ML20024H624
| ML20024H624 | |
| Person / Time | |
|---|---|
| Issue date: | 05/30/1991 |
| From: | Kiessel R Office of Nuclear Reactor Regulation |
| To: | Berlinger C Office of Nuclear Reactor Regulation |
| References | |
| RTR-REGGD-01.084, RTR-REGGD-1.084 GL-89-10, NUDOCS 9106070105 | |
| Download: ML20024H624 (42) | |
Text
_. _. _
.. _. _.. _ ~. _ - _.. _ _
((go atog h-UNITED STATES
/
_ ',g -
NUCLEAR REGULATORY COMMISSION.
g;
- r,;
- -l '
W ASHtNG T oN, D, C. POSbs
,,,*...*f v
May 30, 1991 MEMORANDUM FOR:
Carl H._Berlinger, Chief Generic Communications Branch Division of.0perettonal Events Assessment Office of Nuclear Reactor Regulation FROM:
Richard J. Kiessel, Senior Reactor Engineer Generic Communications Bronch Division of Operational Events Assessment Of fice of Nucitar Reactor Rtgulation
SUBJECT:
TRIP REPORT - MEETING OF AMERICAN SOCIETY OF MECHANICAL ELGINEERS (ASME) BOILER AhD PRESSURE YESSEL (Bl.PV)
COMMITTEE'S WORKlHG GROUP ON VALVES (WGV) ON MAY 21 AND 22, 1991-1 Listed in enclosure 1_ are those WGV agendo items-considered to be of special interest to the NRC.
Information on the complete-agendo is available in my files.-
This meetirig of' the WGV was held concurrently with the ASME Comittee on Qualification of Mechanical Equipment Used in Nuclear Power Plants' Subcommit-tee on Qualification of Valve Assemblies-(SQVA). Tne purpose of the joint meeting was to permit the members _.toLbecome familiar with the current activi-ties _of each_of the groups. While both the SQVA and the WGV are interested in the ASME' Code's requirements for valves useo_in nuclear power plants, each has
=its own special areas of interest.
Specifically, the WGV is primarily involved
-in the-requirements associated with the design, fabrication -and inspection during construction of. valves, particularly the structural-and pressure reten-tion capabilities of-the v61vt, itself. However, the SQVA is:primtrily in-volved!inthedemonstrationofthecapabilityofthevalveassembly(i.e.(s)he t
valve and_its actuator taken as a unit) to perform its specified function over tne expected range of internal and external operating conditions. HIGregg
.(Region 1) is the NRC's: representative on the SQVA.
Althcuah the main purpose of this trip report is to discuss the-WGV portion of-the joint meeting, I have enclosed two handouts from the SQVA portion of the meeting.
Both are letters received by HRSonderegger (SQVA Chairman).
CCNTACT:
Richard J. Kiessel, NRR 492-0840 9106070105 910530 PDR ORG NRRBPDR p/
l e
d by 30, 1991 I
e Carl H. Berlinger,
1.
The first (enclosure 2) is EABake's (a new member of SQVA) letter of April 2,1991, which forwbrded a copy of his paper, "Dasign Basis Qualification of Equiwedge Gate Valves for Safety-Related MOV Applicatiuns."
2.
The secund (enclosure 3) is ALMacKinney's (NUMARC) letter of February 27, 1991, which forwarded a copy of their February 7, 1991, letter sent to their Administrative Points of Contact and its enclosures "MOV Activities Update" and NUf' ARC 91-01
" Industry Guidance in Respondina to NRC Generic Letter 89-10, ' Safety-Related Motor-Operated Valve Testing and Surveillance.
7
// k Richard J. K esse 1
,enior Reactor En9ineer 3
Generic Communic 6tions Branch Division of Operational Events Assessrent Office of Nuclear Reactor Ptsulation
Enclosures:
1.
Selected WGV Agenda Items, May 21-22, 1991, Meeting 2.
EABoke's April 2, 1991, letter to HRSonderegger w/ enclosure 3.
ALM 6cKinney's February 27, 1991, lettu to NUMARC Codes and Stanoords DISTRIE,UTION w/ enclosures FJMireglie, NRR WTRussell. NRR CERossi, NRR CHBerlinger, NRR RJKiessel, LRR CGrin4es, NRR AChaffee, NRR KRWicknun, NRR G6Georgiev, hRR TGScarbrough, NRR RJEosnak, RES AJMurphy, RES JTaylor, E00 POR Central Files
,% DCB & i 00EA R/F OGCE R/F i
RKiesst.1 R/F Doce t Name:
RJLIE5SEL MEM0 TO CHBERLINCER OGCB:
L.4 ' R RJKlessel CE/2f/91 -
May 30, 1991 SELECTED WGV AGENDA ITEMS MAY 21-??, 1991 MEETING 1.
WGV 91 Incorporation of Code Case N31-1 into the Code This cooe case, which pertains to elastomer diaphragm salves, h6d contin-gent approval in the previous revisions of Regulatory Guide (RG) 1.84,
" Design and fabrication Cod 6 Case Acceptability. ASME Section 111 Division 1."
During the discussion, ont of the nuclear utility represen-tatives indicated that the limitations to the use of the code case had been dropped by the NRC.
I offered to verify this and to provide the proposer of this action with the lattst version of RG 1.84, 2.
KGV 91 Code Case on use of B-101 Copper Nickel in valycs This material has recently had its allowable stress values developed for use in other nuclear components. how that this duta is available, the proposer will be able to compare it to niaterials for which pressure ratinas have been developed ano propose pressure rating f or the code case.
I will work with GBGeorgiev to ensure that proper values are assigned.
V Itep401 s
Design Basis Qualification Of m
Equiwedge Gate Valves For Safety-Related MOV Applications The Equiwedge Solution By E. A. Bake, Research Manager, Edward Valves Inc.
Introduction Nuclear Regulatory Commission (NRC) Generic Letter 1
8910 has had a significant impact on the U.S. nuclear power industry, and its testing requirements willlead to the need for repair or replacement of many safety-related motor operated valves (MOVs), Nuclear plant ABSTRACT engineers face two challenges: how to quality existing In NRC Generic Letter 8910, the yatves for performance under design basis event (DBE) blowdown' conditions and how to solve problems U.S. Nuclear Regulatory Commission discovered during qualification tests. NRC sponsored states that performance of safety-tests 2 of representative gate valves like those used in related MOVs cannot be reliably many U.S. nuclear power plants have shown that some extrapolated from static or low flow alu e or d ma e de c'nd ns Thus, problems test results. Valves that have passed wiii certainly be encountered witn some conventional these tests have often failed under gate va'ves in the course of tests required by 8910.
realistic DBE test conditions. MOVs The nuclear plant MOV concems have raised legitimate that cannot meet new standards must questions about the re:iability of conventional wedge -
be upgraded or replaced. Equiwedge gate valves. This paper discusses an afternate design.
The Equiwedge gate valve was developed and intro-gate valves, which combine duced by Edwaro Valves,Inc. (formerly part of Rockwell t
advantageous features of both wedge internationai) in 1975, so it is not a *new" design. Since it and parallel slide valves, provide a was not offered for nuclear service until after orders had third alternative to these other been piaced for valves for most U.S. nuclear plants, relatively few Ecutwedge MOVs exist in these plants.
MOV designs. These gate valves from Nevertheless. the Equiwedge design was subjected to Edward are designed and test proven extensive DBE testing to quahfy it for service as a main to perform reliably in safety related steam isolation valve (MSIV), and Equiwedge MSivs are employed in nuclear power plants on three continents.
MOV apph. cations.
Except for use of linear actuators, the MSIV application involves basically the same DBE operational require-ments as an MOV. Two test programs on Equiwedge MOVs have demonstrated reliability similar to that which had been proven in MSIV simulation tests.
Background
NRC concerns with safety related MOVs can be traced back many years but they were intensified by the Three.
Mile island inciaent in 1979, where valves were held partially responsible for the problems. All types of valves have been stuo'ec. but the emphasis has been on gate EdwardValves.nc varves recentiy tecause iney are very wideiy used in o ETR company nuclear c! ants
Of the many basic types of isolation (stop) valves, two old Edw rd Gs:a v:rve resrures types - globe and gate - predominate in high pres-(
~
sure, high temperature applications in power plants.
g i'
There are many applications where globe valves have advantages, such as for throttling and flow control; however, gate valves have straight through flow pas-
)
sages which offer an advantage over conventional globe
~
valves in minimizing flow resistance.
Gate vatves also offer an at least theoretical advantage over globe valves in lower operating force requirements.
e-While a globe valve disk has to buck the full differential i:z p
pressure force head-on to either open or close, the gate J
[
in a gate valve appears to need only overcome frictional forces induced by pressure pushing it against the guides or seats as it slides across the flow stream.
These advantag6s led to widespread use of state of-the-Figure f art gate valves in fossil-fueled power plants early in the twentieth century, and many of the same valves found Equiwedge Gets vs/ve Features their way into applications in nuclear power plants starting in the late 1950s, While the valves for nuclear plants required a
- pedigree
- and sometimes an N stamp, many were of designs that were far from new. Some problems that were only nuisances in typical fossil-fueled plant applications became serious in safety related moter-operated gate valves in nuclear plants.
One major difference in the applications in older fossil-fueled plants was that many gate valves were used for
(.
component isolation and did not require operation against high flow or differential pressure. Some were only opened or closed under shutdown conditions when there was little or no differential pressure, Many had bypass lines with globe-type bypass valves that could be used to " balance
- differenbal pressure before opening or Figure 2 closing of the main valve. Traditional gate valves gener.
ally performed reasonably well in such applications; at hensive technical paper 3. The Equiwedge (Figure 2) is worst, a motor actuator might require an assist from the one of the newest basic gate valve designs in the world, manual over ride to unstick a jammed valve. However, but it has already accumutated an excellent service some of the safety-related nuclear plant applications
- h. story of successful performance in thousands of involve operation under postulated accident conditions applications worldwide. These applications range from that require opening or closing under high differential common manually operated component isolation valves
' blowdown" conditions.
in fossil-fueled power plants through safety related motor operated valves (MOVs) in nuclear power plants.
