ML20210N989
| ML20210N989 | |
| Person / Time | |
|---|---|
| Site: | Rancho Seco |
| Issue date: | 04/25/1986 |
| From: | Harold Denton Office of Nuclear Reactor Regulation |
| To: | Stello V NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| Shared Package | |
| ML20210N994 | List: |
| References | |
| FOIA-86-290, RTR-NUREG-1195 TAC-60462, NUDOCS 8605060359 | |
| Download: ML20210N989 (11) | |
Text
-i_# F April 25,1986 HEMORANDUM FOR: Victor Stello, Jr.
Executive Director for Operations FROM:
Harold R. Denton, Director Office of Nuclear Reactor Regulation
SUBJECT:
STAFF ACTIONS RESULTING FROM THE INVESTIGATION OF THE DECEMBER 26, 1985 INCIDENT AT RANCHO SECO (NUREG-1195)
Your memorandum dated March 13, 1986 identified and assigned responsibility for generic and plant-specific actions resulting from the investigation of the Rancho Seco incident as documented in NUREG-1195.
Per your request, Enclosure 1 provides a sumary of the plans, schedule, and status for each issue assigned to NRR. The information requested for those issues assigned to Region V is also enclosed (Enclosure 2) per discussions with Region V.
Also enclosed (Enclosure 2) is an NRR draft action plan showing the major milestones relating to the restart of the Rancho Seco plant.
Region V's response to your memorandum was sent to your office as a copy of a memorandum from D. F. Kirsch to F. J. Miraglia dated April 8, 1986.
Originst signed bl' Waro. R. Dentenld R. Denton, Director g
Office of Nuclear Reactor Regulation
Enclosures:
As Stated
\\
gW5dd,'2ri/
o Mk
STAFF ACTIONS ASSIGNED TO NRR RESULTING FROM THE INVESTIGATION OF THE DECEMBER 26, 1985 INCIDENT AT RANCHO SECO
(
REFERENCE:
NUREG-1195) 1.
Issue: Adequacy of the Auxiliary Feedwater System.
i Action:
I a.
Verify the acceptability of the existing initiation and i
control of the AFW system (Section 7.2).
1 b.
Determine the status of any licensee commitment to install the EFIC system, and determine the acceptability of the current 1
schedule for installation. (Principal Finding #11).
Status:
a.
Prior to startup from the current outage, SMUD has committed to modify the control logic of the atmospheric dump valves (ADVs), the turbine bypass valves (TBVs), and the auxiliary feedwater ICS control valves. On loss of the ICS, the ADVs Lnd TBVs will close and the auxiliary feedwater ICS control valves will go to a preset opening to assure an adequate flow of feedwater to the steam generators. However, manual control of the ADVs and the auxiliary feedwater ICS control valves, separate from the ICS, will be provided in the control room. With these changes implementsd prior to startup, the staff concludes that the schedule for the installation of EFIC as proposed in SMUD's March 3,1986 letter is acceptable. SMUD's Recovery plan is expected to include l
details of these modifications and will be submitted within a month for staff review and approval.
b.
By letter dated March 3, 1986, the Sacramento Municipal Utility District t
committed to install the majority of the EFIC actuation and control i
systems at Rancho Seco during the next refueling cutage prior to Cycle 8 (1987). The EFIC installation will be completed during the refueling outage prior to Cycle 9.
This cannitment was in response to a staff request in October 1985 regarding improvement to the EFIC schedule. The installation of the EFIC system at Rancho Seco has been impacted by changes needed in the control room to accommodate EFIC, Regulatory Guide 1.97 requirements, the Control Room Design Review, and the requalification prngram for two new TDI diesels needed to meet power requirements of NUREG-0737 modifications including EFIC. The staff has concluded that this schedule is acceptable in light of the interim modifications described above.
L l
i s
-a-
,-e--
.-,e-.
.------n--
- - ~ -,
me.
w-
. With regard to the generic aspects of this issue, all other B&W designed plants except THI-1, have completed the safety-grade upgrade of the AFW (EFW) systems in accordance with Item II.E.1.2 of NUREG-0737. Regarding TMI-1, the short tenn actions required by the restart hearing were completed before the restart in October 1985. The long term items that will result in a safety grade EFW system will be completed during the end of the scheduled December 1986 refueling outage.
In addition, the EFW systems will be addressed as part of the B&W Design Reassessment Program. We expect that this effort will be completed in late 1986 assuming the B&W Owners Group completes their activity in a timely fashion.
2.
Issue: Completer.ess of various staff and licensee actions associated with control systems.
Action a.
