ML030640563
| ML030640563 | |
| Person / Time | |
|---|---|
| Site: | Point Beach  |
| Issue date: | 02/06/2003 |
| From: | Kendall T - No Known Affiliation |
| To: | Office of Nuclear Reactor Regulation |
| References | |
| FOIA/PA-2003-0094 | |
| Download: ML030640563 (5) | |
Text
SUMMARY
OF POST-TMI AFW LICENSING CORRESPONDENCE This summary was compiled by T. C. Kendall after reviewing the cited correspondence and is solely his work. It is intended as a starting point for further research or review, and does not constitute the licensee's position, or make any representations for the licensee. The compiler's notes and observations are inserted in italics.
5/11/79 NRC and WE representatives meet for a detailed discussion of the AFW system at PBNP. Minutes are transmitted in letter on 5/15/1979. The list of materials requested/provided is comprehensive, and includes system descriptions, power supplies, drawings, procedures, etc. The scope appearsto have been equivalent to an SSDI or EDSFI. It appearsthat the informationprovidedformed the basisfor the 9/21/1979 site specific requirements.
9/21/79 NRC transmits site specific requirements for AFW system to PBNP. Enclosure 1 contains a system description, reliability evaluation, etc. developed by the NRC during the course of the NRR Bulletins and Orders Task Force review. Of significance:
Page 2: "SWS supply is initiated in the control room by a opening motor operated valve in the SWS to each AFW pump suction. The system is arranged such that a failure of either of the two diesel generators on site will not prevent water from being supplied to the AFW system for either unit." Because the one remainingmotor driven pump can provide flow to either unit, this is a true statement. Flow can be supplied to either unit. However, it is misleadingin that it could not (andcannot) be providedto both units at the designflow rate simultaneouslywithout invoking manual action outside of the control room.
This appearsto have been an oversight at the time of the review, and it has carried forward to the present time. Because the event of concern necessarily entails a loss of off site power (andtherefore reliance on the EDGs), it is, by the definition now containedin the FSAR, a dual unit transient.
All 4 MOVs have always been poweredfrom vital AC busses. Since there were 4 valves and only two EDGs at the time of the writing, the only way that the statement could be true is if the SWS MOVs of one MDAFP and one TDAFP were driven off of each vital bus. Initialresearch of the modification history shows this has never been the case; 3 of the MOVs have always been powered off of the B train of vital power. Since the installationof two additionalEDGs (mid-1990s), the reliabilityof the facility in this respect has been significantly improved because the normal alignment separatesthe three SWS MO Vs powered by the B train into two separatebusses suppliedby separate EDGs.
Ideally, the power suppliedto the SWV valve associatedwith each turbine driven AFW pump should derivefrom separatetrains. Just which valve should derivefrom which train will depend upon the other supportingpower requirementsfor the AFW pumps.
In the short term, it isjudged that ample time is availableto locally open the associated pumps' SW supply MO Vs.
Page 3 (X.11.1.1): "Since all valves in the flow path to the steam generators are normally open and fail as-is (with the exception of PCV-1 and 2 which fail open) a loss of A-C or D-C power does not require valve manipulation...". Here is it apparentthat the concern wasfor deliveringwater to the steam generators,and that the recirculation line AOVs (whichfail closed) were not of concern. Furthermore,a loss ofDCpower was consideredacceptableprovided thatflow to the steam generatorswas not precluded or interrupted This conditionstill exists; while a loss of DCpower would cause losses of various indicationsandfunctions, andgreatly complicate diagnosisand mitigation, it would not interruptor preclude the ability tofeed the steam generators from AFWV Page 4 (X. 11.1.3): "Power sources for all instrumentation and controls are taken from the emergency buses which are supplied by the safety related diesel generators or safety related station batteries". Even when expandingthe scope to include the recirculation A OVs (which apparently were not consideredat the time), this statement remains true.
A subtlety is that the recirculationline AOV logic has a portionthat, while powered from the safety relatedEDGs, is routedthrough some non-safety grade circuitry.
Page 4 (X. 11.1.4.1): "All controls for the active components of the auxiliary feedwater system can be operated from the control room". The minimum flow recirculationline AOVs have never been controllablefrom the control room. Only valve position indicationis in the control room. Therefore, it appearsthat the review did not consider the minimum recirculationlinefunction essentialfor system operation/operabilitygiven the concerns at the time.
Page 14 (X. 11.3.1.6; GS-7): "The licensee should verify that the automatic start AFW system signals and associated circuitry are safety-grade". Here, and in the subsequent sub-items, the emphasis is placedsolely on automatic initiationof the system, not subsequent manual control or manipulationof the system.
Page 16 (X. 11.3.2.1): "The licensee should provide redundant level indications and low level alarms in the control room for the AFW system primary water supply to allow the operator to anticipate the need to make up water or transfer to an alternate water supply and prevent a low pump suction pressure condition from occurring. The low level alarm setpoint should allow at least 20 minutes for operator actions, assuming that the largest capacity AFW pump is operating".
Page 18 (X.1 1.3.3; GL-4): "Licensees having plants with unprotected normal AFW system water supplies should evaluate the design of their AFW systems to determine if automatic protection of the pumps is necessary following a seismic event or a tornado".
