IR 05000412/1987064

Insp Rept 50-412/87-64 on 871010-1120.No Violations Noted. Major Areas Inspected:Licensee Actions on Previous Findings, Site Activities,Followup on Loss of Offsite Power & LER Review

ML20237E502
Person / Time
Site:	Beaver Valley
Issue date:	12/11/1987
From:	Lester Tripp NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML20237E497	List: IR 05000412/1987064 ML20237E495
References
50-412-87-64, NUDOCS 8712290032
Download: ML20237E502 (15)

Text

,

.

,

!

, . .

,

.

V. S. NUCLEAR REGULATORY COMMISSION

REGION I

Report N /87-64 Docket N ,

License N NPF-73 Licensee: Duquesne Light Company

,

'

Nuclear Group Post Office Box 4 Shippingport, PA.15077 Facility Name: Beaver Valley Power Station, Unit 2.

Dates

October 10 - November 20, 1987 l

l Inspectors: J. E. Beall, Senior Resident Inspector S. M. Pindale, Resident Inspector Approved By: h. h Uowell E. Tribb, Chief, Reactor Projects

/k:(/67

' Dhte Section No. 3A Inspection Summary: Inspection No. 50-412/87-64 on October 10-November 20, 1987 Areas Inspected: Routine inspections by the resident inspectors (330 hours0.00382 days <br />0.0917 hours <br />5.456349e-4 weeks <br />1.25565e-4 months <br />)

of licensee actions on previous findings, . site activities, followup on loss of offsite power and LER revie Results: No violations were identified. One unresolvid item was ' opened to address the inability to successfully transfer to offsite power following a load rejection. Two NRC open items were closed during this inspection.

l

.

8712290032 871214 PDR ADOCK 05000412 a DCD l

- - - _

s

'

-

OA u

L

. - .

.

I 'a DETAILS 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 activitie ~

" Project Status Summary During the inspection period, the licensee completed the power. ascension testing progra Major test milestones accomplished include initial-achievement of 100% power on October 14, 1987 and load rejection from 100%

power on October 24. The load rejection test resulted in a reactor trip and the ' licensee restarted the plant and commenced the 100 hour0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> warranty run at 100% power. The warranty run was interrupted by a ructor trip on October 29. Reactor coolant system leakage approaching Technical Specifi-cation limits occurred on October 31, the plant was subsequently shut down a s until November 10 for repairs, maintenance and testing. The warranty ru was subsequently completed at 7:52 a.m. on November .17, and the Mnit was declared in commercial -operation at 9:00 a.m. on November 17. ' The unit tripped at 2:06 p.m. on November 17, and remained shutdown for investiga-tion and corrective action at the end of the perio . Inspection Program Status Summary '

e Preoperational test program inspection is complet Inspection of the l ,

'

licensee's startup and power ascension program =was being compl.eted as .I documented in Inspection Report 50-412/87-67 which was in progress at, the ,'? q close of the inspection perio The current status of t.b e st grtup'

inspection program is as follows: [ ,

% INSPECTION COMPLETE AREA END OF THIS PERIOD END OF LAST E RIO Overall Program 95 80 Procedure Reviews: 100 95 4

,

Test Witness: 100 80 Results Review: 90 75

}

't The completion af the licensee's testing program and declaration o g commercial operation marked the initial implementation of the operatiom i phase of the NRC inspection progra ,

" '

,

,

( .-

-

u

- __ -_-___- iY - - - - - - - - - - -

-. y

• 5 .i a . ,6 ;

m i

2 3

'

o

,

d '

+

Licensee Actions on Previous inspection Findings

. ' '

i (Closed) Unresolved Item (87-54-01): The licensee war, to investigate a timing discrepancy for the turbine trip fran reactor trip signal The licensee's investigation determined that the basis for' the above signal I is the verification of reactor trip breaker and bypass trip breaker open-ing times. The licensee confirmed that the opening of the reactor trip breakers was unaffected by the signal delay. Additionally, in subsequent reactor trips, the computer address us within expected values. An engi- y neering evaluation determined that auxiliary relays associated with the A circuit do not significantly affect total circuit response . time The licensee - speculated that the cause for the delay in receiving the signal

