ML16139B086

Event Review Group Rept on Performance of NRC During SG Dryout Event at Oconee Unit 3 on 940810

ML16139B086
Person / Time
Site:	Oconee
Issue date:	12/11/1995
From:	Borchardt R, Zimmerman R INSTITUTE OF NUCLEAR POWER OPERATIONS, NRC (Affiliation Not Assigned)
To:
Shared Package
ML16139B085	List: ML16139B084 ML16139B086
References
NUDOCS 9512140184
Download: ML16139B086 (30)

Text

OCONEE EVENT REVIEW GROUP REPORT ON THE PERFORMANCE OF THE NRC DURING THE STEAM GENERATOR DRYOUT EVENT AT OCONEE UNIT 3 ON AUGUST 10, 1995

-OR 0

DECEMBER 1995 U.S. NUCLEAR REGULATORY COMMISSION 9512140184 951213 PDR ADOCK 05000287 K

P POR

SUBJECT:

OCONEE EVENT REVIEW GROUP REPORT EVALUATION PERIOD:

September 14, 1995, throu gh November 21, 1995 TEAM MEMBERS

~tE Di or for p--tobr tI~a Nil 4WitT otf/~-

R644er Rgulaion Eftery Opertiont. Inc.

ell N ler Regulatory Comiso arvinf IWOM A 'Lntstratio Smoor AS Or nine Ifttj o t W arPlat pertins sfsO Dvision a AtMI Safe Analiis Iltaitu, o Nulea Plnt pert~eMice ; of Muc Car %eactor equlat ion U.S I~clt'Regulatory Co.Qmjsjan Chitff itkten Proram Dreh D~i.

$6o Inspecti on k support Progras Offie ofNuclear Reattor Reglation U.S. Nulear %Yqlatory e.mission

INTRODUCTION Oconee is a three-unit station operated by the Duke Power Company. Each unit includes a pressurized-water reactor designed by Babcock and Wilcox and is rated for 885 Mwe. Unit 1 was placed in commercial operation in 1972; Units 2 and 3 were placed in commercial operation in 1973.

Each reactor has two loops; with two reactor coolant pumps and a once-through steam generator (OTSG) per loop. Each OTSG supplies a separate steam header containing a turbine stop valve and two turbine bypass valves (TBVs). In the automatic mode, each set of TBVs is controlled by a header pressure error signal derived from a single pressure setpoint and the individual header pressure. In the manual mode of operation, the TBVs are operated by a demand signal entered at the bypass valve control station in the control room.

On August 10, 1994, at 4:25 a.m., Oconee Unit 3 experienced a reactor trip from 100% power when power to the Integrated Control System (ICS) was temporarily lost. When the ICS power was restored, the TBVs randomly repositioned to 11% open for steam generator (S/G) 3A and 22% open for S/G 3B, resulting in diverging steam generator pressures. The operators responded to these symptoms as a steam leak, entered the Emergency Operating Procedure (EOP) section on Excessive Heat Transfer and isolated S/G 3B. A Notice of Unusual Event (NOUE) was declared at 4:57 a.m. (an NOUE is the least significant of four levels of emergency event classification).

The plant was stabilized with S/G 3B isolated and considered dry with only a steam atmosphere. Decay heat was removed through S/G 3A using the TBVs and emergency feedwater. The licensee elected to delay refill of S/G 3B until after the 6:30 a.m. shift change and until their following concerns had been addressed: 1) feeding a dry S/G with "cold" emergency feedwater, 2) TBV operability, and 3) steam line integrity. During this time there were conversations between the senior resident inspector (SRI) and NRC Region II and between the SRI and licensee management concerning a proposed NRC-licensee conference call to discuss the event and Oconee's recovery plans. By mutual agreement, the call was scheduled for 10:00 am. The conference call was held as scheduled. Following the call the licensee restored the level in S/G 3B and at 1:37 p.m., the NOUE was exited. A detailed chronology is provided later in this report.

During the time between the shift change and conference call, a licensee staff member expressed his understanding to an NRC resident inspector that the NRC was interfering with plant operations by not allowing the licensee to commence refilling S/G 3B until a conference call with the NRC could be held.

The resident inspector immediately informed the SRI, who in turn took immediate steps to affirm with licensee management that the NRC did not have a hold on refilling the steam generator and that the licensee has the responsibility and authority to take all actions they deem appropriate to safely operate the plant.

During an Institute of Nuclear Power Operations (INPO) routine evaluation at.

Oconee in January 1995, the concern of possible NRC inappropriate action during the August 1994 event again surfaced, prompting a further review by INPO.

Subsequently, on March 8, 1995, Joe Colvin, Executive Vice President, Nuclear Energy Institute and Zack Pate, President, INPO, placed a telephone call to James Taylor, NRC Executive Director for Operations.. In that call

0 0

2 Mr. Pate expressed concerns about the NRC's actions on August 10, 1994. NRC documented its investigation of INPO's concern in a report dated July 10, 1995 (attachment 1).

Following issuance of the NRC report, Mr. Pate informed Mr. Taylor that INPO had a continuing concern about the NRC's actions during the licensee's event recovery efforts on August 10,1994.

In response to INPO's continuing concern, Mr. Taylor, directed that an event review team be formed. The five member review team consisted of senior NRC and industry representatives. The purpose of the review team was to gain a clear understanding of the INPO concerns, assess the NRC's actions, and develop recommendations, if appropriate. A copy of the review team's charter is provided as attachment 2.

Previous reviews of technical aspects of the event have been conducted, and are documented in Licensee Event Report 287/94-02, NRC Inspection Reports 50 287/94-24 and 50-287/94-28, and INPO Significant Event Report 10-95. Although the review team made no attempt to duplicate or verify the technical accuracy of those prior reviews, a member of the team did assess (1) whether there was a safety impact associated with the delay in feeding the dry S/G, and (2) whether any technical concerns associated with the S/G recovery warranted additional followup (attachment 3).

In addressing the specific concerns raised by INPO, the review team primarily focused on whether the NRC, by the role it played, impacted the licensee's response to this event.

CONDUCT OF THE REVIEW During the weeks of October 16 and October 23, 1995, the event review team performed an in-depth review of the NRC's role during the August 10, 1994, reactor trip and S/G dryout event.

The team reviewed pertinent documents, including NRC and licensee correspondence and INPO documents associated with the event. NRC headquarters personnel, NRC Region II staff, the NRC resident inspectors assigned to Oconee during August 1994, INPO management, INPO personnel involved with the Oconee evaluation, and Oconee station management and staff, including available personnel from the operating shifts on duty the morning of the August 10, 1994, event, were interviewed.

