ML20214J245

Forwards Response Addressing 13 Items Identified in NRC 860918 Request for Addl Info Re Inservice Testing Program. Encl 2 Provides Mod to 850816 Relief Request PV-17.Revised SER or Written NRC Approval Requested Re Testing Program

ML20214J245
Person / Time
Site:	Sequoyah
Issue date:	11/17/1986
From:	Gridley R TENNESSEE VALLEY AUTHORITY
To:	Youngblood B Office of Nuclear Reactor Regulation
References
NUDOCS 8612010225
Download: ML20214J245 (43)

Text

_-___--- __ ______ _______ _ _.___ - ____- _ _ _ _ _ _ -

15,

.r TENNESSEE VALLEY AUTHORITY CHATTANOOGA. TENNESSEE 37401 SN 157B Lookout Place Director of Nuclear Reactor Regulation NOV 17 886 Attention: Mr. B. Youngblood, Project Director PWR Project Directorate No. 4 Division of Pressurized Water Reactors (PWR)

Licensing A U.S. Nuclear Regulatory Commission Washington, D.C. 20555

Dear Mr. Youngblood:

In the Matter of ) Docket Nos. 50-327 Tennessee Valley Authority ) 50-328 SEQUOYAH NUCLEAR PLANT (SQN) - INSERVICE TESTING PROGRAM In response to your September 18, 1986 request-for information concerning SQN's Inservice Test (IST) program, TVA is submitting the following '

information to address each of the 13 items identified in your request. This supplements the information that was sent to you in the August 16, 1985, June 6, 1986, and August 28, 1986 submittals. This submittal should provide the additional information and justification requested for preparation of a revised safety evaluation report (SER) for SQN's IST program.

A thorough description of SQN's position regarding these 13 items has been included as enclosure 1. Enclosure 2 provides a modification to relief request PV-17 of SQN's August 16, 1985 submittal. Enclosure 3 includes several simplified drawings which supplement portions of enclosure 1.

As previously mentioned in the June 6, 1986 submittal, completion of the revised SER is requested to support SQN startup. If preparation of the revised SER cannot be completed before restart, a written response documenting NRC's review and concurrence with the SQN Pump and Valve Program would be beneficial. Upon receipt of either a revised SER or written NRC concurrence, SQN will change its pump and valve program to be consistent with NRC-approved position.

If you have any questions regarding this subject, please call M. J. Burzynski at 615/870-6172.

Very truly yours, TENNESSEE VALLEY AUTHORITY

/

R. ridley, irector Nuclear Safety and Licensing Enclosures p cc: See Page 2 f; 4 j F612010225 861117 PDR ADOCK 05000327 P PDR An Equal Opportunity Employer

r Director of Nuclear Reactor Regulation NOV 17 su cc (Enclosures):

U.S. Nuclear Regulatory Correission Region II Attn: Dr. J. Nelson Grace, Regional Administrator 101 Marietta Street, NW, Suite 2900 Atlanta, Georgia 30323 Mr. Carl Stahle, Sequoyah Project Manager U.S. Nuclear Regulatory Commission 7920 Norfolk Avenue Bethesda, Maryland 20814 Mr. G. G. Zech Director, TVA Projects U. S. Nuclear Regulatory Commission Region II 101 Marietta Street, NW, Suite 2900 Atlanta, Georgia 30323

-- ., -. ,,,.,-.,__----_--.-,...-c - . - . . - . , , . , . . - - - - - - , , , , , - - . . . - . . .

G t

ENCLOSURE 1 RESPONSE TO NRC'S REQUEST FOR INFORMATION CONCERNING 1 SEQUOYAH'S INSERVICE TEST (IST)

B. J. YOUNGBLOOD'S LETTER TO S. A. WHITE DATED SEPTEMBER 18, 1986

.1. DIESEL FUEL OIL TRANSFER PUMPS The ASME Section XI (Code) requirements for the diesel fuel oil transfer pumps include measurement of inlet and differential pressure. These requirements are imposed to provide trend data which will allow prediction of pump degradation. The following discussion will describe the existing equipment configuration and testing and how SQN's current

program meets the intent of the Code by

a. Ensuring pump performance is adequate to meet its accident / event mitigation role;

b. Ensuring that degradation is identified before the pump would fail to meet its flow requirements.

System Description

The diesel generator fuel oil system consists of four embedded storage tank assemblies (one for each diesel generator set) and associated pumps, valves, and piping. These tanks are embedded in the Diesel Generator Building substructure and provide each diesel generator set with approximately 68,000 gallons of fuel in reserve.

Two engine-mounted, motor-driven, 15 spm pumps are provided for each diesel generator set to transfer fuel from the embedded storage tanks to the two 550-gallon engine-mounted day tanks. Each of these pumps is capable of supplying fuel to both day tanks through nonisolable cross connections.

Two sets of level switches are provided for each day tank. The level switches initiate the transfer pumps and are arranged so that one pump will be the primary pump and the other a backup. ,

i Each pump has a capacity of approximately three times the rate of fuel consumption of the diesel generator. These pumps are powered from the diesel generator auxiliary boards. Thus, they will be powered if the associated diesel generator is accepting load.

The skid-mounted diesel engine fuel transfer pumps each take suction through flexible hose from the embedded storage tank. Their discharge lines combine into a common header which then splits to connect to the two-day tanks through flexible hose. This system has no provisions for measuring pump inlet pressure or pump differential pressure.

I

,,,v.,,--e.,,-w--.,-.,-m.-.-w.--,, y,-y- -y-,- ----ye ,--y-, _...v----,.y- - - , - - _ _ - . - -

=

4 Background / Requirements A review of testing records from May 1983 to July 1986 was performed to

~

identify cases of poor pump performance. Pump E2 was deficient at the initial test in May 1983. Subsequent testing on May 25, 1983 was performed and the pumps performance was verified to be greater than 10 gPm above the consumption rate. This deficiency was attributed to initial test methods and did not require pump maintenance.

In summary, these pumps have each met or exceeded the diesel generator fuel consumption by 10 spm without any failures through the 16 test intervals. This demonstrates that the pumps can reliably meet the stringent SQN test criteria.

Position / Current Testing SQN Surveillance Instruction SI-7 (SI-7) revision 35, " Electrical power System: Diesel Generators," is the procedure used to verify the operability of the diesel generators. Requirement 2.2 states that

". . . each diesel engine fuel oil transfer pump shall be demonstrated operable per specification 4.0.5 by measuring flowrate and vibration at least once per quarter." Specification 4.0.5 is the technical specification which imposes ASME Section XI testing.

With the diesel generator running and loaded to 4,000 kW, the instruction requires flow and vibration be measured for each pump while the other pump is off. The day tank fuel level change rate is used to calculate a flow rate. Each pump is required by the acceptance criteria to exceed the diesel generator consumption rate by greater than or equal to )

10 gpm. For each pump " acceptable" and " alert" vibration values are also specified.

To perform these tests personnel observe tank levels and vibration. Any unusual pump performance that could be visually detected would also be identified and would lead to corrective action.

This testing is supplemented by technical specification surveillance requirement SR 4.8.1.1.2.a which verifles tank levels and also requires exercising the fuel oil transfer pumps on a monthly basis.

Justification The purpose of ASME Section II is to provide a set of testing / surveillance standards for maintaining pumps and valves. The development of this code was a cooperative effort between the regulator and the regulated. Realizing that these requirements would not always be appropriate,10 CFR 50.55a paragraph a.3 allows alternatives, if authorized by the Director of the Office of Nuclear Reactor Regulation, i

o o

. This paragraph requires the licensee to demonstrate that (i) the proposed alternative would provide an acceptable level of quality and safety; or ,

(ii) compliance with the specified requirements would result in hardship l or unusual difficulties without a compensating increase in the level of i

quality and safety.

The code requires periodic measurement of pump inlet and differential pressures. The intent of this data collection is to allow trending of 4 pump performance. Ostensibly, this enables prediction of a point when the pump will no longer perform acceptably. l

$ As outlined in the system description above, the existing equipment l configuration at SQN does not include inlet or differential pressure instrumentation. As an alternative, SQN defined an acceptance criteria for flow rate which imposes considerable margin beyond the required flow from a single pump. Each pump is required to exceed the fuel consumption i rate by greater than or equal to 10 spm. Since the diesel generator's i fuel consuraption rate is five spm per set, the 10 spm required limit ensures that fuel consumption rate is exceeded by a safety factor of two.

It would presumably be acceptable to the Code to have a small margin and rely on trend data to detect pump degradation before performance is considered unsatisfactory. This method of operation is shown by j hypothetical curve A on figure 1 of enclosure 3. Using this method, sample points 1 and 2 would clearly be acceptable. At point 3 however, l an evaluation would'be required to determine the cause of the decreased i flow measurement for possible corrective action. Assuming that

! instrument accuracles were reliable and that the pump was, in fact, degrading at a constant rate, then continued operation of the pump without corrective action would in time (i.e., by point 4) result in unsatisfactory pump performance.

Curve B shows a similar pump performance trend. Assume at sample point 3 j a decrease in pump performance does not result in corrective maintenance. Using SQN's system, however, the pump performance must decrease by greater than 66 percent in one period before it would fail to meet the diesel generator fuel consumption. TVA's method results in corrective action for gradual pump degradation. It also provides enough margin so that more significant failures do not necessarily render the l

diesel generators incapable of meeting their safety function.

Conclusion Although system design does not provide for pump suction or differential i

pressure measurement, SQN has implemented alternative testing which

. provides an acceptable level of quality and safety. This methodology is simple and conservative and is further enhanced by technical

! specification surveillance requirement 4.8.1.1.2.a.3 and existing i

9

4

_4_

design redundance. Before a diesel generator would be deprived of sufficient fuel one of two redundant pumps would have to degrade by l greater than 66 percent during one quarter while the other was rendered inoperable. The operating data indicates that the pumps are highly j reliable and have not failed to meet the criterion in SI-7 fer pump flow. This stringent SQN criterion will lead to preemptive corrective

! maintenance and eliminate interpretation errors.