Edward Valves, Inc., was historically a manufacturer of Combined with Edward stored energy actuators,160 primarily globe valves, but a line of conventional solid large Equiwedge gate valves have been fumished as wedge gate valves (Figure 1) was made from approxi-quick-closing main feedwater isolation valves and main mately 1944 to 1970. While it employed some unique steam isolation vanes (MFIVs and MSIVs) or for other manufactunng processes to provide precision in guiding similar safety-related apphcations in nuclear power the gate and seating, it was not a design that would be plants worldwide, Table 1 lists the power stations where recommer.ded today for operation at high differqntial these valves are employed, along with the year of actual pressure. While it was relatively trouble-free and was or scheduled plant startup.
liked by many users, production was discontinued because it was not suitable for the applications that we'e beginning to develop twenty years ago.
When the Equiwedge product line was first conceptual-ized. Edward engir us literally began with a clean Equlwedge Gate Valve Development sheet of paper." Not a single part from the previous Development of the Edward Equiwedge gate valve Edward gate valves was used in Equiwedge gate valves.
product line was undertaken in 1972, and its introduction Until tne new design could be developed and proven by in 1975 was accompanied by pubhcation of a compre-testing, Edward offered only globe-style isolation valves, 2
t COMMITTEE CORRESPONDENCE Committee:
ASME SQVA Reply to:
E. A. Bake Edward Valves, Inc.
Subject:
Memoership P. O. Box 1961 Raleigh, NC 27602 Date: April 2, 1991 cc: R.
D.
Norden W.
N.
McLean Mr. Henry R.
Sonderegger Grinnell Research & Development Center 1467 Elmwood Avenue Cranston, RI 02911
Dear Mr. Sonderegger:
I appreciated your call last week inviting me to participate in your Subcommittee on Oualification of Valve Assemblies (SQVA).
This will require close review because of the amount of time commitments I already have in ASME and MSS.
Regardless, I will be participating in the joint meeting that is scheduled in May as a member of the Workino Groun on Valves.
I am very interested in qualification of valves for safety-related nuclear power plant applications.
As a member of the Motor Operated Valve Users Group, I.am well aware of the NRC sponsored gate valve tests by Dr. Bob Steele of INEL.
The publicity associated with this testing gave some people in the utilities the impression that valve manufacturers had given little or no attention to qualification.
In fact, significant gate valve work was done by us several k9 T
The enclosed paper, DESIGN SASIS QUALIFICATION OF EQUIWEDGE GATE VALVES FOR SAFETY-RELATED MOV APPLICATIONS, provides a condensed summary.
Since much of this work was done to meet Electricite de
' France requirements, it was not well-known in the U.S.
You were quick to recognize the name of one of my predecessors, Bert Milleville.
He worked extensively on early drafts of B16.41, and he was somewhat frustrated that our qualification work was driven by individual customer requirements rather than a recognized standard.
We have follcwed B16.41 in more recent work, but we foresee that Dr.
Steele's work and other new developments will lead to changes.
In one way_or another, we would like to be part of that activity.
4 l
Very *ruly yours,
]{
- y))
]'
4.../
/
Earl TG Bake APR 0 8 1991 334ESEAPCH 6 DEVELOPMENT GR!NNEL'. C0VOR AT10N I
There was no preconceived direction that the new valves had to be of the wedge type. An innovative group of perefter siso, care verve
'N experienced valve engineers studied all types of gate valves produced worldwide, and their only goal was to produce a better valve that overcame the weak points of prior designs. Parallel-slide and one-piece
- flexible y
wedge" design coacepts were carefully considered and studied before the final Equiwedge design approach was q
decided upon.
CW Conventional Wedge Gate And Parallel-
//
9 WMM#Am P~~"i\\ NWM Slide Gate Valve Features Prior wedge gate valves ana parallel-slide valves each had design strengths and weaknesser.. The fact that y
both gate valve styles had been used for many years showed that users must have been able to tolerate the g
j
,: f
,r,f,,7,77,,,,
weaknesses of each type in many industnal and power f,,,g,y,7 -
g plant applications. However, selection of either type
~
required compromises, u
While both types had advantages that made them suitable for many services, the following features are principal disadvantages of each type:
Figure 4
. Parallel Slide Gate Valves (Figure 4) have the funda-
.Conventiona! Wedae Gate Valves have a bad reputa.
thental disadvantage that the gate and body seating tion for sticking in the closco position. Sofid Wedge gates, surfaces are in continuous contact over the full trave; of can bind due to pipe loading reactions or due to thermal the valve, from open to closed. Tnus, the seating sur-effects (c. g., body contraction due to cooldown after
/ aces are also the guiding surfaces, and any scratches closure with hot fluid). Also, low or high pressure leak.
or gouges that result from wear are a source of leakage age can develop if pipe load reactions or thermal effects when the valve is closed.
create
- gaps
- between the gate and the body seats, Since most such valves in fossil-fueled plants were Valve designs with one piece ' flexible wedges * (Figure normally operated with very low differential pressures
- 3) clearly offer improved resistance to sticking and (using a bypass), this characteristic usually resutted in leakage, but the necessity of meeting acceptable stress only minor scratches; however, some users indicate that criteria in high pressure valves imposes a senous parallel slide gate valves require more frequent seat limitation; one piece wedges designed for high pres.
refinishing than wedge gate valves, sures tend to be rigid, regardless of shape.
The continuous sliding of seating surfaces is e more l
A problem with many traditional wedge gate valves is serious concem in valves that must operate at very high that the gate guides, which must support loads on the differential pressures (such as safety related MOVs, gate before seating, are often too weak to support high MSIVs and MFIVs in nuclear power plants). Since the diff erential pressure loads. Some are too short to high differential conditions may not be encountered in support overhung loads encountered when a gate closes normal operation the problem might not be evident until under blowdown conditions, and some have excessive a design basis event requires a safety related valve clearances that will not provide a smooth transition of operation. With very high contact stresses at gate-to-loads between the guides and seats. These features are seat interfaces, wear damage may result in galling that not a concem in valves operated at low differential can cause large leak paths and possibly even prevent I
pressures using bypasses, but these charactenstics can full valve travel, produce problems in operation under high differentia!
flowing conditions.
Another limitation of conventional parallel slide valves is Flexible Wedge Gare Valve that seat loading is often inadequate for sealing at low differential prestures. Valves that use only spnngs to
&[O.
b assist the pressure load on the gate have very low seat
/
contact stresses at low pressure and are notoriously poor in seahng under such conditions. Some vatves l
employ internal weoge mechanisms to impart additional loading in the closed pcsition, but this requires additional 7--
complexity; generally, they employ large wedge angles that provide load;ng only so long as stem load is main.
w' s
G
'v ~
ta ned.
Figure 3 3
i ~
~
Edward Equlwedge Gate Valve Features
!! Was recognized that these factors had a major bearing The Edward Equiwodge concept development 'bor.
on the successful opera' ion of a wedge gate vatve with
)
rowed' the best feature of normal parallel-shde valves -
reasonable and predictable operating forces, particularly
('
two separate gates (Figare 5) that permrt better inherent under high differential pressure conditions. The design-flexibility and freedom oi each gate to abgn property with ers of the Equiwedge addressed all five factors and its companion body seat. However, by using the tradi.
evolved a design w'th strong pusde rails on each gate tional wedge seat arrangement in the valve body, the half, engaging rupped guide prooves af each side of the j
Equtwedge has a major advantage in low pressure seat valve body (Figure 6). On opening or closing, there is l
t.ealing relative shding motion between the gate and body se ting surfaces for only a Very smallportion of the valve l
nexible Equlwedge Gate travel. Outside this range, all sliding mouon occurs between the hardfaced surfaces on the gate rails and the guide groove in the valve body.
l Unkke many conventional wedge gate vatves, the guide surfaces in Equtwedge gate valves are designed (and j
test. proven) to support high differential pressure load, even when the valve is not seated. Further, the strength and geometric arrangement of the guide system assures 6
a smootn sliding transition of gate loads from the body I
grooves to the seat on closing and from the seat to the body grooves on opening. Signrficantly, even if wear or scratching occurs on these surfaces, there is no adverse effect on seat seahng. Figures 7,8 and 9 illustrate senous jamrring prcblems that can develop in wedge gate valves with ;nadequate guiding systems.
1 Gate. lamming Our To Guide Dietortion Figure 5 l
Equlwedge Gulding System
\\
\\
b!
\\ <\\
!i i
.Z I
l l
_--+
~~}-- a. -
Figure 7 Gate Jamming Due To Upstream HooMng i
i i
Figure 6
~~
i
^
l The most important advantage of the Equiweage gate
-] j l %
valve over conventional wedge gate valves and any parallel shde gate valve is in its gate guiding system
_ ""*. j _ L.
Whi e some parallel shde valves employ suppiementary l
gate guides to minimize cocking, most depena pnmaniy l _.i_ j on a varying contact pattem between the gates anc
~~ ~ NT
- -O"~'
seats to maintain gate ahgnment Wedge gate valves require separate guide systems, but studies of pnce designs revealed that there are five key factors in wedge gate valves guice system cesign, as hstec in Tabte 2.
Figure 8 4
4 9
Hundreds of mrt:riti coupb sampi:s were 1:sted to oste.samming Due To overttung Loads evaluate materials for vanous valve seating and guiding applications at contact stresses to 15,000 lb/in2, This
'q I{'
testing permitted combinations with serious galkng s
tendencies to be
- weeded out/ and numerous
- repeat g (it W,/
tests" were conducted with successful combinations to f
l assure consistent results. Since even the best of friction i
i/
coefficient data have significant scatter, a statistical
_,j\\
" lf1 analysis was required to develop the proper coefficients L.. Ml for use in actuator sizing calculations. The test program lI led to the use of stellite 2f hardfacing on allgate and Ql body seating and guiding surfaces in Equiwodge gate valves with high flow, high drfferential pressure operating
_.__ m I
mquirements in nuclear safety related applications (the combination is standard in all stainless steel valves, regardless of the appkcation).
