IR 05000269/1989003

Augmented Insp Team Repts 50-269/89-03,50-270/89-03 & 50-287/89-03 on 890104-13.No Violations or Deviations Noted. Major Areas Inspected:Selective Exams of Procedures & Representative Records & Interviews W/Personnel

ML16127A290
Person / Time
Site:	Oconee
Issue date:	02/02/1989
From:	Peebles T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
To:
Shared Package
ML16127A288	List: IR 05000269/1989003 ML16127A285
References
50-269-89-03, 50-269-89-3, 50-270-89-03, 50-270-89-3, 50-287-89-03, 50-287-89-3, NUDOCS 8902270199
Download: ML16127A290 (52)

Text

pR~ REGUZ UNITED STATES NUCLEAR REGULATORY COMMISSION

REGION II

101 MARIETTA STREET, NX ATLANTA, GEORGIA 30323 U.S. NUCLEAR REGULATORY COMRISSION

REGION II

AUGMENTED INSPECTION TEAM Report Nos. 50-269/89-03, 50-270/89-03, and 50-287/89-03 Licensee:

Duke Power Company 422 South Church Street Charlotte, NC 28242 Docket Nos.:

50-269, 50-270, and 50-287 License Nos.:

DPR-38, DPR-47, and DPR-55 Facility Name:

Oconee Units 1, 2, and 3 Inspection Conducted: January 4 -

13, 1989 Team Members: W. L. Jensen, Events Assessment Branch, NRR R. C. Jones, Section Chief, Reactor Systems Branch, NRR H. N. Pastis, Oconee Project Manager, NRR C. J. Paulk, Reactor Inspector, RII P. H. Skinner, Senior Resident Inspector, Oconee D. C. Ward, Reactor Inspector, RH L. D. Wert, Resident Inspector, Oconee, Assistant Team Leader Team Leader:

2-:2z -7'

T. A. Peebles", Chief Date Signed Projects Section 3A

X 17.)>C Di-~

~v TABLE OF CONTENTS Pag I.. Introduction -

Formation.and Initiation of AIT Background

'B. Formation of Augmented Inspection Team (AIT)

C. AIT Charter -

Initiation of Inspection, CAL Letters

D. Persons Contacted

E. Design Descriptions 1. 6900/4160V Electrical Distribution System

2. Main.Feedwater and Auxiliary Feedwater Flowpaths (Final

Feedwater portions)

3. Pressurizer Auxiliary Spray System

.

4. Thermal Shock Operating Region Description

II. Description of Events Event Description

1. January 2 Trip 2. January 3 Event B. Detailed Sequence of Events -January 3

III. Firefighting and Fire Protection Aspects

IV. Equipment Status and Role in Event Normal Supply Breaker to 6900V Switchgear ITA

1. Equipment as Found Status of Investigation 3. Corrective Actions B. Cabling Above Breaker (Potentially Damaged)

1. Equipment as Found Inability of Licensee to Quickly Resolve/Identify Cables 3. Significance of Potentially Damaged Cables 4. Other Drawing Deficiencies 5. Corrective Actions C. Main Feedwater Valves 29 Failure of Automatic Switchover to Upper Nozzle Flow Path 2. Significance of Delay in Shutting Valves to Lower Nozzle Ring 3. Main Feedwater Block Valves Corrective Actions

D. Reactor Coolant System and Auxiliary Pressurizer Spray 31 Initial Attempt to Use Auxiliary Spray 2. Manual Valve Found Shut 3. Consequences of Differential Temperature Limit Violation-E. Reactor Coolant Pump 1A2 Delayed Recovery

F. Other Equipment Failures

V. Natural Circulation Natural Circulation Events

B. Evaluation of Natural Circulation Effectiveness 3 Summary

V Operator Action a.nd Management Involvement During Event and Recovery 38 VII. Findings Reportability Review

B. Open Items and Concerns

VIII.Conclusions

IX. Exit

X. Acronyms and Abbreviations

Attachment 1 - 6900/4160V Electrical Distribution Systems Attachment 2 - Main Feedwater and Auxiliary Feedwater Flowpaths (Final Feedwater Portions)

Attachment 3 -

Normal Containment Cooldown Limits Attachment 4 -

Letter of Concurrence Attachment 5 -

Reactor Coolant System Arrangement Attachment 6 -

OTSG Level Ranges

Introduction -

Formation and Initiation Of AIT Background Oconee Units 1, 2 and 3 are Babcock and Wilcox Pressurized Water Reactors rated at 860 megawatts electric (2568 MWt). The units are identical with the exception of some shared auxiliary systems. Their onsite emergency power source is the two unit (87,500KVA each) Keowee Hydro Station. The facility is located on the shore of Lake Keowee in Oconee County, about 30 miles west of Greenville, South Carolin Unit 1 was licensed February 6, 1973, Unit 2 was licensed on October 6, 1973,.and Unit 3 was licensed on July 19, 197 On January 2, 1989, at 4:30 p.m.,

the licensee notified the NRC headquarters duty.officer of the following event:

Reactor Trip on Unit 1 from 100%.

During the normal surveillance on RPS channel "0", the I&C technician failed to follow procedure and bypass RPS channel "A".

Channel "A" was tripped due to a failed temperature sensor which feeds the reactor high temperature trip logi Following trip, high feedwater rates to "A" S/G caused level to reach hi-hi trip set point generating.a feedwater pump trip and an auto start of the emergency feedwater system. There was no overcooling of the primary loo EFW level control valve "A"- appeared to fail in the open position and was placed in manual contro Plant stable and in hot standb Licensee investigating EFW valve control failure and feedwater flow rate problem. RI informed by license On January 3, 1989 at 8:14 p.m.,

the licensee notified the NRC headquarters duty officer of the following event:

At 1919 hrs. with Unit 1 at 25% power, an explosion and fire occurred in the "ITA" 6900V feeder breaker generating a turbine

.trip and a trip of the "1A1" RC The fire spread into a potential transforme RX power reduced to 15% following the RCP trip and the unit was verified as being stabl At approximately 1930 hrs. the RX was tripped (Rx trip was actually 8:00 p.m. with RCP trip just after) and the plant was verified'

as stable. At approximately 1945 hrs. all RCPs were secured and the plant was stabilized in natural circulatio At 1945 hr an unusual event was declared. The fire affects only.Unit On At approx. 2013 hrs. the fire was declared out (some smouldering was still present). The licensee has positioned a team at the fire to assess damage and to prevent reflas No offsite fire support was needed. There were no injuries associated with this event. The SRI responded to the sit UPDATE 1:22 a.m. on 1/4/89 NOUE was terminated at 01:1 Plant is out of natural circulation mode with the 1B2 RCP restored to service. They are at 485 F and cooling down to cold Shutdow Formation of AIT On the morning of January 4, 1989, the Acting Regional Administrator, after a briefing by regional staff and consultation with senior NRC management, directed the.dispatch of an AIT. The team was to include the resident inspectors who had responded to the January 2 trip and January 3 fir C. AIT and Charter The charter for the AIT was prepared on January 4 and 5, 1989. The Oconee Section Chief (Reactor Projects,' RII) arrived at the Oconee site at 8:30 a-.m. on the fourth and was designated team leader when the AIT was formed later that mornin An entrance meeting was held with site staff and the team leader later in the afternoon of January 4. Other AIT members arrived onsite between January 4 and the evening of January 5. (An additional member reported onsite on January 11, to provide specific B&W technical knowledge to the team's evaluation of the natural circulation details).

A CAL was issued on January 4., 198 This CAL documented the understanding between the licensee -and the NRC that all equipment related to the ITA switchgear fire would be maintained in such a manner that-would not destroy or cause to be lost any evidence which may be needed to investigate or reconstruct the event. The AIT team leader would be informed prior to any troubleshooting activitie The charter for the AIT specified that the following tasks be complete.

Develop and validate a detailed sequence of events associated with the shutdowns of Oconee 1 on January 2 and January 3, 1989, and review the equipment failures that led to the shutdowns and those failures that occurred during and subsequent to the shutdown This should include but not be limited to:

pressurizer auxiliary spray; 6.9Kv breakers; main feedwater valves; reactor coolant pump recovery; and the management attention given to corrective action prior to the reactor startup on January.

Evaluate the significance of the equipment failures with regard to radiological consequences, safety system performance, safety significance, and plant proximity to safety limits as defined in the Technical Specification.

Evaluate the potential for fraudulent material as a contributor to the breaker failur. Evaluate the accuracy, timeliness, and effectiveness with which information on these failures was reported to the NR.

For each equipment malfunction, to the extent practical, determine:

a. Root cause b. If the equipment was known to be deficient prior to the event c. If equipment history would indicate that the equipment had either been historically unreliable or if maintenance or modifications had been recently performed Any equipment vendor involvement prior to or after the event e. Pre-event status of surveillance, testing and/or preventive maintenance The extent to which the equipment was covered by existing corrective action programs and the implication of the failures with respect to program effectivenes. Evaluate the effect of the failures on Units 2 and 3, if any, and the licensee's respons.

Evaluate the licensee's action taken to verify equipment operability on Units 2 and. Identify any human factors/procedural deficiencies related to the failure.

Through operator and technician interviews, determine if any of the following played a significant role in each failure:

plant material condition; the quality of maintenance; or the responsiveness of engineering to identified problem.

Evaluate operator action during the Unit 1 shutdown of January 2 and subsequent equipment recover.

Evaluate management involvement during the Unit 1 shutdowns and the subsequent recover.

Provide a Preliminary Notification upon initiation of the inspection and an update on the conclusion of the inspectio.

Prepare a special inspection report documenting the results of the above activities within 30 days of the start of the inspectio Persons Contacted

Licensee Employees at Exit Meeting M. Tuckman, ONS Station Manager C. Boyd, ONS.Design Engineering Group K. Brown, ONS Onsite Safety Review Group D. Deatherage, ONS Operations R. Gill, DPC Compliance P. Guill, DPC Compliance C. Harlin, ONS Compliance Engineer E. Legette, ONS Compliance J. McIntosh, ONS Station Services Superintendent B. Millsaps, ONS Mechanical Maintenance F. Owens, ONS Operations Coordinator G. Ridgeway, ONS Shift Operating Engineer Other Licensee Employees T. Cout~u J. Glynn L. Hawkins W. Knight M. Patrick R.

Repko Ml Scharf Design. Description. 6900/4160V Electrical Distribution Syste (FSAR Section 8.3.1)

The 6900V buses and associated switchgear for Unit 1 are located on the third floor (ground level) of the turbine buildin The, 6900V bus supplies power t~o the 6900V sw,itchgear TA and T This switchgear supplies power to only the reactor coolant pump Attachment One contains an elementary diagram of the Unit 1 6900V and 4160W system During unit startup and under shutdown conditions, startup transformer CT-i (located in the transformer yard, adjacent to the turbine building) supplies power to the 6900V buse It also supplies power (under these conditions) to the 4160V Main Feeder Buse Each of the two redundant MFB provide power to each of the redundant 416dV switchgear TC, TD TE. When the unit is at power (greater than about 200 MWe carried by unit generator),

the "normal'

power supply is utilized to provide power to both the 4160V MFB and the 690cV buses through the unit auxiliary transformer (IT).

Attachment One also shows that a power path is available to provide startup power from any unit's startup transformer; (This path was not being utilized and did not play a role in this event.)

The fire occurred in the normal supply breaker just after the operators had shifted the supply

to the 6900V switchgear from the startup transformer (CT-1)

to the output of the main,.generator (1T).