This review was conducted more than a year after the event, and most of the people interviewed-thought the NRC-licensee interactions were unremarkable and had trouble remembering the exact details and timing of events. However, aspects of the NRC-licensee conference call were considered noteworthy by nearly all participants.-

3 The following individuals were interviewed:

Institute of Nuclear Power Operations (INPO)

Zack T. Pate, President, Terence J. Sullivan, Executive Vice President and COO Claude C. Cross, Vice President Ronald'P. Thurow, Evaluator Calvin E. Goslow, Event Analysis Nuclear Requlatory Commission Robert Carroll, NRC Region II Project Engineer Ellis Merschoff, NRC Region II, Director, Division of Reactor Projects (DRP)

Pierce Skinner, Acting DRP Branch Chief Paul Harmon, Oconee Senior Resident Inspector Lee Keller, Oconee Unit 3 Resident Inspector Keith Poetner, Oconee Unit 1 Resident Inspector*

Gus Lainas, Assistant Director for Region II Reactors, Office of Nuclear Reactor Regulation (NRR)*

,Herb Berkow, Project Directorate 11-2, NRR Dick Wessman, Chief, Mechanical Engineering Branch (MEB), NRR Ed Goodwin, Chief, Section A, Events Assessment & Generic Communications Branch (EAGC), NRR Len Weins, Oconee Project Manager, NRR Jay Rajon, Senior Mechanical Engineer, MEB, NRR Nick Fields, Reactor Systems Engineer, EAGC, NRR

• position on August 10, 1994 Oconee Nuclear Station Jack Peele, Station Manager Gary Rothenberger, Operations Superintendent Wayne Morgan, Senior Technical Specialist Henry Lowery, Senior Technical Specialist Fred Owens., C-Shift Operations Shift Manager Mike Hill, C-Shift Unit Supervisor (SRO)

Sam Lark, C-Shift Reactor Operator Rich Gerner, C-Shift Reactor Operator Joe McCollum, E-Shift Operations Shift Manager Larry Evans, E-Shift Unit Supervisor (SRO)

Barry Honeycutt, E-Shift Control Rouim SRO LJ Davis, E-Shift Reactor Operator Larry Wilkerson, E-Shift Reactor Operator Marshall Loudermile, E-Shift Reactor Operator Note: C-Shift on duty at the time of the dryout event; E-Shift relieved about 7:00 a.m. on August 10.

4 The team met Monday, October 16, 1995, at the INPO office in Atlanta, Georgia and interviewed key INPO personnel. The interviews included INPO senior management and personnel that participated in the INPO evaluation of the Oconee station in January 1995. In addition to gathering facts, the review team clarified concerns expressed by Mr. Pate. The concerns were summarized in the following two objective/problem statements.

A. Review the Oconee S/G dryout event and determine if the action/level of involvement by the NRC resident inspectors during an emergency operating procedure is consistent with the direction and expectations of the NRC Executive Director for Operations and what is needed to optimize plant safety.

B. Examine the Oconee S/G dryout event and determine if timing of the conference call and/or the conference call information gathering technique is appropriate with a nuclear station in a declared emergency state and in the process of accomplishing an emergency operating procedure.

The team's interactions with INPO management significantly improved the members' understanding of the areas of concern and led the team to conclude that the NRC's prior understanding of INPO's concerns was not fully accurate. Therefore, the NRC's earlier review effort (attachment 1), aimed primarily at determining whether a resident inspector exceeded his regulatory authority by taking over operational control of the event, did not fully address the INPO concerns provided to the team.

On Tuesday, October 17, 1995, the team met at the NRC Region II offices and interviewed Region II personnel involved with the August 10, 1994, Oconee event and the followup investigation of the appropriateness of NRC actions.

On Monday, October 23, 1995, and Tuesday, October 24, 1995, the team met at the Oconee Nuclear Station. The team interviewed the NRC resident inspectors and station personnel, including the available shift operators from the crews that were on duty August 10, 1994, at the time of the reactor trip and the events that led to the refilling of S/G 38.

During the period November 14-21, 1995, an NRC team member interviewed NRR staff who participated in the August 10, 1994, conference call with the licensee. The interviews were conducted in the NRC headquarters office.

CHRONOLOGY August 10, 19 94 (times are approximate) 4:26 Unit 3 tripped when power was momentary lost to the Integrated AM Control System. This loss of power tripped the main turbine and both main feed pumps by providing a false high S/G level signal and caused the turbine bypass valves (TBVs) to fail partially open (22% for S/G 38 and 11% for S/G 3A).

The main turbine trip caused an

5 anticipatory reactor trip. The TBVs' partially open positions were not immediately known to the control room-operators but the lower S/G 3B steam pressure provided immediate indication of a possible steam leak.

4:27 Operators isolated S/G 3B, per the EOP, because of a suspected steam leak. Motor-driven emergency feedwater pump 3B was shut down as part of the.isolation process.

4:42 Unit 3 was stabilized in a hot-shutdown condition with decay heat removal via forced reactor coolant system (RCS) flow. S/G 3A was fed by emergency feedwater pump 3A and the turbine-driven emergency feedwater pump. S/G 3B was isolated and essentially dry.

4:57 An NOUE was declared due to secondary-side depressurization and entry into the EOP for excessive heat transfer.

5:13 NRC involvement began when the licensee placed an event notification call to the NRC headquarters operations officer and shortly thereafter a phone call to the NRC Senior Resident Inspector (SRI) at his.home.

5:30 SRI arrived in the control room, observed activities, scanned the control room panels, and had a brief discussion with the shift supervisor to ascertain the plant status and the licensee's immediate plans.

6:00 Operators were preparing to feed S/G 3B IAW EOPs but delayed to determine operability of the TBVs and verify integrity of the B steam header. The shift manager was concerned about thermal shock to the S/G tubes and the potential.for radioactive material release if "cold" emergency feedwater was used to feed the dry and isolated S/G.

The plant had been operating with elevated fission product activity in the reactor coolant. The Operations C-shift staff decided to conduct the shift turnover before proceeding with any further evolutions. The bases for the decision included: (1) the plant was considered stable and there was no urgency on the part of C-shift to refill S/G 3B (2) the operators felt they had reached an appropriate stopping place in the EOP to seek management direction, and (3) it was common practice near the end of a shift to await shift turnover before proceeding with a new evolution.

6:15 Shift turnover started.

6:30 The Station Manager and Operations Superintendent arrived in the control room.

The Unit 1 NRC Resident Inspector (RI) arrived in the control room to help the SRI monitor control room activities.

6 The Operations Superintendent supported C-shift's decision to wait until after the shift turnover to fill S/G 3B. Licensee concerns remained about the potential for cold-water shock to the S/G tubes if emergency feedwater (EFW) was used, about the integrity of the steam header, and about the potential for further erratic.operation of the TBVs. The licensee decided to try to use the warmer main feed system to fill S/G 3B but first wanted to discuss the situation with B&W Nuclear Technologies. Work continued to evaluate TBV operability and steam line integrity.

The SRI discussed S/G 3B recovery plans with the Operations Superintendent. Topics discussed included licensee assessments of methods for minimizing stress on the S/G tubes during refill, establishing the operability of the TBVs, and establishing the condition of the main steam piping. The SRI agreed with the Operations Superintendent's assessment that the plant was in a stable condition and did not challenge the licensee's current or proposed actions.

