Text

Informal,Interim Rept Recommendations to NRC on SER, Inservice Insp & Testing Program for 1970-1980 Period. Areas Covered Include Class 1,2 & 3 Insp & Testing of Pumps & Valves.Program in Compliance,As Modified by Rept
	ML19269C754
Person / Time
Site:	Perkins
Issue date:	12/31/1978
From:	Coppola A, Randy Hall, Restivo T; BROOKHAVEN NATIONAL LABORATORY
To:	Cheng C; Office of Nuclear Reactor Regulation
References
	CON-FIN-A-3117 BNL-NUREG-25451, NUDOCS 7902120151
	Download: ML19269C754 (37)
	v • d • e

IMITED DISTRIBUTION L-NUREG --25451 FORMAL REPORT RECOMMENDATIONS TO THE STAFF ON THREE MILE ISLAND UNIT #1 A. COPPOLA, T. RESTIVO AND R. E. HALL ENGINEERING AND ADVANCED REACTOR SAFETY DIVISION DATE PUBLISHED - DECEMBER 1978 DE?ARTMENT OF NUCLEAR ENERGY BROOKHAVEN NATIONAL LABORATORY UPTON, NEW YORK 11973 AL1C Research and Technica N Assistance ileport

}

Oc of Nucl or Re c or Re ulot on d 2./ ta d [j; Contract No. EY-76-C-02-0016 lb] OU)I$

7902120/S/

g

/

NOTICE Tlus reix rt w pequrett as an an ount of work aponmtett by the Unitril State.

Gmetnment. Neither the Uniteti States nor the Unitril.%tates Nutlear Rrgulatory Commiwnin, nor any..I t heir emplo> ren, nor any of t heir mnirat tors, sulx ontration, e.r their emplo>cca, makes any warranty, esprew or impliest, or awumes ans legal habihty or resinnubility for the an *acy, romph teness or uwfulness of any infor ma-tion, appas atus, pr<wim i or pnw css diu lomt, ea repic=-nti that its uw would not infringe privately ownr.1 rights

BNL-NUREG-25451 INFORMAL REPORT LIMITED DISTRIBUTION Recommendations to the Staff on Three Mile Island Unit #1 A. Coppola, T. Restivo and R.E. Hall Engineering and Advanced Reactor Safety Division Department of Nuclear Energy Brookhaven National Laboratory Upton, New York 11973 December 1978 Prepared for U.S. Nuclear Regulatory Commission Washington, D.C.

20555 Under Interagency Agreement EY-76-C-02-0016 NRC FIN No. A-3117

NOTICE: This document contains preliminary information and was prepared primarily for interim use. Since it may be subject to revision or correction and does not represent a final report, it should not be cited as reference without the expressed consent of the author (s).

TABLE OF CONTENTS Executive Summary..........................................................

1 In s e rvi ce I n.s pecti on P ro g ram............................................... 3 Pumps - Inservice Test Program.......................,.....................

9 Valves - Inservice Test Program............................................ 14 Conclusion.................................................................30 i

BROOKHAVEN NATIONAL LABORATORY RECOMMENDATIONS TO THE NRC STAFF ON THE SAFETY EVALUATION REPORT OF THREE MILE ISLAND #1 METROPOLITAN EDISON COMPANY INSERVICE INSPECTION & TESTING PROGRAM FOR THE 1977-1980 PERIOD (SUBMITTAL DATED AUGUST 17, 1977 and SEPTEMBER 30,1977)

Executive Summary At the request of the Nuclear Regulatory Commission's Division af Opm ating Reactors staff, the Reactor Engineering Analysis Group of Brookhaven hadonal Laboratory (BNL) has conducted a review of the Inservice Inspection and Testing program (ISI/IST) of the Three Mile Island #1 Docket No. 50-289. This evalua-tion is based upon the ISI/IST program as described in Metropolitan Edison Company's submittals of August 17 and September 30, 1977.

In addition to the above documents, a meeting was held with the management of Three Mile Island #1, the NRC staff and BNL, October 18 & 19, 1978 to review the requirements of ASME Section XI.

Mr. A. Coppola, T. Restivo, and R.E. Hall were the principals involved in this evaluation and have based their conclusions on numerous discussions with the NRC' staff so as to :chieve a program wide consistence of review.

The review covers five major areas and is sumnarized as follows:

1.

Inservice Inspection of class 1 - In this area, the licensees submittal is a departure from the ASME Code.

It is termed the " focused approach" and relies heavily on analysis to reduce the number of examinations planned. This was found to be unacceptable and it is recommended that the program for this area be disapproved and revised as soon as practical.

2.

Inservice Inspection of class 2 - In this area, the submittal meets the code both as to the number and types of examinations, except in five in-stances which were reviewed as relief requests. Of these, four are recommended for approval and one is an open item pending additional information.

3.

Inservice Inspection of class 3 - In this area, the submittal meets the code except for one relief request. This particular relief request is an example of the problems to be expected in older plants (pre 1974 code designs),

since it involves examinations of underground concrete pipes (material not covered by ASME Code), and the inspections on tests required to ascertain their condition.

In this case, while it is not possible to demand strict compliance with the code, it has been recommended that the licensee be requested to pro-pose alternate surveillance methods, since the piping is essential for plant safety.

~

1

4.

Inservice Testing of pumps - Again, in this area the problems with older plants are apparent. The program proposed follows the ir, tent of the code requirements, (monthly performance tests), but because of a lack of com-plete instrumentation, six (6) relief requests were reviewed. Of these, five (5) are recommended for approval and one is not.

5.