2. Main Feedwater and Auxiliary/Emergency Feedwater Flowpath Attachment Two is a simple diagram of the pertinent portions of the Emergency Feedwater and Main Feedwater System flowpaths to the two OTSG Each OTSG is 70 feet high with the auxiliary feed nozzles located 59 ft. above the lower (cold leg) primary piping bend. Feedwater (either MFW or EFW)

entering the OTSG through the auxiliary nozzles is sprayed directly on the tubes at a high elevation in the OTSG to provide a cooling effec The main feed nozzles are located 19 feet below the auxiliary nozzles and feed into the downcomer of the OTSG and is at 95%

on the OR. Fifty percent (50 %) on the OR is 11 ft. below the main nozzles. The OR begins 102 inches above the lower tube sheet or 96 in. on the startup rang % OR corresponds to about 250 inches on the startup range which is just under 40% of OTSG full. rang Natural circulation is normally achieved at the Oconee plant through 'operator manual actions and some automatic feature When.all RC pumps are tripped, the MFW system flow path is automatically realigned by the ICS such that MFW is provided to the OTSG via the auxiliary feedwater (upper) inlet header. This realignment of the MFW system is a unique feature of the Oconee plan In the event that the MFW pumps are tripped, the emergency. feedwater (EFW)

system is automatically actuated to provide feedwater to the OTS When feedwater is being supplied by the auxiliary nozzles and regardless of whether the feedwater is provided by either the MFW or EFW system, the feedwater is sprayed directly into the tube region of the OTSG at an elevation roughly 50 feet above the lower tube shee Feedwater flow is continued until the OTSG level is raised to 50% on the-operating range, which corresponds to an OTSG level of approximately 20 feet above the lower tube sheet, and thereafter is controlled to maintain this leve The purpose of spraying high in the OTSG is to raise the thermal center in the OTSG, thereby enhancing the cold driving head for natural circulatio If the system functions as designed, the natural circulation flow would result in a core delta T of approximately 40 to 50 degrees F following a trip from 100% powe During normal operation above 15% power the startup block valves (FDW-33,42) and the main feedwater block valves (FDW-31,40) are open and main feedwater flow is controlled by the ICS. The ICS regulates feedwater flow to match electrical megawatt demand by controlling the main control valves (FDW-32,41)

and controlling main feedwater pump speed to maintain a constant 35 psid across the main control valve At less than 15% power, the startup

feedwater control valves (FDW-35,44) control OTSG level at about 25 inches on the startup range which is the programmed LEVEL setpoint).

The main feedwater block valve functions are interlocked with the startup control valves when the block valve control switch is in automatic. Then whenever the startup control valve demand is less than 50%,

the block valves will shut and when the startup control valve demand is greater than 80% (open),

the main blocks ope This means that the main feedwater block valves should close on a RX trip with their control switch in automati This is caused by the startup valves being automatically positioned by the ICS to their LEVEL set poin Upon a loss of all RCPs (assuming main feedwater is still available) in order to promote natural circulation the ICS should automatically cause the following configuration changes: The startup line valves to the auxiliary feed nozzles (FDW-38,47) ope i The startup line isolation valves (FDW-36,45) shut securing flow to the lower or main feedwater nozzle i ii. Main feedwater will 'feed into.the auxiliary feed nozzles and control OTSG level at 50% on the O i During the feedup of the OTSG to this 50% OR level, the startup control valves are automatically positioned at approximately 40% open to rapidly increase the leve During this event, the four valves discussed above that should have automatically shifted did not shift positio Additional details are contained in paragraph IV C of this repor The Emergency Feedwater System has two automatic start signals:

both main feedwater pumps with low hydraulic oil pressure or both main feedwater pumps with. low discharge pressur Either of these signals starts all three EFW pumps (one turbine driven and two motor driven).

The OTSGs are fed via the auxiliary nozzles by the Emergency Feedwater System (through valves FDW-315,316) on its separate level control circuitry. On a loss of all RCPs,. if EFW is initiated, it will feed up the OTSGs to 50% OR level which is the natural circulation setpoin.

Reactor Coolant System and Auxiliary Pressurizer Spray The Oconee reactor coolant system is composed of two OTSGs each being force circulated by two RCP A pressurizer is located off of the B hot leg and the normal charging flow enters both of

the A loop cold leg The relative elevations are shown on Attachment Pressurizer levels are: 260 inches-high alarm; 220 in. -

normal operating level; 200 i low alarm; and 80 i heater cutof The normal pressurizer spray line in Unit One is connected to the discharge line of the 1A1 RCP. The Oconee units also have an auxiliary pressurizer spray capability. This auxiliary spray originates at the High Pressure Injection pump discharge lin The HPI pumps also serve as the normal charging pump The auxiliary spray was designed to be used during the later stages of a normal. cooldown and during an emergency cooldown when normal pressurizer spray is lost. The control of this auxiliary spray formerly required manual operation of a valve located in the penetration room and a Reactor Building entry to manually align valve Recently a Station Modification was completed which permits control of auxiliary spray from the control roo The manual valve in the penetration room was replaced with a fail-closed air operated globe valve (HP-355) which is operable from a manual loader located on Unit Board One in the Control Roo No reactor building entry is now normally. required to initiate auxiliary spra This modification has provided the operators another means of reducing RCS pressure and has facilitated the use of auxiliary spray during plant cooldown Technical Specification 3.1.2.6 prohibits use of pressurizer spray if the spray -

pressurizer differential temperature exceeds.410 degrees F. Since the LDST temperature is normally about 100 degrees F, this TS effectively limits the use of auxiliary spray to pressures less than about 740.psig. For most of this event, auxiliary pressurizer spray was unavailable due to a mispositioned manual isolation valve in the reactor building. Additional details of auxiliary spray involvement in the event are contained in paragraph IV. D of this repor. Thermal Shock Operating Range Attachment 3 of this report is a copy of the Normal Containment Cooldown Limits including the Thermal Shock Operating Range (TSOR) Instruction This is an enclosure to the Oconee Emergency Operating Procedures. This TSOR is the mechanism by which the licensee ensures that the Pressurized Thermal Shock Limits (PTS) are not exceede Basically, PTS involves the reactor vessel undergoing a severe overcooling so that the RCS temperature is excessively low relative to the reactor system pressure. If the overcooling is significant, some risk of brittle fracture of the vessel exists especially if a repressurization occurred. PTS is of concern to

the Oconee units since they are older plants and have a larger degree of radiation induced embrittlemen Relatively newer plants which have less copper and phosphorus in the reactor vessel welds are less susceptible to embrittlemen If the plant conditions dictate, then a.three hour hold in the TSOR is required. This hold or soak allows the temperature gradient which exists across the vessel wall to equaliz This equalization of the gradient reduces the stress-on the vessel wall thus decreasing the chance of a preexisting flaw propagating through the vesse II. Description of Events Event Descriptions 1. January 2, 1989 Reactor Trip At 3:23pm on January 2, 1989, Oconee Unit 1 tripped from 100 percent power. The trip was caused by personnel error durin Reactor Protection System (RPS)

functional testin (Procedure IP/1/A/305/3:. Nuclear Instrumentation and RPS Channel Calibration and Functional Test ).

A test of RPS channel D was in progres Channel A of RPS was placed in the tripped condition during most of the testing.(as required by TS) because two dummy bistables were in place in the channel A circuitr The dummy bistables were installed in the high hot leg temperature circuitry and the temperature-pressure circuitry in channel A due to a failure of a hot leg RT Installation of dummy bistables is permitted by Technical Specification A channel containing dummy bistables is required to be tripped prior to placing another channel in a "bypass" condition for testin Before starting the portion of the test involving the channel D Control Rod Drive Breakers, channel A should have been taken out of the tripped condition but was no As a result when channel D was tripped by the technician, this resulted in a 2 out of 4 channels tripped condition, which caused the reactor to trip as require While the licensee's investigation into the cause of the trip is still in progress, indications at this point are that the principle cause of the trip was a personnel error by the I&E technician that was conducting the tes Additionally, it appears that some changes could be made to the procedures involved which may have prevented this erro This testing had been completed successfully several times with the existing RPS configuration and procedures. While resolution and corrective actions to prevent reoccurrence are important, any equipment failures/malfunctions that may have played a role in the events

of January 3 were the areas of emphasis to the tea Although the plant was stabilized quickly, several equipment problems of concern did occur:.

One Main Steam Relief Valve did not reseat until pressure was reduced by the operators to about 960 ps Apparently the valve became unseated again when the operators let pressure increase and consequently was reseated by the operator Due to the testing that was in progress, the Reactor Demand and Feedwater Demand portions of the Integrated Control System (ICS)

were in the manual mod After the trip was initiated, this would have prevented the normal automatic reduction in feedwater flow by the ICS in the tracking mode. Since the BTU limit portion of the ICS is located between the Feedwater Demand Station and the Feedwater valve, control circuitry, it should runback the feedwater flow signa (BTU limit is basically a superheat monitor,,

utilizing OTSG pressure, hot leg temperature, feedwater temperature and RCS flow to ensure superheating of steam is occurring. It will decrease feedwater flow if inadequate superheating is presen BTU limits are in effect when reactor power is less than 25 percent).

While the B

feedwater was successfully decreased by the B BTU limit feature, the A

side BTU limit was not actuated and subsequently did not reduce the A side feedwater flow signa Investigation by Instrument and Electrical personnel revealed a.fault.in a multiplier module in the A BTU limit circuitry. 'The module was replace Since the A OTSG feedwater flow remained high, the OTSG rapidly reached its high level setpoin This caused a trip of both MFW pumps and an Emergency Feedwater (EFW)

System actuatio The A OTSG EFW flow control valve (1FDW-315)

opened fully and remained open when it should have been shut due to the high OTSG leve The operators immediately recognized this situation, selected manual on 1FDW-315 and controlled EFW flow. 1FDW-315 is controlled automatically by an OTSG level circuit. This level control circuit has 'two channels; primary and backu The backup channel was selected (switch in control room) prior to the event because a spurious fault in the primary level circuitry had been observed about a week earlie During this event the operators did not consider the primary level control channel to be reliable and consequently remained in manual control of 1FDW-315 vice selection of the primary level channel when the backup channel faile A malfunctioning circuit card in the backup level control circuitry of 1FDW-315 was found to be the cause of the problem. The card was replaced and the affected portions

of 1FDW-315 control circuitry were successfully retested before Unit 1 was returned to powe The A main and startup feedwater block valves did not shut automatically on the trip as they should hav The operators noted this and shut the valves by operation of the control room switche The exact mechanism of this, failure was not understood prior to the restart of Unit 1, but maintenance and management personnel had responded positively to the resident inspectors' questions in this area and were actively pursuing the issu This problem had recently occurred following two trips of Oconee unit 3 on November 14, 1988. After the failure had occurred on Unit 3 (prior to Unit 3 restart) the inspectors had bee informed by the licensee that the automatic closure of these block valves was not a safety related feature nor was it required for accident mitigation considerations. After the fire and subsequent trip of Unit 1 on January 3, more progress was made on a resolution of this issu Further details are discussed in section IV.C of this repor Unit 1 was returned to criticality on January 3 at 2:00a.

January 3, 1989 Event On January 3, 1989, an event occurred at Oconee Unit 1 which began due to a fire in switchgear 1T At about 7:16pm, with the reactor at approximately 25% full power and the main generator output at 200 MWe, the operators transferred the KV auxiliaries (switchgear 1TA and 1TB)

from their offsite source (CT-1)

to their main generator output feeder (1T)

in accordance with OP/1/A/1107/0 Approximately one minute after the transfer was completed the following indications were received: the turbine tripped on a main transformer (1T)

lock out; RCPs 1A1 and 1B1 tripped; a fire alarm annunciator lit; and reports of fire and smoke in the turbine building were phoned into the control room. The fire brigade was dispatched to the scen (Units 2 and 3 supplied most of key brigade members). 1TA was isolated from 6.9 KV sources and the reactor ran back to 14% (generator tripped). _After some review of electrical diagrams available to the control room operators, DC power was secured to 1TA (approximately 10 minutes later).

