IR 05000155/1988011
| ML20150B385 | |
| Person / Time | |
|---|---|
| Site: | Big Rock Point File:Consumers Energy icon.png |
| Issue date: | 06/28/1988 |
| From: | Danielson D, Huber M, James Smith NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20150B377 | List: |
| References | |
| 50-155-88-11, IEB-85-003, IEB-85-3, NUDOCS 8807120008 | |
| Download: ML20150B385 (13) | |
Text
o
'.. '
'
'
..
.
U.S. NUCLEAR REGULATORY COMMISSION
REGION III
Report No. 50-155/88011(ORS)
Docket No. 50-155 License No. OPR-06 Licensee:
Consumers Power Company-212 West Michigan Avenue Jackson, MI 49201 Facility Name:
Big Rock Point Nuclear Plant Inspection At: Big Rock Point Site, Charlevoix, Michigan Inspection Conducted: April 21 through May 9 and May 16 through June 7, 1988 od/6 b u L!h->~
/
/
2E M Inspector M. P. Huber d'/
/
N Date oblYwd
/t U
/v J. F. Smith 62#
.
Oste hfhtU C#
'
Approved By:
0. H. Danielson, Chief-d (
Materials and Processes Section Date Inspection Summary Inspection on April 21 through May 9 and May 16 through June 7,1988 (Report No. 50-155/88011(ORS))
Areas Inspected:
Routine unannounced safety. inspection of the licensee's activities with respect to NRC Bulletin No._85-03 and Supplement 1, both titled "Motor-0perated Valve Common Mode Failures During Plant Transients Due to Improper Switch Settings," (25573), implementation of the Pump and Valve Inservice Test (IST) Program (73756) and followup of previously identified items of noncompliance (92702).
Results: Of the areas inspected, one apparent violation was identified (Paragraph 5.e).
8807120008 880629 PDR ADOCK 05000155 Q
PDC y
. _,
--
_
_ _
__
-_
.
.
,
,. -
.
.
DETAILS 1.
. Persons Contacted Consumers Power Company (CPCo)
-+T. Elward, Plant Manager R. Abel, Production Superintendent
+M. Acker, ISI Coordinator
+R. J. Alexander, Technical Engineer
+E. M. Evans, Senior Engineer
+J. J. Fremean, Director, Nucler.r Safety
-+D. Moeggenberg, Engineering Supervisor
+L. Hanshar, Quality Assurance Superintendent
-+G. C. Withrow, Engineering Maintenance Superintendent-J..Tosky, Engineering-T. Fisher, Quality Assurance
+ Denotes those who attended exit meeting held on May 20, 1988.
-Denotes those who attended telephone exit on June 7, 1988.
'
2.
Licensee Action on Previous Inspection Findings (Closed) Violation 155/87022-01):
Failure to correct administrative weaknesses of the !ST Program. The NRC inspectors reviewed two procedures, No. 3.1.5.1 titled, "Inservice Testing of Valves" and No. 3.1.5.2 titled,
"Inservice Testing of Selected Safety-Related Pumps" both dated November 24, 1987.
These procedures established the testing program for the valves and pumps in accordance with the ASME Boiler and Pressure Vessel Code,Section XI. Requirements were defined to ensure that the appropriate performance and evaluation of the testing was completed by the responsible personnel.
These administrative procedures were created in order to ensure that the chance of missed requirements and an overall inadequate IST Program would no longer exist.
This item is considered closed.
3.
Licensee Action on IE Bulletins (IEB)
{0 pen) TI 2515/73 and IE Bulletin No. 85-03 (155/85003-88) and Supplement 1 (155/85003-18)): _ Motor-0perated Valve (MOV) Common Mode Failures During Plant Transients Due to Improper Switch Settings.
The bulletin requested licensees to develop and implement a program to ensure that switch settings on certain safety-related MOV's are selected, set, and maintained correctly to accommodate the maximum differential pressures expected on these valves during both nor. al and abnormal events within the design basis.
In NRC Inspection Report No. 155/87022(DRS),
the NRC inspectors review of the plant's safety-related core' cooling system (Core Spray) concluded that certain M0V's would be required to open against full reactor pressure.
