ML20214S851

Forwards Revised Responses to Sqrt Audit Issues,Per 870522 & 27 Requests.Util Considers Supplemental Info Sufficient to Resolve All Remaining Concerns Re Audit Issues

ML20214S851
Person / Time
Site:	Beaver Valley
Issue date:	06/05/1987
From:	Carey J DUQUESNE LIGHT CO.
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
2NRC-7-155, TAC-62895, NUDOCS 8706100013
Download: ML20214S851 (11)

Text

.e a

'Af 2NRC-7-155 Beevor Vaney No. 2 Unit Project Organization (412) 641 5200 Telecopy (412) 64}5200 Ext.160 I.o',"s,"$"'

Shippingport. PA 15077 June 5, 1987 United States Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555

SUBJECT:

Beaver Valley Power Station Unit No. 2 Docket No. 50-412 BVPS-2 Seismic and Dynamic Qualification of Safety-Related Mechani-cal and Electrical Equipment Gentlemen:

Please find enclosed Duquesne Light Company's (DLC) revised responses to Seismic Qualification Review Team (SQRT) Audit issues. These responses reflect the information requested by the NRC during telephone conversations on May 22, and 27, 1987. DLC considers this supplemental information sufficient to resolve all remaining concerns associated with these remaining audit issues.

Should you have any questions, please contact E. T. Eilmann at (412) 393-7895.

DUQUESNE LIGHT COMPANY By Vd. J. Carey Senior Vice President ETE/ffr

( NR/ETE/SEIS/0YN Attachment i AR/NAR cc: Mr. P. Tam, Project Manager (w/o/a)

Mr. S. Hou, NRC (w/a)

Mr. J. Beall, NRC Sr. Resident Inspector (w/a)

Mr. L. Prividy, NRC Resident Inspector (w/a) l Mr. J. N. Singh, EG&G Idaho, Inc. (w/a) l , INP0 Records Center (w/a) l A

8706100013 870605 PDR ADOCK 05000412 A PDR. I I

'O SQRT Generic Item 3.10.1.3(4)

The applicant is to inform the staff of the completion of the seismic and dynamic qualification program. The completion and confirmation must occur before fuel load.

Response

Duquesne Light has completed their review of the qualification documentation for the-new Atwood and Morrill wye pattern globe Main Steam Isolation Valves (MSIV's) . and have found these valves to be qualified for BVPS Unit 2.

' Attached is a revised copy of the Summary section (Tab 4) of the seismic qualification file which describes the qualification methodology and docu-mentation. The installation of the valves is complete. Equipment and sys-tem tests are ahead of schedule. Criticality will not occur until required testing has been completed.

Status: Closed (May 27, 1987 telephone call with NRC). -

l l

BEAVER VALLEY POWER STATION - UNIT NO. 2 '

EOUIPMENT SEISMIC AND QUALIFICATION REVIEW PROG PUMP AND VALVE OPERABILITY REVIEW PROGRAM SORT /PVORT REVISION'/ ISSUE NUMBER: O DESCRIPTION:

D M th) 6 h 1%LATid \ LLt/ES MARK NO.

2>M 35440Y 1014,B<'C SPEC NO. _ 2BVS- 261 VENDOR: P.O. NO. _ 2BV- 26/

4 f'ucob )) ObrutiLL d.o, IL1C..

s /

Sumary of Qualification approach, methods and conclusions:

See Attached

~

Dates indicateof outstanding open item reports reouiring resolution:(if none, none).

! The cualification evaluated. documents for the above equipment have been re these qualification documents have been found onstration to provi bility reouirements of the Nuclear Regulatory u cation Com Review Team (SORT) and/or Pump and Valve Operability Review .

Signec4(A h $8hWM te 41 29-Ao

(/% & y' gar /'

/ paring Engineer)

Supervisor Initials Date

(

ninnti APR 2 9 m7 Nf.Ssupervisor)

4 3

Page 1 of 5 ATTACHMENT 2BVS-261 Summary of Qualification approach, methods and conclusions:

The Main Steam Isolation Valves are designated as ASME III, Class 2 valves and are located in the Main Steam lines carrying saturated steam fran three steam generators to the BVPS-2 turbine.