The fire brigade initially tried to extinguish the fire with C02 and dry chemical extinguishers. An Unusual Event was declared at 7:45p During this time the operators were reducing reactor power by manually driving control rods inwar At 15 % power, T-cold is 579 F and at 0 % power, T-cold is 532 At about 7:55pm, the decision was reached to activate the TSC and OS At approximately 8:00 p.m.,

it was decided that water would be used to put out the fir The reactor was manually tripped (from about 4% power) and then RCPs 1B2 and 1A2 were tripped in order to de-energize the 1TB switchgear prior to using water on the fire in ITA. The operators had not tripped the reactor earlier as they were minimizing the cooldown tran sient by slowly reducing reactor power and assuring the main feedwater pumps were shifted to auxiliary stea ICS did not actuate the valves to shift main feedwater flow to the emergency feedwater flowpath on the loss of the RCPs. (The upper nozzles in the OTSG, sprays feedwater on the tubes to promote natural circulation.)

The operators positioned these four valves via Control Room switche Additionally, the startup feed control valves did not feed up the OTSGs to the 50% OR level automati cally. The operators took control of these valves and raised OTSG level, as they recognized it was important to get level up to promote natural circulation and to limit the (increasing)

RCS pressur The operators had caused a second HPI pump to start (opening valves HP-24 and HP-26 caused the pump to start automatically) as pressurizer level decreased to 170 inches just before the trip (as power.and temperature decreased).

At about 8:05pm, station management was calle At about 8:25pm, the fire was reported ou The plant appeared to be stable on natural circulatio At about 8:30pm, when the operators placed the main feedwater control valves in manual, indications are that the A MFW control valve backed off its seat and feed to the A OTSG was.increased. Both MFW block valves had to be shut by operating their breakers since some problems were encountered when attempts were made to shut them from the C At 8:36pm, both MFW pumps tripped on A OTSG high level and EFW was actuate No problems with EFW were note Natural circulation continued with the A loop Cold leg Temperature (Tc)

decreasing from 460 to 400 degrees F as the TBVs were opened to steam the OTSGs. The high OTSG levels (A-95%, B-50%

OR), RCP seal inleakage, and HPI flow all added to the coolin The residents arrived onsite at about 8:50pm, within 15 minutes of notification of the even In-core exit thermocouple readings slowly decreased to about 524 degrees F at 9:00pm, 512 degrees F at 9:45pm, 501 F at 10:.03pm, 493 F at 10:45pm based on resident inspectors observation Pressure remained high at about 2100 psig. At about 9:00pm, 1TB was energized from CT-After careful verification of prerequisites RCP 1B2 was restarted at 10:03p Tc was noted at about 488 degrees F immediately after the pump was starte Thermal Shock Operating Range (TSOR)

discussions were held in the TSC at about 9:30-10:00pm. The EOPs discussed using T-cold as the only parameter to monitor for the cooldown, but the procedure basis suggested that with natural circulation conditions and excessive HPI flow that incore exit thermocouple temperature was the recommended parameter. As a precaution, a three hour soak was started at Tc of 488 degrees F, pressure 2142 psig. Auxiliary spray was not yet available. (In parallel with the above actions this was an option being investigated).

Since auxiliary spray was not available and the 1B2 RCP could not provide sufficient "normal" spray flow, combined with management's reluctance to open the PORV, the soak was completed above the TSOR envelope. (Another factor in this decision was that the controls for the PORV are in a cabinet in the back of the CR.)

Due to problems reestablishing RCP seal return flow, the recovery of the 1A2 pump was delayed until about 3:25a The unusual event was.terminated at about 1:20am after a RB entr disclosed that manual valve 1HP-340 was incorrectly shut in the auxiliary spray flowpath. At about 2:30am, the operators did utilize auxiliary spray (for less than 5 minutes) to reduce RCS pressur (1980 to 1870 psig).After the 2nd RCP was restarted, a normal cooldown was commenced to cold shutdown. (A cooldown of about 25 degrees F/HR had been initiated earlier).

Throughout most of the event the resident inspectors monitored licensee actions. The resident inspector observed CR operations while the SRI was in the TSC and coordinated with licensee management and NRC personne B. Detailed Sequence of Events - January 3 Event 1-3-89 7:16pm

-

Rx. Power approximately 26%, 200 MWe-6.9 KV auxiliaries transferred from CT-1 to 1T in accordance with OP/1/A/1107/02, shortly thereafter received transformer 1T differential phase and generator lock ou :16:37

-

Turbine Tripped 1A1 & 1B1 RC Pumps Tripped (The feeder breaker to 1A1 RCP tripped open; the Operators manually opened the feeder breaker to. the 1B1 RCP since zero amps to the RCP was indicated in the CR.)

Received a fire alarm annunciato Received telephone reports of a fire and explosion in the vicinity of the 6.9 KV switchgear in the Turbine Buildin The reactor ran back to 14% powe :17

-

The Fire Brigade was dispatched to the 6.9 KV Switchgea Control Room personnel verified that 1TA was isolated from 6.9 KV Source :29

-

Removed ITA DC control power @ 1D1A and ID1 :33 -

7:41

-

Used C02 extinguishers in an unsuccessful attempt to put out the fire. Then an attack was made on the fire using dry chemical extinguishe This too was unsuccessful due to high heat causing reflas :45

-

Shift Supervisor (Emergency Coordinator) declare Unusual Event because the fire had not been extinguishe :49

-

Shift Supervisor (SS). made decision to activate the Technical Support Center (TSC) and Operational Support Center (OSC) to obtain additional technical suppor :56

-

The second (lA) HPI Pump started automatically, on low seal injection flow after operator throttled open HP-26. Rods were driven inward to reduce power from 15% to about 4%. This was causing a reduction in RCS temperature and inventor Pressurizer level was about 200 inches and decreasin HP-120 operation appeared to be sluggish and errati :57

-

SS requested personnel support from off-duty shifts as a backup to the ONS Fire Brigad :58

-

1HP-24 was opened, 1HP-26 was opened mor Pressurizer level was still lo :59

-

The Fire Brigade leaders with concurrence from the Shift Supervisor decided it would be necessary to fight the fire with wate :59

-

The reactor was manually tripped in preparation for de-energizing all KV Switchgear for personnel safety of the Fire Brigade and to prevent the possible lock out of CT-1.(prior to using water to fight the fire).

8:02

-

Pressurizer spray valve opened at 2205 psig setpoin Pressure was apparently continuing to increase due to HPI flo (Since RCP s were secured no spray flow occurred.)

8:02:22

-

Tripped the 1B and 1A RCPs S-De-energized 1TB 6.9 KV Switchgear (manual disconnect)

1. The feedwater valves from the Main Feedwater System to the SG Auxiliary feedwater nozzles did not open automaticall. The Start-up Feedwater Control Valves did not automatically increase OTSG level to 50% on the Operating Rang :04

-

RPS high pressure trip setpoint (2355psig) was reache :04

-

Turbine bypass valves placed in manual and throttled about 10% ope :05

-

The operators opened 1FDW-38 and 1FDW-47, shifting Main Feedwater to the Auxiliary Feed ( upper) nozzle Valves FDW-36 and 45 incorrectly remained open and feed actually could enter both sets of nozzle :05

-

Turbine bypass valves placed in auto and went close (MS pressure was low, about 800 psig)

8:05 -

8:11

-

The Operators used the Startup Feedwater valves to increase OTSG level to 50% on the operating range. In retrospect they fed at a higher than required rate in order to limit peak RCS pressure. It appears that RCS makeup rate was highe.r than necessary, contributing to a peak RCS pressure of about 2395 psig at 8:07pm Inadequate Core Cooling Monitor (ICCM)

thermocouples read about 570 degrees F and were fairly stead Indicated cold leg temperatures decreased from 545 degrees F to 485 degrees :06 1HP-26 was throttled to the full open positio :06:56

-

1HP-26 was closed (the opening was unintentional).

8:06:57

-

TA HPI pump was taken off and placed in Aut (Pressurizer level was about 192 inchesat this time.)

8:10

-

The NCOs at tempted to use HPI Aux. Pressurizer Spray to reduce RCS pressure, but could not get flo (Manual valve later found shut.)

8:10 dA HPI pump was restarted automatically, apparently due to the sluggish operation of HP-12 :12 mA HPI pump was stopped and placed in Aut Feedwater flow had been reduced due to reaching 50%

level on OR and RCS pressure was increasing *as a result of the second. HPI pump on with the normal makeup valve (1HP-120) ope HP-26 was not utilized

at this tim RCS pressure reached 2300 psig during this evolutio :15

-

Fire was reported to be out, additional personnel began arriving at sit :20

-

The control room operators shut FDW-36 and FDW-45 to isolate feedwater flow to the main feedwater nozzle :22

-

Turbine bypass valves placed in manual and throttled to 15% ope RCS pressure was stabilized at about 2100 psi :23

-

1HP-24 was close :11 -

8:31

-

The Operators maintained 50% SG level on the OR with the Startup Feedwater valve :29:45

-

A&B Main FDW Control valves were placed in manua When "A" Main FDW Control valve was placed in manual, it apparently backed off its closed seat and overfeed of the "A" SG bega :31 - 8:43

-

The NCOs pressed the close pushbutton on both Main FDW Block valves (FDW-31 and FDW-40 both valves started to clos FDW-31 (A MFW Block) started to close and travel stoppe The valve was closed from the breaker at about 8:36:4 FDW-40 cycled closed and immediately began to reope This valve cycled open and was closed by the NCOs two more time Then the NCO noticed that the valves

"Open" pushbutton was stuck in the depressed positio The valve was closed and the breaker opened to prevent trave :36

-

The "lA" Main Feedwater Pump tripped on high "A" SG level (94% OR)

stopping the overfee The 1A and 1B MDEFW Pumps and the TDEFW Pump started automaticall The emergency feedwater flow control valves (1FDW-315 and 1FDW-316) functioned correctly. RCS conditions at this time were 462 degrees F cold leg and 2150 psi :36 -

10:05

-

The NCOs maintained RCS natural circulation flo Indicated cold leg temperatures slowly decreased from 460 degrees F to 400 degrees F due to Turbine Bypass Valves being open to 10%, the high SG levels ("A" SG @

95% OR and "B".

