ML20059J598
| ML20059J598 | |
| Person / Time | |
|---|---|
| Site: | Fermi  |
| Issue date: | 11/04/1993 |
| From: | Rosenthal J NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD) |
| To: | Holahan G NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD) |
| Shared Package | |
| ML20059J601 | List: |
| References | |
| NUDOCS 9311120281 | |
| Download: ML20059J598 (5) | |
Text
{{#Wiki_filter:7 ~___..__.-.a a ~ r neg y* UNITED STATES j ,y NUCLEAR REGULATORY COMMISSION $5 WASHINGTON, D.C. 20555 4001 %...../ NOV 4 1993 MEMORANDUM FOR: Gary M. Holahan, Director Division of Safety Programs Office for Analysis and Evaluation of Operational Data FROM: .ack E. Rosenthal, Chief Reactor Operations. Analysis Branch Division of Safety Programs Office for Analysis and Evaluation of Operational Data
SUBJECT:
HUMAN PERFORMANCE STUDY REPORT - FERMI UNIT 2 (8/13/93) On August 13,1993, Fermi Unit 2 was operating at 93 percent power, when an equipment operator removed tape residue from a reactor instrument calibration / vent valve shaft and caused a spurious reactor pressure vessel (RPV) Level 8 signal. This lead to automatic trips of the turbine, main generator, feedwater pumps and reactor. Although a reactor operator (RO) initiated the standby feedwater system, the reactor water cleanup system (RWCU) isolated, recirculation pumps tripped, and the high-pressure coolant injection (HPCI) and reactor core isolation cooling (RCIC) systems actuated automatically as reactor water decreased past Level 2. There were many operator distractions during the post-event recovery, including: several equipment failures, smell of burning insulation in the control room, and about 100 control board lamps burnt out either prior to or during the event. Following the reactor scram, operators did not take manual actions required to operate the gland seal steam, resulting in loss of condenser vacuum and subsequent main steam isolation valve (MSIV) closure. The shift supervisor declared an Unusual Event due to HPCI injection. For about 1-% hours, operators maintained reactor pressure control by. cycling nine safety relief s ves (SRVs) and using standby feedwater and condensate booster pumps for RPV level control. The shift supervisor assigned a low priority to restoring RWCU until about 1-% hours into the event. RPV thermal stratification caused the RPV temperature difference to exceed the technical specification (TS) limit of 146* F and prevented restarting the recirculation pumps. The couldown rate exceeded the TS limit of 100* F/hr by several degrees. l As part of the AEOD program to study the human performance aspects of operational events, a team was sent to the site on August 17,1993, to study the August 13,1993, event. The AEOD team leader was Robert Spence; other team members were I Dr. Susan Hill, David Prawdzik, ani William Steinke of the Idaho National Engineering d Laboratory ONEL). Thg team was casite for 2-% working days and gathered data from U/ 9311120281 9312'J4 PDR ADOCK 05000338 P PDR
m Y ,.. = Gary M. IIolahan discussions, plant logs and recordings; review of operations and training materials; and interviews with control room operators and other station staff. Enclosed is the report prepared of the results of the human performance study. Specific human performance aspects of this event are summarized in this memorandum: ~ Procedures and Training Although the gland seal steam system is intended for automatic control, the system had been operated manually for the past 4 years because of a design problem. Neither the plant procedures nor simulator training were revised to reflect manual operation of this .j system. Manual operation of the gland seal steam system was not addressed in reactor or turbine trip response procedures. Teamwork l The licensee had no guidance for conduct of volunteers reporting to control room shift supervision, receiving instructions, or waiting quietly in standby. Clear lines of control board responsibility were not always delineated when more than two ROs were used. In l this event, the third RO took over many of the actions of another RO. This initially i resulted in short-term, ineffective operator performance. The licensee did not provide j training or guidance on how extra ROs should assist dming an event. i i An operator stated that failed equipment rates have recently improved. Ilowever, the -l number of operator distractions from various equipment problems was high in this event. r November 1992 Event The AEOD team learned of a November 1992 loss of feedwater and reactor trip that f offered a comparison between operator response in two similar events. Therefore, the team interviewed the principal operators in that event also. In this earlier event, a loss j of feedwater caused a reactor trip and s..silar RPV Level 2 actuations, but prompt operator action restored gland seal steam and mechanical vact.um pumps to maintain condenser vacuum. Thus, the MSIVs did not close and SRV cycling was not necessary. l This event was less complicated than the August 13,1993, event because of successful operator actions. i This event was handled more successfully by another crew through better teamwork command, control, and communications; and experience. The operatons throttled IIPCI/RCIC/ standby feedwater flow instead of allowing Level 8 trips and throttled CRD 'i flow to minimize thermal stratification and RPV cooldown rate. RWCU was restored l within I hour of the scram. l i This crew had fewer distractions, such as alarms, burnt out light bulbs, or additional ROs coming into the control room. The NASS primarily relied on only 2 ROs, as in the ) . simulator, and used the 3rd RO for discrete duties. i ~