i was due to computer problems that were experienced during. Ow' time of the -

l even The engineering evaluation recommended that' testing and/or troubleshooting activities should be used to determine the tw;e if the address time is abnormally high on future reactor trips. Future turbine trip from reactor trip signal times, and licensee correctiv? actions (if necessary) will be reviewed by the inspector during' subsequent routine inspections. This item is close (Closed) Unresolved Item (87-03-03): Inconsistencies were ' identified be- i tween Station Administrative Procedure No. 10 (SAP ip), Onsite Safety Committee (OSC), and the plant Technical Specifications (TS). The'T weree changed prior to issuance of'the full power license ari M P 10 was revised 4 to reflect i'S requirement The inspector reviewed W Tatgtt SAD 10 1 '

revision and the current 15. One discrepancy was identif M f a - that, TS V 6.8.2, Procedures, charges the OSC to review plant procedures and admin ,

istrative policies and changes thereto, however, both SAP 10 e.nd 'l5^ 4

'

6.5.1.6, OSC Responsibilities, charge the 0SC to review only changes af inten Additionally, Unit 1 TS 6.8.2 charges the OSC to review 'only intent changes to procedures and administrative policies. .The licensee stated that this was an omission during the TS change process and tha TS l 6.8.2 would be changed via a formal TS change request at a future. dat l l This change will provide TS consistency regarding procedure changes re-quired to be reviewed by the OSC. The above administrative TS change will I be reviewed during a subsequent inspection. This item is closed, i 5. Site Activities

>

Throughout the inspection ped od, the inspectors tusredJ the licensee facilities. General work activities were observd incfuding construction, surveillance, testing and maintenance. The irspectors uso monitored the licensee's housekeeping, security: and radiation control activities. In particular, the inspectors monitored the licensee's progress towards

,

achieving full power and commercial operation including review of the - '

following events:

t'

.

/

k

,

,.

)

- _ _ _ _ _ _ _ _ _

, - - . _ _ ___ --_-- _ -- _- _ _ _ _

Q

'

..

. . 2,.

,o ,.

f

,

-

5.1 Turbine Trip / Reactor Trip Due to Low EHC Pressure

~

On October 14, a turbine / reactor trip occurred from full power due

,

"

to a turbine electro-hydraulic control (EHC) system low fluid press-lure signal. Prior tc> the trip, plant operators received a control

- N om a brm indicating low EHC fluid pressure. An operator was sub-sequently rii'./pa tched to the turbine to open the manual isolation

,

valves (#) for the associated pressure switches in order to restore sy iten pre s sure . The pressure switches had previously been pressur-ired and isobted after oscillating EHC fluid pressure during routine operations caused an automatic turbine trip in August, 1987. The 11censee and the turbine vendor (Westinghouse) had been unable to vpee on a course of corrective action to the EHC pressure fluctua-tions and the licensee elected to isolate the pressure sensors as a shart urm correr:tive action. Upon receipt of the control room alarm, operators speculated that the low pressure signal was caused by leak-rige of the trapped and pressurized EHC fluid leading to the pressure switches. Oprning the pressure switch isolation valves was expected to restore Mmal EHC pressure to the switches. However, before the operator reached the turbine to open the isolation valves, a second pressure switch reached its low setpoint, and the coincidence for.the low EHC fluid pressure turbine trip (2/4) was satisfied. A reactor trip was automat' tally initiated due to the turbine trip as per plant desig Plant crerators immediately implemented station emergency operating proe dures and stabilized the uni Plant response was

' normal. The NRC was notified of this event in accordance with 10 CFR 50.72 reporting' requirements, h

Licensee :orrective action for this event includes electrically I jumpering out the EHC pressure switch automatic turbine trip func-tio Westinghouse concurred with this action. The alarm function

'. was unaffected Special operating instructions were issued to plant operaters which informed them that the low pressure trip had been defeated und directs them to manually trip the . unit if an actual low O1C pressure condition occurs, and continues to decrease or remains kw for greater than one minut Therefore, turbine protection is ornided altheegh the pressure switches remain defeated. This is an 1 3r.terim sclution, and pending further licensee review, a long term normanent fix (design change) is being considered by both the licen-

, see and Westinghouse. The inspector had no further questions at this cime. The effectiveness of these interim corrective actions will be reviewed during subsequent resident inspection .