SG at 50% OR),

RCP seal inleakage from HPI, and normal HPI makeup. Incore T/C temperature decreased slowly from 570 degrees F to 530 degrees F after the RC Pumps were stoppe The NCOs further

opened the Turbine Bypass Valves. to lower SG levels and the I/C T/C temperature slowly decreased to 510 degrees :28

-

Transmission Dep commenced checkout of 1TB and determined that 1TB was not affected by the fir :50

-

The Resident Inspectors arrived in the control roo :03

-

1TB 6.9 KV switchgear was re-energized from CT- :1 The TSC was established and the Station Manager relieved the. Shift Supervisor as Emergency Coordinator.. A discussion was started in the TSC about whether or not operation within the TSOR envelope was required due to the decrease in indicated Tc following the point at which the RCPs were secure At 2150 it was decided that operation in the TSOR was not required. It was decided that a 3 hour3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> soak would be performed at stable RCS pressure and temperature after RCP restar :03

-

1B2 RC Pump was restarted. RC Pressure stabilized at 2100 psig and indicated Tc increased to 490 degrees 10:10 I/C thermocouples read approximately 491 degrees F, Tc 488 degrees F, Th 489 degrees F, and RC Pressur psig. These conditions were maintained-for about 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> to stabilize the RC :05am

-

The three hour soak was completed and a gradual cooldown of the RCS was initiate Pressure control of the RCS was difficult due to lack of driving head to the normal pressurizer spray line with only the 1B2 RC Pump operatin :18

-

The Unusual Event was terminated by the Emergency Coordinato :00

-

1A Condensate Booster Pump was started to feed the SG and the EFW pumps were secure An entry was made into Unit 1 R Bld and valve 1HP-340 was found shut which had prevented flow to pressurizer auxiliary spray. The valve was reopene :30

-

The NCOs opened 1HP-355 and established HPI auxiliary spray flow to the pressurizer for about five minute The RCS pressure decreased from about 1980 psig to 1870 psi Pressurizer auxiliary spray was then isolate /3

-

1/4 10:00-3:25am

-

Attempts to reestablish seal return flow to 1A2 RCP continue HP-277 was opened to reduce seal back pressur The LDST was degassed to further reduce back pressure. The seal return block valves from the other two idle pumps (1A1 and 1B1) were close Cycling the AC and DC oil lift pumps on 1A2 RCP finally established sufficient seal return flow to start the pum :25am

-

1A2 RCP was started and a normal cooldown commence III. Firefighting and Fire Protection Aspects Fire Fighting Techniques The inspector reviewed the fire fighting actions taken by the station fire brigade to extinguish the fire in electrical bus ITA by interviewing station personnel involved in the fire fighting activities. In addition, the inspector reviewed the station Fire Plan which provides the required response to a fire in any plant are The following paragraph describes the fire fighting activities of the fire brigade between 7:16pm and 8:15pm on January 3, 198 At approximately 7:16pm, the Fire Brigade Captain (FBC)

and fire brigade members assigned for the night shift were notified of a fire in the Unit 1 Turbine Building. The fire was detected by a smoke detector alarm which annunciated in the Control Room and a concurrent telephone call from a worker in the Turbine Building who heard an explosion in the area of the 6.9 KV switchgear. At this time the FBC responded to the Unit 1 Control Room and was told the location of the alar The FBC then went to the fifth floor elevator lobby fire equipment storage area where the remainder of the fire brigade was assemblin After donning fire protective clothing and obtaining necessary equipment, the FBC responded to the 3rd floor elevator lobby where the initial command post was establishe At this time the FBC and two other brigade members entered the Turbine Building to investigate the fir The FBC and the two brigade members proceeded to electrical bus IT Approximately 100 feet from the fire source, this team was confronted with heavy smoke which extended from the floor to the ceiling of that elevation of the Turbine Buildin At this time, the FBC also heard steam being discharged and was concerned that a steam leak existed. (The steam noise was from the steam jet air ejectors which were open.)

Upon arriving at electrical bus 1TA, the FBC discovered a fire burning in the normal supply breaker to 1TA. The door of the panel was open and fire and heavy smoke was coming from the compartmen The FBC decided to attempt to extinguish the fire at this time by using carbon dioxide (C02). A 150 pound wheeled C02 fire extinguisher was brought to the scene and the 1-1/2 hose line closest to the panel was charged with wate The team then discharged the entire contents of the C02 extinguisher into the cabinet. Initially the flames of. the fire were

extinguished by the C02, however, reflash occurred. The FBC and the two members brought a second 150 pound C02 wheeled extinguisher to. the are The FBC instructed the team members to use this extinguisher to contain the fir At this time additional fire brigade members were used to relieve the original two members and FBC who were involved in the fire fighting to that point. The FBC requested that a 150 pound dry chemical wheeled fire extinguisher be brought to the fire scene. At approximately the time the second C02 extinguisher was emptied, the dry chemical extinguisher was placed in servic It was at about this time that the FBC requested permission to use water from the 1-1/2 inch hose line to extinguish the fir The dry chemical was also ineffective in extinguishing the fir Subsequently, with the approval of the Shift Supervisor, (the reactor was tripped and the RCPs secured to de-energize 1TA) the FBC directed the brigade members to extinguish the fire using water. The water was very effective and the fire was quickly extinguished at approximately 8:15p Approximately seventeen employees responded to the fire scene and assisted in the emergency. The fire burned for about one hour and assistance from off-site local fire departments. was not requeste During the firefighting activities, communications between the fire scene and Control Room were conducted using portable radios. Initially when the FBC and two fire brigade members entered the Turbine Building,, they were unable to communicate with the Control Room due to the high noise in the area and some dead spots where radio communication was not possibl During this period of time when communication between the FBC and Control Room was not possible, the Safety Officer (SO),

who is second in comman at the fire scene, moved the command post from the 3rd floor elevator lobby to the Unit 1/Unit 2 Block House. The SO moved the command post in order to establish better communications with the Control Room and to stage additional fire fighting equipment closer to the fire scene. During communications between the Control Room and FBC/SO, the status of the fire fighting activities, the potential impact on the plant of fire fighting activities, and fire fighting equipment/personnel needs were discusse Overall the fire brigade performance in fighting the fire was satisfactor The FBC performed his duties well and in accordance with his trainin However, the following observations were noted:

-

Initially upon entering the fire scene the FBC was unable to communicate with the Control Room or the remainder of the fire brigad The FBC should have considered establishing a means o communication, (runner system) during the early stages of the fir A total of seventeen fire brigade members responded to this fire emergency. Of these seventeen, the FBC only involved two brigade members in the initial fire fighting. This may have been the result of the FBC inability to communicate with the Control Room and S The FBC should have used additional manpower during the initial

firefighting to man the backup hoseline and stage additional fire extinguishers at the fire scene so firefighting activities could continue uninterrupte As part of the licensee's review of the fire event, a critique of the FBC and fire brigade-members performance will be conducte Any deficencies noted during that critique will be resolved by addressing them in the quarterly fire brigade training. This should be adequate to resolve the few minor discrepancies noted by the inspecto The inspector did identify two apparent programmatic weaknesses -in the Fire Protection Program during the review of this fire even These two programmatic weaknesses are discussed in the following paragraph In discussing the fire fighting activities with the plant personnel, the inspector expressed the concern that there seemed to be a significant delay in applying water to the fir The FBC and Shift Supervisor (SS)

stated that they delayed using water based on the following safety and operational considerations:

-

..The FBC was attempting to contain the fire until it was verified that power (6.9KV) had been removed.from the electrical bu The FBC and SS were concerned that if water was used, damage could have occurred to electrical bus 1TB which supports the two remaining RCPs which were still runnin The SS stated that he did not want water damage to 1TB to force him into natural circulation cooldown, but instead he wanted to enter natural circulation through a controlled process by manually stopping the two remaining RCP In addition the SS was concerned that water damage to 1TB could cause transformer CT1 to be locked out. This would require switching ove to the Keowee Hydro Station for backup power and a short period of time without power to Unit 1 during the transfe The FBC and SS decision to delay using water based on the above concerns and the the fire brigade performance under the circumstances of this particular fire was appropriat However, in discussions with plant personnel and management it appeared to the inspector that the licensee's staff may be overly apprehensive about the use of water on any fire involving energized electrical equipment or cables. This concerned the inspector since in some fire situations it may not be possible to de-energize equipment or circuits prior to initiating fire fighting or the delay resulting from attempting to de-energize equipment or circuits may allow the fire to grow significantl Based on this concern, the inspector reviewed the licensee's fire brigade training program to verify adequate training.is provided on fighting fires involving energized electrical equipment. The licensee's training program includes a classroom discussion, Lesson Plan GOFP-008, which specifically covers the use of fire streams on electrical curren This training is provided during the initial training for all fire brigade members and is

covered during requalification training conducted over a two.year perio Actual hands-on fire fighting of fires involving energized electrical equipment is not provide The NRC guidance which describes the fire brigade training requirements for Oconee are contained in:

Appendix A to Branch Technical Position APCSB 9.5-1 and Nuclear Plant Fire Protection Functional Responsibilities, Administrative Controls, and Quality Assurance, August 19, 1977 lette Neither of the.se documents specifically require licensee's fire brigade training programs to. include practice on live fires involving energized electrical equipment. However, Attachment N, Fire Brigade Training, of the August 19, 1977 letter states that practice on actual fir extinguishment should be provided for fire brigade members on proper methods of fighting various types of fires similar in magnitude, complexity, and difficulty as those which could occur in nuclear power plants. All licensee's were requested to compare their programs with this guidanc By letter dated January 16, 1978, Duke Power compared their program to the guidance in Attachment No. 2 of the August 19, 1977 letter and concluded the Duke program met the NRC guidance with the exception that practice sessions would be provided every two year The NRC accepted the licensee's comparison and this is documented in the NRC Safety Evaluation Report (SER) dated August 11, 197 In reviewing this. event, the inspector also questioned the licensee's personnel as to why the off-site fire departments were not notified that their assistance may be require The inspector was informed that per procedure off-site assistance from local fire departments is only requested for fires outside the protected area fence. The inspector also found that the Fire Plan section 5.C specifically states offsite fire departments will only be used for fires outside the protected are In addition, the inspector found that approval of.this provision of the Fire Plan appeared to be granted in the NRC SER dated August 11, 197 The acceptance of this provision is. clearly not in conformance with the guidance provided in the NRC letter dated August 19, 1977. Attachment N of this guidance addresses the need for Fire Fighting Procedures which coordinate fire fighting. activities with off-site fire department The guidance in Attachment N is provided to insure coordination between off-site departments and the licensee's operations staff in fire situations which can not be brought under control by the station five man fire brigade. This coordination ensures that in these fire situations additional manpower from the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> manned off-site fire departments can be readily brought onto the plant site to assist the station fire brigade if necessary. In addition, the inspector found that no procedures were in

place outside the Fire Plan which would provide a means of readily getting off-site fire departments into the protected are By letter dated January 16, 1978, the licensee compared their Fire Plan with the guidance in the August 1977 NRC documen In this letter the licensee states that it is their policy that off-site fire departments will not be called for fires inside the protected area fenc Subse quently, the August 11, 1978, SER was issued by the NRC which in Section 6.0 accepted the licensee's January 16, 1978 submittal (Note:

The August 11, 1978 SER contains a typographical error in paragraph 6.1. The date of January 16, 1978 appears as January 6, 1978).

In response to this concern the licensee generated a: revision to Section 5.C of the Fire Plan to state that off-site fire department assistance may be. used at the discretion of the Shift Superviso The licensee also stated that onsite training has been conducted with the off-site fire department IV. Equipment Status and Role in Event Normal Supply Breaker to 6900V Switchgear 1TA Equipment As Found After the fire was extinguished,. initial inspection of the 1TA switchgear showed heat and smoke damage to the external components. Additionally, the door to the 1TA Normal Supply Breaker compartment was attached by two fastening screw The hinges had been blown off by the explosio Internal inspection of the breaker compartment revealed extensive fire damage. The control section for the breaker was also heavily damaged by fire and hea The adjacent cubicles also suffered fire and heat damag The 1TA Normal Supply Breaker compartment also suffered mechanical damage as a result of the explosion. Although there was an explosion and fire, the breaker remained close After the breaker was removed from the cubicle, severe damage was note The insulators and load side connectors had vaporized and there were large holes. in the metal separator between the buses as well as the fire and heat damag.

