ML20237J612
| ML20237J612 | |
| Person / Time | |
|---|---|
| Site: | Brunswick  |
| Issue date: | 08/13/1987 |
| From: | Zimmerman S CAROLINA POWER & LIGHT CO. |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| References | |
| NLS-87-138, NUDOCS 8708180254 | |
| Download: ML20237J612 (5) | |
Text
-
1,
-Carolina Power & Light Company SERIAL: NLS-87-138
/UG 131987 10CFR50, A.ppendix 3 United States Nuclear Regulatory Commission ATTENTION: Document Control Desk Washington, DC 20555 BRUNSWICK STEAM ELECTRIC PLANT, UNIT NOS.1 AND 2 DOCKET NOS, 50-325 & 50-324/ LICENSE NOS. DPR-71 & DPR-62 LLRT OF ISOLATION VALVES IN NON-LOCA DIRECTIONS
Dear Sir:
' On October 17,1985, the NRC issued Inspection Report 85-31 for the Brunswick Steam -
Electric Plant regarding inspections performed on September 9-13 and September 23-26, 1985. During the course of these inspections, four valves were identified which were being local leak rate tested in non-LOCA directions, but were not properly identified as such in the study of valves tested in non-LOCA directions. As a result of the inspection,.
Carolina Power & Light Company committed to submit a report to the NRR which would include justification and evaluation of testing isolation valves in the reverse direction and include the four valves identified, if necessary. The testing procedure for the four valves identified in Inspection Report 85-31 has been revised so that the'se valves are tested in the LOCA direction and these valves have been successfully tested in the LOCA direction. Accordingly, valves Ell-F015A, Ell-F015B, E21-F005A, and E21-F005B do not appear in the attached report.
The issue raised in the inspection report referenced above will remain open with Region 11 pending NRR review. Please refer questions regarding this matter to Mr. Stephen D. Floyd at (919) 836-6901.
Yours very truly, Z'//facsc~
S.
. Zimm rman er Nuclear Licensing Section MAT /ppo (4091 MAT)
Attachment cc:
Dr. 3. Nelson Grace Mr. W. H. Ruland Mr. E. Sylvester a1%$b p
_,s,,....__
9
- m - mmmm - -mm
ATTACHMENT 1 TO SERIAL: NLS-87-138 LOCAL LEAK RATE TESTING OF VALVES IN NON-LOCA DIRECTIONS i
i l
l l
l l
(4091 MAT /ppo )
L.
t
. Local Leak Rate Testing of Valves
.+
in Non-LOCA Directions Summary inspection Report 85-31, dated October 17,1985, identified four valves (Ell-F015A, Ell-F015B, E21-F005A, and E21-F005B) which are local leak rate tested in non-LOCA'-
directions, but were not properly identified as such in the study of valves tested in the non-LOCA direction. Since the issuance of the inspection report, the Company has revised the testing procedures so that these valves are now tested with the test pressure applied in the LOCA direction. Therefore, these valves do not appear in this study.
There are a total of 58 valves currently being tested in non-LOCA directions at Brunswick. The following is a summary of the number and types of affected valves:
Valve Type Unit 1 Unit 2 Gate Valves 12 14 Globe Valves (Beneath Seat) 7 7
Globe Valves (Above Seat) 2 0
2 Butterfly Valves 6
6 Total 29 29 Gate Valves The following gate valves are currently being tested by reverse direction LLRT at Brunswick:.
B21-F016 E41-F002 B32-V22:
E41-F079 (UNIT 2)
B32-V30 E51-F007 Ell-F009 E51-F066 (UNIT 2)
Ell-F021 A G16-F003 i
Ell-F021B Gl6-F019 Ell-F022 G31-F001 These valves incorporate a wedge-shaped disc which seats on both seating surfaces simultaneously. The reverse direction LLRT verifies the leak-tight integrity of the seating surface furthest from primary containment. Therefore, the only remaining leakage concern is escape of containment atmosphere through the valve packing.
i' Six of these valves are located inside containment (B21-F016, Ell-F009, Ell-F022, E41-F002, E51-F007, and G31-F001). Any packing leakage would remain inside containment. The LLRT, as presently performed, verifies the integrity of the possible i
containment leak paths for these valves.
i l
Two of the gate valves (B32-V22 and B32-V30) are on 3/4-inch lines which connect to the I
reactor recirculation pump seals. As the stems on these valves are only 3/8-inch, a severe packing leak would create only a very small leakage path. These valves are located on the 20-foot elevation reactor building which is the entrance level to the building. Any packing leakage would readily be evident to operations personnel during
)
1 (4091 MAT /ppo )
4 shif t plant inspections. The normal operating pressure to which these valves are exposed is in excess of 1,000 psig, therefore, any leakage noted during operation of the plant should be of much greater magnitude than that which would occur at the peak accident pressure of 49 psig. Therefore, appropriate packing integrity would limit any leakage not identified during the LLRT.