, Additionally, compliance with code-specified requirements would result in j unnecessary hardship - a plant modification would be required - without a j compensating increase in the level of quality and safety. Therefore, both criteria (i) and (ii) of 10 CFR 50.55a(a)(3) are satisfied.

l 2. TRENDING OF CONTAINMENT ISOLATION VALVE LEAKAGE RATES i

ASME'Section XI (the Code) paragraph IWV-3426 requires that leakage measurements be compared with previous measurements and with the

, permissible leakage rates specified by the plant owner for a specific '

i valve. The resulting trends are to be used to assess the current l performance of category "A" valves (i.e., containment isolation valves)

and predict their future performance.

I i SQN's current testing and maintenance practices use substantially target 4 leakage rates to provide assurance of containment isolation valve performance. The following discussion will describe the existing testing and maintenance practices, and how they meet the intent of the code by:

1 I

i a. Ensuring containment isolation valve leakage is maintained at satisfactorily low levels

b. Performing corrective uaintenance on valves to minimize containment leakage.

System Description

Each' unit at SQN has 207 containment isolation valves which are covered by the leakage surveillance program. These valves range from 3/8-inch valves on the postaccident sampling system to 24-inch vacuum breaker valves on the containment steel shell.

l SQN's containment leak test program, in accordance with 10 CFR 50 Appendix J ' consists of the following three types of tests:

Type Al tests are intended to measure the primary reactor a.

containment overall integrated leakage rate.

t 1Since type A leakage limits are totally separate from the type B and C j leakage limits, any future references to leakage limits or Code requirements

will apply only to type B and/or type C tests.

1 5

O-

b. Type B tests are intended to detect local leaks and to measure leakage across penetrations such as air-lock door seals, gaskets, electrical penetrations, resilient seals, etc.

c. Type C tests are intended to measure containment isolation valve leak 4

rates.

SQN's technical specifications require that leakage from all penetrations and valves subject to type B and C tests be limited to less than or equal to 0.60 L a. SQN's La is based on a Westinghouse ice condenser containment. L ais a leak rate equal to .25 weight percent of the containment air per 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> at 12 psig.

4 Background / Requirements

)

The SQN technical specifications place several requirements on the Operating License to ensure that the containment will be able to perform its safety function. The applicable specifications which implement 10 CFR 50 Appendix J are summarized below.

a. Limiting Condition for Operation (LCO) 3.6.1.1 requires the maintenance of containment integrity. To verify this, penetration closure devices are verified to be in the correct position and/or operable every 31 days. The personnel air locks are verified operable. Penetrations subject to type B tests must be leak tested if opened following a type A or B test, to ensure that the combined

! leakago limits for type B and C penetrations are met, j Where these requirements are not met, one hour is provided to restore containment integrity. When this condition cannot be met, the unit is required to be placed in cold shutdown.

b. Lc0 3.6.1.2 requires that leakage from all penetrations and valves subject to type B and C tests be limited to less than or equal to ,

0.60 L .

i In the event the type B and C leakage rates exceed .60 L.,

LCO 3.6.1.2 limits reactor coolant system temperature to less than 2000 F.

c. LCO 3.6.3 imposes requirements specifically for the containment isolation valves. The valve isolation times are checked against specific values contained in the technical specifications. This LCO also places specific leak test requirements on the purge isolation valves (i.e. , test within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of closing or every 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> for multiple cyclings). The leakage when combined with the other type B and C penetrations must again be less than or equal to 0.60 L a-

~

- - , - . - , --.,.,..-,-,-..--._,n-,. ,._-.,.,,-._,,,n__,, , _ , , , , . . - . . . ~ . . - n---__ .,.c,,-,,,,--.,-,.,,,,_,,,-.n ----.,,.,e ,,. - . , , ,._ , . ..,

o o

If any of the containment isolation valves are not operable, the action statement of LCO 3.6.3 allows four hours to repair the valve.

When the valve cannot be restored within four hours, the affected penetration must be isolated. For those penetrations where isolation is not feasible, the action requires placing the unit in cold shutdown.

Position / Current Testing / philosophy SQN has selected a total reference leakage of approximately 40 percent of 0.60 L a. Each containment isolation valve is assigned a specific reference leak rate such that the sum of the individual reference leak rates (type B and C penetrations) approximates 40 percent of 0.60 L .

This is a conservative value. During refueling outage testing, every reasonable effort (i.e., corrective maintenance / replacement) is made to maintain individual valve leakages below the reference value. During plant operations, SQN imposes an additional administrative restriction.

When leakage testing is performed during operation, individual valve leakage is limited to 20 percent of 0.60 L . Any penetration which exceeds 20 percent of 0.60 L is repaired before it is returned to service. This philosophy is formalized in the acceptance criteria section of SQN's Surveillance Instruction SI-158.1, revision 25,

" Containment Isolation Valve Leakrate Test," and reads as follows:

Each testable leakage path has been assigned a reference leak rate which is used to help ensure that the total allowable leakage is not exceeded. The individual reference leak rates are not to be considered as acceptance criteria, however all reasonable efforts should be made to maintain individual reference rates to ensure 0.6 La is met. A measured leak rate that execeds the reference leak rate is acceptable provided the Technical Specifications and Appendix J of 10 CFR50 are satisfied. In addition, maximum single penetration leakage shall, at all times, be limited administratively to 20 percent of 0.6 L . Sequoyah plant management (cognizant supervisory personnel) or a designated senior level engineer will determine whether a leakage path that exceeds its reference leak rate may be left as-is.

Any "As Left" leak rate that exceeds its reference leak rate will be noted on a deficiency los and included in the data package with justification provided and signed by one of the personnel noted above. These determinations shall be made by considering effects on overall leakage and possible effects on adjacent piping and components, as well as considering the time, cost, and radiological exposure required for corrective measures.

Any evaluation must adequately support any valve's continued ability to meet the 0.60 La technical specifications limit.

• e - - - - < ~ .w, -

,-----+---*-v,-

= _ - _ _ _ . _-

5 Justification

. ASME Section XI Article IWV-3426 requires the comparison of leak rate measurements to previous test values. This article is linked to Article IWV-3427 which requires corrt.ctive actions when trending projections indicate exceeding the permissible leak rates.

SQN's leak rate program simplifies the increasing leak rate trend t approach by imposing a procedural limitation on valve leakage. The following provides justification to support SQN's leak rate program in

, lieu of the trending requirements of the Section XI code.

4

s. Any valve which exceeds its assigned reference leak rate value would be identified for corrective action to return it below its reference value. Under this system of controls trending has no predictive value.

b. Trending of individual valve leak rates is meaningless since the reference leak rate for each of the 207 containment isolation valves is extremely low (e.g., .0225 scfh for a 3/8-inch valve up to 1.44 sefh for a 24-inch valve.).

c. The permissible leakage limit provided in IWV-3426 is 7.5D standard cubic feet per day where D is the nominal valve size in inches. The reference leak rate for SQN's containment isolation valves is approximately 1.44 D standard cubic feet per day. This margin of conservatism (1.44 versus 7.5) would result in corrective action long before trending projections would indicate changes in a valve's leak

rate performance.

I

d. For valves which experience significant changes in performance, it is impossible for trending to serve as a tool for corrective action since assigned reference leak rates provide conservative limits for triggering corrective action. A " single point" type of valve failure cannot be predicted either by the trend approach or by SQN's procedural method.

Conclusion As described above, SQN has developed a program which assigns specific reference leak rate values for each containment isolation valve. By i

procedure these valvec are tested and compared to very stringent acceptance criteria. These stringent criteria result in corrective actions before trend data would indicate a need for corrective action.

This system of controls is formalized by procedures required by the j technical specificaions. SQN' believes this program is simpler and more conservative and as such, is an effective alternative to trending and i corrective actions required by IWV-3426 and IWV-3427.

t i

e

- - - - , - - , , - - - - . ,,,,,a r,m-~~-,,--,,,,,_-w-,.-,-,-,-,--,--y.,,--,-,,,e-w.,n,-,

- . - - - - ~ ,

--v ,,n,,---,, , , , , - - , - - - - , - - , - - - , , , - - . - - -

. . _ - . -, _. -_ = _ _ . - . -. --

i

3. CORRECTIVE ACTION FOR VALVES FAILING EXERCISING TESTS m"

System Description

Pressure isolation valves (PIVs) at SQN consist of primary and secondary check valves on the emergency core cooling system (ECCS) injection lines, two power-operated valves on the upper head injection (UHI) test system piping and two power-operated valves on the residual heat removal (RHR) suction line. Technical specification 3.4.6.2 limits leakage from any PIV to one spm at a reactor coolant system (RCS) pressure of 2,235 i

20 psig. Testing is actually performed in mode 4 or 5 at reduced RCS pressures and the leak rates extrapolated to an equivalent leak rate at j 2,250 psia. With the exception of the RHR suction line valves, the PIVs

! are arranged such that often several valves are tested in parallel.

Additionally, the leak test system is such that numerous leak test lines 4

may feed into the leakage path. The result of these considerations is that measured leak rate reflects the leak rate from more than a single valve and the change in the assigned leak rate for a valve from surveillance to surveillance is not indicative of a change in the leakage i characteristics of the valve in question. (For illustration, reference SQN FSAR figures 6.3.2-1 and 6.3.2-15.)

Background / Requirements In the November 1982 SQN submittal for the Section XI Pump and Valve

{

Inservice Test (IST) Program, section 3.4, SQN requested relief from the corrective action requirements of articles IWV-3417 and -3423 of Section j XI of the ASME Boiler and Pressure Vessel Code. The basis stated was

-that requirements.for corrective action for components in safety systems are adequately covered in limiting conditions for operation (LCOs) contained in technical specifications. With these brief statements, NRC responded in the April 1985 safety evaluation report (SER) for the SQN Pump and Valve IST Program that the licensee had not provided adquate information or justification for not meeting the requirements of the

subject articles and that relief would therefore not be granted. The relief request and statement of basis were ill-presented in that technical specifications require equal or more restrictive than those

. specified in the Code; the requirements of IWV-3417 and -3523 are complied with. This clarification was provided in section 3.2.2 (enclosure 2) of SQN's August 1985 submittal. Additionally, SQN therein requested generic relief from the requirements of IWV-3427(b) for PIVs and provided detailed discussion of the basis for such relief. It is this relief request for which the following discussion is provided.