Following issuance of the inspection
- a
.
i
_
.
._.
..
.
.
-
report, the NRC staff issued a Request for Additional Information (RAI)
and Supplement 1 of the bulletin and established the position that the requirements of IEB 85-03 apply to Big Rock Point.
At the time of this inspection, the RAI and Supplement had just been received by the licensee and response had not yet been completed.
The NRC-inspectors discussed the proposed response to the bulletin with the licensee and determined.that M0-7071 will be included within the scope of the licensee's bulletin program. However, MO-7061 will not be included in the program.
Initially, this was not clear to the NRC inspectors and the licensee indicated that the response weuld be clarified to explicitly state this fact and provide adequate justification for not including the valve in the program.
The licensee's reasoning was that the spool piece between the valve (M0-7061) and the down stream check valve would be verified to be depressurized following stroke testing of valve. As a result, MO-7061 would not be exposed to the differential pressure and would not be subject to the bulletin requirements.
The NRC inspectors reviewed Procedure T30-22 titled, "Emergency Core Cooling System Valve Tests" and agreed with the licensee that depressurization of the spool piece and non-leakage of the valve, at that time, was verified; however, there is a potential for leakage to develop during reactor operation and pressurite the spool piece. Under this condition, depressurization would require the valve to open against a large differential pressure and therefore the bulletin would be applicable to Valve MO-7061.
't The bulletin will remain open until the valve testing issues are resolved and review of the program is completed.
4.
Pump and Valve Inservice Testing (IST) Program Implementation The licensee's pump and valve IST Program implementation was inspected to verify compliance with Appendix B of 10 CFR Part 50; 10 CFR Part 50.55a(g);
and Subsections IWp and IWV of Section XI cf the ASME Code (1977 Edition with Addenda through Summer 1978).
The inspection included a review of administrative controls, selected surveillance procedures, test results, and documentatlon.
Several concerns were identified in NRC Inspection Report No. 155/87022(ORS)
with respect to possible deficiencies in the licensee's IST Program.
The li:ensee's actions taken in response to these concerns was used by the NRC inspectors as the basis for the inspection.
,
a.
Inservice Testing of Pumps Although the NRC had been informed that no pump bearing temperatures and flow measurements would be taken, the licensee had not requested relief from these requirements. The NRC inspectors determined that
,
the relief requests had not been completed prior to the end of this
'
l
.
-
- -
-
.
-
-
-
-
- -
--
-
-
-
--
- -
- -
-
-
' '
.
.
.
inspection, when the licensee indicated that they would be completed by the 1988 refueling outage.
On June 24, 1988, the licensee submitted a request to the NRC for relief of the testing requirements.
The limits of alert and action ranges defined in the ASME Code for allowable pump operating parameters are specified by the pump tes t procedure. Corrective action shall be initiated once the parameters fall outside the predetermined limits. No problems were identified with this approach and no violations or deviations were found, b.
Inservice Testing of Valves (3) Valve Position Indicator Verification The ASME Code Subarticle IWV-3300 requires valves with remote position indications to be verified every two years. The NRC inspectcrs reviewed procedures to determine the adequacy of the tests and that valves required to be tested were included in the program. Procedures reviewed include:
TV-09, Revision 12, "Emergency Condenser Valve Operability Test," dated January 6, 1988.
TSli-03, Revision 7, "Cold Reactor Depressurization System (RDS) Isolation Valve Testing," dated February 23, 1988.
TR-52, Revision 24, "Sphere Isolation Test," dated June 18, 1987.
TR-88, Revision 2, "Core Spray and Enclosure Spray Valve Initiation and Operability Test," dated June 5, 1984.
The NRC inspectors determined the procedures to be adequate and no problems were noted.
(2) Valve Stroke Timing The licensee established upper stroke time limits for the valves listed in the IST program.
The NRC inspectors found the criteria that was used to be acceptable and all valves had limiting stroke times in accordance with the licensee's established criteria.
The NRC inspectors did note that additional work needed to be j
performed by the licensee.
The books used by the licencee to i
record valve stroke times and trend the valve performance had not yet been updated to include the revised times.
