IR 05000352/1985043
| ML20205J168 | |
| Person / Time | |
|---|---|
| Site: | Limerick  |
| Issue date: | 01/10/1986 |
| From: | Beall J, Gallo R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML20205J136 | List: |
| References | |
| 50-352-85-43, NUDOCS 8601300112 | |
| Download: ML20205J168 (19) | |
Text
.
.
U. S. NUCLEAR REGULATORY COMMISSION
REGION I
Report No. 85-43 Docket No. 50-352 License No. NPF-39 i'
Licensee: Philadelphia Electric Company 2301 Market Street
!
Philadelphia, Pennsylvania 19101 Facility Name:
Limerick Generating Station, Unit 1 Inspection Conducted: November 1 - 14, 1985 Inspector:
E. M. Kelly, Senior Resident Inspector f[h
/)o[gd,
- **
f E. Beall,' Project Engineer date
,
Approved by:
'
R. M. Gallo, Chief date Reactor Projects Section 2A, DRP
,
Summary:
Areas Inspected: Special inspection conducted by the senior resident inspector (49 hours5.671296e-4 days <br />0.0136 hours <br />8.101852e-5 weeks <br />1.86445e-5 months <br />) concerning the discovery by an auxiliaries operator of both Emergency Service Water (ESW) system Loop 8 pump discharge valves in a locked and throttled-
,
closed (not fully open) position on October 31, 1985.
The inspection also addressed the discovery of an unlocked (but fully open) ESW pump A discharge valve on November 14, 1985.
These valves are required by station drawings and approved operating procedure checkoff lists to be normally locked in the full-open position.
Results: The cause, duration and extent to which the Loop B valves had been throttled were not established, in that, no record of throttling or evidence of
a fully-closed condition was found. Surveillance test results, valve lineup
'
verifications conducted prior to October 31, and the documented as-left valve positions at noontime on October 31, would indicate that the valves were in'a normally locked full-open position, contrary to their as-found positions at 8 p.m. on October 31, 1985. A licensee conducted safety evaluation implemented to verify "as-found" system operability, concluded that ESW Loop B was capable of supplying design flow and therefore remained operable. A violation was identified regarding locked valve controls and proper lineup for the ESW pump discharge valves. (Details 4.5 and 7.2)
9601300112 860123'
ADOCK05%32 PDR
.
-
-
-
-
-
-
-
-
.--
-
- -
-
-
-
-
.
"
.2 Unresolved items associated with the administrative control of locked valves, the practice of independent verification of manual valve positions, the revision'
of IST procedures which may throttle non-automatic valves (Detail 5.2), and the supply of HPCI by only the B Loop of ESW were also identified during this inspection. (Detail 6.1)
i
.
.
DETAILS 1.
Persons Contacted G. Leitch, Plant Manager J. Franz, Superintendent of Operations J. Doering, Operations Engineer P. Duca, Technical Engineer T. Shea, Test Engineer Shift superintendents, supervisors, l' censed and non-licensed plant opera-i tors were also contacted and, in some cases, interviewed during this inspection.
2.
Event Description
.
On October 31, 1985, at approximately 8 p.m., a non-licensed Auxiliaries Plant Operator (AO) informed control room supervision that the discharge valves on both the "B" and "D" emergency service water (ESW) system pumps (Loop B) were fully closed and locked.
These valves are required to be normally locked and full open, and keys to those locks are under the ad-ministrative control of shift supervision. The full closure of both of those valves would have rendered ESW Loop B inoperable, which affects the operability of HPCI, the 012 and 14 emergency diesel generators, and the
"B" and "0" residual heat removal (RHR) and core spray (CS) pumps, as well as the "B" control room chiller.
(See Attachment 1 for simplified Loop B schematic.)
The A0 had been dispatched to the spray pond pump house (SPFH) for the performance of the remaining portions of a surveillance test of the ESW pumps which had begun on October 28, 1985, and which had been suspended at approximately 1:00 p.m. on October 31.
In preparation for the surveil-lance, the A0 had obtained a key, under authorization of shift supervi-sion, in anticipation of having to unlock and throttle the pump discharge valves to establish the desired test flow condition as specified by the test procedure. The procedure had been revised that afternoon, changing the test flow criterion from 3000 to 2100 gpm.
The A0 found both ESW loop B pump discharge valves to be locked and appar-ently closed; a condition other than the normally required alignment of locked and fully-open. The A0 was directed by shift supervision to open both valves, and an entry was made in the shift supervisor's log at 8:15 p.m. on October 31, 1985 that the valves were found closed and were opened. The surveillance test was then conducted, running the "B" ESW pump between 8:24 p.m. and 9:27 p.m. and the "D" ESW pump between 9:33 p.m. and 10:20 p.m.