..?

n g e t

m m ___m--_m_ - . - - - - - - - - - - - - - - - - - - - - - - -

_ - _ _ _ _ _ __ _ _ _ - - _

.

.

5.2 Turbine Trip / Reactor Trip Due to Erratic Feedwater Regulating Valve Operation On October 15, while at 65?; and increasing reactor power, a turbine trip occurred due to high water level i n t h e " A'" steam generator (SG). Since reactor power was above the P-9 interlock (49?s) , a reactor trip also occurre Prior to the trip, all three feedwater regulating valves (FRVs) were operating in automatic. At approxi-mately 3:14 p.m., plant operators noticed erratic operation of all three FRVs, including full travel range oscillations in the "A" and

"B" FRVs. No apparent system perturbations were introduced which may have caused the transient. The operators immediately placed the "A" and "B" FRVs in manual in an attempt to restore the feedwater control system to normal; however, feedwater level and flow stability could not be restored in that manual adjustments to one FRV adversely affected the response on the remaining lines and SGs. At about 3:15 p.m., the water level in the "A" SG reached its high level turbine trip setpoint and initiated the automatic turbine trip. The plant was subsequently stabilized in Mode The NRC was notified per 10 CFR 50.72 reporting requirement .

The licensee attributed the cause of the erratic FRV operation to be valve over-responsiveness to small SG water level changes. The licensee subsequently adjusted the gains in the FRV control system to stabilize valve operatio Additional planned corrective action ir.cludes adjustment of the FRV response times to further stabilize the syste FRV anomalies continue to cause operational problems at Unit Additionally, similar erratic FRV operation occurred. earlier on October 15. However, the licensee speculated that due to the opera-tion of only one main feedwater pump and a lower power level, plant operators were able to stabilize the control system after placing the ,

FRVs in rnanual . The inspector had no further questions at this tim !

The effectiveness of the licensee's corrective actions will be '

reviewed during subsequent routine inspection .

5.3 Manual Reactor Trip Due to Turbine Building Fire On October 16, plant operators manually tripped the reactor from full power due to a fire in the vicinity of the No. 2 turbine bearin Prior to the plant trip, the control room was notified that smoke was issuing from the Unit 2 high pressure turbine enclosure, and the emergency squad was immediately dispatched to the turbine buildin The control room supervisor subsequently directed plant operators to manually . trip the reactor and turbin The fire was extinguished within 10 minute .

_ _ _ _ _ . _ _ . _ _ _ _ . _ _ . . _ _ . _ _ .- __

-_ . _ _ _ _ _ _

.

.

.

. 5 The licensee notified the NRC of the plant trip 'in accordance with 10 CFR 50.72 reporting requirement Subsequent-licensee'investiga-tion identified that the fire 'was caused by oil soaked insulation on the No. 2 gland piping. Licensee followup inspecti,ons identified no significant equipment damage. The licensee' determined that the No. 2 bearing leaks oil whenever the lid on the lubricating Loil' reservoir is removed and negative pressure is lost in the turbine pedesta The reservoir lid is removed by plant operators to physically inspect a basket strainer for impurities or debri Interim corrective actions included providing instructions to plant operators to main-tain the time that the oil reservoir lid is removed to a minimum dur-ing periodic strainer inspection Additionally,- a sheet metal deflector and trough were installed beneath the No. 2 bearing to pre-vent oil from spraying onto adjacent insulatio The licensee and turbine vendor (Westinghouse) are also investigating _long term cor-rective actions to provide an internal oil seal system so that admin-

'strative controls for the reservoir lid would not be necessary. The inspector had no further questions at this time. The inspector will monitor the effectiveness of the licensee's corrective acti on s'.