F/gur,p Equlwedge Gt.te Valve Qualification Tests Special Ec ulwedge Qualificetion Tests For And Experience Nuclear (MSIV And MFIV) Applications Substantial testing was done befute the Edward After the Equiwedge design and laboratory inction tests Equiwedge gate valve product kne was introduced in were complete, it remained necessary to prove that the 1975. The initial test programs 3 clearly demonstrated combination would work in an actual valve under high that Equiwodge gate valves overcame the primary differential operating conditions. Before quoting these disadvantage attnbuted to prior wedge gate valves, valves for safety-related nuclear apphcations, a size f 6 Rigorous tests showed that these valves do not
- stick"in valve with a prototype Edwardlinear MSIV actuator 6 the closed postriorr, they require an opening force to was subjected to flow interruption tests stalting in 1977, "unwedge" the gates, but this force is predictable and Since Edward had supplied special balanced Flite-Flow within the capabikties of vane actuators sized for closing MSIVs for BWRs and PWRs, there was a naturalinterest (this wcdging action is an adeantage, because, once the in exploring apphcability of Equiwodge gate valves in this valve is closed, sustained stem load is not essential for market niche.
O seat sealing). The testing program also clearly demon-4 in extensive testing, the size 16 valve was closed strated thet Equiwedge gate valves provide excellent seat sealing at both low and high d/fferential pressures.
repeatedly under *Dlowdown" conditions, typically in from 3
three to five seconds, discharging air from a 300 ft,
In addition, demanding flow tests were conducted with 1500 osi reservoir. Instrumentation showed that the stem water at pipe flow velocities well above the normal thrust required for closing was consistently within operating range. These tests showed very stable flow calculated predefions based on valve dimensions and performance with freedom from flow inauced vibration, friction coefficient data (friction testing of the selected The fully open gate assembly has clearances in the valve trim matenals had previously proven that valve guiding system, but there was no evidence of any testing with air was comparable to testing with steani),
excitation or *ratthng' due to fluid flow.
Seat letkage tests with air after numerous valve clo-sures at high differential pressure gave excellent results i
l While the initial Equiwedge quahfication program was at both high and low pressure (6. 5 SCFH at 1500 psi, very ccmplete, it was recognized that further testing was less tnan O. 2 SCFH at under 100 psig - essentially necessary before these valves Could be offered with the neghgible air leakage in & large metal seated valve);
confidence necessary for safety related nuclear apphca-seating surfaces show ed negkgible damage, proving that tions. First, since a very large portion of the thrust the Equiwedge gate guiding system performed as required to open or close a gate valve is frictional, it was designed.
necessary to have friction coefficient values that could be used with confidence for actuator sizing-The successful testing of the size 16 proto*voc MStV led to initial acceptance of tht. Edward Equiw".uge gate Use of a
- valve factor" suggested by an actuator manu-valve and stored energy actuator for usi as main steam facturer was not considered to offer the necessary and feedwater valve appkcations in man ' of the nuclear l
rehacihty. Since pubhshed data on friction coefficients of plants h3ted in Table 1. Nevertheless, the increased l
vanous valve tnm matenals in hot water and steam concem for vane safety and reliabihty that emerged in l
enviro,'ments were inadequate when the Edward the early 1980s led to the demand for additional testing Equiwedge gate valve was developed, a fnction test
- on a full scate MSIV in the 1980-1982 time period, a program was undertaken to develop an incependent size 28x24x28 MSIV was subjected to two flow interrup.
database. A special test fixture was designed, permit-tion test programs. This vatve was closed repeatedly 4
ting 'shder bar" triction tests in a pressure vesse'in hot unoer blowdov n conditions discha ging air from a 3
pressunzed water and in saturated steam at tempera-reservoir that had been expanded to over 900 ft. In this l
tures to 550cF.
prog'am6, the salve cemonstrate: consistently success-l 5
1
= _.
fuloperction with shutoff dificrtntiil prcssurcs up o Crose-Section Of Pat'asdes PORV SJoch Vafve 1180 psig, even while sub}ected to simulated linebleak pipe loading and seismic loads. Again, required actuator force was within calculated predictions,.
A second full. scale Equlwed0e gate valve test program j---
was conducted in France 7 on the first size 30x24x30 MSIV constructed for 1300 megawatt Electricite de France PWR plants. These tests did not involve flow f [/
interruption testing, but they included hot functional tests (including thermal transients) that demonstrated smooth opening and closing operation through 600 full-stroke cycles and 400 exercise cycles. As in previous
(
Equlwedge tests in the U.S., there was no evidence of
(
valve ' sticking" in the closed position. The tests in x
V --
France also included static loading, seismic and reso-i, nant frequency search testing that provided further O'M y a tw 6
4 assurance of safe valve operation under extreme
)
conditions, Equlwedge MOV Qualification The initial Edward interest in applying Equiwedge gate valves in safety-related nuclear service was in MSIVs and MFIVs, because this had been an Edward niche, af gf u
C g
Since Edward did not offer its older gate valve design for f
h nuclear applications, the Edward participation in the
/
=
MOV market was primarily with globe valves, Neverthe-less, the' success of the Equiwedge design in MSIV
'j h N g %gj quahfication stirred interest in using these valves with electric motor actuators Of course, Equlwedge gate
?
ir valves had been used with motor actuators in fosil-4 3 4
mfg fueled plants from the time the product line was intro.
QK duced, so there was good commercial MOV experience to draw upon, yNC The application of a gate valve as an MOV invnives one k fy '
subtle difference from applications with linear actuators v
like those used with MSIVs, in the MOV, the stem thread I bS i
).g friction torque must be reacted somewhere to produce t
linear stem motion. In the Equiwedge gate valve, the Cxx N w' Y sxx%
torque reaction is intemal, and analyses and tests show that the effect is minima! with proper gate guide system geometry. Equiwedge gate valves have shown excellent performance in tests simulating requirements for block Figure 70 valves for power operated relief valves (PORVs).
The PORV block valve application is quite similar to the The Palisades PORV block valve, Figure 10, was MSIV application, except that the block valves are subsequently required to be subjected to proof testing on smaller and the PWR primary system block valves must steam, and it operated successfully in blowdown tests at operate at up to 2500 psi differential pressure as com.
Wyle Laboratones (Norco, CA) in 198910; stem thrust pared to about 1000-1200 psi for a BWR or PWR MSIV, measurements showed that the required closing force
. A standard size 3 Equiwedge gate valve (except for was within the bounds of Edward calculations. Figure 11 hardfaced body guide grooves) was subjected tp PORV illustrates a plot of actual measured closing force versus block valve simulation tests (sponsored by EPRI) at the time in one of the Wyle tests, compared to a calculation Duke Power Marshall Steam Station in 1960, and results based on Edward friction data. The required force during were quite successful 8. Following those tests, special much of the valve travel was much less than the calcu-size 3 PORV block valves were fumished for Duke lated force due to dynamic effects that cannot be rehed Power Catawba Units 1 and 2. In addition, a special size upon to assist in the final closure of a gate valve, but the 4 Edward Equiwedge PORV block valve was built, required force to complete closure compares favorably quahfied and supplied for the Consumers Power with the calculation The valve closed fully and sealed off 9
Palisades Nuclear Plant.
tightly with a satisfactory closing force margin.
6 t
Valve Tht'ust x 10*3 lbs 12 Calculated Thrust gg 10 8
Nessured Thr'uet
/
g
,,',e,
a.
7
/
6
/,e*
5 4
3 1
0 O
2 4
6 8
10 12 14 16 18 20 Closing Time Inittel Pressure = 2l00 pel Terminal Pressure a is00 pel Saturated stees et 700 degress F
~ Figure 11 Knowledge developed in recent years has shown that' service. Although much of the U.S. nuclear plant con-electric motor actuator sizing and torque switch settings struction was already committed to other products, require consideration of rate of loading and similar ~
nuclear plant construction continued in other parts of the 4
effects, but the basic Equlwedge gate valve design that world, and the Equiwedge design has become widely :
. was well-qualified for MSIV and MFIV applications has accepted for entical MSIV and MFIV applications.-
- been proven reliable for MOV applications as well. The.
Equiwedge has developed a solid track record, including PORV block vatve tests were considered as 'confirma.
tests in safety related MOV applications.
tory" to Edward engineers who had been involved with
_ previous Equlwedge gate valve tests in the U.S. and As a result, Edward can offer an existing product to meet France.-
the challenge of NRC Generic Letter 89-10. With Equiwedge, nuclear plant erigineers can replace ques.
Summary And Conclusions tionable safety related MOVs, within the required Conventional wedge gate valves, while time proven for compliance period with a product ahedy proven and fossil fueled plant applications. have been found inad.
qualified to satisfy the most demanding requirements.
equate for many critical safety related MOV services in nuclear facilities. Unlike fost,il fueled services, where 3
- these valves are generally cycled under low differential -
pressures, these nuclear applications require reliable operation with high differential pressure. This change
. required new onswers from the valve manufacturing i
irdustry, and Equiwedge is Edward's answer.
In addition to the results of the test programs described in this report and in the references, field experience with Equiwedge gate valves has been exco!!ent After well
_ over a decade of service in both fossil fueled and nuclear power plants. Edward Equlwedge gate valves have earned an excellent reputation for trouble tree performance.
Edward spent most of the 1970s evaluating gate valve design options and proving the performance of the I-Equiwedge valve design before offering it for nuclear
.7-
i Table 1 i
Edward Equiwedge MSIV, MFIV And Similar Applications In Nuclear Power Plants i
l Valve Startup Size Pressure Class Plant Country Year Quantity 5
,)..
LV.
~
18x16*18 2500 Superphenix France 1983 8
g
- - r l
8C f'I' I ' 1, d '.