In light of the ongoing B&W generic review, assess the need to reevaluate the actions taken by the staff and by the licensees in reJponse to the findings, conclusions and recommendations associated with BAW-1564;Bulletin 79-27; N'JREG-0667; the February 1980 loss of NNI power at Crystal River; the March 19, 1984 partial loss of NNI at Rancho Seco; and BAW-1791.
(Principal Finding #15 and Other Finding #11).
a
- b. Assess the need to expand the scope of USI A 47 to include additional consideration of frequent. events with understrable ccnsequences even if the consequences of a particular event are bounded by the FSAR analysis, and the degree to which events that are not significant at the referenced plants might be significant at other plants (Principal Finding #15h).
Status:
a.
This item will be partially addressed as part of the B3W Design Reassessment Program. During the April 8 meeting, the BWOG stated that they will be perfonning a comprehensive ICS/NNI evaluatien. This activity will also include a review of previous PRAs, FMEAs, standards, NUREGs, and IE Bulletins in order to produce a list of combined recommendations. We-will monitor and review these activities to assure that the recomendations from these reports are appropriately addressed. We expect that this effort will be completed in late 1986 assuming the EV0G ccmpletes their activity in a timely fashion.
With regard to the retrospective nature of the question, we will review staff and licensee actions taken in response to the varicus documer.ts listed in 2a, to assess completeness and acceptability in light of today's kncwledge.
In addition, this review will be perfonned to identify and propose solutions to any program 3 tic deficiencies which may be uncovered. We expect to complcte this effort by September,1986.
. b.
The A-47 review did evaluate the frequency of events with potentially undesirable consequences even though the events were bounded by the FSAR analysis. A review of LERs and Nuclear Power Experience (NPE) publications was conducted to identify important control system failure sequences not detected by the FMEA analysis or the simulation studies perfomed for the reference plant. The operating history period for the LER and NPE review for each NSSS Vendor type ranged between 3 and 10 years. The objective was to identify control system failures that could (a) cause transients or accidents to occur more frequently than the NRC guidelines established in the SRP,(b) detrimentally affect operator actions, (c) result in protection system actuation or cause Technical Specification Safety limits to be exceeded.
These experience reviews were also used to determine if the events found might be a concern on other plants.
Our review did not identify any major control system failure scenarios occurring at an excessively high frequency. Although a number of potentially significant events had occurred, corrective action to improve the reliability and availability of the plants had been taken by the utilities.
The A-47 review was performed on four reference plants, one from each of four NSSS vendors. The B&W reference plant had a model 721 ICS system.
Singic, double and selected multiple independent control failures were studied. Selection criteria to identify control systems whose failure (s) could potentially lead to steam generator or reactor vessel overfill, overcooling, overpressure and/or overheating events were established.
For these evaluations, a minimum number of safety-related systems were assumed operable." The A-47 review also assumed that requirements imposed by the staff through IE Bulletins and Orders, and NUREGs (e.g., NUREG-0800 (SRP), NUREG-0667, NUREG-0737) hid been implemented on the individual plants. Our preliminary assessment of the reference plant evaluation was that the conclusions drawn could be generally applied to B&W plants with both 721 and 820 systems. This conclusion was based on the above stated assumptions.
The recent event at the Babcock and Wilcox Rancho Seco plant consisted of complex multiple failure sequences involving not only control failures but also delayed operator actions and errors, including the valve misalignment which resulted in the failure of the makeup pump. These events were compounded because Rancho Seco had not in fact fully implemented j
the above stated requirements.
It is our belief that if the existing staff requirements and design modifications committed to be implemented by the utility had been fully implemented, the events initiated by control system failures at the Rancho Seco plant would not have been significant.
l l
I i
. We believe that the A-47 analyses on the reference plant can be generally applied to plants with the model 820 ICS system. However, to confirm that control system failure initiated transients are not significantly different than on 721 system plants, we have initiated additional work to review a B&W "820" system design. This additional review will include the effects of loss of restoration of power to the 820 ICS system. Consistent with the coordination of this effort with the B&W Owners Group, we estimate that this additional work will be completed by the end of 1986.
Further expansion of the scope of A-47 does not appear appropriate, since A-47 analyses to date have shown that the consequences of events initiated by control system failures (such as the Rancho Seco event) do not result in core melt or risk significant consequences, and therefore would not support any proposed new generic requirements based on the criteria specified in 10 CFR 50.109 (Backfit Rule).
Independent of A-47, the Division of Human Factors Technology and the Operating Reactors Assessment Staff is conducting a study to develop indicators that describe the safety performance of nuclear plants from commonly available data. The number of reactor trips and ESF actuations resulting from maintenance, operator error, and equipment failures are being evaluated in order to determine the frequency of challenges to the protection systems and to identify major contributors. Also, an independent study is being conducted by MITRE for NRR to assess the significance of challenges to the protection systems from balance of plant systems and to recommend corrective action if deemed to be necessary. We anticipate that these
~
additional ac'tivities will provide more insight on potentially frequent events with undesirable consequences even if the consequences of a particular event are bounded by the FSAR analysis. We believe that these activities should continue separate from USI A-47.