Inservice Testing of valves - This area far outweighs the others in number of relief requests. The review is divided into general and specific commentaries. The general subjects covered (6) were developed with the NRC staff and are applicable to all plants. They outline the accepted code inter-pretations. Of the remaini19 21 relief requests,16 are recommenced for accep-tance, 3 are recommended to be denied, and the remaining 2 are acceptable addi-tions and deletions to the valve program.

In summary, it has been found that the program, as reviewed and modified by this analysis is in compliance to the extent possible with the requirements set forth in Section XI of the 1974 Edition and Addenda through the Summer 1975 of the ASME Boiler and Pressure Vessel Code as required by 10CFR50.55a(g), ex-cept in the area of class 1, Inservice Inspection.

BNL has evaluated requests and recommended relief from specific require-ments which were determined to be impractical for this facility because of limited access, design, geometry, and materials of construction of some com-ponents.

Several other requests for relief from the require.ments should be denied.

Thie report includes the relief request specific evaluations that are recomm aded to be included in the NRC's Safety Evaluation Report on the stb-ject af ISI/IST for the Three Mile Island Nuclear Station Unit 1.

These recsmmendations are a result of the above described review and do not con-st stute a completeness evaluation of the program.

e 2

Three Mile Islano #1 Rec ">. dations for SER based on meeting cf October 18 and 19,1978 and e revie t of Licensee's Submittal dated August 17, 1977 and September 30, 1977 1.

INSERVICE INSPECTION PROGRAM 1.1 Class 1 Relief Request: The licensee has presented a unique program with signifi-cant deviations from the ASME code. By analysis, che licensee has chosen specific welds which he considers more likely to fail (because of higher stress and other considerations) and concentrates his inspection in these areas. He calls this the " Focused Approach" and requests the approval of the examinations called for his table A-1 in lieu of the examinat;ons called for in ASME tables IWB 2500 and IWB 2600.

Code Requirements: The major areas of deviation from the code tables IWB 2500 and IWB 2600 in Section XI are exceptions to the following re-quirements:

a.

ASME Items Bl.2, B2.1 and B3.1 examination Category B-B, rquires 10 percent of the longitudinal and 5 percent of the circumferential welds to be inspected during or near the end of each inspection interval (volumetrically) b.

ASME Items Bl.4 and 82.2, examination Category B-D, requires all noz-zles welis to be examined 100 percent during the inspection interval (volumetrically).

c.

ASME Items Bl.6 and B4.1 examination Category B-F, requires volumetric and surface examinations of 100 percent of all safe end welds during the inspection interval.

d.

ASME Item B4.5, B4.7 and 84.8 examination Category B-J (piping), re-quires 25 percent of circumferential joints including adjoining one foot sections of longitudinal joints (volumetric exam) and 25 percent of the pipe branch connection joints (volumetric exam over 6" and sur-face exam under 6").

Basis for Requesting Relief: The inspection program detailed in Table A-1 of the submittal follows the Code, except that inspections are focused on those areas which engineering analysis indicates are subject to relatively more critical conditions of stress, fatigue, radiation, and/or thermal cy-cle.

Inspections are also required of those areas which had recordable indications during the preservice baseline examination.

It is considered that inspection of areas subjected to relatively more critical conditions or which have pre-existing indications will provide good assurance of iden-tification of any potential problems before significant flaws develop in the Class 1 component pressure boundaries.

3

Fundamentally, the approach taken by Meted in regard to the inservice inspection program has always been that the inservice inspection effort should be directed at those areas of the plant which are most likely to de-velop problems, and that areas for inspection should not be selected in a random basis. Meted's reasons for taking this focused approach have been as follows:

By a more judicious selection of inspection locations, the effective-ness of the inspection is improved. For example, experience indicates that welds subjected to the highest fatigue and stress conditions are more likely to degrade than welds subjected to milder conditions in the same environment. 1.ikewise, experience indicates that defect growth of ten initiates at existing flaws. Accordingly, the focused aproach concentrates the planned ISI inspections on the higher stressed and fatigued welds and on areas with known flaws.

The use of the focused approach is consiaered to provide at least the same degrea of protection against undetected defect growth as the code approach, while requiring a reduced number of inspections. Meted ex-pects this to result in significantly reduced radiation emsure to personnel, which is considered to be highly desirable.

The MetEa ISI program for Class 1 components has always been barad on the focused approach.

It was originally developed in 1968 and 1969, and has been in the TMI-1 & 2 technical specifications since their original issue.

Over the years the program has been updated to include relatively minor changes to reflect new information. The type of information used to update the program has included final values of calculated stress and fatigue us-age factors, locations of recorded but acceptable indications in welds based on preservice inspection, and results of inspections at other plants.

Meted anticipates that further updating will be required in the future as experience at TMI and other plants is obtained.

Evaluation: The licensees program significantly reduces the number of ex-aminations required by the ASME code. The reductions in ce tegory B-D (noz-zie to vessel welds) for instance, are equal to 75 percent of the required examinations. These reductions are not accompanied by any iccreased amount of examinations in ar) cther category. The net result from tPis " Focused Approach" is less assurance that the Class 1 pressure boundary will not ex-perience weld failures, thac would be achieved if the ASME code required examinations were performed.

It is therefore recommended that this request for relief be denied, and th e tha '.icensee be directed to revise and re-submit the Class 1 ISI program as soon c practical.

1.2 Class 2 The program for ISI-Class 2, as outlined in the licensees submittal (table B-1, Revision 1) is in compliance with the ASME code both in the number and type of examinations planned, with exceptions noted in table B-2.

These exceptions are all acceptable as outlined below.