The A0 throttled the discharge valve on each pump to achieve the flows required by the revised test procedure and, at the end of the testing, he returned both valves to a fully-open position and locked them.
This condition was independently verified by another A0 at 10:40 p.m. that same night, and documented in the locked valve log.
.
.
,
3.
Emergency Service Water (ESW) Design Description ESW provides cooling water from the spray pond (ultimate heat sink) to essential Unit 1 equipment at the following design flow rates:
Emergency diesel generator heat exchangers; 4 diesels (total), 700
-
gpm per diesel
-
HPCI pump room unit coolers; 2 coolers (normal and standby); 100 gpm per cooler RCIC pump room unit coolers; 2 coolers (normal and standby), 80 gpm
-
per cooler
-
Control Room chillers; 2 chillers (one normally in-service), 350-815 gpm per chiller Core Spray pump room unit coolers; 4 pumps total), 2 coolers per pump
-
(normal and standby), 100 gpm per cooler RHR pump room unit coolers; 4 pumps (total), 2 coolers per pump (nor-
-
mal and standby), 230 gpm per cooler RHR pump seal water and motor oil coolers; 1 each per pump, 48 gpm
-
per pump The system consists of two independent loops designated "A" and "B", each of which supplies an identical complement of the above equipment with the exception of HPCI and RCIC room coolers. All four diesels may be aligned to either loop; however, normal alignment is.for ESW Loop A supply to the D11 and 13 engines, and ESW Loop 8 supply to the D12 and 14 engines.
Technical Specifications currently prohibit alignment of all four diesels to either loop until confirmatory flow testing is performed to verify ade-
.quate flow to the RHR pump motor oil coolers. A significant feature of the ESW system is that ESW Loop A supplies RCIC room coolers, while Loop 8 supplies the HPCI room coolers.
Flow demands on either ESW loop during normal alignment and operation, such as occurs during ST-6-011-232 testing, are variable. This is due to-the fact that:
pump room unit coolers draw flow when their respective fans are run-
-
ning, which in turn occurs upon either a pump start or high room tem-perature.
Further, during a pump start, the normally selected fan-cooler combination will start; the standby fan-cooler unit initi-atos only if the normal fan-cooler unit fails to start. Thus, there is no flow to a unit cooler during a typical IST surveillance.
flow to a control room chiller varies with inlet service water tem-
-
perature, and will range from 350 - 815 gpm. Flow is controlled by a
,
valve set to maintain an 85 degree F outlet water temperature.
-
-
-
-
_.
.
e
~The chillers are alternated with only one chiller. typically selected as in-service. As an example, during an ESW Loop 8 IST with the "A" chiller in service, the minimum 350 gpm flow is constantly drawn; the same test with the "B". chiller in-service would draw a load varying from 350 -~815 gpm.
Each loop of ESW consists of two 50% capacity pumps capable of supplying a rated flow of 6400 gpm at 240 ft. head or approximately 110 psig discharge pressure. The shutoff head of each pump is about 130 psig. The pumps are located in the Spray Pond Pump House, and take suction from the spray pond and discharge through 20-inch piping to a common underground header for that loop.
Each pump has a 20-inch manually-operated discharge gate valve, required to be in a locked and full-open position by ESW system operating procedures and Administrative Procedure A-8.
The valves, tagged 11-0002A through D for the respective pumps, are ' located directly at each pump's discharge.
Each loop of ESW also serves non-essential loads such as the Reactor Enclosure and Turbine Enclosure cooling water systems, as a backup to normal. service water, and can provide emergency makeup to the spent fuel pool. The system will automatically initiate upon the respective diesel generator start signal (e.g., diesel D13 running will initiate "C" ESW pump), which will automatically isolate required loads.from normal service water.
ESW flow is returned, via common RHR service water piping, to the spray pond.
FSAR Table 9.2-3 specifies that each RHR pump requires 36 gpm for seal.
>
water cooling and 12 gpm for motor oil cooling.
Since no means currently exists to verify these design flows, an operating restriction was placed on the ESW system that no more than two diesel generators could be aligned to an ESW loop. The operating restriction addressed the potential for flow starvation to the RHR motor oil coolers, and was added as a note in Technical Specification 3.7.1.2 until the completion of confirmatory flow testing. The note prohibits aligning all four diesels to a single loop if one entire. loop of ESW were lost. Operating procedure S11.1.A and Operations Engineer Control # 11 - 13 are more restrictive than the Tech-nical Specification note, and cautions against aligning more than two die-sels per ESW loop.