The licensee agreed with the NRC inspectors concern; and agreed to incorporate the limiting stroke times into the trend books by September 1988.
i
-
- - -
-
-
.
-
-
,
.
,
-
.
In addition, post maintenance trending of stroke times was accomplished by a licensee operations: order.
Instructions had been provided to the Shift Supervisors with regards to the routing of the valve stroke times following maintenance for incorporation into the trending books. The licensee indicated that their maintenance procedures would be revised by the end of 1988 tu reflect this practice.
(3) Pressure Isolation Valves An NRC letter niated October 15, 1986, transmitted information pertinent to the licensee's IST program.
It identified eight valves which appeared to perform a pressure bouncary function, but which were not Categorized "A" or "A/C," as appropriate.
The licensee's response to that letter, dated December 27, 1986, agreed with the NRC position and indicated that the valve classification would be corrected. The NRC inspectors reviewed current IWV t.esting tables to confirm execution of_this commitment and found that Core Spray _ Isolation Valves M0-7061 and M0-7071, which were included in the group of eight pressure boundary valves, were still incorrectly classified as Category "B".
Discussion with the licensee disclosed that_the improper classification was the result of an inadvertent idministrative oversight. Prior to the end of the inspection, the licensee had completed Quality Review Form, Log No. 726-88 dated May 20, 1988, to correct the anomalous entries. _-The changes to TV-30, the document in which the IWV testing tables are located, included reclassifying Valves M0-7061 and M0-7071 to valve Category "A" and specifying required inservice testing for the vtives. The NRC inspectors noted during the inspection that relief wnuld H o,Z.ed from leak testing Valve M0-7071.
The licensee has nor finite means for deter.i niro the leakage i
rate through the vahe - '~ 4.-ing that vo'VE C,lerdtion during the coui '.c F Lr,e,.'..%,+eration demonstrMes functionslly adequate seat tightness. Yne licensee agreed with the NRC inspectors observations and will submit a request for relief from this testing.
(4) Relief Valve Testing Testing of safety and relief valves is defined in Paragraphs IWV-3511 through IWV-3514 of the ASME Code. The licensee's Test Procedure TR-28, Revision 29, titled,."ASME Boiler and Pressure Vessel Code Section XI, Steam Drum Relief Valve Testing" tests only for the setpoint where the valve opens and cannot be revised to include flow capacity and reseating pressure using the bench testiri method and equipment used by the licensee. The NRC inspectort 2xpressed concern that the licensee's current test method, which involves testing the valve on a test stand using a limited pressure source, was
<
.
.
.
._
.
.
.
,
.
.
.
,_
,
inadequate. However, af ter further review the NRC inspectors concluded that operational readiness is determined by the licensee's bench testing methods, as defined in Section 4.091(c)
"Bench Testing" of Performance Test Code 25.3-1976, which allows bench' testing with stands having limited accumulator volume and/or pressure source capacity.
No violations or deviations were identified.
5.
RDS Valve Testine a.
During this inspection period, the NRC it 3pectors reviewed'the activities of the licensee to address problems-with Reactor Depressurization System (RDS) depressurization valves.. The depressurization valve is designed to open upon energization of the pilot valve, either manually or automatically in response to specified plant conditions. Energization of the scienoid valve opens the pilot valve, venting the area above the main piston and pennitting system pressure (normally 1330 psig) to fully lift the main disk off its seat.
On April 21, 1988, Depressurization Valve "B"-(SV-4985) was. tested-at a laboratory with unsatisfact(
esults.
The valve was initially verified to be leak free in the i stand when pressurized to 1350 psig. After reducing pressure to 1000 psig for full stroke functional testing, repeated attempts to stroke the valve by energizing the solenoid were unsuccessful and resulted in no valve stem movement although the pilot valve was verified to be actuating properly.
In the' initial unsuccessful tests, the solenoid was energized for approximately two seconds. Licensee data for testing of other RDS valves in 1985 showed the delay tin;e from solenoid energization to commencement of stem travel to be 600 milliseconds for initial lifts and approximately 1500 milliseconds for subsequent lifts when a two phase steam / water mixture was present in.the discharge piping.