The three main steam lines each contain a 28" wye pattern globe type main steam isolation valve and are located in the Main Steam Valve Building immediately outside the Reactor Containment. If a pipe ruptures upstream or downstreaa of the valve, a signal, such as that developed by excess fl ow, causes the valve to close. This valve closure prevents steam flow out of the systen in either the forward or reverse flow direction depending on the location of the pipe rupture.

DLC purchased three Category I Main Steam Line Isolation Valves from Atwood and Morrill Co., Inc. ( A&M) located in Salem, MA. A&M was one of the industry's first valve manufacturers (in the late 1960's/early 1970's) to take 'special interests in the~ specifications for main steam isolation valves. These speci-fications are: low pressure drop, tight seating, the ability to open against a differential pressure, closing time adjustable between 3 to 10 seconds, means for testing through partial stroke and return, and simplicity and reliability of operation.

A&M has fabricated and installed main steam isolation valves at f acilities in Tarapur, India, Nine Mile Point Station Unit 1 of Niagra Mohawk Power Co., and the Oyster Creek Station of Northern States Power Co.

Consequently, the A&M wye type main steam isolation valve which has had more experimental testing than any other design, and its dependable operating experience with a history of successful installations, provides the highest level of confidence in its use.

The information contained in this package adequately reflects the requirements of Specification 2BVS-261, " Main Steam Line Isolation Valves" (Tab 8).

The following mark numbers are qualified by documentation contained in this package:

Main Steam Isolation Valve: 2 MSS *A0V101A Main Steam Isolation Valve: 2 MSS *A0V101B Main Steam Isolation Valve: 2 MSS *A0V101C

i 2BVS-261 ATTACHMENT Page 2 of 5 O

The following documents were used in the qualification of the main steam isola-tion valves:

SWEC File No. Tab No. Document Description 8 2BVS-261 Purchase Specification 2606.510-261-018A 14A 503-13824 MSIV Air Control Panel Test Plan 2706.510-261-002A ISA 13824 MSIV Air Control Panel Test Report 2606.510-261-017A 148 511-13952 Valve Actuator Test Procedure 2706.510-261-003A 15B 201-39500 Valve Actuator Test Report 2706.510-261-004A 15C 44 -

MSIV Bend Test Report 2706.510-261-005A 150 302-15579-00 MSIV Design Report 12241-NP(N)-X2H,R.6 11 Main Steam Pipe Stress Calculation 2006.510-261-048C 10A 15579-03 28" "Y" Type MSIV with Seal and Sh. 1 Poppet 2006.510-261-049C 10A 15579-03 28" "Y" Type MSIV with Seal and Sh. 2 Poppet 2006.510-261-050C 10A 15579-03 28" "Y" Type MSIV with Seal and Sh. 3 Poppet The roadmap to the information used to qualify the equipment can be found in the Database (Tab 7). The Database will allow cross-referencing of Mark No. to Vendor I'.D. No. _

There are two panels associated with the MSIV operation. The air control panel (Specification 2BVS-261 MSIV) which is located on the valve and contains no Class 1E components and an electric panel located remote fran the MSIV. that controls the air control panel on the valve and provides valve position indica-tion. The electric panel is qualified under specification 2BVS-211 (which is the qualification package for the original MSIV). The qualification of the air control panel is documented in the seismic qualification package for specifica-tion 2BVS-261.

The location of the air control panel is shown on attached sketch No.1. The solenoid valves and terminal blocks on the air control panel are not required to electrically function for valve closures. Hence, the panel is not required to be qualified to IEEE 323-1974.

The seismic test plans for the air control panel (Tab 14A) and valve actuator (Tab 148) were prepared by Atwood and Morrill Co., Inc., for AVC0 Environmental Testing Services and American Environments Company, respectively.