4.
Surveillance Testing 4.1 ESW System Testing 4.1.1 Testing on the Morning of October 31, 1985 Surveillance test (ST)-6-011-232 is required by Technical Sjpecification 4.0.5 to be performed quarterly to verify operability and test the performance of the "B" and "D" ESW pumps, as well as to exercise, stroke-time test and fail-safe test more than 40 different valves associated with operation of ESW Loop B.
The surveillance test is intended
to satisfy the inservice testing (IST) requirements of ASME
.
.-
l
.
.
.
Code Section XI.
The running of both ESW Loop B pumps had been performed during the period of 9:30 a.m. through noon on October 31 as part of the original ST.
Steps 6.4.6 and 14 of the test procedure call for incal manual throttling of gate valves 11-00028 and D (for the B & D ESW pumps, respectively) to establish the desired flow condition of 3000 gpm..However, upon starting an ESW pump with normally aligned equipment loads in-service, flow had been previously observed to be consistently less than 3000 gpm, so that throttling of the normally-locked full open discharge valves was unnecessary.
Additional loads are usually required to be placed on ESW to achieve the higher flow, but, on the morning of October 31, shift supervision interpreted the procedure as only allowing for ESW Loop B valves to be lined-up for normal operation in accordance with approved operating procedures.
The ST was therefore suspended at approximately 12:30 p.m.,
since flow greater than 2400 gpm could not be achieved without a procedurally unauthorized valve lineup change.
The test results were analyzed by licensee engineers and, at approximately 2 p.m.,
it was decided to partially fail the earlier test (pumps only), but to change the pump flow criterion to a lower value. ASME Code Section XI allows for such a change, and specifies that data be evaluated within 96 hours0.00111 days <br />0.0267 hours <br />1.587302e-4 weeks <br />3.6528e-5 months <br /> after test completion before an operability decision is required. A partial ST was prepared under Test Procedure Change (TPC) 85-1387 to demonstrate ESW pump operability at a new reference flow of 2100 gpm. based on flow to loads normally in service to provide a repeatable ISI criteria per ASME Section XI.
4.1.2 Suspension of the Morning Testing and Shift Turnover The non-licensed A0 assigned on day. shift to monitor pump.
performance locally (and to possibly throttle the valves)
remained inside the SPPH from 8:38 a.m. to 11:52 a.m. on October 31. No direction was given by licensed control operators to throttle the pump discharge valves, since no reason for throttling existed. The A0 returned to the con-trol room between noon and 12:30 p.m., and informed the control room operator supervising the ST that both dis-charge valves had been left in their original fully-open locked positions. The A0 had not been directed to throttle either discharge valve, nor did he unlock either valve, although an LV-100 key had been brought out to the SPPH by a shift supervisor earlier at approximately 10:45 a.m. in anticipation of having' to throttle the valves. During an
,
interview of the A0 by the NRC senior resident inspector on November 5, 1985, the A0 stated that the LV-100 key was
.
i never used and remained in his pocket, and that the as-left -
position of the valves was unchanged from their as-found position of fully-open and locked. That condition was doc-umented on the. Independent Verification of Restoration (IV0R) data sheet attached to procedure ST-6-011-232 upon the A0's return to the control room at about Noon on Octo-ber 31, 1985. However, the valve positions were not inde-pendently verified until later in the evening of October 31, upon resumption of the surveillance test.
The A0 assigned to day shift was never again involved in the conduct of either the failed ST or the later' partial ST.
He did, however, keep the LV-100 key on his person until his shift relief arrived and briefly worked with him from approximately 2:30 - 3:00 p.m.
The dayshift A0 turned over the LV-100 key to his shift relief, who in turn re-turned the key to the control room at 3:15 p.m.
4.1.3 Retest on the Evening of October 31, 1985 As a result of a test procedure change (TPC 85-1387),which was necessary to demonstrate EWS pump operability at a new reference flow level the nightshift (3 - 11 p.m.) A0 was informed by shift supervision that he would be later involved in performance of the partial ST at the SPPH. The A0 signed-out an LV-100 key at 7:00 p.m. in anticipation of having to throttle the EWS pump discharge valves. The A0 arrived at the SPPH at about 7:30 p.m.,
forgot his vibration meter which he had to return to the plant to retrieve, and re-turned to and badged in at the SPPH at 7:58 p.m.
Shortly thereafter, the A0 reported via telephone to the control room that he had found both ESW Loop B discharge valves in an apparent closed position and locked.
