ML20244E577

From kanterella
Jump to navigation Jump to search

Discusses Inservice Test Programs Re Full Stroke Exercising of Normally Closed Swing or Tilting Disk Check Valves Per 10CFR50.55a & Section XI of ASME Code
ML20244E577
Person / Time
Site: Davis Besse, 05000000
Issue date: 06/24/1982
From: Norelius C
NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To: Eisenhut D
Office of Nuclear Reactor Regulation
Shared Package
ML16342B348 List: ... further results
References
FOIA-87-121 TAC-48561, NUDOCS 8701150292
Download: ML20244E577 (2)
v • d • e


Text

.

UNIT ED STATES yo r%q#

4 0

g NUCLEAR REGULATORY COMMISSION S REGloN ill f j no moostvcLv noAo n

G LEN ELLYN, ILUNOIS 60m

.f JUN2h.882 l

i MEMORANDUM FOR:

D. Eisenhut, Director Division of Licensing l

FROM:

C. E. Norelius, Director, Division of Engineering and Technical Programs

SUBJECT:

INSERVICE TEST PROGRAMS - FULL STROKE EXERCISING OF NORMALLY CLOSED SWING OR TILTING DISK CHECK VALVES Licensees conducting inservice testing of valves in accordance with  !

10 CFR 50.558, and applying the ASME Code,Section XI, edition and' addenda combinations allowed by 10 CFR 50.55a(6)(2)(1) are required to full stroke exercise normally closed check valves whose safety function is to permit flow upon reversal of differential pressure j (e.g. check in ECCS injection line). The ASME Code requires that, f for swing or tilting disk check valves, if the test is made by use of fluid flow through the valve (as opposed to using a mechanical exercisor), the pressure dif ferential for equivalent flow shall be no greater than that observed during the preoperational test.

During a recent inspection of Toledo Edison Company's Davis Besse Nuclear Power Plant, Mr. K. Connaughton of this of fice determined that flow through the check valves was the only parameter used.to .

satisfy the full stroke exercising requirements referenced above.

This practice is not uncommon among licensees. Acceptance" criterion vary. Some licensees may specify " design flow" while'others use l

" minimum flow for check valve lift" or merely " observe flow." Under design conditions, observation of design flow may be adequate to demonstrate that a valve is capable of fulfilling its safety function.

Corrections to measured flow for off-design conditions (e.g. check in ECCS injection line being tested in the Refueling Mode) must be made in order to provide a similar demonstration. The observation of

" minimum flow for check valve lift" (presumably established for a particular model of valve when new) or some arbitrary flow does not, in our opinion, provide any information as to the condition of the valve unless, of course, the valve has failed in some manner as to deny practically any flow.

N 0011Toaci A i I

i M

_ _ n . -. - . . . . . _ , . . _ _ _ _ . . , \

UN 2 4 667*

D. Eicanhut We are not aware of any licensees having submitted written requests for relief from this particular requirement. While we realize that preopera-tional test data on differential pressure across individual valves at reference flows is all but nonexistent, we feel that the use of differ-ential pressure as the test parameter may have greater technical merit than current practice. Please inform us of the staff's positions on:

1) the applicability and enforceability of the requirement considering:

the lack of preoperational test data and; the absence of requests for relief from this requirement.

2) whether or not licensees should be required (as an alternative to the test method and acceptance criterion described in the ASME Code) to:

demonstrate that under design conditions, the valves will permit design flows and are therefore capable of performing their intended function or; establish, by test and measurement, appropriate baseline differential pressures at reference flows which can be used in place of nonexistent preoperational test data.

On May 17, 1982, W. D. Shafer and K. Connaughton of this office discussed these matters, via telephone, with J. Page of your staff (MEB). Your efforts in resolving these concerns are greatly appreciated.

./

/

C..E. Norelius, Director Division of Engineering and Technical Programs l

, , r: ~ % c. . - ., ..,-;~-' -- '.O v...,?? A3" 9 NAs-.v-..~.