Position / Current TestinR/ Philosophy IWV-3427(b) addresses the corrective actions which must be taken as a result of trending of leak rates from Category A valves. Briefly stated, an increase in leak rate which reduces the margin between the measured i

4

,_ , _ _ _ _ . _ . . , ,__- ____ -- _ _,_ , _ ____ _ _ _ __ _ _,.._. _ _ _ , _ f _ __ ,,_ ._. _ _ -,_ __ _ _ ___

4 4

1 1

leak rate and the maximum allowable leak rate by 50 percent or greater requires doubling the test frequency, and a projected leak rate which would exceed the allowable limit by greater than 10 percent (based on three or more tests) requires replacement or repair. SQN is requesting relief from the requirements of IWV-3427(b) for PIVs (with the exceptiot.

of RHR suction line motor-operated valves) due to,the inability to assign a realistic leak rates to individual valves. A detailed discussion of the reasons for this problem and justification for relief is provided below.

As previously mentioned, PIVs at SQN are often tested in combination due to parallel system arrangements. When a combined leak rate is measured from more than one valve, (e.g., two check valves in parallel branch lines) the distribution of leakage between the two valves is unknown.

This means that the total leak rate for the two valves cannot exceed the 1 one gpm technical specification limit allowed for one valve since it is impossible to determine whether at worst case one valve'is leaking the full amount measured. The total measured leak rate for both valves is, therefore, assigned to each of the valves. Obviously tho assigned leak rates for the valves are not representative of the true leak rate characteristics of the individual valves.

The second factor is the arrangement of the leak test system. Tha leak test system at SQN consists of numeroes 3/4-inch test lines which tie into the main ECCS branch lines upstream of the primary and secondary check valves. All of the lines tie together, with air-operated globe valves isolating the individual test lines from the ECCS branch lines (reference FSAR Figure 6.3.2-1). Any leaka6e into the test system leakage path aligned for a valve under test, would create an indeterminate contribution to the measured leakage for the valve being tested. The actual amount of this additional leakage can vary from test to test depending upon the plant conditions, the use of the test system valves, and the pressure on the back side of the PIV(s) being tested.

These conditions further affect the assigned leak rates for individual valves by introducing numerous sources of potential indeterminate contributions to the measured leak rates.

The third f actor contributing to the problem of representative assigned leak rates is the effect of test pressure. Technical specifications require the leak rate from any PIV to be limited to one spm at an RCS pressure of 2,250 psia (2,235 1 20 psig). Due to operational and industrial safety considerations and the impact of necessary test configurations on LCOs (i.e., an ECCS supply valve closed to perform test which isolates both trains of the ECCS subsystem), testing of pIVs is conducted with RCS pressure well below 2,250 psia (typically 200 to 600 psig). This provides the test pressure for backseating the primary pIVs. Testing of the secondary PIVs is performed generally by use of l

l l

l l

, _ _ . . _ _..~f_._....-..__. _ , , _

. . . . = - .- .- - - - - _ -. .. - . .- -.

i  !

RHR pumps or cold leg accumulators, neither of which produce a test ,

l. ~ pressure much beyond 400-500 psig; the UNI secondary PIVs are backseated with the positive displacement pump for which speed is adjusted to 4

maintain a discharge pressure slightly less than RCS pressure so that ,

flow is not injected into the RCS. Testing of check valves in this manner (i.e., extrapolating leak rate data taken at low pressures to a corresponding leak rate at high pressure) is very conservative as there is no way to account for the additional closing (seating) force exerted on the CV by the higher pressure. Therefore, the extrapolation assumes the position of the CV disc relative to the valve seat stays constant as

? pressure is increased--no improvement in seating--and the leak rate increases as a function of the increased differential pressure across the j same size opening. The effect of test pressure can vary from test to

, test depending upon the specific test condition and the seating j characteristics of the valve.

Although each of these three factors result in adding conservatism to the assigned leak rate for each valve, each contribute to masking the actual leak rate for an individual valve. Experience has shown that leak rates j for PIVs at SQN exhibit leak rate histories which could indicate valve j improvement as often as valve degradation, i.e., leak rates for

! individual valves can change from test to test in either direction. For this reason, trending of leak rates for PIVs is meaningless for the

{ purpose of monitoring valve degradation.

4 Justification i The Code intent of monitoring valve leak rates is to assess operational readiness, detect component' degradation, and predict / project incipient valve failure so that preemptive repairs or replacement may be initiated before the valve reaching a point where it cannot perform its intended

function. Leak rate testing of PIVs ac required by technical specification 4.4.6.2.2 verifies operational readiness of both the individual valves and the associated intersystem pressure boundary by

applying conservative allowable limits for each PIV, (i.e., even if all PIVs leak one spm the total leakage rate will be well within the i relieving capacity of the associated relief valves). If a PIV is found to leak in excess of one spm, for example two spm, the valve is not at the point where either it or the pressure boundary interface cannot perform its intended function, the relief valve capacity is still well

', above the now assumed total leakage rate; however, preemptive repair or

! replacement must be taken in accordance with the technical specification and the intent of the code is satisfied.

I i

1

)

- _ - . . - _ , . _ . - , . - - . , - _ _..-.~m._--,,-_..._, - -,_- , ,.---._ __._-.. _ - -

I o .

j. .

1

- 11-4 Conclusion 4

In summary, the alternate testing proposed for PIVs, leak rate testing and performance of repairs / replacement, as required by technical specification 4.4.6.2.2, will meet the intent of article IWV-3427(b) and j is in accordance with 10 CFR 50.55a paragraph a.3 for alternative testing to show an acceptable level of quality and safety. Additionally, compliance with the requirements of IWV-3427(b) will provide meaningless and possible misleading information and can result in unnecessary increased frequency testing / repair / replacement of valves which are well within their operational requirements and are not experiencing real degradation.

4. TESTING OF VALVES. PREVIOUSLY OUT-OF-SERVICE SQN is seeking to eliminate an inconsistency in ASME Section XI. At issue is how to properly demonstrate that ASME Category A and B valves are operable. This relief request is sought from Article IWV-3416 so that a valve can be considered to meet IWV-3416 if it has been exercised in accordance with the frequency of IWV-3411. The intent of this request 4 is to make it clear that a Category A or B valve that is "in frequency"

! need not be retested unless it has been subjected to maintenance, repair,  ;

or modification which could alter its ability to function. l Background / Requirements Sequoyah technical specification Surveillance Requirement (SR) 4.0.5

requires that

Inservice inspection of ASME Code Class 1, 2 and 3 components and inservice testing of ASME Code Class 1, 2 and 3 pumps and valves shall be performed in accordance with Section XI of the ASME Boiler 4

and Pressure Vessel Code and applicable Addenda as required by 10 CFR 50, Section 50.55a(g), except where specific written relief has been granted by the Commission pursuant to 10 CFR 50, Section 50.55a(g)(6)(1).

Surveillance intervals specified in Section XI of the ASME Boiler

and Pressure Vessel Code and applicable Addenda for the inservice

~ inspection and testing activities required by the ASME Boiler and Pressure Vessel Code and applicable Addenda shall be applicable as

follows in these Technical Specifications

< e, r

4 M 2= hae C --d h e e er - w .a+wr'em --m--+vmi e* Meg--Yw M* Mew e- &-Je p- += e-ys-gy-e--m--m-+---rqyiv+ey?e Se. ypg- vga y-gwy y,tre ysu w w

• ypymwyyyw*- 9 -+%-

. I ASME Boiler and Pressure Vessel Code and applicable Required frequencies Addenda terminology for for performing inservice inservice inspection and inspection and testing testing activities activities Weekly At least once per 7 days Monthly At least once per 31 days Quarterly or every 3 months At least once per 92 days Semiannually or every 6 months At least once per 184 days Every 9 months At least once per 276 days Yearly or annually At least once per 366 days Technical specification SR 4.0.1 and SR 4.0.3 provide a clear statement of NRC's opinion of the necessity of testing to verify equipment operability. SR 4.0.1 states:

Sueveillance Requirements shall be met during the operational modes or other conditions specified for individual Limiting Conditions for Operation unless otherwise stated in an individual Surveillance Requirement.

SR 4.0.3 states:

Failure to perform a Surveillance Requirement within the specified time interval shall constitute a failure to meet the OPERABILITY requirements for a Limiting Condition for Operation. Exceptions to these requirements are stated in the individual Specifications.

Surveillance Requirements do not have to be performed on inoperable equipment.

ASME Section XI Article IWV-3411 " Test Frequency" states that Category A and B valves shall be exercised at least once every three months, except as provided by IWV-3412(a), IWV-3415, and IWV-3416. IWV-3412(a) provides relief from full-stroke testing where this is impractical.

IWV-3415 allows testing fail-safe valves during refueling outages when a quarterly test is not practical. IWV-3416 allows a cessation of testing if the associated system is inoperable or not required to be operable.

IWV-3416 goes on tn require a valve exercise test within 30 days before return of the system to operable status and a subsequent resumption of the test schedule.

Justification The technical specifications are incorporated into the SQN operating license. They impose minimum equipment operability requirements and incorporate the ASME Section XI requirements into the license. The general technical specifications in section 3/4.0 clearly require

1 .

)

that surveillance requirements must be met before equipment can be  ;

l. considered operable. This means that SQN pumps and valves must be tested l

in accordance with ASME Section II before they can be used to meet the

, requirements of the technical specifications. 1 i

I Section XI sets a general valve test frequency of three months. Relief from this frequency can be obtained for fail-safe valves and valves that l cannot be tested at power. Testing can also be deferred when a valve is i

in a system which has been taken out of service. This avoids extraneous testing of. equipment when it is inoperable or not required to be operable.

Section II Article IWV 3416 requires that a valve must be tested within j 30 days before its return to service. This is not consistent with the standard three month frequency. While IWV-3411 requires a test every 92 days, IWV-3416 can effectively increase the frequency to 30 days.

This occurs at the end of refueling outages when many systems have been out of service for maintenance. In these cases the testing must be

conducted during a 30-day window before restart. If the restart date is changed, the window moves and retesting is then needed to bring all valves within the 30-day limit. This has the undesirable effect of t motivating the test staff to wait until as close as possible to the restart date, to minimize their wasted efforts.