The A0 stated, as confirmed'in two subsequent interviews with the NRC senior resident inspector, that he was direct-ed between 8 and 8:15 p.m. by the control room operator (CO) to open the valves.
The A0 did not attempt to deter-mine whether the valves were fully closed or partially closed, by throttling down the as-found condition. The valve stem threads--the only visual means of verifying valve position--were described by the A0 as showing mostly threaded stem, and therefore the valves were assumed ce fully closed.
Each pump discharge valve is a 20-inch manually-operated gate with a~ rising stem that is visible within a yoke between the valve body and an upper pipe cap, into which the stem penetrates and which acts as a dust cover. When the valve is fully open, no stem threads show within the 19 - 20 inch high open yoke space. When the valve is fully closed, the stem shows complete thread. Any intermediate or throttled position will show some " bare" stem and some portion of threaded ste.
.
The stem thread positions described above were verified by the inspector at the SPPH on November 3, and again on No-vember 13, 1985 with the A0 night shift at which time various valve positions were recreated and the attached pictures were taken.
It was determined that, without physically verifying position of the valves by manipulation, a 90%
throttled valve would be visually indistinguishable from a fully closed valve.
4.2 Previous ESW Loop B Testing The quarterly IST for ESW Loop B had been performed on four occasions since October 1984. The results of~these tests were reviewed by the inspector.and discussed with licensee personnel.
Each test contained a procedural step requiring restoration of ESW Loop B to " normal flow" or alignment by opening each pump ' discharge valve to the "FUL-LY" (as emphasized in the procedure) open position. Return to a fully-open position is a sign-off step in the procedure, and is re-quired to be independently verified.
The IST for ESW Loop B was last performed on July 28, 1985. The in-spector interviewed operators who conducted that test. The responsi-ble Chief Operator stated that he did not remember directing an A0 to throttle the pump discharge valves, nor did he remember having to add additional -loads (specifically another set of diesel generator cool-ers), to provide sufficient flow to pass the test. Documented test results showed 3000 gpm for each pump (flow is indicated in the main control room), with locally measured discharge pressure of 114 psig for the "B" pump, and 112 psig for the "D" pump. The A0 stationed at the SPPH to record discharge pressure during this test stated that he could not recollect whether he had to throttle the ESW pump discharge valves. The A0 also stated that he could not recollect whether the valves were administrative 1y controlled with locks at that time.
In order to determine when the valves were initially locked, the inspector reviewed measures to administrative 1y control locked valves. The inspector could not verify whether the valves were controlled with chain locks during the IST on July 28, 1985, since-the associated IV0R data sheet for the valves listed them as "0 PEN" for both the as-found and as-left conditions. Those conditions were independently verified with the initials of the Chief 0perator who supervised:the test, and who stated that he had sent another A0 to check on their as-left open status. The inspector could not find records of the actual date when these valve handwheels were chain-locked to their respective valve bodies.
The Assistant Operations Engineer found a note dated June 20, 1985, which documented his intent to add these valves to the Administrative Procedure A-8, Appendix 2, Locked Valve List.
Based on Procedure A-8 Revision 2 approved on August 29, 1985 which did contain locked valve controls for all four ESW 11-0002 valves, the inspector concluded
.
.
that the locks were most probably added between June 20 and August 29, 1985. No entries could be found in the Locked Valve Log on July 28, 1985 (or any other dates from January through October 1985), which would indicate throttling or otherwise positioning a locked ESW Loop B discharge valve. Therefore it can be concluded that on August 29, 1985, all four ESW 11.0002 valves were locked in the full open posi-tion.
No violations were identified.
4.3 Monthly ESW Loop B Alignment Verification Surveillance test (ST)-6-011-452-0 is performed monthly to verify the proper alignment of all ESW Loop B valves.
Step 6.3.1 of the ST nro-cedure requires visual verification of the "B" and "0" ESW pump sit -
charge valves in the full open position.
The last performance of this surveillance occurred on October 17, 1985, and the 11-0002 8 and D discharge valves were both found in
'
their required full-open positions, as initialled in the approved ST procedure. Twelve similar surveillances were completed between'0cto-ber 1984 and October 1985, as required by Technical Specification Limiting Condition for Operation (LCO) 4.7.1.2.a which satisfactori-ly verified the correct positions of all 65 ESW Loop B valves includ-ing the "B" and "D" pump discharge valves.
