ML20096B510

Provides Response to Request for Addl Info Re Insp Rept 50-443/91-81 & Motor Operated Valve Program Controls Re Switch Setpoint Error Analysis & Step Friction Coefficients

ML20096B510
Person / Time
Site:	Seabrook
Issue date:	04/30/1992
From:	Feigenbaum T PUBLIC SERVICE CO. OF NEW HAMPSHIRE
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
GL-89-10, NYN-92058, NUDOCS 9205120162
Download: ML20096B510 (12)

Text

_

New Hampshire Y'hhkhh Ted C. Felgenbaum President and Chief bewtive Officer NYN 92058 April 30,1992 United States Nuclear Regulatory Commission

-Washington, DC 20555 Attintion:

Document Control Desk.

l(cierences:

a)

Facility Operating License NFP 86, Docket No. 50 443 l

b)

USNRC Generic Letter No. 8910, dated June 28,1989, " Safety Related i

Motor operated Valve Testing and Surveillance" c)

USNRC letter dated February 2,1992," Motor Operated Valve inspection at Seabrook Station Inspection Report 50 443/91-81,* M. W. Ilodger to

- T. C. Felgenbaum

Subject:

Response to a Request for Additional Information Gentlemen:

Inspectioa Report 50 443/91 81 l Reference (c)] requested additional information regarding New 11ampshire Yankee's (NilY) motor operated valve program controls related to r, witch setpoint error analysis and stem friction coefficients. The requested information is provided in the Enclosures.

Shocid you have any questions regarding this matter, please contact Mr. James M.

Peschel, Regulatory Compliance Manager, at (603) 474 9521,-extension)772.m

,/

Very trt y yours,

(.h' Qf /

es%

8 Ted C. Feigenbaum

'l CF:TG P/a et Enclosure fr'r -.M TIA0 s

t lldd: t9EaO/0SflG 13a et, s2 *t I

t Y0l-p u M, d.~ 5'-

s I

f

• lI NU/AffEmfD I

AES MtQglil NM A u'E 13C i

V 3

l l

9205120162 920430 PDR ADOCK 05000443 G

PDR j

O80Ja"O New Hocnpshiro Yonkee Division of Public Service Company of New Hampshire

\\

P.O. Box 300 e Seabrook, NH 03874

• Telephone (603) 474 9521

. =

.a--.

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

- Uhited States Nuclear Regulatory Commission April 30,1992 Attention:

Document Control Desk Page two i

cc:

hit. Thomas T. Martin Regional Adminleasuior U. S. Nuclear Regulatory Commission llegion 1 475 Allendale Road:

King of Prussia, PA 19406 I

Mr. Gordon IL Edison, St.- Project Manager Project Directorate 13.

t Division of Reactor Projects U.S. Nuclear Regulatory Commission Washington, DC _2055$

Mr, Noel Dudley

-NRC Senior Resident inspector P.O. Ilox 1149 i

Scabrook, Nil 03874

.-a.._,J..,

.L.....,....L--...-.

___...__,m.,..-,--,,i,._._-,.-,..m_.._,-_._~,.m._-

,_.,,_.,__-......J,._,,_,,...,,,i.,.---,#-

. ~ ~. -. _ _. -

4 v

New llampshire. Yankee April 30,1992 k

I?NCl.OSUIZE 1 TO NYN 92058 SWITCII SP.TPOINT IIRI?OR ANALYSIS t

?

s o

.m

..___:.... + -. _,, _. _ _ _ _.., -.,,

.,--.,.,,,,,,-.,...-r...

,, _. ~,. _,

NRC Reauest Motor Operated Valve Testing New llampshire Yankee (NilY) needs to:

• 1!stablish design control measures for error analysis by April 30, 1992 and describe i

the method for determining the overall accuracy of the control switch setpoints."

{

N11Y Response l

4 Technical Support Group calculation 92-CALC 0003 addresses the measurement accuracy of the various diagnostic techniques used by NilY. This calculation has been independently reviewed and approved and is available for review at the station.

+

i The NilY MOV diagnostic testing program includes two major valve groupings, rising stem MOVs and butterfly MOVs. The diagnostic scst method differs for these two groups as the former involves a valve stem thrun mr.asurement whereas the second involves a valve stem torque measurement.

