ML19312E682

Forwards Info Re Implementation of 10CFR50,App J,Containment Penetration Testing Requirements,Per 800411 Request

ML19312E682
Person / Time
Site:	Prairie Island
Issue date:	05/30/1980
From:	Mayer L NORTHERN STATES POWER CO.
To:	Office of Nuclear Reactor Regulation
References
NUDOCS 8006090400
Download: ML19312E682 (8)

Text

0

~

NSF NORTHERN 5TATES POWER COMPANY M IN N E A f*O Ll e. M l N N E S OT A 95401 May 30, 1900 Director of Nuclear Reactor Regulation U s Nuclear Regulatory Commission Washington, DC 20555 PRAIRIE ISLAND NUCLEAR GENERATING PLANT Docket No. 50-282 License No. DPR-42 50-306 DPR-60 Information on Implementation of 10 CFR Part 50, Appendix J In a letter dated April 11, 1980 from Mr A Schwencer, Chief, Operating Reactors Branch #1, Division of Operating Reactors, USNRC, we were requested to provide information related to implementation of the contain:aent penetra-tion testing requirements of 10 CFR Part 50, Appendix J.

The information requested by Mr Schwencer is provided in the enclosure to this letter. Please contact us if you have any questions related to the information we have provided.

L 0 Mayer, PE Manager of Nuclear Support Services cc: J G Keppler G Charnoff Enclosure LOM/DHM/jh 8006090hf00

e Enclosure Director of NRR May 30, 1980 Northern States Power Company Response to Request for Additional Information, Prairic Island Station Units 1 and 2, Implementation of 10CFR50, Appendix J Containment Leakage Testing Item 2.1 Fan Coil Isolation Valves NSP stated that Fan Coil Unit isolation valves are in systems " sealed" from containment atmos phe re . To justify an exemption from type C testing requirements of Appendix J for these valves, provide the following information.

1. A description of how the system is sealed (e.g., safety class system designed to remain full of water following a postulated LOCA).

2. A description of how the seal integrity is verified (e.g. , hydro-static test to demonstrate that the system will remain full of water for 30 days following a postulated LOCA).

3. A discussion of how the " sealed" system will function to prevent atmospheric leakage.

Response

Fan coil units inside containment are provided with water from the plant cooling water system when they are operating in their safeguards mode.

Portions of the cooling water system serving the fan coil units are free from single failures, designed as Class I seismic, and are missile protected. Cooling water system pressure exceeds maximum postulated containment accident pressure. There is no potential for leakage of radioactive material out of the containment via the cooling water system.

The cooling water supply lines to the fan coil units are provided with a check valve inside containment and a manual motor operated gate valve outside containment. Return lines are provided with a manual motor operated gate valve inside and outside containment.

In the event of accident, the cooling water supply and return isolation valves remain open to satisfy their safeguards function. In the event of a fan coil unit or associated piping rupture the containment manual isolation valves would be closed to prevent the entry of non-borated water into containment. Pressure against the closed isolation valves is maintained by 1/2-inch equalizing lines. The water supply for this " seal" is provided by the cooling water system pumps (3 motor driven and 2 diesel driven) which take suction from the Mississippi River.

o Response (continued)

A lapse of integrity of the cooling water system piping or fan coil units inside containment would be indicated immediately by an increase in contain-ment unidentified leak rate. Leakage would also be evident during bi-weekly inspections of the containment. System p; essure tests are conducted in accordance with the Prairie Island ASME Code Section XI Inservice Inspection and Testing Program at least once each inspection interval (10 years).

Item 2.2 Extrapolation of Airlock Door Seal Leak Test Results The request for exemption from testing the airlock door seals at a reduced pressure within three days af ter being used is acceptable for operating reactors. However, the test results at this reduced pressure must be extrapolated to full pressure to determine the acceptability of the test.

Submit a description of the extrapolation method you propose to use.