Further, although not required by Technical Specifications or formal-ly documented in a surveillance procedure, an A0 routinely checks SPPH general conditions once per shift. A1.though the shift log entry is termed "SPPH Cleanliness and Housekeeping," several A0s who were interviewed by the inspector stated that part of their roucine obser-
'
vations include such items as ESW pump motor bearing oil levels and discharge valve positions. The licensee concluded that the 'above represented evidence supporting the position that the valves remained in the required open position prior to October 31, 1985 except when administrative 1y controlled by procedure.
.
No violations were identified.
4.4 Diesel Generator Operability Test Runs More than 40 diesel generator test runs have been completed, during the 12 month period covering October 1984 - October.1985, involving the D12 and D14 emergency diesel generators. Approximately half of these have occurred since June 1985, and one for each engine (012 on October 3 and D14 on October 26) took place in the month preceding the performance of.the ESW IST on October 31, 1985.
H I
.
.
The diesel surveillance tests, performed in accordance with ST-6-092-312 or 314, are significant in that the in-series air cool -
er, lube oil and jacket water heat exchangers on each engine require ESW flow for operability.
Each diesel test calls for verification of a respective ESW pump start within 55 seconds of engine initiation.
An ESW pump (either "B" for 012 or "D" for D14) will automatically start, following time-out of the 55-second load sequencing timer, after the engine reaches greater than 200 rpm. During the test, en-gine cooler temperatures are recorded, along with inlet and outlet ESW pressures.
Performance of ST-6-092-314 for the D-14 diesel generator occurred on October 26, 1985. The inspector reviewed the initialled step (6.4.4)
in that procedure which verified that the "D" ESW pump started. The pump subsequently ran for a minimum of one hour during the test with acceptable engine cooler temperatures and ESW inlet and outlet pres-sures of 120 and 60 psig respectively. These test results, along with other previous tests of both the D-12 and 14 diesel generators, indicate that the ESW Loop B pump discharge valves could not have been fully closed for the entire period of January October 1985, when the plant was periodically operated at progressively higher power levels.
No violations were identified.
4.5 ESW Loop A Testing The ESW Loop A IST, conducted quarterly in accordance with ST-6-011-231, was last performed on October 6, 1985.
As in the Loop B test, the pump discharge-valves are specified to be throttled in order to establish the required 3000 gpm. Also, the valves are required to be returned to the full-open position. The
.
valves are not specified in the procedure as locked, although the valves were controlled via-locks at that time. Normal alignment of all four ESW pump discharge valves, per Operating Procedure S11.1.A is "open and locked" and the valves are indicated as locked open on ESW P&ID M-11, Sheet 1, Revision 27. Restoration of both valves was verified as open after completion of the ST on October 6, 1985.
An ESW Loop A lineup verification was performed in accordance with monthly ST-6-011-451 on October 7, and again on November 7, 1985. The Loop A pump discharge valves were visually verified on both occasions as being in the full-o' pen position. However, during routine A0 rounds in the SPPH at.12:30 a.m. on November 14, 1985, the 11-0002A discharge valve was discovered to be full open but not locked. The chain was in place and draped around the valve's handwheel and yoke in a position which appeared to be secured. The lock was affixed to one end of the chain link, but not the other. Shift supervision ex-amined the valve's co'ndition, and directed that the chain be locke.
.
This instance is in noncompliance with Administrative Procedure A-8 which implements methods used to control valves, such as ESW dis '
charge valve 11-0002A, which are required to be locked. This in-stance, together with the findings of October 31, addressed in detail 7.2, constitutes a violation (50-352/85-43-01).
5.
~ Licensee Corrective Actions 5.1 Completed Actions 5.1.1 The licensee interviewed operators associated with the conduct-on October 31, of ST-6-011-232 as well as those involved in previous testing of ESW. Loop B on July 28, 1985.
Those interviews concluded that the documented as-left posi-tions of the ESW Loop B discharge valves at approximately noon on October 31 were not consistent with the observed as-found posittor.s at 8 p.m. that same night.
All four Residual Heat Removal Service Water (RHR SW) pump discharge valves were immediately verified to be in a full open position on October 31.
Those four valves, which were not required to be locked at that time, have subsequently been secured with chain locks and will be added to the locked valve list.
A review of other unlocked ESW system manually operated valves was completed by the inspector, and ESW valves 11-1006 and 1008 were identified as essential to safety-related equipment cooling. The valves are planned to be secured and added to the locked valve list.
The licensee initiated a special procedure to verify full system lineups of all loops of RHR SW and ESW at approxi-mately 7:00 p.m. on November 6.
These verifications were completed satisfactorily by November 7, 1985.
