ML20236B941
| ML20236B941 | |
| Person / Time | |
|---|---|
| Site: | Beaver Valley |
| Issue date: | 09/30/1987 |
| From: | Lester Tripp NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML20236B936 | List: |
| References | |
| 50-412-87-54, NUDOCS 8710260422 | |
| Download: ML20236B941 (21) | |
See also: IR 05000412/1987054
Text
{{#Wiki_filter:_. ._ . . U. S. NUCLEAR REGULATORY COMMISSION REGION I Report No. 50-412/87-54 1 Docket No. 50-412 License No. NPF-73 Licensee: Duquesne Light Company Nuclear Group P. O. Box 4 Shippingport, PA 15077 l Facility Name: Beaver Valley Power Station, Unit 2 l Dates: July 11 - July 30,1987 and August 8 - August 28, 1987 Inspectors: l J. E. Beall, Senior Resident Inspector L. J. Prividy, Resident Inspector S. M. Pindale, Resident Inspector, Unit 1 ( S. Barber, Reactor Engineer (Examiner), Region I . . 30 87 Approved by: / L. E. Trim , Chief, Reactor Projects / Da'te Section 3A Inspection Summary: Inspection No. 50-412/87-54 on July 11 - July 30, 1987 and August 8 - August 28, 1987. i , Areas Inspected: Routine inspections by the resident inspectors (474 hours) of licensee actions on previous findings, site activities, startup test program , I implementation, misalignment of instrument isolation valves, reactor coolant pump underfrequency relays and turbine trip ci rcui try , licensed operator training program, security program and review of LERs. Results: During the inspection period, the licensee received a full power i license, achieved initial criticality, completed low power physics testing and i conducted some portions of the power ascension testing phase. Two unresolved l items were identified as a result of inspection concerns subsequent to two reactor trips (details 5.4.1 and 5.7). A third unresolved item was opened to address identified weaknesses in the licensee's security program. There were no violations. 8710260422 h h 12 PDR ADOCK PDR G - _ _ _ - _ - _ _ _ _ _ _ _ - _ _ - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ A
. ,
j DETAILS , 1. Persons Contacted During the report period, interviews and discussions were conducted with members of the licensee's management and staff as necessary to support inspection activities. . 2. Project Status Summary l This inspection overlapped Inspection Report 50-412/87-56 which was a l special inspection conducted during the approach to initial criticality l and low power operations from July 31 - August 7, 1987. During the in- ' spection period, Unit 2 entered Modes 4 and 3 (July 15 and July 17, 1987, respectively) and initial criticality (Mode 2) was achieved at 6:25 a.m. on August 4, 1987. The licensee completed low power physics testing and received a full power license on August 14, 1987. Entry into Mode 1 (>5%) was accomplished on August 15, 1987. The generator was placed on the grid on August 17, 1987. For the remainder of the inspection period, the licensee was conducting initial startup testing up to the 30% plateau. Approximate dates for power ascension testing, as last announced by the licensee are listed below. I - Station Blackout (30%) September 8, 1987 - Load Rejection Trip (50%) September 10, 1987 - Achieve 100% Power October 4, 1987 - Turbine Trip (100%) October 9, 1987 - Load Rejection Trip (100%) October 10, 1987 - Commercial Operation October 17, 1987 l l 3. Inspection Program Status Summary Preoperational test program inspection is complete. Inspection of the licensee's startup and power ascension program is continuing with recent l activities of test witnessing and test results review occurring (Inspec- tion Reports 50-412/87-52 and 55). AREA % INSPECTION COMPLETE Overall Program 60- Procedure Reviews 90 Test Witness 60 Results Review 40 _ _ _ _ _ _ _ _ _ _ _ . _. . _ _ _ _ . _ --_ __ - -
. .. .q . . ! l l 2 l ) The current inspection status is consistent with the applicant's startup program progress. At the end of this inspection period, there were , approximately 20 open NRC inspection items as listed below i NO. OF OPEN INSPECTION ITEMS TYPE OF ITEM END OF THIS PERIOD END OF LAST PERIOD i Bulletins 2 2 ~ i 1 1 Violations 1 1 Deviations 0 0
Construction Deficiency i Reports 0 1 . I Unresolved 17 21 l i ' TOTAL 20 25 4. Licensee Actions on Previous Inspection Findings j (Closed) Unresolved Item (86-13-03): Ultrasonic examination report form did not specify the acceptability of the inspected weld. .This item was last updated in NRC Inspection Report 50-412/87-47 and left open pending inclusion of a revised form in General Procedure 101, " Requirements for Data Recording." The inspector reviewed GP-101, Revision 3, and noted that the new form, previously reviewed, had been inc cporated in this revision. This item is closed. (Closed) Unresolved Item (85-26-01): Information' Notices 85-52, 83-72, and 84-78 identified deficiencies associated with environmental qualifi- < cation test failures, including some involving Limitorque operators. Licensee response to these notices included replacing limit switches, replacing motors having unqualified insulation, rep 1_ acing unqualified terminal strips, replacing improper electrical connectors with ring lugs, i j and inspecting all Limitorque operators. The licensee did not provide an evaluation to support an initial conclusion that these apparent defects ! I were not reportable in accordance with 10 CFR 50.55(e). The inspector reviewed additional Nonconformance and Disposition Reports 15995 through 16007 which documented inspection of numerous Limitorque operators result- ing from the initial Report of a Problem (ROAP) and Construction Deficiency Report 86-04 which documented the above. The inspector had no further questions; this item is closed. 1 1 l w