1.2.1 Relief Request: Penetration Pressurization System as shown on ISI DWG 300-015 be excused from hydrostatic pressure test requirements of IWC 2412 and IWC 2500.

4

Code Requirements: ASME Section XI, paragraphs IWC 2412 and 2500 re-quires that both exempt and non exempt components be hydrostatically tested at 1.25 times their design pressure, at least once during the inspection interval.

Licensee Basis for Relief: This system is an air system used for pressurization. The introduction of water for pressure testing will be harmful to the piping and other components and will impair and de-grade its subsequent operation. This system will be pneumatically pressure tested at normal operating pressure and leak checked by the

~

soap bubble method.

Evaluation:' The staff agrees with the licensee's basis for not intro-ducing water into this system. The licensee was requested to give ad-ditional justification for not testing at 1.25 times the system design pressure. Until this information is received, this should be con-sidered an open item.

1.2.2 Relief Request: The following systems be excused from the hydrostatic pressure test-requirements of IWC 2412 and IWC 2500 and an equivalent pneumatic test be performed instead:

.1 Waste Gas Disposal System

.2 Nitrogen Supply System

.3 Hydrogen Purge System

.4 Service Air System

.5 Containment Monitoring System Note: The system limits are givei, on page 2 of table B-2 of Licensee's submittal.

Code Requirements:

same as 1.2.1 above Licensee Basis for Relief: These are systems which contain gas. The introduction of water for pressure testing will be harmful to them and their associated components, and will impair and degrade their sub-sequent operation. These systems will be pneumatically pressure tes-ted to 1.25 times design pressure and leak checked using the soap bub-ble method.

Evaluation: The staff agrees with the licensee's basis and the test proposed is structurally equivalent to the ASME test (for gaseous sys-tems).

It is therefore recommended that this relief be granted.

1.2.3 Relief Request: Decay Heat piping (14") from DH-V6 A/B to Re-actor building sump be excused from testing per table IWC 2600, Item 2.1 and table IWC ?;20 Category C-F.

Code Requirement: The tables cited above call for volumetric ex-amination of 100 percent of selected pipe welds in pressure retaining piping, pumps, and valves in systems which circulate reactor coolant.

5

ihe welds selected include the following:

a.

circumferential butt welds at structural discontinuities b.

circumferential butt welds in piping with 3 pipe diameters of the ceaterline of rigid pipe anchors, or anchors at the penetration of the primary reactor containment, or at rigidly anchored com-ponents.

c.

longitudinal weld joints in pipe fittings (i.e., in tees, elbows, reducers).

d.

branch connection weld joints.

e.

pump casing and valve body weld joints.

This includes the weld metal and base metal for one-wall thickness beyond the edge of weld.

Licensee Basis for Relief: This piping is encased in concrete under the Reactor Building floor and therefore cannot be inspected volumet-rically. The butt welds on DH-Y6A and B cannot be inspected since these valves are located in a welded valve container and are not accessible.

Evaluation: The piping section for which relief is requested can be considered exempt from the requirements cited above for the following reasons:

1.

It does not normally contain or circulate reactor coolant.

In long term accident situations, it will be used to circulate re-actor coolant cc any other fluid which has been deposited in the reactor building sump.

2.

Under conditions of use, the highest pressure will be containment pressure, (below 200 psi), and below 275*F in temperature.

3.

It is also exempt from the hydrostatic pressure test requirements of IWC 5000 because it is an open ended system, open to the re-actor building sump.

The request is therefore considered within the intent of the code and no relief is required.

1.2.4 Relief Request: Decay heat piping elbow (10") immediately up-stream of DH-V4 A/B (one butt weld and two longitudinal welds), be excused from the requirements of table IWC 2600, Item 2.1 and table IWC 2520, Category C-F.

Code Requirement:

same as 1.2.3 above Licensee Basis for Relief: This section of decay heat pipe is con-tained within a 14 inch guard pipe and is not accessible for inspec-tion.

It will be subjected to hydrostatic test per IWC 500.

6

Evaluation: The staff agrees with the licensee's basis of inac-cessibility. Evidence of leakage during hydrostatic test is not re-stricted and will be regarded as sufficient cause for the removal of the guard pipe and closer inspection of the covered portion of the pipe and welds. The guard pipe and the weld attachments to the valve in this case should be considered under class 2 piping supports.

Accordingly, it is recommended tnat this relief request be granted.

1.2.5 Relief Request: The following pumps be excused from the hydrostatic pressure test requirement of IWC 2412 and IWC 2500:

1.

Decay Heat Removal Pumps A/B 2.

Building Spray Pumps A/B -

3.

Make-up Pumps A/B/C Code Requirement: same as 1.2.1 above Licensee Basis for Relief: Pressure testing pumps at 1.25 times sys-tem design pressure will damage pump seals Evaluation: These pumps have undergone proof pressure testing prior to operation or installation in the system. They are also under frequent visual surveillence during monthly inservice tests, and the elimination of this particular hydrostat (once in ten years) will not significantly increase the risk of failure of these pumps. Accord-ingly, it is recommended that this relief request be granted.

1.3 Class 3 The program for ISI Class 3, as out. lined in the licensee's submittal (table C-1) is in compliance with the ASME Code both in number and type of examinations planned, with the exception of certain under-ground piping as noted on table C-2.

These are discussed below.

1.3.1 Relief Request: Underground concrete piping in the nuclear service river water, decay heat river water, and reactor building emergency cooling systems be excused from the testing requirements of IWD 2410(b).

Code Requirement: The paragraph cited requires that all class 3 com-ponents be subjected to a hydrostatic test at 1.1 times the system de-sign pressure at least once every inspection interval (10 years).