4 The 30-day requirement of Article IWV-3416 is also not consistent with the philosophy of the technical specifications. These specifications

! impose the survelllance requirements only when the equipment is required, and they only require that the surveillance be in frequency before equipment is returned to service.

i The inconsistency between IWV-3411 and IWV-3416 has apparently been identified by the American Society of Mechanical Engineers. They have addressed the problem in their most recent draft revision of the testing  !

standards, "American National Standard, In-Service Testing of Valves"  !

I ANSI /ASME OM-10-1986 Draft 8. Section 5217 of Draft 8 corrects the j inconsistency and reads as follows:

For a valve in a system declared inoperable or not required to be  !

operable, the exercising test schedule need not be followed. Within l three months prior to placing the system in an operable status, the  ;

f valves shall be exercised, and the schedule followed in accordance with requirements of this Standard.

{-

Based on the precedent of the technical specifications and Article IWV-3411, it is reasonable to assume that quarterly testing of Section XI valves providesuan acceptable level of quality and safety. Further, when restart schedules are moCified, the 30-day requirement can result in additional testing of equipment that is not yet required to be operable, i This extraneous testing is costly, and it is contrary to the ALARA j '

._, - . _ _ _ _ _ _ _ _ _ - - . . . . . _ - _ _ __.._1__.-_-_ _ _ _ _

t

- 14-principle of minimizing personnel exposure. As such,' the 30-day test i j frequency of IWV-3416 constitutes a hardship without a compensating ,

increase in the level of quality or safety. i conclusion f

The SQN technical specifications and the new draft ASME standard both provide equipment test frequencies. Both documents require the surveillance-to be "in frequancy" before equipment is considered operable. IWV-3416 is not consistent with IWV-3411, draf t standard

, OH-10-1986, or the technical specifications in that it imposes a much

! more restrictive test frequency on equipment that is being returned to

! service. To properly return Category A and B valves to operable status

, the valves need only be "in frequency" as defined by IWV-3411, and l satisfy technical specification surveillance requirements.

The lack of consistency with the other controlling documents, results in additional economic burden and increased personnel exposures without a corresponding increase in the level of quality or safety. For these reasons, SQN believes this relief request satisfies the criteria of 10 CFR 50.55a(a)(3) and an exemption from the 30-day requirement of ASME Section XI Article IWV-3416 is warranted.

5. PRESSURIZER AUXILIARY SPRAY LINE CHECK VALVE (62-661) AND POWER-OPERATED AUXILIARY SPRAY VALVE (62-84)

The following discussion provides a review of SQN's position regarding removal of the auxiliary spray line valve FCV-62-84 and exclusion of the auxiliary spray line check valve 62-661. Justification has been provided and is based on Article IWV-1100 which specifies the scope of the application of rules and requirements for inservice testing of certain class 1, 2, and 3 valves which are required to perform a specific function in shutting a reactor down or mitigating the consequences of an 1 accident.

System Description

i A flow path from the chemical and volume control system (CVCS) charging j line from the regenerative heat exchanger to the pressurizer spray line l provides capability of auxiliary spray to t.he vapor space of the pressurizer for backup RCS pressure control during cooldown. The subject valves within this flow path (62-661 and FCV-62-84) are ASME Class I.

(Reference FSAR figures 9.3.4-1 and 5.1-1) fmye- e--p--- ----i..--v--,-- rw -- , - - o, um- p----,--yy y--y----sp. --gw-g,,..-_+-i-,-.i-s .

9---,,e-e 7 -gipy 7y-q.-*- -- + r9y 9 -m . m y=.g -

y7 er w-

Background / Requirements FCV-62-84 was also evaluated during the referenced meeting, and the conclusion was reached that this valve did not perform a pressure isolation function and was not utilized as safety grade equipment to take the plant to cold shutdown. In a subsequent revision to SQN's pump and valve submittal dated October 8, 1982, SQN chose to add FCV-62-84 to the IST program. At that time SQN's pressurizer power-operated relief valves (PORVs) were not considered safety grade pressure control equipment. The reasoning at that time was that FCV-62-84 could provide a possible backup to the PORVs. FCV-62-84 was stroke tested under Surveillance Instruction 166.3 for both units from March 1984 until April 1986. Test data was collected and recorded in accordance with IWV-3400.

SQN has maintained that check valve 62-661 is nonsafety-related, was outside the scope of IWV-1100, and has, therefore, never included this valve into its ASME Section XI Pump and Valve IST Program. This valve was evaluated during the June 1982 NRC-TVA working meeting and determined at that time to be a nonsafety-related component. NRC identified in Appendix D of the April 1985 SER the subject check valvo for inclusion to SQN's IST program. In Section 3.9.1 (attachment 2) of SQN's August 1985 submittal,62-661 was determined to be nonsafety-related and was not added to the IST program. SQN's justification was that the loss of the auxiliary spray check valve would provide no compromise to safety or affect any safety analysis. In this response to NRC's September 18, 1986 letter requesting additional information, SQN is providing further justification for not including check valve 62-661 in their IST program.

Justification FCV-62-84 was added to SQN's IST program following the 1982 NRC-TVA working meeting at the suggestion of the licensee. At that time, SQN's pressurizer PORVs were not safety grade valves. The previous PORVs were air operated and would fail during a phase B accident when control air is isolated to containment. Since that time the PORVs have been changed out ,

to faster, solenoid actuated valves to meet the more stringent j requirements for cold overpressurization protection. FCV-62-84 was later l removed from SQN's IST program per the August 16; 1985 submittal. l Justification for removing this valve is as follows:  !

a. FCV-62-84 receives nonessential air and, therefore, does not meet single failure criteria (A train /B train air supply).

b. Following a phase B event, FCV-62-84 fails closed due to isolation of control air to containment making auxiliary spray unusable during the accident.

c. Charging and letdown is isolated upon a safety injection signal making auxiliary spray unusable until letdown and charging is reestablished during the postaccident recovery stages,

d. Normal sprays and pressurizer PORVs provide functional diversity and redundancy for pressure reduction capability in a postaccident condition,

e. FCV-62-84 does not serve a pressure isolation function.

The IST program addresses valves which are safety-related as defined in section 3.1.6 of the April 5, 1985 SER. SQN maintains that check valve 62-661 is not safety-related for the following reasons:

a. It does not form an interface between a high and low pressure system and, therefore, does not perform a pressure isolation function.

b. It does not serve any specific function in shutting down the reactor to the cold shutdown condition or in mitigating the consequences of an accident. (Reference similar justification for FCV-62-84 above.)

Conclusion The pressurizer spray line check valve 62-661 and the power operated auxiliary spray valve FCV-62-84 are both contained within ASME Class I piping but do not serve to perform a specific safety function. Loss of auxiliary spray would provide no compromise to safety or affect any safety analysis. For this reason these valves fall outside the scope of IWV-1100 and will not be included in SQN's IST program.

6. NORMAL CHARCING LINE CHECK VALVE 62-543

System Description

The normal charging line for SQN is supplied during normal operation by the centrifugal charging pumps (CCPs) which also serve to provide high head safety injection in the event of an accident through an alternate emergency core cooling system (ECCS) injection path (see FSAR Figure 9.3.4-1). The normal charging path is automatically isolated on receipt of an SI signal by two motor-operated valves aligned in series outside containment. The subject check valve,62-543, is located in series with the motor-operated valves on the normal charging line inside containment.

. . -- - . . . - -~ .- .- .

The penetration is continuously pressurized (outside containment) postaccident by the operating CCPs, (i.e., a high pressure water seal is indefinitely provided on the penetration preventing any leakage of containment atmosphere through the penetration to the environment). From this system description and additionally noting that this check valve  ;

does not perform a pressure isolation function, the only safety-related i function of this valve is to provide the inboard containment isolation i barrier as required by 10 CFR 50 Appendix A GDC 55. l Backtround/ Requirements l l This check valve has not previously been included in the SQN Pump and Valve IST Program. Appendix D of the April 5, 1985 SER on SQN IST Program for pumps and valves identifed this check valve for new inclusion to the SQN IST Program and for testing in secordance with Code requirements. The August 1985 TVA submittal (Section 3.9.4.2.1) provided a response to the SER and provided basis for past and future exclusion of this valve from the IST program. NRC has requested additional clarification of TVA's position regarding Section XI requirements for this check valve, and specifically the basis for not including this valve from the SQN Pump and Valve Program.

1 Position / Current TestinR/ Philosophy i

As reflected in SQN's Pump and Valve Program submittals, EG&G Idaho, Inc. , (original program reviewers for SQN) questions and answers, early NRC draft Regulatory Guide (MS 901-4) positions and section 3.1.4 of the

, -April 5, 1985 SQN Pump and Valve Program, all containment isolation valves that are Appendix J, type C leak tested must be and are included in the IST program for SQN. The subject check valve (although designated as the inboard containment isolation valve in accordance with 10 CFR 50 GDC 55) is not type C leak tested in accordance with Appendix J. Neither the outboard motor-operated valves nor the inboard check valve are subject to type C testing due to the water seal provided postaccident by the CCP(s) at a pressure greater than 1.1 Pa, with a 30-day water seal

. inventory provided by recirculation from the containment sump. This

, water seal precludes a release path, preventing any leakage of containment atmosphere through the penetration to the environment. Any i (water) leakage through the motor-operated valves would be into

, containment. The water seal is assured since at least one-CCP will

, always remain in service postaccident, and the seal will be provided even

with consideration of a single active failure. Accordingly this check i

~

valve is not needed to mitigate the consequences of an accident and, therefore, has not previously been included in the SQN Pump and Valve IST Program.

i l

p 4

-e . - , , - , - - .. . , - - n ,- - ---,.,,y m,-- - , - - - , - ~ - - - - - - .- - - -

Conclusion In summary, TVA does not believe that this check valve falls within the scope (IWV-1100) of Section XI since the valve is not needed to perform a specific function in shutting down the reactor or in mitigating the i

consequences of an accident. The containment isolation function for preventing release of containment atmosphere to the environment is provided by the above described water seal provisions for the SQN penetration and system design. Therefore, TVA believes the subject check valve should not be added to the Section XI IST program.