. . 3 l (Closed) Construction Deficiency Report (87-00-14): Lack of documentation certifying isolating a reactor coolant system (RCS) loop while in Mode 5
or 6. The RCS design includes isolation valves intended to allow draining of a loop of the RCS without lowering reactor vessel level with the core installed. The licensee was notified by the RCS vendor of a documentation problem which potentially impacted the safety analysis of operating the RCS with the loop stop valves shut. In lieu of the supporting documenta- tion, the licensee deleted that configuration from their license request. The current technical specifications do not allow RCS loop stop valve closure with any nuclear fuel present. The licensee intends to correct ' the documentation problem and submit a change request to the technical l specifications at a future date. This item is closed. 1 (Closed) Construction Deficiency Report (87-00-18): This item involved the discovery, during testing, that certain safety-related fans could not be . started from the alternate shutdown panel. Use of spring--eturn switches prevented the seal-in of the start signal. The switches were replaced by two position switches and testec; satisfactorily. This item is closed. (Closed) Unresolved Item (87-02-01): Demonstration of correct low press- ure operation of RCS-PCV-4550 and 456. During preoperational testing, these valves failed to stroke at low pressures of 80 - 200 psig. Disas- sembly revealed no significant damage and retesting was conducted after modifications were made to increase the orifice diameter on the bottom part of the solenoid actuator. Inconsistent stroke times were again i I experienced during Mode 5 testing. The licensee removed the orifices entirely and the valves stroked satisfactorily but with faster cycle times than assumed in the piping loads analysis. The piping analysis was reperformed and the increased stress loads were found to be within the calculated design limits of the existing hardware. The inspector had no further questions; this item is closed. 1 ' 5. Site Activities Throughout the inspection period, the inspectors toured the licensee l facilities. General work activities were observed including surveillance, testing and maintenance. The inspectors also monitored the licensee's housekeeping, security and radiation control activities. In particular, the inspectors monitored the licensee's progress towards meeting the pre- requisites to conduct power ascension testing at increasingly higher power plateaus. Emphasis was made on tech spec compliance and licensee correc- tive. action and overall response to the reportable events that occurred. i , I . _ . _ . _ . _ _ _ - . _ _ _ _ . _ _ _
, - , - ,- - - - - - . Y ' .. , \\. J; 'NI ' ,', ' i r ' % , 'g- j ' ., , - 4,o O i s. . , , . 5.I' ' Inoperable _ Mash Steam Isolation Valves N ,s . On July 17,3 af ter stroking all three main steam .'i solition valves- l (MSIVs),' the. licensee declared t)pt ' A' and 'C' MSIVs :inoderable when I - their actuatW arms traveled only L2 of_ the required.14. inches be- i tween the apen and closed limit svitches. The problem was. attributed- l to mechanical binding. in .the dive act0ator section between the actuator stanchion. and - the. stanctdon .' guides. The licensee took. 'immediate corrective action ' to restore all MSIVs to L an . operable - status by lubricating the interfering. surfaces _ and adjusting :the stanchion guides tU provide additional-- clearance between - the guides. and the actuatort st'anchion. Subsequently ( o r. Ju.ly 20)' 'all .MSIVs demonstrated satisfactory closure performance at design: conditions in i accordance" with Test Procedure. P0-2.21A.03,. "MSIV Operability iTest." ) It should be noted ,that- this vas 'the first heatup with these valves installed as ' the Y pattern va.nes areplaced the original ball-type, valves after hot fanctional ' testing was completed. The resident inspectors will review riSIV performance following' the next cooldown- heatup cycle. 3 No deficiencies in ths licensee corrective. actions were identified. 5.2 Pressurizer Safety Valv);ulifts ' On July 25, while in Mode 3, pressurizer safety valve, 2RCS-RV551C, lifted twwe at RCS pressures (2175 and 2180 psig) significantly below the ipecified set pressure o' 2485 PSIG - 1% . The valve reseated properly, the licensee On. ared 'it inoperable and a plant cooldown to Hot Shutdown (Mode 4) y ditions was . immediately commen- ! ced. A containment entry was made and the bolt'used for safety valve ' set pres sure .. adjustment was found to be improperly, secured. The valve set pressure was adjusted an'd the valve was satisfactorily retested and declared operable on July 26. At this _ time, the licen- see also checked the 'A' and 'B' pressurizer safety valves and deter- mined that their set pnssures were correct and" their set pressure adjusting belts were properly secured. While investigating the pressurizer safety valve' status, the licensee ' noted that the 'B' and 'C' . safety valves appeared to .bc. . leaking as - - evidenced by their tailpipe temperatures.of approximately 240 250F, ' The 'A' safety valve tailpipe . temperature read, approximately '180 - 190F which ' indicated that it was- not leaking. During.the inspection period,. the inspector noted no_ substantial upward trend in these temperatures or in. identified RCS leakage. '_The licensee ' has - pur- chased new . safety valves which' are on ' site and scheduled- to be installed in place of 'B' and 'C' in an. upcoming one week maintenance: .' outage. No violations were identified. _ _ _ _ _ - _ . - - _ .
