |
| Person / Time | |
|---|---|
| Site: | Robinson  |
| Issue date: | 09/14/1989 |
| From: | Dance H, Garner L, Jury K NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML14176A771 | List: |
| References | |
| 50-261-89-18, NUDOCS 8910030472 | |
| Download: ML14176A772 (7) | |
Text
UNITED STATES 0 PNUCLEAR REGULATORY COMMISSION
REGION II
101 MARIETTA STREET, ATLANTA, GEORGIA 30323 Report No.:
50-261/89-18 Licensee:
Carolina Power and Light Company P. 0. Box 1551 Raleigh, NC 27602 Docket No.:
50-261 License No.: DPR-23 Facility Name:
H. B. Robinson Inspection Conducted: August 14 -
September 8, 1989 Inspector:
f=/
L. W. Garner, Senior Resdentnlhspector Da egned ry Fesident Inspecton Approved by: /c
'
r H. C. Dance, Section Chief DatSigbed Division of Reactor Projects SUMMARY Scope:
This special -announced inspection was conducted to review the circumstances surrounding inoperability of the Auxiliary Feedwater (AFW)
system, which resulted in a plant shutdown on August 22, 198 Results:
Two violations were identified:
The first violation, which was previously identified in Inspection Report 89-11 as violation 89-11-01, concerns the failure-to have three AFW pumps operable since January 6, 198 The second violation concerns the failure to take corrective action to assure conditions adverse to quality were promptly identified and correcte No notice of violation is being issued since these items are under consideration for escalated enforcemen.003047 PD R ArinCK
~
e
REPORT DETAILS 1. Persons Contacted C. Baucom, Senior Specialist, Regulatory Compliance C. Coffman, Project Engineer, Onsite Nuclear Safety
- 0. Crook, Senior Specialist, Regulatory Compliance
- J. Curley, Director, Regulatory Compliance
- R. Dayton, Project Engineer, Technical Support
- C. Dietz, Manager, Robinson Nuclear Project
- J. Eads, Project Engineer, Nuclear Licensing W. Flanagan, Manager, Modification Projects W. Gainey, Supervisor, Operations Support
- S. Griggs, Technical Aide, Regulatory Compliance E. Harris, Director, Onsite Nuclear Safety R. Johnson, Manager, Control and Administration
- C. Martin, Principal Engineer, Onsite Nuclear Safety, Brunswick Steam Electric Plant
- A. McCauley, Principal Engineer, Onsite Nuclear Safety
- J. Moon, Senior Engineer, Technical Support
.*R. Morgan, Plant General Manager D. Nelson, Maintenance Supervisor, Mechanical
- M. Page, Acting Manager, Technical Support
- D. Quick, Manager, Maintenance
- J. Sheppard, Manager, Operations
- R. Smith, Manager, Environmental and Radiation Control
- R. Snaider, CRESAP
- D. Stadler, Onsite Licensing Engineer
- H. Young, Director, Quality Assurance/Quality Control
- Attended exit interview on September 8, 1989 Acronyms and initialisms used throughout this report are listed in the last paragraph of the inspection repor.