Response

The Prairie Island Technical Specifications require airlock door leakage to be less than the design leakage of the door seals reported by NSP in Supplement No. I to the Initial Unit No. 1 Reactor Containment Building Leak Rate Test Report. The value reported was an arbitrary 1 cc/ min / lineal inch of resilent seal at test pressure (P = 10psig). There is no need to correct leakagefromtestpressuretopeakaccident pressure (P ) since the leakage acceptance criterion is not stated in terms of full,preasure.

An attempt to clarify this issue was made in NSP's Prairie Island License Amendment Request dated August 7, 1975. Refer to Exhibit A, Item 1(c). No action has been taken on this request by the NRC Staff.

If extrapolation were necessary, the following method could be used. If P is the gauge test pressure used and P is the gauge pressure the test results are to be corrected to, a c$nservative factor to apply to the leakage measured at P would be P,/P .

Item 2.3 Direction of Test Pressure NSP stated that there is no assurance that the testing of certain valves in a direction opposite to that of their safety function will produce results equivalent to or more conservative than results obtained from testing in the direction of the post-accident pressure. In that case, testing in the direction of the post-accident pressure will be required.

However, should these valves be analyzed such that you can provide a basis for determining that reverse direction testing will produce results equivalent to or more conservative than testing in the direction of the safety function, reverse direction testing would be acceptable. If a basis for justification of continued testing in the reverse direction cannot be provided, submit a plan of action to conform with the requirements of Appendix J, including a schedule for the accomplishment of the plan.

l l

Response

Table 1 lists the containment penetration leakage tests which are conducted ,

in the reverse direction.

l Testing of blind flanges involves pressurizing the piping between the inboard an6 outboard flanges. The inb ard flange is pressurized in the reverse direction. This is a conservac.ve test since test pressure acts to unseat the inboard flange. Testing of airlock overall leakage is, for the  !

same reason, a conservative test since the inner door is pressurized in the reverse direction which tends to open the door.

Testing globe valves in the reverse direction is acceptable if this results in applying pressure under the seat. Testing butterfly valves in the reverse direction is also acceptable if they are constructed for sealing in either direction. Reverse direction testing should also pressurize the

, valve stem seal (if any) or the integrity of the stem seal should be l' verified in some other manner. This position is consistent with the requirements of IWV-3423 of Section XI of the ASME Boiler and Pressure j Vessel Code, 1977 Edition, Summer 1978 Addenda, which is used in conducting the inservice inspection program valve leakage tests. Testing gate valves in the reverse direction is generally unacceptable.

Each of the penetrations tested in the reverse direction will be reviewed to determine if it meets the above criteria. Procedure changes or modifica-tions will be made to allow testing in the direction of post-accident pressure if they are found to be feasible. This review will be completed prior to the conduct of the 1981 refueling outage type B and Type C tests, i

i f

I l-

a Table I Containment Penetrations Tested With Dif ferential Pressure in Reverse Direction Penetration No.

(Unit 1 Designation) Description Type Valve or Barrier 9 Ilot leg to RHR inboard Motor operated gate valve (2) isolation valves 10 RilR inlet Motor operated gate valves (3) 11 totdown inboard Control (globe) valves (3) isolation valves 17 Loop B Sample inboard Control (globe) valve isolation valve 18 Fuel transfer tube Blind flange with double gasket 19 Service air Blind flange 20 Ins t rument air inboard Control (globe) valve isolation valve 25A Containment purge exhaust Air operated butterfly valve inboard valve 25B Containment purge supply Air operated butterfly valve inboard valve

28A Safety injection inboard Motor operated gate valves (4) t isolation valves Control (globe) valves (2) 31 Nitrogen to accumulators Control (globe) valves (3) inboard isolation valves 35 SI test line inboard Control (globe) valves (4) valves isolation 41A & 42B Containment inboard Air operated butterfly valve vacuum breaker valves 42B Containment inservice Air operated butterfly valve purge supply inboard isolation valve 42C & 42F Steam heat Blind flanges (6)

o Table 1 (continued)

Penetration No.