^ ;
(
(
j i
l l
{ ~~ >.,y. ~,
i 1
a x$
s Y-
,\\
4.
y gq.
g J.,
..\\.
t I
.t o
. < ' 'd.
..l.
l
^ '.
7.'~
'g ;'t; -
y
.r.
,4..
y
}
~
)
l s,
T.
s..-
4:
.1
.c
~
,.i.f.-
l g
32x24x32 900 Quangdong China 1992 6
l 1
i n
20 2500 Superphenix France 1983 4
l
- -.y
^
4%i,4 -
,t
... g. e i,a;= <
- wn.
y.iuve. -
4.,,...,
- p 4
l 16x14x16 900 Kori2 Korea 1983 2
e Kori 3 Korea 1984 3
Kori 4 Korea 1986 3
Yonggwang1 Korea 1987 3
Yongg*ang 2 Korea 1982 3
~' _
' ' s'q;Wa$ 6-
< # m. r.m ;
c.
~
4 p%r.
4*
paqh,g.g,
~-
$e y
,,.-t.
e 24 600 Bruce A Canada 1986 4
Bruce B Canada 1997 4
=
- **h * #,[
%f.e,,, *'
'e h
-. 4.,'
.. $nw
.,,;,.],,,,.gg,,,
3
~
...A.
r.
....t x
l
/
i l
8
. -. - ~. -. -. -..
I
+
l r
Table 2 Key Factors In Wedge Gate Valve Guide System Designs i
- 1. Guide Strengm - Where a high differential pressure
- 4. Gui@_LCDgm - The length of the guides and the exists across the gate when it is not in contact with the elimination of ' overhanging loads
- Is important to avoid body seats, the guidos must support the full differential gate cocking and
- locking
- in the guides.
pressure load.
S. Guide Surf ace Matelia'$ - The inction coefficient l
2, GatcStrenath - When all of the differential pressure developed between these surfaces in a hostile fluid reaction loads are on the guides, the gate becomes a environment affects operating forces, and indentation simply supported beam.
under high localtzed load may lead to jamming.
l
- 3. Guide Precision -The precision of the guide system NOTE: Parahel shde pate valves depend on seating
- determines how the gate breaks and makes contact with surfaces to provide most of these entical functions.
the seats during opening and closing.
References l
1 Sa'ch;.Relgig12a!ygJritc;u,n1.htyedan g Washington, DC: U S. Nuclear Regulatory Commission, Genonc Letter y
89 t0, June 28,1969
.A s
2 Egtu!ts of NRC-Soonsored Testinc of M; tot Onerated Yaac5 Washington. DC: U S. Nuclear Regulatory Commis-sion. NRC Informat on Notice No. 90-40. June 5.1990 3 Roger D Norden. ' Development of the Rockwell Edward Equiwooge Gate VaNo
- Edward Technical Article V. Rep 75 5.1975 3
4 E, A. Bake and R. L Clapper,*Ouick Closing lse:1 tion 1
3 e'
Valves-The Equiweoge Attemative
- Edward Technical Ilr E.A Bake Article V-Rep 78 4.1978 i
5 E A Bake *The Rockwell Stored Energy Actuator -
Deve:opment and Quahficat.on
- Edward Technical Article 1
V Rep 80-3,1980 6 E. A. Bake and J. D Gallagher,"Outck Closing Equiwodge Isolation Vatves.. Ongoing Qualification. Edward Technecal e
I Article V Rep 82 2.1982 7 E. A. Bake and Did er Thevenet,'OuicbClosing Equtwedge Isolation Valves.. Global Oaahfication." Edward Technica Article V Rep 85 2.1985.
8 EPR! Marsha" Ele;tnc Motor Ocerated Va!ve @ock Va!ve)
(nggggy;tgeprd Pato Alto. CA Electnc Power Research Institute. Report NP 2514 LD, July 1982.
9 Je'frey H. Bond and Joseph B Gallagher. *Ouatrfication Program for a Motor Operated Gate Vatve
- EPRI Power Piant Va've Symposium. Charione, NC August 2123,1990 101c1812z111;_O3 Narco. CA. Wyle Laboratories.
December t 8.198
~
9
m 3
l l
l EdwaroValvesinc o BTR company 1900 Soutn Saunders Street Ra!eigh, North Carolina 27603
- 800/225 6989 919.'832 0525
NUCLEAR MANAGEMENT AND RESOURCts COUNCll N W e Sa e 300
- Womnyon DC 20006 244 v
5776 (g Street
(?C2) 672128:1 february 27, 1991 To:
NUMARC Codes and Standards Contacts Since some of your committees are interested in motor operated valve testing and survelliance, I am enclosing a letter which Tom Tipton sent to the NUMARC Administrative Points of Contact transmitting a NUMARC report relating to the NRC Generic Letter on this subject. A copy of the guidance document, NUMARC 91-01, Industry Guidance in Resoondino to NRC Generic letter 89 10
" Safety-Related Motor Ocerated Valve Testino and Surveillance" is also attached.
If you have any questions on this subject, please contact the individuals listed in Tom's letter or me.
Sincerely,
- k. !., J Arland L. MacKinney Senior Project Manager g,i g n ; l y *k. m,
- ~
r
.?
t
- .=a o5 t%1
.is 2:SEArcu7. Mytteptf[M 0T' T,'C N DFI.il0N
f.--=-
~
4 NUCLEAR MANAGEMENT AND RESOURCES COUNCIL 4776 tye $tteet. N W e s/te 300
- Wosrvrport DC 2000c 24%
(202) B721260 February 7, 1991 TO:
NUMARC Administrative Points of Contact
SUBJECT:
NUMARC Guidance Document for Generic Letter 89-10, " Safety Related Motor-Operated Yalve Testing and Surveillance
- The purpose of this letter is to provide you with copies of guidance document NUMARC 91-01, Industry Guidance in pescondino to NRC Generic Letter 89-10 "Safetv-Related Motor Ocerated Valve Testino and Surveillance
- and to bring you up to date on other motor-operated valve (MOV) related activities.
Following distribution of the draft Guidance Document addressing Generic Letter (GL) 89 10 and receipt of industry comments, we have been working with the industry and the Staff to resolve issues that were raised during the NRC sponsored wcrkshops on the issue. As a result of interactions with the industry and the Staff and the NRC's issuance of Supplement 1 to GL 89-10, the guidance has 'een finalized.
Enclosed for your use and information are four copies of the final guidance document addressing GL 89-10. As noted in the introduction, the document is offered for your voluntary use as appropriate.
Please ensure that 1
the appropriate personnel within your organization, dealing with GL 8910, receive a copy of the enclosed guidance and MOV activities update.
As you know, EPRI is continuing to move forward with their MOV Performance Prediction Program.
Several short term activities have been completed (e.g., the review of the INEL test program) and others are on going.
Planning for and
. activities for long-term work is also well underway. We have worked very closely I
i with EPRI in structuring the program to focus on the additional information to
. satisfy the recommendations of GL 89 10.
Because of the importance of this l
issue, EPRI is making this program an open participation activity available to all nuclear utilities regardless of their membership status with EPRI.
Should you have any questions regarding the guidance or the enclosed Activities Update, please contact Clive Callaway or Warren Hall.
Si
- ely, t
tjfM]IaD omas E. Tipton D rector Operations, Management and Support Services Division TET/WJH:amw Enclosures cc: NUMARC Board of Directors (w/o enclosure)
HUMARC Executive Points of Contact (w/o enclosure)
2/M/91 MOV ACTIVITIES UPDATE In October 1990, a meeting was held in Germany to exchange information on pumps, check valves, and motor operated valves (MOVs) with the German regulator and several German utilities. The majority of these discussions focused on HOVs. The German test effort has seen results very similar to the
.NRC's test program; i.e., higher than expected friction factors and tilting.
They are also aware of the hydraulic lock up and spring pad. relaxation issues, but indicated that their actuator design precluded these problems. One item of interest is tnat the Germans seat all of their gate valves using limit control, not torque. The limit control is backed up by a series torque switch.
They have concluded that the only reliable application for wedge-type gate valves is in low pressure systems where leak tightness is necessary.
At their new Konvoi plant, out of 250 safety-related MOVs, approximately 20 are gate valves and most of those are parallel disk gates, it appears that
'neir older plants continue to use wedge gate valves in similar configurations as in the United States.
It is our understanding that the NRC will be requesting from licensees, during the audits on Generic Letter 89 10, documentation to support the The Motor-accuracy of the diagnostics used in their plant for HOV testing.
Operated Valve Users Group (MUG) is preparing to perform validation testing of the major vendors MOV diagnostic equipment.
Members of the MUG will write the The validation test program which will be reviewed by Kalsi Engineering.
The testing will be performed at the Idaho National Engineering Laboratory.
purpose of the testing is to validate the accuracy of the MOV diagnostic The results of the testing will be documented in a MUG report and equipment.
made available to the industry. We anticipate that this validation testing i
should satisfy the NRC request.
The NRC sponsored an NRC/It!EL Gate Valve Test Program (Phases I and ':1) designed to test the capability of MOVs commonly installed in BWR High Pressure Coolant Injection (HPCI), Reactor Core Isolation Cooling (RCIC), and Reactor Water Cleanup (RWCU) systems to close under postulated blowdown isolation conditions.
As part of the EPRI MOV Performance Prediction Program, l l
f-W The purpose EPRI sponsored an independent review of the NRC's test program.
of the EPRI program was to review the NRC/INEL test data with the major focus being on the evaluation of the " disk factor" for valves being stroked under differential pressure and to physically inspect the valves, observe the damage and take physical measurements of valve clearances. Copies of the final EPRI report were sent to the NRC on January 21, 1991, for their review in prepa-The results ration for a meeting with EPRI and NUHARC to discuss the results.
of this EPRI evaluation will aide in developing lessons learned from the INEL tests to better plan future activities for their MOV Performance Prediction
- Program, l
l r-l 4
2-
,.,w
--c
.,.m
1 l
l Industry Guidance in Responding to NRC Generic Letter 89-10
" Safety-Related Motor-Operated Valve Testing and Surveillance" l
January 1991 i
Nuclear Management and Resources Council, Inc.
