ML20215G994
| ML20215G994 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 06/12/1987 |
| From: | Hukill H GENERAL PUBLIC UTILITIES CORP. |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| References | |
| 5211-87-5115, GL-87-06, GL-87-6, NUDOCS 8706230476 | |
| Download: ML20215G994 (4) | |
Text
.
i GPU Nuclear Corporation Nuclear
- eme ar8o-o s
Middletown, Pennsylvania 170$7 0191 717 944 7621 TELEX 84 2386 Writer's Direct Dial Number:
June 12, 1987 5211-87-2115 U. S. Nuclear Regulatory Commission j
J Document Control Desk Washington, D.C.
20555
Dear Sir:
Three Mile Island Nuclear Station Unit 1, (TMI-1)
Operating License No. DPR-50 Docket No. 50-289 GPUN Response to Generic Letter (GL) 87-06 NRC GL 87-06 dated March 13, 1987 requested that GPUN submit specific information in accordance with 10 CFR 50.54(f) concerning TMI-1 Pressure
. Isolation Valves (PIVs).
We were requested to provide a list of all PIVs in the plant along with a description of the periodic tests or other measures performed to assure the integrity of the valve as an independent barrier, acceptance criteria for leakage, if any, operational limits, if any, and the frequency of test performance.
THI-1 tests the following four Event V PIVs:
a.
CF-VSA and B (Core Flood and LPI tilting disc check valves to the Reactor Yessel).
b.
DH-V22A and B (LPI tilting disc check valves to the Reactor Vessel).
These valves are independently leak tested per T.S. 3.1.6.10. at a RCS pressure of 510 to 699 psig. Acceptance criteria is per T.S. Table 3.1.6.1.
Since the testing is performed at reduced pressure, measured leakage at the reduced differential pressure is adjusted to a functional maximum P.
The operational limits are per T.S. 3.1.6.10 and ASME Section XI, IWV-3420. The frequency of testing is:
a.
Prior to achieving hot shutdown following a cold shutdown of greater than 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> duration unless testing has been perforrr.ed within the previous 9 months, and b.
Prior to achieving hot shutdown after returning the valve to service following maintenance repair or replacement work.
8706230476 870612 PDR ADOCK 05000289
\\ (O P
pyg A
t k k()
GPU Nuclear Corporation is a subsidlary of the General Public Utilities Corporation
5211-87-2115 June 12, 1987 1
l In addition to CF-V5A and B and DH-V22A and B, two non-Event V PIVs, CF-Y4A and B (Core Flood Tank to Reactor Vessel), are independently tested for the pressure isolation function per ASME Section XI requirements. The failure of CF-V4A and B woJ1d not lead to a LOCA outside containment; therefore, they are not WASH 1400 Event V valves. Testing is performed at a RCS pressure of 510 to 699 psig.
Acceptance criteria is 5.0 gpm and measured leakage at reduced differential pressure is adjusted to functional maximum AP.
Operational limits and additional acceptance criteria are per ASME Section XI, IWV-3420 (1980 Edition with Addenda _ through Winter 1980). The frequency of testing for CF-V4A and B is the same as for CF-VSA and B and DH-Y22A and B.
The following TMI-l valves were previously determined not to be WASH 1400 Event Y valves and were not subject to the 1981 NRC Order. Their pressure boundary function is determined in combination but they are not independently
. tested for the closed function. These valves include:
1.
Motor operated gate valves DH-Y1 and DH-V2 on the Decay Heat Removal Drop Line. These valves are electrically interlocked closed when the RCS is i
>400 psig.
In addition, when the RCS is >450*F the breaker for DH-V2 is
{
open. This precludes the possibility of inadvertent actuation or
{
mispositioning of these passive valves. Leakage past DH-V1 and 2 opens relief valve DH-Y37 at approximately 495 psig.
DH-V37 is piped to the RC Drain Tank which is alarmed for high level and high temperature.
This provides assurance that DH-V1 and 2 leakage would be observed and operator action would pin point the source of the in-leakage to the RC Drain Tank.
In addition, both DH-V1 and 2 were disassembled in 1976 and 1983 for maintenance (see GPUN letter 5211-84-2143 dated June 13,1984 to J. F. Stolz from H. D. Hukill for additional details).
On both occasions the inspection showed that the pressure isolation capability of DH-Y1 and 2 was satisfactory. Therefore, TMI-1 experience and continued operational monitoring indicates that DH-V1 and 2 will continue to perform their l
pressure barrier function, 2.
High Pressure Injection (HPI) Valves a.