6:45 The SRI called the Region II office from the RI's office to tell them about the event and the licensee's plans. They discussed the fact that the licensee was considering a procedure change to the EOP to allow feeding the isolated 3B S/G with main feedwater rather than emergency feedwater. The SRI said that the plant was considered stable and the licensee was not in a hurry to fill S/G 3B. Region II officials informed the SRI that, if time permitted before the licensee planned to feed S/G 3B, it would be desirable to have a conference call with the licensee to discuss plant status, ongoing activities, and S/G 3B recovery plans.

7:15 The SRI conveyed the request for a conference call to the licensee (recollections differ regarding with whom in the licensee's organization the SRI arranged the call; however, it was most likely a supervisor in the Regulatory Affairs department).

7:45 After participating in the call to the Region II office, the NRC's Unit 3 RI arrived in the control room and relieved the Unit 1 RI.

Licensee discussions continued about the use of main feedwater, the thermal-shock concern, high fission products in the primary coolant, and the possible need for a procedure change to the feed system valve lineup procedure.

The Station Manager departed from the control room for the 8:0; a.m.

Plan-of-the-Day meeting and was informed by a RI that Region II would like to conduct a conference call before the S/G was returned to service.

The Station Manager acknowledged the request for a conference call.

l I0 7

8:15 Instrumentation & Control technicians determined that the TBVs for S/G 3B were operable. No main steam line leakage was identified.

The licensee remained satisfied that the plant was in a stable condition and that there was no urgency to refill S/G'3B before thoroughly evaluating all conditions and options. The Operations Superintendent directed the control room staff to call him if the pressure in.S/G 3B dropped from approximately 800 psig to 400 psig.

This directive was based on a concern for differential pressure across the S/G tubes.

The SRI returned to the control room and discussed the licensee's current plans with shift and Operations department management in order to establish a proposed conference call time that would not impact or delay the licensee's efforts to refill the S/G.

Based upon the information provided, the SRI proposed 10:00 a.m. to the Regulatory Affairs department. The time for the call was apparently agreed to based on the licensee's 11:00 a.m. time estimate to refill S/G 3B. The SRI and Regulatory Affairs arranged for this proposed conference call in the manner they normally did at Oconee. It was not standard practice for the resident staff to coordinate conference calls with licensee line management (Operations Superintendent, Station Manager, etc.) and they did not do so on this day.

8:30 The Station Manager returned to the control room to further evaluate the most appropriate course of action.

Duke staff contacted Babcock & Wilcox Nuclear Technologies (BWNT) to discuss feeding a dry S/G with EFW. BWNT responded that it was acceptable to slowly "trickle-feed" the dry S/G with EFW.

8:45 The Unit 3 RI was approached by a licensee staff member who expressed concern that the NRC was preventing the licensee from refilling S/G 3B by putting a "hold" on refill until the NRC-licensee conference call had been completed. He further stated his view to both the NRC and licensee management that the S/G should be refilled as soon as possible and that it was not appropriate to leave it in an abnormal dry and isolated condition.

(Later interviews determined that no control room operators believed that there was an NRC hold on any plant operations and that the delay was to establish plant conditions for refill and to consult with BWNT.)

The RI responded that he was unaware of any such hold and suggested that the individual pursue his concern with licensee management.

The RI immediately informed the SRI of the perceived "hold."

In-the control; room, the SRI ensured that licensee management clearly understood that there was no hold on any licensee operations.

10:00 A conference call was held among the licensee, the resident inspectors, Region II staff, and NRR staff. The Station Manager and the Operations Superintendent discussed the event, current plant status, and recovery plans, including supplying feedwater to S/G 3B.

Although most of the time was spent discussing plant conditions and

8 plans for refilling S/G 3B, a series of questions from the NRR staff were considered by many of those involved to have detracted from the productivity and efficiency of the call.

Specifically, NRR staff asked the licensee several questions about whether existing analyses adequately addressed feeding a dry steam generator. (From interviews, the questions appear to have stemmed from a concern that feeding S/G 3B with "cold" water could place the S/G tubes in tension. If the integrity of the S/G tubes was dubious (because of circumferential cracks), there was a concern whether existing analyses accounted for potentially degraded tubes.)

The questions were viewed as challenging the appropriateness of the previously approved EOPs and their bases. Detailed responses to these questions would have required significant additional research. The difficulties presented by this line of questioning, which could have delayed refilling the S/G while the licensee researched calculations and/or performed additional review of the EOP. bases, were recognized by NRC management on the call.

NRR told the licensee that the NRC was not questioning Oconee's EOPs or Duke's authority to implement them. The conference call was concluded with an understanding that S/G 3B would be refilled with main feedwater to minimize thermal shock to the S/G tubes. The call lasted about 40 minutes.

10:40 Unit 3 control room personnel informed the Operations Superintendent that S/G 3B had depressurized to 400 psig. Shortly thereafter, the operators were instructed to prepare to feed S/G 3B with normal feedwater.

While in the Unit 3 control room, an Operations support procedure writer informed the Operations Superintendent that the S/G tubes-in-compression differential temperature limit had been exceeded. The S/G tube-to-shell differential temperature limit of 60 degrees was exceeded by approximately 22 degrees.

11:15 Operations attempted to place a main feedwater pump back in service but experienced problems with the steam supply valve.

12:05 The licensee began feeding S/G 3B with a combination of main (hot) and emergency (cold) feedwater. Actually, emergency feedwater flow was established first when plant personnel realized that the S/G tube-to-shell differential temperature limit was exceeded; main feedwater flow was then established.

1:00 3B S/G was refilled.

1:37 The Unusual Event was exited.

DISCUSSION OF EXISTING POLICY AND GUIDANCE

9 NRC responsibilities and guidance to the NRC staff regarding oversight of nuclear power plant operations are derived from the Energy Reorganization Act of 1974 and can be found in numerous documents ranging from the Code of Federal Regulations to specific inspection procedures and training course outlines. Section 1.43 of Title 10 of the CFR (10CFR Part 1.43) states in part that the office of Nuclear Reactor Regulation (a) Implements regulations and develops and implements policies, programs, and procedures for all aspects of licensing, inspection, and safeguarding..;

(b) Identifies and takes action regarding conditions and licensee performance that may adversely affect public health and safety, the environment, or the safeguarding of nuclear reactor facilities; (c)

Assesses and recommends or takes action regarding incidents or accidents.

Section 50.72 of 10 CFR requires the licensee to notify the NRC Operations Center via the Emergency Notification System of the declaration of any of the classes of emergency specified in the licensee's approved Emergency Plan. This was the basis for Oconee's initial notification of the NRC.

