JPN-96-048, Forwards Response to RAI Re USI A-46, Seismic Qualification of Equipment in Operating Nuclear Power Plants

Forwards Response to RAI Re USI A-46, Seismic Qualification of Equipment in Operating Nuclear Power Plants

ML20134Q008
Person / Time
Site:	FitzPatrick
Issue date:	11/25/1996
From:	William Cahill POWER AUTHORITY OF THE STATE OF NEW YORK (NEW YORK
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
REF-GTECI-A-46, REF-GTECI-SC, TASK-A-46, TASK-OR JPN-96-048, JPN-96-48, NUDOCS 9612020049
Download: ML20134Q008 (12)

Text

. _ ._.. ..

l l

• . 123 Main street l White Plains. New York 10601 I i 914-681 6840 l 914 287-3309 (FAX)

)

. 1

, #> NewYorkPbwer

& Authority wiiii. ;. c.hii,. s,.  ;

l Chief Nuclear Officer '

November 25,1996 JPN-96-048 l U.S. Nuclear Regulatory Commission  :

l Attn: Document Control Desk l Mail Station P1-137 l Washington, D.C. 20555 )

i  :

l

SUBJECT:

James A. FitzPatrick Nuclear Power Plant l

Docket No. 50-333 l Response to Request for Additional mformation Regarding USI A-46 " Seismic Qualification of Eauipment in ODeratina Nuclear Power Plants" l

l l

References:

NYPA letter, W. J. Cahill, Jr. to USNRC (JPN-95-049) dated l November 15,1995 regarding " Summary Report for Resolution of l Unresolved Safety issue (USI) A-46" l

Dear Sir:

I l i l Attached is the Authority's response to a recent request for additionalinformation l regarding Unresolved Safety issue A-46," Seismic Qualification of Equipment in Operating

! Nuclear Power Plants" at the James A. FitzPatrick Nuclear Power Plant. Six questions regarding the FitzPatrick Summary Report (Reference) were recently telecopied to the Authority by the NRC Project Manager.

l If you have any questions, please contact Ms. Charlene D. Faison.

Very truly yours,  ;

. . }

a00104

~

William J. ahill, Jr.

f j Chief Nuclear Officer

< Attachme nts: As stated I

b / .

9612O20049 961125 PDR ADOCK 05000333 i P PDR  ;

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

9 t

l :.

.i i

+

cc: Regional Administrator . 1 1 U.S. Nuclear Regulatory Commission ,

475 Allendale Road  ;

King of Prussia, PA 19406

l. . ,

l. Office'of the Resident inspector-

' U.S. Nuclear Regulatory Commission  !

P.O. Box 136  ;

Lycoming, NY 13093 j l

Ms. K. Cotton, Acting Project Manager  !

' Project Directorate 1-1  ;

Division of Reactor Projects 1/11' -(

U.S. Nuclear Regulatory Commission i Mail Stop 14 B2 )

' Washington, DC 20555  ;

l 1

i l^

l t

l l

+

+

k 1

l I'

1 i

i i . . - -

Attachment I to JPN-96-048

Response to Request for AdditionalInformation Regarding USI A-46 4

Question 1 Table 3.1 in Section 3 of Enclosure 2 of the licensee's submittal provides a summary of instances where the intent rather than the letter of certain caveats, as described in Appendix B of the Generic

, implementation Procedure, Revision 2 (GIP-2) was met. Based on the information provided by the

, licensee, it is unclear as to how some equipment were determined to meet the intent of the stated j caveat. Listed below are specific areas that f all in this category for which we are requesting additional information.

i

a. For the Liquid Nitrogen Tank A and Tank B Relief Valves 27RV-101 A, and 27RV-101B, the ,

licensee stated that the valve operator cantilever length is 97 inch and it exceeds the i distance given in Figure 8.7-1 of GlP-2 (FOV/BS Caveat 5). Also, the valve is mounted on  !

a pipe that is smaller than 1 inch in diameter (FOV/BS Caveat 4). No mention was made regarding the weight of the operator. The concerns relating to these two caveats are that a 3 valve with a heavy operator on a small line may cause an overstressed condition in the

adjacent piping, and that long valve operator cantilever length may lead to excessive valve