5.4 RCS Leakage Anomalies 5. Notification of Unusual Event due to RCS Leakage An Unusual Event was declared at 11:20 am on October 20, due to reactor coolant system (RCS) unidentified leakage exceeding 1 gp Data from the operations surveillance test (OST) No. 2.6.2, RCS. Water Inventory Balance, indi-cated a 1.33 gpm unidentified leak rate. During review of the OST calculations, plans were made for a containment entry at power to identify primary system leaks. A crew-of maintenance and operations personnel entered containment, however, the leakage sources were not positively identified during the e n t ry.. The Unusual Event was subsequently declared in accordance with plant procedures and the appro-priate state, local and federal notifications were mad The leaks could not be terminated within the time allowed by plant Technical Specifications (TS), therefore, a con-trolled plant shutdown was initiated in accordance with TS requirement The NRC was notified of the initiation of the plant shutdown in accordance with 10 CFR 50.72 report-ing requirements. Mode 3 (Hot Standby) was reached at 5:21 pm on October 20. The licensee subsequently identified the primary source of the leakage to be a body-to-bonnet leak on the RCS Loop A cold leg bypass isolation valve (RCS-44).

Additional leaks were also identified on the same valve for the Loop C bypass flow line (RCS-46, packing leak) and in the Loop C cold leg instrument manifola outlet isolation

.

_ _ _ _ _ _ _ _ _ _ _ _ _ . . . . _ _ _ . . _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _ _ . . _ _ _ _ _ _ _ _ - . _ __ . _ _ _ _ _ _ _ __ _m___ _ _ ____._m

- _ _ _ _ _ ______ ____ - __ _ _ _

.

. . . ,

.

, 6 valve (RCS-37, body-to-bonnet leak). The Loop A bypass

.

I flow line valve was repaired and the unidentified leak rate was reduced to less than 1 gp The Unusual Event was terminated at 7:30 am on October 21. , Additionally, the limiting conditions for opera ti_ on for the associated TS were satisfied since the unidentified leakage wa's. brought to within the specified limits. The plant remained in Mode 3 while the other two leaking valves were being repaire The reactor was subsequently restarted on October 2 A common root cause for .these valve leakages- cculd not be

. identified. Discussions with the ~ valve vendor indicated'

that . the valves can withstand pressures greatly in excess of any that they may have been subjected to in the course of _ normal operations or testin The licensee planned to replace the two loop bypass flow line valves during the first refueling outag . Plant Shutdown to Repair Leaking RCS Valves On October 30, while operating at full power, the licensee made a containment ~ entry to search for suspected ~ primary system leaks. Two leaks were identified; the Loop C cold leg instrument manifold outlet isolation valve (RCS-37),

and an auxiliary feedwater system valve (FWE-99). RCS-37 was one of the valves that was leaking on the October 20th Unusual Event. On October 31, OST 2.6.2 was performed and RCS unidentified leakage was again greater than 1 gpm (1.26 gpm). The licensee was able to quantify various leaks and reduce the unidentified leak rate to. less than 1 gpm, including an estimated 1/2 gpm leak on RCS-37 based on visual observation. Licensee management was notified of the leak rates as measured by the OST and the quantified and estimated leak rate values. The. licensee subsequently determined that the unidentified leak rate was .less than 1 gpm and an Unusual Event was not declared. The plant shut down to Mode 2 on October 31 so that radiation levels in the area of RCS-37 would be substantially reduced to facil-itate valve repair. Two additional valves were identified to be leaking slightly (the loops A and C cold leg bypass isolation valves, Nos. RCS-44 and RCS-46, respectively).

These valves were also leaking during the October 20 Unusual Even OST 2.6.2 was performed on November 1 and the unidentified leak rates were again slightly greater

.

m_ _ _ _ _ _ _ _ _ _ _ __ _ . _ _ _ _ _ __ _ _ _ _. _ _ _ _ _ _ _ _ . _ _ _ _ _ _ . _ _ _ _ . _ ._ _ _ . . . . . _ . _ _ ____-.___________.______.____.____m _-_.m___._a

- _ - _ _ _ - _ - _ = - _ - - -

.

. .

.