Qualification Summary - Seismic Testing / Analysis:

Qualification of the Main Steam Isolation Valves was accomplished by seisnic testing and analysis.

The qualification test procedure for the 32" MSIV air control panel (Tab 14A) was prepared by Atwood and Morrill Co., Inc. The procedure was reviewed and approved in accordance with specification 2BVS-261 (Tab 8). The air control panel tested is identical to that purchased for use at BVPS-2 (per the telephone memorandum of April 14, 1987 located in Tab 90).

O 2BVS-261 ATTACHMENT Page 3 of 5 The qualification test report for the MSIV air control panel (Tab 15A) was prepared for Atwood and Morrill Co. Inc., by the Environmental Testing Labora-tory of AVC0 Systems Division. Environmental Aging was not performed prior to seismic testing since no Class 1E safety-related components are located on this air control panel. Seismic aging is considered complete due to the length of time and amplitude to which the panel was vibrated on the seismic test table.

The panel was subjected to resonance search testing and single axis continuous sinusoidal vibration.

Single axis testing was allowed, in accordance with IEEE-344-1975, since no cross-coupling was indicated during the resonance search. Single frequency testing was also justified since no resonances occurred below 50 Hz. The report was reviewed and approved in accordance with specification 2BVS-261 (Tab

8) and the-test procedure (Tab 14A).

The test report demonstrates the air control panel's ability to withstand the prescribed seismic events without any structural degradation or compromise to functional capability with the following anomalies.

The slight peak-notch response indicated at 26-34 cps, in both' X and Y axes during resonance search testing, was attributed to low-level input plus some rotation excitation by the shaker / fixture suspension and was not associated with the test panel structure. Deviation from unity response, whether peak or notch, during the seismic vibration test, was assessed and attributed to CG location, fixture suspension, shaker stiffness, and/or noise. The deviations, being low in level (+ 2.0 db) and occrring above 36 cps, did not affect the reliability of the seTsmic level tests.

The qualification test procedure for the pneumatically operated safety-related valve actuator and accessories (Tab 148) was prepared by Atwood and Morrill Co. , Inc. The procedure was reviewed and approved in accordance with 2BVS-261 (Tab 8).

The qualification test report for the valve actuator (Tab 15B) was prepared by Atwood and Morrill Co., Inc. The qualification of the 28" MSIV actuator assembly is based on the generic test program of a 28" BWR MSIV actuator assembly. The prototype unit was seismically tested at American Environments Co., Inc. The report was reviewed and approved in accordance with 2BVS-261.

The test methodology employed is in accordance with IEEE-323-1974, 344-1975, 382-1972 and the qualification test procedure.

2BVS-261 ATTACHMENT Page 4 of 5 The valve actuator that was tested is smaller and lighter than the valve actuator for the BV-2 Power Station. However, the items are very similar by design as itemized below:

BWR-Tested Item PWR BV-2 Item For 26" Line Size For 28" Line Size WT = 6,000 lbs. WT = 10,000 lbs.

Production Cover Production Cover Yoke Rods Yoke Rods Springs Springs Spring Plates Spring Plates 20" Dia. Air Cylinder Ass'y 28" Dia. Air Cylinder Ass'y Stem Stem Packing Packing

. Speed Control Speed Control Packing Gland Follower Packing Gland Follower Namco Limit Switch Namco Limit Switch EA 740-20100 EA 740-50120 CFP, Air Control Assembly CFP, Air Control Assembly with Solenoids The valve actuator was seismically tested -in three stages; the first stage was a low amplitude resonant frequency search of the actuator assembly, the second state was a biaxial dynamic seismic simulation test of the limit swi tch assemblies and air manifold assembly components, and the third stage was a static bend test of the actuator assembly. In addition to the testing, the BVPS-2 actuator has been analyzed (Design and Seismic Report, Tab 150).