F - ] " g u 5 - 5.3 . Reactor Trips from Dropped Control Rods j On August 7, 1987, a manual reactor trip . was . initiated when the , reactor was in Mode 2 due to four ' dropped control rods'(Control Bank ~ ' . l D, Group 2). .This reactor. trip is discussed in further detail ~in' NRC . Inspection Report 50-412/87-56. The licensee subsequently- replaced the suspected circuit card 'with a moveable gripper'. coil for'. the affected control ' rods. The reactor-was . subsequently taken critical- at 4:38 am on August 10 to resume initial startup testing. However, ! at 4i43 am on. August 10, the same four control rods' (Control. Bank D, ! Group 2) fell into the core. . The- reactor was 'again. manually tripped as per Abnormal Operating Procedure No. 2.1.5, Dropped RCCAs. Plant systems responded as' designed d Ping the event. The NRC was notified of the event per 10 CFR 50.72 reporting-requirements. 1 Since this event was identical.to t'he. August 7th reactor trip, it'was'- i apparent that the root cause of the -event had not ' been identified. Based upon both previous Unit 1 experience and . vendor - recommenda- tions, however, the suspected circuit card had been'a likely cause to J the rod control system anomalies. Since no facilities existed to l test the card on site, it was replaced and sent off site-for testing. The testing results later identified that the suspected card 'was undamaged. Additionally, following the circuit card replacement,' the licensee wa able to successfully move the affected control rods. ! Following the second event, the licensee did not plan to restart the reactor until the source-of the problem was identified and corrected. After extensive troubleshooting and rod exercising,- an intermittent malfunctioning silicon controlled . rectifier (SCR) was identified as the apparent cause of both reactor trips. The SCR was found to fail i intermittently' and after a " soak" time of.several minutes. The fail- j i ure was subsequently reproduced in the B control rod bank'by jumper- ' l ing the affected SCR. The SCR was replaced'in the rod control system and was subsequently tested satisfactorily- on August 11, 1987.' Cri- ! ticality was subsequently achieved later that day to perform in plant radiation surveys to gather baseline data. No further'similar events occurred during this inspection. No deficiencies in the licen sees final corrective actions were i identified. 5.4 Reactor Trips Due to Low-Low Steam Generator Levels 5.4.1 On August 15, at 2:37. am, an automatic reactor and turbine trip occurred. from approximately 5% thermal power due to low-low water -level in. the "A" steam generator .(SG). Plant- operators had initiated an increase in reactor power, how- ever, the steam dump system did not respond while control- ling automatically in the steam pressure mode' of operation. Therefore, the plant operator placed the steam. dump system . _ _ - . _ _ _ _
r- . . 6 i f r. manual and inserted a 3% demand on the system. The steam dump system was then returned to automatic and the demand quickly increased to about 18%. In response to the relatively large steam flow, the SG water levels " swelled" in all three SGs, and then rapidly decreased. During the ! transient, the feedwater level control syst ca was in the > l manual mode of operation, controlling with the feedwater regulating bypass valves. After the decreased SG water l level was recognized, plant operators attempted to increase j feedwater flow, however, SG water levels continued to drop j due to the addition of relatively cold feedwater. The i reactor tripped when the "A" SG reached its low-low water t level setpoint. The licensee subsequently determined that a i contributing cause for the trensient was that during steam ' dump system operation, the reactor operator manipulated control rods in an attempt to minimize the secondary plant
transient. However, in the existing plant configuration, the controlling rods (Bank ..D) exhibited unusually high worth, and primar> system temperature changes were exag- gerated due to the control rod manipulation. Therefore, control rod movement under that plant configure. ion, amplified the transient. This phenomenon concerning con- trol rod high worth was not initially identified by the licensee as a contributing cause to the event, however, high control rod worth was specifically noted as a root cause to the reacter trip later on August 15. For details concerning the nature of the high control rod worth problem and licensee corrective actions, see Section 5.4.2. The resident inspector observed trip recovery actions. j Operator response to the trip was in accordance with plant j procedures. This event was reported via ENS per 10 CFR 50.72 reporting requirements. During the event followup investigation, the licensee de- l termined that the turbine trip initiation from the reactor ! trip as indicated from the computer printout showed that the turbine trip signal exceeded the maximum expected delay interval. The actual time for initiation of the turbine trip due to the reactor trip signal indicated .711 seconds on the computer printout versus 'a .167 second vendor spec- ification. The licensee subsequently reviewed both Unit I and Unit 2 plant drawings. For Unit 2, while the signal is generated on the reactor trip breakers and bypass reactor trip breakers, the computer input is driven by auxiliary relays. However, auxiliary relays are not used at Unit 1, but the required time for the turbine trip from reactor
I m _ ___ - -__ D
R E ( - ~ l 1 I 7 trip signal is the same for both units (.167 seconds). In response to the above, the licensee initiated an engineer- irg memorandum (EM), No. 62552, to address and resolve the discrepancy. The licensee's initial engineering review concluded that there was no immediate safety concern, how- t ever, the issue will be further investigated. Pending ! resolution of the timing discrepancy and review of the licensee's evaluation of the potential safety concern, this item is Unresolved (50-412/ 87-54-01), s.4.2 On August 15, at 9:33 pm, an automatic reactor trip occur- '. red from approximately 20% power due to a low-low water level in the "A" steam generator. The low-low water level was reached following an attempt by plant operators to initiate a power increase by withdrawing control rods about 3-1/2 steps. Immediately following the control rod with- drawal, steam dump system . valves opened, and the steam r generators exhibited a slight level " swell", followed by a sharp level reduction. The feedwater level control system was operating in manual. Operators responded to the sharp level reduction by increasing feedwater flow to the SGs. However, the addition of relatively cold feedwater, in con- junction with the low water levels, did not prevent reach- { ing the low-low water level reactor trip setpoint. Plant ^ systems responded normally following the reactor trip. The licensee made the required notifications for the event. Several contributing factors led to this reactor trip. The plant configuration at the time (Bank D control rods at about 20 steps) provided unusually high control rod worth not normally encountered during power operations (about 3 times higher than normal) and the moderator temperature coefficient (MTC) worth was lower than would be experienced at full power operation. Consequently, any power and/or temperature changes were exaggerated due to control rod high worth, and minimal dampening effect occurred due to current MTC values. Other factors included a relatively low initial steam generator level (about 30% narrow range), and a relatively cold feedwater, whose addition inherently causes level shrinks. The licensee.- subsequently directed plant operators to ,nintain SG water levels at approxi- mately 50% to allow for anticipated level shrinks while operation'at low power with the feedwater system operating in manual. Also, the operators were cautioned not- to wait for movement of the SG 1evel traces .to initiate feedwater flow adjustments when level changes are expected. There- fore, anticipated water level changes could be adjusted and maintained and plant operators would not have to ' chase' - _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ - _ - - _ - _- _.