. 7 than 1 gp The licensee identified and quantified the leakage from RCS-44, and . determined l that an - Unusual Event'

and Technical Specification Action Statements were no appropriate at the tim However, ' a plant ' shutdown was initiated for maintenance purposes, so that all affected -

valves could be repaire Mode 5 (Cold Shutdown) was reached on November The valves were subsequently repaired and tested, and a plant startup . commenced' on November 10, 198 Inspection coverage of the activities. associated with resolution of '

the above RCS leakage problems included examination, repair., mainten-ance and testing effort The inspector reviewed the licensee's methods for quantifying leak rates and discussed industry ' technique with licensee representative No significant concerns were identifie ,5 Reactor Trip During power Ascension Testing On October 24, 1987, a reactor trip occurred from 100% power due to low-low water level in the steam generators .during the 100% . power complete load rejection test. Upon opening the main generator output breakers, steam generator water levels shrunk rapidlyfand the reactor tripped approximately six seconds into the transient. The trip was

~

not entirely unexpected but was not part of the preplanned test. NRC inspectors were present in the control room and observed the test-and trip response. Operator actions were prompt and correct. Emergency Operating Procedures were used. to stabilize the plant in ~ Mode Control room supervision was in clear control managing the actions of the operators. The NRC was notified of the plant trip per 10 CFR 50.72 reporting requirements. The electrical transient following the load rejection resulted in an interruption of' power from one off site source and the loading of the associated diesel generato Addi-tional details are presented in Section .6 Reactor Trip Due To Low-Low Steam Generator Level On October 29, the reactor tripped from full power due to a low-low -

steam generator (SG) water level condition on the "B" SG. During routine plant operations, several high and low feedwater heater level alarms annunciated. One feedwater heater reached its extreme high setpoint and automatically isolate The heater drain pumps began indicating swings in electrical current levels and the main feedwater pump low suction pressure alarm annunciate The heater drain . pump

.

---

._.a A

_ _ - _ _ _ _ -

,

. ..

,

)

.

8 i i

l output flow to the suction of- the main feedwater pumps was substan- 3 tially reduced until the "A" main feedwater pump automatically tripped due to. low suction pressure. An emergency plant shutdown was initiated as per station procedures. The "A" main feedwater pump was successfully restarted,' however, it automatically tripped again after about 10 seconds. SG water levels drif ted downward until the low-low ]j setpoint was reached on the "B" SG, which1 initiated the reactor tri All emergency systems functioned normall The NRC was notified of the trip in accordance with 10 CFR 50.72 reporting -requirement s Licensee corrective actions ' for this event included . verifying high and extreme high level setpoints, and physically inspecting the nor-mal and high level valve controllers. All feedwater heaters were included in the above verifications. Minor. discrepancies were iden-tified and corrected, including minor valve actuator problems. Addi-tionally, a feedwater system integrity walkdown was performed subse-quent to the trip. No significant damage to equipment was identi-fied. Technicians remained on site during the subsequent plant startup and power ascension to troubleshoot any potential problem l The startup .and power ascension was deliberate to assess the effec-  ;

tiveness of the equipment adjustment No~ significant anomalies; j occurre The system was " tuned" during power operations and further adjustments are still planned as the licensee is currently monitoring feedwater system performance and correcting minor control problem The inspectors had no further questions at this time. The effective-ness of these actions will be reviewed further during routine inspection .7 Reactor Trip on Low-Low Steam Generator Level On November 10, 1987, a reactor trip. occurred from 11% power'due to low-low steam generator levels following a valve failure in the con-densate system. The valve was a 24 inch butterfly type valve in the bypass line around the condensate demineralizers. The valve was a major flow path because most of the condensate demineralizers were 'l not availabl The demineralizers, as originally designed and  ;

installed, were found to pass resin at. high flows and modifications I were required to the demineralized septums. At the time of the trip,.

only two of the five demineralizers had been modified. with the rest 9A awaiting parts from the vendo This valve also failed during preoperatior.al testing about a year ago when it was in the closed position during the start of the condensate pum Design review identified that the valve disk was limited to about 70 pounds differential pressure which was adequate for a . low flow bypass line, but which would fail when subjected to the conden- i sate pump's 600 pound shutoff head. The valve also experienced some i

.

_ _ . . _ . _ _ . _ - . -

- _ - _ . ._ _-

.

.