Status of Investigation The affected breaker was the Normal Supply Breaker to the 1TA switchgear. This breaker was manufactured by ITE in 1969 and was a Type 7.5 HK Air Circuit Breake The breaker was rated for KV and 2000 amps and its Serial Number was 44349-CD1-1-4C. It was located in the second cubicle of the switchgea The licensee had vendor representatives from ASEA Brown Boveri (ABB)

inspect the breaker and switchgea The determination by the vendor representatives was that the fire lasted too long and the damage was so severe that a single cause could not be identified. Some possible causes were developed and are under evaluation. The most prevalent theory is that there was a cracked insulator which led to the failur A review of the maintenance history, as far back as 1981, revealed that preventive maintenance had been performed as schedule In September 1987, Doble insulation tests were performed and did not indicate any cracking or degradation of the insulator No maintenance had been performed on the breaker that required any replacement parts and no modifications had been made to the control circuitry for this breake Cable damage was also present in the RCP 1A1 cubicle adjacent to the Normal Supply Breaker. Some of the damaged cable was used to provide a signal to the IC-The damage appears to have provided erroneous information that one RCP was running, thus preventing the ICS to function properly (see Section IV C).

Inputs from the RCPs to RPS may have been affected also, however, none would have interfered with the ability ofthe RPS to function. No other interfaces have been identified to have been adversely affecte.

Corrective Actions Due to the extensive amount of damage to the ITA Normal Supply Breaker and its.cubicle, no root cause will be able to be determine The lack of performing maintenance, incorrect maintenance, or the use of improper parts may be ruled out as contributing factor The licensee is continuing to evaluate the situation in order to determine to the most likely cause of the even The licensee has commenced corrective actions for the repair/replacement of the switchgea ASEA Brown Boveri of Texas is fabricating the cubicle based on the original desig Should any component require replacement due to obsolescence or unavailability, design Engineering will evaluate and approve such changes. The replacement breaker and cubicle are scheduled

for installation on January 16, 1989, and testing will commence later that wee The control section of the 1A1.RCP is also being refurbished with new components as well as being totally reviewe This work is being done by the license Cabling Above The ITA Normal Supply Breakers 1. Equipment As Found There are four cable trays that run directly above the ITA switchgea These cable trays are B-103, B-104, B-105 and B-115. There was no identified damage in tray B-10 Trays B-104 and B-105 had heavy soot deposits making it difficult to identify any damag The( licensee's initial inspections identified seven safety related cables in tray B-105 but did not identify any apparent cable damage. Meggar tests on the cables in tray B-115 did not reveal any damage. The meggar test was performed because the cables were in conduit and not accessible for physical inspectio Inspection of the cables was stopped at the request of Operations. The request was made for safety and operational considerations. The inspection was scheduled to resume after the refueling canal was filled on January 12, 198.

Inability of Licensee to Quickly Identify Cables Although the event took place on January 3, 1989, the licensee was not able to provide a list of any type.to the inspector on January 4, 1989, and a complete list of cables in the cable trays was not provided until the morning of January 10, 198 This delay was due to inaccuracies in "As-Built" drawing Drawing N F, Re,

Electrical Equipment Layout Turbine Building Below E '

+ 0" Cable tray Sections, did not include all cables in the affected trays and indicated that additional cables should have been in the tray In order to determine what cables were in the trays, cable sheets had to be used, which was a time consuming proces The licensee identified eight safety related cables in tray B-105 by visual inspection and then verified them with the cable sheet From the information -provided, there are no additional safety related cables in trays B-105 and B-10 Had the drawings been accurate, the licensee would have been able to quickly identify potentially affected cables and commence an evaluation several days soone The licensee would have performed a physical inspection to verify the cables and to ascertain any damage to the cable. Significance of Potentially Damaged Cables The following is a list of the safety related cables in tray B-105:

CABLE N FROM TO 1EB1T513 B1T-1 UNIT 2 ES NORM. CONT. CA ETC103 1TC-1 B2T IETC102 1TD-1 B2T 1ETE103 1TE-1 B2T IEB1TI15 B1T-1 1EPSLPI 2EB1T809 B1T-8 UNIT 2 ES NORM. CONT. CA EB1T1110 B1T-11 2EB4 Damage to any of the above cables was evaluated not to be of safety significance. Damage to 2EB1TI110 could possibly result in involving Technical Specifications for Unit 2, but would not affect Units 1 and. Other Drawing Deficiencies While disc.ussing how the DC Control.Power to the 1TA switchgear was identified, drawing discrepancies were identified by the inspecto The discrepancies were differences between the one-line drawings and the elementary wiring drawings (OEEs).

The licensee reviewed all switchgear OEEs, load center OEEs, and safety related 600 volt and 208 volt Motor Control Center OEE From this review, eighteen errors were identified. The licensee has initiated SPRs to correct the identified discrepancies and to review the non-safety drawing This is identified as Inspector Followup Item IFI 50-269,270,287/89-03-03, Discre pancies Between One-Line Drawings and Elementary Wiring Drawing. Corrective Actions The licensee has committed to replacing and/or splicing all safety related cables above the ITA switchgear in tray B-10 Additionally, all non-safety related cables in tray B-105 will be meggared to ensure no damage exists that could potentially affect the safety related cable Based on the results of the testing of the cables in B-105, a determination will be made as to what actions, if any, will be taken for the cables in B-10 This is identified as Inspector Followup Item IFI 50-269,270, 287/89-03-04, Damaged Cable Repair/Replacemen The licensee has also committed to initiate a SPR by January 16, 1989, to make the cable tray section drawings "For Information Only."

They will also be annotated that they must be used in conjunction with the cable sheets which are the controlling

documents. The licensee committed to perform an evaluation of the problem and reach a decision by mid-February 1989 as to what actions will be taken. This is identified as Inspector Followup Item 50-269,270,287/89-03-05, Inaccurate As Built Drawing There does not appear to be any significant damage to the cables in the trays above the ITA switchgea Evaluations indicate that there would be no safety significance if any of the cables were damaged and the licensee has initiated actions to identify, any damage and repair/replace as necessar C. Main Feedwater Valves Failure of Auto Switchover to Upper Nozzle Flow Path The Motor Operated valves (1FDW-36,38,45 and 47) required to change positions automatically on loss of all four RCPs did not operate as a result of the fire. The cable from the 1A1 RCP to the ICS was damaged by the fire and the wires were shorted, indicating that the 1A1.RCP was operatin The valves were still able to be operated remote-manuall. Significance of Delay in Shutting Isolation Valves to Lower Ring The operators very quickly observed that the four valves did not shift position as expected. The operators opened the two valves opening a flow path to the upper nozzles (FDW-38,47),

but failed to shut valves FDW-36 and 45 until about 20 minutes late For these 20 minutes, a flowpath was open to both feed rings. It is probable that most of the feed flow entered the OTSG through the lower ring during this time due to the lower flow resistance of that pat This resulted in a loss of the cooling effect caused by feeding directly on the OTSG tubes high in the QTSG. This significantly effected the establishment of the thermal driving head for natural circulation flo (See paragraph VI.)

Due to the significant role the positioning of these valves plays in the establishment of natural circulation, the team concluded that the Emergency Operating Procedures should be revised to include requirements to verify the correct positioning of these valves on a loss of all RCP This is identified as Inspector Followup Item 269,270,287/89-03-06:

EOP Revision Requiring Verification of Upper Nozzle Flowpat. Main Feedwater Block Valves As discussed in the narrative of the January 2 trip, some problems with the main feedwater block valves had been observed previous to this even The malfunctions that had occurred after two Unit 3 trips in November 1988, and also occurred after the Unit 1 trip on January 2, 1989 were caused by a problem in

the control circuitry of the valve. Although this particular problem did not occur during the January 3 event, other difficulties with the valves necessitated closing them from their breaker The problem in the control circuitry to the valves is being corrected by the licensee as the cause has been determine Recently, on all three Oconee un:ts the control room switches for the block valves had been replaced by a Nuclear Station Modificatio The replacement switches occupy less control board space and are of a pushbutton desig Paragraph I.E.,2 explains how the valves should operate if placed in automati On several occasions, after reactor trips it was noted that these valves had shifted to manual and.remained open instead of automatically shuttin The installation of t~he new switches added a new relay to the circuitry (relay 1RXB on diagram OEE-145-24)

which "seals in" the valve in automati It was determined that this is a sensitive' relay and is de-energizing, causing the valves to shift to manual and stay open. The cause of the relay dropping out was traced to the momentary loss of power a.s the power to this circuitry (panel 1XGB) is automatically swapped to the unit startup transformer on unit trip I&E is pursuing replacement of the block valve switches. to eliminate this proble Currently this is planned to be completed during each units next scheduled refueling outag The problems with these block valves which occurred during the January 3 event were not caused by the above malfunction. The switches for the block valves were still in manual since the unit was so low in powe However, when the CR operators did try to shut these valves, they did not properly function. In addition to the operator burden of dealing with these valves during the event, their failure was significant in that if they would have shut properly when first activated, the overfeeding of the A OTSG would have been terminated much earlie In fact the operators indicated that if the valves would have shut properly, the main feedwater pumps would not have been lost and emergency feedwater would not have been initiate (Emergency Feedwater did perform as expected.)

During power escalation earlier on January 3, some problems were observed with the A

Main Feed Block Valve (1FDW-31). It had to be manually operated to obtain a fully opened positio FDW-31 also caused a feedwater swing which delayed power increase on the afternoon of January 3. A priority 1 work request was written,,the stem was lubricated and, since the valve was still difficult to cycle manually, the packing was adjuste The main yoke bearing of the operator was lubricated and the valve cycled electrically satisfactorily. A decision was made to continue with plant startu A work request (WR 051348G)

remained outstanding for more extensive work on the valve operator during the outage

scheduled (at that time)

to begin after about two weeks of operatio Plans were (and still are) to replace the valve stem, reduce the packing load, check the stem and yoke bearings and perform required maintenance. The operator on FDW-31 had been refurbished in September 1987 and was repacked/stroked on September 8, 1987. During the event, FDW-31 was finally shut by an operator manipulating the breaker supplying power to the valv This effectively bypassed the torque switch which enabled the operator to shut the valv FDW-40 ( B Main Feed Block Valve)

apparently failed to shut due to the pushbutton being stuck in the open positio This problem had not previously occurred with this valve and is being investigated by the licensee. Since a similar type of pushbutton switch is also utilized in the Engineered Safeguards Control Console Rz modules, this is-identified as an Inspector Followup Item 269,270,287/89-03-07:

Evaluation of FDW-40 Stuck Control Pushbutton Corrective Actions While the main feedwater block valves are not safety related and are not required for accident mitigation purposes they are important to safety and played a role in this event. During its investigation into the problems with these valves the team concluded that the licensee was working the problems and, in fact, on the control circuitry issue had made significant progress toward resolutio In reaching the decision to continue plant startup with possible problems existing in the FDW-31 operator, the licensee probably considered that operation was scheduled for only two weeks before a scheduled refueling outag Auxiliary Pressurizer Spray Initial Attempt Early in the event (about.8:10pm) the operators tried unsuccessfully to initiate auxiliary pressurizer to reduce RCS pressure. (Manual isolation valve in the RB was later found shut).

The team could not find any approved procedures which directed the operators to do thi Discussions with the operators indicate that intentions were to reduce RCS pressure to establish plant conditions in the TSOR region. The Emergency Operating Procedures direct the operators to establish conditions in the TSOR if the entry criteria are me The EOP does not tell the operators how to get to those condition Discussions with plant personnel indicated that the auxiliary spray had been presented.in training as one available method to reduce RCS pressur Use of auxiliary spray with spray differential temperature above 410 degrees F violates OP/0/A/1103/05: Pressurizer Ope.ration and-TS 3.1.2.6. The team

did note that OP/1/A/1102/01; Controlling Procedure for a Unit Startup, does state if spray differential temperature exceeds 250 degrees F (during a heatup),

spray should be used only if required for plant safet Several sections of the EOP also discuss auxiliary spray for RCS pressure reductio.