The discharge valves on both ESW Loop A pumps were visually verified to be locked and fully-open shortly after finding the B Loop valves closed at 8 p.m. on October 31 and again on November 7 per the special procedure.
However, as dis-cussed in detail 4.5,.the "A" ESW pump discharge valve was later discovered on November 14, 1985, to be unlocked.
5.1.2 As a result of preliminary (ST)-6 011-232 testing on the morning of October 31, 1985, system flow (ESW Loop B) was observed as 2400 gpm in the as-found system load condition vs. the IST test criteria.of 3000 gpm. The procedure pro-vides for the throttling of Loop B discharge. valves assuming initial flow greater than 3000 gpm but does not recognize an initial low flow condition. A safety evaluation was presented during PORC Meeting #85-115 on November 19, 1985,
.
O
.
_
which addressed the operability of ESW Loop B based upon the observed system flows. Data from previous system flow balancing during July 1984 preoperational testing was utili-zed, along with design pressure drops across ESW equipment loads.
Loads not in-service during the October 31 test included the ECCS room cooler ring header and HPCI room coolers. The licensee's assessment relied upon the observed increase in ESW Loop B system flow from 2400 to 3000 gpm on the morning of October 31, when an additional diesel cooler loop (engine D13) was added.
The PORC concurred with the conclusion of the evaluation that sufficient "B" (or "D") pump capacity existed to ade-quately supply the essential ESW Loop B loads. The PORC review of this evaluation was attended by the resident in-spector. No unacceptable conditions were identified.
5.2 Proposed Corrective Actions The licensee outlined additional actions to be undertaken, during meetings held on November 6 and 13, 1985. These are proposed to include:
5.2.1 Provision of more positive means of verifying manual valve conditions by visual observation of stem position including possible replacement of the upper. stem covers on all ESW and RHR SW pump discharge valves with a transparent cover.
using open/ closed marks.
5.2.2 Revision of the independent verification of restoration (IV0R) data sheet used in the ST-6-011-231 and 232 proce-dures.
The IV0R signatures for as-left valve positions would be listed on a separate page and more-clearly distin-guishable from the personnel performing the surveillance.
Similar revisions may be extended to other surveillance test IVOR methods. Also to be addressed is the timeliness of the IV0R, especially for surveillance.such as the ESW IST which requires more than one day to complete and yet does not require the system or loop. to be declared inopera-ble during the test.
5.2.3 Plant Management direction to Operations shift supervision regarding investigation of plant problems.
In the instance of the A0's' discovery on October 31 of the ESW Loop B dis-charge valves being apparently fully closed, no direction was provided to determine whether the valves were full or only partially closed, nor did shift supe'rvision proceed to the SPPH for a direct verification of that observed condi-tion. While.the immediate safety concern was addressed in that the A0 was directed to open the valves, the questions of how much the valves were throttled and how they cace to be throttled were not immediately considere.
.
5.2.4 Revision of ST-6-011-231 and 232 to include throttling of valves other than the pump discharge valves. Similar revi-sions may be extended to other surveillance procedures; however, preliminary licensee review has indicated that the ESW IST is the only IST which throttles valves which would not automatically-return to the required safe position if the system were to be automatically called upon to initiate.
5.2.5 Generation of a new routine test (RT) procedure, proposed as a quarterly reminder of Operations staff policies re-garding such topics as IV0R, Operating Aids, and manual valve controls.
5.2.6 Initiation of a new surveillance procedure, and annual A0 training on the proper method to check required positions of manually-operated safety-system locked valves.
NRC review of the completion of above proposed corrective actions is
~
considered an unresolved item (352/85-43-03).
5.3. November 13, 1985 Flow Simulation The ESW "B" pump was run from 2:20 p.m. until-3:30 p.m. on November 13, 1985, to simulate the test conditions of October 31. The '!A" control structure chiller was in-service at the time, and being fed from normal service water; therefore, the "B" control structure chiller received a minimum preset flow of ESW which was locally de-termined to be approximately 500 gpm. Normal Service Water flow was isolated (via a manual block valve) from ESW Loop B and the return path, as on October 31, was via the winter bypass lines to the spray pond. A normally-aligned system configuration was used with 2 of the 4 diesels lined-up, and no ECCS pump room coolers running. This flow check was witnessed by the Senior Resident Inspector, along with the Plant Manager and Operations Engineer.