., 1

vibration under high flow conditions. The licensee revised operatin procedures to require the valve to' be manually opened during conden-sate pump starts to preclude another failure and ~ concurrently initi-ated the procurement of a redesigned replacement , valve. Following the November 10th f ailure during operation, the valve was replaced with the' redesigned replacement. The new valve is much more massive (6000. pounds versus about 1200 pounds),- has a much larger operator, and is designed to withstand the full differential pressure of a con-densate pump star The inspector observed the operation of'the valve under flow conditions and noted no ~ significant vibration or other operating abnormalitie Following valve replacement, no recurrent problems of this nature were observed during the perio The inspector had no further question .8 Reactor Trip on Spurious Thrust Bearing Trip On November 17, 1987, a reactor trip occurred' from'100% power due to a turbine trip on thrust bearing wear. A technician was ~ working on a recorder at chest ' aight and' inadvertently bumped a toggle switch at about waist height turning off the . power supply to . .the turbine rotor position module. The technician: then turned the power ' supply back on and a very brief voltage spike 'was generated which tripped the turbine and the reactor. The electrical . transient fullowing the

.

trip was complex and included ' motorizing the generator through the 4 -

.

KV buses, tripping the reactor coolant pumps, the loss of .all offsite power for 17 seconds, and the auto start and loading of both diesel generator An NRC specialist team was dispatched to the site to investigate the event and review the licensee's corrective action These actions were completed and the plant was placed on the grid on

-

November 22, 198 Additional details are presented in. Section . Followup on Loss of Offsite Power

.

l On November 17, 1987, a reactor trip occurred which resulted in a brief loss of offsite power (see Section 5.8). An NRC specialist team was dis-patched to the site to investigate the event and review the licensee's corrective actions. A previous trip on October 24, 1987, during a plant challenging startup test had resulted in a partial loss of offsite power (see Section 5.5). The October 24, 100% load rejection test had been initiated by opening the main generator output breakers'while leaving the generator field breaker shut and the generator energized supplying onsite load _

_.,___.__-_.a..---- - - - - - - - - " -

.

.,

The onsite loads consist of four 4 kV buses, two on one onsite transformer  ;

(2C) and the other two on a different onsite transformer (2D). The,4 kV '

buses supply normal operating equipment such . as reactor coolant pumps, feedwater pumps, condensate pumps and service water pumps and are the normal power supply for the two safety equipment bu se's . The alternate supply for the 4 kV buses is offsite power with two transmission lines each supplying a different transformer (2A and 28) which then each supplies two of the 4 kV buses. Each safety equipment bus also has a dedicated emergency diesel generator which starts automatically to provide power on loss of voltage to its associated emergency bu .1 October 24, 1937, Partial Loss of Offsite power Opening the generator output breakers during the October 24 test detached the generator and the site loads from the offsite grid and a consequent loss of frequency synchronizatio About six seconds after test initiation, a reactor trip occurred on low-low steam generator levels due to the ' transient. The reactor trip caused a turbine trip and, after a 30 second time delay, the trip of the  ;

generator field breake !

The reactor coolant pumps, unlike the other-4 KV loads, have under-frequency (UF) trip devices to insure RCS natural circulation is not ,

impeded by an electrically braking pump. . These pumps tripped on UF '

shortly after the reactor trip / turbine trip. .The '4 KV buses at-l tempted to transfer to their alternate power sources, the offsite I

supplied transformers, due to undervoltage when the generator field breaker opened. The frequency phase differential between the onsite i buses and the offsite grid as well as the current inrush of the re- '

l accelerating motors caused one of the offsite transformers'(2A) to be

! unable to accommodate the transfer. The associated 4 KV bus breakers tripped on overcurrent and the dedicated diesel automatically sup-plied power to the safety loads which had lost power (both diesels had been started manually and were running unloaded prior to the I test). The other two 4 KV buses successfully transferred to offsite '

power. Operators manually placed the de-energized 4 KV buses on off-  ;

site power, restored forced reactor coolant flow, and secured the j diesel generator The trip of the reactor coolant pumps was as designed. The licen-see's position was that the partial loss of offsite power was due to the loss of synchronization between the 4 KV buses and offsite power caused by the detachment of the still energized generator from the  ;

grid and the subsequent reactor trip. The unit did -not trip during lesser load reject transients during the startup testing phase and '

the licensee's analysis in the FSAR is that 100% load rejection is a low probability even It should be noted that the 4 KV buses are .

powered from the offsite sources during reactor startup until after I the generator is on line and the plant at about 50% powe .

_ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ - - _ _ _ -

--_

.

.

.