Followup of Significant Events (93702)
On August 22, 1989, at.1:46 a.m., the unit was placed in hot shutdown in accordance with TS 3. Technical Specification 3.4.5 requires that with two AFW pumps inoperable, either restore one inoperable AFW pump to operable status within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> or be in hot shutdown within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />. At 12:00 a.m. on August 22, 1989, the licensee determined that one of the MDAFW pumps was inoperable as hydraulic calculations indicated insufficient NPSHa for the MDAFW pumps under certain postulated design condition The calculations showed that with the CST at the maximum
- guaranteed level, the MDAFW pumps at rated flow would experience
insufficient NPSHa before delivering the 35,000 gallons to the S/Gs required by TS 3.4. The maximum guaranteed level is approximately 7.5 feet, the elevation at which lines from the CST are not seismically qualifie Based on previous calculations, the SDAFW pump was declared inoperable on August 16, 1989, at 12:55 p.m.,
because of a NPSHa inadequacy. These calculations determined that two MDAFW pumps at rated flow would have sufficient NPSH However, when the SDAFW pump is operated at rated flow with two operating MDAFW pumps, none of the pumps have sufficient NPSH Two sequences of events, which were integrated into the DBD effort, eventually resulted in the AFW system operability problem identificatio One sequence involved a licensee self-initiated AFW SSFI by CN Research and inspection effort was completed on the internal AFW SSFI by December 198 The final draft and final-report was issued in March 1987, and August 1987, respectivel Both documents identified that there was no evidence that simultaneous three pump operation flow requirements were adequately considered and factored into the original AFW system design. Calculations determined that each AFW pump met suction pressure requirements at 35,000 gallons CST leve However, the reports identified that at 1,200 gpm, deficiencies would occur with SDAFW pump NPSHa before the 35,000 gallon CST level was reached. In September 1987, this concern was assigned a RAIL number to be tracked until resolutio In July 1988, to resolve concerns, identified in the RAIL item, Technical Support requested that analysis be included in the DBD to address the plant concern centered around the MDAFW and SDAFW pump suction piping size and NPSHa, for "continuous multiple pump operation without damage".
The other interrelated sequence involved observation of abnormal equipment performance during a reactor trip. On December 22, 1986, an AFW SSFI team member observed that with all three AFW pumps supplying water to the S/Gs, each pump delivered less than demanded flow. Total flow was approximately 300 gpm less than desig The post trip review did not address reduced AFW system flow, nor operability considerations for the observed condition On January 14, 1987, at Technical Support's request, ONS initiated special project 87-01 to investigate the reduced flow phenomena. On February 18, 1987, the 87-01 report was issued, concluding that the reduced flow was not a function of low NPSH nor too small suction piping..The report determined that at the time of the event (CST level 19.5 feet), if all three AFW pumps were providing 1200 gpm total flow, there would be 2 feet NPSHa to the SDAFW pump; NPSHr is 23 fee However, the calculation was flawed, in that, approximately 4 feet of the 24.7 feet is not available since the SDAFW pump shaft is elevated approximately 4 feet from the CST bottom; thus, inadequate NPSHa could have been a proble The report failed to identify this fact or that the analysis was not a worse case, bounding type calculatio For example, the calculations did not consider the effect of valves, tees, and elbows when calculating the head loss due to pipin The report suggested the problem could be with the
pumps themselves, restrictions to the flow in the piping, improper controller settings on the SDAFW pump, problems with the flow monitoring equipment, or any combination of these factor The report recommended running an AFW system flow test under conditions similar to those experienced during the 12/22/86 even The inspectors were not provided information which would indicate that the potential problems listed above were actually evaluated. or investigate On March 2, 1987, Technical Support issued PIR 87-027/00, requesting a study (including field testing)
to determine system conditions when all three AFW pumps are operating at different CST levels, and to recommend appropriate corrective actions if neede Reference was made to the fact that total flow to the S/Gs was sufficient to satisfy accident requirements,
"but the cause for the reduction has yet to be determined and could point to possible design problems with the system".
The PIR was accepted for implementation on April 23, 198 The PIR was incorporated into the five year schedul Work was to begin January 1988 and complete by December 31, 1988; however, this effort never materialize At some point in 1988, a decision was apparently made to incorporate the PIR into the DBD effort. Phase I of the AFW DBD, compilation of design information, was scheduled to be completed prior to January, 198 On December 30, 1988, the AFW DBD was issued for preliminary use; however, the DBD had not been field validate On March 28, 1989, to support the validation phase (phase II)
and to address PIR 87-027/00, the DBD phase II project manager requested that the NED Mechanical/Nuclear unit prepare an AFW system hydraulic calculatio Thus, by the first quarter of 1989 (possibly earlier), both the SSFI event sequence and the December 22, 1986 trip event sequence had become integrated into the AFW DBD phase II proces Hydraulic calculations were performed between June 2 and July 13, 1989, and were issued on July 19, 1989, as RNP-MN/MECH-1034, revision 0. The report concluded that the CST volume and suction pipe sizing seemed to be adequate for 600 gpm system flow (i.e., that the MDAFW pumps or SDAFW pump were capable of operating until the CST level dropped near the AFW nozzle).