The 11-0002 B discharge valve was. initially in the fully-open locked condition, with no stem ~ thread visible in the 20 inch high open space within the. collar. The pump was started and flow was stabilized at 2600 gpm and 120 psig discharge pressure. The discharge valve was then throttled closed until flow was observed to change from the sta-bilized 2600 gpm value. The change in flow occurred after the dis-charge valve was more than 90% throttled--an observed 1-1/2 to 2 inches of unthreaded stem remaining ~ visible (see Attachment 2)--at an observed flow of 2350 gpm and 123 psig discharge pressure. After sufficient time at this throttled condition, an extra diesel genera-tor (013) was added to ESW Loop B loads and flow was observed to in-crease to 2600 gpm at a slightly. lower discharge pressur.
.
.
After shutdown of the ESW Loop B pump, the discharge valve was momen-tarily taken to the full closed position to ascertain how much more stem travel would be required. The valve stem was observed to. travel about one inch before being full-seated. The stem was observed to show practically all thread in the full-closed position (see Attach-ment 2), except for approximately 1/2 to 3/4 inches of unthreaded -
stem surface.
The inspector concluded from the above that the A0 who initially reported fully closed discharge valves on the evening of October 31, could have observed valves throttled by as much as 90%, yet still capable of passing design flow rate in an ESW loop.
No violations were identified.
5.4 ESW System Maintenance Licensee review of corrective maintenance performed on ESW Loop B indicated that Maintenance Request Forms 85-5481 and 5482 were initiated between June 3 - 5, 1985 to repair two valves, HV11-044 and 074, whose seats were leaking. ESW Loop B was drained between May 29 and June 3, and the system was blocked and tagged under permit #11-0105.
Part of that block involved closing both the "B" and "D" pump discharge valves.
The block was cleared on June 5, and all tags were documented as be-ing removed by 4:55 a.m. on June 6.
An IV0R was completed at 6:17 a.m. on June 6, indicating that the discharge valves were open. The plant was in a cold shutdown condition at this time.
No later record of maintenance on the ESW system, which would have
affected these valves, was found. No unacceptable conditions ~were
-
identified, and the inspector concluded that the valves were returned to the open position on June 6, 1985,.and have.not been repositioned for maintenance since.
No violations were identified.
Technical Specification Implications and pRA Insights 6.1 Technical Specifications Technical Specification LCO 3.7.1.2 requires that during oparational conditions 1, 2 and 3, two independent ESW system loops shall be op-
.
erable. Each loop is to be comprised of two operable-ESW pimps, and a flow path capable of suction from the pump wet pits to assec.iated safety-related equipment. With one ESW loop inoperable, all diesels are to be aligned to the available-loop (after the confirmatory test-ing described in detail 3 of this report), and all other equipment aligned-to the'ir. operable' loop is to be declared inoperable. The i
~ inoperable loop must then be restored within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />, or the plant
'
is to be in hot shutdown within the next 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.
L
. - - - _ _ _ _
.
.
The loss of ESW Loop B also causes HPCI to be inoperable, along with two RHR and two Core Spray pumps. Similarly, the loss of ESW Loop A causes RCIC to be inoperable, along with two RHR and core spray pumps.
Since Technical Specification 3.5.1.C requires all_ low pres-sure ECCS systems to be operable if HPCI is inoperable, Technical Specification 3.0.3 would require action to be initiated within one hour to begin a plant shutdown. Therefore, although TS 3.7.1.2 alices for.an inoperable loop of ESW for up to 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />, the loss of ESW Loop B appears to require more restrictive action. The effect of the Limerick ESW system design for HPCI and RCIC room coolers is identi-fied as an unresolved item.
(352/85-43-02)
6.2 BNL Review and PRA Insights The NRC staff evaluated the Probab'ilistic Risk Assessment (PRA) con-ducted by the licensee for Limerick, and presented their review insights and safety conclusions in NUREG-1068 published in August 1984. Subsection 8.2.2 of that report. discussed dominant internal event contributors, and in particular the risk importance associated with, for example, a loss of HPCI or RCIC room. cooling.
Also, a report of a Technical Review conducted by Brookhaven National Laboratory (BNL) for NRC Region I in August 1984 at Limerick was pro-vided by-letter to the licensee on October 4, 1984. The review cov-ered ESW design, preoperational testing and operating procedures, and focused upon the ESW system because of its PRA importance.
The BNL review recommended a careful review of ESW valve alignments, especially to the diesel generators. Also the review recommended that a program be established for locking system valves. The locked valve program was implemented prior to fuel load, and was verified.in De-cember 1984. Additional valves have been included to the locked valve list in Administrative Procedure A-8 periodically since that date.
7.
Summary and Conclusions The following highlights the'significant findings and conclusion relative to ESW surveillance testing, from the licensee's event. followup and the senior resident inspector's independent review.