The adequacy and extent of the' licensee's review of this event were also examined by a team of NRC specialist inspectors following the November 17, 1987 event (Inspection Report 50-412/87-68).

6.~2' November 17, 1987, Loss of Offsite Power The. spurious ~ thrust bearing wear trip signal (see Section- 5.8) was very brief, less than two 60 hertz cycles in duratio This machine protection trip was unique in that it was not a " seal-in" trip. These two facts were critical to the event in that a machine - protection

,

trip occurred and then cleared before the autamatic transfer of the 4 KV breakers to offsite power was complete.

l The trip signal opened the generator output breakers, opened the generator field breaker, forced the traiisfer of the 4 KV buses to the offsite power sources (the 2A and 2B transformers), and generated reactor trip and turbine trip signals. The 4 KV bus transfers oc-

~

curred by the simultaneous movement of all the onsite bre:1kers (tripping open) and the offsite power breakers (closing). The breakers take two cycles to open' and four cycles to clos The simultaneous movement is necessary to minimize the interval that no breaker is shut (in this case two cycles) so that the loads see. no interruption .of powe Another design feature prevented a 4 KV i breaker from receiving a signal to close if the other 4 KV breaker j on that bus was already closed.

l The problem that had not been anticipated in the design involved the two cycle period when one breaker has already . tripped open and the i other breaker has not yet closed (although already in motion). Dur-l ing that period, no breaker appears to be shut and the tripped breaker can receive a signal to reclose. Because the November 17th trip signal cleared quickly and because residual . magnetism still remained on the generator, the onsite power source still had suffic-ient voltage to allow breaker reclosur The four 'onsite 4 KV breakers reclosed and for several cycles all eight 4 KV breakers were shut. Each 4 KV bus provided a current path from the.offsite grid, through an offsite transformer, onto the 4 KV bus, through an onsite transformer and onto the main ger.erator (motorizing the main gener-ator).

Oscilloscope traces show that about 6,000 amps motorizing current was seen on the 22 KV generator bus which indicates that well over 20,000 ,

amps current was split among the 4 KV buses and their transformer windings. All the breakers tripped on overcurrent with the exception of the offsite supply breaker to one transformer (2A) which tripped more quickly on phase differential (an offsite ' transmission line

.

- - - _ - . _ . - - _ _ . . . . - - - . ._ . - - _ _ . - _ _ - _ _ . . .

___-_

- .

12 i protection relay). All' the 4 KV buses were d'e-ene rg i zed , both diesels ' automatically started, and both diesels provided power to i their safety buses. Af ter about 17 seconds, the phase differential i relay timer reset and'offsite power was automatically restored to one .j transformer (2A); the associated 4 KV but, breakers 'then automatically j closed restoring power to two of the four 4 KV buses. The licensee secured one diesel, inspected the de-energized transformers, con-firmed that no damage had occurred, restored power to the remaining ,

,

two 4 KV buses, and secured the remaining diese .3 Corrective Actions I The licensee analyzed the sequence of events and other data concern-ing the two loss of offsite power events as did the- NRC specialist team and resident inspectors, and initiated two -modifications .in-

_

I tended to preclude a repetition of the November 17th event. The l

licensee conducted a review of the existing trip circuits and made modifications as necessary to make all trips " seal-in" in nature The licensee also changed the 4 KV bus transfer logic, effectively making the automatic fast transfer feature only capable of switching from onsite power to offsite power (and not the reverse). Further details on the modifications and post-modifications are presented in Inspection Report 50-412/87-6 l An additional concern was identified during the review of the October

, 24th event. The loss of synchronization with the grid following l generator output breaker opening challenged the ability of.the 4 KV buses to automatically transfer to offsite power (the grid) following L

a subsequent reactor / turbine trip. The time delay before attempting  !

the transfer also causes the reactor coolant pumps to trip on U The present arrangement keeps the 4 KV buses on onsite power to pro-vide some load for the generator and assist the steam dumps in smoothing out the transient and therefore. improve the ability of the i unit to avoid tripping on load rejectio The. licensee-is currently re-exaniining this issue (Engineering Memorandum 62948) with the possibility of changing the logic to immediately transfer the 4 KV buses to offsite power upon tripping of the generator output-breakers. This item will be tracked (Unresolved Item 87-64-01) and the resolution reviewed in a future inspectio . Surveillance Testing l

The inspector witnessed / reviewed selected surveillance tests to determine whether properly approved procedures were in use, details were adequate, test instrumentation was properly calibrated and used, technical specifi-cations were satisfied, testing was performed by qualified personnel and test results satisfied acceptance criteria or were properly dispositione The following surveillance testing activities were reviewed:

.