3.
Issue: Adequacy of the design of the integrated control system (ICS).
Action:
l Address the adequacy of the design of the ICS. Particular features i
to be included are:
a.
Whether the ICS is sufficiently reliable to assure that the frequency of unnecessary safety challenges is acceptably low (i.e., is the ICS properly classified as a nonsafety-related system) l b.
Loss of remote (i.e., hand) power coincident.with loss of automatic control.
(Principal Finding #2).
c.
Results of SMUD's analysis to date of the power supply monitor.
(Principal Finding #1).
. d.
Results of SMUD's contractor analysis of the power supply monitor.
(Principal Finding #1).
e.
Role of the power supply monitor as a potential single failure in the ICS and/or NNI.
(Principal Finding #1).
f.
Results of a study of the response of the ICS upon restoration of power.
(Other Finding #3).
g.
Acceptability of the failure mode of meters and recorders that are affected by a loss of ICS power.
(Other Finding #4).
Status:
As noted in response to issue 2a, the BWOG will be performing a comprehensive review of the ICS/NNI. Thus, we expect that many of the specific areas identified above will be addressed. However, to assure that these items are considered, we will be providing the above list of specific concerns to the BWOG as part of the NRC comments on the BWOG plan.
4.
Issue: Adequacy of the maintenance program for manual isolation valves.
Action:
b.
Evaluate the adequacy of industry standards and NRC requirements regarding periodic testing and maintenance of manual valves in safety-related systems (Principal Finding #3).
Status:
The industry does not have a regular maintenance program that applies to every manual valve. The NRC does not have a requirement for maintenance and testing of convenience valves such as the locked open manual valve discussed in NUREG-1195.
Section XI of the ASME Boiler and Pressure Vessel Code specifies inservice inspection, testing, repair, and replacement of valves that are components in systems classified ASME Class 1, 2, or 3.
In a nuclear power plant there are many components, including manual valves, that are not classified into any of the above categories and, consequently, not subject to the requirements of Section XI.
i
. The NRC requirements for valve testing are contained in 10 CFR 50.55(a)(g) which incorporatesSection XI of the ASME Boiler and Pressure Vessel Code.
Therefore, regulatory requirements for valve testing extend only to valves that are within the inservice testing program.
The NRR evaluation of licensees' maintenance programs does not consider valve classification.
It is the licensee's responsibility to perform the testing, repair, and maintenance on the valves that are within their inservice testing and maintenance programs.
The staff is preparing a memo to IE to request the issuance of an IE Information Notice regarding the lack of a regular industry maintenance program that applies to manual valves in safety systems.
In addition, the staff plans to prioritize this issue in accordance with NRR Office letter No.
- 40. We expect the prioritization of this issue to be completed by September 19,86.
5.
Issue: Adequacy of procedures and training.
Action Evaluate the adequacy of the procedures and operator (licensed and nonlicensed) training, particularly with regard to:
a.
The degree to which event specific procedures (e.g., loss of ICS, station blackout) are needed to quickly recover from events that have been diagnosed by the operators, and to mitigate such events if the initiating condition cannot be immediately corrected.
(Principal Findings #4 and #7).
c.
The consistency between E0Ps and NRC approved procedure generation packages.
(Note. The ED0 assigned responsibility for this item to IE. NRR also has input to this issue which is provided below).
d.
The adequacy of procedural guidance concerning:
(1) when to trip
- auxiliary feedwater pumps and (2) the relative priorities of avoiding the PTS region and maintaining pressurizer level.
(Principal Findings #5 and 6).
Status:
a.
Finding No. 4 notes that the generic procedure guidelines supplied by the Owners' Group do " include an explicit procedure..." for the i
event that occurred, but that the plant failed to incorporate this procedure in their plant-specific Emergency Operating Procedures.
Finding No. 7 goes to the inadequacy of plant-specific training.
Therefore, NUREG-1195 supports a concern with the plant-specific adaptation of generic procedure guidelines and with the cor-responding operator training, but does not suggest a generic issue nor a need for mcre event-specific procedures.
. The function-based procedures called for in the TMI Action Plan and Supplement 1 to NUREG-0737 are intended to enable operators to identify and take appropriate steps without first diagnosing the cause of the event, lest an erroneous diagnosis lead to a pattern of systematic operator error. However, both the industry and the NRC staff have been careful to avoid procedures that lead to round-about approaches to readily identifiable faults. Most of the generic procedure guidelines lead to event - or problem-specific procedures quite directly.