, yoke stress. The licensee is requested to provide additional information concerr'ng the weight of the valve operator and to demonstrate that the pipe stresses / valve yoke stress are ,

low as described in Appendix B of GIP-2. j

b. For the Inboard and Outboard Main Steam isolation Valves 29AOV-80A-D and 29AOV-86A-D, the licensee stated that these are large piston-operated AOVs on a 24 inch pipe and the valve operator cantilever length is 130 inch. These valves are larger that AOVs in the ,

experience data base as shown in Figure B.7-2 in Appendix B of GIP-2. The licensee is  !

requested to provide additional information including the weight of the valve operator to demonstrate that the selsmic capacity of these AOVs may be represented in the generic seismic testing equipment class.

4 Response 1a The two valves, Liquid Nitrogen Tank A and B relief valves 27RV-101 A and B, are id scal. The relevant features are:

. The valve is a pressure control valve; the valve is relatively lightweight because it does not have a heavy operator such as a piston-driven air-operator or a motor operator. Based on the walk-down and valve drawings, the weight of the valve is estimated to be less than or equal to 50 lb. j

• The valve is mounted on a less than 1 in_h diameter pipe. )

. The valve measures 26 inch from the carder of the line to the top of the valve.

• The valve is anchored to the liquid nitn:gan tank via a welded steel bracket that is attached to the valve at a point 18 inch below the top of the valve (8 inch above the center line of the pipe). The liquid nitrogen tank is a large, stainless steel horizontal tank that was separately evaluated as part of the A-46 program.

These valves do not meet the wording of the GIP Bounding Spectrum (Reference 1) caveats for )

fluid-operated valves (equipment class 7) because:

. The valves are mounted on a pipe which is less than 1 inch in diameter (FOV/BS Caveat 4).

1 I

. Attachment I to JPN-96-048 Response to Request for AdditionalInformation Regarding USI A-46

. The valves do not meet the cantilever length requirement in GIP Figure B.7-1 because the Figure does not provide a maximum length for valves mounted on pipes less than 1 inch in diameter (FOV/BS Caveat 5).

As pointed out in the request and quoting GIP Appendix B, page B.7-3, a ..The concern is that valves with heavy operators may cause an overstressed condition in the adjacent piping.. ". The Seismic Review Team (SRT) judged the valve as meeting the intent of the caveats because the welded steel bracket securing the valve to the tank prevents the valve from stressing the pipe. This judgment is consistent with GIP guidance - again quoting from GIP page B.7-3, a ..There is no l concern if the valve, the operator, and the line (if smaller than 1 inch) are well-supported and i anchored to the same support structure."  ;

Response Ib I The eight valves, inboard and Outboard Main Steam isolation Valves 29AOV-80A,B,C,D and 29AOV-86A,B,C,D, are identicat The relevant features are:

. The valve is a Rockwell 1612 JMMNY air-operated stop valve with a Ralph A. Hiller SA series piston operator. The operator weighs 1450 lb.

. The valve is mounted on a 24 inch diameter pipe and measures 130 inch from the center line of the pipe to the top of the operator.

. The yoke consists of four carbon steel tubes at ute corners of a 13 inch square. Each tube is 48 inch long, 3 inch diameter, and 5/8 inch thick.

These valves do not meet the wording of the GlP Bounding Spectrum caveats for fluid-operated valves (equipment class 7) because they exceed the weight / cantilever length limits of 750 lb. /

100 inch in GlP Figure B.7-2.

In accordance with GlP page B.7-4, a 3g load evaluation was performed. This evaluation is documented in the SEWS and repeated below:

Under lateral load, the steel tubes bend as guided cantilevers:

I = 3.14159 r' t = 3.14159 x 1.5' x 0.625 = 6.6 in' S = l / r - 4.4 in' k = 12El/L' = 12 x 29000 x 6.6 / 48' = 21 k/in stress = 3WL / 8 / S = 3 x 1.45 x 48 / 8 / 4.4 - 5.9 ksi displacement = 3W / 4k = 3 x 1.45 / 4 / 21 = 0.05 inch Based on the low stress and displacement, the SRT judged the valves to be acceptable.

