IR 05000454/1986029

Safety Insp Rept 50-454/86-29 on 860721-0808.Violations Noted:Installation of Defective Component as Fission Product Barrier & Failure to Control Identified Nonconforming Matl to Prevent Inadvertent Use

ML20206N974
Person / Time
Site:	Byron
Issue date:	08/14/1986
From:	Forney W NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
Shared Package
ML20206N967	List: IR 05000454/1986029 ML20206N963
References
50-454-86-29, NUDOCS 8608270044
Download: ML20206N974 (10)

Text

_ _ .. . . ._ _ _ _ ___

'

.

I U.S. NUCLEAR REGULATORY COMMISSION

'

REGION III

i

Report No. 50-454/86029(DRP)

Docket No. 50-454 License No. NPF-37

, Licensee: Commonwealth Edison Company Post Office Box 767 Chicago, IL 60690

.

Facility Name: Byron Station, Unit 1 Inspection At: Byron Station, Byron, IL Inspection Conducted: July 21 through August 8, 1986

..

.

Inspectors: P. G. Brochman j J. M. Hinds, Jr.

WhL W. L. Forney % ief ' -

Approved By: F//////6 Reactor Projects Section 1A Date'

Inspection Summary i Inspection on July 21 through August 8, 1986 (Report No. 50-454/86029(DRP))  !

Areas Inspected: Special, unannounced rafety inspection by the resident i inspectors to review the circumstances 1.urrounding the installation of a nonconforming pressurizer code safety vtiv i

Results

One apparent violation of NRC requirements was identified. This i apparent violation was the aggregate of three problems: (1) installation 1 of a defective component as fission product barrier (Reactor Coolant System Pressure boundary); (2) failure to control identified nonconforming material-to prevent its inadvertent use; and (3) a failure in management oversight of i the implementation of the Quality Assurance and Maintenance program In i total, these problems constituted a potential threat to the public's health  :

,

and safet '

i i l

l

,

)

I l

!

i

!

.

"

8608270044 860818 PDR ADOCK 05000454 F G PDR

.-- ,A . - - - . - - - - . , , . , . . . , , .,, ,.. , , . _ - ~, _ - - - , ,

-

. . - . - . - . . . _

.

, .

,

>

DETAILS

'

1. Persons Contacted Commonwealth Edison Company

• R. Querio, Station Manager ,

• R. Pleniewicz, Production Superintendent

• R. Ward, Services Superintendent

• L. Sues, Assistant Superintendent, Operating

• Schwartz, Assistant Superintendent, Maintenance

• T. Joyce, Assistant Superintendent, Technical Services .

T. Tulon, Operating Engineer, Unit 1

• A. Chernick, Regulatory Assurance Supervisor

• F. Hornbeak, Technical Staff Supervisor

. R. Williams, Technical Staff

• E. Zittle, Regulatory Assurance Staff

'

The inspector also contacted and interviewed other licensee and contractor personnel during the course of this inspectio * Denotes those present during the exit interview on August 8,198 . Purpose This inspection was conducted to review the circumstances surrounding the

'

i installation of a nonconforming valve in Byron Unit 1 as Pressurizer Code i

Safety Valve IC (IRY8010C); and to review the licensee's investigation and corrective action for this even . Description of the Event

'

On July 2,1986 Byron Unit I was shutdown due to excessive Reactor Coolant System (RCS) leakage. That event is discussed in Inspection Report No. 454/86025(DRP). On July 4, 1986 the unit was placed in Cold Shutdown (Mode 5). The leakage was identified as coming from a valve which was subsequently replaced. Additional work was performed on the RCS during this outage, including replacement of the IC Pressurizer Code Safety Valve IRY8010C, which was believed to be leaking approximately 0.5 gp On July 17, 1986 heatup of the RCS to normal operating conditions was commenced. At 0420 on July 18, IS86, upon reaching 1750 psig in Hot Standby (Mode 3), 1RY8010C actuate The normal lift setpoint for this valve is defined in Technical Specification 3.4.2.2 as 2485 i 1% psi The closed indicator light extinguished and an immediate 50 psig pressure

drop in pressurizer pressure was observed, as was an increase in the Pressurizer Relief Tank level, pressure, and temperature. However, the open indicator light for 1RY8010C did not illuminate nor did the

" Pressurizer Safety Valve Actuated" annunciator alarm. The valve reseated but began to leak excessively. Operations department personnel J

.

