ML20149J486
| ML20149J486 | |
| Person / Time | |
|---|---|
| Site: | Trojan File:Portland General Electric icon.png |
| Issue date: | 02/11/1988 |
| From: | Cockfield D PORTLAND GENERAL ELECTRIC CO. |
| To: | Martin J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| References | |
| NUDOCS 8802230079 | |
| Download: ML20149J486 (6) | |
Text
.
p: emq m
assume r c, it I_
!!!!E C
r3 FC0 i h APB David W. Cockfield Vice President, Nuclear February U.., 1988 Trojan Nuclear Plant Docket 50-344 License NPF-1 Mr. John B. Martin Regional Administrator, Region V U.S. Nuclear Regulatory Counissica 1450 Maria Lane, Suite 210 Walnut Creek CA 94596-5368
Dear Mr. Martin:
t Pesponse to Request for Information Regarding Design Basis Documents ard the December 1987 Turbine Runback Mvant Your letter of December 22, 1987 requested a description of our plans for prioritizing and resolving open items identified during the development of Design Basis Documents. Attachment 1 discusses our process for handling these open items. The appropriate resolution of Design Basis Document open items is a high priority objective for PGE.
Your letter also r'equested our evaluation of the turbine runback event at Trojan on December 6,1987 and corrective actl.ons rer;arding maintenance activities. Attachment 2 provides this inforn tion. Ac discussed in, an action list ia; been developed from oue critique of this event. Thes. actions will receive significent mat amant attention to ensure appropriate completion.
Sincerely, Attachments c:
Document Control Desk Nuclew Re 'Jiatory Commission U.S.
a Mr. William Dixon State of Oregon Department of Energy Mr. R. C. Barr NRC Resident Inspector Trojan Nuclaar Plant 8802230077 880211 PDR ADOLK 05000344
$ gJ P
\\ \\
e sw saw:n saa rmo ov;,:n 972c >,
^
i Trojan Nuclear Plant Mr. John B. Martin Docket 50-344 License NPF-1 February 11, 1988 Page 1 of 2 DESIGN BASI 3 DOCUMENT OPEN ITEM ACTION PLANS Design Basis Document (DBD) open it3ms will be processed in accordance with Nuclear Division Procedure (NDP) 200-6, "Design Basis Documents".
During the preparation of each DBD, a list of the design basis open items is developed. This list of open items is included in the DBD so that the user of the document can determine which information is not known or has not 'oeen resoir d.
These lists are reviewed with the cognizant engineer-ing personnel to prioritize the open items, to assign an action party, to assign a due date, and to determine what information is required.
The action items are assigned to one of four different categories.
The first category is those open items which represent significant deficien-cle.s in design basis information, They are given the highest priority and generally have action duc dates of about six months or less. The second category includes the open items which constitute information which is useful to the design engineers and the operating personnel but which 10 not necessary in the short-term to answer questions about the design of the existing system; they are classified in a second lower priority. These items are assigned action due dates of approximately one year for completion.
The third category is those open items which, in the judgement in the group reviewing the open items, do not need to be resolved at the present time and will be pursued when a design change would occur which would require the information to be developed.
An example of this type of information woul6 be the calculations which show the design margin for vendor-supplied heat exchangers.
Some DBD open items have not been related to design bases or can be readily answered with the information available.
In thic case, these itens have been resolved and closed.
The fourth category involves thoso cases where a significant deficiency is uncovered during the development of a DBD which may raise questions regarding the operability of a system.
In this case, an Event Report in accordance with NDP 600-3, "Event Reports", is initiated to obtain an engineering evaluation of the deficiencies.
To date, two Event Reports have been initiated because of questions of operability of the design of certain systems.
One Event Report evaluated the operability of the Containment Spray System on a single failure of a sodium hydroxide tank suction valve to shut after educting sodium hydroxide into the system.
The other report reanalyzed the operability of the Auxiliary Feedwater System in the event of u umall line break where the individual line high flow isolation may not occur.
In both cases, the existing design wa.
determined to be adequate, and additional documentation has been developed for the DBDs.
o, s
Trojan Nuclear Plant Mr. John B. Martin Docket 50-344 License NPF-1 February 11, 1988 Page 2 of 2 After all the design basis open items for a system have been classified and prioritized, the DBD Project Manager publishes an action plan for the resolution of the open items. These action plans are reviewed and approved by the General Manager of Technical Functions. The DBD Project Manager tracks the resolution of open items.
Periodic updates to the open item action plans are published so that management can monitor the progress being made to resolve the open items.
The DBD will be revised as appropriate to include the new information.
This process is bains used to addreso the open items which have resulted from the development of DBDs for the first 13 systems selected and will continue to be used for subsequent systems.
DLN/kal 2288P.288
a e
Trojan Nuclear Plant Mr. John B. Martin
-Docket 50-344 License NPF-1 February 11, 1988 Page 1 of 3 TURBINE RUNBACK EVENT EVALUATION AND CORRECTIVE ACTION PLAN, I
(ht. December 6,1987 Trojan experienced a continuous turbine runback due to the failure of a turbine electrohydraulic control (EHC) switch.