pary M. Holahan P P However, an RO took HPCI out of service by placing the IIPCI auxiliary oil pump in a " pull-to-lock" for about 15 minutes after its RPV Level 2 initiation signal cleared, necessitating operator action for potential reactuation,instead of immediately restoring it to standby. j The licensee's post trip or human performance evaluations of this event did not identify any of the issues addressed above. Summarv The two events studied were, overall, benign. The safety significance of the August i event was increased because MSIV closure resulted in loss of the normal heat sink and .l could have lead to a stuck open SRV. 1 r During the August 1993 event, loss of the gland seal steam and condenser vacuum resulted in MSIV closure and SRV pressure control. MSIV closure occurred in the l August event because the operators did not respond to the turbine panel annunciators. In the November event, the low condenser vacuum alarm prompted operator action. l Licensee procedures and simulator training did not address manual operation of the gland seal steam system or timely restoration of RWCU and CRD throttling after a loss of forced circulation. During the November 1992 event, an RO took HPCI o'11 of service, necessitating operator action for potential reactuation,instead of ihediately restoring it to standby. The licensee's human performance evaluation of the event had not identified the need for guidance, simulator fidelity, or simulator training on gland seal steam, RWCU and CRD system operation, and natural circulation after a reactor trip. This report is being sent to Region III for appropriate distribution within the region. [ sen e Reactor Operations 1.alysis Branch [ Division of Safety Pregrams. Office for Analysis and Evaluation of Operational Data ~ ~
Enclosure:
As stated cc w/ enclosure: Mr. Doug Gibson, Senior Vice President -Nuclear Generation Detroit Edison Company 6400 North Dixie Highway Newport, MI 48166 -{ Distribution: See a uched list
- See p vious concurrence R
C:ROAB f k JRose a Cj#[/93 11/ 11/9 /93
Gary M. Holohan ' Summary The two events studied were, overall, benign. During the August 1993 event, unnecessary loss of the gland seal steam and condenser vacuum resulted in unnecessary MSIV closure and SRV pressure control. ; Licensee procedures and simulator training did not address timely restoration of RWQ, CRD throttling or natural circulation to minimize RPV thermal stratification after a loss of forced circulation. i The comparison of the thermal stratification data between the two epents promptedThe licensee to question-General Electric regarding the accuracy of-th/ bottom head fluid hy temperature measuremennddahe stratification mechinism, timpg, and extent. The - j pO' recently developed calculational-meth~oDIhermal tifica ' n should result in a - more thorough understanding of this phenomenon and appr peQ 3 responses.-Th6RC should follow this analysis with intere t. g Analysis of this_atid other events suggests that the NR ahould encourage simulator requalification training to7ncompass-awarg6f cre sizes and a few examples of gNs varying degrees of aberrant ~ behavior to better rEpyesent-ahaQperators may encounter. a g p dun ualdEnts. During the November 1992 event, an RO to ' HPCI out of service, after its RPV t dg operator action for potential reactuation, Level 2 initiation signal cleared, necessita)dby. The licensee's human performa instead of immediately restoring it to stan evaluation of the event had not identifId the need for guidance, simulator fidelity, or simulator training on gland seal stea, RWCU and CRD system operation after a. reactor trip. This report is being sent to Rey, ton III for appropriate distribution within the region. Jack E. Rosenth ' Chief Reactor Operations Snalysis Branch' Division of Safety Programs Office for Analysis and Evaluation of Operational Data
Enclosure:
stated cc w/encio re: Mr. Dou/Gibson, Seniorj@ Edison Company ice President -Nuclear Generation Dettpit 6400 North Dixie Highway Ndwport, MI 48166 / Distribution: See attached list ./ / ROAB Alf6 ROAB C:ROAB RSpence:rgz GLanik JRosenthal 10/d/93 10/ /93 10/ /93 . =_
y o "g.- -Gary M. Holahan Distribution: DCD/ Central File ' PBaranowsky_ REckenrode, NRR KRaglin, TTC Public ~ ' JKauffman WKropp, SRI ROAB R/F _ ETrager _ EGreenman, Rill RSpence JIbarra FCoffman, RES GLanik AChaffee, NRR JLieberman, OE JRosenthal TColburn, NRR J Segala, OE EJordan LMarsh, NRR WSteinke, INEL Dross PEng, NRR Shill, INEL - VBenaroya FHebdon, NRR DPrawdzik,'INEL LSpessard CThomas, NRR MPhillips, RIII KBrockman BBoger, NRR RSavio, ACRS SRubin WDean, NR R MTaylor, EDO i L}}