. . 8 1 the level changes. Other licensee corrective action for j this event included notifying all nuclear control operators l about the high control rod worth with Control Bank D l inserted. Additionally, cautions were documented and pro- ! vided to plant operators concerning awareness to the l unusual conditions and how to deal with them. No addi- l tional operational events of this nature recurred, i No violations were identified. 5.5 Auxilia ry Feedwater Pump Auto-Start Due to Improper System Lineup On August 16, while in Mode 2 at 2% power, an automatic start of the l 'B' auxiliary feedwater (AFW) motor driven pump occurred. The' pump l automatically started during the performance of an Operations Sur- veillance Test on the turbine driven AFW pump (TDAFWP). Ten seconds following the start of the TDAFWP, the motor driven pump automati- cally started due to low turbine driven pump discharge pressure. l Each (of two) motor driven pumps are designed to automatically start ' following a predetermined time following a start signal of the TDAFWP coincident with low discharge pressure. The associated 'B' pump pressure root stop valve switch (2FWE-PS1588) was found to be valved out, therefore, preventing detection of the discharge pressure. On August 17, the licensee had completed additional safety related instrumentation valve lineup checks and determined that additional instrumentation valves were incorrectly positioned. Flow switch, QSS-FIS1028, was valved out. This flow switch provides a no-flow signal for stop valve 20SS-SUV102B to close to isolate the chemical addition tank from the refueling water storage tank and thus prevent cross-contamination upon loss of quench spray flow. Also, pressure instruments (SIS-PT101A&B) which provide computer alarms for low head safety injection pump low discharge pressure were found to be valved out. While these valve lineup checks were being conducted, an unre- lated reactor trip occurred and in view of the questionable nature of the instrumentation valve lineups, the licensee committed to make extensive additional safety system status checks prior to startup. These extensive additional checks included lineups of instrument root l and isolation valves and instrument and control switch alignment for safety systems such as auxiliary feedwater, emergency diesel gener- ators and low head safety injection. These extensive additional l checks were completed on August 19, and no other significant discrep- ancies were found. l
. . ! 9 , i Concerning the root cause offthe problem, the licensee concluded that I the discrepancies probably occurred during the preoperational test H program prior to full implementation of permanent station work pro- cess controls. The inspector noted that 2MSP-4.03-1 "ESF and Mis- cellaneous Safety-Related Instrumentation Valve Alignment and Cali- bration Verification" had not been performed when changing from Mode: 5 to Mode 4 since the existing' revision.(Rev. 0) of this MSP was not considered technically accurate and hence, not worthwhile to' perform. l A similar MSP is used at Unit 1. to assure proper valve alignment of ! key ESF and miscellaneous safety-related instruments after an l extended outage. . Implementation of 2MSP-4.03-I. might have prevented I the misalignment of pressure ' auxiliary. feedwater switch 2FWE-PS1588 since it is to be included in the MSP. Current ' licensee planning calls for implementing 2MSP-4.03-I by the end of Septembe r, . c 1987. No other concerns were identified. L 5.6 Reactor Trip Due to Personnel' Error On August 18, with the . reactor. at 20% power, negative rate power range signals caused a reactor trip. Personnel error caused .the trip l when all power was lost for the 24-volt power supply for the 1AC rod .) control power cabinet which caused the affected rods sto drop. I&C- technicians were replacing an overvoltage protector in the #1 power supply and while attempting to mount the protector, they grounded it against an adjacent terminal board which shorted the #2 power supply. l Systems responded as designed to place the plant in Hot Standby (Mode , l 3). The licensee notified the NRC of the ~ event in. accordance with 1 10CFR50.72 reporting requirements. The responsible I&C personnel i were counseled concerning the need for adequate precautions which l should be taken to prevent arcing across terminals of different power l supplies. A ~ new overvoltage protector was satisfactorily installed prior to the subsequent startup. No other concerns were identified. a 5.7 Reactor Trip From Turbine Induced Transient 1
On August 25, the reactor automatically tripped.from 10 percent power ' due to low water level coincident with a steam /feedwater flow mis- match on .the "C" steam generator- (SG).- . An inadvertent = turbine trip initiated the transient, which' was caused by a spurious = signal from - the turbine electrical overspeed pickup device. The reactor trip was _ not a direct result of the turbine trip, however, about six seconds following the turbine trip initiation, the reactor tripped when the i low water level was reached on;"C" SG. The bistables for -the steam / ) - feedwater flow mismatch circuit were tripped due' to maintenance work i on an ' inoperable transmitter. Control . room operators immediately ' l 1 j i ' i 1_ _ _ _ __ _ _ - - _ _ - - _ _ _ . _ _ _ _ _ _ _ _ _ _ __ __
I 10 i implemented station Emergency Operating Procedures. Approximately 27 seconds following the turbine trip, the main generator output break-