Section 50.72(c)(3) goes on to require that the licensee maintain an open, continuous communication channel with the NRC Operations Center upon 5

request by the NRC. This continuous communication channel was not requested, during the Oconee event but this requirement indicates the importance placed on timely and accurate information. For events such as the August 10, 1994, event at Oconee, NRC management commonly relies on both the resident inspector staff and the licensee for real-time information about events. RIs are expected to develop their own independent assessment of licensee performance and plant conditions through personal inspection of plant conditions and discussions with licensee personnel, but NRC management also relies on direct discussions with licensee management to ensure that there is a common understandin' of tK:

situation and that the licensee clearly understands the reasons for th current and proposed regulatory response.

Several documents, such as the "NRC Incident Response Plan" (NUREG-0728) and "Concept of Operations" (NUREG 1471), discuss NRC responsibilities, organizations, and operations. Although written to discuss events more significant than the Oconee event of August 10, 1994, these documents set forth the agency's.philosophy for responding to various types of operational events. Operational responsibility and the relationship between NRC and the licensee is described in NUREG 0728: "During an incident at a licensed facility, the licensee is at all times responsible for mitigating the consequences of the incident. The licensee is also responsible for providing appropriate protective action recommendations to State/local officials."

NUREG 1471 describes the NRC's independent assessment of plant conditions functions:

10 The NRC monitors the status of a reactor involved in an accident to:

- assess and predict reactor core and containment conditions, assess licensee's understanding of the event and confirm the classification

- provide technical and logistical support to the licensee as requested.

Although the regulations, NUREGS, and other documents vary in their level of detail, there are several consistent messages that relate to NRC involvement in all types of operating reactor events.

First and foremost, the licensee is responsible for the safe operation of the facility. This responsibility includes taking any and all actions deemed necessary to protect public health and safety. Second, the primary role of the NRC is to monitor the activities of the licensee to ensure that appropriate actions are being taken by the licensee. The resident inspection staff is expected to provide the initial site coverage and assessment function. The NRC's goal is to perform its monitoring and assessment function with as little impact on the licensee as is possible and at the same time ensure that NRC's evaluation is timely and accurate enough to take the appropriate regulatory actions. A balance is necessary. The appropriate balance involves numerous variables, including safety significance of the event, the complexity of the event, time constraints, and available staffing.

Reaching the correct balance is dependent on the personal judgment of the key NRC and licensee personnel involved.

ANALYSIS OF RESIDENT INSPECTOR ACTIONS NRC management expectations and policy are that the resident inspector staff will respond to the site during an event to monitor licensee actions and verify that the licensee acts in accordance with the regulations, the license, and EOPs as appropriate. The residents are not granted authority to direct the licensee's operation of the facility.

Each member of the resident inspection staff and the Region II staff interviewed clearly indicated a desire to minimize impacts on the licensee.

The resident staff appeared to consciously limit direct interaction with the board operators so as not to interfere with their licensed duties.

Resident-licensee interactions were mostly with licensee management.

During each licensee interview the team tried to assess the appropriateness and impact of the resident inspectors' actions during the recovery from the S/G dryout event.

Licensee staff (including on-shift control room operators) and management were specifically requested to recall all interactions with the resident staff and to evaluate the residents' impact on the event.

Every control room operator interviewed considered the residents' actions to be appropriate, believed the residents were neither intrusive nor disruptive, and were generally unaware of any concern regarding an NRC hold on operations until the next day.

Licensee operations and station management shared the operators' view about

11 the residents' performance during the event. However, intentionally or not, the resident inspectors left an impression with some members of the licensee's organization that the NRC wanted the licensee to wait until after the conference call before feeding S/G 3B, even though licensee management acknowledges that there was no NRC hold. The team concluded from the interviews with the Region II staff that they did want the message conveyed to the licensee that they desired a conference call, if time permitted before the licensee started to refill S/G 3B, but not if the call would delay the licensee's recovery efforts.

Licensee management may make a spectrum of interpretations when the NRC requests such a conference call.

These "situational" interpretations can, and did, range from a perceived verbal "hold" being made, to a high confidence level on the part of licensee management that they fully understand that there would not be some form of retribution for initiating refill of the S/G without the prior conference call.

Although subtle, this point is important. Because of the authoritative role of the NRC, licensees listen carefully to the residents and may interpret statements, side remarks, or observations as directives or requirements. Consequently, open, clear, and direct communications between the NRC and the licensee are particularly important during events.

ANALYSIS OF THE NRC-LICENSEE CONFERENCE CALL The NRC values conference calls as an efficient method of obtaining accurate and timely information. They promote a mutual understanding of the facts and concerns and help the NRC respond appropriately to events and conditions. Given the licensee's and the NRC senior resident inspector's assessments that the plant was in a stable condition and there was no urgency on the part of the licensee to refill S/G 3B, the Region believed it was reasonable to request a conference call for a time to be agreed to by the licensee.

In requesting the SRI to set up a conference call with the licensee, the Region II staff specifically intended that the call not interfere with the licensee's activities to refill the isolated S/G.

The purpose of the call was to ensure a mutual understanding of the event, the current plant conditions, and the licensee's plans.for recovery, including S/G refill.

Scheduling the conference call was jointly done by the SRI and the licensee's Regulatory Affairs group. This was the standard practice (by the licensee's wish).

As a result, the licensee's line management "inherited" the decision for, and time of, the conference call.

Once the 10:00 a.m. time was established, the conference call was apparently integrated into the licensee's schedule, possibly giving some licensee staff the impression that the conference call was a prerequisite to S/G refill.

At least one licensee staff member was convinced that the NRC had a "hold" on S/G refill until after the conference call.

Licensee management understood that the call was not a prerequisite to refill and the NRC did not have any hold on operations.

Still, some licensee staff

12 believed that the NRC expected that the call would be conducted before refill.

Licensee management involved in the event thought that preparation for, and conduct of, the conference call did divert the licensee's attention and resources to some degree. Licensee management clearly recognized, understood, and agreed with the NRC's need for timely and accurate information to fulfill its mission of protecting public health and safety.

As a result, licensee management desired to support the NRC. Although licensee management indicated they felt they could change the time of the call if they wished, they seemed to have been hesitant to question the need for a conference call or its timing.

The hesitancy to question the need for conference calls is not unique to licensee staffs. Some NRC personnel interviewed believed that more effective use of the resident staff's knowledge would have avoided (or at least delayed) the need for a conference call.

On the other hand, other NRC staff believe that the NRC response decisions should not normally be made without hearing the licensee's views first hand:

NRC event response decisions made without hearing the licensee's views would be too dependent on an individual NRC resident inspector's views (which may differ from licensee's) on the situation, including the licensee's performance and the safety significance of an event.

With regard to the conference call itself, the NRC had the appropriate technical-staff and management.on the call.

Because of the complexities of the event, the staff had numerous questions. The licensee had some difficulty responding to a few questions about system response (i.e. how far pressurizer level dropped) as is not unusual shortly after an off normal event. However, several NRC managers on the call sensed that licensee management was starting to look ahead toward plant restart before having fully considered the ramifications of the event. The Region II Deputy Regional Administrator discussed this concern with the plant manager on August 11.