Two gate valves (G16-F003 and G16-F019) serve as the isolation for the drywell floor and equipment drain sumps. These lines tie into the submersed sump pumps. Even when the sumps are pumped to the low-level point, the pumps still remain under water. A water seal from containment atmosphere is maintained and any packing leakage would consist of water in amounts well within the capacity of the reactor building drains to radwaste.
Of the 14 gate valves which are tested in other than the LOCA direction, only four (Ell-F021 A, Ell-F021B, E41-F079 [ Unit 2], and E51-F066 (Unit 2]) rely entirely on disc-to-seat seal packing integrity to ensure against gaseous leakage from the primary containment. These are the first valves off the primary containment. Two valves (Ell-F021 A, B) are on the lines which connect to the containment spray header and the other two (E41-F079 and E51-F066 [both Unit 2]) are on lines which serve as the HPCI and RCIC exhaust vacuum breakers. Testing in the LOCA direction is not possible since the valves are located on lines that open to the primary containment atmosphere.
Globe Valves (Beneath Seat)
The following globe valves are currently being tested by reverse direction LLRT at Brunswick:
B32-F019 B21-F022D B21-F022A Ell-F027A B21-F022B Ell-F027B B21-F022C These globe valves are pressurized from beneath the disc during the local leak rate test.
Despite being opposite from LOCA pressure direction, this is a more conservative direction because test pressure attempts to lift the disc from its seat rather than compress it into the seat. Therefore, leakage measured during testing is likely of greater magnitude than that which would occur during a LOCA. However, testing from beneath the seat isolates the packing from test pressure. Further justification for any packing leakage is provided below.
The following give globe valves are located inside the drywell:
B32-F019 B21-F022C B21-F022A B21-F022D B21-F022B Any packing leakage would remain inside the primary containment. Therefore, the LLRT as presently performed verifies the integrity of all possible containment leak paths for these valves.
(4091 MAT /poo )
Two valves (Ell-F027A and Ell-F0278) are located on the torus spray header. Testing is possible only in the. reverse from LOCA direction since the lines these valves are i
installed in open to the_ primary containment atmosphere of the torus. Since the packing on the valve is normally exposed to drywell pressure only and the valve is cycled only for testing, total packing failure is unlikely. Additic,al conservatism is provided since the LLRTs for these valves are conducted by pressurizing between the two isolation valves.
Therefore, recorded leakage is at least a factor of two greater than would actually occur during a LOCA. Finally, the integrated leak-rate tests verify the integrity of the packing approximately once per 40 months and test history has not indicated any leakage concerns from the packing of these valves.
Globe Valves (Above Seat)
These globe valves (E41-F079 and E51-F066 both Unit I only) are pressurized from above the disc during the local leak rate test. Although test pressure does not tend to unseat
.l the disc, the test does verify the disc-to-seat seal and the valve packing. As such, testing in this manner is preferred for this application. These valves are on the HPCI/RCIC turbine exhaust vacuum breaker lines. The configuration of the vacuum breaker lines is such that it ties into the turbine exhaust line between the torus penetration and the containment isolation valves for the exhaust line. Any leakage past the referenced valve seats will be bound by the second valve on the vacuum breaker line in addition to the two isolation valves on the exhaust line. Therefore, at least three barriers exist for LOCA pressure at the vacuum breaker line. Hence, packing integrity of these valves is of more consequence than leakage past the valve seat. The LLRT, as
- presently performed, does verify packing integrity and is the most appropriate test for i
l this application.
Target Rock Solenoid Valves The Target Rock solenoid valves are electric solenoid valves and have no leakage paths i
around the stem. Valve bonnets are one piece units which are welded to the valve body.
Valve stroking occurs electrica!!y through the valve bonnet. The only leakage path which exists for these valves is by the seat which is verified presently by the LLRT.
RNA-SV-5261 RNA-SV-5262 j
l Butterfly Valves These butterfly valves have an offset disc-stem arrangement which places the packing on the opposite side of the seating surface from that which LOCA pressure would apply.
Therefore, the packing is verified by the LLRT each refueling outage. Seat leakage is of no concern since butterfly valves are bidirectional and can be tested from either direction.
CAC-V5 CAC-V9 CAC-V6 CAC-V16 CAC-V7 CAC-V17 I
(409tMAT/ppo) i