ers opened and the non-emergency 4 kV buses automatically trans- i ' ferred. At the same time, all three reactor coolant pumps (RCPs) tripped on bus underfrequency signals (2 out of 3 logic). The plant was subsequently stabilized in Mode 3, using natural circulation cooling. All three RCPs were successfully restarted approximately 50 l minutes following the reactor trip. The licensee notified the NRC in i accordance with 10CFR50.72 reporting requirements. The spurious { electrical overspeed signal occurred while technicians were trouble- ' shooting the device at the turbine local trip panel. The automatic j trip of all three RCPs and the spurious overspeed signal are further i discussed is Section 9. 6 No violations were identified. 6. ESF Actuations ( 6.1 Auxiliary Feedwater Pump Automatic Start On July 17, while in Mode 3, an automatic start of all three auxil- iary feedwater ( AFW) pumps occurred due to a low-low water level in ! I the "C" steam generator (SG) and the failure of the turbine driven ! AFW pump to reach a minimum discharge pressure within a required time period. Prior to the event, a reactor coolant system heatup was 1 initiated which resulted in an increased steam pressure. Since pre- I ! operational test No. P0-2.03.01 (Incore Thermocouple and RTO Cross- ! Calibration Data Collection) was in process, plant operators at- ! tempted to maintain stable steam .line flow rates. Consequently, ' i plant operators allowed steam generator water levels to drift down- i ward during the heatup evolution. The operators responded to the decreasing SG 1evels by increasing feedwater flow, however, the addi- 1 . tion of the relatively cold feedwater did not prevent the level ' reaching the turbine driven AFW pump automatic start (low-low level , in any one steam generator). This and subsequent similar SG 1evel l related operational events prompted licensee corrective actions and 1 are discussed in Section 5.4 j Following the automatic start of the turbine driven AFW pump, both j motor driven AFW pumps automatically started. The motor driven AFW j pumps are designed to automatically start following the failure of ~ a the turbine driven AFW pump to reach its required discharge pressure l within a predetermined times. The licensee stated that, . in this ! case, the turbine driven AFW pump could not develop normal discharge ! pressure due to the low steam generator pressure (420 psig). All plant systems responded as designed. This event was reported in accordance with 10 CFR 50.72 reporting requirements. After the steam generator levels were recovered, plant operators shut down the three AFW pumps and returned to normal feedwater control. No violations were identified. ! l ! l - _ - _ _ _ _ _ _ _ _ _ - - _ _ _ _ _ _ _ _ _ _ - _ _ _ _ . _ _ _ _ . _ _ . - __ _ - _ _ - _ _ .
, - _ - - - . . - . . 11 6.2 Inoperability of the Control Room Air-Pressurization System On. July 20, the licensee. discovered that .the Unit 2 control room l emergency bottled air pressurization (CREBAP) system.was ; inoperable, L contrary to . Technical Specification requirements. On June 19,;1987, an equipeent clearance was issued ~for the containment isol.ation, Phase. L B (CIB) actuation signal to the CREBAP- system to prevent an inadver -
tent CREBAP system -discharge .during related maintenance activities. This . action is permitted by plant . Technical Specifications while. in Modes 5 and 6' however, on . July 15, Unit 2 entered Mode- .4. Upon , discovery on July 20, the licensee immediately restored the CIB actuation circuitry and made the required. NRC notifications. Licen - see investigation into the . event determined that Operations Surveil- lance Test No.1/2.44A.1, Control Room Emergency Habitability System Check, had been performed 'on. July 15, prior to entry into Mode 4. 2 However, the procedure was subsequently determined to _ be , inadequate: in that the CIB interposing relays were not verified to be energized. This test procedure is planned to be revised to . include .the required- veri fication . This event occurred prior to initial criticality on Unit 2 and therefore, there was neither decay heat nor radioactive. fission products present. Therefore, there were no cr@ble acci- ! -dents that could require a CIB initiation at that time . . Addition- l ally, the control room emergency ventil.ation system was unaffected by the clearance and would have operated if needed. The .necessary pro- ' l cedure changes will be reviewed during a subsequent routine resident inspection. No violations were identified. 6.3 Feedwater Isolations 6.3.1 On July 22, a feedwater isolation occurred while in Mode.3 l due to improper operation of the steam dump control system. l' At the time of the event, the. RCS was heating up to normal operating temperature after completion of earlier testing which involved a plant cooldown. The plant operator noted- that the 'C' -steam generator (SG) pressure was. approxi- mately 1040 psig which was above the control pressure ' set- point of 1005 psig where steam dumps. should open'and modu- late to control SG pressure. The operator reset the steam 1 dump control circuitry . to clear any blocks and allow the steam dumps to operate. It -was later. determined that the- steam dumps had been manually blocked by the. operator by placing the. steam dump control interlock selector switch-in~ the OFF position. Therefore,lwhen.the operator reset the l steam. dump control circuitry 'and placed this- switch in' the ' , ON position, the- large steam pressure error allowed nine ! steam . dump valves to open of which three popped full open
1 1 . . i 12 1 'instead of modulating open as designed due to misaligned valve controllers. This resulted in " swelling" in all three SGs to greater than 75% on the SG narrow range level ! instruments and a feedwater isolation' occurred. All plant i systems responded as designed. Operators immediately l closed the steam dump valves and restored SG pressure con- trol using the residual heat release valve. Feedwater control and SG 1evel were returned to normal. ! The licensee adjusted the misaligned controllers for the l three steam valves and all steam dump valves were verified j ' as properly operating. However, the inspector expressed concern that a major contributing factor to.this event was l operator unawareness of the current operational status of 4 l the steam dump control circuitry. It is not clear whether j l this unawareness was due to bad communication with other ' operators ' or individual operator error / distraction due to other operational duties at the time. Root cause analysis of this event and corrective action by the licensee were considered weak and contribute 'to the general observations concerning LERs made by the inspector in Section 13. This ! event was reported in accordance with 10CFR50.72 reporting ' requirements. 6.3.2 On August 16, with the plant operating at approximately 2% power, a feedwater isolation (FWI) occurred . due to high water levels (75% narrow range). The high level setpoint was reached as a result of an inadvertent operation of the i steam dump system (SDS) while performing a preoperational , I test (POT). During the performance of the POT, with the ! SDS in the steam pressure mode of operation, the reference l temperature circuit card was removed and an analog signal was inserted. The SDS selector switch was then placed in