Both Region II and NRR management concluded that several of the staff's questions aimed at confirming the adequacy of the EOP bases, were not appropriate at that time. NRC action was taken promptly to discontinue that line of questioning, which had the potential for delaying refill of S/G 3B while the licensee researched the answers to the questions. The NRC's reasons for challenging the adequacy of the EGPs did not warrant delaying the licensce's-planned refill activities.

13 CONCLUS IONS

1)

The actions and level of involvement of the NRC, resident inspectors were consistent with the NRC's mission and policies, and were appropriate for the reported plant conditions. The NRC resident inspectors did not exceed their authority. However, the fact that some members of the licensee's organization were left with the impression that the NRC wanted the licensee to wait until after the conference call before feeding S/G 3B (even though the licensee acknowledges there was not a NRC hold on them) shows the importance of open, clear, and direct communications between the NRC and the licensee, particularly during events.

2)

The NRC resident inspectors were responsive in correcting with Oconee management a misperception by a licensee employee that the NRC had put a hold on refilling S/G 3B until after the 10:00 a.m. conference call.

3)

Although the licensee had declared a Notification of Unusual Event (NOUE) and was in the process of completing plant recovery, licensee management considered the plant stable and felt no sense of urgency to refill the dry S/G. Nonetheless, the residents were mindful of minimizing impact on licensee staff while they gathered information and assessed the potential safety significance of the situation.

4)

Based on plant conditions as reported by the licensee, the fact that an NOUE had been declared and EOPs were still in use had no bearing on the appropriateness of NRC's'request for a conference call.- The NRC's attempt to arrange the conference call for a time which would not delay the licensee in refilling the S/G was appropriate:

5)

Although talking.directly with the licensee helped the NRC accomplish its mission, the conference call did divert some licensee management attention from S/G recovery and may have slightly delayed S/G refill.

Based on the licensee's assessment that plant conditions were stable and the fact that there was no hold on licensee actions, the NRC's request to hold the conference call was appropriate. However, greater use of the resident inspection staff to,provide information to regional and headquarters staff could have minimized the impact on the licensee. For example, one of the resident inspectors could have prebriefed NRC staff planning to participate in the call so that the call could have been shorter.

6)

Both Region II and NRR managemnent *n the conferatce call concluded that several of the staff's questions aimed at confirming the adequacy of the EOP bases, were not appropriate at that time. NRC action was taken promptly to discontinue that line of questioning.

However, no form of NRR corrective action (discussion in staff meetings, memorandum, etc.) was taken to communicate more broadly to the NRR staff "lessons learned" from the conference call in an attempt to sensitize the staff and improve overall performance during future conference calls.

14 RECOMMENDATIONS As a result of its review, the team has the following recommendations for NRC action:

1)

Review and modify, as appropriate, NRC guidance such as inspection procedures and NRC Manual Chapters to ensure expectations are clear for NRC staff who interact with licensees during regulatory response to off-normal events. Also, develop guidance on scheduling and conduct of NRC-licensee conference calls.

Include guidance on the timing of conference calls and on deciding whether a conference call is needed. Train staff on changes to procedures and reinforce expectations.

2)

Identify and use appropriate forums for discussion between NRC and industry counterparts to ensure a good level of understanding and agreement regarding each other's needs and expectations during off normal events. The role of the resident inspectors, considerations associated with conference calls, and revisions to NRC guidance documents should be discussed. Suggested forums for discussion include routine meetings between resident inspectors and appropriate licensee managers and staff, routine NRC-licensee management meetings, and appropriate industry meetings and workshops.

3)

Review/modify the NRC's procedure, "Guidance for Management Resolution of Inappropriate Regulatory Actions by NRC Staff," to ensure that concerned individuals and the NRC have a common understanding of the details of the concern(s).

4)

Evaluate the sufficiency of existing BWNT analysis and/or test data to ensure that adequate bounding parameters (S/G tube condition) have been considered. (Attachment 3)

15 ATTACHMENTS:

1.

Letter fromJ. Taylor(NRC) to Z. Pate(INPO), dated July 10, 1995

2.

Memorandum from J.Taylor(NRC) to W.Russell(NRC), dated September 14,1995 - Team Charter

3.

Safety Assessment associated with Dry S/G Recovery

Mr. Zack T. Pate President and Chief Ex ive Officer Institute of Nuclear P Operations 700 Galleria Parkway Atlanta, Georgia 30339-5957

Dear Mr. Pate:

On March 8, 1995, you informed me of a concern in which you believed an NRC Resident Inspector exceeded his regulatory authority by taking over operational control of the recovery actions at Oconee Unit 3, following a reactor trip on August 10, 1994. In response to your concern, I directed the NRC staff to conduct a review of the reactor trip and the circumstances surrounding the subsequent recovery actions.

Based upon a comprehensive review of available records and interviews with the key personnel involved in the event, including NRC staff and the Duke Power Company personnel, we have concluded that the Resident Inspector did not exceed his authority. While it was determined that some members of the Oconee Operations staff believed that the NRC delayed Oconee's response to the event, this perception was apparently due to a mis-communication by the licensee's management. I have enclosed the staff's detailed review of this event for your information (Enclosure 1).

Additionally, jou indicated that there were two other events in which the NRC staff's actions may not have been appropriate. These issues will be addressed with Mr. Joe Colvin of the Nuclear Energy Institute (NEI) in separate correspondence.

I believe that these concerns serve to underscore the importance of clear communications between the NRC and licensees. The Commission recently issued a policy statement to establish its expectations in this regard. It is enclosed for your information (Enclosure 2).

If you have any questions, please call me or Mr. James L. Milhoan.

Si ncer1 afl signed by JaI it. Taylor James M. Taylor Executive Director for Operations

Enclosures:

As stated REVIEW OF NRC INVOLVEMENT IN OCONEE STEAM GENERATOR DRYOUT EVENT

1. Concern A concern was expressed that on August 10, 1995, a Resident Inspector at Oconee exceeded his regulatory authority by taking over operational control of an event. Specifically, it was stated that following a reactor trip with a stuck open turbine bypass valve, which led to dryout of one of the two steam generators, a Resident Inspector told Oconee operators not to refill the steam generators until after a conference call was held with the NRC staff.

2.

Approach A comprehensive review of this concern was conducted by reading all available records of the event and by interviewing the key NRC and licensee personnel. Specifically, a review was conducted of Duke Power Company's 50.72 report, licensee event report (LER), and the NRC's Preliminary Notification, inspection report, Notice of Violation, and related NRC/Duke Power Company correspondence.