There are four distinct techniques,,ach-with a separate-accuracy capability. Th e r.c techniques are:

Strain Gauge Method (Tensile)

Rising Stem Valves i

Load Cell Method - Rising Stem Valves Load Cell Comparison Method Rising Stem Valves Strain Gauge Method (Torsion) liutterfly Valves The accuracy for each case is determined by the Square Root Sum of Squares Method

.in accordance with ANSI /ISA S67.04 1988. provides additional information on the methods used to determine the overall' accuracy of the control-switch setpoints.

1 r,,. ~ +..,

r,.

..L-,------.-,

~

e

.J.,

4 4

w New flampshire Yankee April 30.1992 i

l

}

l 4

h I

l' ATTAC11 MENT 1 TO ENCLOSURE 1

^

t 4

+

P 4

{.

i

. h t

t 6

+

6 h

?

I c

4 h

4 V

,,-.a.

,,.,,,, 'A.4,,

y

+.

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

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

,,,m_,.y,,,

..%_,,,,,,,,,m-,

.,,m4.,,,

,,.,,,,,...,,cy

.p-,,.mm,,..,,.

4' N,

l N81Y Mctbod for Determinine the Overall Accuraev of Control Switch Setooints 9

'escribed in the NilY NRC Generic Letter 89-10 Program, an ongoing design oasis review i

og performed for Scabrook Station safety related MO\\"s. The design basis review for er en MOV includes a calculation that determines the minimum and raaximum thrust or torque 3

value for that s als e.

The thrust / torque values represent the design range and are locumented in a controlled NHY desit i document,1-NI{Y 250000.

n3

%=

-ian range does not take into account iorque switch repeat + y and diagnostic test

curacy. These factors vary with the test method and ar.onsidered on a valve

.)

is by NiiY Technical Support Group enginects when an MOV enters into the

.b 7e Letter 89-10 test program.

(,j%,'1 developed to determine the T

a Support Group Calculation, 92-C A LC-0003, was accuracy associated with torque switch repeatability, data acquLition, and data d

^

f:,

a: curacies.

'I he combined accuracy value is then used to reduce the design range

~

snectfu in 1-NilY-250000 to a mot restrictive set of values called the " target range' In nis way,

'.f the inget range is ruaintained for a given measureroent, the design tange will g

at be execeded.

fL rhe accuracy factor foi,'

MOV actuator's torque switch repeatability is based on actuator vendor recommendations. The data acquisition and data processing accuracies take into account the accuracies a.sociated v ith:

measur' ; and test equipment, (M&TE),

physical constants for converrJon of strain measurements to thrust or torque, an' cort w ita echnie,ues that associate load cell measurements on the valve open stroke it spring pack displac.nent on the valve's closing stroke.

M V employs tou cencral techniques to test MOVs depending on ll'c valve's configuration and physical arrangement in the plant as tollows:

Strain Gauge Method (Tensile) - Rising Stem Valves Lead Cell Method - Rising Stem Valves i ad Cell Comparison Methoct - Rising Stem Valves Strain Gauge Mer'iod (Torsion)

Butterfly Valves Calculation. 92 CALC-0003 determines accuracy factors for the various data acquisition / data processing configurations used in these techniques for each individual valve.

9LCALC-0003 combines the toique switch repeatability factor with the data acquisition / data processing accuracy facte: asing the " square root sum of the-squares" method described in ANSI /IS A S67.04 - 1988.

1

As mentioneo aove, the resuiting combined accuracy is then applied to the design thrust or torque range to provide the target range valu-s. The target range values are tabulated iri approved station procedure, ES1850.003. Diagnostic testing is performed to confirm that h10V control switch setpoints result in control switch actuation within the specified target

}

range for thru:t/ torque. As the design basis review verification is completed, this process will be carrica out and continued for each $10V included in the NiiY NRC Generic Letter 89 10 i'rogram.

In December 1990, the NiiY A10Y usagnostic system was tested at the Idaho National Engineering Laboratory, (INEL), This testing confirmed the accuracies expected for the NilY diagnostic system. NHY and INEL presently have an on-going program to standardize the diagnostic test system accuracy results to the h10V Users Group TEh1V format.