7. RCP SEAL INJECTION CHECK VALVES (62-560. -561. -562. -563)

System Description

The RCP seal injection lines for SQN are supplied during normal operation by the centrifugal charging pumps (CCPs) which also serve to provide high head safety injection in the event of an accident through an alternate (ECCS) injection path (see FSAR Figure 9.3.4-1). The seal injection lines are not isolated postaccident, but instead continue to be supplied by the CCP's to provide seal flow to the RCP's for postaccident availability. The subject check valves are located on each of the four seal injection lines inside containment.

The penetrations are continuously pressurized postaccident by the operating CCP's (i.e. , a high pressure water seal is indefinitely provided on the penetrations preventing any leakage of containment atmosphere through the penetration to the environment). From this system description and additionally noting that these check valves do not l perform a pressure isolation function,'the only safety-related function of these valves are to provide the inboard containment isolation barriers as required by 10 CFR 50 Appendix A GDC 55.

Background / Requirements These check valves have not previously been included in the SQN Pump and Valve IST Program. Appendix D of the April 5, 1985 SER on SQN IST Program for Pumps and Valves identifed these check valves for new inclusion to the SQN IST Program and for testing in accordance with Code requirements. The August 1985 TVA submittal (section 3.9.4.2.2) provided a response to the SER and provided basis for past and future exclusion of these valves from the IST program. NRC has requested additional clarification of SQN's position regarding Section XI requirements for these check valves, and specifically the basis for not including these valves from the SQN Pump and Valve Program.

l l

_ - - ._ __1

Position / Current Testing / Philosophy As reflected in SQN's Pump and Valve Program submittals, EG&G Idaho, Inc., (original program reviewers for SQN) questions and answers, early NRC draft Regulatory Guide (MS 901-4) positions and section 3.1.4 of the April 5, 1985 SQN Pump and Valve Program, all containment isolation valves that are Appendix J, type C, leak tested must be and are included in the IST program for SQN. The subject check valves (although designated as the inboard containment isolation valves in accordance with 10 CFR 50 GDC 55) are not type C leak tested in accordance with Appendix J. These inboard check valves are not subject to type C testing due to the water seal provided postaccident by the CCP(s) at a pressure greater than 1.1 Pa, with a 30-day water seal inventory provided by recirculation from the containment sump. This water seal precludes a release path, preventing any leakage of containment atmosphere through the penetration to the environment. The water seal is assured since at least one CCP will always remain in service postaccident, and the seal will be provided even with consideration of a single active failure.

Accordingly, these check valves are not needed to mitigate the consequences of an accident and, therefore, have not previously been included in the SQN Pump and Valve IST Program.

.t Conclusion In summary, SQN does not believe that these check valves fall within the scope (IWV-1100) of Section II since the valves are not needed to perform a specific function in shutting down the reactor or in mitigating the

~; consequences of an accident. The containment isolation function for preventing release of containment atmosphere to the environment is provided by the above described water seal provisions for the SQN penetration and system design. Therefore, SQN believes the subject check valves should not be added to the Section II IST program.

i

8. REACTOR COOLANT SYSTEM CHECK VALVE.68-559 TO THE PRESSURIZER RELIEF TANK Paragraph IWV-3522 imposes requirements for exercising class 1, 2, and 3 safety-related check valves. NRC, by their letter to TVA dated September 18, 1986, requested additional justification for our request for relief from IWV-3522 for check valve 68-559. It is SQN's position that quarterly exercising the subject check valve during operations or 4

cold shutdown is hazardous to test personnel and requires complex and unusual system configurations which places the plant in a degraded condition. In consideration of NRC's concern for valve operability, 1

1

- - . - - - y , , , ..,,,-,_,,,y,.,_.. -

. , . _ , , , - . , , - , , - , - , ~ . , , ,

. t l

SQN is willing to compromise the hardships involved in testing 68-559 and

- is requesting relief to disassemble and inspect 68-559 every third refueling outage beginning with cycle 3 for both units. This constitutes a change to the existing relief request PV-17 of the August 16, 1985~

submittal to add an alternate test and frequency. (Reference enclosure 2 for amended relief request PV-17.)

System Configuration Each of the ECCS and containment spray pump suction and discharge relief valves located within the Auxiliary Building relieve into a single l four-inch line that penetrates containment and merges with other relief valve discharge lines. All of these relief valves are connected by common piping to the pressurizer relief tank (PRT). The PRT is a quench tank designed to accept steam flow from the pressurizer and quench it in primary grade water. The tank is provided with level and pressure instrumentation and is equipped with spray and drain capabilities to

maintain nominal level and pressure. A blowout dine provides overpressure protection at 85 psig.

check valve 68-559 located on the inboard side of containment is a Copes Vulcan Model C42 swing-type check valve within ASME class II piping. The valve is normally closed with backseat pressure limited to 85 psis. In the closed position it precludes leakage out of containment. The valve is required to open to provide a relief path from the emergency core cooling and containment spray relief valves in the Auxiliary Building.

. There are'3/4-inch preoperational test connections upstream and downstream of check valve 68-559.

l The downstream side of the valve is' subjected to discharge from the following:

A. Pressurizer Code Safety Valves (3) - lift setpoint, 2,385 psig B. Pressurizer PORVs (2) - lift setpoint, 2,335 psig C. Residual Heat Removal Pump Discharge Relief Valve - lift setpoint, 1 600 psig -

D. - Residual Heat Removal Pump Suction Relief Valve - lift setpoint, 450 psig E. CVCS Letdown Relief Valve - lif t setpoint, 600 psig ,

F. CVCS Seal Return Relief Valve - lift setpoint, 150 psig 4 G. Reactor Head Vent Valves - lif t setpoint, manual H. Miscellaneous Valve Leakoff Lines

{

The upstream side of the valve is subject to discharge and must pass flow from these additional sources:

e l

l

A. Containment Spray Pump Suction Relief-Valves (2) - lift setpoint, 100 psig B. Safety Injection Pump Discharge Relief Valves (3) - lift setpoint, 1750 psig C. Centrifugal Charging Pump Suction Relief Valve - lift setpoint, 220 psig D. Safety Injection Pump Suction Relief Valve - lif t setpoint, 220 psig E. Residual Heat Removal Pump Discharge Relief Valves (2) - lift setpoint, 600 psig This configuration is shown on FSAR figures 5.1.1, 6.2.2.2, and 6.3.2.1.

A composite sketch has been provided in figure 2 of enclosure 3.

Background

NRC identified check valve 68-559 in their April 1985 Safety Evaluation Report (reference item 1 of Appendix D) for inclusion into SQN's IST Program. Subsequently, in the August 1985 submittal, this valve was added to SQN's IST program. Relief was requested from the IWV-3522 requirement of exercising 68-559 quarterly at full flow. The basis for this relief (reference PV-17) stated that a full flow exercise would not be possible due to the nature of the system and that a part stroke exercise presented unacceptable personnel hazards unless various systems required for operation were removed from service. Removal of the affected systems from service would require entry into LCOs and an overall degraded plant condition regardless of operating mode.

An alternate test was not proposed due to the fact that complex plant manuevers would be required and unnecessary safety risks would be imposed on test personnel.

ASME Section XI Article IWV-3522 Requirements ASME Section XI Article IWV-3522 requires a quarterly test to verify that a normally closed check valve's disk moves away from the seat when:

A. The closing pressure differential is removed and flow is initiated, or B. A mechanical opening force is applied.

Confirmation of disk movement is to be by:

A. visual observation, or B. electrical position indication, or C. system pressure indications, or D. other positive means.

, If the test is made without flow, the mechanical force used must be limited to 10 percent of the equivalent force represented by the minimum emergency condition pressure differential acting on the valve, or 200 percent of the force required to open the valve when it is new.

l Paragraph IWV-3522 requires that check valves be full stroke exercised quarterly during plant operation, "unless such operation is not practical." The article goes on to define less stringent requirements in an effort to make the testing requirements reasonable when quarterly, full stroke tests are not practical. Generally, as a minimum, the Code requires a full stroke test during cold shutdown (mode 5) unless three months have passed since the last shutdown exercise.

Justification A. Factors Supporting Reliability of CV-68-559 The intent of ASME Section XI paragraph IWV-3522 is to assess operational readiness and detect check valve degradation or failure. For check valve 68-559, the Code-required test would demonstrate the valve's ability to provide a relief path from

, associated safety systems. Typically, check valve failures result 3: from excessive corrosion, flow or vibration induced wear, or vibration induced mechanical failure of subcomponents. Corrosion is not expected to occur since this is a stainless steel valve in a primary grade water system with chemistry limits defined in the technical specifications. The valve is located on a four-inch

, relief header which is not' expected to experience any flow under normal conditions. The valve should alco not experience any backseat forces, however, if a downstream relief valve did to actuate, backseat pressure would be limited to 85 psig by the PRT blowout disk. If any one of the relief valves upstream of CV-68-559 operated, it would subject 68-559 to an opening force of 1,000 to 18,000 lbg.

B. Difficulties Involved in Exercising CV-68-559 As shown in figure 2 of enclosure 3, CV-68-559 is subject to

discharge from 19 relief paths with setpoints ranging from 50 to 2,485 psig. SQN's concern is that without isolation valves, testing CV-68-559 would be a very complex activitly requiring highly unusual system configurations and would pose unacceptable personnel hazards in modes 1 through 5.

1 (1) Full Stroke Testing

i. Flow Test This test method requires establishment of maximum flow through the valve to fully open it. Typically, the tested valve is in a normal or injection flow path and an existing pump can be used to provide the maximum required flowrate. Generally, existing instrumentation is used to verify flow or test connections can be used to connect temporary instruments.

Check valve 68-559, however, is within an abnormal configuration and is integral to a four-inch pipe that is a common discharge header for nine one-inch, two-inch, and three-inch relief valves. Establishing full flow through 68-559 would require disabling a minimum of two relief valves and compromising the associated emergency core cooling systems. To recover from the test would require resetting the relief valve setpoints by bench test and reinstalling the relief valves.