, - , - ,-r- .- --- -- ,, - e ,

- -

-

\

!

'

.

entered containment and determined that both valves 1RY8010C and 1RY8010A l were leaking. At 0700 on July 18, 1986 1RY8010C was declared inoperable '

and a cooldown to Mode 5 was commenced. The unit entered Hot Shutdown I (Mode 4) at 1238 and entered Mode 5 at 1630. Valve IRY8010C was removed and inspected. The licensee's inspection revealed that the valve's disk insert was missin By July, 23, 1986 both 1RYS010A and 1RY8010C had been replaced and the unit was returned to powe The licensee began an immediate investiga-tion of this event on July 18, 198 . Chronology of Events The licensee's investigation developed the following chronology of events:

July 1985 Twa Pressurizer Code Safety Valves (IRY80108 and IRY8010C) were replaced due to excessive seat leakage. These two valves were placed in the Hot Shop. The Hot Shop is an area in the Auxiliary Building where contaminated components may be decontaminated, repaired, rebuilt and tested. The serial nuabers of the two valves and their associated Limit Indicators Switch Assemblies (LISA) were as follows:

Valve No. N56964-00-0031/LISA No. 65322-00-0023 Valve No. N56964-00-0032/LISA No. 65322-00-0024 Both valves had QA hold tags hang on them. However, the QA hold tags listed only the LISA serial numbers, not the valve serial numbers. The valves were stored in a common roped off area, of the Hot Sho September 1985 Nuclear Work Requests (NWR) were written to repair the two valves, however, the NWRs and their attached Test Report Packages only listed the LISA serial numbers, not the serial numbers of the valves themselve October 1985 Work commenced on the valves per the NWRs. The Mechanical Maintenance (MM) technician performing the work identified a discrepancy between serial numbers on the valves and the serial numbers listed on the NWRs and brought this to his supervisor's attentio The matter was resolved verbally and no changes were made to the test report packages or NWRs, because the supervisor believed that the LISA serial number was a unique identifier and was sdequate to identify the valv October 7, 1985 The MM placed valve No. 32 on the test stand in the Hot Shop and removed the LISA. During testing the LISA'is required to be removed from the valve. The

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

.

.

licensee believes that only one valve was disassembled and only one LISA was removed at a time, while both valves were in the Hot Shop. Valve No. 32 failed its leak test, was removed from the test stand, and returned to its position on the Hot Shop floo October 8, 1985 The MM placed valve No. 31 on the test stand in the Hot Shop and removed the LISA. This valve also failed its leak tes There was some confusion on the part of the MM as to how to properly annotate the test report package to indicate the test failure and provide for the required maintenance. Valve No. 31 was left on tFe test stan October 15, 1935 The problem with how to indicate the test failure on the test package remained until October 15, 1985 when a new supervisor was assigned and directions for the valve repair were provided in the NWRs and test package October 16, 1985 The technical representative from the valve manufacturer (Crosby) arrived onsite to repair the valves. Valve No. 31 which had remained on the test stand since October 8, was disassembled and the disk insert was discovered to be badly steam cu The representative ordered new parts and reassembled valve No. 31, with the disk insert removed. Valve No. 31 was removed from the test stand and valve No. 32 was installed. The NWR for valve No. 31 should have been segregated (pigeon-holed) with other NWRs for components which were awaiting parts. However, the NWR for valve No. 32 was " pigeon-holed" in erro October 18, 1985 Valve No. 32 vf; repaired and satisfactorily teste However, the '-st Report Package for the incomplete valve No. 31 w - 5 erroneously obtained and incorrectly annotated to indicate that successful testing had been completed on valve No. 31. The MM, his superviser, and 1.he QC int pector all failed to verify the serial numb'r on the test package against that of the valve beir; teste The test package was not actually taken into the Hot Shop, but remained outside. This was done because of a concern over contamination of the NWR and test package, due to high contamination levels in LPe Hot Shop. The valve No. 32 was then '

returred to the Hot Shop floor, alongside valve No. 31. The valves remained in the Hot Shop Hold area until July 1986 because there were no storage facilities for repaired contaminated part ,

. -- . .-

.