the average temperature mode of operation. Upon restora- l tion of the reference temperature signal, the analog signal ' was removed without placing the selector switch back in the steam pressure mode. This resulted in the SDS controller l sensing a large temperature deviation and consequently, opening all SDS valves. . The transient caused a steam flow ' increase and associated level ' swells' in all three steam generators, initiating the FWI signal. The FWI signal tripped the running main feedwater pump and closed the main feedwater ' isolation valves. The SDS was subsequently placed in the OFF positica and feedwater flow was returned to normal . The root cause of the event was determined to be an inadequate procedure in that the POT procedure did l l l
. . 13 not instruct testing personnel to place the SDS selector switch back to the steam pressure mode prior to removing the analog temperature signal. This test is not planned to be used again, therefore, no procedure change was initi- ated. This appeared to be an isolated occurrence. This ' event was reported to the NRC in accordance with 10 CFR , 50.72 reporting requirements. The inspector will review the licensee's event review associated with this event when it is submitted. 6.4 Inadvertent Safety Injection On July 30, with the plant in Mode 3 (Hot Standby), an inadvertent
safety injection (SI) actuation occurred. The SI was initiated when , two (out of three) independent steam pressure bistables on the "C" steam generator (SG) were tripped. Two separate station groups were performing work at the time of the event, each tripping a separate steam pressure bistable. Meter and control repairmen (MCRs) were performing Maintenance Surveillance Test No. MSP-21.09-I, Loop C , Steam Line Pressure Protection - Channel IV, and Test Group personnel were performing System Operability Verification Test No. SOV-2.24C.01, Automatic Steam Generator Water Level Control . As a result of the SI Actuation, approximately 2000 gallons of Refueling Water Storage Tank water was injected into the reactor coolant system. A main steam line isolation also occurred as a result of the low steam line press- ure signals. Both emergency diesel generators started automatically and all other plant systems operated as expected. Control room operator response to the SI was in accordance with plant procedures. ' The NRC was notified of the event in accordance with 10 CFR 50.72 reporting requirements. Licensee investigation into the event identified that personnel error was the primary contributor to the cause -of the SI. Test Group per- 1 sonnel identified that the initial conditions for the test were not satisfied in that one steam pressure bistable was tripped. This information was apparently miscommunicated to the control room Nuclear Shift Supervisor, who subsequently approved initiation of the test based upon inaccurate information. It was also determined that the pre-test briefing for S0V-2.24C.01 was not of sufficient detail, The licensee stated that test performance in this area would be strengthened. Further inspector review into this event idatified that, while systems engineers are assigned to and specialized on specific plant systems, the test engineer for this test was unfami- liar with the associated process rack logics. Initial test condi- tions were listed, which reflected the neces sary pre-test lineup. However, due to the miscommunicated information, the test engineer assumed that deviating from the initial conditions was acceptable. The licensee stated that the test engineer should have documented a formal test change prior to starting the test for approval. __-_____-____ _ __ __ -_ _ _ ____ __ _-___-____ - -__ -
. , 14 i i l i i Several root causes associated with this event have been identified; ! (1) personnel error by Test Group personnel, (2) improper communica- tion between plant personnel and Operations staff, (3) failure to use system experts on sensitive tests, (4) the need for operations per- sonnel to probe station personnel requests so that all pertinent details of plant evolutions are understood, and (5) inadequate pre- test briefing. Short term corrective actions taken by the licensee j included advising station personnel to research and verify verbal
requests and formalizing pre-test briefings. The Licensee Event ' Report associated with this event had not been completed at the end of the inspection, however, it will be reviewed for adequacy of corrective actions when submitted. l No violations were identified. 7. Control Building and Cable Vault Relief Dampers 1 During performance of CO2 testing in the Control Building and Cable ! Tunnel, overpressure conditions (approximately 10" W.G. ) occurred as a j result of the C02 dumping into relatively confined areas. This situation ! created the. potential for equipment damage and personnel injury. The licensee conducted a study to identify corrective action which would be compatible with both building pressure and fire protection requirements. The engineering solution adopted by the licensee was to install relief dampers at key locations in Control Building ductwork and Cable Vault < exterior walls to vent off any overpressure condition upon a CO2 dump without compromising CO2 concentration in the affected space. Missile and tornado proof openings with grated bars for security protection were pro- vided in the cable vault structure at the 735' and 755' elevations to accommodate these relief dampers. Engineering details for the structural and relief damper work were provided in several Engineering and Design Coordination Reports and the revised construction work was installed and inspected to similar procedures that were used for the original cable vault structure. The inspector monitored the i s tallation of these dampers during the inspection period. The licensee plans to functionally test these dampers before the actual performance of the CO2 discharge concen- tration test. The inspector discussed with licensee personnel the need to periodically check these dampers to ensure operability consistent with similar checks performed on fire dampers. No deficiencies were identified. 8. Surveillance Testing Observation On August 13, 1987, the inspector witnessed the satisfactory performance of OST 2.24.2 " Motor Driven Auxiliary Feed Pump (2FWE*P23A) Monthly Test." The inspector noted that the operators conducting this surveillance test were knowledgeable concerning the necessary operational details for proper test performance e.g. , establishing and maintaining proper communica- tions with the control room, observance of precautions in the procedure, _-______-. _. - -_- ____ _ -