Licensee Basis for Relief: These sections of piping are underground and therefore cannot be visually inspected. They also cannot be isolated with tig"t closing valves as required by IWD 2600(b). There-fore, during a pressure test, leakage through installed valves would invalidate the results of the test.

Evaluation: The ASME Code Section XI testing covering Class 1, 2 & 3 ISI is applicable only to those components constructed to the re-quirements of these classes outlined in Section III of the Code. The 1974 ASME Code for Class 3 does not cover components of concrete 7

construction. The 1.1x design pressure hydrostatic test is therefore not applicable in this instance. The intent of the code, to uncover leaks and potential breaks in the pipe is applicable however, and some alternate means of survei'llance should be proposed and implemented.

It is therefore recommended that this specific relief be granted, but that an alternate surveillance method be adopted.

O e

e 8

2.

PUMPS - INSERVICE TEST PROGRAM In general, plants constructed prior to the issuance of the 1974 ASME Codes do not have provisions for the pump tests required by that Code.

In particular, the flow measurement capability is missing in most cases. The staff expressed its view that the fact that a pump circuit did not presently include instru-mentation for measuring flow did not make flow measurement impracticable within the meaning of 10 CFR 50.55a(g)(4), unless the installation of such instrumenta-tion was impractical, and that each case would be considered on its own merits.

Accordingly, each relief request was disposed of as follows.

2.1 Relief Request:

Bearing temperature (T ) will not be measured, nor b

will lubrication level be observed on the following pumps:

a.

Reactor Building Emergency Cooling (RR-P1A, RR,P18),

b.

Screen Wash (SW-P1A, SW-PIB),

c.

Screen House Ventilation Equipment (SW-P2A, SW-P2B),

d.

Decay Heat River Water (DR-P1A, DR-P1B),

e.

Nuclear Service River Water (NR-P1A, NR-P1B, NR-P1C).

Code Requirement: The bearing temperature of all centrifugal pump bearings, and main shaft bearings of reciprocating pumps shall be measured at points selected to be responsive to changes in the tem-perature of the bearing.

Lubrication level or pressure is to be observed during testing.

Basis for Requesting Relief: These pumps are vertical deep well type pumps with the pump submerged under water at all times. Pump bearing are lubricated by the water being pumped. There are no installed means to measure bearing temperature, and the pump design / installation makes it impractical to measure bearing temperature by any other means.

Evaluation:

These pumps are constantly submerged during operation, and lubricated by the working fluid. The staff agrees with licensee that there is no practical means to measure bearing temperature. Also, the roquirement to observe lubricant level or pressure is not applicable to these pumps.

Therefore, it is recommended that relief be granted as requested.

2.2 Relief Request: Pump bearing temperature (T ) cannot be measured.

b a.

Control Building Chilled Water ( AH-P3A, AH-P3B),

b.

Spent Fuel (SF-P1A, SF-P18).

Code Requirement:

The temperature of all centrifugal pump bearings and main shaf t bearings of reciprocating pumps shall be measured at points selected to be responsive to changes in the temperature of the bearing.

Oil temperature prior to the oil entering a cooler shall be considered the bearing temperature.

Basis for Requesting Relief: Pump bearing temperature cannot be measured on this pump since the bearings are located deep inside the pump casing and are surrounded by an oil reservoir. An exception is 9

2

requested per 10 CFR 50, 50.55 a (g)(4) in that measurement of para-meter Tb is not practical within the limits of design of this pump.

Evaluation: Measurement of pump bearing temperature on these pumps is considered impractical based on their design. A monthly check on the lubrication level and signoff is presently practiced.

Based on this, it is recommended, that relief from Code requirement of measuring Tb be granted.

2.3 Relief Request: Vibration measurements will be made on the pump motor and not on the pumps for the following pumps:

a.

Reactor Building Emergency Cooling Pumps (RP.-P1A, RR-P1B),

b.

Screen Wash Pumps (SW-P1A, SW-P1B),

c.

Screen House Ventilation Equipment (SW-P2A, SW-P28),

d.

Decay Heat River Water Pumps (DR-PIA, DR-PIB),

e.

Nuclear Service River Water (NR-P1A, NR-PIB, NR-P1C).

Code Requirement:

The location of vibration measurement shall general-ly De on a bearing housing or its structural support, provided it is not seperated from the pump by any resilient mounting.

Basis for Requesting Relief:

These pumps are vertical deep well type pumps with the pump submerged under water at all times.

It is not pos-sible to measure vibration on pumps in this type installation. Past operating experience has shown that motor vibration is indicative of pump mechanical problems in this type installation. Therefore, motor vibration will be measured in lieu of pump vibration.

Evaluation: The staff agrees with the licensee's basis, and the proposed alternative testing is acceptable.

It is therefore recom-mended that this relief request be granted.

2.4 Relief Request: Flowrate (Q) cannot be measured during inservice tests of individual pumps, for the following pump sets:

a.

Screen Wash (SW-P1 A, SW-PIB),

b.

Screen House Ventilation Equipment (SW-P2A, SW-P28),

c.

Control Building Chilled Water, ( AH-P3A, AH-P3B),

d.

Makeup and Purification (MU-P1A, MU-PIB, MU-P1C),

e.

Nuclear Service Closed Cooling (NS-P1A, NS-PIB, NS-P1C).

Code Requirement:" Flowrate (Q) and AP, shall be measured in a variable resistance system.

In a fixed resistance system, pump AP or flowrate is required but not both.

Basis for Requesting Relief: There are no flowmeters installed in the flow paths ot"these pumps, therefore, flowrate (Q) cannot be measured.