Manual Valve Found Shut The reason the initial attempt at using auxiliary spray was unsuccessful was traced to a closed manual valve, 1HP-340, in the RB which isolated the HPI discharge to the spray lin Incorrectly positioned manual valves have not been a problem at Oconee and this appears to be an isolated cas The licensee expended a dedicated effort toward resolving why this valve was shu On October 30, 1987, the auxiliary spray system was verified isolated in preparation for unit startup by Enclosure 4.1 of OP/1/A/1102/01 which requires ensuring valves 1HP-355 and 1HP-472 were shut. Sometime between October 24 and October 31, 1987, valve 1HP-340 was checked and signed for as open on OP/l/A/1104/02:

High Pressure Injection System Valve Checklis On November 1, 1987, a test (weld inspection) was completed which required pressurization of the auxiliary spray lin During the test, 1HP-355 was also successfully functionally tested with a positive indication of flow and pressure change in the control room. Between November 1, 1987 and January 3, 1989, only one activity in the RB which could possibly have effected 1HP-340 was identified. On November 9, 1987, the HPI warming line flow was being adjusted by use of a Removal and Restoration procedur Part of this evolution involved operating valve 1HP-240. While it is possible that a mistake caused valve HP-340 to be shut, it is not probable since the two valves are in very different locations and are not the same type of valv (Team members found 1HP-340 is located on the first grating level on the east side of the RB,. HP-240 is inside the D ring by the RCP and has a long chain operator installed). While the reason for the valve being mispositioned will probably never be determined, the team is convinced the licensee made a genuine effort to resolve the issue. The team considers this mispositioned valve an isolated cas. Consequences of Differential Temperature Limit Violation Late in the recovery, after valve 1HP-340 was opened, the operators utilized auxiliary pressurizer spray to reduce RCS pressure. Spray flow was initiated for less than 5 minutes, reducing RCS pressure from about 1980 psig to 1870 psig. Since LDST temperature was about 100 degrees F and the pressurizer fluid was at about 640 degrees F, this action violated the 410 F differential temperature limit specified in Operating Procedures

and T While no approved procedures specifically direct the, use of auxiliary spray under these conditions, the operators training has presented the spray as an available means of reducing primary pressur In response to the teams' questions on the consequences of the use of spray -(approximately 550 degrees F differential temperature), the licensee obtained a letter from B&W evaluating the spray nozzle transien The letter stated that a previous analysis, performed after ANO-1 initiated spray with a 61 degrees F differential temperature, bounded the Oconee transient of January 3, 1989. At a 610 delta T, the analysis stated that 250 cycles was permissible without consequenc The letter concluded that the transient had a negligible impact on. the fatigue life to. the pressurizer spray -nozzle component Initiating pressurizer spray with differential temperature above limits is part of Unresolved Item 269,270,287/89-03-02:

Failure to Follow Procedures Concerning TSOR and Auxiliary Pressurizer Spray. More discussion of the teams concerns in this subject are in paragraph VII A2 RCP Loss of Seal Return Flow During the event, the delay in restoration of the 1A2 RCP played a significant role. The first (1B2).RCP was restarted at 10:03pm, but did not provide a significant amount of pressurizer spray, since the normal spray line is connected to the discharge of the 1A1 RCP. The 1A2 RCP was not restarted until 3:25am. Auxiliary pressurizer spray was not operable until 2:00am as the shut manual valve had to be opened. This resulted in the inability to significantly reduce primary plant pressure (without PORV actuation) until the 1A2 RCP was restored. This second RCP provided sufficient driving head to permit RCS pressure reduction through.use *of the normal pressurizer spray lin The delay in restoration of the 1A2 RCP was due to a loss of all seal return flow from the RCP after it was secure From approximately 10:00pm to 3:25am, attempts were made to re-establish seal return flow from the pump. 1 HP-277 (Unit 1 RCPs seal return valve) was opened to reduce back pressure on the seals. The seal return stop valves from the 1A1 and 1B1 RCPs (previously secured) were close The LOST pressure was decreased to reduce seal back pressure. I&E personnel verified that the loss of seal flow was not due an instrumentation problem (verified actually lost flow not simply failed indication).

The AC and DC oil lift pumps were cycled in attempts to establish sufficient seal return flo At about 3:10am these attempts were successful and at 3:25am the pump was restarte These corrective actions were observed by the resident inspectors during the event and were conducted i~h a professional and logical manne Procedures OP/1/A/1103/06:

"RCP Operation" and

OP/1/A/1104/02:

"HPI System" were reviewed and discussions were held with the involved personne The licensee's actions to restore the pump did not violate these. procedures. The team found no specific written guidance on methods to recover seal return flow. The loss of seal return flow upon securing the pump is not a rare occurrence with Westinghouse RCP While the delay in pump restoration was signifi cant in the overall sequence of events, the team found that the licensee acted reasonably during attempts to restart the pum It appears that these actions were the best available means to expeditiously restore the pum F. Other Equipment Failures/Malfunctions During review of the event, several other equipment problems which occurred during the event were noted. The licensee is taking action to correct or resolve these problems and the team judged these problems as having less critical roles in the event than the previously discussed malfunction. Problems with the Transient Monitor equipment resulted in the inability to obtain data which could have been very useful in evaluating RCS natural circulation characteristics and other aspects of the even The transient monitor computer hard disc head crashed during the transient. The cause of the failure is unknown and the system was returned to operatio. The team noted during a review of the licensee's Post Trip Review Report and during interviews with the operators that the operation of valve 1HP-120 (Normal Makeup Control Valve)

appeared sluggish and erratic during the even A work request was generated to investigate possible problems with the valve on January 10, 198 Natural Circulation Issues Natural Circulation Events Following the manual tripping of the reactor coolant (RC) pumps at 8:02:22pm and prior to the restart of the 1B2 RC pump at 10:03pm, natural circulation was used to remove core decay hea The AIT investigated the natural circulation performance of the plant and the operators'

actions during this perio (See Section 1.E.2) Of particular interest was:

Core delta T, as measured by the differences between core exit thermocouples and Tc, was75-100 degrees F during portions of the transien This is significantly. larger than the 40-50 degrees F delta T normally expected during natural circulatio After the A OTSG level had increased to about 90% level on the operating range (8:36 to 10:03pm),

the loop A Tc indicated erratic behavior and decreased to 400 degrees This raised

concern as to whether natural circulation was occurring in the A-loo. During the January 3 event, the reactor was manually tripped at 7:59pm from a power level of approximately 4%. At 8:02:22, the operators tripped the remaining two RCP Due todamage caused by the fire (as discussed in Section IV. ),

automatic realignment of the MFW system to the auxiliary feedwater inlet header did not occur. At 8:05, the operators manually shifted MFW to the EFW header. (auxiliary nozzles).

However, the two startup block valves in the MFW systems were not closed and the feedwater flow path to the OTSG via the MFW header remained open. At about 8:20 p.m.,

the operators di.scovered the open valves in the MFW system and closed them thus directing all feedwater flow via the EFW heade The effect of these actions on the plant response during this, period can be summarized as follow From 8:05 to 8:11 p.m.,

levels in the OTSG rapidly increased to 50% on the operating range and were maintained at roughly this level until 8:31 As a result of this rapid filling, OTSG pressures decreased from approximately 1000 psig to 500 psig. Hot leg temperatures were maintained between 540 degrees F to 555 degrees F; cold leg temperatures decreased from 545 degrees F to 460 degrees Indicated core delta T ranged from 75 degrees F to 100 degrees The team investigated the system responses and operator actions over this time period (8:05 to 8:31pm). As a result of the open valves in the MFW system, the team concluded that the majority of the feedwater delivered to the OTSG during the filling process was via the MFW header and thus resulted in a decreased driving head for natural circulation relative to that normally expected. It is the teams conclusion that the large indicated core delta T was the direct result of this decreased driving hea IFI 269,270,287/89-03-06:

EOP Revision Requiring Verification of Upper Nozzle Flowpath (See paragraph IX.C.2.)

will be used to follow EOP revision to help prevent thi The team also notes that the rapid filling, and the resultant depressurization, of the OTSGs over this time period was apparently performed to limit peak RCS pressur This was directly responsible for the overcooling of the RCS which occurre Due to the low decay heats present during this transient, there was no apparent need for this rapid filling and it appears that a less severe cooldown would have occurred if the operators had limited the rate of feedwater additio The natural circulation performance of the plant over this period was examined in detail by the team. Due to the low decay heats present, it was difficult to determine with certainty

whether natural circulation was fully established. Decay heat was calculated to range from 20 to 14 MWt during the natural circulation perio Based upon the close coupling of the indicated cold leg temperatures to the saturation temperatures of the OTSG and the relatively stable hot leg temperatures, it appears that natural circulation was occurrin However, the team also noted that when the operators opened the turbine bypass valves on the OTSG to check whether natural circulation was established, hot leg temperatures and OTSG pressures increased. This response is apparently the result of hotter fluid in the reactor vessel being circulated through the syste Thus, thermal gradients apparently existed which indicates that natural circulation may not have been fully established.< The team believes, however, that even if natural circulation cooling was not fully established at this time, adequate core cooling was being maintained. Additionally, the team believes it is likely that natural circulation cooling would have been fully established without further action required by the operato Of further note is the operator actions concerning verification of natural circulation over this perio The operators were concerned about the large indicated core delta T and whether natural circulation was established. The operators had expected to see core delta T on the order of 50 degrees To check natural circulation, the operators opened the turbine bypass valves while monitoring the core exit thermocouples for response. Because the incore temperatures decreased in response to this action, the operators concluded that natural circulation was occurrin The team believes that this action was reasonable and prudent given the operators expectations and trainin. At 8:31pm, the operators placed the MFW control valves in manual and, as discussed in section II.B, the "A" MFW control valve backed off its shut sea This resulted in a rapid filling of the A-OTSG with 140 degrees F feedwate The A-OTSG level increased from 50% to 94% on the operating level (30 feet from the lower tube sheet) from 8:31 to.8:36 p.m. At 8:36 p.m., the MFW pump was automatically tripped and the auxiliary.feedwater pumps were started. Due to the high levels in the A-OTSG, no EFW was. delivered, while EFW was delivered to the B-OTSG to maintain its level at 50% on the operating rang Due to the high levels in the A-OTSG, the operators opened the Turbine Bypass Valves to steam the A generator and decrease leve However, little level decrease actually occurre This situation was maintained-until the 1B2 RC pump was restarted at 10:03p System response from 8:36 to 10:03 was difficult to determine due to the problems with the transient monitoring system discussed in section I Incore T/C temperature

3 gradually decreased from 570 degrees F to 510 degrees F over the period.. A-0TSG and B-0TSG levels remained at 95% and 50% on the operating range, respectivel The B-OTSG 'pressure. was maintained between 440 and 540 psig using the turbine bypass valves and EFW; B-loop cold leg temperatures ranged from 455 to 480 degrees F and tracked the saturation temperatures of the B-OTSG. The A-OTSG ranged from 440 to 640 psi The A-loop cold leg temperatures showed a gradual decline from 470 degrees F to 400 degrees F. However, two rapid temperature rises of roughly 50 degrees F to 480-490 degrees F occurred at approximately 9:00 p.m. and 9:14 p.m. These rapid temperature rises were immediately followed by a gradual reduction of the cold leg temperature B. Evaluation of Natural Circulation Effectiveness The team investigated the system response during this perio With respect to the B-loop behavior, it is clear the natural circulation was continually maintained. In addition, with the shift to feeding of the B-OTSG via the auxiliary feedwater header, core differential temperatures decreased to 50-60 degrees F as opposed to the 75-100 degrees F observed while feeding via the MFW heade The behavior in.the A-loop is much more difficult to determine. The gradual cooling indicated by the A-loop cold leg RTDs is believed to have been caused by seal injection flowing into the cold leg pump suction piping.. Charging flow was increased by the operators to maintain pressurizer level and injects into both A loop cold leg Further, since the A-loop cold leg temperatures do not track the A-0TSG saturation temperatures and the hot leg RTD temperatures did not track the core exit thermocouple indications, the team concluded that natural circulation generally did not occur in the A-loo The team questioned why natural circulation didn't occur in the A-loop, especially given the high level in the A-OTSG? The team concluded that the rapid addition of MFW resulted in a subcooled pool of water in the bottom of the A-0TS The natural circulation flow initially present during the filling period-transported cold fluid into the suction pipe of the reactor coolant pump which retarded the cold driving head and terminated the natural circulation flo The temperature spikes at 9:00 and 9:14pm are indicative of a temporary reestablishment of natural circulatio However, this also transported additional cold fluid into the suction piping which then terminated natural circulation. It was unclear to the team whether natural circulation could ultimately be recovered in the A-loop given the low decay heat present. It was also unclear to the team whether the rapid introduction of subcooled water into the OTSG could have had a negative impact on the OTSG structural integrity. This should be further examined by the license This is IFI (269, 270, 287/89-03-08)., Summary

The team concluded that the low decay heat present during this event was a significant contributor to the overall system response. Plant decay heats were calculated to range between 20 and 14 MWT during the natural circulation period. At these levels, heating of the reacto coolant to establish the delta T required for natural circulation would be slow (on the order of 1-2 degrees F per minute).