.

July 6, 1986 The NWR for valve No. 31 was obtained and indicated that the valve was ready for use. Valve No. 31 was then installed on the pressurizer as valve IRY8010C, without a disk inser July 17, 1986 The unit was heated up, and entered Mode 3 at 193 July 18, 1986 At 0420, at 1750 psig, 1RY8010C actuated. An immediate 50 psig pressure decrease was observed, as were high temperature, level, and pressure indications on the Pressurizer Relief Tank. The closed indicating light for 1RY8010C was observed to extinguish, but the open indicating light did not illuminate. The " Pressurizer Safety Valve Actuated" alarm did not annunciat Operators entered containment and subsequently identified that IRY8010C and 1RY8010A were both leaking. At 0700 1RY8010C was declared inoperable and a cooldown was commenced. Mode 4 was entered at 1238. The cooldown was continued to Mode 5 and both valves were removed. The licensee's inspection of 1 the valve installed at 1RY8010C indicated that its disk insert was missing. Additionally, the RTDs for

the temperature indicators downstream' of each safety valve were inspected and it was discovered that the indicators for the A and C valves were switche Therefore, the wrong (non-leaking) valve had been removed originally. The RTD miswiring was believed

,

by the licensee to have occurred during original

construction.

July 19,1986 Two new valves were tested and installed as 1RY8010A and IRY8010 July 20, 1986 A Technical Staff engineer was reviewing the test packages for the valves which were installed on the 19th and discovered that the setpoint tolerance was wrong. Technical Specification 3.4.2.2 specifies a

setpoint of 2485 i 1% psig for the pressurizer code safety valve The valves were set using a generic maintenance procedure which specified a tolerance of i 3%, per the manufacture's instructions. A post calibration check of the test gage indicated that it read five psig high and the adjusted lift pressure for 1RY8010A was within the 1% tolerance rang However, the adjusted lift pressure for 1RY8010C was not within the 1% tolerance range. Since IRY8010C was not set ia accordance with the limits of Technical Specification 3.4.2.2, it was removed from the pressurizer, taken to the Hot Shop, reset,

retested, and reinstalled on the pressurize !

.

.

July 23, 1986 Folicwing completion of these repairs the unit was heated up to normal operating conditions, taken critical and returned to rated powe . Evaluation of the Event This event can be described as the aggregate of three problems:

(1) failure to control nonconforming material to prevent its installation and use in the reactor plant; (2) installation of a defective component as a Fission Product Barrier / Reactor Coolant System Pressure Boundary; and (3) failures in management of the implementation of the Quality Assurance and Maintenance program These three problems characterize the aggregate of five apparent violations of NRC requirements as follows: CFR 50, Appendix B,Section XV, as implemented by the Commonwealth Edison Quality Assurance Manual, Quality Requirement 15.0, requires that items or conditions involving maintenance and modifications which are found nonconforming to engineering requirements or specifications will be controlled to prevent their inadvertent use or installatio Nonconforming items are to be identified, documented and segregated for disposition. The failure to maintain control over a ncnconforming pressurizer code safety valve, number N56964-00-0031, and prevent its inadvertent installation and use in the plant as a Fission Product Barrier / Reactor Coolant System pressure boundary is an apparent violation of 10 CFR 50, Appendix B, Section XV (454/86029-01a(DRP)). Technical Specification 3.0.4 requires that entry into an Operational Mode shall not be made unless the conditions for the Limiting Condition for Operation are met without regard to the Action Statements. Technical Specification 3.4.2.2 requires that three Pressurizer Code Safety Valves shall be operable in Modes 1, 2, and 3 with a lift setpoint of 2485 i 1% psig. A Pressurizer Code Safety Valve is not operable with its disk insert removed. The failure to have three operable Pressurizer Code Safety Valves upon entry into Mode 3 is an apparent violation of. Technical Specification 3. (454/86029-01b(DRP)). Technical Specification 4.0.5.a requires that inservice testing of ASME Code Class 1 valves shall be performed in accordance with Section XI of the ASME (American Society of Mechanical Engineers)