Following the review of the documents, interviews were conducted in Region II, at the site, and at NRC Headquarters in Washington with the following people:

Region II Region II Deputy Regional Administrator Reactor Projects Section Chief Reactor Projects Project Engineer for Oconee NRC Headauarters Director, Division of Reactor Projects -

I/II, NRR Deputy Director, Division of Engineering, NRR Director, Division of Reactor Controls and Human Factors, NRR Oconee Project Manager, NRR Oconee Site Oconee Site Vice President (Duke Power Company)

Oconee Plant Manager (Duke Power Company)

Senior Resident Inspector (RH)

All three Resident Inspectors (RH)

The results of the document review and interviews are provided in Sections 4 and 5 of this attachment.

-2

3.

Background

At 4:25 a.m., on August 10, 1994, Oconee Unit 3 experienced a reactor trip from 100 percent power. The cause of the trip was the loss of both main feedwater (MFW) pumps. The automatic trip of both main feedwater pumps occurred as required when the associated Integrated Control System (ICS) temporarily lost power (less than 1 second). The temporary loss of ICS power-occurred when fuses internal to the 3KI inverter blew.

Power to the ICS was restored when the "ASCO" transfer switch, immediately downstream of the inverter, automatically transferred to the alternate power source (AC Regulated Power System). When the ICS was repowered, the Turbine Bypass Valve (TBV) went to manual and randomly positioned at 22% open for the steam generator 3B valves and 11% open for the steam generator 3A valves. The operators did not immediately recognize the temporary loss of panel board KI, or the status of the TBVs. Due to the subsequent divergence of steam generator pressures (i.e., steam generator 3B at 600 psig and steam generator 3A at 800 psig) and Reactor Coolant System (RCS) cooldown, a steam leak was suspected.

This prompted the isolation of the 3B steam generator, which was completed at 4:27 a.m., with the 3B steam generator pressure at 550 psig. A Notice of Unusual Event (NOUE) was made at 4:57 a.m.,

due to the secondary side depressurization, which required entry into the "Excessive Heat Transfer" section of the Emergency Operating Procedures (EOP).

The lowest post-trip RCS temperature was 524 degrees fahrenheit. The 9

lowest pressurizer level was 35 inches. After the isolation of the 3B steam generator, RCS average temperature (Tave) was maintained at approximately 538 degrees fahrenheit with the 3A Motor Driven Emergency Feedwater (MDEFW) pump and the Turbine Driven Emergency Feedwater (TDEFW) pump feeding the 3A steam generator. Additionally, a small amount of emergency feedwater was leaking past isolation valve FDW-316 into the 3B steam generator. This leakage was not enough to establish any level in the 3B steam generator, but did maintain steam generator p,,sQsure at approximately 800 psig on 3B (saturation pressure for water at 538 degrees fahrenheit is 946.88 psia).

The steam generator pressure readings indicated that the 3B steam generator was dry with only a steam atmosphere. The leakage into the 38 steam generator only became apparent much later (7:51 a.m.), after the TDEFW pump was secured and the 3B steam generator pressure began to decrease.

With the primary plant stable, and the 3B steam generator maintaining approximately 800 psig, the operators elected to maintain the 38 steam geverator isolated until after shift turnover (6:30 a.m.),

This was due in part to the perceived risk of feeding this essentially dry steam generator with relatively cold emergency feedwater.

Although guidance existed in the licensee's EOP for recovering a hot/dry steam generator, due to the amount of time the SG had been isolated, the licensee contacted the steam generator vendor (Babcock & Wilcox) to

-3 consult over concerns of possible thermal shock to the tube sheet region

. upon initiation of feedwater. B&W indicated that there were no additional concerns associated with recovering the steam generator and that they should follow their existing EOP guidance. At approximately 7:51 a.m., the TDEFW pump was secured which caused the 3B steam generator pressure to slowly decrease. After the 3B.steam generator pressure began decreasing, the differential temperature between the steam generator shell and tubes began to increase. As the licensee prepared to recover the 3B steam generator, they became aware that there was a limit of 60 degrees for the differential temperature (tubes hotter than the shell) listed in the Babcock & Wilcox Technical Basis Document.

At 11:33 a.m., the 3B MDEFW pump was restarted,iwhich due to the leakage past 3FDW-316, allowed some recovery of pressure (and differential temperature) in the 3B SG. At approximately 11:50 a.m.,

MFW pump B was restarted and used to feed SG 3A. At 12:02 p.m., a combination of MFW and EFW was used to feed SG 3B through the auxiliary feedwater ring. At approximately-1:00 p.m., the level and pressure in steam generator 3B were recovered. At 1:37 p.m., on August 10, 1994, the licensee exited the NOUE and remained in hot shutdown to conduct their post trip review.

The timeline of the event and recovery actions is attached.

4.

Results of the Review of the Concern The Senior Resident Inspector was notified of the event by the licensee, and responded to the site at approximately 6:30 a.m. At this time, the event had occurred two hours earlier, the steam generator was still dry, and the licensee had decided to wait until after shift change to refill it.

The Senior Resident Inspector called his supervisor in the Region at approximately 7:00 a.m., leaving a Resident Inspector in the control room to monitor the event. During the 7:00 a.m. call to the Region, the decision was made to request a conference call between the licensee, NRR, and Region II in order to assure the event was clearly understood by the NRC. The Senior Resident conveyed this request to the Plant Manager, and 10:00 a.m. was agreed upon as a convenient time.

Subsequent discussions with the Plant Manager as part of the review of the concern confirmed that he understood this request did not place an operational hold on refilling the steam generator. In-fact, the Plant Manager welcomed the opportunity to brief all the involved parties within the NRC at one time, understahding that the quality and currency of the information on the.event used by the NRC would influence the decision made regarding NRC's response to the event.

By 7:30 a.m., the Resident Inspector assigned lead responsibility for Unit 3 had relieved the Resident Inspector in the Unit 3 Unit Control Room who was monitoring the recovery activities. Shortly after his arrival in the control room, the Unit 3 Resident Inspector was told by a S1

member of the licensee's operations support staff that NRC had placed a hold on refilling the steam generator until after a 10:00 a.m.

conference call.

The Resident Inspector immediately contacted the Senior Resident who confirmed with the Plant Manager and Operations Manager that no hold had been placed on.Oconee with regard to their response to the event. Both the Plant Manager and the Operations Manager stated they understood and had not inferred otherwise. In subsequent discussions with the Plant Manager as part of the review of this concern, he indicated'that he and the Operations Manager understood the reason for delaying the refill of the steam generator was based solely on Duke Power Company's decision to proceed cautiously, but his instructions to his subordinates regarding his agreement to a 10:00 a.m.

conference call with NRC had been misinterpreted by some to constitute a hold point in the refilling evolution.

At 10:00 a.m., the conference call was held as scheduled between Region II, NRR, and Duke Power Company. During this call, an NRR reviewer from Mechanical Engineering Branch began asking questions about the EOPs, their basis, and the appropriateness of following them if the result is to trickle feed a dry steam generator with cold emergency feedwater.

Shortly after this line of questioning emerged, the Region II Deputy Regional Aministrator contacted the Director of the Division of Reactor Projects -I/Il, NRR (DRP I/II) on a separate telephone line to assure that the operational situation was clearly understood, and that any concerns with Oconee's-EOPs or their plan to implement them were conveyed at the appropriate management level.