Also as shown on sheets 11 through 13 of Table 16.2-7 of the FitzPatrick UFSAR (Reference 2),

the MSIVs were evaluated for the combined effects of operational, dead weight, and DBE seismic loads.

2

Attachment I to JPN-96-048 Response to Request for AdditionalInformation Regarding USl A-46 Question 2 >

In Attachment C of Section 3, as listed in the table for Safe Shutdown Equipment List of the licensee's submittal, the screenings and evaluations for some equipment were noted as relying upon operation actions. The licensee is requested to confirm that a controlled procedure that prioritizes the operator actions exists and that it will preclude any conflicting or competing events which could lead the operator to not perform timing actions.

Response 2 Existing procedures have been used to implement and control operator actions identified as a part of the resolution of USl A-46. These operating procedures are controlled using existing processes.

The actions were reviewed by the Operations department and are acceptable. i The GIP does not require the development of procedure (s) specifically for USI A-46. Section 3.2.8 ,

of Revision 2 of the GIP states:

" Procedures should be in place for operating equipment selected for safe shutdown and operators should be trained in their use. it is not necessary to develop new procedures specifically for compliance with the USI A-46 program. Existing plant procedures can be used.

If a SSE occurs, it is not necessary to use only the safe shutdown equipment identified in for tae USl A-46 program. The operator may attempt shutdown using other available systems and equipment provided these other means of shutting down do not prevent later use of the safe shutdown method identified for the A-46 program.

The plant procedures for shutting down should be reviewed by the Operations Department of the plant to verify that the procedures are compatible with the identified method of safe shutdown and that they do not preclude the use of the safe shutdown equipment if some other method of shutting down is attempted first. See Section 3.7 for suggested methods of performing this review.  ;

The shutdown procedures which are associated with the use of the USl A-46 safe shutdown equipment should be procedures which are available to the operator as a result of his following approved normal and emergency operating procedures (EOPs). Note that normal plant shutdown procedures would be used for any deliberate, planned shutdown; EOPs would be used for a plant trip or emergency situation."

l l

3 l

. Attachment I to JPN-96-048 Response to Request for AdditionalInformation Regarding USl A-46 Question 3 For your plant structures containing equipment in the USl A-46 scope:

a. Identify structures which have licensing-basis floor response spectra (5% critical damping) for elevations within 40-feet above the effective grade, which are higher in amplitude that 1.5 times the SOUG Bounding Spectrum.

b. Provide the response spectra designated according to height above the effective grade identified in item 4.a above and a comparison to 1.5 times the Bounding Spectrum.

c. With respect to the comparison of equipment seismic capacity to seismic demand, indicate which method (Method A or Method B in Table 4-1 of GIP-2) was used to address the seismic adequacy of equipment installed on those floors as identified in item 4.a above.

Response 3a and 3b A gra,nhical comparison and a discussion of the LBFRS compared to 1.5BS was provided in Section 3.1.1 of the USl A-46 Seismic Evaluation Report (Reference 3) previously submitted. The figures (Figure 3.1 and Figure 3.2) from that report are attached.

Plant grade is elevation 272 ft. Figure 3.1 shows the envelope of the horizontal Turbine Building floor response spectra (FRS). FRS are available for two elevations in the Turbine Building - 272 )

and 300 ft. Both are within 40 ft. of grade, and both are enveloped by 1.5BS. Note that the Turbine Building at FitzPatrick encompasses the Control Building (battery rooms on 272 ft. el., relay room .

on 284 ft. el., and control room on 300 ft. el.), the Electrical Bay (272 ft. el. , contains 480V  !

switchgear, transformers, and MCCs), the Diesel Generator Building (272 ft. el.), and the Service {

Water Pump Area (255 ft el.).

Figure 3.2 shows the available Reactor Building FRS for elevations cord.ning A-46 equipment -

272 ft. el.,326 ft. el., and 344 ft. el.. Elevation 272 ft. is grade; there is also A-46 equipment below elevation 272 ft. In the reactor building crescents and the torus compartment. Only elevation 272 ft. (and below) is within 40 ft. of grade. The FRS for 272ft. el. is largely enveloped by 1.5BS except at about 15 Hz.