The licensee then reduced pressure to zero, energized the solenoid
,
and steadily increased pressure in a minimum pressurization test.-
'
With zero pressure downstream, a steam pressure of 650 psig was required to lift the disk from the seat.
Inspection revealed the stem and stem guide area to be corroded 'with a black oxide and the stem displayed drag marks and g::uges. The corrosion which inhibited stem travel during lift tests also prevented reseating of the disk on the seat and testing was terminated. Scheduled replacement of a cracked seat, top assembly ebuild and machining of main disk as required was comenced.
l Initial tests were conducted in the horizontal position.although
the valve's system orientation is vertical.
This testing variable was retroyed when the laboratory modified the test rig for vertical testing.
6
.
- -
-
-
'.
'
.
,
The NRC inspector expressed the following concerns to the licensee on April 28, 1988:
(1) Based on the unsuccessful as-found testing of SV-4985, one of four RDS trains was apparently inoperable, or at least seriously degraded, for some in6eterminable period prior to shutdown on April 9, 1988. The inability of SV-4985 to perform its safety function would seriously degrade the licensee's ability to respond to a plant emergency.
Failure of two valves during a LOCA of the size against which RDS was designed to protect would preclude achieving the required rete of depressurization, according to BRP Technical Specification bases, and according to the licensee's analysis, result in core damage.
(2) The licensee's implementation of the quarterly surveillance test required by Technical Specification 4.1.5 has apparently been inadequate to verify valve operability.
The surveillance, which discharges a small volume of gas upon actuation, verifies the pressure decay of the test gas and brief loss of full closed indication on the valve's main stem and disk.
Full open indication is not received during the test, indicating the stem travels some distance less than full stroke. Stem travel of only 1/8 to 1/4 inch is required to lose closed indication, according to the valve's vendor.
(3) Based on the findings on SV-4985, the condition of des msurization valves in the A, C, and D treins was in question.
(4) Determination of the source of corrosion ari the rate of corrosive oxide buildup un the stem surfac6. should be evaluated against the apparent inadequacy of the surveillance.
This evaluation should be factored in.o the licensee's pending change to Technical Specifications.
(5) Maintenance practices should be enhanced to ensure craftsperson's awareness of the need to verify that the stem travel is not degraded by corrosion. Based on the NRC inspector's observations of previous valve disassembly and inspection activities and interviews with maintenance department supervisors, it is apparent that maintenance personnel do not include as part of their evaluation a check for corrosive binding of the stem.
The NRC inspector's review of Maintenance Procedure MRDS-2,
"Replace, Repair and/or Inspect All or Part of the RDS Depressurizing Valves" performed in September and November 1987 detemined that the procedural steps which might have identified stem bir. ding were marked "N/A" by the repairman and not perfomed.
-
.
.
-
-
-
-
.
.
..
.
,
(6) The NRC inspector clearly conveyed the expectations of Regional and NRR Management that an adequate licensee response to the NRC inspector's concerns would require testing of all four depressurization valves in a manner that ensures acceptable valve performance and full valve stem travtl. Testing approaches other than live steam testing at a laboratory could be acceptable provided the test methodology yielded comparable results.
On April 29, 1988, the NRC inspector attended a meeting of the Corrective Action Review Board convened to examine the failure of SV-4985 and devise a course of action for future testing. The licensee elected to send SV-4984, the "A" train depressurization valve, to the steam laboratory fr testing.
On May 6-9, 1988, the NRC inspector reviewed testing of RDS Depressurization Valve SV-4984 performed May 4,1988.
The inhibited stem travel observed during SV-4985 testing was not in evidence and the valve operated with a delay time of 260 milliseconds on the initial lift. A second test resulted in a delay time of 880 milliseconds. However, on the third attempt, the valve failed to opere 'e after 35 seconds of solenoid actuation. The licensee was unabie to establish an apparent cause.
Overheating or other problems with the pilot valve solenoid were apparently ruled out when the licensee immediately performed three successful minimum pressurization tests in which only 12 psig was required to lift the main disk off its seat, indicating pilot valve operability.