1

,,6 4 x-2 i

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

4

+

New llampshire Yan c:

Apri' 9, -1992

.i ENCI.OSt1RE 2 TO NYN-92058 STEM FRICTION COEFFICIENTS b

5 y

?

I r y:-

{:..

k l

':(

';c l',

, - o a

...;...,,,~

r

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

_s i-NRC REOUEST

'New llampshire Yankee (NilY) needs to:

" Validate the assumed valve factors or friction coefficients using the design basis test results and justify the use of 0.15 as a stem friction coefficient."

- NilY RESPONSE The New if ampshire Yankee response to the first part of the NRC request to " validate the assumed valve factors ot friction coefficients using the design basis test results" is provided as follows:

s The NHY NRC Generic ~ Letter 89-10 program is being revised to include a methodology which will use the MOV diagnostic field test results to corfirm the design basis calculations. This methodology will feed back actual test measurements

- to recalculate valve factors / friction coefficients and will verify that the values assumed

- in the. design calculations are conservative. This methodology will be implemented

~

by ' Dece mber 31, 1992.

The New 11ampshire Yankee response to the second part of the NRC request concerning

~

-.the use of a 0.15 stem to stem nut coefficient of friction for selected valves, is provided as follow:

' Discussion The ' required torque outpt t of a - valve operator 'is calculated by multiplyirg the required thrust for valve -operation ~ by the stem factor. The stem factor is based on the physical-parameters of the stem (stem diameter, pitch, lead) and the assumed coefficient-cf friction between the stem and stem nut.

Valve manufacturers have

. provided -'the. recommended stem / stem nut coefficient of friction to be used for determining the thrust output for valve actuators. The valve manufacturers' usually provide a coefficient of friction of either 0.15 or;0.20. A 0.20 coefficient of friellon is generally used to take into account the possibility of poor maintenance on the stem threads. In Inspection Report 91-81 the NRC has requested New Hampshire Yankee.

to justify the use oi a 0.15 coefficient of friction.

Presently motor operated valves at Seabrook supplied by Y!estinghouse and Velan and-l, valves having' Rotork actuators use a. coefficientL of friction.less than 0.20.

A basis for the use of the lower friction factors is described below.

Westinchouse Valves n

i Wes9ghouse recommends a 0.15 coefficient of friction. Following the EPRI Marshall ElectricE Motor-Operated Valve (PORV Block -- Valve) Testing, in which the Westinghouse valves experienced higher than expected stem thrusts, Westinghouse l.

. initiated a valve testing program to determine the causes o[the higher than expected stem thrusts. The Westinghouse testi, consisted of three subprograms: 1) water flow 1

i f-i

testing, 2)[ valve - internal-friction testing and 3) valve seat friction factor testing.

Following these tests separate effects testing was performed on the rnotor operators and on the valve design, During thrust / torque testing the coefficient of friction was verified less than the 0.15 design value of coefficient of friction.

Reference Westinghouse proprietary test report WEMD 5672 " Motor operated Gate Valve Closure Problem Flow Testing Program" dated September 23,1o82. This report has a typical graph of the friction coefficient which shows 'a friction coefficient less than _-0.12 for multiple tests (63) under flow and differential pressure conditions.

Velan Velan recommends a 0.15 coefficien; of friction for their valves.

. Velan recommendation is based on testing performed at Velan, their experience and other sources (Rotork coefficient of friction testing).

Velan Test Report No. R D-406 analyzed Velan proprietary test repoi

'o determine the coefficient of friction between the stem and the stem nut. The proprietary testing had bcen performed by Velan to qualify gear actuators. Stem torque and thrust readings were obtained from stem mounted strain gages at 0, 500,1000, 2000 and 2500 stem open/close cycles. Stem diameters of 1-5/8",1-7/S", 21/8" and 21/2" were tested. The stems were lubricated prior to _ testing.

Velan Report RD-406 concluded that the coefficitat of friction experienced during testing was less than 0.13.