(Unit 1 Designation) De sc r ipt ion Type Valve or Barrier 43A Containment inservice Air operated butterfly valve purge exhaust inboard isolation valve 48 Low head safety injection Motor operated gate valve inboard isolation 49B Demineralized water Blind flanges (2)

Airlocks Airlock inner door Double gasketed door 1

e 8 e

Item 2.4 Water Testing of Isolation Valves

, NSP stated that the Technical Specifications at Prairie Island Station permit a large number of isolation valves to be hydrostatically tested in lieu of the pneumatic testing required by Appendix J and that where hydrostatic testing is utilized, an appropriate air / water correlation factor is applied.

This approach is theoretically acceptable if a justification for the j validity of the air / water correlation can be provided. However, since the j leakage involved is generally quite low and also since the leakage path characteristics are essentially unpredictable, it is extremely difficult to correlate water leakage to equivalent air leakage accurately. To date, no acceptable correlation has been demonstrated in this area.

Nevertheless, hydrostatic testing of these valves may be acceptable where the system is designed to remain intact and liquid-filled following a ,

postulated loss-of-coolant accident. For these systems, testing with water as a medium is more appropriate than testing with air on nitrogen since it i more closely approximates the post-accident environment. If hydrostatic testing is to be employed, justification must be provided that the available fluid inventory is sufficient to maintain a water seal on the isolation valves during and following an accident based on the leakage-rate limit of the test.

Response

The Prairie Island Technical Specifications currently allow water tests of the following penetrations:

4 RHR Supply and Return Charging Line Reactor Coolant Pump Seal Supply and Return Safety Injection Containment Spray Containment Sump ECCS Suction i Low Head Safety Injection All of these lines will remain water filled and intact outside containment

)

following a loss of coolant or steam line break accident. Present practice is to apply an air / water leakage scaling ' factor of approximately 60. This scaling factor effectively limits the permitted water leakage to a total of a few liters / min to permit the overall containment penetration leakage rate criterion to be satisfied. Leakage rates of this magnitude do not raise serious questions concerning available makeup inventory of any of the systems involved.

4 Item 2.5 3 raining of Systems for Type A Test Item 4 of Table 1, concerning Appendix J,Section III. A.1/d), states:

The primaryLsystem is vented to the containment atmosphere, but coolant is not drained to expose systems communicating with the primary system to the air test pressure. Each system is, however, subjected to Type C test if practicable.

o Item 2.5 (continued)

Although you indicated in your submittal that this item does not comply with Appendix J, no request for exemption has been submitted. Review the affected systems and submit requests for exemption, along with justification for each request, where the following requirements of Section III.A.1(d) of Appendix J are not satisfied:

1. Systems required to be drained to the extent necessary to assure exposure of the containment isolation valves to containment air test pressure and to assure they will be subjected to the post-accident dif ferential pressure.

2. Valves required to be Type C tested.

kesp'nse Sect ion III. A.l(d) of Appendix J requires ". .. portions of the fluid systems that are part of the reactor coolant pressure boundary and are open directly to the containment atmosphere under post-accident conditions and become an extension of the boundary of the containment shall be opened or vented to the containment atmosphere prior to and during the test. Portions of closed systems inside containment that penetrcte containment and rupture as a result of a loss of coolant accident shall be vented to the containment atmosphere". We believe this requires the draining and venting of those systems inside containment which may communicate with the post-accident atmosphere either through design.or due to failure of non-seismic or non-missle protected piping. Systems designed for the accident environment (such as the seismic, missile protected fan coil units and the secondary side of the steam generators) need not be drained and # anted.

Section III. A.l(d) does not require systems which are normally operating and filled with water following an accident to be vented to the containment.

Type C tests of isolation valves in these systems are required, however.

Prairie Isla.id does not conform to the requirements of Section III. A.l(d) since many of the isolation valve tests are oerformed using .ater and not air as required by Section III.C of Appendix J. As discussed in item 2.4, however, we believe water tests are more appropriate for these isolation valves. Also, as discussed earlier, tests are performed in some cases with pressure applied in a directior opposite to post-accident pressure. The validity of this testing will be reviewed as noted in our response to item 2.3.