An exception is requested per 10CFR 50.55a (g)(4) since measurement of flow would require a design change to this system, and therefore not be within the limits of the current design.

10

Evaluation:

2.4.1 Screen Wash Pumps (SW-P1A, SW-P1B):

These pumps are used to wash silt and small debris from the river water inlet screens. One of the two pumps is always in operation during normal plant operation with redundancy provided by the second pump. The pumps are typically automatically cycled at 15 minute intervals. Local alams are sounded if the pump is not started when cycling is commanded.

The licensee has proposed to measure the required pressure para-meters, and has stated that a visual check can be made (on a monthly basis) at the screens to determine if the pumps are perfonning as required to effectively wash down the screens.

During short and long term emergency conditions, the performance of the pump can also be visually monitored in this fashion at higher frequencies. Ultimate measures for washing down the screens could be hand held hoses.

For these pumps, since a qualitative visual check can be made of flowrate whenever these pumps are operated, it is recommended that the relief request be granted.

2.4.2 Screen House Ventil ; ion Equipment (SW-P2A, SW-P2B), Control Building Chilled Water ( AH-P3A, AH-P3B), Nuclear Service Closed Cooling (NS-P1A, NS-PIB, NS-PIC):

The licensee has indicated that these pumps are part of a variable resistance fluid system (control valves in operation). As such, the Code requires that AP and flowrate be measured and compared to reference values to evaluate the pump's perfomance. The licensee, in requesting relief from measuring flowrate as required by the Code, and has not proposed any acceptaole alternative to evaluating the pump's performance.

The licensee should propose an alternative method with tne rationale to support the method proposed. Until an acceptable alternative is available, it is recommended that the request for relief be denied.

2.4.3 Make-up and Purification Pumps (MU-PIA, MU-PIB, MU-P1C):

Discussions with the licensee pointed out that one pump is run continuously for normal make-up and purification during plant operation. The other 2 pumps are turned over and recirculated thru mini-flow loops every 30 days. The mini-flow loop did not provide for flow measurement and was estimated to flow at about 10 percent of pump capacity.

The licensee was asked if any of the subject pumps were tested flowing to the RCS during refueling outages. Personnel present at 11

the meeting believed some testing was accomplished at refueling, but were not sure of the specifics. The licensee was requested to, and agreed to provide the details of these tests. Until the specifics of these tests are available for evaluation, it is recommended that the request for relief be denied.

2.5 Relief Request:

Individual pump tests cannot be accomplished monthly in the manner required by the code, because the flowrate (Q) thru individual pumps cannot be measured, for the Nuclear Service River Water Pumps (NR-P1A, NR-PIB, NR-PIC).

Code Requirement: Measure flowrate (Q) as one of the parameters to be measured during individual pump tests conducted every 30 days.

Basis for Requesting Relief: Flow metering for this system is located in the common discharge lines from all three pumps. Plant operating requirements dictate t operation of at least two Nuclear Service River Water pumps during plant operations, thereby making it impossible to measure flow for a single pump. Pump flow will be measured for each pump during plant shutdown when operation of only one pump is required.

Evaluation: The licensee has indicated that operation of each pump could be checked to Code at plant shutdown, when operation of only one pump is required. The frequency of these individual pump tests was stated as being approximately one year.

In order to bring the test frequency more in line with the Code re-quirement of 30 days, the licensee was asked if perfonnance criteria and supporting rationale could be established for the 2 pump system operation using existing instrumentation. The licensee agreed to look into this type test approach or other alternatives for 30 day test-ing/ evaluation cycle.

Until this is accomplished, it is recommended that the request for re-lief be denied.

2.6 Relief Request: Pi (inlet pressure) cannot be directly measured, but can be determined by calculation for the Spent Fuel P. umps (SF-P1A, SF-P18).

Code Requirement: Measure Pi and AP (Reference ASME Code,Section XI, Sum. 75, Table IWP-3100-1).

Basis for Requesting Relief: Since there is no pump inlet pressure gauge installed, test quantities Pi and AP cannot be directly measured.

An exception is requested per 10 CFR 50, 50.55 a (g)(4) since measure-ment of pump inlet pressure would require a design change to this sys-tem and therefore not be within the limits of the current design. How-ever, pump inlet static pressure will be calculated based upon the dif-ference in elevation between f -..r taction and the source of pump suc-tion fluid.

12

Evaluation: Measuring the static head of a fluid on the inlet side of a pump Dy recording the height of fluid over the pump inlet is an acceptable alternative to a pressure measurement, where the reservoir is large in area compared to pump flow area.

It is therefore recom-mended that this relief request be granted.

13

3.

VALVES, INSERVICE TESTING PROGRAM

3.1 General

The scope of this review is limited to those valves which perform a safety related function. Safety related valves, for the purpose of IST, have been defined as those valves that are necessary to function to safely shutdown the plant and/or mitigate the consequences of an accident. As a minimum, all valves that receive a containment isolation signal or a safety injection signal shall be included in the IST program.

The following guidelines were developed after review of some initial IST programs.

3.1.1 Leak Testing of Yalves which Perform a Pressure Isolation Func-tion There are several sdety systems connected to the reactor coolant pressure boundary that have design pressures that are below the reactor coolant syste9 operating pressure.

It is required that there be redundant isolation valves forming the interface between these high and Icw pressure system to prevent the low pressure systems from Wing subjected to pressures which exceeds their de-sign limits.

In this role the vaives are performing a pressure 150iation function.

The staff considers it necessary to provide assurance that the condition of each of these valves is adequate to maintain this re-dundant isolation and system integrity. For this reason it is be-lieved that some method, such as a leak testing, should be used to assure their condition is sufficient to maintain this pressure isolation function.