This Design Report provides the basis for ASME Section III code acceptance for the 3-28" Class 600 wye type Main Steam Isolation Valves on BVPS-2. A stress analysis was performed in accordance with Specification 2BVS-261 (Tab 8), this analysis provides the necessary evaluations and calculations to assure acceptance to the BVPS-2 design requirements. Also included in the Design Report is a static seismic bend test (operability report) which indicates a natural frequency of 35.4 Hz and the conclusion that the valve will operate under seismic conditions.

The test report demonstrates the actuator's ability to withstand the prescribed seismic events without any structural degradation, switch continuity /

discontinuity, or compromise to Class 1E safety function with the following anomal ies.

During the resonant frequency search, resonant frequencies below 33 Hz were noted in the limit switch-junction box assemblies. These assemblies were considered as non-rigid structures and subjected to the second stage of testing described above. The resul ts of the dynamic shake table test were satisfactory, thus resolving the anomaly. At the crvnpletion of the static seismic test, air leakage in excess of the specified maximum allowable rate was noted. This leakage was traced to the 1/8" NPT cennection between the 90 degree elbow tube and the solenoid stack adaptor. Examination of the connection revealed cross-threading. Both the elbow fitting and the adaptor were replaced and the connection was sealed with hydraul ic sealant. Air

4 2BVS-261 ATTACHMENT Page 5 of 5 leakage was again measured and found to be within the acceptable limits. This anomaly was attributed to the incorrect thread engagement of the 1/8" NPT con-nection and was resolved by the corrective action taken by the manufacturer (properly engage the 1/8" NPT connection).

The purpose of the MSIV bend test (Tab 15C) was to show that while applying various pipe end loads to the valve nozzles and an equivalent seismic pull load to the operator assembly, the valve would perform its function as an active valve and close. The end loads applied were to simulate those expected to occur during a guillotine break in the main steam piping while the seismic pull load is the equivalent of a typical design basis earthquake for a MSIV.

The bend test report (Tab 15C) concludes that the ASME III, Class 2 MSIV operated successfully through several cycles of motion under full pipe and seismic load conditions. The closure times under fully loaded conditions and no-load conditions were measured and found to be below the five second closure limit imposed by Specification 2BVS-261 (Tab 8). Stresses in the valve body were recorded during maximum load conditions, and found to be below maximum allowable values defined in the ASME code. (See Appendix II of Report in Tab 15C). The stresses in the valve actuator are determined in Tab 150 and are shown to be below allowable limits for all applicable load conditions.

The MSIV design and seismic report (Tab ISD) provides the bases for ASME Section III Code acceptance for three 28" Class 600 wye type Main Stean Isola-tion Valves for the BV-2 Power Station.

A&M has performed this analysis in accordance with the requirements of Design Specification 2BVS-261, Revision 1, (Tab 8) and criteria defined in the appli-cable ASME Boiler and Pressure Vessel Code. This analysis provides the neces-sary evaluations and calculations to assure acceptance to the design require-ments stated in Specification 2BVS-261, Revision 1.

The Namco EA-740 limit switches are generically qualified for seismic and environmental conditions in Namco Reports QTR 111, Revision 1, QTR 112, Revision 1 and QTR 127, Revision 1. These reports are available under SDOF 2707.660-651-0018.

Nozzle End Loads and By-Pass Piping Loads The MSIV's are installed in-line with 32" diameter main steam piping. The end loads developed on the MSIV's are considered acceptable due to the maximun stress levels in the process piping being less than code allowables. The 2" by-pass piping loads developed on the valve body are also considered acceptable based on maximum stress levels in the by-pass piping being less than code allowables. The above end loads are contained in Tab 11. In addition, the nozzles located on the air intake manifold of the control panel utilize flex-ible pipe connections. Therefore, the attachment piping to the manifold will not transmit end loads to the nozzle.

Conclusions Based on the above information, SWEC verification of test input, design analy-sis, resolution of anomalies and nozzle end load review, it has been determined that the MSIV's are seismically qualified for use at BVPS-2.

o a

SK ETCH NO. 1 MSIV AIR CONTROL P ANEL BO ARD LOC ATION

, \

g ox.

s tie.o

/' s AMBm. .