7.1 Summary of Findings 7.1.1 Quarterly in-service flow testing of the ESW pumps is difficult to duplicate due to the. variability of system
. alignments and equipment loads. This finding is discussed in detail 4.1.1.
7.1.2 Methods utilized to identify the required conditions of
-
manual valves, and methods of independently verifying.those conditions, are not clearly specified in the ESW IST proce-dures.
This finding is discussed in detail 4.1.2.
.
-
-
..
.
7.1.3 Full closure of both ESW Loop B pump discharge valves, from approximately noon until 8 p.m. on October 31, was unlike-ly. The cause and duration of their throttling, (more than 75% closed), was not determined as of the end of this~ in-spection.
This finding is discussed in detail 4.1.3.
7.1.4 An engineering evaluation of the flovs experienced on the
_
morning of October 31 concluded that--without knowing the exact position of the discharge valves but assuming them left in the throttled position (90% closed)--sufficient capacity existed with either the
"B" or "D" pump operating to supply design loads. The evaluation is discussed in detail 5.1.2.
7.2 Conclusions The inspection found that the ESW system Loop B discharge va.lves were open during the ST performed on morning of.0ctober 31, 1985 and demon-strated by measured flows experienced. The valves were assumed not to be repositioned following the morning test, as based on interviews with plant operators involved with the test.
The same valves were reported to be closed by a plant operator approximately eight hours after the test had been suspended at noontime on October 31. However, the operator did not attempt to verify whether.the valves were fully closed or in a throttled position. A subsequent flow check conducted on November 13, 1985, demonstrated that the valves could be throttled-by as much as 90% closed--a condition visually indistinguishable from fully closed-yet capable of passing the flows experienced on the morning of October 31. The operator who reported the valves as closed later stated that the condition he observed on the evening ~of October 31 could have been a 90% throttled condition. ESW Loop B inoperability wculd render half of all low pressure ECCS systems as well as HPCI inoperable, requiring a plant shutdown in accordance with Technical Specification 3.0.3.
A licensee safety evaluation of the operability of ESW Loop B on October 31 was presented to the PORC during meeting #85-115 on November 19, 1985. The evaluation was based on the as-left system lineup experienced during the test performed on the morning.of October 31, 1985,- and during which system flow was demonstrated. The analysis assumed that the valves were not repositioned after the morning test, and that the ESW Loop B discharge valves were suffi-ciently open to supply essential cooling loads. However, the dis-
.
charge valves were found in a position other than fully open as
-
required following surveillance testing in accordance with ST-6-011-232. This is a violation of System Operating-Procedure S11.1.A which requires an ESW system alignment with the pump discharge valves in a fully open position. (50-352/85-43-01)
.
,
L
8.
Unresolved Items Unresolved items are those for which more information is required to as-certain whether they are acceptable or constitute a deviation or viola-
'
tion. Unresolved items are discussed in details 5.2 and 6.:1 of this report.
9.
Exit Meetings The inspector met with the Station' Manager on November 6 and 13, 1985 to discuss the findings of.this inspection. Also present were the Superin-tendent of Operations and the Operations Engineer. At these meetings, the.
Station Manager and the Operations Engineer outlined the corrective action described in detail 5 of this report.
Also at these meetings, the inspector requested that the licensee identify any material which was discussed which may be considered proprietary in
,
nature.. None of the items discussed in this report. involve proprietary
'
information, nor was written material provided to the licensee.during this inspection.
,
I a
-
.
.. + -
-
.
l N]O C-67dlC t:':::.
d h
,
Loo,B-Sadnak SLJa-
,a
'
Ms ?(av (l ts, (3l23 p Ma2)
i
'
SPkW, Pond
,y y
(~)9Ne s-f~
"
o-gg ea se YY R* yyk" N j
,
V ff
.El42i 5%, ~
, saw
&
Cant:
u g,, g 9 r, cop 6 51Oky'
ED?*'
Smtos 0xe?.C cNo?beuawc# % '"
~~a v
,s Dies mc Iqwc pm l> (zaDbit]
!
,
HVG 200c p I
taw Cods l
"
.
ATTACHMENT 1 t
-_
f
,
I
-
.
.
-
'
,
.
_
,
i
. [
'
l q
t jA.
~
"
l
-
! k.j in 4,.
g.
'
-
-
.
l
,.
w..
-
.
ff j
'
-
.
,
L
,
e,
gi
^
.
,
l
-
.:
-.
.=
')t
,,, i 's
- <t er;e
'tlvr>
t
.
I' I I O^ h IN I Y
, y t.r
,1 3 1 -) ] tj
--