Valves with Rotork Motor-Onerators Rotork provided - St m Factor Test Summary of Rotork Technical Reports TR-219 and TR 209. : The tive of these Technical Reports was to determine actual stem factors of two typice threaded stems with' appropriately tapped aluminum bronze stem nuts. Torque and thrust were simultaneously measured under varying conditions of lubrication. Rotork concluded that the Rotork published stem factors (based on a coefficient of friction of 0.14) are conservative for lubricated stems. Rotork found that lack of lubrication had a much greater effect on stem factor than load did on thread efficiency. Rotork performed an additional stem factor test (reference Rotork Technical Report TR-3027 which concluded that the Rotork assumed coefficient of friction of 0.14 is-conservative. These ten results and field caperience uro shown that there is margin between the recommeaded theoretical coefficient of friction aed the actual coefficient of friction.

Rotork ' stated that during the recently completed thrust uprate for the Rotork actuators the observed coefficient of friction was 0.12 - 0.13.

Limitoroue Limitorque Corporation's sizing ~ procedures -include tables of stem factors for two coefficients of friction,0.13 and 0.20 in SEL-10. The Limitorque gate and globe valve selection procedure (SEL-1) states " Conservatively,0.20 is generally used to take into account the possibility of _ poor maintenance on the stem threads.

The actuator 4

2

-.a

_.._m m._ _ _ _. _.. _. _._.

i manufacturer has no control: of the cocificient of friction.

Lower coefficient of iriction may be used if the stem and stem nr1 maierial as well as the surface finish and lubrication warrant, however, the valve manufacturer shall make this judgement?

- EpR1' MOV Aonlication Guide The EPRI Application Guide states

• A thread friction coefficient of C.15 is suggested as a nominal value for in-service applications in the absence of specific test data for a particular application."

Industry Testinc During the recently ccmpleted Thrust Rating increase 'lesting of Limitorque SMB.

000, - S M B.00, SMB-0 and SMB-1 performed by Kalsi Engineering, coeffic~ient of friction was found in the rang: of 0.0S and 0.14 with typical values equal ta 0.12.

Proper lubrication _was maintained during the testing of four differ::nt sizes of actuator stem / stem nut combinations tested to 4,000 cycles each.

Seabrook Testinn in general, the measured thrust output during valve diagnos'.ic testing was greater than

~

the caldulatedi hrust output based on the torque switch setting curves. _New t

Hampshire Yankee practices a formal stem maintenance and stem lubrication program in ' accordance with Station Procedures. NHY plans to measure thrust and torque-during future diagnostic testing where practical to facilitate calculating as tested stem factors. Additionally, NilY performs as found testing, where practical, which provides information on the adequacy of the lubrication prograni.

Co'nclusion The'use of higher coefficient of friction to attempt to bound all eventualities such as stem damage, unlubricated stems ete could result in valve or actuator damate due-to_.

thrusts much higher than calculated being produced. New Hampshire: Yankee follows

~

good valve stem maintenance and valve stem lubrication practices in accordance with Statinn Procedures. NHY is testing motor operated valvesJo verify that the required thru, output _'is achieved. - Future testing will include torque and thrust measurements where. practical to verify that appropriate stem factors are being used.

For the reasons stated above, -NHY will-continue: to use' coefficient or friction values recommended by valve manufacturers -unless : these ' friction factors are proven-unacceptable by valve testing or indt.stry experience proves that a 0.15 coefficient of friction is ' unacceptable.

1 L

3 L

L

... - - ~

. ~.

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

REFERENCES i)

EPRI Report NP-666f' D: " Application Guide for Motor-Operated Valves in r4uclear Power ' Plants 2)

EPRI Topical Report TR 100449:"EPR't MOV Performance Prediction Program Motor-Operated Valve Margin Improvement Guide" 3)

Westingbouse proprietary test report -- WEMD 5672

• Motor-Operated Gate Valve Closuse Problem Flow Testing Program" dated September 23, 1982.

4)_

Velan Test Report No. RD 406: " Analysis of Stem / Nut Coefficient of Friction'.

5)

Rotork

• Stem Factor l Test Summary

• from Rotork Technical Reports TR-219 and TR-209 of-April 1975 6).

Rotork Technical Report TR-3027: " Stem Factor Investiga*, ion"

- 7)

Limitorque Corporation " Gate and Globe Valve Selection Procedures" e

T 4

i i

[

4-1 4k 1

I4-

--m...

m