1776 Eye Street, N.W.
Washington, DC 20006-249E l
.l Industry Guidance in Responding to NRC Generic Letter 89-10
" Safety-Related Motor-Operated Valve Testing and Surveillance" l
January 1991 a
I L
ACKNOWLEDGEMENTS The Industry Guidance Document in Responding to Generic letter 89 10, ' Safety-Related Motor-Operated Valves Testing and Surveillance' was developed by the NUMARC Ad Hoc Advisory Comittee on Motor-Operated Valves.
NOTICE This information was prepared in connection with work conducted by the Nuclear
- Management and Resources Council,.Inc. (NUMARC).
Neither NUMARC nor any of its employees, members, participants or. consultants make any warranty, expressed or implied, or assume any legal'11 ability or responsibility for the-accuracy, completeness or usefulness of any information, apparatus, product or process disclosed in this document, or represent that its use would not infringe privately-owned rights.
~
.g -
TABLE OF CONTENTS Section Eggt I
Introduction.........................
5 1.0 Scope of the Guidance Document Z.n' Scope-of' Generic Letter.................
7 3.0' -Design Basis Review and Determining Correct Switch Settings..-..;............. 9
-i 4.0 Verification of Switch Settings...-.........
21 5.0 Post' Maintenance and Periodic Testing...-......
25 6.0 Documentation....................... 31 7.0 S c h ed ul e.........................
3 3 i'
References...,...,......................
37 i
L.
l:
i
~. _ _ _ _ _ _ _ _ _ _ _ _.. _ _ _ _ _. _. _
l,, '
INTRODUCTION j-In Generic Letter (GL) 89 10, ' Safety Related Motor.0perated Valve Testing and Surveillance" and Supplement 1 of GL 8910, the Nuclear Regulatory Commission (NRC) recommends that
- sensees develop a program to coordinate the design basis review, baseline testing and periodic testing of motor. operated valves (MOVs). The generic letter limited its recessendations to the lic.ensee's design basis.
Supplement 1, however, stated that the recossendations of the generic letter may be beyond the design basis of most i
, plant.s and recognized that the guidance provided constituted a generic bakfit in the following areas:
Consideration of the operability of position changeable valves o
located in safety related systees; HOV testing beyond Section XI of the ASME Boiler and Pressure o
Code; o
MOV testing at design basis cor,sitionst t
l Preparation or revision of procedures; and o
Maintenance of records of test methods, failures, and corrective o
actions regarding failures.
Supple. ment I changed the following recommendations that wer. :9ed during the NRC workshops:
MOVslocatedinductwork(i.e., dampers)arenowexc1dedfrom a
the scope; Weir gates and sluice gates are now excluded from the siope; o
MOVs that do not change position during any accident scenario and I
o are inhibited from being mispositioned electrically may be excluded from the scope; c
1
,- w-m wer yv,,,,,.rw-y,..v,-,=--c,v,w,,,,w,-e-yws-%,m,.c
..c y-
,- rw-. m m me -,-=,,v mmew - - w r me. e
_,-,ww.w-w..
..-w---eww-e--ye-r y-1
o, MOVs whose position has no bearing on any system operation during o
a plant transient may be excluded from the scopes and Similar type test data along with static testing are acceptable o
alternatives to design basis testing if sufficient technical justificationisprovided.
The NRC is encouraging the industry to test as many MOVs as practical at design basis conditions. The NRC prefers dynamic testing to predictive analysis to detemine the proper setpoints for switches. However, alternatives to testing can be acceptable even for testable valves.
Therefore, developing qualified methods of justifying data will ht key to reducing costs and establishing an acceptable MOV program.
The Electric Power Research Institute (EPRI) is perfoming research
- aimed at developing a methodology to predict MOV performance. A predictable meth'odology would reduce the need for testing at high differential pretsures.
The EPRI program will examine what effects the differential pressure, flow and temperature have on the required thrust to operate the valve.
In addition.
l dynamic effects on available actuator output such as variations in the rate-of-loading are under evaluation. Expected results from the EPRI 'Perfomance, J
Prediction Program
- ares l
c Writing an in $ttu Test Guide; identifying presently predictable valve designs / applications; o
Developing criteria for determining similarity between valves o
o Developing guidelines for justifying test dat:1 Developing criteria for predicting anomalous valve behaviors and o
Compiling an MOV characteristics database.
o As noted in-the generic letter, the NRC's intent is to ensure that an effective MOV program exists.at esch nuclear power plant.
Because such a-
~
program is complex, involving various departments and organizations, each utility may consider establishing a central point of contact with overall-responsibility for the pr0 gram. Valuable infomation on program control, training, and other MOV program elements can be found in the ' Motor Operated i
l 2
+
~m-s-,v.,m..w.y,.m..,_J
.-,,,,,_,.,m
,_,,,,..,...romm
,,..m_m.v.,
,,,.,,.4,,.,,,,,,,mm..,,,,.,,
,,,, ~,,,
.,,.-_,,,,r-,,,
.m.,
,-c-,--,,.---,--,-,
_. y,_
.p,, r ~ _, _, -. - _. _ -
Valve Performance Update,' issued by the Institute of Nuclear Power Operations
(!NPO) in October 1988.
The INPO report provides a compilation of MOV program elements used at selected utilities with low MOV failure rates.
To allow time to complete the research, the generic letter pemits a two stage approach for testing MOVs.
In the first stage. the industry is to I
use current analytical and testing methods and best available data to determine and verify switch settings of MOVs.
In the second stage, the industry 15 to use the results of the ongoing research programs to further correct the setpoints for the MOVs, where necessary. Figure 1 illustrates the approach being utilized.
NUKARC developed this document to assist utilities in responding to Gl. 89-10. The interpretations and positions taken in this document were developed by the NUMARC Ad Hoc Advisory _ Comittee on Motor Operated Valves.
The objective is to address the action items of the generic letter and to elaborate on methods that the utilities can use to follow its recomendations.
These guidelines are offered to NUMRC members for their voluntary use as appropriate.
(
6 r
3
..:..I-i :L. ~- ::-~ - ~ z:: ^
- =-
= =. =. - - ----
4 0
g m
twEs f
ssE8 g
e s
8 g
82wB e
I
.lt y
,W}
R "g 8
sn E
., e gN g
If wf[
g!Et IN I"
a 8
wI E
g l
5 I
I I
e
!"5
!?E E
E!
sg!< g!
N@
E a
E
-s a
mj
[sg s
I gr m
a
-r I
s 5
9 em gg E
g e
-)
eg-5 e
{g
.lll lggl..g E
.Eg j
E g
s
-s r,a
- -l a
m g
g*,s e
x.
g
=E rg legBE
- Er 5 !
Es W
E.
EEEE 15
"~
c s
a E
W E...
lE Es ga g
e
=
I e
o e
e i
g E_!,
v
-. ~
EN b3 85 W "E Il Ep WW s.
s 8
I I
4 m
e-r
---g
-y.
--9 m
enm -
4,-.,e
.eH m.
w,,,
,-,=,.e-q
1.0 SCOPE Of THE GUIDANCE DOCUMENT 1
This guidance document consists of the following sections:
o GENERIC LETTER SCOPE: Clarifies the scope of the generic letter.
Provides criteria that can be used to exempt certain HOVs from the program.
o DE51GN BA5!5 REV!EW AND DETERMINING CORRECT SWITCH 5ETTING5:
Outlines methods for detemining the design basis conditions, ensuring the compatibility of the valve with the actuator, and verifying the appropriate control circuit logic and cable sizing. Discusses approaches for selecting and setting the switches and describes the application of each switch in the
- actuator, o VERIFICATION 0F SWITCH $ETTINGS: Discusses methods for demonstrating correct switch settings for MOV reliability at design basis conditions, o POST NAINTENANCE AND PERIODIC TESTING: Provides appropriate periodic and post maintenance testing requirements and intervals. Discusses trending and root.cause analysis, o DOCUMENTATION: Outlines documentation, procedural recomendations and reporting requirements for generic letter comp 1iance.
o SCHEDULE: Describes factors that may influence the schedular recomendations of the generic letter.
5 1
N e
e a
s j e
I i
I l
1 l
9 6
9 e
f G
i t
s 6
g, y
n.y..$
p p+v"#7' i
,pe,y qy V."WW
~
_,_.,,.py-da-ge 1
M"T
- N N p,q-m.eq-3et***'T
- T#T
2.0
$ COPE OF THE GENERIC LETTER The NRC intends that each utility consider in its program only those MOV operations and events that are included in the existing approved design basis. However, the NRC does recomend that utilf tles include in their programs consideration for sispositioning of passive MOVs.
NUMARC believes that MOV sispositioning, coincident with a single active failure, say be beyond a plant's design basis. Each utility should consider ANS $8.9, ' Single Failure Criteria for Light Wate:-
Reactors Safety Related fluid Systems' to determine its applicability to an MOV's design basis.
1 2.1 MOVs included in the scope:
Active valves that have both open and close functions during o
design basis-accidents (DBA);
o Active valves that must respond in only one direction during DBA; 6
- Passive safety-related valves that are normally in their o
proper position for DBA; and
- Motor operated valves that do not receive a safety actuation o
l signal but are located in a safety related system.
Z.2 MOVs that can be excluded:
l
- Passive safety-related valves that are in their proper o
position for DBA-And are inhibited from inadvertent electrica)
[
mispositioning from the control room; e
Indicate valves that need only be evaluated if a utility chooses to consider mispositioning.
j 7
l
--w-v
,, c
-w
,,f.
-,m
.-,-ym*.,
.-,w.,-w g-c.
,e-p..,-
5--. - - -
w---g--
y r--
. +,
' Valves located in a safety related system that do not receive o
a safety actuation signal lad are inhibited from inadvertent electrical mispositioning from the control room; Valves whose position is not relied upon during a DBA or in o
emergency operating procedures; Motor operated dampers located in air duct systess; and o
o Sluice gates and weir gates.