Response 3c The GIP does not specify whether Method A or Method B is preferable. At FitzPatrick, Method A was applied first, and Method B was applied if the elevation (within 40' of grade) or frequency requirements (greater than 8 Hz) associated with Method A could not be met. Of the 443 items on the seismic Safe Shutdown Equipment List (SSEL), Method A was used for 315, Method B was used for 124, and other methods were used for 4 items. Of the 428 items on the seismic SSEL that are mounted within 40 ft. of grade, Method A was used for 315, Method B was used for 111, and other methods were used for 2 items (note that only 15 items are mounted more than 40 ft. above grade).

4

Attachment I to JPN-96-048 i Response to Request for Additionalinformation Regarding USI A I Question 4 i' Under the title of Amerace (Agastat) E7022PK on page 1/2 of the Stevenson & Associates Review l of Outlier Relays, the report states "This construction justifies using the SSE as the IRS (rather than l t 2.25x SSE), which reduces the demand to 1.6g peak /1.1g ZPA, well below the capacity of 4.0g / i 1

1.6g ZPA." What is the basis for using the SSE to replace the IRS? The SSE is the free field )

- ground motion specified for the plant and the shape and amplitude of the spectrum is independent of the structure, while the IRS is the in-structure-response spectra at particular elevations in the building. The amplitudes of an IRS are generally higher than those of the SSE due to amplification and filtering by the building structure.  ;

Question 5 a

Please submit information that will verify the demands for relays Amerace (Agastat) E7022Pk,  ;

Barton 289, and GE 12HFA151 A2F are 1.6g peak /1.1g ZPA,1.1g peak / 0.7g ZPA, and 1,1g peak

/ 0.7g ZPA, respectively.

Response 4 and 5 l The original relay capacity versus demand evaluation proceeded as follows:

1. In accordance with GlP Section 6.4.2, Screening Level 2, the demand at the base of all of I the cabinets housing the subject relays was derived using 1.5x SSE as the IRS (In-structure j Response Spectrum), with an additional f actor of conservatism of 1.5 as required by GIP 1 Table 6-1. Tnis resulted in a seismic demand at the base of the cabinets of 0.53g peak / I 0.34 ZPA.

2. The SRT designated the cabinets housing the Agastat relays as high amplification, and the cabinets housing the Bartons and GE relays as medium amplification. Per GIP Table 6-2, the base demand is multiplied by 7 for the high amplification cabinets and 4.5 for the medium amplification cabinets to arrive at the relay seismic demands.

3. The resulting relay seismic demands were compared to the EPRI Relay GERS. The data is summarized in the table below:

Relay Seismic Capacity Seismic Demand (peak /zpa) (peak /zpa)

Agastat E7022PK 4g /1.60 3.70 / 2.40 Barton 289 3g /1.20 2.40 /1.5g GE 12HFA151 A2F Sg /1.2g 2.4g /1.5g The relays were designated outliers because the ZPA demand exceeds the ZPA capacity.

Note that in all cases the peak capacity exceeds the demand, but the ZPA demand exceeds  ;

the capacity. '

in the Stevenson & Associates Review of Outlier Relays, the seismic demand calculated above was i reconsidered by more closely examining the construction of the structures where the relays are located.

5

1

. j Attachment I to JPN-96-048

Response to Request for AdditionalInformation Regarding l USl A-46 I 4

The Agastat relays are located in the diesel generator engine control panels (93ECP-A,B,C,D).

These panels are mounted on the diesel generator pedestals next to the diesel generators on elevation 272' of the Diesel Generator Building. The pedestals are 31.5' long x 10.5' wide x 7' thick reinforced concrete, are founded directly on rock at elevation 265'. (See FitzPatrick reinforced concrete drawings, FC-38 series.) The rest of the Diesel Generator Building floor at elevation 272' consists of a 12" thick reinforced concrete slab supported on a lean concrete fill down to the top of rock at elevation 265'. Based on this construction, it was concaded that the in-structure response at the base of these cabinets would be essentially identical to the response at the top of the rock,  !

which is represented by the SSE. Using the SSE, rather than 1.5 x 1.5 x SSE, as the demand at  ;

the base of the cabinet, reduces the seismic demand on the relay to 1.6g peak /1.1g ZPA, which is  :

, well below the relay's seismic capacity.

l The GE relays are mounted in the diesel generator auxiliary relay panels (93AURP-01,02) which are mounted on a reinforced concrete wall in the Diesel Generator Building, about 6' above the floor slab at elevation 272'. Considering the construction of the Diesel Generator Building, the IRS at elevation 272' should be close to the SSE, rather than 1.5 x 1.5 SSE, which would reduce the relay demand to 1.1g peak / 0.7g ZPA.