Some open issues remain in the staff review and approval of generic procedure guidelines where this is not the case. Therefore, the staff sees no basis for the generic concern expressed in Subissue 5.a.
Also see Subissue 5.c.
The staff is proceeding with a plant-specific resolution of the E0P and training problems at Rancho Seco prior to restart, and is continuing to work with Owner's Groups to close open items in the review and approval of generic procedure guidelines.
c.
Audits comparing plant-specific emergency operating procedures with the NRC reviewed and approved plant-specific commitments for the creation, validation, training, and maintenance of emergency operating procedures, contained in Procedures Generation Packages (PGP) do suggest a generic problem. The findings of the inquiry into the Rancho Seco Incident reinforce the conclusion that many licensees are.not following their own PGPs.
The staff is taking three actions to respond to this evidence of a generic problem:
1.
An IE Notice is being prepared to alert the industry to the problem and to the other staff actions. A draft is under concurrence review at this writing.
2.
NRR is continuing with an expedited program to perform intensive audits of E0Ps at ten selected plants to better define the extent and severity of the problem. Three audits have been completed, one is scheduled for May 1986, and the remaining six will be performed at the rate of one every 21/2 months.
3.
A Temporary Instruction is being prepared jointly by NRR and OIE to guide a one-time inspection of E0Ps at all operating plants.
A draft is being evaluated and pilot-tested in the Regions now, d.
As part of the BWOG program, the Owners' Group intends to perform a review of the procedures. At this time, the detailed scope of the progran has not been provided to the staff. We will therefore identify this item as a specific NRC concern as part of our response to the BWOG program plan presented on April 8, 1986. We expect to complete our review, assuming timely input from the BWOG, by the end of 1986.
m
7.
Issue: Adequacy of the FSAR accident analyses Action:
Verify the adequacy of the Rancho Seco FSAR accident analysis, particularly the degree to which credit is given for the non-safety related ICS and the nonsafety-related Main Steamline Failure Logic.
(Section 9) and (Principal Finding #14).
Status:
Based upon the description of the issue, we believe this is a plant specific issue rather than a B&W Generic Review effort. The licensee will submit a recovery plan which the staff will review prior to restart to determine if more needs to be done. Therefore, this issue will be resolved as part of the Rancho Seco restart review.
8.
Issue: Adequacy of Required Staffing Action:
Evaluate the adequacy of plant staffing to deal with expected operational transients.
(Other Finding #7).
4 Status:
As noted in response to issue 5, the Owners' Group will be performir,g a review of plant procedures.
In addition, as part of their information gathering effort the Owners' Group will be interviewing plant operators.
i Since the detailed BWOG program has not been provided to the staff, we are unable to determine whether this issue will be addressed. Therefore, we will identify, as part of our comments on the BWOG plan, this issue.as an NRC concern which should be included as part of their overall evaluation.
We expect to complete this task as part of our overall reassessment program, by the end of 1986.
l NRR DRAFT ACTION PLAN RANCHO SECO EVENT I.
PLANT SPECIFIC ACTIONS Lead Item Responsibility Date Status / Comments 1.
Transmit Initial Staff PD#6/DPL-B 2/10/86 Complete concerns to SMUD.
Provided at 2/10/86 meeting with licensee, NRR and Region V.
2.
SMUD's submittals regarding N/A 2/19/86 Initial staff the 12/26/85 event.
review completed.
Hold for staff review of forthcoming SMUD Recovery Plan (Item 8) 3.
Transmit NUREG-1195 to E00 2/21/86 Complete' SMUD for informatior, and coment.
4.
Respond to SECY's March 6, PD#6/DPL-B 3/21/86 Complete 1986 memo (partial response see item 11).
5.
Site meeting with SMUD N/A 3/24-25/86 Complete by Directors, NRR, 0IE and Regional Administrator, Region V.
6.
Respond to ED0 March 13, PD#6/DPL-B 4/14/86 Complete 1986 memo.
7.
Meeting with licensee 4/18/86 Complete to discuss their recovery plan outline.
8.
Receipt of Licensee's N/A Ea rly Basad on i
Recovery Plan July 1986 licensee's current estimate l
J
4
) Lead Item Responsibility Date Status /Coments 9.
Develop NRR review plan, PD#6/DPL-B One week review assignments, and after schedules.
receipt of recovery
{
plan.
- 10. Transmit requests for PD#6/DPL-B One month additional information from Item 8 to SMUD.
- 11. Meeting with licensee to PD#6/DPL-B Five weeks review outstanding issues.
from Item 8.
- 12. Commission meeting.
PDf6/DPL-B TBD i
PD#6/DPL-B TBD
- 14. Restart readiness review.
Region V TBD I
~
l i
f i
i 4
[
i i
l i
i t
,