11. Mechanical Exercise (Disassembly)

Check valve 68-559 has an enclosed design which lacks an external lever which might be used to perform an IWV-3522(b) type test. To perform a full stroke mechanical exercise test, a team would be required to enter containment, disassemble the valve and stroke it.

This test method would put the personnel at risk in the same manner as recovery from a full flow test. In addition, this work would be open to high temperature and pressure exhaust / leakage from the reactor head vents, pressurizer safety valves, PORVs, and the CVCS and RHR system relief valves.

We do not believe the benefits of exercise testing this valve warrant the safety risks to personnel. We have no evidence that stainless steel valves in clean water systems, with no flow and small backseat pressures will fail. For the valve to fail it would have to withstand a minimum opening force of approximately 1,000 lbg.

(2) Part Stroke Testing As shown on figure 2, the relief header is equipped with 3/4-inch test connection. To perform a part stroke test, clean water would be pumped into the header upstream of 68-559.

After overcoming any backpressure within the PRT, the water should begin to flow freely to the PRT.

_ _ _ _ _ ____r___._.

The PRT volume is approximately 13,000 gallons from 0 to 100 percent of the level instrumentation range. With two percent instrument inaccuracies, at least 260 gallons of water would be needed to change PRT water level enough to demonstrate flow through CV-68-559.

This large amount of water precludes using a high pressure test tank. To supply primary grade water into the accumulator room, a hose would have to be connected to the primary water system and run to the test vent upstream of CV-68-559.

This test method also puts test personnel at risk for some of the same reasons. With any of the systems running, the relief valves pose an immediate threat to personnel safety when the test vent isolation valves are opened. The technical specifications and safety analysis cannot allow removal of all these systems except during a defueled condition.

(3) Modifications During the design stages, SQN identified valves requiring testing and provided means to perform the tests. Ideally, each valve or a set of valves can be isolated using manual valves.

For check valve 68-559, this philosphy would have placed manual valves upstream and downstream of the vent connections near the valve. This configuration, however, is prohibited by ASME Section III, Class 2 Section NC-3677.3 which states that there will be no intervening stop valves between relief valves and their relief points.

(4) Technical Specification Requirements Testing of check valve 68-559 while in mode 1 through 5 would have to be performed within the constraints of the SQN technical specifications. The following technical specification LCO ere barriers to safely performing the IWV-3522 exercise test on CV-68-559;

1. LCO 3.1.2.2 - requires two CCPs in modes 1 through 4.

2. LCO 3.1.2.3 - requires one CCP in modes 5 and 6.

3. LCO 3.1.2.4 - requires two CCPs in modes 1 through 4.

4. LCO 3.4.1.3 - requires two RCS loops or two RHR loops in mode 4. <

5. LCO 3.4.1.4 - requires two RHR loops to be OPERABLE with one running in mode 5.

6. LCO 3.4.2 - requires at least one pressurizer safety valve to be operable in modes 4 and 5.

7. LCO 3.4.3.1 - requires all pressurizer safety valves to be operable in modes 1, 2, and 3.

f

-, __, ,__-m -,. _---._,..._,--__..-_m, - _ _ - - _ _ . = _ . . . - - - - - ... ....r _ . ,,., . - - _ . ,, m..---~m_ - - , _ - . . . ,.. ..

~

8. LCO 3.4.3.2 - requires both PORVs to be operable in modes 1, 2, and 3.

9. Additional specifications that will place additional operability requirements on the PORVs below modes 1, 2, and 3 were submitted to NRC in response to NUREO 0737.

10. LCO 3.5.2 - requires two ECCS subsystems in modes 1, 2, and 3.

11. LCO 3.5.3 - requires one ECCS subsystem in mode 4.

Conclusion Check valve 68-559 is a stainless steel swing-check valve which does not experience routine flow, vibration, or differential pressure. In normal operation the valve environment consists of reactor grade water that should preclude corrosion. When the valve is required to open, it will have an opening force of between 1,000 and 18,000 lbg delivered by the system which actuates the relief valves. For these reasons, there is a high probability that the check valve will be capable of adequately relieving overpressure events for the life of the plant.

To test the valve safely will require establishing administrative control over eight safety-related pumps and/or 19 relief valves. The plant technical specifications and FSAR do not allow rendering these systems inoperable, in combination.

SQN believes that the circumstances surrounding this valve provide confidence that the valve will operate as designed and that routine testing will not increase the level of safety and quality. There is not sufficient cause to warrant exposing the test personnel to the risks, or alternately exposing the plant to complex manuevers to test this valve.

As provided by 10 CFR 50.55a(a)(3)(ii), relief from exercising this valve is warranted since the difficulties of testing outweigh the benefits derived while in modes 1 through 5. SQN will perform an alternate test to disassemble and inspect 68-559 every third refueling outage beginning with cycle 3 for both units.

9. REACTOR HEAD VENT VALVES FSV-68-396 and -397 The following discussion describes the alternate test method SQN has chosen to meet the requirements of IWV-3300 for the reactor head vent throttle valves FSV-68-396 and -397. The same alternate test method was employed for the reactor head vent block valves FSV-68-394 and -395 due to their similarity in valve design.

System Description

The reactor head vent system consists of a one-inch line leading from one of the four UHI penetrations on top of the reactor vessel to the  !

pressurizer relief tank. Within this flow path are two sets of i

1

• l

parallel flow solenoid valves. The first set of valves, FSV-68-394 and

-395, act as block valves (full open/ full closed) with green and red indicating lights located on the M-4 panel in the control room. The

-second set of valves, FSV-68-396 and -397, serve as throttle valves with a 0 through 100 percent thumbwheel position controller also on M-4. Both sets of valves are ASME Class II, category B Target Rock, totally enclosed solenoid valves.

The purpose of the system is to vent noncondensible gases from the reactor head region during an accident and prevent gases from impeding reactor coolant circulation flow through the core.

Power supplies for these valves are as follows:

FSV-68-394 125V Battery Bd I (panel 4)

FSV-68-395 125V Battery Bd II (panel 4)

FSV-68-396 125V Battery Bd II (panel 4)

FSV-63-397 125V Battery Bd I (panel 4)

Background / Requirements

, Article IWV-3300 of the Section XI Code requires that valves be visually observed every two years to verify remote valve indications accurately reflect valve operation. Since these valves cannot be visually observed, SQN included in their latest pump and valve IST submittal (reference August 16, 1985 submittal) a request for relief, PV-18, which provided the basis for alternate testing under the provisions of 10 CFR 50.55a(g)(5)(iii) .

Position / Current TestinR/ Philosophy Surveillance Instruction (SI) 166.41 entitled, " Reactor Head-Vent Valve Stroke Test Every 18 Months During a Refueling Outage," has been written and is presently in the plant review process for approval. Portable accoustic monitoring equipment is used to provide an indirect means of verifying valve position. An accoustical " trace"1 is made for each valve as it is opened and closed. Valve position is noted on the trace as a function of time. This SI is scheduled to be performed before startup for both units and will be conducted while the unit is in mode 5 with RCS pressure sufficiently high to effect a blowdown to the l

pressurizer relief tank.

1A trace is a graphical recording of noise level caused by changes in flow turbulence.

--, , , . , , ._, e -----.r- - . . - - ,,.,n-..- .,. ---n--.,.-n,--- - . , - - , - - - - - - - - , - , - - , - - + - + - -

l l

Justification Visual confirmation of valve position as required by IWV-3300 is not possible due to the totally enclosed design of the Target Rock solenoid valves. The accoustic monitoring method was chosen as an indirect means of confirming valve position to meet the intent of the IWV-3300 requirements under the provisions of 10 CFR 50.55a(g)(5)(iii).

Conclusion An accoustical test method will serve in lieu of visual observation to ,

verify that the remote position indicators of the reactor head vent '

valves FSV-68-394, -395, -396, and -397 accurately reflect valve operation. The basis for the request for relief from the IWV-3300 code )

requirement is in accordance with 10 CFR 50.55a(g)(5)(iii), and is similar to the relief request submitted and approved for SQN's PORVs.

]

10. POSTACCIDENT SAMPLING SYSTEM The following discussion describes the alternate test method SQN has chosen to meet the requirements of IWV-3300 for the Postaccident Sampling Valves FSV-43-250, -251, -287, -288, -307, -309, -310. -317, -318, -319,

-325, and -341.

System Description

The Postaccident Sampling System (PASS) is made up of various 3/8-inch sample lines which are capable of taking air and liquid samples from within containment following an accident. These sample lines pass through six different containment penetrations with each penetration having inboard and outboard containment isolation valves (CIVs). All 12 CIVs are ASME Class II, Category A, totally enclosed Target Rock solenoid valves. These valves are equipped with remote and local handswitches.

The remote handswitches are located on the M-10 panel in the control room ,

and act to either " block" or " permit" the use of the local handswitches located in the postaccident sampling facility. Red and green lights on I the local handswitch and control room handswitch provide indication of valve position. The power supply for these 12 valves are the 120V ac vital instrument power boards.

Background / Requirements During modifications for installation of the postaccident sampling facility, SQN identified 14 PASS valves (two check valves and 12 flow solenoid valves) asSection XI valves and made a revision to their IST program in February 1984. These valves were placed in section 6.8C of the FSAR in Amendment 3. Section 6.8C included a remark which stated

~

l l

that testing these valves would begin following the installation of the PASS facility. Installation was completed on April 14, 1984 for unit 1 and on December 18, 1984 for unit 2.

Article IWV-3300 of the Section XI Code requires that valves be visually observed every two years to verify remote valve indications accurately reflect valve operation. Since the subject valves are totally enclosed and cannot be visually observed, SQN included in their latest pump and valve IST submittal (reference dated August 16, 1985 submittal) a request for relief PV-15 which provided the basis for alternate testing under the provisions of 10 CFR 50.55a(g)(5)(iii).