Boiler and Pressure Vessel (B&PV) Code and applicable addenda as required by 10 CFR 50.55a(g). The licensee has committed to the 1980 Edition, Winter 1981 Addenda of the ASME B&PV Code per their inservice inspection program. The ASME B&PV Code, Winter 1981 Addenda,Section XI, Division 1, Subsection IWV-3200 requires that when a valve has been replaced or repaired or has undergone 1 maintenance that could affect its performance it shall be tested to demonstrate that the performance parameters which could be affected 1 by the replacement, repair or maintenance are within acceptable l limits. The ASME B&PV Code, Winter 1981 Addenda,Section XI, Division 1, Subsection IWV-3512, " Test Procedures for Category C

l l

,

.

.

Valves," requires that safety valve set points shall be teste Technical Specification 3.4.2.2 requires that an Operable Pressurizer Code Safety Valve shall have a lift setpoint of 2485 1% psi Pressurizer code safety valve number N56964-00-0031 was installed as IRY8010C on July 6, 1986 without the required tests having been performed. The failure to test pressurizer code safety valve number N56964-00-0031 and verify its lift setpoint was within 2485 i 1% psig is an apparent violation of Technical Specification 4.0.5 and the ASME B&PV Code, Winter 1981 Addenda,Section XI, Division 1, Subsection IWV-3200 (454/86029-01c(DRP)). CFR 50, Appendix B,Section V, requires that activities affecting quality shall be prescribed by documented instructions, procedures, or drawings of a type appropriate to the circumstances and shall be accomplished in accordance with these instructions, procedures, or drawings. Technical Specification 3.4.2.2 requires that the setpoint of the pressurizer code safety valves be 2485 t 1% psi The setting of the pressurizer code safety valve's lift setpoint is an activity affecting qualit The failure of maintenance personnel to specify the appropriate tolerance in the work instructions for 1RY8010C is an apparent violation of 10 CFR 50, Appendix B, Section V (454/86029-01d(DRP)). CFR 50, Appendix 8,Section V, requires that activities affecting quality shall be prescribed by documented instructions, procedures, or drawings of a type appropriate to the circumstances and shall be accomplished in accordance with these instructions, procedures, or drawing Control and Instrumentation Diagram (C&ID) M-2060, sheet 9, Revision G, indicates that the Resistance Temperature Detector (RTD; for Temperature Indicator ITI-464 should be installed downstrean of 1RY8010C and that the RTD for Temperature Indicator ITI-466 should be installed downstrean of IRY8010A. The installation of the Pressurizer Safety Valve temperature instruments in accordance with approved drawings is an activity affecting quality. The installation of the RTD for ITI-464 downstream of 1RY8010A and

, the RTD for ITI-466 downstream of 1RY8010C is an apparent violation of 10 CFR 50, Appendix B, Section V (454/86028-01e(DRP)).

6. Corrective Actions Initiated by the Licensee Short Term The licensee has established a task force to review the entire maintenance program, including the QA aspects and to develop a Conduct of Maintenance Improvement Program (CMIP). This program, when developed, will have specific goals and timetables for the accomplishment of those goals. The Institute of Nuclear Power Operations (INPO) has been requested by the licensee to conduct an assistance visit to review the aspects of the maintenance program related to this event and to recommend improvement item _ _ _ . _

O

~

i Improvement items recommended by INP0 will be reviewed by .the licensee and may be inccrporated into the CMIP. The licensee has begun building contaminated storage facilities for repaired components and is improving the. housekeeping and segregation of materials in the Hot Sho Long Term The licensee will implement the CMIP and track the accomplishment of these goals. Management involvement will be increased in the maintenance are . Conclusions This event is of serious safety significance and is indicative of programatic failures of personnel to accomplish the maintenance and

quality assurance programs as they are written. As the age of Byron Station increases, more and more reactor plant components will require repair or replacement. These repaired / rebuilt components must bc