, l

, /

s *,/ A. ~

lO3.0 R G II s 'I" s ,

s \

\

s 2.50

+- G U-* \\

,,gg(

t.w 0

'

• G 50 -Pj J

1' c

'~l 17 4.0

• , f AFPflo d .

G+

e

~~ 1

--(4 1 - -

G+ l y

- / s l Ef* PDSS 2* SCH to GE.o RE.F +

^

149.o spex. A

~,

O

su SORT

+

DLC Letter - December 24, 1985 Issue 142 i

Provide the method for considering low-cycle fatigue effects resulting from seismic excitation on electrical and instrumentation equipment and their related supporting structures.

Discussion:

I This was identified by_ the NRC as an 'old issue" in the meeting of 4/1/87 in Bethesda. The NRC's request for this information was made with reference to-i DLC's response to item 42 identified in DLC's letter 2NRC-5-157 - dated 12/24/85 which was written in response to the BVPS-2 SER (NUREG-1057) dated i

10/85.

Response

Low-cycle fatigue effects, which require a significant strain range typical-3 ly involving plastic deformation and malfunction in approximately 1000 -

cycles or less, have been considered in seismic qualification testing and analysis.

All1 Class 1E electrical and instrumentation equipment have been qualified by test. In seismic . testing, the number of tests associated with the 5 OBE events and the 1 SSE event and their duration and amplitude, along_ with any natural frequency scan testing, when utilized, adequately addressed low-

, cycle fatigue affects. Each seismic test was 30 seconds (not 15 seconds as

! actually anticipated) in duration. An approximate minimum duration of 3 minutes was associated with low-level sine sweep testing in _each axis when natural frequency scan testing was employed.

4 In most cases the electrical and instrumentation equipment support struc-tures were also included in the seismic testing. When the qualification of these low carbon steel supporting structures _was by analysis, material

stress allowables of 70% of minimum yield strength for the OBE design condi-tion and the lesser of 100% of minimum yield or 70% of ultimate strength for i- the SSE design condition were utilized. Fatigue design data such as the i conservative design curves of the ASME Boiler and Pressure Vessel Code, dem-onstrate these stress allowables are sufficient to prevent low-cycle fatigue._ Also the design of the cabinets and their supporting structures

generally required the first natural frequency to be above the amplified region of the specific amplified response spectra (ARS). This typically resulted in stresses much lower than the above allowables. Therefore they 2 may be considered as " seismically rigid" structures. (See attached clarifi-l: cation regarding fatigue cycling effects.)

Status: (Response modified as agreed upon in May 27, 1987 telephone conver-i sation with NRC) Closed

e SQRT 4

Issue #42 Additional Clarification - Fatigue Cycling Effects The design basis for the support structure contains inherent margin against -fatigue due to seismic excitations. Typical peak amplitudes for the BVPS-2 amplified response spectrum occur at about 10 Hz. Since each seismic event is considered to be of 15 seconds duration,150 cycles is the maximum number of cycles that could occur at peak amplitude. Since the cabinet supports respond rigidly to this excitation for 5 OBE events and 1 SSE event the total number of cycles is conservatively 900 cycles of which no more than 150 cycles would be at SSE amplitude. Typical support materials have minimum yield strengths of 33 ksi and minimum ultimate strengths of 45 ksi. Using the ASME fatigue curves which are developed for pressure boundary materials as an accepted basis for fatigue evalu-ation, then from the ASME fatigue curve (Figure I-9.1) for low carbon steel with UTS < 80 ksi, the number of cycles that correspond to 70% of yield (0BE) is iBout 30,000 cycles, and the number of cycles that corres-pond to 100% yield (SSE) is about 15,000 cycles. Therefore,.by ensuring that the stress levels are below the limits specified above, fatigue due to seismic excitation is precluded.

As an alternative approach, using the same fatigue curves, the ' stress allowable (Sa) of 90 ksi for 900 cycles provides an adequate design margin when compared to the maximum design allowable of 33 ksi (yield).

. . _ _