Acceptable methods for inhibiting inadvertent electrical 2.3 mispositioning include:
1
\\
Elcetrically blocked in the safety position, e.g., locking open o
the breakeri or Positioning requires dual operator action, e.g., key-lock o
switcli on the control board with key administratively controlled or cover over control board switch.
Active safety related MOVs that respond in only one direction are 2.4 only required to be tested in the response direction. However, if a utility chooses to consider mispositioning, one should verify by testing or analysis that the MOV will not be damnged by mispositioning to the non accident position.
Indicate valves that need only be evaluated if a utility chooses to consider mispositioning.
8
- l DESIGN BASIS REVIEW AND DETERMINING CORRECT SWITCH SETTINGS 3.0 3.1 Design Basis Review Action items a. and e. of the generic letter reconnend that utilities review and document the design basis for each MOV included in the Suggested items for review should include the following:
scope.
I Determining the design basis pressure, temperature, flow, o
differential pressure, and the valve's normal and safety positions; Determining the " weakest link' between the motor operator and o
the valve; Reviewing Control Circuit Logic; o
o-Determining Degraded Voltage; Reevaluating Seismic Calculations; o
Reviewing Overload Sizing; and o
o Reviewing Cable Sizing.
I i
Determining the design basis pressure, temperature, flow, 3.1.1 differential-pressure,' and the valve's nomal and safety positions.
I Review pump and system curves to' detemine saximum differential pressure (delta P) and flow for the valve. Relief valve setpoints wt11 also determine limits.
Rev'iew normal Plant bperating Procedures and Emergency Operating Procedures and determine the maximum delta P which would result from implementation of these procedures.
9 i
a.
4 Review the Final Safety Analysis Report (FSAR) ard other appropriate j
licensing and design documents for maximum delta P, pressure, temperature, and flow used in the design basis accident analysis.
Care should be taken to look at the resultant delta P at the valve (component) versus the system to obtain the correct delta P for the specific valve.- If the MV's design function is to isolate a break in the line, this should be considered in the analysis.
Review valve orientation and physical configurations that say result in flow disturbances.
Review valve and actuator vendor documents for valve design criteria and compare this information with the results from above. Valve vendor component drawings and Bills of Material routinely contain design delta P, pressure, and temperature. Vendors may have used system design pressure (piping class) for component design delta P.
Additionally, fluid type and phase should be known and documented.
NRC and industry research may reveal that fluid type an,d phase have an effect on determining final thrust requirements.
3.1.2 Determining the ' weakest link" between the motor operator and the valve.
Valve limitations can be obtained from valve manufacturers. The valve manufacturer can typically supply information on the weakest link component for both valve opening and closing. The stem, stem disk connection, valve seats and bolting are comon limiting valve components.
For rising stem valves, some utilities have based the valve structural limit on the calculated stem thrust at the valve code pressure rating.. Original valve factors and stem factors are used in this calculation.
This method typically yields conservative results that could prove adequate in many situations.
10
i j
Ensure that the vendor is providing weak link information for pressure retaining parts based on ASME code allowable stresses, not i
yield stress or ultimate stress limits. The year and addenda of the code must be consistent with the original code data package on the valve to avoid code data package changes. When analyzing non-pressure retaining parts, some reasonable allowable stress should be utilized such as the AISC Steel Construction Specification or ASME code Section NF, which is a slightly modified version of the AISC Specification, l.ine pressure is an input to these calculations and should be verified to envelop any higher line pressures that may result from considering valve mispositioning. Actuator limits are based on the vendor's rating for nuclear service.
3.1.3 Reviewing Control Circuit logic.
NOTE:
Industry wide standards are not established for MOV switches.
Therefore, it is suggested a convention for each MOV application be adopted. The convention should be justified and documented.
Ensure that the logic for the torque switch and position limit switch is. appropriate for the application, e.g., butterfly valves are rarely
- torque-seated; they'are usually limit opened and limit closed.
Document whether the MOV motor operation is stopped by the torque switch or the position limit switch.
Ensure that the circuitry for the interlock and indication limit switch circuitry is appropriate.for the application.
Position indication.for MOVs with a two rotor control scheme will be erroneous if the torque bypass limit switch is set to bypass the torque switch-for a significant portion of the valve stroke.
11 l
L i
.aw, a
3.1.4 Determining Degraded Voltage Action item f. of the generic letter recommends that each utility evaluate the capabilities of the motor at design basis degraded voltage.
it is acceptable to use in the analysis the steady state degraded voltage condition at the time the MOV must operate within the accident scenario. This value may be different from the prescribed design basis degraded voltage. An actual test at degraded voltage is not necessary.
3.1.5 Reevaluating Seismic Calculations Stem thrust values may have been included in seismic calculations.
Evaluate the thrust / torque values used in the design basis for seismic calculations.
3.1.6 Reviewing Overload Sizing Ensure overloads are properly sized or configured. (Refer to Reg.
Guide 1.105, R...
1, and NUREG 1296.)
Some plants have the overloads jumpered out completely while other plants have the overloads jumpered out only under certain situations.
The setpoint or configuration of the overloads should be documented and justified by each utility.
3.1.7 Reviewing Cable Sizing Ensure the power supply and cable sizes are adequate to provide the high starting currents needed for MOV operation. (Refer to the August 1988LisitorqueMaintenanceLetter.)
Caution should be used when sizing cable for DC motors. The motor's L
reversing action causes the armature current to be carried by four 12
lengths of cable. Additionally, in some cases, motor cabling has been sized based on motor nameplate full load amps (running amps) rather than locked rotor amps.
NOTE:
Items 3.1.2 through 3.1.7 should be reviewed again after I
initial determination of switch setpoints to ensure that the original design criteria have been maintained.
I 3.2 Determining Correct Switch Settings i
Action item b, of the generic letter states that one purpose of the generic letter is to ensure that a program exists for selecting and setting valve actuator switches.
No industry wide standards exist for_ selecting and setting the actuator switches._ Therefore, more than one correct method may exist for setting these switches. This section describes the functions of the switches listed below and presents examples of various switch setting methodologies.
Torque Switch Bypass Limit Switch o
o Valve Stem Position Indication Limit Switch o Valve Position Limit Switch o Torque Switch NOTE:
For a detailed evaluation of detemining required stem thrust refer to the NMAC/EPRI, " Application Guide for Motor Operated Valves in Nuclear Power Plants.'
l i-13 l
3.2.1 Torque Switch Bypass Limit Switch (TSBLS)
Function The function of the TSBLS is to prevent torque switch actuation from interrupting the actuator motor operation. This is accompitshed by placing the TSBLS in parallel with the torque switch in the control circuit. As long as the TSBLS is enabled, torque switch actuation will not cause the control circuit to drop out the actuator motor contactor. However, there is no locked rotor protection from the torque switch if motor stall occurs while the TSBLS is enabled.
Ocen Direction Loaic - The TSBLS may be set to bypass the torque switch during the valve unseating sequence to prevent a spurious trip.
Initial valve unseating loads are affected by internal sliding friction, differential pressure and seating loads induced from previous valve closure. The TSBLS may be enabled to bypass the torque switch up to the entire valve opening sequence.
Close Direction Logic The TSBLS may be enabled up to approximately 98% of the valve closure stroke until seat overlap occurs. At this point, the TSBLS is disabled and the close torque switch is enabled..
This logic ensures that the actuator motor operation will not be interrupted before flow has been stopped.
Considerations For TSBLS Settinas - Whenever the TSBt-5 disabled, 8
the torque switch setting should be high enough % v
<alve operation at actual design basis conditions.
Since initial unseating loads (hamerblow and cracking) may be.
unpredictable, it is recommended to bypass this portion of the stroke.
Loads due to, flow and differential pressure will differ from the static situation.
Butterfly valve torque requirements cue to hydrodynamic effects typically increase at some mid stroke position. Depending on the 14 l
valve design and process conditions, the hydrodynamic effects may require more actuator torque than the hydrostatic effects, i.e.,
seating and unseating.
-l i
When the TSBLS is enabled for the entire valve stroke, some fom of diagnostic monitoring and/or thermal overload protection should be considered. Without diagnostics, an increase in required thrust due j
to degradation will not be immediately apparent.
Two rotor limit switch designs create many limitations in TSBLS
}
setting logic, e.g., it does not allow a bypass of the open torque switch on limit seated valves to overcome hamerblow.
3.2.2 Valve Stem Position Indication Limit Switch (VSPILS)
Function The function of the VSPILS is to provide remote indication of the valve stem position.
Indication is typically provided for the fullyopenorfullyclosedposition(redonlyandgreenonlylights, respectively). The exact valve stem intermediate position is unknown because both the red and green indication lights are either off or illuminated during the entire transient portion of the stroke.
In addition to remote indication lights, the VSP!LS is frequently used for computer point indications, pemissives or interlocks with other plant equipment.
Ocen Direction Loaic Nomally, correct indication of open position l
is less critical from a functional viewpoint than correct closed position indication. This is because the ficw rate and pressure drop curve for a typical gate, globe or quarter turn valve is relatively flat during the last portion of the opening stroke. However, if ASME Section XI stroke time testing uses the remote indication lights as a basis for valve stroke time, then the correctness and consistency of i:
the open position-indt' cation becomes important.
L 15 l
p a.
Close Direction Loaic - Correct indication of the close valve stem positto is critical for MOVs with throttle circuitry.
This is especially true where flow indication is not available to detect a partially open valve. Substantial leakage can pass through any valve that is partially open.
Considerations For VSPILS Settinas - Correct position indications are critical for HOVs with throttle logic. Correct position indication is less critical for H0Vs with seal in contactor logic (i.e., once the button is pushed, the motor contactor remains energized until interrupted by the control circuit).
Limitorque 2 Rotor Systems Limitorque actuators equipped with 2 rotor limit switch mechanisms have difficulty providing correct VSPILS indication where turque switch bypass protection is used.