Following this conversation, the Director, DRP I/II called the Deputy Director, Division of Engineering, NRR (DE) away from the conference call in progress, discussed the issue, and the DE returned to the call and clarified that NRR had no technical problems with Oconee's EOPs. The Deputy Regional Administrator then stated that Region II was neither questioning Oconee's EOPs nor Duke's authority to implement them.

Subsequent discussions with the Plant Manager as part of the review of thiz issue revealed that the brief discussion and questions about the basis of Oconee's EOPs was a minor distraction, but did not result in any lasting confusion, nor was it a particularly memorable occurrence.

What was memorable to the Plant Manager was that this had been his first event related conference call with the NRC as Oconee Plant Manager and he felt he could have done a better job of organizing and presenting the facts of the event and the recovery plan. In fact, the Plant Manager recalled that the Region II Deputy Regional Administrator had called him the following day to inform him that Duke managers had left the impression of rushing to restart the unit, rather than pursuing a careful and deliberate analysis of the initiating failure, its cause and consequences. The Plant Manager felt this feedback was useful and has helped him improve subsequent communications with the NRC.

5.

Conclusions The concern that a Resident Inspector at Oconee exceeded his regulatory authority by taking over operational control of an event was reviewed and found to be without basis. While this perception was held to be true by some of the Oconee operations.staff, it was the result of a miscommunication on the part of Duke management, rather than inappropriate actions by the Resident.

Attachment:

Oconee Steam Generator Timeline

OCONEE STEAM GENERATOR DRYOUT TIMELINE August 10. 1994 0426 Unit 3 tripped due to a momentary loss of the Integrated Control System's power. This loss of power tripped both main feed pumps by providing a false high S/G level -signal and caused the turbine bypass 'valves to fail partially open (22% for S/G 3B and 11%,for S/G 3A).

0427 Operators isolated S/G 3B due to a suspected steam leak.

0442 Unit 3 was stabilized in hot shutdown with decay heat removal via forced RCS flow with 3A S/G being fed by emergency feedwater and 3B S/G, isolated and essentially dry.

0457 NOUE declared due to secondary side depressurization and entry into EOP for excessive heat transfer.

0513 50.72 Report made which stated, "the Resident Inspector will be notified." Resident(s) subsequently respond to the site.

0630 Residents arrive on site. Shift change occurs. The transient was discussed with station management by the Resident Inspectors. A conference call was requested which would include Oconee site management, NRC regional management, NRR, and the Residents to brief the NRC on the event. As a result, some licensee participants concluded the NRC position was that 3B S/G should not be fed without prior NRC concurrence. When the Resident realized this impression was given, he clarified that the licensee had the authority to decide on actions to cope with plant transients.

1000 The conference call requested by Region II management to discuss Oconee's plan to trickle feed the dry S/G with emergency feedwater is conducted.

1115 Licensee completed S/G Recovery Plan development after contacting the S/G vendor (B&W) and considering the potential problems with cold feeding a S/G.

1205 Licensee begins feeding 3B S/G with a combination of main (hot) and emergency (cold) feedwater.

1330 3B S/G is ref iled.

1337 NOUE is exited.

EnciOSure z NRC POLICY ON COMMUNICATIONS BETWEEN THE NRC AND LICENSEES In 1991 the Commission established the NRC Principles of Good Regulation, a copy of which is attached for reference. As noted in the Principles, the Commission believes that good regulation must be transacted publicly and candidly and that open communications must be maintained with Congress, other government agencies, licensees, and the public. Regulatory actions should always be fully consistent with regulations and should be promptly, fairly, and decisively administered.

The Commission encourages and expects open communications at all levels between its employees and those it regulates. Licensees should feel unconstrained in communicating with the NRC. The Commission also *expects that the NRC staff exercise initiative in maintaining open lines of communication and ensure that its regulatory activities are high quality, appropriate and consistent. The Commission recognizes that honest, well intentioned differences in opinions between the staff and the licensee will occasionally occur, and therefore encourages open communications to foster an environment where such differences receive constructive and prompt resolution.

Open communication also extends to the reporting of perceived inappropriate regulatory actions by NRC staff when dealing with licensees. The Commission encourages licensees to provide specific information regarding such concerns.

The NRC will not tolerate inappropriate regulatory actionsv by the NRC staff, nor will it tolerate retaliation or the threat of retaliation against those licensees who communicate concerns to the agency. NRC staff whose actions are found to be contrary to this policy could be subject to disciplinary actions in accordance with the NRC Management Directive 10.99, "Chapter 4171, Discipline, Adverse Actions and Separations," or in accordance with the Collective Bargaining Agreement Between the U.S. Nuclear Regulatory Commission and National Treasury Employees Union.

Inappropriate regulatory actions include activities which exceed the agency's regulatory authority, involve improper application of agency requirements, or adversely affect the agency's regulatory functions.

Examples of inappropriate regulatory actions include, but are not limited to, unjustified inconsistent application of regulations and guidance by NRC staff or management that significantly affect licensee activities and inappropriate action on the part of NRC staff and management that disrupts effective communication with the licensee.

NRC PRINCIPLES OF GOOD REGULATION INDEPENDENT. Nothing but the highest possible standards of ethical performance and professionalism should influence regulation. However, independence does not imply isolation. All available facts and opinions must be sought openly from licensees and other interested members of the public. The many and possibly conflicting public interests involved must be considered. Final decisions must be based on objective, unbiased assessments of all information, and must be documented with reasons explicitly stated.

OPEN. Nuclear regulation is the public's business, and it must be transacted publicly and candidly. The public must be informed about and have the opportunity to participate in the regulatory processes as required.by law. Open channels of communication must be maintained with Congress, other government agencies, licensees, and the public, as well as with the international nuclear community.

EFFICIENT. The American taxpayer, the rate-paying consumer, and licensees are all entitled to the best possible management and administration of regulatory activities. The highest technical and managerial competence is required, and must be a constant agency goal.

NRC must establish means to evaluate and continually upgrade its regulatory capabilities. Regulatory activities should be consistent with the degree of risk reduction they achieve. Where several effective alternatives are available, the option which minimizes the use of resources should be adopted. Regulatory decisions should be made without undue delay.

CLEAR. Regulations should be coherent, logical, and practical. There should be a clear nexus between regulations and agency goals and oDjectives where explicitly or implicitly stated. Agency positions should be readily understood and easily applied.

RELIABLE. Regulations should be based on the best available knowledge from research and operational experience. Systems interactions, technological uncertainties, and the diversity of licensees and regulatory activities must all be taken into account so that risks are maintained at an acceptably low level. Once established, regulation should be perceived to be reliable and not unjustifiably in a state of transition.. Regulatory actions should always be fully consistent with written regulations and should be promptly, fairly, and decisively administered so as to lend stability to the nuclear operational and planning processes.