Plant Operations personnel initiated a manual turbine and reactor trip as a result of this event.
Subsequent investigation revealed the causn of the runback was a load decrease push button switch that had failed in _ s decrease position.
The problems that occurred during and following this event have been carefully reviewed and an action plan was developed to correct the weak-nesses that were identified. The details of this plan were discussed with members of the NRC Region V staff in meetings on December 9, 1987 and January 4,1988.
One of the major concerns arising from our review l
of this event was the conduct and management control of the repairs that were made to the EHC system. Several weaknesses in the control of
' maintenance activities and in procedural compliance were apparent.
t Followice the trip, Plant Maintenance personnel were called to the Plant to invesiagate the problem that caused the runback.
Also onsite were the l
Duty Plant General Manager, the Duty Technical Services Manager, the Maintenance Supervisor, the. Instrumentation and Control (I&C) Supervisor, and two system engineers who are knowledgeable in the electrical and I&C areas.
Possible troubleshooting techniques were discussed by these personnel, however, no formal work instructions or Maintenance Requests were issued to document and control these activities. At that time, this practice was consistent with our procedure (Administrative Order 3-9) which allows activities to proceed on "urgent" work items prior to the initiation of a Maintenance Request.
Our critique of the event indicated that better control over "urgent" work was necessary, and we have since limited work that can be performed on an "urgent" basis prior to issuing a Maintenance Request.
Now, only "urgent" work may proceed on problems that directly affect reactor safety or pose an imminent danger to the public or Plant personnel. Work items that affect power production, reliability or other areas will not ba classified as "urgent" without approval of the Duty Plant General Manager and written work instructions will be required before proceeding.
In either case, the work performed will be fully documented.
The troubleshooting performed on December 6 determined that the cause of the problem was a load decrease push button switch that had failed in the closed position. A spare push button switch was not available so a decision was made to use an installed push button that does not serve a function in the EHC system. The need for a temporary modification was identified and the system engineer prepared the modification with the assistance of I&C Department personnel. The modification was installed e
s Trojan Nuclear Plant Mr. John B. Martin Docket 50-344 License NPF-1 February 11, 1988 pago 2 of 3 on the af ternoon of December 6,1987.
The push button switch was removed and was installed in place of the failed load decrease push button switch.
Basic installation cheeks to verify continuity were performed, with fur-ther functional testing to be done when the turbine was synchronized to the grid.
The Plant was ready to return to power operation on the evening of December 7, 1987.
Engineering and Maintenance personnel were onsite to observe the operation of the turbine controls after the generator was placed on tha grid. Upon closing the generator output breaker, the generator failed to pick up electrical load and the output breaker was reopened. Troubleshooting was again performed under an "urgent" priority without a written work plan or work instructions.
It was found that the load increase push button switch was not performing its function. The cause of this problem was determined to be an improperly installed temporary modification. When the push button switch was removed on December 6, 1987 to replace the load decrease push button switch, the I&C personnel determined that the temporary modification did not specify what was to be done with the unused leads that were formerly connected to the switch. The I&C Supervisor instructed the technicians to leave two cf the leads disconnected and taped and terminate a third lead at a common point in the circuit. He noted this change on the Temporary Modification record but failed to obtain the proper approvalc from the engineer who initiated the modification. Although not readily apparent froe system drawings, this termination cauced the load increase push button switch to be inoperable.
The cause of this problem is the failure of the Supervisor to follow the temporary modification instructions.
A change to the instructions was made without proper engineering approvals.
As a result of this problem, a hold was placed on startup by PGE Management until an independent investigation could be conducted and the results reviewed.
The subsequent independent investigation confirmed the problem with pro-cedure compliance, control of troubleshooting activities, and temporary modification activities.
It also identified another procedure compliance violation in that an independent technical evaluation was not performed when the temporary modification was revised as required. The results of the report were presented during the December 9 meeting.
A significant action plan was developed as a result of this investiga-tion.
Both long-and short-tere. actions were identified. The short-term acticas were completed prior to the Plant restart on December 24, 1987.
Longer term actions are underway and will continue to receive a high level of emphasis until completed. A summary of these actions include:
1.
Develop guidelines for urgent Maintenance Requests approvals.
(Completed on December 12 1987.)
i i
- e
,e
,i l
Trojan Nuclear Plant Mr. John B. Martin Docket 50-344 License NPF-1 February 11, 1988 Page 3 of 3 2.
Independent review of management controls and oversight.
3.
Develop a specific training program for all personnel on procedural compliance.
(Completed on December 15, 1987.)
4.
Conduct procedure compliance training for Plant Managers and Supervisors.
(Completed by December 20, 1987.)
5.
Develop maintenance troubleshooting guidelines.
6.
Develop a ready-for-startup procedure to address Plant trip and forced outages.
7.
Revise instructions for temporary modifications.
8.
Review post-maintenance testing guidelines for adequacy.
9.
Review / improve communications with the NRC.
10.
Conduct a QA review of past temporary modifications for procedure compliance.
(Completed on December 15, 1987.)
11.
Develop a policy letter which specifically defines system engineer responsibilities.
(Completed on January 19, 1988.)
DLN/kal 2288P.288
-,