Further, because of the low decay heats, it is likely that subcooling of the A-OTSG occurre Further, stored energy and metal heats could represent significant fractions of the overall heat required to be removed by the OTS Also, heat removal via the letdown system, which continued to operate during the event, represented a signi ficant fraction ( 4 MWt) of the overall heat removal from the RC All of these factors made detailed understanding of the plant response by the team difficul In addition, these factors led to a plant response which was not expected by the operators and was a source of confusio However, overall operator response was prudent and assured that adequate core cooling was maintaine In summary, the team concluded that adequate natural circulation was maintained throughout the event. Due to equipment malfunctions and the low decay heats present during the event, the operators were presented with plant parameters which were not consistent with their experience and training which led to preprogrammed responses such as the initial rapid filling of the OTSG to the 50% operating leve However, the operator actions were reasonable and prudent and assured adequate core coolin V Operator Actions and Management Involvement During the Event and Recovery During the event the operators were following the -Oconee Emergency Operating Procedures. These are derived from B&W owners group Abnormal Transient Operating Guidelines (ATOG). ATOG is designed to treat symptoms of abnormal occurrences rather than require operators to diagnose the particular event. During the event, operators utilized only the immediate action and subsequent post trip sections of the Oconee Emergency Operating Procedures EP/1/A/1800/ Operators did not believe that reactor conditions or symptoms met the entry criteria for other sections of the Emergency Operating Procedures. However, their actions were influenced by training in other Emergency Procedure section The AIT reviewed operator actions during the event and compared them to the EOPs. This was done to identify possible weaknesses which should be considered for future procedural upgrades. One specific weakness is the lack of procedures for planned entry into natural circulatio IFI 269,270,287/89-03-09 Operators manually fed up the OTSGs to the 50%.

level on the operating range after the reactor coolant pumps were tripped

. when the automatic system failed due to fire damage. (see Section IV.C.1)

Had the automatic ICS fill system functioned and opened the startup valves to 40 % open, overcooling might have been more severe as the operators had the startup valves. opened to only 30 % ope The valves are designed to open to 40% demand,

"lock-in" (ICS psuedo mode)

at this position, and

rapidly increase OTSG leve The licensee had previously initiated a SP to remove the "lock-in" feature on these valves on a loss of RCPs to permit operator control from'the control board if require The rate of manual OTSG feeding caused rapid reductions in reactor system temperature and affected pressurizer level, mainly due to the low decay heat. The loss in pressurizer level caused the operator controlling charging flow to rapidily increase makeup, allowing too much flow to the "A" loop cold legs which may have adversely affected natural circulation flow. Additionally, the licensee should consider providing procedures for a smoother planned transition into natural circulation giving consideration for establishing the proper steam generator levels before the reactor coolant pumps are tripped. Guidance on control of charging flow during transition to natural circulation should be provided to minimize unnecessary injection of cold water into the reactor system, which could inhibit natural circulatio During natural circulation operation, the pressurizer PORV and the auxiliary pressurizer spray are both means available to the operator to reduce reactor system pressure. EOPs for loss of heat transfer (Section 502) and steam generator tube leak (Section 504) specify' use of the POR EOPs for the subcooled cooldown (CP-605)

specify that auxiliary spray be used. In the event that RCS pressure must be reduced because of excessive cooldown, a three hour soak in the thermal shock operation 'region (TSOR)

is require The means of pressure reduction to the TSOR is not specified. The procedures should provide consistent guidance for pressure reduction while in natural circulatio These concerns are included in URI 269,270,287/89-03-02:

Failure to Follow Procedure Concerning TSOR and Auxiliary Pressurizer Spray (see Section VII.B.2).

The procedures provide that following a cooldown rate in excess of 100 degrees F within a one hour period that the reactor system pressure be reduced to the thermal shock operating region (TSOR)

for a three hour soak. Which temperature measurement to be used is not specified and this caused some confusion during the event among the operating staf Only the cold leg temperatures in one coolant loop exceeded this criterion and temperature indications elsewhere in the reactor system did not decrease as rapidl There was also disagreement as to what starting temperature to use. A second entry condition into the TSOR is if RCS temperature decreases below 500 degrees F and all coolant pumps are off and HPI has operated in the ECCS mode. This criterion may have also been met during the event since the cold leg RTDs were cooled below 500 degrees F and charging flowrate equivalent to safety injection flowrate was introduced into the "A" cold leg These instructions were also a source of confusion to the operating staff since the temperature sensor to be utilized was not specified and instructions were not provided relating the equivalency of high charging flow to safety injectio The instructions for entry into the TSOR were subsequently clarified after the licensee discussed these aspects with B&W. The licensee issued Training Package 89-02 which specifically states the entry criteria for TSO During the event the plant staff did not reduce reactor system pressure into the TSOR even though both the entry criteria had been me The actual cooldown was about 120 degrees F in an hour and the minimum cold leg temperature was 400 degrees F based on loop "A" cold leg RTD Station management manning the Technical Support Center decided against reducing reactor system pressure but instead performed the 3 hour3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> soak at the pressure at which the plant had stabilized. Pressure was not reduced since the auxiliary pressurizer. spray was inoperable and the plant staff was concerned that operating the pressurizer PORV would unnecessarily violate the reactor system pressure boundar An additional concern was expressed that the pressure reduction could cause formation of a reactor vessel head bubble which would further complicate recover One reason that the plant staff was reluctant to depressurize the reactor coolant system to the TSOR region using the PORV is that the* PORV control is within a cabinet at the rear of the control roo An operator manipulating the PORV manual control switches could not directly observe the effect on reactor system pressure or pressurizer level. The licensee had previously committed to relocate the PORV control to the control console during this refueling outag The licensee believed that the requirements for entering.the TSOR were too conservative and should be unnecessary in the case of the Oconee-even The licensee did not know the basis for selecting.the pressure range of the TSOR or the basis for selecting the entry criteri The AIT requested' that this basis be provide The basis for TSOR should be evaluated to ensure that unnecessary reactor system depressurizations are not required and that th reactor vessel is adequately protected from thermal shoc Operators should be aware of the basis of procedural step URI 269,270,287/89-03-02:

Failure to Follow Procedure Concerning TSOR and Auxiliary Pressurizer Spray will be followed to track these issues (Section VII.B.2).

The reactor.system experienced an excessive cooldown in the even Conditions for entry into the TSOR region were satisfied based on cold leg RTD indications both from the 100 degrees F in an hour criterion and the below 500 degrees F criterio The three hour vessel soak that is required at the TSOR pressure was performed at a higher pressur The licensee requested B&W evaluate the consequences of over cooling on vessel crack growth and thermal fatigue before further operation. (See paragraph VII.B.1)

VII. Findings Reportability Review Unit 1 tripped on January 2, 1989, at 3:23 This was reported using the NRC red phone at 4:30 as required by 10 CFR 50.72 Section (a)(2)(ii)

(Four-hour reporting requirement).

This occurrence was also provided to the resident inspector at approximately 5:30 p.m. on that da The fire that occurred in the 1TA switchgear occurred at 7:17 p.m. on January 3, 1989. The Unusual Event was declared at 7:45 in accordance with the licensee's Emergency Pla The NRC was notified at 8:14 as required by 10 CFR 50.72 (a)(1)(ii)

(one hour reporting requirement).

The resident inspector was notified at home at approximately 8:35 In both of these events the NRC was notified as required by regulations. However, the time lag between the initiation of the incidents and the time the resident inspector was notified is of concern to the tea This is important since it allows an NRC representative familiar with the specific operation of the unit to evaluate the licensees actions, the consequences of the actions.and provide input to NRC management concerning the' effect of the occurrence on public health and safet B. Open Items 1. Oconee TS 3.1.2.1 requires that the reactor coolant system pressure and system heatup and cooldown rates shall be limited in accordance with Table 3.1-1, Figures 3.1.2-1A, and 3.1.2-2A (Unit 1).

Table 3.1-2 states that whenever reactor coolant temperature (cold leg temperature if on natural circulation cooldown or any RCPs are operating)

is greater than 280 degrees F, the maximum cooldown rate (as measured by cold leg temperature)

should not exceed 50 degrees F in any half hour period. During the event on January 3, 1989, the RCS cooldown rate exceeded this limi This item is identified as an Unresolved Item 269,270,287/89-03-01:

RCS Cooldown Rate in Excess of TS Limits, pending review of this issue by regional staf In response to the teams questions on the consequences of this overcooling transient the licensee requested an analysis from B& By letter dated January 13, 1989, B&W stated that preliminary results of the overcooling analysis (utilizing the most limiting weld properties and Regulatory Guide 1.99, Rev.2)

indicate that the integrity of the reactor vessel was not challenge. During the inspection, the team identified several examples where approved operating procedures were not followed. The team concluded that the significant examples of inadequate procedure were:

a. Step 5.17 (of the subsequent actions after a reactor trip portion of the Oconee Emergency Operating Procedure (EP/1/A/1800/01)),

requires the operator to refer to the appropriate enclosure in Enclosure 7.1A: "Normal

Containment Cooldown - Limits" requires the operators to stabilize RCS conditions in the TSOR and perform a three hour soak at those condition During this event discussions in the TSC resulted in a management decision not to enter the TSOR of the enclosure. A three hour soak was performed with the RCS conditions above the TSOR region. (Details of this are discussed in section VII of this report.)

OP/0/A/1103/05:

Pressurizer Operation requires that pressurizer spray not be initiated if spray-pressurizer differential temperature exceeds 410 degrees At about 2:30 the operators initiated auxiliary spray to reduce RCS pressure from about 1980 psig to about 1870 psig. This action violated both the operating procedure and TS 3.1.2.6 which prohibits use of pressurizer spray

-

with a differential temperature of 410 degrees F. or greate The above two examples of inadequate procedure are identified as an Unresolved Item 269,270,287/89-03-02:

Failure to Provide Adequate Procedures Concerning TSOR and Auxiliary Pressurizer Spray, pending review of this issue by regional staf The team has several concerns relating to these procedural problems;

-

The questions that causedmanagement to believe that entry into TSOR may not have been required, (what temperature indications should be used, what setpoint is to be used to calculate the change in temperature)

have been answered by the issuance of Training Package.89-02 (see section VII).