However, it stated that to support higher system flow rates, especially those resulting from common auto start signals (i.e., 3 pump operation),
very high CST levels (97.3%)
are required to provide NPSH to the SDAFW pum This conclusion was considered to be a potential significant deficiency affecting operability; therefore, on July 2, 1989, it was entered into the discrepancy resolution program as item AFW-D2 On July 21, 1989, Modification Projects Manager and Technical Support Personnel were notified of this potential significant deficienc The report also noted that the SDAFW pump NPSHa was marginal with a 600 gpm flow rate and both MDAFW pumps secure The NPSHa would have been sufficient until the CST level dropped to one foot below the pump shaft elevation or approximately. one foot above the CST AFW nozzl This conclusion left very little margin for calculational erro Plant review of RPN-MN/MECH-1034, revision 0, -revealed two potential errors:
(1) the recirculation flow had not been properly incorporated in the calculations; and (2) the maximum assumed temperature of the CST was
calculated at 120oF instead of the 115 0F utilized in accident analysi The first would reduce the NPSHa while the second would increase the NPSH Thus, the errors were mistakenly considered to compensate for.one another. However, an adjustment of the head loss for actual recirculation flow, using the-relationship that head loss is proportional to the square of the flow rate, would show that the increased flow rate would signifi cantly overshadow any vapor pressure change from 120'F to 1150F. Apparently neither NED nor Technical Support performed such. an estimation of the impact of the potential errors on the conclusions of RNP-MN/MECH-1034, revision In addition, the temperature statement in (2) above was incorrect, in that the NPSH calculation had actually been performed using vapor pressure at 100'F, thus correction to 1150F would increase vapor pressure and decrease the calculated NPSH Independent of NED's calculation efforts, ONS was developing their own computer model for determining NPSHa to the AFW pumps. On July 25, 1989, ONS benchmarked their model, with good agreement, against field data taken on June 20, 198 On the same day, ONS reviewed NED's calculation and determined that it was non-conservative. On July 27, 1989, ONS informed (via memo)
the NED engineer of this conclusio Subsequently, ONS had verbal discussions with the NED engineers concerning their result On August 16, 1989, revision 1 to RNP-MN/MECH-1034 was issue Revision 1 calculations determined that with a full CST, all three AFW pumps would not have sufficient NPSHa if all three pumps were delivering rated flow to the S/G Additionally, the SDAFW pump would have insufficient NPSHa at less than 5.4 feet (at 1150%)
or 3.4 feet (at 40'F) above the pump shaft (i.e., at CST levels below 9.4 feet and 7.4 feet, respectively).
Both MDAFW pumps operating together would have sufficient NPSHa at CST levels below the CST AFW outlet nozzle. 'Based upon these results, on August 16, 1989, at 12:55 p.m.,
the licensee declared the SDAFW pump inoperabl Revision 1 addressed increased flow due to recirculation and a potential CST temperature of 115'F, as well as refining the calculation for piping friction losses; however, this resulted in no significant change in head los NED had not attempted to verify its model using plant data as ONS ha ONS performed calculations for the different pumping configurations addressed by NED in RNP-MN/MECH-1034, revision 1. On August 16, 1989, ONS concluded that the NED calculations still appeared to be non-conservativ This concern was transmitted to the plant staff during a project review meeting on August 18, 198 At this time, based upon NED's calculations, the plant was developing a modification, a JCO, and an emergency TS relief request, to return the SDAFW pump to service prior to the seven day LCO expiration at 12:15 a.m., on August 2 On August-20, 1989, ONS concluded that the approved calculations (revision 1) would not support operation of the AFW pumps (as proposed).