! Position / Current Testint/ Philosophy SQN Surveillance Instruction (SI) 166.4, " Remote Valve Position Indication During Refueling Outage." contains instructions for testing the PASS valves. This instruction is presently in the plant review

] process for approval. Test methods involve monitoring changes in pressure during valve operation. A simplified diagram (see figure 3 and 4 of enclosure 3) displays how pressure is used as an indirect means of verifying valve position. Test pressure used for testing each set of valves is as follows:

Valve Number Test Pressure Range 4

FSV-43-250/251 RCS Pressure 100-350 psig FSV-43-309/310 RCS Pressure 100-350 psig FSV-43-287/288 N2 Pressure 75-100 psig (Regulated)

FSV-43-318/319 N2 Pressure 75-100 psig (Regulated)

FSV-43-307/325 N2 Pressure 75-100 psig (Regulated)

FSV-43-317/341 Service Air 20-30 psig (Regulated)

Justification The PASS valves are similar to the reactor head vent Target Rock solenoid valves. Visual confirmation of valve position as required by IWV-3300 is not possible due to the totally enclosed design. The pressure method, as described in the current test section of this response, was chosen as an indirect means of confirming valve position to meet the intent of the IWV-3300 requirement under the provisions of 10 CFR 50.55a(g)(5)(iii).

Conclusion A pressure monitoring test method will serve in lieu of visual observation to verify that the remote position indicator of the PASS i

valves FSV-43-250, -251, -287, -288, -307, -309, -310, -317, -318 -319,

-325,-and -341 accurately reflect valve operation. The basis for the request for relief from the IWV-3300 Code requirement is in accordance with 10 CFR 50.55a(g)(5)(iii).

11. MAXIMUM ALLOWABLE STROKE TIME FOR POWER-OPERATED VALVES Requirements By the letter from B. J. Youngblood to S. A. White dated September 18, 1986, NRC stated their position on stroke time measurements of power-operated valves. These measurements must be trended so the information can be utilized to monitor valve degradation and predict valve failure. This trending requirement may be exempted by NRC for rapid-acting, power-operated valves. NRC also states that relief from the corrective actions of IWV-3417 presents no safety concerns since variations in the stroke times will be affected by the response times of the test personnel. This relief is restricted to valves with maximum limiting stroke times of two seconds or less.

Current Testing Program At the present time, SQN has 20 rapid-acting, power-operated valves on each unit with maximum allowable stroke times of two seconds or less in the Section XI program. These category A and B valves are routinely stroke tested, and they have been trended by the same program used for all valve trending. Stroke time data indicates that test personnel response times vary enough to warrant not basing corrective actions on the trends.

Proposed Trending / Corrective Action Program Category A and B, power-operated valves that have a maximum allowable stroke time of two seconds or less will not be trended. Corrective actions will be initiated if the valve's test stroke time exceeds two seconds, when measured to the nearest one-tenth of a second. This treatment will not be extended to any valves other than Category A and B power-operated valves with maximum allowable stroke times of two seconds or less.

Justification Stroke time data is collected and trended to aid the operating personnel g in detecting valve degradation and predicting valve failures. Where valve stroke times are very'short, the testing error becomes large enough to render the data useless. For rapid-acting valves there is a significant variance in stroke times that is caused by differing test personnel reaction times. This leads to false trends that indicate performance improvements and/or degradations.

The NRC position on rapid-acting valves allows an exemption from the trending requirements of IWV-3417(a), 1980 Edition. Before this relief is granted the licensee must assign a maximum limiting stroke time of two seconds.

4

- . , . . - - ~ - -

5 SQN procedures assign a maximum allowable stroke time of two seconds for its 20 rapid-acting, power-operated valves. When a valve's test stroke time exceeds two seconds, corrective action is required. This fulfills the intent of IWV-3417 and is in line with NRC's position. For power-operated valves with maximum allowable stroke times of greater than two seconds, the times will be trended and corrective actions will be based on the trends.

Conclusion Basing corrective actions on trend data for rapid-acting power-operated valves is not an appropriate application of trending and can lead to

! unnecessary maintenance mandated by inappropriate trending. To mitigate the random effects of personnel response time differences, a single value

' of two seconds will be used to trigger corrective action. This is consistent with NRC's position on rapid-acting valves and meets the intent of Article IWV-3417(a), 1980 Edition. Accordingly, relief from trend requirements for power-operated valves with maximum allowable stroke times of two seconds or less is appropriate since the alternative

-corrective action guidelines will maintain the quality and safety of these valves.

12. PARALLEL UHI AND SIS CHECK VALVES

System Description

The ECCS at SQN is comprised of low, intermediate and high head safety injection pumps, and cold leg and upper head passive delivery systems (accumulators). ECCS flow is delivered to the RCS through various hot leg, cold leg, and UHI lines, with each delivery path containing primary and secondary check valves (reference FSAR figures 6.3.2-1 and 6.3.2-15). These check valves perform safety-related functions in both the open and closed positions, opening to provide ECCS injection and closing to serve as pressure isolation valves (pIVs).

l In all ECCS delivery systems, with the exception of the cold leg accumulator, multiple parallel injection lines are supplied from a single source, e.g., the intermediate head safety injection system (SIS) pumps supply four parallel injection lines to the four RCS cold legs from a single common supply line. Plant instrumentation exists to indicate injected flow, but plant instrumentation is not provided on the individual injection lines.

Background / Requirements

. In SQN's 1982 submittal of the Section XI Pump and Valve Program, relief was requested to part stroke many ECCS injection check valves during cold 2

shutdowns as it was not practical /possiblo to full stroke the check I

i

- . . - - . - . - - - . - -_ .- - _-.--._-_- = , - - --

valves either quarterly at power or during cold shutdown with the reactor head installed. In the case of RHR check-valves, relief was requested to 4

full stroke the check valves during cold shutdown. In the case of the boron injection tank (BIT) injection check valves, the_ relief request (PV-6) additionally noted that the four branch line check valves are arranged in parallel and individual branch line flows, would only be measured during refueling system performance testing since plant instrumentation is not available to measure the individual branch flows and test instrumentation is installed only for the extensive system performance testing conducted during refueling outages. It was specified that the combination of parallel valves would .be part stroked during cold shutdown testing and each check valve would be full stroked (verified) during refueling outage system performance testing. This relief request ,

was approved by NRC in the April 1985 SER (section 3.4.2.4.4) for the SQN Pump and Valve IST Program.

During a subsequent procedure and program review, SQN noted that the same situation existed for many other ECCS (SIS and UHI) injection check valves--part stroking is performed for the combination of parallel valves during cold shutdown testing due to lack of plant instrumentation--but t that relief requests had not specifically noted this situation as had been done for the BIT check valves. This relief was then requested in SQN's June 6, 1986 submittal of program revisions by clarifying the associated relief requests for the affected valves (PV-5, PV-8, PV-ll) and revising the corresponding section af the narrative provided in enclosure 2 (SER response) as appropriate.

NRC has asked that SQN further discuss the basis for not quarterly exercising the UHI and SIS check valves individually. The basis and justification for relief is provided as follows. ,

Position / Current Testing / Philosophy In SQN's 1982 submittal, SQN stated that the subject ECCS check valves could not be stroked during plant operation, and specifically identified the affected valves and bases as allowed by IWV-3522 (reference associated relief requests PV-5, PV-8, and PV-ll). The submittal further requested relief to only part stroke the subject check valves during cold shutdowns, with the exception of certain RHR check valves, due to

, inadequate letdown capability with the reactor vessel head installed, i

reactivity transients which would be involved and potential damage to the fuel and core internals. In the case of PIVs, relief was also requested to part stroke the check valves on a nine-month cold shutdown basis to coincide with the frequency for required PIV leak testing so as not to c

impose an additional unnecnssary leak testing burden. (The check valves

must be leak tested in accordance with technical specification 4.4.6.2.2 i

4

l following each opening of the valve.) Approval for these relief requests was granted by NRC in the April 1985 SER, sections 3.4.1.1, 3.4.1.2, 3.4.1.3, 3.4.1.4, 3.4.2.4, and 3.7.1.1.

, The additional relief now being requested, item 4 of the Summary Description of SQN's June 6, 1986 submittal, involves verification of stroking of individual ECCS check valves which are arranged and stroked in parallel combination as described under the previous system description. Plant instrumentation is available to verify injected flow but is not available on individual parallel injection lines; therefore, parallel check valves are verified to part stroke in combination on a cold shutdown basis (nine month). The following provides discussion of

additional testing performed to verify operability of the check valves under discussion.

During system performance testing conducted each refueling outage, test instrumentation is installed inside containment on individual injection lines from the intermediate and high head safety injection pump delivery systems and full stroking of associated individual check valves is verified, with the exception of the SIS /RHR/ accumulator primary check

valves. These check valves are individually verified to part stroke

during this system performance testing and one of four of these check valves is disassemled and manually full stroked during each refueling outage on a rotating basis (reference SER section 3.4.1.4, and PV-16 from enclosure 1 and section 3.4.1.4 from enclosure 2 of SQN's August 1985 program submittal).

The check valves from the low head safety injection system cannot be individually verified to stroke (they are arranged in parallel combinations of two) but reasonable assurance is provided that they do

. and can indeed function as required due to the frequent verification of high total injection flow through only the two injection lines, i.e. ,

approximately 2,500 gpm is normally supplied through two cold leg injection lines from an RHR pump whenever providing RHR shutdown cooling and greater than 1,988 spm is verified through these lines on a cold

shutdown basis (nine-month). A flow rate greater than 500 spm is verified on the RHR hot les on a cold shutdown basis (nine month).

During each refueling outage, the four primary UHI check valves are manually full stroked when the UHI spool pieces are removed, and one of the two secondary check valves is disassembled and manually full stroked each refueling outage on a rotating basis (reference SER section 3.7.1.1, and PV-11 from enclosure 1 and section 3.7.1 from enclousre 2 of SQN's August 1985 program submittal).

f-4 4

. , _ _ _ - _ _ _ . . . _ ~ . _ - - ~ , _ _ . _ _ . _ . _ _ _ _ _ _ _ _ _ u. . . _ _ . _ _ _ -

Conclusion In summary, relief is requested to part or full stroke (as applicable to the specific valves) the subject ECCS check valves in parallel combination on a cold shutdown basis (nine-month) due to lack of instrumentation on individual injection lines. Alternate testing as described above and in the individual relief requests (reference SQN's June 6, 1986 submittal) will provide reasonable assurance of individual valve operability as intended by the Code. These relief requests are in accordance witn 10 CFR 50.55a(a)(3) and 10 CFR 50.55a(g)(5)(iii).