Repeated testing eventually resulted in pilot valve disc / scat degradation and the vendor comenced repairs to SV-4984.
In an effort to address the NRC inspector's concerns over the ability of the qu6rterly surveillance to verify full stroke stem travei as required by applicable codes, the licensee designed and constructed a test rig installed cn a spare top assembly which employed multiple reed switches operated by movement of the magnetized rod atop'the main stem. With the test rig installed on the recently rebuilt and reinstalled SV-4985, a quarterly surveillarice as described earlier was satisfactorily performed on May 8,1988. Although control room closed position indication was lost and regained, thus satisfying the surveillance's acceptance criteria, strip charts attached to the various test reed switches did not indicate upward stem tra'!el.
These results lend support to the NRC inspector's position that stem travel is very slight, possibly in the range of 1/8 to 1/4 inch, and is sufficient only to depressurize the limited volume of gas used as the test medium.
Previous performances of the quarterly surveillance indicates that total time to depressurize the gas is 0.8 - 1.2 seconds.
Based on the inability of the special test to verify full stroke travel, the licensee ele:ted to recove Depressurization Valves SV-4986 and SV-4987 on the "C" and "D" trains for full stroke testing at the laboratory facility.
It should be noted that other problems existed with the 'tDS valves.
Pilot valve leakage was of more immediate concern. The principal problem experienced by the licensee was the inability of the pilot
m
-
.
.
-
,
.
.
.
.
..
.
,
valves to seat after operation.. They appeared' to hold pressure -
until partial stroking is'done and then failed to reseat properly.
The reasons that l' 4kage occurs is partly due to pitting of the disk face and M a drawing brought about by the leakage induced by reseating on pitted. areas.
Leakage of the pilot valves is serious because it represents Reactor
'
Coolant system leakage which goes directly.to the containment.- It also can prograss to a point where the RDS valve'would self-actuate,-
possibly lead to personal injury and/or blowdown of the. reactor. The
. licensee.has addressed this problem by inodifying the pilot valve'and submitting a proposed technical specification change to prevent unnecessary stroking of these valves, b.
When the NRC inspectors reviewed the possible causes for ~the failure of the RDS "B" valve the following was found:
(1) The weld overlay of Stellite No. 6 on the main disk and stem was severely pitted and blackened by steam and by the. water condensed from steam, both of which were.in the valve simultanevJsly. The most serious damage seemcd to.be done at the steam / water interface.
The next most serious corrosion was evident in the area under water. The area subjected to stagnant
.
steam was the next most seriously attacked.- The least damage
was evident in the area below the disk / seat contact area; here, there was no evidence of corrosion damage.
(2) The damaged Ask and stem from a similar valve were sent to Battelle for analysis.
Battelle destructively analyzed the sample and concluded that the stellite had~ pitted as a result of exposure to a two phase water / steam system which-was known to produce corrosion in stellite. Their metallugraphic sections conclusively showed pitting of the stellite and incidentally, showed a vetry straight line interface with the staialess steel base metal. This ordinarily is indicative of an overlay
'
process which has very little dilution of the overlay metal by the base metal. Their x-ray intensity indicated the following approximate compositions:
'
ELEMENT * RCoCRA** BASE METAL GEN"L PROD. SCALE PITS MET X-SECT
-
-
GRAY MATFL A
B TWO ARFAS
.
l Cr 26-32 38.4 24.8-25.4 27.8 34.1 26.5 Fe
26.8 49.0 4.8 7.6 55.9 68.0 Co Rem.
34.0 12.5 59.9 61.4
--
--
Ca 6.5 2.4 3.1
---
--
--
--
No
6.8
--
--
--
--
--
W 3-6 7.5
--
--
--
--
--
Si
4.2 5.2
--
--
--
--
C 0.9-1.4
--
--
--
--
--
--
Ni
--
--
--
--
--
--
- Percent; where single values are given, the numbers are maximum values,
- Typical composition for Stellite 6 overlays.
,
,
-
- -
- -
-
-
.
.
.
A
.
_
.
~
..
..
.
..
,
At the time of the inspection, the root cause for the valve failure (including possible problems with materials) had not yet been determined.