In the event that leak testing is selected as the appropriate procedure for reaching this objective the staff believes that the following valves should be categorized as A or AC and leak tested in accordance with IWV-3420 of Section XI of the applicable edi-tion of the ASME Code. These valves are:

a.

DH-V22 A/B, Decay Heat Removal b.

CF-V4 A/B, Core Flooding c.

CF-V5 A/B, Core Flooding d.

RC-V4 or equivalent, Decay Heat Spray e.

RC-V23 or equivalent, Decay Heat Spray We have discussed this matter and identified the valves listed above to the licensee. The licensee has agreed to consider leak testing these valves in accordance with IWV-3420 of the applicable edition of the ASME Code and to categorized these valves with the appropriate designation.

In addition valves DH-V1 and DH-V2 on the Decay Heat suction line should also be considered in this category.

If after considering these valves for leak testing, the licensee finds that no leak testing is necessary, a detailed basis for the decision shall be provided to the NRC for evaluation.

14

In addition to valves at the junction of high and low pressure systems, there are valves which perform the same function between systems of equal operating pressure, but different ASME class con-struction. These should also be considered as performing a pres-sure isolation function since they prevent flow to a system or piping that is constructed and tested to a less stringent part of the ASME Cooe. Valves in this group should also be reviewed for possible inclusion in Category A.

For Three Mile Island Unit #1, the valves in the make-up system which interface with the reactor coolant system fit this description. Two out of 3 valves in each leg of the make-up system used for high pressure injection should be considered as performing a pressure isolation function and dealt with accordingly.

3.1.2 Containment Isolation Yalves In our review, we have found deviations from the containment isolation criteria set forth in General Design Criteria (G.D.C.)

and the requirements of Appendix J, particularly, the type C test.

We have determined that it is not in the scope of the IST program to evaluate the licensee's Appendix J program or the containment isolation criteria set forth in G.D.C.

The Appendix J review is a completely separate review, however, the detenninations made by that review are directly applicable to the IST program. The pres-ent IST submittal is acceptable, until the Appendix J review is completed, at that time, we will require the licensee to amend his IST program to reflect the conclusions of the Appendix J review for his plant.

In particular, the following valves are recommended for review and possible classification as containment isolation valves:

a.

PP-V101, Penetration Pressurization System b.

PP-V102, Penetration Pressurization System c.

PP-V133, Penetration Pressurization System d.

PP-V134, Penetration Pressurization System e.

NI-V27, N2 Supply f.

SA-V2, Service Air g.

IA-V6, Inst. Air h.

WDL-V304, L.W. Disposal

i. DH-V69, Decay Heat Removal *

j. DH-V64, Decay Heat Removal

3.1.3 Category A Valve Leak Check Requirements The staff's present position is that all Category "A" valves shall be leak tested to Section XI requirements. The leak test re-quirements and exceptions for Category A valves are explicitly stated in ASME Section XI.

In principle 10 CFR 50.55 a(g) Setion XI is separate and different from the requirements of other valve

These valves are additional to those on the list presented at SER meetiag.

15

testing requirements in the CFR such as Appendix J.

The test re quirements of 10 CFR 50.55 a(g) are to establish operational re-adiness at system function differential pressure.

In general:

a.

For Category A valves which communicate only with the con-tainment atmosphere, i.e. containment purge, hydrogen purge, Appendix J leak testing results are sufficient for Section XI requirements.

b.

For Category A valves which communicate with the primary cool-ant system, the licensee must perfonn the leak test at system function differential pressure. Relief to test at system function differential pressure are specified in Section XI and in those cases tests at lower pressure, such as those es-tablished for Appendix J requirements, are acceptable provided the licensee can satisfy sub-paragraph IWV-3420 C5 of Section XI.

c.

Those valves that perform both a pressure isolation and con-tainment isolation function shall be leak tested to meet Sec-tion XI of the applicable edition of the ASME Code in addition to Appendix J of 10 CFR 50 requirements.

3.1.4 Valve Exercising Requirements The ASME Code requirements for valve exercise tests for category A, B, and C valves allow certain deviations from the prescribed 3 month period if it is " impractical" to exercise the valves during plant operation.

It is the licensees opinion that these de-viations do not constitute relief from the Code requirements.

It is the staff position that these deviations must be reviewed in order to insure that proper and consistant criteria are used to determine impracticality (i.e.. where the failure of a valve in the test position would decrease the availability of a safety sys-tem). Accordingly, while there is no relief requested in these cases, a basis for the deviation and an evaluation will be included in this review, to document the criteria used to de-termine impracticality. When a valve is not exercised at 3 month intervals, it must be exercised at cold shutdowns unless relief is granted. A guideline used by the staff to define the duration of a cold shutdown during which valve testing is required is as fol-lows:

Valve testing should commence not later than 48 hours5.555556e-4 days 0.0133 hours 7.936508e-5 weeks 1.8264e-5 months after shut-down (reactivity 0 or negative), and continue until complete or plant is ready to return to power. Completion of all valve test-ing is not a prerequisite to return to power. Any testing not completed at one cold shutdown should be performed during suD-sequent cold shutdowns to meet the Code specified testing fre-quency.

In the case of valves exercised less frequently than cold shutdown (i.e., refueling), relief from the Code requirement must be re-quested. These cases are treated as such in this review.

16

3.1.5 Category E Valves The licensee submittal is deficient in that there are no category E valves listed in the IST program. A list of " locked" and

" sealed" in position valves was submittled at the SER working meeting of October 18 and 19,1978. This list is a general list for valves throughout the plant, and does not include those valves which are administratively locked in position.