. . 15 i 1 adherence to the procedures and proper recording of data. During the test, the inspector noted that a clamp insert was missing from a unistrut support for the 1/2-inch lube oil supply line to the outboard pump bear- ing. The inspector pointed out the deficiency to one of the operators and it was repaired in accordance with station procedures. No cause was identified for the absence of the small support component. No other concerns were identified. 9. Plant Instrumentation Anomalies Certain technical problems were identified and investigated as a result
of the Turbine / Reactor trip that occurred on August 25. The first problem l was the spurious signal from the turbine electrical overspeed pickup ' device that occurred while technicians were troubleshooting the device at the turbine local trip panel. The second problem was that all three j reactor coolant pumps (RCPs) automatically tripped on bus underfrequency signals concurrent with the automatic fast transfer of the 4KV non- emergency buses. These two problems are detailed below. 9.1 Two electrical overspeed (EOS) circuits are used.at Unit 2, requiring a one out of two logic to initiate a turbine trip. On August 25, station technicians were electronically troubleshooting one of the ) EOS circuits. This particular EOS circuit electrically translates turning gear pulses into turbine speed, while the other EOS circuit receives input from the main turbine Electro-Hydraulic Control (EHC) System (at the turbine pedestal). Since the EOS pickup device was non-functional and there appeared to be no input into the trip cir- cuit, the associated bistable was not disabled for the troubleshoot- i ing activities. However, when a portable pulse counter was connected i to the pickup device, a turbine trip signal was generated. The { licensee subsequently investigated the associated circuit character-
istics and determined that there was an "open" in the circuit. Addi- tionally, it was identified that excessive amounts of electrical 1
noise (both AC and DC) existed on the circuit. This, in combination l l with the additional electrical spike that occurred when connecting the pulse counter to the pickup device, contributed to the turbine trip signal. Licensee immediate corrective actions prior to plant , startup were to disable the associated circuitry for the EOS turbine -{ trip and to connect a recorder to the circuit to monitor electrical i noise during subsequent turbine operation. While some electrical ! noise has been identified, the levels were not in excess of the value ! which would, by itself, cause a reactor trip. The licensee plans to l repair the EOS during the planned mini-outage in mid-September, i ! l - 1 . 1 !
. , 16 9.2 Approximately 27 seconds following the turbine trip on August 25, the main generator output breakers opened and the non-emergency 4KV buses automatically transferred as designed. At the same time, all three RCPs automatically tripped due to bus underfrequency (UF). There is an UF sensing relay connected to each of three RCP buses. . Signal s from any two UF relays will trip all three RCPs. The purpose of UF RCP trip is to ensure that adequate RCP kinetic energy is available for full coastdown and that natural circulation is not inhibited. l The unit was operating in a configuration such that station loads ! were being supplied via the two unit station service transformers (USST). Therefore, when the generator was tripped, automatic fast' bus transfers occurred to switch the station 4KV distribution system supply to the two ' sys tem station service transformers (SSST). The l SSSTs are the paths which provide offsite power to the station. The { automatic bus transfer is designed to occur such that bus electrical l l degradation.does not occur. At the end of the inspection period, the ' licensee had not yet identified the cause of the UF signals. Some of the licensee's investigation and action items include sending the UF. relays offsite for vendor testing, reviewing the preopera- tional test data for the automatic bus transfer, and realigning the electrical distribution system such that the SSSTs supply the 4KV 3 loads. The results of the vendor testing showed that the UF relays l l were in good condition. A preoperational test was satisfactorily performed on the automatic bus transfer feature, however, loading configurations were different since the generator was being backfed . ) from the system. The licensee plans to perform additional testing i using a normal operating configuration to further evaluate and deter- .l mine the root cause of the event. For the interim, the main gener- l ator will not supply electrical power for the 4KV buses, thus the automatic bus transfer feature will not be needed. This issue is currently being evaluated by both licensee and vendor engineering personnel. Pending licensee resolution of the above technical issues, this is Unresolved Item 50-412/87-54-02. 10. Incorrect Bolting Material for Refueling Water Storage Tank Manway Cover i l The inspector observed that approxim' tely half of the bolting associated a with the ground level manway cover for the Refueling Water Storage ' Tank - (RWST) showed evidence of corrosion. Subsequent licensee examination- confirmed the presence of carbon steel bolting which is ~ unacceptable. The licensee -was unable to determine from available personnel and documenta- tion how the unacceptable material became installed. Specifically, 15 unidentifiable carbon steel bolts and 13 SA194 GR2H nuts had been installed - _ _ _ _ - _ - _ - - - - _ _ _