This concern was discussed with the Technical Support Manager and NED engineer associated with the DBD progra It was decided that this item would be discussed and resolved with NED the next da On August 21, 1989, several discussions and a meeting occurred between ONS and NED concerning the different calculational models as
compared to actual plant flow dat That evening, NED informed plant management that two MDAFW pumps running at rated flow would experience insufficient NPSHa at low CST level This resulted in the plant decision to declare one MDAFW pump inoperable and shut the unit down as required by T Based on the above sequences of events/actions, the inspectors concluded that with aggressive analyses and sufficient resolution of the following opportunities, a timely realization of AFW system inoperability would have resulte These opportunities include, but are not limited to, the following:
(1) Data from reactor trips on, and subsequent to, December 22, 1986, indicated reduced AFW flow with three pumps operatin Occurrences of this phenomena were not addressed by the respective post trip review (2) Incomplete/faulty analysis contained in Special Project 87-01 and RNP-MN/MECH-1034, Revisions 0 and 1, respectivel (3) Significant AFW system deficiencies identified in the AFW SSF (4) Test data was taken on June 20, 1989, during operation of the SDAFW pum The data was taken to access the differential -pressure and vibration performance of the pum The data was apparently not reviewed for other pump operating characteristics (i.e. NPSH).
A relatively simple comparison of the suction pressure and CST level would have indicated that levels above 7.5 feet (non seismic lines are.attached at 7.5 feet) would be required to maintain adequate NPSHa to the SDAFW pump when running at rated flow (no MDAFW pumps in service).
(5) During SDAFW pump surveillance testing at 600 gpm flow to the S/Gs, the installed suction pressure gauge pegs below the zero indicatio A below zero indication does not necessarily indicate a problem with NPSH; however, apparently there was no attempt to determine relationship to NPS The above instances afforded the licensee numerous opportunities to recognize the significance of reduced AFW system performance. Aggressive identification of the significance and correction of these anomalies would have resulted in a timely resolution of AFW system concern This is a violation of 10 CFR 50, Appendix B, Criterion XVI, in that, items (1)
through (5) are examples of failures to promptly identify and correct conditions adverse to quality (89-18-01).
On January 6, 1983, TS amendment No. 74 was issued to require operability of all three AFW pumps above 3500 The failure to maintain three AFW pumps operable since January 6, 1983, as required by TS 3.4.5, is a violatio This was previously identified in Inspection Report 89-11 as violation (89-11-01).
No notices of violations are being issued since these items are being considered for escalated enforcemen. Exit Interview (30703)
The inspection scope and findings were summarized on September 8, 1989, with those persons indicated in paragraph 1. The inspectors described the areas inspected and discussed in detail the inspection findings listed below and in the summar Dissenting comments were not received from the license Proprietary information is not contained in this repor Subsequent to the exit, a member of the internal AFW SSFI team indicated that to the best of his recollection, the three pump operation concern was incorporated in the SSFI report as a result of the December 22, 1986, reduced flow operation observatio No documentation is available confirming this belie The inspectors considered this additional information and determined that it would alter the sequence of events, but would not affect the inspectors conclusions-concerning the Criterion XVI violatio Item Number Description/Reference Paragraph 89-18-01 VIO -
10 CFR-50 Appendix B Criterion XVI Failure to Promptly Identify and Correct Conditions Associated With the AFW System, paragraph VIO -
Failure to Maintain Three AFW Pumps Operable as Required by TS 3.4.5, paragraph. List of Acronyms and Initialisms AFW Auxiliary Feedwater CNS Corporate Nuclear Safety CST Condensate Storage Tank DBD Design Basis Document F
Fahrenheit JCO Justification For Continued Operation MDAFW Motor Driven Auxiliary Feedwater NPSH Net Positive Suction Head NPSHa Net Positive Suction Head available NPSHr Net Positive Suction Head required NED Nuclear Engineering Department ONS Onsite Nuclear Safety PIR Plant Improvement Request RAIL Regulatory Action Item List SDAFW Steam Driven Auxiliary Feedwater S/G Steam Generator SSFI Safety System Functional Inspection