In the next sub-mittal the licensee should include all the safety related valves which are locked in position either physically or by written procedure and are required to keep that position during an em-ergency. Th:se valves should be listed as Category E on the IST valve list by system as are the category A, B & C valves.

Although IWV-1300 of the 1974 Edition of Section XI of the ASME Boiler and Pressure Vessel Code excludes valves used for operating convenience and maintenance only from testing requirements, it is the staff's opinion and recommendation that any such valve which is in the nonnal or alternate flow path of cooling water of en-gineered safety systems, from the source to the reactor coolant pressure boundary or containment atmosphere, should be included in the valve testing program.

If the valve is normally locked open or closed, it should be reflected in the program and designated

" Category E".

This recommendation also applies to engineered safety systems which are designed to remove decay heat from the reactor core following a loss of coolant accident.

3.1.6 Corrective Action for Inoperative Valves Relief Request:

Inoperable valves will be evaluated using TMI No.

I technical specifications to detennine when an inoperable valve will limit plant start-up from a cold shutdown condition.

Code Requirement: ASME Section XI Paragraph IWV 3410 (g) and IWV 3520 (c) concerning corrective action required when a valve fails an exercise test, both state the following: "If the condition is not or can not be corrected within 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months , the valve shall be declared inoperative. When corrective action is required as a re-suit of tests made during cold shutdown, the condition shall be corrected before start-up. A retest showing acceptable operation shall be run following any required corrective action before the valve is returned to service."

Basis for Relief: Constraints and limitations on plant start-up with an inoperable valve depend on many specific plant design features and conditions. The limiting conditions for start-up and operation have been analyzed and are described in TMI No. I tech-nical specification, subsection 3, pages 3-1 to 3-60 (FSAR volume 5).

Evaluation: The staff agrees that the conditions and constraints set forth in the licensees technical specifications, in particular subsections 3.3 Emergency Core Cooling, Reactor Building and 3.2 Makeup and Purification and Chemical Addition Systems, give a more 17

detailed basis for preventing plant start-up than the ASME Ccode.

These bases have been previously reviewed and approved by the NRC and therefore it is recommended that this relief request be granted.

3.2 Particular Valve Relief Requests:

3.2.1 Reactor Building Spray System:

3.2.1.1 Category C valves:

Relief Request: Valves BS-V21 A/B, BS V30 A/B and BS V 52 A/B be excused from the exercising requirements of the ASME Code.

Code Requirement: ASME Sec. XI paragraph IWV 3520 states,

"(a) Test Frequency - Check valves shall be exercised at least once every 3 months, with the exceptions as defined in paragraph IWV-3520(b).

I, b) Exercising Procedure - Check valves shall be exercised to the position required to fulfill their function unless such operation is not practical during plant operation.

If only limited oper-ation is practical during plant operation the check valve shall be part-stroke exercised during plant operation and full-stroke dur-ing cach cold shutdown."

License? Basis for Relief: There is no practical way to test these normally closed check valves during reactor operation or shutdown.

In addition, the following objectives apply to particular valves:

BS-V30 A/B - Testing this valve in the manner suggested imposes undue risk on the plant in that water may be discharged from the Building Spray system spray nozzles.

BS-V21 A/P - Testing this valve in the manner suggested may tend to introw.a Sodium Thiosulfate into the Reactor Coolant System which contributes to corrosion and/or metallurgical problems.

BSOV52 A/B - The same situation exists with BS-V52 A/B which may tend to introduce Sodium Hydroxide into the system.

The maximum flow that can be obtained through a vent or drain con-nection is only sufficient to verify that the disc just leaves the seat.

It is felt that the possibility of introducing dirty water or sodium compounds does not warrant the performance of a test with insignificant results.

Evaluation: For all the valves listed above, there exists ar. up-stream test connection which should enable limited testing of these valves, (i.e., lifting off the seat or partial stroking).

Immediately upstream of these test connections, there exists a normally closed motorized valve which should allow testing without 18

the contamination problems implied above.

It is also noted that a test showing that these check valves are operable ana not " frozen" in their seated position is very desireable as compared to no tes-ting at all over the life of the plant. The licensee shou'd re iew these valves and propose such tests. After revaiew of these proposals, it may be necessary to grant relief from the exercising frequency or full stroke requirement, however entil such proposals are submitted, it is recommended that this relief request be de-nied.

3.2.2 Decay Heat Removal System:

3.2.2.1 Cateaory B valves:

Relief Request: Valves DH-V6 A/B be excused from the quarterly and cold shutdown exercising requirements and tested per Code at refueling.

Code Requirement: ASME Sec. XI papagraph IWC 3410 states,

"(a) Test Fraquency - Category A and B valves shall be exercised at least once every 3 months, with the exception as defined in IWV-3410(b)(1),(e), and ( f).

(b) Exercising Procedcre - (1) Valves shall be exercised to the position required to fulfill their function unless such operation is not practical during plant operation.

If only limited oper-ation is practical during plant operation the valve shall be part-stroke exercised during plant operation and full-storked dur-ing each cold shutdown: in case of frequent cold 9.utdowns these valves need not be exercised llore often than once every 3 months.

Normally closed valves that t annot be operated during normal plant operation shall be specifically identified by the Owner and shall be full-stroke exercised during each cold shutdown:

in case of freqecnt cold shutdowns these valves need not be exercised more often than once every 3 months.

Licensee Basis for Relief:

These valves are normally closed and allow the recirculction cf Reactor Building Sump water when re-quired under emergency ar accident conditions. The Reactor Buil-ding Sump contains watt.r that does not meet the cleanliness re-quirements of the decay heat system, and exercising DH-V6 will allow the entry of that water, thereby contaminating the Decay Heat System suction line, if tested during operation or at cold shutdown.