There is still a question on whether PORV operation should have been used to go into the TSOR. During discussions with plant personnel. it was emphasized to the team that the decision to deviate from the approved procedures was made by management personnel in the TSC, not the operators. The existing procedures and training guidance would dictate the operators actions if similar conditions were to develop (and the TSC is not manned).

It is not clear what action would be taken by management if similar conditions evolved on Unit 2 tomorro (The relocating of the PORV controls to the unit board from a cabinet in the back of the CR has not yet been completed on Unit 2.)

-

The issue of whether it is so important to get into the TSOR region that the TS limit on pressurizer spray should be intentionally violated, should also be resolve Some discussions with the team indicate that the significance of not going to the TSOR envelope on this event was very low and that the entry criteria (requiring the TSOR soak) are overly conservativ If it is decided that criteria is too conservative, the team believes the criteria should be revised

to be more realisti This would significantly reduce operator burden and may prevent unnecessary depressurization action The licensee has realized for some time-that, in a scenario like that of January 3, -auxiliary pressurizer spray would be one means available to reduce RCS pressure. TS 3.1.2.6 essentially eliminates this optio The team found documentation that indicates the licensee had raised the issue (for some resolution or guidance) at B&W Owners Group meetings back in October of 1988. A letter received by the licensee from B&W indicates that the thermal cycle caused on the spray system by-the transient on January 3 was of low significance. As the spray can withstand a certain number of cycles at a specific.differential temperature, the team. feels. that the appropriate TS should be re.vised to permit thi. IFI 269,270,287/89-03-03:

Discrepancies Between one-line Drawings and Elementary Wiring Drawings. (Section IV.B.4)

4..

IFI 269,270,287/89-03-04: Damaged Cable Repair/Replacement. (Section IV.B.5)

5. IFI 269,270,287/89-03-05:

Inaccurate As-built Drawing (Section IV.B.5)

6. IFI 269,270,287/89-03-06:

EOP Revision Requiring Verification of Upper Nozzle Flowpath. (Section VI.) IFI 269,270,287/89-03-07:

Evaluation of FDW-40 Stuck Control Pushbutto.

IFI 269,270,287/89-03-08:

Evaluation of OTSG rapid cooling (Section V.B) IFI 269,270,287/89-03-09:

Procedure change for entry into natural circulation VII Conclusions The firefighting techniques were adequate and did not further endanger the uni Due to the extensive amount of damage to the ITA Normal Supply Breaker and its cubicle, no root cause was able to be determine A lack of performing maintenance, incorrect maintenance, or the use of improper parts may be ruled out as contributing factor The licensee is continuing to evaluate the situation in order to come to the most likely cause of the even C. Based on the results of the testing of the cables in B-105, a

determination will be made as to what actions, if any, will be taken for the cables in B-10 These actions willibe followed by the Regio D. The main feedwater block valves are not safety related and are not required for accident mitigation purposes, but they are important to safety and played a role in this even The startup decision following the January 2 reactor trip was made without completely understanding their behavio The problems with these block valves which occurred during the January 3 event were not caused by the earlier malfunctio The operators indicated that if the valves had shut properly, the main feedwater pumps would not have been lost and AFW initiate Therefore, the two reactor trips had different' causes and the equipment malfunctions from the first trip did not contribute to the second trip or its recover The team concluded that management was extensively involved in both the equipment repairs after the January 2 RX trip and the event recovery on January 3/ During this event, discussions in the TSC resulted in a management decision not to enter the TSOR regio A three hour soak was performed with the RCS conditions above the TSOR region. (Details of this are discussed in section VII of this report.)

The operators initiated auxiliary spray to reduce RCS pressure. This action violated both the operating procedure and TS 3.1.2.6 which prohibits use of pressurizer spray with a differential temperature of 410 degrees F or greate The Team's concerns expressed in Section VII.B must be addresse Once further clarification of these issues is made, procedural changes should incorporate this information to provide better guidance to the operato IX. Exit The inspection scope and findings were summarized on January 13, 1989, with those persons indicated in section I.D. After subsequent discussions of regulatory issues by regional staff, the inspector summarized the open items with station compliance personnel on January 18, 198 No dissenting comments were received from the license The areas of concern will be addressed as action on the Unresolved Items is determined after further regional staff evaluatio The licensee did not identify as proprietary any of the material provided to or reviewed by the inspectors during this inspectio X. Acronyms and Abbreviations This list is intended only for the terms most generally use All terms which are used only in one section are.defined therei TA and TB 6900V switchgear CT-1 startup transformer 1T auxiliary transformer from the main generato MFB Main Feeder Buses TC, TD, TE 4160V switchgear HPI high pressure injection HP-20 seal leak off return valve HP-26 emergency injection to-the 1A loop HP-24 suction valve for A HPI pump from the BWST HP-25 suction valve for BHPI pump from the BWST HP-120 pressurizer level control valve HP-31 RC pump seal injection valve HP-340 manual isolation valve from charging line to au spray AIT ATOG Augmented Inspection Team ATOG Abnormal Transient Operating Guidelines B&W Babcock and Wilcox BWST borated water storage tank CAL Confirmation of Action Letter CR Control Room Delta T Differential Temperature ECCS Emergency Core Cooling System EFW Emergency Feedwater EOP Emergency Operating Procedure FBC Fire Brigade Captain FDW Feedwater HPI High Pressure Injection I&E Instrument andElectrical ICS Integrated Control System LDST Letdown Storage Tank MDEFW Motor Driven Emergency Feedwater (Pump)

MFB Main Feeder Bus MFW Main Feedwater System MWt Megawatts thermal NCO Nuclear Control Operator NSM Nuclear Station Modifications ONS Oconee Nuclear Station OR Operating Range (OTSG.level indication)

OSC Operational Support Center OTSG Once Through Steam Generator PORV Power Operated Relief Valve PTS Pressurized Thermal Shock RB Reactor Building RCP Reactor Coolant Pumps RCS Reactor Coolant System RPS Reactor Protection System

RTD Resistance Temperature Detector RX Reactor SPR Station Problem Report SO Safety Officer S Shift Supervisor Tc Temperature of the RCS Cold leg TBV Turbine Bypass Valves T/C Thermocouple TDEFW Turbine Driven Emergency Feedwater (Pump)

TSC Technical Support Center TSOR Thermal Shock Operating Range WR Work Request XSUR Extended Startup Range (OTSG level indication)

230KV A/W L 230 KV 19KV IT CT-I lT 4.66KV A 4.16KV GEN 6.9KV 6.9KV The fire started in this bk.9 K V 6.9KV ITA ITB 0T1 RCP RCP RCP RCP IAI IBI IA2 1B2 OC-EL-MPD-21 8-16-85 Main Power 6900V Distribution System 0-702 Distribution DMC/ARD I-TRAINING USE ONLY

MAIN STEP-UP TRANS 230 KV UNDERGROUND 100 KV LINE FEEDER (LEE STEAM STATION)

IT CT1 FROM KEOWEE J

HYDRO STATION

/

CT-4 CT-5

"EEGEC STATU '"BUS SKI FL STANDBY. BUS ISLI SL2 ( Ni ( N2 El (E2 (ISI (,S2 STANDBY MFB BUS I MFB ITC-1 ITC-2 Tm

(

( ITD-I ( ITD-2ITE-1 (ITE-2(

I0C ITD ITE HPSW HPSW (B)

(A)

OC-EL-MPD-3 8-19" M ain Pow er

-70?.

4160 V Distribution System 0-702 DistributionDM

/A B,

TRAINING USEF ONLY

EMER FOW --3-100 3S

33GENERATOR

32 37I 7 4 TO UPPER

SURGE TANeK EE rOWW 3.16 318 EMER 1DW

47

76 AGIRB TYPICAL FOR UNITS 1.2,3 ALL VALVES

'FOW'

EXCEPT AS NOTED ATTACHMENT TWO

Unit 1 Page 1 of 3 EMERGENCY OPERATING PROCEDURE EP/1/A/1800/01 ENCLOSURE 7. 1A Normal Containment Cooldown Limits 100 200 300 4?

500 600 700 2200

2000 (- SO

,

1800

1600

-

ui

00 C

2 000i,

',E

,

12002000 o0 OO

.

OU IIT!

f;'

1800 400Ui

-=

I'SCIO

0 100 200)

300

50 600 17 0 EATORCONTAINERTR (F ATCMENTTHE SO when:

,

7u

.

h.a s en

..,!.

-2

occaure.

E I n Cs r

-

tabilizeaca

0 conditions and even. any777V 14a. lopr-100 pranon r us o

o r-r 6-cro-4 z

200*

FSLC O

T

.A*.n con5inou

.acATTACHMENT THE

ATTACHMENT 4 UNITED STATES NUCLEAR REGULATORY COMMISSION 101 MARIETTA ST., ATLANTA, GEORGIA 30323 Docket No. 50-269 JAN 0 4gg License No. DPR-38 Duke Power Company ATTN: Mr. H. B. Tucker, Vice President Nuclear Production Department 422 South Church Street Charlotte, NC 28242 Gentlemen:

SUBJECT: OCONEE NUCLEAR STATION UNIT 1, CONFIRMATORY ACTION LETTER This letter is to confirm our understanding of commitments made during the telephone discussions between us on January 4, 1989. The telephone discussions related to the fire on January 3, 1989, in the "iTA" switchgear, subsequent equipment malfunctions, a mispositioned valve in the pressurizer auxiliary spray system, and several additional equipment problems of concern which occurred following the reactor trip on January 2, 198 It is our understanding that you will maintain all equipment related to the

"iTA" switchgear fire in such a manner that it can easily be kept or placed in the "as found" condition. Therefore, you should minimize any actions, other than those necessary to protect the health and safety of the public, which would destroy or cause to be lost any evidence which would be needed to investigate or reconstruct the even You will advise the AIT team leader, Thomas Peebles, of this office prior to conducting any troubleshooting activitie Such notification should be soon enough to allow time for the team leader to assign an inspector to observe the activitie You should make available to the AIT all relevant written material related to the installation, testing, maintenance, and/or modifications to the ITA switchgea Please let us know immediately if your understanding differs from that set out abov

Sincerely, Malcom L. Ernst Acting Regional Administrator CAL 50-269-80-01 cc: M. S. Tuckman, Station Manager State of South Carolina ATTACHMENT 4

ATTACHMENT 5

36 NEAR SIDE AR SIDE

tIR A

SI'

~38

us 36 FAR SIDE NEAR SIDE

35

33 NEAR SIDE NEAR SIDE 23

.

IA l

A SDEFAR SIDE

1 EAR SIDE FAR SDE

AURILIARY FEEDWATER (L I. 8135-12'I I E

MEAD ER 11LET ELE /-EL

.

823'-6 1/4 ILEY V

6 -

11/1 P UMP 1 82)0 IM FEDWATER NOZZLE 2 RED

NEAR A FAR SIDE ELEI. 765-15" filV 777'6 15 1.

Reactor CooLant System TypicaL ELevation ATTACHMENT 5

ATTACHMENT 6 100%

648" 631" 625" Upper Tobe Sheet 620" 600" 600 S1OTSO LEVEL A

INDICATION 500 500 70%

60%

100%

400 400394 394 388 388" 90%

50%

w

.60%

70%

o

0%

30 %

-40%

50%

250 40%2 0%.2000 200 200 30%

200 20%

20%

10%

0%" -

0%

108 100 z

1000 100" 10%00

0%

J A D Lewer Tube Sheet u nUn OC-CM-S l12 2-13 -85 ONCE-THROUGH STEM B SW TECH MANUAL GENERATOR OTSG LEVEL RANGES v

ATTACHMENT 6