?v*- -v'~~ W-9-:9^ --*F "

,_ .. .. + . . -a... ~~.;- a ..me. +

[ $

DISTRIBUTION: ' '

JAli1 ? Igm SEiiiiingglBrRO Dziemann, DaFS -

LBf2 Reading BSheron, DSI EHylton DWessman, DL ABournia -

ASchwencer  : i TMNovak DGElsenhut Tippolito. ORAB HEMORAWOUM FOR: Richard Spessard, Director 01 vision of Engineerinn & "

Technical Programs Region III NRC -

FRON: Darrell G. Eisenhut, Director  :

Division of Licensing, NRR  :

)

l

SUBJECT:

STOPWATCHES CAL 1 BRAT 10H AND ASME REQUIREMENTS OH 'l TRANSIENTS ANALYSES i-In your letter of Septenber 7,1982, you requested NRR assistance to evaluate the I use of uncalibrated timing devices in safety-related applications and to clarify [

aQ apparent inconsistency between transient analyses and ASME reoutrements regarding f:

valve closure times which are defined in the Technical Specifications. Our Reactor i:

Systems Branch evaluated the main steam ifolation valves (MS!Y) closure, and its i '

impact on minimum critical power ratio (HCPR) as determined in the FSAR analyses.

Our Procedure and Test Review Branch provided the requirements for the stopwatches.

This review was performed pursuant to TIA 82-58.

?

The ASME code allows rounding of measured closure times for MSIV's to the nearest 1.0 second or the use of a timing device with an occuracy of 1.0 second. This makes it possible for a ceasured closing time of 2.0 seconds to appear as 3.0 seconds which would satisfy the allowable technical specification value and the FSAR analyses. l

~

We evaluated the impact of a 2.0 second closure time versus a 4.0 second closure time on vessel pressure and EPR for the MS!V closure event. In the analysis, an 9 MSIV positten switch scram was assumed for the effect en MCPR and a high flux scram was assumed for the assessment of the effect on overpressure protection.

The results of these analyses indicate that for closure times of 2.0 seconds or j greater, the impact on MCPR and vessel pressure is insignificant and will not li challenge safety limits. For the limiting cases, vessel done pressure was calcu- 1 lated to increase 170 pst and the MCPR was calculated to be four percent of the il initial NCPR valve. For a typical boiling water reactor operating at 1100 psig y and an initial MCPR of 1.24, this results in a eak vessel pressure (bottom nf I

Dressure) of less than 1300 psig and a HCPR OF .19. The respective safety Ilmits aret pressure must he less than 1375 plig and HCPR must he ortater than )\)" k'I N 1.06 On this basis, we conclude that the interpretation of the ASME code which allow as ev:b as 1.0 second errnr in elv closure time is of no safety cr,ncern (h for plants with a technical specification mintmm allowable "$1Y closure time of  ;"f *}l [ d' 3.0 inconds or orester. q ,# [

'320 399 03011 DMO.3.T. .. h b 7 CF $}QV. .

o" el j ,

i i o --

~ ~ '^' ~ ~ " - --E

a ,

Richard Spessard Vith respect to the use of stopwatches for measur. ment of tirie intervals for validating operability of safety-related equipment, this method is acceptable where it can be shown that this method of measurement provides the reouired accuracy. Therefore, the results of the above analyses indicates that a 1.0 second total deviation on MSIV closure time is acceptable. The stopwatches mst be calibrated and controlled as required by 10 CFR Part 50, Appendix B, Article X11, and the the required accuracy is detemined from the technical specification bases. ANSI /ANS-58.4-1979, "American National Standard Criteria for Technical Specifications for Nuclear Power Plants," paragraoh 5.1.(6) states:

" Errors, frw instrumentation or other sources, assumed in the development of the technical specification limits shall be dis-cussed in the bases to provide a clear relationship between the technical specification and the safe +y analysis values."

Our technical specification basis does not provide a discussion of measurement error, and the limit does not include an allowance for measurement error. it is necessary to include a measurement error allowance in the surveillance test acceptance criteria.

For the operator error component of the error associated with the use of stop-watches, we consider 700 milliseconds to be an acceptable assumption.

We trust that the information provided is responsive to your concern, and 19 WRR responsibilities under TI A Ho, R?-58 have been comoleted.

Ori ginal signed t'y :

Darrell G. Eisenhut, Director Division of t.icensino Of fice of Huc1rar Reactor E* qui stion cc: 9. Starottu.ki. 0-1 J. 01shinski, 0-!!

J. c.agliardo, P-1V J . Crews , 0-V e

_f e-

, 5

Retrieved from "https://kanterella.com/w/index.php?title=ML20244E577&oldid=3273097"