, maintained at the same high quality levels as new components to ensure l the safe operation of this plan The consequences of this valve failing can be limited to two possibil-ities, the valve fails to open within the required pressure range which results in the RCS being overpressurized or the valve fails to reseat which results in a Loss of Coolant Accident (LOCA). Due to the design of the valve, it is not considered likely that normal operating pressure could be reached. Therefore, the overpressure scenario is unlikel The possibility of the valve failing to reseat is of a much higher probability. Again however, due to the valve design it is unlikely that normal operating pressure could be reached. Therefore, a LOCA would not occur in Power Operation (Mode 1), but in Mode 3, with consequently less severe effects on the reactor core. Nevertheless, a LOCA in any mode is

a severe accident.

'

During this inspection the following management failures and weaknesses were identified. Some of these failures / weaknesses were direct contributors to this event; others were not, but they have been

, identified for their potential to cause similar problems:

I The failure to specify the valve serial number on the hold tag and

'

the NWR, rather than the LISA serial number, was not a direct contributor to this event. However, it is indicative of a failure

of the Quality Assurance program to properly identify nonconforming material and of the Maintenance program to properly identify which

.

j l reactor plant components were worked on.

I The failure to identify all subcomponents, when they are l disassembled, of a nonconforming component as being nonconforming is a programatic weakness in the implementation of the Quality

'

j Assurance program. During tours of the Hot Shop and general maintenance shop the inspector observed two safety related

,

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

.

.

centrifugal pumps which had been disassembled and were lying in a roped off area, Labeled as a QA hold area', and not all the

components were QA hold tagged. This is considered a management

weakness in implementing the Quality Assurance and Maintenance

, programs.

!

This problem is exacerbated by the lack of permanent physical

^

i barriers to segregate different nonconforming components. The i

inspectors recognize that it may not be possible or practical in

all cases to utilize permanent barriers, but believe that they may l be used effectively in some application , The poor radiolngical controls and housekeeping practices performed

'

in the Hot Shop were a direct contributor to this event. The i

conflict between the desire to keep work packages uncontaminated, I which resulted in their remaining outside the work area, and the 1 desire to minimize the generation of radioactive waste material by

'

not making copies of important documents, for example, was created by poor radiological practices. During a tour of the Hot Shop the inspector identified numerous contaminated components which were not properly bagged or for which the poly bagging was torn. Additionally, the loose poly bags were piled two to three feet high in some areas of the Hot Shop. The identification of the QA hold area was very poor. The only sign identifying the test stand as a QA hold area

'

, was face down on the floor and partially obscured by a piece of poly, t The lack of appropriate, controlled storage facilities for repaired

, or rebuilt components which are also contaminated is a management failure which directly contributed to this even The failure to provide appropriate procedures and training to  !

maintenance personnel on how to annotate test report packages for a failed test is a maintenance progra.n weakness which did not directly contribute to this even ' The failure to expeditiously close out completed NWRs is a management weakness which directly contributed to this event.

l The licensee's interpretation of the ASME B&PV code to allow a i safety valve to be tested satisfactorily, placed in stores for up to 18 months, and then installed in the reactor plant without any further testing may meet the letter of the code. However, the inspectors believe that the intent of Subsection IWV-3200 is to require the testing of safety valves a reasonable amount of time before installation. This aspect of the Maintenance program was a i direct contributor to the even ,

, The failure to provide approved written procedures which specified the tolerance values for components for which Technical Spec.ifications were more restrictive than industry codes or vendor instructions is a management weakness. This activity should not be left to the discretion of the work analyst and was a direct contributor to this event.

,

9

, _ _ _ - -

_ _

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

- - -

. . . .. .

.

. .

'

.

9 The failure to control and-segregate the NWRs for those components which are awaiting parts is a management weakness which directly contributed to this even . Exit Interview (.?0703)

The inspectors met with licensee representatives denoted in Paragraph 1 at the conclusion of the inspection on August 8, 1986. The inspectors summarized the purpose and scope of the inspection and the finding The inspectors also discussed the likely informational content of the inspection' report with regard to documents or processes reviewed by the inspectors during the inspection. The licensee did not identify any such documents / processes as proprietar i I

.

,

10