There are several factors.to be con-idered in properly addressing this issue. The NRC inspectors have li..'ormed the licensee that the testing report and root cause would be necessary to properly address this issue. The licensee understood this, however, contributing factors and the root cause will not be dete' mined for some time, c.
During the inspection period, the NRC inspectors researched licensing correspondence to assess the licensee's position on full stroke testing of the depressurization valves.
It was-found that the licensee had requested relief from full stroke testing these valves due to their installed configuration. The following is a listing of the correspor.dence between the NRC and the licensee and other actions taken by the licensee'in response to the issue:
(1) Relief from full stroke testing of the depressurization valves
,
was requested by the licensee, as part of their IST program
,
for pumps and valves, in a letter dated February 1,1979, to the NRC.
(2) NRC letter to CPCo requesting specific technical basis for not full stroke testing of the depressurization valves dated April 27,1979.
(3) Resubmittal of relief to NRC for full stroke testing in letter dated Decemter 7, 1979. No revision of the request or additional technical basis was provided as requested.
(4) CPCo internal correspondenc'.: dated January 9, 1981, describes concern that Target Rock valves should be full stroked, but
)
discontinued further investigation,
(5) Jun<. 1, 1983, letter to NRC from CPCo describing the licensee's l!
Integrated Assessment and Scheduling Approacn for Issue Resolution Program, including Issue No. 83, the full stroke
,
"
testing of RDS valve issue.
This was identified as an issue that is essential for safe and reliable future operation of the plant.
(6)
Integrated Assessment of Open Issues and Completion Date for
,
Issue Resolutions was submitted on February 2,1984.
Issue
'
No. 83 was identified and work was to be completed on this issue by the end of March 1985.
(7) Letter from NRC to CPCo dated March 2,1984, requested additional information from CPCo.
It was identified in the RAI that it was unacceptable to never full stroke exercise the valves and alternate test methods should be investigated.
i
.
.
- -
.
.
-
.
..
-
,.
(8) The Big Rock Point Integrated Plant Safety Assessment SEP dated May 1984 was the final report of NUREG-0828.
Section 5.3.1.3.
"Full Stroke Testing of RDS Valves," identified the necessity for the RDS valves to be reliable and the necessity for performance of full stroke testing. The licensee proposed to review several aspects of the valve design and test method to determine if the test 1. valid by March 1985.
It should be noted that the NRC concluded the pilot valve leakage evaluation should take precedence over the full stroke testing issue.
(9) May 14, 1984, CPCo internal correspondence discussed the full stroke issue as identified in a letter from the NRC dated March 2, 1984. Again, it was concluded that full stroke testing was being considered as identified in Issue No. 83 of the integrated plan.
It was also identified that the current test was only a "pop-test" and was inadequate.
(10) A letter from CPCo to NRC dated October 4, 1984, responded to clarifications made to NRC Request for Additional Information stating that full stroke testing is being addressed as part of the Integrated Plan.
(11) Letter from NRC to CPCo dated October 15, 1986, transmitted results of the contractor review of the IST program in a Technical Eva'uation Report. The results stated that not full stroke testing of the reference valves was unacceptabl.s.
(12) Letter from CPCo to NRC dated December 22, 1986, submitted the same request for relief from full stroke testing, although it was clearly known to the licensee that this approach was unacceptable.
The NP^ inspectors presented the following concerns to the licensee, based ca the above:
(1) The need to full stroke test depressurization valves in order to fully comply with the ASME Code for inservice testing was known to the licensee in 1978, based on the December 22, 1978, reliet request.
(2) The substantial safaty significance of the RDS dapressurization valves and the licensee staff's concerns over adequacy of the surveillance to verify valve operability has been well documented since June 1983.
(3) The NRC staff clearly stated their expectation that a means should be developed to conduct full stroke testing and has been known by t.he licensee since March 1984.
(4) The NRC staff's conclusion that never full stroke testing the valves is unacceptable.
.
-.
..
,.
(5) Although the need to full stroke test has been carried on the Integrated Plan for several years, a review of licensee activity indicated virtually no licensee effort toward that end.