The Bartons are mounted on a rack on elevation 242' of the Reactor Building. The Reactor 4 Building was constructed by excavating from the top of the rock at 265' down to 222*. The base  !

mat is 5' thick, so the bottom of the torus compartment and the crescent areas (the annular areas surrounding the torus compartment) are at elevation 227'. . There is a massive concrete drywell support pedestal in the center of the reactor building from rock up to about elevation 250'. As a result, the lower part of the reactor building is essentially a concrete monolith embedded in a rock foundation. As above, based on this construction, it was concluded that the in-structure response at the base of the racks would be essentially identical to the response at the top of the rock, which is represented by the SSE. Using the SSE, rather than 1.5 x 1.5 x SSE, as the demand at the base of i the cabinet, reduces the seismic demand on the relay to 1.1g peak / 0.7g ZPA. This is well below the relay's seismic capacity The RAI raises the point that the SSE is the free-field ground motion, and the IRS is generally higher than the SSE due to amplification induced by the building structure. In general this is true, however,' for a rock site like FitzPatrick, the SSE represents the motion at the rock, so when a  !

portion of a structure is essentially founded on rock - like elevation 272' of the Diesel Generator Building and the lower Reactor Building - it is reasonable to conclude that the IRS at this location I would be nearly the same as the SSE.  !

6

. Attachment I to JPN-96-048 Response to Request for Additionalinformation Regarding USl A-46 l

~

1.5

. . . . . . . . . . . . .. j

- 1.0 - -*- -.* . .ri . 1.5x Bounding Spectrum rs,t

'. B . . . . . ... . , , . . . ..

C . . i . . ... . . . . . . .. ,

.j . . . . . ... . .. . . ..

g . . . . . ... . . . . .

.e . . . . . ... . . . . . . .. .

3 . . . . . ... . . . . . . .. i 8 . . . . . ... . . . . . ..  ;

< 0.5

'2 *--*

• - * ' . - = . = . = .'= L .2. . .. . * .L 2. .* 2  ;

Envelope of Horizontat 7 '

t I

. Tur.t>ine.Bui.ldin.9 F. R.S.' . '. '..

. . ... . . . . . . .. ~ . t 0

)

1 10 100 Frequency (Hz)  ;

Figure 3.1 1.5x Bounding Spectrum vs. the Envelope of the Turbine Building Floor Response Spectra (both curves 5% damped)  ;

i 1.5 >

. . . . . . . . 1,5x Bounding Spectrum  ;

--- Reactor Building 344 '

- -- Reactor Building 326 )

---

Reactor Building 272

. . . ,l. .i_, .. . . . . . . . .

_ g .o ...... . .. ,.

.t

- . . .. . . . . ... .,.. e. .

. , .s 3 , . .e .\.. . . . . . . . .

c . . .

l- .\, . * * = **

3m , . .. s.---, . . . . . . .. }

, , . . .\ ... , i . . . . . . ..

g . . . . ..\. .  ; . . . . . . ..

3 . . . . ..g. l . . . . . . ..

o , .  ;

.ll

.. ,. .\.

. . -*--N.

,-.,g ' j N.

0.5 - --,- ,- .--. . . .. ,

l--,... q, s g . . . . . ..

. . ,.i.., . . . . . . .. i

.f ,'... N,, s .-r, -

. J ../. ,, . . '. s ,/ . ',. . N on

_--e,,,,._f.. . . . ... . '?. . . z. . . . .