In these applications, consideration should be given to installing a 4 rotor limit switch mechanism.
3.2.3 Valve Position Limit Switch (VPLS) f.gnetion - The function of the VPLS is to open the motor contactor in the open or close valve positions, for rising stem valves, an open VPLS typically controls actuator motor shutoff in the open direction to prevent backseating. An open and close VPLS typically is used for quarter turn valves and in certain rising stem valve applications.
NOTE:
Except for three way valves and other unique situations where torque backseating is desired, most valves are de energized in the open direction using the VPLS.
Open Direction Loaic - Valves using the open VPLS usually de+ energize the actuator motor at'a position where the valve is capable of passing full flow but before it would coast into the backseat or
- stops, i
l 16 l
3 l.
Close Direction Looit Typically the VPLS is used in quarter turn, parallel disk gate, and diaphragm valve appiteations. A close VPLS logic has been used successfully in rising stem gate and globe valves where it is desirable to control inertial overshoot, in addition, close VPLS logic, with the torque switch fully bypassed, allows full usage of the actuator margin (up to motor stall capability), if needed, to close the valve.
Considerations For Vpl$ Settinos It may be advantageous on high i
speed valves to set the open VPLS initially a conservative distance from the backseat. The backseat is then approached using trial operations. Some additional margin between the test condition stopping position and the backseated position is desirable for applications where line pressure or flow could create additional opening forces and coast.
Due to 'the potential for valve or actuator damage in rising stem valves using close VPLS logic, detailed procedures and technician training are recommended. HOV diagnostic systems are also valuable in verifying proper switch setup. Additionally, without diagnostics, if the torque switch is not in the closing circuit, an increase in the required thrust due to degradations will not be apparent.
Actuators using close VPLS logic equipped with stem nut compensating spring packs (Limitorque SB/SBD or Rotork NAX models) can be set to actuate at a certain compensating spring pack deflection, e.g., the Westinghouse method.
3.2.4 Torque Switch function - The function of the torque switch is to limit the actuator output torque as sensed by worn movement against a spring pack which is preloaded to a set.value. Only actuator loads sufficient to overcome the spring pack preload can be sensed by the torque switch.
17
/
Onen and Close Direction Looig The torque switch is active unless bypassed by the torque switch bypass limit switch (ISBLS) or disabled entirely using some other method. Rising stem applications typically control valve closing using the torque switch, i.e., torque seated to a specified value. The open torque switch is only for protection in the event of torque overload. However, there are some types of valves that require torque seating into the backseat, such as three-way valves. Most quarter turn valve applications have the torque switch in the circuit for torque overload protection only.
Onsiderations For Toroue Switch Settinas for Risina Stem Valves.
Opening and closing requirements for this application are typically expressed in applied valve stem thrust. The specific valve thrust requirement can be obtained from the valve manufacturer, in house calculations and from test data. Since the resulting stem thrust produced by a given applied stem drive nut torque is dependent on the conversion efficiency (stem factor), allowances should be provided for potential stem factor variations.
Provided the torque switch is in the circuit, some means must be used to either calculate or measure the actuator stem thrust output at a prescribed torque switch setting. Methods used in the industry to i
accompitsh this are discussed below, Reliance on the torque output versus switch setting chart o
provided by the actuator manufacturer. These charts are specific for a given worm shaft, spring pack and torque switch.
Manufacturing tolerances and spring pack relaxation can drastically affect the actual torque values obtained.
Appropriate stem factors and allowances for stem packing drag variations should be applied.
18
o ' Reliance on torque bench test data to determine actuator torque output at a given torque switch setting. Appropriate stem factors and allowances for stem packing drag variations should be applied.
o in situ torque measuring systems can measure the actuator or gearbox torque output at the MOV location.
Included in this category are torque wrenches through complete diagnostic systems.
Some degree of inertial overshoot should be considered using this method. Also, appropriate stem factors need to be applied and stem packing load variations should be accounted for if they cannot be directly measured.
o Sub component testing involves estabitshing unique force displacement relationships between actuator spring pack displacement and torque switch setting. These two relationships are integrated to predict actuator torque switch setting versus torque output. As mentioned above, appropriate stem factors and allowances for stem packing drag variations should be applied.
o Diagnostic testing can directly or indirectly measure stem thrust. Appropriate stem factors and allowances for stem packing drag variations should be applied. This is the only technique that can effectively detennine inertial overshoot.
NOTE: Regardless of the technique used to establish the torque switch setting, allowances should be made for differential pressure, flow, dynamic correction factors -and instrument and set up uncertainties.
Caution should also be exercised to ensure no actuator or valve limits are exceeded.
Considerations for Toroue Syltch Settings For Oyarter Turn Valvel -
Opening and closing requirements for this application are typically expressed in applied valve stem torque.
The specific valve torque 19 i
requirement can be obtained from the valve manufacturer and from test for butterfly valves, flow velocity, flow disturbances, and
- data, elastomeric seat material hardness are key considerations.
j Methods used in the industry to either calculate or measure the actuator torque capability at a prescribed torque switch setting are similar to those explained for rising stem valves when calculating or measuring thrust for rising stem valves.
i 4
k l ',
I-l' 20
(
~. - _ _ _. -, _ _ _
t 4.0 VERiflCATION Of $ WITCH SETTINGS NOTE:
Although diagnostic tools are not specifically recomended by the generic letter, this document assumes the use of diagnostics to obtain trendable parameters. The data obtained according to this section can be used for post maintenance testing and trending during periodic tests.
4.1 Baseline Static Testing (No Differential Pressure or flow)
All MOVs should be tested and set to ensure all switch setting parameters are initially established and documented as detemined in the preceding sections. Baseline static testing should be performed to ensure that the MOV is capable of operating within the limits of the calculated parameters. Testing under differential pressure conditions may not be practical or desirable for mary valves.
4.2 Oynamic Testing at or Hear Design Basis full Differential Pressure / flow Dynamic testing performed as close to full design basis r'equirements as possible is desirable. When unable to test at full design basis delta P or flow, current analytical methods and best available data should be used to determine the switch settings.
NOTE: Additional analysis may be required if an HOV falls to stroke fully under design basis conditions.
For example, evaluation of the resultant leak rate may be required to determine that leakage is acceptable to meet design basis requirements should the MOV fail to close completely.
4.3 Switch Setting Parameter Hargin
]
One should integrate the minimum setting required based on the calculations to overcome delta P and flow and the maximum setting 21 l
f
_. _ _ = --_ -. _-
allowed based on MOV assembly limits, i.e., weakest link.
The assembly limits are based on operating and seismic load requirements.
When full delta P and flow cannot be obtained, each assembly's setting should be reviewed on a case by case basis and adjusted to obtain as much margin as reasonable to achieve maximum reliability.
As additional data is obtained from industry testing, currently in progress, settings should be reevaluated.
)
i 4.4 Alternatives to Design Basis Testing 1
It is not necessary to test every testable valve at design basis conditions. When testing at design basis conditions is not performed, adequate reliability margin should be the basis for the justification. Recent NRC research indicates that valves behave more predictably at lower pressures.
Adequate margin may be determined based on good engineering analytical techniques using results from similar valve test data, either in house or industry. Determining similar valve characteristics is the subject of the EPRI ' Performance Prediction Program.'
If practical, partial differential pressure testing should be perfomed to demonstrate sufficient margin.
If an adequate confidence margin is not available, switch setting logic may be modified to provide full actuator output in the design direction (e.g., valve position limit switch logic used for closure instead of torque seating).
Refer to Section 3.2.3.
Equations for assessing the effects of differences between design-basis conditions and those tested may be obtained from the valve vendors and the NKAC/EPRI 'Appitcation Guide for Hotor Operated Yalves in Nuclear Power Plants.'
22
\\
The EPRI ' Performance Prediction Program" overall objective is to develop the technical justification for eninimizing high diff&rential pressure testing.
1
,W e
4 l
23 l
.a m
e s
l i
l i
1 e
f F
4 J
t e
i h
I t
t 4
3 I
s f
+
i i
)
}
--N-w-
i 5.0 POST MAINTENANCE AND PERIODIC TESTING Action item j. of the generic letter recommends that each MOV be tested after maintenance or adjustment and periodically thereafter to identify any degradations or misadjustments. Satisfying the periodic f
testing requirement can be achieved in conjunction with post j
maintenance testing if the station procedures allow the flexibility.
5.!
Post Maintenance Testing The following is a list of maintenance activities and the appropriate retests. This list of activities and tests is a guide and not meant to be all inclusive. The selection of testing will depend upon the r
scope of the maintenance perforced and the requirements of any applicable codes, standards, and Technical Specifications.
for this guidance document, post maintenance testing is assumed to be done using diagnostic monitoring techniques. These tests are divided into two categories:
Baseline Test The initial set up verification includes o
measuring a set of parameters to ensure their values provide adequate assurance that the MOV will perform its intended function.
The parameters should include thrust or torque, switch position, motor current, power, and motor slip.
I Verification Test + This test monitors some of the specific o
parameters measured to' establish the baseline. This testing requires comparison of these parameters against the baseline
' test results to ensure they have remained within the allowable' operational margin. An example of this type of test would be a motor current signature taken after packing adjustment, i
25 l
l-
4 5.1.1 Retest following Preventive Maintenance Normally, preventive maintenance involves routine cleaning. stem lubrication, and inspection of gearbox lubricant. This does not include the replacement of any of the motor operator parts, other than the limit switch compartment cover gasket. The retest for the routine preventive maintenance should include as a minimum, stroking the valve twice from the control room to detemine operability.
5.1.2 Valve Mechanical Work An appropriate test for nearly all mechanical work on the valve would be a general leakage inspection and any testing required by ASME codes. All retests on the motor operators should include, as a minimum, stroking the MOV twice-to verify proper function.
5.1.3 Valve Packing Adjustment or Replacement The packing drag on the valve stem has a direct impact on the thrust / torque required to open or close the valve.