One depressurization valve was tested in 1985 following maintenance performed by the vendor at the test lab, but that testing was to verify post maintenance operability and was not part of any program to investigate full stroke test methods or satisfy regulatory or code requirements, d.
The NRC inspectors discussed the history of the RDS valves including testing and maintenance activities performed on the valves, and Subarticle IWV-3400 of ASME code Section XI, which requires full stroke testing of the valves.
It was also noted that RDS depressurization valve reliability has a d' rect impact on the reliability of the entire RDS. This includes the evaluation of the pilot valve leakage problem as well as the full stroke testing of the valves. A review of the licensee PRA indicated that RDS valve reliability needed improvement in order to reduce the risk of core melt.
To some extent, maintenance and testing will improve reliability and reduce risk. However maintenance performed in the past failed to identify the degraded condition of the valve.
Partial stroke testing of the valve was performed to demonstrate that the system would perform satisfactorily, however, it in fact did not adequately demonstrate the operability of the valves. A partial stroke test was performed in November of 1987 and met the acceptance criteria.
The failure of valve SV-4985 to stroke during testing performed in April of 1988 demonstrates the inadequacy of the partial stroke test. The valve stem had been so badly corroded that no valve movement could be achieved, and the corrosion of the valve occurred over a time period of approximately 12 years, the period the valve was in service. Accelerated corrosion may have been present at the end of life however it is likely that the mechanism occurred over a time frame of one to two years.
The partial stroke test was inadequate and virtually no full stroke testing (only 1 valve was full stroke tested) of the valves was performed over the life of the system.
The reliability of the system was in question for an indeterminable time. One valve was inoperable for an indeterminable time, and the reliability of the others was in question.
e.
In conclusion, the reliability, if not operability, of the RDS to perform its function was in question. The ASME code Section XI Paragraph IWV-3412 defines the exercising procedure for valves and requires the full stroke testing of the valves.
The licensee
,
f ailed to develop a testing program in resnonse to IST requirements, NRC initiative,, and the licensae's own program (Big Rock Point Integrated Plan) for a period of 6t least sevm years.
The surveillance test in place to ensure valve operability was merely a "pop-test" and inadequate in that valve operability was not proven by this type of test.
l
.
. - -
.
.-
...
-
.-.
.
-..
-
.
.
-
,
.
,
The inadequate test program, correction of the program, and full
' '
stroking of these valves is considered a failure to meet Technical Specification 9.0 "Inservice Inspection and Testing."
(155/88011-01(DRS)).
The licensee has modified, refurbished, and tested the
"A", "C", and
"0" valves (SV-4984, JV-4986, and SV-4987) prior to installation at the plant. The "B" valve (SV-4985) will be refurbished by replacernent of main valve internals with new parts, the pilot valve modified as recommended by: the manufacturer and full stroke tested prior to installation.
In order to meet requirements of the code and maintain a higher degree of reliability in the future, the licensee has-
,
committed to full stroke test four of the valves and physically inspect the internals of one valve each refueling outage.
Additional corrective actions are pending further analysis to be conducted by the licensee.
The complexity of the issue and.many variables iivolved with the root cause and meaningful long-term corrective octions dictates that resolution will take time.
The licensee's analysis of the problem and long term corrective actions will be considered an open item (155/88011-02(DRS)).
6.
Open Items Open items are matters which have been discussed with the licensee, which will be. reviewed further by the inspector, and which some action on the part of the NRC or licensee or both.- An open item disclosed during the inspection is discussed in 5.e.
7.
Exit Interview The NRC inspectors met with licensee representatives (denoted in Paragraph 1) on May 20,1988 to discuss the scope and findings of the inspection. The licensee acknowledged the statements made by the inspector with respect to items discussed in the report.
In. addition, a telephone exit interview was conducted on June 7, 1988 with the licensee as a result of the inspection effort conducted at the NRC Region III office.
The inspectors discussed the likely informational content of the
>
inspection report with regard to documents or processes reviewed by the
,
inspectors during the inspection and the licensee did not identify any
such documents or processes as proprietary.
>
i
l
.
.
-
-
-
-
-
-
-
-