}

0 1 10 100 Frequency (Hz) i i

,,='M u

~ . - 1 Figure 3.2 1.5x Bounding Spectrtm vs. the Reactor Building Floor Response l Spectra (all curves 5% damped) i 7

1 i

l I

i w.~,

a 4: eame,a, , -- m. w - 4.- M -4 a 2--

I Attachment I to JPN-96-048 Response to Request for AdditionalInformation Regarding

! USl A-46 Question 6 l In its submittal, the licensee stated that all the identified outliers will be resolved no later that l startup from the Refuel 13/ Cycle 14 refueling outage (currently scheduled to begin October 30, 1998). The licensee is requested to provide it's justification to support the proposed schedule for resolving all the identified outliers, to ascertain that the resolution of most safety-significant outliers l will be completed in a timely manner.

Response 6 i

Schedule for Outiler Resolution l At this time, the Authority has resolved 44 out of 62 SSEL outliers. Of the remaining 18 outliers one outlier (10SOV-68B) will be resolved during the 1996 Refuel Outage, currently ongoing . The remaining outliers have been prioritized by the Authority's Reactor Engineering / Nuclear Systems Analysis (RE/NSA) Group in descending order of core damage frecuency contribution by examining the J.A. FitzPatrick Individual Plant Examination (IPE) model. The RE/NSA Group has determined t , that the resolution of the outliers should be performed in the following order:

l t

Relav outliers located in Diesel Generator canels - Failure of these relays, as a result of a l seismic event, would affect the starting and running capabilities of the diesel generators and result in a dominant contributor to the station blackout event.

Emeroency Core Coolino System Trio Cabinets (09-95. 09-96) -- ATTS actuation and interlock relays are located in these two panels. In case of f ailure of these two panels as a result of a seismic event, the automation control capabilities of the reactor pressure and level inventory function will be lost. This would result in the highest contribution to the FitzPatrick core damage f requency.

600V Motor Control Center (MCC) 71MCC-161 - In case of failure of this MCC, as a result of a seismic event, would result in loss of division "B" of RHR containment spray mode and )

"B" RHRSW cross-tie capability. I 1

l Uninterruotable Bus MG Set 71UPS-1-- The f ailure of 71UPS-1 would result in turbine trip ,

l !nitiator. l Other outliers -- These outliers have no major contribution to the FitzPatrick core ohmage frequency.

The resolution of the outliers will be performed in the order stated as above, without interruption of l plant operation. The Authority believes that all outliers can be resolved during plant operation. If resolution of the relay outliers require modifications, then these modifications will be installed during the first plant outage of sufficient duration.

! All remaining outliers will be resolved during plant operation in 1997/1998 and the refueling outage in 1998. Outliers will be resolved by either modification, replacement, testing, or analysis. During these analyses and testing, additional modifications beyond those anticipated may be identified and the above stated schedule may change. The Authority will inform the NRC if a schedule change becomes necessary.

4 4

8 4

=

4 Attachment I to JPN 96-048 Response to Request for AdditionalInformation Regarding USl A-46 This schedule assures that the remaining outliers are resolved in a timely manner and is consistent with the schedule included with the USl A-46 summary report (Reference 3) submitted in November 1995. As stated in that letter (Reference 3), outliers were reviewed for operability and no I operability concerns were identified. In addition, there were no significant or programmatic deviations from the GIP (Generic implementation Procedure). Further schedule improvements are not possible without an additional scheduled plant shutdown. An additional scheduled shutdown is not warranted at this stage of the resolution of USl A-46.

I l

l l

i 9

j 1

. .- Attachment I to JPN-96-048  !

Response to Request for Additionalinformation Regarding  !

USI A-46 )

! l l

l References i

l 1. SOUG Generic Implementation Procedure for Seismic Verification of Nuclear Power l Plant Equipment, Revision 2, dated February 14,1992.

2. James A. FitzPatrick Nuclear Power Plant, Updated Final Safety Analysis Report,  !

l Section 16.2 " Nuclear Steam Supply System Component Structural Loading Criteria '

! and Design," Table 16.2-7.  !

l l 3. NYPA letter, W. J. Cahill, Jr. to USNRC (JPN-95-049) dated November 15,1995

regarding " Summary Report for Resolution of Unresolved Safety Issue A-46" i l

4. NYPA letter, W. J. Cahili, Jr. to USNRC (JPN-95-028) dated May 31,1995 regarding " Updated Status Report on Seismic Verification -- Unresolved Safety issue (USI) A-46" l

l l

l l

i l

I l

10 l