_ . . , i 17 compared to the vendor drawing requirements which specified SA193 GRB6 bolts and SA194 GR6 nuts (AISI Type 410 stainless steel). The bolting was restored to vendor drawing requirements by replacing the affected bolts / nuts on a one-by-one change out. Subsequently, on 8/28/87, the licensee advised the inspector that their inspection of the lower manway covers for ' all tanks (total of 6) fabricated by the same vendor verified the correct installation of bolting materials on these tanks. Based on this inspec- tion, it was concluded that the incorrect bolting material on the RWST manway cover was an isolated problem. No other deficiencies were identified. 11. Review of Licensed Operator Training Program A Region I inspector conducted a review of the licensee's proposed train- ing program for initial licensed operator training to determine if it met regulatory requirements and licensee commitments. Discussions were con- ducted with the training staff regarding the conduct of their initial licensed operator and simulator training programs, applicable portions of Volumes 1 and 2 of the Nuclear Group - Training Administrative Manual were reviewed, and the training process for plant design changes was investigated. There is only one training program to license initial reactor operator (RO) and senior reactor operator (SRO) candidates at Beaver Valley 2. For initial licensed operator training, the applicable INPO accredited Unit 1 program is completed by an individual desiring a Unit 2 license. Then the Unit 1/2 cross training program is completed before a candidate will sit for a license exam. This program emphasizes the differences between Units 1 and 2. An upgraded training program exists for both Units 1 and 2 operators to upgrade to an SR0 license. The procedures used to administer the Unit 1 Training Program will also be used for the Unit 2 systems. Training on plant design changes is initiated when the licensee's engi- neering department notifies their training department of the impending change. The training department, upon formal notification, initiates action to change the plant's training material and schedule training on the change. When the required operator training is completed, the engi- neering department is notified to complete the process. The simulator is used as an integral part of the.requalification training program. Operators are evaluated on their operating ability eight days a year using the simulator. Although the simulator is designed as a plant reference machine for Unit 1, significant training benefit is gained by its use for training on Unit 2. Operations and training management use the feedback provided by simulator evaluations to improve the training program.
.. . . _ _ _ _ - _ _ - _ _ _ - _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _____ _ _ -_-____. _ _ _ _ _ _ _ - _ _ . , 18 An informal system exists for operations personnel to recommend imp.ove- ments to the training program. These recommendations are made by letter, phone call or by verbal request. These requests are documented on the training department action list. This tracking system ensures the appro- priate disposition of the request. Although the program structure makes for a lengthy training period since operators must first complete the appropriate Unit 1 training program prior to the Unit 1/2 cross training, the inspector concluded that the present initial operator training program is adequate. 12. Security Program Weaknesses l During this period, the inspector identified certain . weaknesses in the i licensee's security program. The- licensee's immediate corrective action J was effective in correcting the specific identified items. The licensee initiated a multi-disciplinary study to evaluate if other similar weak-
i nesses might exist; this study was sti}l in progress at the close of the inspection period. Additional details are provided in Attachment 1 which is withheld from public disclosure due to its Safeguards content. This item is Unresolved (50/412-87-54-03) pending NRC security inspector review of the licensee's completed study. 13. Inoffice Review of Licensee Event Reports (LERs) The inspector reviewed LERs submitted to the NRC Region I office to verify that the details of the event were clearly reported, including the accur- I acy of the description of cause and adequacy of corrective action. The inspector determined whether further information was required from the . licensee, whether generic implications were ' indicated, and whether the l event warranted onsite followup. The following LERs were reviewed: ' -- LER 87-01, Reactor Trip Due to Loss of Vital Bus, i -- LER 87-02, Inadvertent Prec9erational Safety Injection. -- LER 87-03, Reactor Trip Due to Loss of Simulated Steam Generator Level Signal. -- LER 87-04, Operation in Violation of Technical Specifications. l LER 87-04-01, Revision to LER 87-04. -- l LER 87-05, Inadvertent Auxiliary Feedwater Actuation. -- LER 87-06, Inoperable Main Steam Isolation Valves. -- , . . _ _ _ _ _ . . _ . _ . _ _ . _ .__.____. _
- . li 19 ! ! > LER 87-07, Inoperaole Control Room Pressurization System. -- . LER 87-08, Failure to Perform Surveillance Test Within the Required l -- Frequency, i f -- LER 87-09, Inadvertent Feedwater Isolation. I I LER 87-10, Reactor Cooldown Due to Inoperable Safety Valves. l -- The above LERs were reviewed with respect to the requirements in 10 CFR 50.73 and the guidance provided in NUREG 1022. Some of the LERs reviewed did not include a specific, . narrative descriptio'n of the events i (LER Nos. 87-03,87-05,87-08); only brief abstracts describing the event were included. Two LERs failed to clearly identify the corrective actions taken or planned to prevent recurrence (LERs Nos. 87-05, 87-07). Addi- tionally, one LER (LER No. 87-09) did not address all apparent root causes of the event. Specifically, a contributing factor to this event seemed to be operator error / unawareness to plant conditions. However, the LER only addressed the equipment problem aspect of the event. i i LERs are used to collect, collate, maintain and evaluate information con- ! cerning licensee events. Providing accurate information is necessary for proper classification of operational events. The inspector expressed the above concerns to the licensee, who acknowledged the inspector's comments. The adequacy of subsequent LERs will continue to be monitored by the l inspectors via the routine inspection program, J 14. Exit Interview l Meetings were held with senior facility management periodically during the course of this inspection to discuss the inspection scope and findings. A i summary of inspection findings was further discussed with the licensee at i the conclusion of the report period. 1 l 1 l , 1 } !
.. . ATTACHMENT'1 .. THIS PAGE CONTAINS SAFEGUARDS INFORMATION- AND IS NOT FOR PUBLIC DISCLOSURE, IT IS INTENTIONALLY LEFT BLANK. .- l , ' I l s + % i
1 / l 1 l 1 i }}