13. CONTAINMENT PRESSURE TRANSMITTER ISOLATION VALVES The following discussion describes the alternate test method SQN.has chosen to meet the requirements of IWV-3300 for valves FSV-30-134 and

-135.

System Description

An open-ended 1/2-inch line penetrates containment into the annulus to enable operation of a differential pressure transmitter (PdT-30-133) for containment / annulus WP indication in the control room. Within this line are two Target Rock solenoid valves (FSV-30-134 inboard and FSV-30-135 outboard) which isolate upon a phase A containment isolation signal. Both solenoid valves are totally enclosed and are ASME Class II, Category A.

Background / Requirements In a letter from R. Gridley to B. Youngblood dated August 28, 1986, SQN submitted a relief request to their Pump and Valve IST Program to alternately test FSV-30-134 and -135. This letter resulted from Licencee Event Report (LER) SQRO-50-327/86031 which identified SQN's failure to visually observe the subject valves in accordance with the ASME Section XI IWV-3300 requirement. Since these valves are totally enclosed and position cannot be visually observed, a flow test method was chosen as an alternate means of verifying remote indications accurately reflect valve operation. (Reference PV-21 of the August 28, 1986 letter.)

Position / Current Testing / Philosophy SQN SI 166.4, " Remote Valve Position Indication During a Refueling Outage," contains instructions for testing FSV-30-134 and -135. This instruction is presently in the plant review process for approval.

Test methods involve monitoring changes in air flow during valve operation. An air supply is connected to the open-ended line inside

containment. A 1/4-inch cap is then removed from the bottom of

. PdT-30-133 in the annulus to provide an air flow path. With both valves in the open position, air flow is then established and verified in the annulus. Each valve is stroked to the closed position, and an appropriate change in air flow is verified. This provides an indirect means of verifying valve position.

Justification Valves FSV-30-134 and -135 are similar to the reactor head vent and postaccident sampling Target Rock solenoid valves previously discussed.

Visual confirmation of valve position, as required by IWV-3300, is not possible due to the totally enclosed design. The air flow test method, as described in the current test section of this response, was chosen as an indirect means of confirming valve position to meet the intent of the IWV-3300 requirement under the provisions of 10 CFR 50.55a(g)(5)(iii).

Conclusion An air flow test method will serve in lieu of visual observation to verify that the remote position indicator of FSV-30-134 and -135 accurately reflect valve operation. The basis for the request for relief from the IWV-3300 Code requirement is in accordance with 10 CFR 50.55a(g)(5)(iii).

0405c l

l l

\

I

\

ENCLOSURE 2 MODIFICATION OF RELIEF REQUEST PV-17 WITHIN SEQUOYAH'S IST PROGRAM

PV-17 System: Reactor Coolant System Valve: 68-559 Class: 2 Category: C Function: Opens to admit SIS and CS relief valve discharge to the PRT.

Impractical Requirement: Exercise quarterly at full flow.

Basis for Relief: Full ficw exercising, by definition, is not possible due to the nature of the system. Partial stroke exercising would present unacceptable personnel hazards unless all interacting systems were removed from service; these systems include RCS, CVCS charging / seal flow / letdown, RHR, SIS, and CS systems. Removal from service of the affected systems would required entry into LCO's and an overall degraded plant condition, regardless of operating mode.

Plant operating experience has shown that this CV does function to provide release to the PRT as indicated by PRT temperature, pressure, and level increases.

Alt.

Testing: Disassemble and inspect Frequency for Alt.

Testing: Every third refueling outage beginning with cycle 3 for both units.

0391h

ENCLOSURE 3 GRAPH AND SYSTI}I DRAWINGS (Supplement to Enclosure 1)

Figure 1 18 17 - B 16 n .

4- SI-7 acceptance criteria pump is dec Lared i E 13 - INOP]:RABLE below thi.; line.

h 12 -

g> 11 -

n 10 -

D 9-

f. 8-7-

g 6

0 N

/ / ~~///////

4- / Fuel consumption rate / of /////

a diesel -l 3- .

genei ator at full load.* -

0 l l O 1 2 3 4 TEST INTERVALS A - Hypothetical, Code allcyable test program results.

B - Hypothetical application of TVA program to performance similar to Curve A.

• - The diesel generators consume fuel at approximately 5 gpm when loaded. SI-7 acceptance criteria is equal to this consumption rate plus 10 gpm.

/ *g W W O O o

=~ --. = ~ a:: - - - <I: . yh~

W W O O WO@ N~N' M ' M Wpm #D C D

.n. m...-_..-_

m m

c n-m e

e o

V .Q * . .%}# $- . - .

... . _ . . , . n . n .- . m -. _ n . - . - - . - - . . . - . . - .

m,

.% ma' -a" ~m n.

, D' *** p-* Q. N 2 O O g .

n ..__p g _ g .. ._-----....-.'

• CQ m 3 -

o n

-- C _ ._ . - . _ _ _ _ _ . _ . - . . - - - - . - - _ -

M D

OW U O O Y O

1. M W '

_ _ _ _ .._ - O .s s O e - '..

%@ % Ut % O  % , - ---

W p

, O W

+

.N _. . t/s . . . m __9 , 7 - . _ . _ . _ - . - _ . _ . e i

.. ___.21.._. O._ D___...._....._...__. - - - - . -

S. C

+ . + -

4

- - - - - _ - - . l M, ll ~_- . ~ ,

{ y; g . . . _ _ _ _ . _ _ . . _ _. _.

m n

-. . co -..._..__..._.og-g __ .

v o

%It n

. . - . _ - - . . _ . . . . g . *

........_...-n - - . . _

e c

- .. G yy; . . _ _ . . . . ? - .. . . q -._ .

....._._..._.._._.,e. . _ . . . _ _ _ _ _ .

a e

m

• M

.Mt; o "

m

. . _. _ ._ . . C- -

M g

. . _ _ . . _ . ~ . . . _ . . . . - - . _ . _ _

>*= .

n a

_. t M  :

' g. i I i 4 . _ . _ _ _ . ._. ...-. .., - - '

~..

_ _ - __ _ _ . 3 '

I >**r . I

', m .a - _ . . . _ . . _ . . a i.

. . _ _ _ _ . . . . _ . _ . _ . . . m:

_ . . . - . . . _ . - 't Inside Containment 9

I - ---- n -

. . . _ - . - . . _ - . .b o

A - ,

,. Outside Containment r

-i m mm '

ubW 190: DO t

, ,4. o.C ,

. t i t

l m. m @ m o m g __ . 9 _ _ y C

" e l l

• 6 w . . _ w . _ _ yI ._. _. m a e m .- _ ._ y . _ . . g . . ___ g _..

i l l e a e e 4

l @. @. ( 11 (#1 L/t + '

e g . i t i Ul t/s j

n - -- y ..

N

w. a u . -- un u p r _ t/s4,

( 11 a i  ! I l f l l*

CD m

>* L/t A tit W n j j 6 ,

g. --

g.....g _- g .. } i { . . ,

.g. __.g__g__.g_..___. * '

Z- = M O. M M m N- M- T N N N M. m'.  ! I i i i

N;_ +

l

• l,

. e- o m m e o- e e c ,

s , t 4 m. m c c m m m n i ,

i t 03 03 n -- n -- tr) m , tri- n~ -

n . l l l

n n n n n n n pa.

n l . } }

3 p m'. -. -. 2" M

m M

.o-o 3'

g, 3".

g- _ - g' D O g O

-- g i

{

. I l

• I

'4 e

!*g 00 3 M M M 00 00 00 i j

. 00 w i

, m 3..~ n . u 9 n _. _ n _.. - a .. -. oO... _ n>. I

' i i I e $ N N.

m< m O O O

-[

O - .-.  %

• O r  % --- w w  % .

i  : .,

O O ,  % w- ,

t/i t/s Lit '

O O O -

(

%  %  %  ? . 'I k l .

.h .' l -

l.h l l-  ! l l.. f.

., p. w. -

,. ,.* ,N- . -. . . . .. .

,,~' .--

1

. -. .~

~.

• t ' '#i

.n - .

. . . . . . Figure 3 ........._...

e I,. .: ,

t

. ". :, l[. .) I d r --'!b E' Ir I /.. k- -' - - . . ,L ..

i

-i;:-$.,h'l'7:

' l .l.ds,!

l Of CS HOT LE'G .. - ..

.e

... . -w ;.. , 1 NTM

.~S U M N -- -~~~~'~~~~~~~~~ ,

4.l.

. ,y .m.I ' ' - - ' ' -

.  : :- J: .

t. ,,;,3 ti,.

,o - -

.. . - [ '. . . '

251X,,,X310 i

6 . , ,. l ..

7 i.-. con TAidMENT l

- - .. _ ..L. _ . . - i

.. _... AuxiLARY BLOG

_ _ Q._ Q - J 341 l 317 2fo 309 WA-RHR

_l_ -

pq_ g es n- , ,.

~ -._ , .

i . . . flush vuves t .-

l.

a......

I SAM PLE CH) LL.C,R. .

{ X

h l r---'--L-- 1- -

lllQUlO j PZ J

._ . . . . . PA S .

l-- $ AM PLE I COL L ECTcR. i PAN": L I DF Al N

~

TA N K '_--*

" l m ~,. DT. .

T~

WATER ~

. . GUffLY

.._~......v . . . . _ _

L.

-.--- TRtT.I.A.TED - . . . - - . . . . .

DRAIN a . . . . .. .

TAW K fi '

1

o ,

Figure 4 s

INLET - -

A OUTI.E T B

UPPER! A B l LO!NER

)(93-2SS X 43- 311

'.43-307 X X 93 325 j (43-297 )(43-318

• CON ; AI N .M ENT

---

) - --- - - - ._- _ _ _ , _ _ , __

AuxI t. ARY B un.DIMC--

g

~

edudor

^

.. m.

T I* 3I 3 d 4b sk j Z -

n n . m u y r

" }

+

N CM L'd m 9W :4 -

RCT casa cosa co3x p3

- 3 4 ut.E r ourter }N.

n n '

. i ByfAbs

• a

^

• N 2 i Wlu m e. .

l S e.e. '

• I SAMPLE l CASK j i

• - - .