PV' UNITED STATES 49 NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20555-4001 September 14, 1995 MEMORANDUM TO:

William T. Russell, Director,R FROM:

James M. Taylor Executive Dire for Op ations

SUBJECT:

ESTABLISHMENT OF OCONEE EVENT REVIEW TEAM On March 8, 1995, Institute of Nuclear Power Operations (INPO) President Mr.

Zack Pate, informed me that he believed an NRC Resident Inspector exceeded his regulatory authority by taking over operational control of the recovery actions associated with the steam generator dryout event at Oconee Unit 3, following a reactor trip on August 10, 1994. In response to INPO's concerns, I directed the staff to conduct an independent review of the reactor trip and the circumstances surrounding the subsequent recovery actions. I informed Mr.

Pate of the results of this review in a letter dated July 10, 1995.

Recently, Mr. Pate informed me that INPO has continuing concerns related to this event. Although the staff found that the Resident Inspector did not exceed his authority, INPO questioned the thoroughness of the staff's review.

Mr. Pate also expressed concern about the technical and procedural issues associated with the delay in the licensee's recovery actions. INPO believes that these issues are not well understood and appear sufficiently complex to warrant further evaluation by both the NRC and INPO.

In light of these concerns, you are directed to form an, event review team led by Roy P. Zimmerman, Associate Director for Projects, to conduct an indepth evaluation of the licensee's and the staff's activities.related to this event.

The event review team should include other NRC staff, as you deem appropriate, and provide for participation by one or two INPO personnel and a senior licensee manager (preferably from a B&W plant).

Mr. Zimmerman should coordinate directly with Mr. Bill Kindley, INPO's VP for Government Relations, to facilitate the participation of the INPO and industry representatives.

While this event and the associated issues do not meet the threshold for an Incident Investigation Team (IIT), the event review team should follow Section 1.7 of the IIT procedure with respect to participation by industry organizations.

In carrying out this assignrent, the team shouild be guided by the following:

1.

Maintain independence and objectivity and avoid undue influence by previous staff actions and findings.

2.

Gain a clear understanding of INPO's concerns. Review and evaluate all relevant information, including the INPO SER on this event, and hold discussions with INPO representatives.

William T. Russell

- 2

3.

INPO indicated that the delay in the licensee's actions to recover the dry steam generator raises technical and procedural issues.

Pursue this matter with INPO and the licensee to understand the basis for these concerns.

4.

Gain a clear understanding of the staff's actions related to this event. Evaluate the results of the staff's review of the reactor trip and the subsequent recovery actions. Hold discussions with those staff involved with the review.

5.

Determine the appropriateness of the Resident Inspector's actions following the reactor trip. Interview whatever staff and licensee personnel you deem necessary; however, you should specifically include the plant operators involved in the recovery actions.

6.

Upon completion of the team's review, recommend what if any additional actions the agency should take regarding this event, including the rationale for the recommendations. If there are any lessons learned from the team's review, identify them. The team should submit its report to the Executive Director for Operations (EDO).

7.

Provide periodic briefings to me of your activities.

8.

Maintain records of the information you collect and document the results of interviews to the extent necessary to capture salient information.

9.

This assignment should begin as soon as practical and be conducted expeditiously.

cc: S. Ebneter, RH Z. Pate, INPO OIG pnp SAFETY ASSESSMENT ASSOCIATED WITH DRY S/G RECOVERY

Purpose:

1) Assess the potential safety impact associated with the delay in feeding the dry S/G. 2) Determine whether any technical concerns associated with the S/G recovery appear to remain which warrant additional follow-up.

Background:

A discussion of the event and reference to related documents is provided in the body of this report.

Approach:

Reviewed background documents and conducted interviews of the following individuals:

Duke Power Company Mike Barley, Licensing Engineer, Department of Regulatory Compliance at Oconee Tony Lee, Senior Technical Specialist, Oconee Operations Group Greg Swindlehurst, Section Manager, Division of Nuclear Engineering Babcock & Wilcox Nuclear Technologies Jeff Brown, Supervisor Engineer, Department of Special Projects

& Integrated Services (DSP&IS)

Rick Coe, Manager for Steam Generator Integrity, DSP&IS John Shepard, Engineer, Department of Non-commercial Steam Generator Service Product Tom Smith, Senior Product Manager, Department of Project Management Carl Thurston, Engineer, DSP&IS Summary: 1) Following the reactor trip at 4:42 a.m., the licensee maintained the plant in a hot shutdown condition with S/G 3B essentially dry and isolated. There were several technical concerns the licensee wanted to resolve prior to refilling S/G 3B.

At about 10:05 a.m., the indicated compressive tube-to-shell delta temperature limit of 60 F (tubes hotter than shell) was exceeded for S/G 38.

This limit was not incorporated into plant procedures because the plant procedure writers did not envision a significant delay in refilling an isolated, intact S/G. The operating crew was unaware of the temperature limit until about 10:40 a.m., when a member of the Operations

2 department informed Operations management. The limit was exceeded by 22 F. The criteria used to determine the maximum differential temperature is based on avoiding permanent.

deformation of the tubes and to prevent growth of existing S/G tubes flaws. The differential temperature was reduced below the limit when the SG 3-B was slowly feed by the "B" motor driven EFW pump shortly after 12:05 p.m.

The licensee sent plant transient data to BWNT for assessment after the event was terminated. By letter dated August 10, 1995, BWNT confirmed that the structural integrity of S/G 3B was not adversely affected. Additionally, the licensee performed a post-trip leak test and found no leaking S/G tubes.

During the team's visit to Oconee on October 23-24, 1995, and from subsequent discussions with BWNT, additional information was made available regarding the margins which exist in the maximum differential temperature limit of 60 F. BWNT stated that there were S/G tube structural integrity tests performed at TMI Unit 1 during 1984.

The results of these tests demonstrated that with a compressive, differential temperature of 100 F, the above stated acceptance criteria were still met.

Experience has indicated that the degradation mechanism of most concern for B&W plants is intergranular stress corrosion

/

cracking (IGSCC) and circumferential, fatigue cracks at the 15th support and at the lower face of the upper tube sheet.

Any significant incremental damage to any such pre-existing cracks would be expected to have caused detectable leakage following the transient, whereas no such leakage was detected.

2) It is the NRC staff's understanding that injection of EFW into a hot, dry OTSG with some pre-existing tube degradation is acceptable. However, the extent and type (circumferential versus axial) of degradation in the S/G tubes which can be tolerated in this scenario has not been specifically studied by the staff.

The NRC staff has not reviewed the limits of acceptability for feeding emergency feedwater into a dry S/G with respect to S/G tube degradation. Consequently, the staff plans to evaluate the sufficiency of existing BWNT analysis and/or test data to ensure that adequate bounding parameters (S/G tube condition) have been considered.

Any impacts on existing EOP guidelines would then be evaluated.

It is the staff's.judgement, that the existing EOPs remain acceptable.