Tilting disc check valve MU-V95, tilting disc check valve MU-V107A and normally closed motor operated globe valve MU-V16A. Two of these three normally closed valves on the HPI Line to RCP A Cold Leg would be available to serve as PIVs.
b.
MU-V94 (normally open tilting disc check valve), MU-V220 (normally closed piston check valve), MU-Y107B (normally closed tilting disc check valve), and MU-V16B (normally closed motor operated globe 4
val ve). Two of these three normally closed valves on the HPI Line to RCP B Cold Leg would be available to serve as PIVs.
5211-87-2115 June 12, 1987 MU-V94 is a normally open check valve which could also close to serve as a PIV.
Similarly, normally o)en check valve MU-Y219 and normally open air-operated gate MU-V18 (w11ch spring closes on E.S. actuation) would also close to serve the pressure boundary function if required.
MU-V18 is leak tested each refueling in accordance with 10 CFR 50 Appendix J.
These normally open valves are not classified as PIVs as defined by GL 87-06.
c.
MU-V868 (tilting disc check valve), MU-V107C (tilting disc check valve) and MU-V16C (normally closed motor operated globe valve).
Two of these three normally closed valves on the HPI Line to RCP C Cold Leg would be available to serve as PIVs.
d.
MU-V86A (tilting disc check valve), MU-V107D (tilting disc check valve) and MU-V16D (normally closed motor operated globe valve). Two of these three normally closed valves on the HPI Line to RCP D Cold Leg would be available to serve as PIVs.
The following HPI valves are passive: MU-V16A/B/C/D, 86A/B, 95, 107A/B/C/D, and 220.
During normal plant operation there is no flow through these valves; they remain closed. Globe valves MU-V16s and the HPI check valves are of a different design; and therefore, are not subject to a common mode failure.
In addition, there are high pressure pump discharge check valves (MU-V73A/B/C) that can act as pressure barriers and there is a normally running HPI Pump which acts as a pressure barrier.
During 1980 all of the HPI check valves (except MU-V220 which was installed after 1980 and is of a different design and manufacturer) were disassembled to resolve manufacturing problems. Problems found in these valves (see LER 80-003/01T-1 & 2 and GPUN letter no. 5211-84-2143 dated June 13,1984 to J. F. Stolz from H. D. Hukill) were not attributed to normal service use, but to poor retaining device design and poor quality control during factory assembly of the valve internals.
These and all similar PIVs have been modified and the problems of poor retaining device design and poor quality control during factory assembly have been corrected by installing modified lock plate seat retention assemblies and one-piece hinge pins. GPUN believes the modified HPI check valves will perform their pressure barrier function.
All Makeup System PIVs experience infrequent and short duration flow and significant wear of parts is not expected.
3.
RC-V4 (normally closed globe valve) and RC-Y23 (check valve). Auxiliary Spray to the pressurizer.
Flow through the Aux. Spray Line is infrequent and of short duration and significant valve wear is not therefore expected.
t j
F 1
5211-87-2115 June 12, 1987 RC-Y4 is procedurally required to be closed when the RCS is >400 psig.
In addition RC-V4 does not automatically open; therefore these valves are passive. Leakage past RC-Y23 and RC-V4 opens relief valve DH-V67 at approximately 520 psig.
DH-V67 is piped to the RC Drain Tank which is alamed for high level and temperature. This provides assurance that RC-Y23 and RC-V4 leakage would be observed and then operator action would pin point the source of the in-leakage to the RC Drain Tank.
==
Conclusion:==
The leaktightness of the RC pressure boundary is verified by the performance of-our current reactor coolant system leakage calculations daily when the RCS is >525'F.
All of the above PIV's except RC-V4, 23 and MU-Y220 have been disassembled and inspected at least once and found to be operable.
MU-V86A and 107C were found to have experienced factory assembly problems which were corrected when the problems were identified.
GPUN believes the above is a complete list of PIVs as defined by GL 87-06 and that the current testing requirements and operational monitoring for the RC pressure boundary are adequate to assure safety. This completes GPUN's response to GL 87-06.
Sincerely,
. D.
u ill Vice President & Director, TMI-l HDH/MRK/spb:0893A cc:
G. Edison R. Conte W. Russell Sworn and subscribed to before me this /.24 day of Ont 1987.
Jim p. M f
Notary Public
$NAR04 P, bH04 GOIAkV PLADLIC MIDDLETOWN BOR0. DAUPMm COUNTY EY CCf4M45Ll0N EXPillis JUNE 12,1989 Ihmber, Pensy!Vcals Assoeurtion of Nctaries i.