It also requires the isolation of the Borated Water Storage Tank which is not allowable during operation or cold shutdown.

Evaluation: The staff agrees that the potential contamination of the Decay Heat System is undesireable. These valves are not sub-jected to high temperatures or pressures during nomal operation and therefore it is recommended that this relief request be granted.

19

3.2.2.2 Category C valves Cold Shutdown Testing of Valves DH-V22 A/B Code Requirement: ASME Sec. XI paragraph IWV 352 states,

"(a) Test Frequency - Check valves shall be exercised at least once every 3 months, with the exceptions as defined in paragraph IWV-3520(b).

(b) Exercising Procedure - Check valves shall be exercised to the position required to fulfill their function unless such operation is not practical during plant operation.

If only limited oper-ation is practical during plant operation the check valve shall be part-stroke exercised during plant operation and full-stroked dur-ing each cold shutdown.

In case of frequent cold shutdowns these check valves need not be exercised more of ten than once every 3 months. Normally closed check valves that cannot te operated dur-ing nomai 11 ant operation shall be specificaly identified by the Owner and sna11 be full-stroke exercised during each cold shut-down.

In case of frequent cold shutdowns these check valves need not be exercised more often than once every 3 months."

Licensee Basis:

It is not possible to exercise low pressure injection valves DH-V22 A/B while the reactor is in nomal oper-ation.

Evaluation: The staff agrees with the licensee's basis, (low pressure pumps cannot provide pressure differential required for opening these valves except during certain acc' dent conditions),

and therefore the indicated testing meets the Code frequency re-quirements.

3.2.3 Core Flooding Systems 3.2.3.1 Category B valves Relief Request: Valves CF-19 A/B and CF 20 A/B be excused from the stroke time requirement of IWV 3410(c) 2.

Code Requirement: ASME Sec. XI paragraph IWV 3410(c) 2 states, "The stroke time of all power-operated valves shall be measured to the nearest second or 10% af the maximum allowable st*oke time, whichever is less whenever such a valve is full-stroke tested."

Li.ensee Basis for Relief: Air operated valves 2" and less have full-stroke times usually less than one second. Thus, the valve stroke time cannot effectively be measured using nomal test equipment such as a stopwatch. Air operated valves 2" and less will be tested to ensure their operational readiness as required by Section XI; however, the valve full-stroke time will not be re-corded for these valves.

20

Evaluation: The staff agrees with the licensee's basis, and the alternate test is acceptable.

It is recommended that this relief request be granted.

3.2.3.2 Category C valves Cold Shutdown Testing of Valves CF-V4 A/B & CF-V5 A/B Code Requirement:

same as 3.2.1.1 above Licensee Basis:

It is not possible to exercise core flooding valves CF-V4 A/B & CF-V5 A/B while the reactor is in normal oper-ation.

Evaluation: The staff agrees with the licensee basis, (these valves opened only during accident or emergency when Reactor pres-sure has dropped below core flood pressure), and therefore the indicated testing meets the Code frequency requirements.

3.2.4 Spent Fuel cooling System:

3.2.4.1 Category A valves Relief Request: Valve SF-V23 be excused from the exercise re-quirements of category A valves.

Code Requirement: The exercise requirements for category A valves are as given above in paragraph 3.2.2.1.

Category E valves need not be exercised. This valve actually is a dual category AE val ve. For these the Code is not explicit but footnote 1 of para-graph IWV 2110 states,

" Combination of categories, such as categories AC are to be used when more than one distinguishing category characteristic is ap-plicable.

In such mses, all requirements of each of the individual categor.;s are applicable although duplication or re-petition of common testing requirements is not necessary."

Licensee Basis for Relief: This valve is a passive (closed) con-tainment isolation valve. The valve will be leak tested, but since it does not change position during an accident or emergency, it will not be exercised for test.

Evaluation: The staff agrees with the licensee's basis. This valve should be reclassified AE, and excused from the exercise requirements of category A valves. Accordingly, it is recommended that this relief request be granted.

3.2.5 Make-up System 3.2.5.1 Category A valves Cold Shutdown Testing of Valves MU-V18, MU-V20, MU-V25, MU-V26 21

Code Requirement: Same as 3.2.2.1 above.

Licensee Basis: All of the above listed make-up valves are in constant operation providing makeup and letdown to the reactor primary system. The valves are not designed wit;1 a part-stroke capability and any error in the frequent part-stroke testing of the valve could jeopardize the continued safe operation of the re-actor coolant system. These valves will be tested per Code at cold shutdowns exceeding 48 hours5.555556e-4 days 0.0133 hours 7.936508e-5 weeks 1.8264e-5 months in duration.

Evaluation: The staff agrees with the licensee basis. Failure of these valves during a test while the plant is in opcration would reduce safety and in some cases cause unneccessary shutdown.

It is therefore agreed that the indicated testing meets the Code frequency requirements.

3.2.5.2 Category B valves Relief Request: Valve MU-V51 be excused from the stroke time re-ouirement of IWV 3410(c)2, Code Requirement: Same as 3.2.2.1 above.

Licensee Basis for Relief: Same as 3.2.2.1 above.

Evaluation: The staff agrees with the licensee basis, and the alternate test is acceptable.

It is recommended that this relief request be granted.

3.2.5.3 Category B valves -

Cold Shutdown testing of MU-V36 and MU-V37 Code Requirement: same as 3.2.2.1 above Licensee Basis: same as 3.2.5.1 above Evaluation: The staff agrees with the licensee basis.