,

_ _ _ _ _ _ _ _ _ . _ _ _ . _ _ _ . _ _ _ _ _ _ _ _ _ . _ _ . . _ _ _ . _ . _ . _ . _ . _ . . _ . _ . . _ . . . ___.__._______________.__m.___.____m

.

. .

, 1S OST 2.1.1 - Control Rod Assembly Partial Movement Test OST 2.6.2 - Reactor Coolant System Inventory Balance OST 2.6.7 - Accident Monitoring Instrumentation Channel Checks OST 2.11.1 - Low Head Safety Injection Pump Test OST 2.13.2 - Quench Spray Pump Test No deficiencies were identifie . Inoffice Review of Licensee Event Reoorts (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-acy of the description of cause and adequacy of corrective actio The inspector determined whether further information was required from the licensee, whether generic implications were indicated, and whether the event warranted onsite followup. The following LERs'were reviewed:

LER 87-20-00: Reactor Trip Following Main Steam Isolation Valve Closure LER 87-20-01: Revision 1 to LER 87-20-00 LER 87-21-00: Zero Drift of RCS Pressurizer Protection Pressure Transmitters LER 87-22-00: -Automatic Start of No. 1 Emergency Diesel Generator LER 87-23-00: Reactor Trip Due to Steam Flow / Feed i}ow Mismatch with l

l Low Level LER 87-24-00: Safety Injection Due to Low Steamline Pressure Signal LER 87-25-00: Auxiliary Feedwater Actuation LER 87-26-00: Reactor Trip Due to Steam Flow / Feed Flow Mismatch with Low Level LER 87-27-00: Missed Surveillance Test LER 87-28-00: Turbine Trip / Reactor Trip on Low EHC Pressure LER 87-29-00: Turbine Trip / Reactor Trip Due to Erratic Main Feedwater Regulating Valve Operation LER 87-30-00: Manual Reactor Trip Due to Fire at #2 Turbine Bearing LER 87-31-00: Unusual Event - Reactor Coolant System Leakage in Excess of Technical Specifications The above LERs were reviewed with respect to the requirements of 10 CFR 50.73 and the guidance provided in NUREG 1022. Inspection Reports 50-412/

87-54 and 60 documented the review of the first 19 LERs and certain weak-nesses were identified although an improving trend was noted. Continued improvement was observed during this period with no specific deficiencies identified. Current LERs were noted to document good event analyses, root cause determinat.uns and corrective actions implementatio <

9. Site Meetings Meetings were held with senior facility management periodically during the course of this inspection to discuss the inspection scope and fir. ding Two announced meetings were also held at the licensee's facility on November 20, 198 .

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ -

--_ -_ -- _ _ _

,

.

.

l

14  !

!

The first announced meeting was held in the Training Conter and addressed certain operator. licensing items. In particular, the NRC's position on simulator exams for Unit 2 operator candidates was presented in light of-the recent changes to 10 CFR 50. In summary, the- lic.ensee must either certify a simulator to be in close ag reenient with Unit 2 or get NRC-approval of alternative training devices or facilitie The meeting: pro-vided an opportunity for the licensee and the NRC staff to discuss the licensee's current ' plans to meet the new regulations . including implemen-tation schedules The other announced meeting was held in the Admini.stration Building and consisted of a review of the unit's operating history to date with. the emphasis on unplanned reactor shutdowns and safety system actuations. The licensee's experience during the startup testing phase was discussed and compared with the recent experience of other facilities during that phase as documented in NUREG 1275, " Operating Experience Feedback. Reports - New Plants". The data indicated that the unit had experienced about an aver-age number of automatic reactor trips, and a less than average number of safety system actuations, Technical Specification violations and losses of safety system function as other comparable, recently licensed facilities had experienced during the startup testing phas )

.

.

- - ______ ___