It is clearfrom these two recommendationsthat the NRC consideredmanualactionfor longer term activities (i.e. those longer than 20 minutes) acceptable,provided that there is adequate instrumentationavailable to the Operatorfor timely recognition and action. However, those situationsthat may create short-term hazards to equipment must have automaticprotection/preventionschemes. One of the options suggestedfor plants with unprotectednormalAFW supplies is an automatic low suctionpressure trip of the AFW pumps. This implies that automatic trippingto protect the pump,followed by a manual action to switch over the suction source was consideredacceptable (this was the solution eventually implemented by PBNP).
10/29/79 WE responds to 9/21/79 NRC letter with various commitments and a status report. It is brief (only 4 pages), and cites the comprehensive nature of the actions and information requested, and the short response time given. Further updates committed. Of significance:
Page 3: In defending the existing configuration of CST level configuration (and refuting the need to install redundant level instruments on each tank), PBNP states that "the water supply for each operating [AFW] pump can be shifted from the primary source to the secondary source by opening a single motor operated valve controllable from the control room", and that "...because of the ease with which the AFW pump water supply can be shifted to the secondary source..." There appearsto have been no consideration,explicit or otherwise, of the potential consequences of a loss of a vital bus in these statements.
12/17/79 WE commits to install additional AFW flow instrumentation (for flow to the steam generators) and to have an Technical Advisor.
2/4/80 WE responds to deferred items, and updates status on active items from 9/21/79 NRC letter. Of significance:
Page 3 (Recommendation GL-4): "The shifting to service water is done by operation from the main control room of one motor-driven valve per pump. Automatic switchover is not necessary on the Point Beach design. The addition of automatic pump trips on low suction pressure has been evaluated as unnecessary...". This began a dissent with frequently contentious tone that continuedforsome time untilfinal resolution.
5/16/80 NRC responds to previous WE correspondence. Finds PBNP response unacceptable on several counts:
GS-1: Must require all 4 AFW pumps to be Operable for 2 unit operation, and 3 pumps (both motor driven and the associated turbine driven) for single unit operation.
GS-5: Emergency procedures must address cooling for TDAFW pump bearings.
GS-6: Requires independent verification of system alignments following testing or maintenance.
GL-3: Proposed modification to supply TDAFW bearing coolers from fire water is unacceptable. "No credit will be given for any operator actions outside of the control room for two hours". NRC position was based on a misunderstandingthat additional manual action would be necessaryfor activation of the fire water cooling. However, the citing of 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />sfor actionsoutside of the control room appearto be extreme, especiallyfor a station blackout event. This limitation does not appearanywhere else, and appearsto have been obviated by the more extensive station blackout rule.
GL-4: NRC re-asserts that protection for the AFW pumps must be provided in the event of a loss of the CST due to a seismic/tornado event.
Page 4 (item D): NRC reasserts that PBNP must submit AFW system flow requirements. PriorPBNP response was that ample information was already available from similarplants, and that PBNP need not provide plant specific information.
7/8/80 WE replies to 5/16/80 NRC letter. Maintains that automatic switchover of AFW pump suctions is not needed, and that the CSTs are seismically rated. No mention is made of the non-seismic structures located adjacent to or above the CST.
11/19/80 Intemal WE memo documenting various technical questions from NRC on AFW. No significant new issues/questions.
1/27/81 NRC issues SER with open items for those that are still incomplete or under contention.
No new positions or resolutions.
4/9/81 WE responds to SER. Commits to install redundant level instrumentation for each CST, and to install low suction pressure trips on the AFW pumps.
7/28/81 NRC issues an RAI limited to the automatic initiation and flow instrumentation for AFW (TMI Action Plan Item II.E.1.2).
9/16/81 WE provides specific responses to RAI of 7/28/81. No new issues identified.
5/21/82 NRC requires submittal of proposed changes for recommendation GS-1 (Technical Specifications upgrade) within 45 days.
5/3/82 NRC issues an SER contingent upon completion of committed modifications & TS changes. SER concludes that, based upon a review of the Franklin Research TER (also enclosed), the PBNP AFWS automatic initiation and flow indication systems comply with the staff's long term safety grade requirements. Note that the emphasis has remainedon automaticinitiation andflow indication.
7/27/82 NRC issues TS amendments to resolve GS-1.
2/11/83 WE updates NRC on status of modifications. No new items.
3/10/83 NRC internal memorandum recognizes and elucidates an operability dilemma posed when feeding one unit with a motor driven AFW pump. The corresponding pump's flow path to the other unit must be isolated, rendering that flow path inoperable. This is a new concern "created" by the new TS requirement that all 4 AFW pumps must be operable whenever both units are in operation. This was an unforeseen consequence of the TS upgrade.
3/24/83 WE dockets correspondence documenting discussions of AFW operability when feeding one unit with a motor driven pump while the other unit is required to have an operable pump. A change to the Tech Specs is proposed.
5/4/83 WE provides payment for the license amendment approval fee, and requests a response to the previous correspondence of 3/24/83.
6/20/83 WE proposes a hardware modification that will automatically align the MDAFW Pump isolation valves to the affected unit and isolate it from the unaffected unit upon receipt of an auto-start signal.
9/15/83 NRC concurs with the proposed modifications.
11/11/83 WE notifies NRC of completion of modifications to permit testing of AFW actuation circuitry.
7/6/84 WE notifies NRC that the modifications proposed in the 6/20/83 letter are completed.
2/24/86 INPO issues a report "Reliability of PWR Auxiliary Feedwater Systems". Among other items, it deems installing automatic low suction pressure trips on AFW pumps a "poor practice" because of the potential for spurious trips due to simultaneous starts of multiple pumps. PBNP implementation includeda time delay to permit transient effects of a simultaneous start to pass before initiatinga trip.