The recomended retest is to monitor the valve running load to ensure it has not increased to the point of reducing the margin of thrust / torque available'under design basis conditions to an unacceptably low level.
5.1.4 Replacement of the Limit Switch Replacement of the limit switch assembly will require verification to-ensure that the limit switches are properly set. This verification should meet or exceed the method used to perform the initial-baseline limit switch verification.
26
,,w,,,
,,,,,-~ --,
mv,e
,n,,,
,-n,.
v.,....,,--
rv -rw a
,,n,-
,,-,-.v--,.<,.
-- - - -,,,w,,re---,,w n---~,,-.mr,,v,--w,--~-----,n e.mr~
s
5.1.5 Torque Switch Replacement Replacement of the torque switch on torque seated valves will require new baseline testing to ensure the actuator output is still within the cstablished '4 ign limits.
t 5.1.6 Motor Replacement Replacement of the motor should include an insulation resistance test and rotation verification. Motor current, motor slip, or power readings should be recorded and oc,mpared to baseline data.
5.1.7 Removal / Reinstallation of the Motor Operator if the motor operator is removed for any reason, verification should be coupleted to ensure that the limit switches are properly set.
This verification should meet or exceed the method used to perform the initial baseline limit switch verification, j
5.1.5 Motor Operator Overhaul if the motor operator requires disassembly for overhaul, then an entire new baseline should be estabitshed prior to returning it to service.
5.2 Periodic Testing The NRC stated in Action item d. of the generic letter that the ASME Section XI stroke timing test alone is not acceptabisi for periodic switch setting verification. An appropriate periodic test and the surveillance interval will depend upon an MOV's importance to safety and its maintenance and, performance history.
As indicated in the post maintenance testing section, baseline N0i' testing should include monitoring of parameters readily available 27
-j
4 from the motor centrol center.
Examples of these are motor current, motor power, motor slip, etc. By having this baseline inform 6 tion available, periodic testing requirements can be satisfied by repeating this measurement and comparing it with the baseline values.
These test results can be trended to provide predictive maintenance capabilities for the plant's HOVs.
Periodic testing should not be perceived as a stand alone program, but rather an outgrowth of existing programs of post and preventive maintenance and surveillance testing. For example, if preventive maintenance is performed on an MV three years after initial baseline testing, obtaining parameter values from the motor control center (HCC) to compare with baseline values would improve the ability of the preventive maintenance to detect MOV degradations. This would satisfy the periodic testing requirements of the generic letter and reset the periodic testing clock, if such values could be obtained through surveillance testing, then this may prevent having special procedures to accomplish this testing.
As with any good predictive maintenance program, allowable parameter value changes should be determined and ' Alert
- and "Act' ion' levels identified. This has the added benefit of identifying degrading conditions early, thereby allowing the HOV to be repaired before failure.
5.3 Analysis of HOV failures Action item h. of the generic letter states that each MOV failure should be thoroughly investigated and actions documented. The intent is to ensure that the licensee captures this data and reviews it for developing trends in MOV reliability.
It should be noted that this effort applies pniv to MOVs that fail (i.e., fati to perform their intended function). Hinor degradations that have no impact on MOV l
l 28 l
l
l
- 4. "_
coerability should not be considered failures. Documentation of-
-failures should include:
o As found deteriorated condition; o Repair / alteration performed; o Analysis of failure (root cause); and o Retest performed.
Evaluation of this information is reccmmended to take place every two years or after each refueling outage after program implementation, whichever is-longer. The evaluation should examine:
o Periodic testing frequency and o Common mode failures.
Although not required by the generic letter, it would be helpful to analyze and trend all MOV defects. Where data shows a problem _or potential problem, the situation should be investigated and resolved.
4 29
. _.. g e,
..7
. _......... ~.,,,. _.., _, _ _... _...
._-.~.._____m._
G 4
e f
k e
T i
e t
30
4 u,
6.0 DOCUMENTATION The design basis review documentation and results of all maintenance and' testing should be retained and be retrievable for industry use and NRC review under existing plant requirements.
6.1 Recomended Procedures and Guidelines o MOV Program Guidelines o MOV Preventative & Corrective Maintenance Procedures o MOV Testing Procedures o Design Basis Review o MOV Setpoint Determinatio:' Lid.
6.2 Recontended Documentation a.m 'r* ific, *.ign Controls Engineering review and appeoval required to change setpoint o
criteria (initial setpid c. teria to include engineering-calculations)
An MOV setpoint control document and/or individual setpoint n
criteria documentation for each MOV
- ~
o MOV f ailure analysis doctimentation o Alternative justification to full design basis parameter L
. testing o Schedule outline
- l 31 L
( ~'
l.
i i
4 6.3 Required Documentation o: Acknowledgerents and exceptions - 12/28/89 Program description available at plant 1/01/91 o
Baseline program completion - 30 days after completion o
future schedule changes and exceptions - as needed o
e 4
4 32 I
.. ~.
7,0 SCHEDULE Ceneric letter 89 10 provides for three refueling outages or five i
years, whichever is longer, to complete engineering and baseline testing.
Periodic testing for all H0Vs is to be completed every three refueling outages or five years, whichever is longer. However, trending and operating experience can justify extending this interval.
The NRC considers the schedule to be achievable based on the following assumptions listed below, A parallel approach is used, e.g., as the design basis review o
and engineering is completed on groups of MOVs the testing can comence while the engineering continues on HOVs still to be re.iewed, The average number of valves to be addressed per plant equals o
- 150, o Working one crew pe).shi f t, tw^ shii6s per day, six days / week, averages five motor-op rated valves comp'eted per week.
o Each refueling ou'sgc rovins
,3" reeks in which HOVs can be workeJ.
7.1 Factors Which can Impact Completion Schedule In addition to the obvious impact on utility resources, there are many other factors which will influence the attainment of the schedule, some of which the utility can control and others which it cannot. These factors,should be considered when establishing l
individual schedules. A partial list of factors that have impacted some utilities' ability to complete programs similar to that required by the generic letter are found below:
33 l
l l
7.I.I Vendor Support Receipt of valve and actuator vendor design information is a potential impact due to the relatively small number of vendors and the potentially large numbers of requests for information they will receive. Additionally, some vendors no longer supply the nuclear industry. On site support of maintenance and testing can also be affected by many requests to relatively few vendors.
7.1.2 Planning and Scheduling Support Due to the difficulty in obtaining work windows, even during a refueling outage, it is useful to provide dedicated planning and scheduling support for HOVs. This will maximize the amount of work that can be accomplished.
Items such as work area tag outs, PNs on motor control centers supplying HOVs, Local Leak Rate Testing and other work that renders NOVs unavailable can all impact an H0V testing schedule.
7.I.3 Spare Parts Experience has shown that implementation of a comprehensive MOV program will require more spare parts than previous history would indicate. This is due to the increased number of HOVs worked in a given pt'iod and the ability of diagnostics to identify problems f
previously not detected until failure. Unfortunately, some spare parts have fairly long lead times for delivery and lack of an "up-front" spare parts effort can adversely impact schedules.
Additionally, consideration should be given for stocking the highest level of spare part or assembly. This can reduce labor cost, exposure, and contamination.
t l
l l
34
- 7.- I 4 Engineering Support While the need for skilled technicians.to perform work and testing on MOVs-is well recognized, the engineering effort required to support them is sometimes underestimated. The required design' reviews, calculations to obtain the limiting differential pressures and flows, and detenntning the proper switch settings are time consuming and resource dependent. Additionally, resolution of problems during testing, root-cause analyses, and final acceptance of. test results also consume engineering resources.
7.I.5 Research Results Research is needed to resolve the calculational uncertainties involved in some valve applications to determine the required thrust.
The research results may dictate a reanalysis of design review and replacement of certain valves. Adequate research may take more than five years to complete.
One way_in which the utilities can minimize the impact on outage schedules is to perform MOV testing during power operations on selected valves.
Periodic testing can also have its impact lessened by _ performance during on line periods.
CAUTION:
This approach needs to be fully considered addressing the impact on plant operation.
The scheduling impact of periodic testing required by the generic letter on a three refueling outages /five-year ~ interval can be minimized by incorporating it with the following existing tests:
o Post-Maintenance Testing Existing Surveill'ance Testing o
o Existing Preventive Maintenance Testing 35
,5h+2e h ed,dvf + @-W Mo4MW J MMA -
---5 615.hhMp2M
,6.--
- - - -- - ~ - " - - -
hh-f' t
4, e
l.
i
! 'i F
J-.
f
- }
t I
i a
1, y
i 4
f y'.
/ -
f i
i-I 9
i t.
I s
9 i
36 c
.... -. - - - ~....
.- - ~
REFERENCES Generic Letter 8910, " Safety Related Motor-0perated Valve Testing And Surveillance," dated June 28, 1989.
IE Bulletin 85 03, " Motor-Operated Valve Common Mode Failures During Plant Transients Due to Improper Switch Settings," dated November 15, 1985.
IE Notice 90-40, "Results of NRC-Sponsored Testing of Motor Operated Valves,"
dated June 5, 1990.
IE Notice 86-29, " Effects of Changing Motor 0perator Switch Settings," dated April 25, 1986.
INPO, ' Motor-0perated Valve Performance Update," dated October 4,-1988.
INPO Significant Operating Experience Report 83-09, " Valve Inoperability Caused By v4otor-Operated failures " dated October 21, 1983.
' Limitorque Maintenance Letter, August 1988.
RMAC/EPRI,." Application Guide for Motor Operated Valves in Nuclear Power
-Plants," dated March 1990.
MMAC/EPRI, " Technical Repair Guidelines for the Limitorque Model SMB 000 Valve l-Actuator," dated January 1989.
l l
NUREG-1296, "Themal Overload Prctection for Electric Motors on Safety-Related Motor-Operated Valves - Generic issue-11 E. 6.1.," dated June 1988.
Reg. Guide 1.106, " Thermal Overload Protection for Electric Motors on Motor-L Operated Vaives," Rev. l., dated March 1977.
37
- - -..