ML20235G591

From kanterella
Jump to navigation Jump to search
Provides Update on Status of Program to Develop Generic Fix for two-stage Safety Relief Valve (SRV) Setpoint Drift Concern Outlined by BWR Owners Group SRV Committee at 851017 Meeting
ML20235G591
Person / Time
Site: FitzPatrick Constellation icon.png
Issue date: 04/02/1987
From: Pickins T
BWR OWNERS GROUP
To: Cherny F
Office of Nuclear Reactor Regulation
Shared Package
ML20235G546 List:
References
REF-GTECI-B-55, REF-GTECI-PV, TASK-B-55, TASK-OR BWROG-8714, NUDOCS 8709300116
Download: ML20235G591 (10)


Text

- _ -__ _____ -_-_ _________ _ __ _ _ _ _ _ _ _

's -,

l Wt3 .* _).h 3

b OUJNERS' GROUP c/o NORTHERN STRTES POLDER CO. e 414 Nicollet Moll e Minneapolis. MN 55401

!Mie;,.<d 27 (6121337-2037 w- se J BWROG-8714 Em 3 1 April 2,1987 C '

'~

, .G o -

._ : 1 .i Mr. Frank C. Cherny E d O Section B Engineering Issues Branch Division of Safety Review and Oversight U.S. Nuclear Regulatory Commission Washington, DC 20555

Subject:

BWR Owners' Group Program to Develop a Generic Fix for the Two-Stage SRV Setpoint Drift Concern (Generic Issue B-55)

Reference:

NRC/BWROG October 17, 1985 meeting, same subject

Dear Mr. Cherny:

This letter provides an update on the status of the subject progran that was outlined by the BWROG SRV committee at the reference meeting.

BACKGROUND The BWROG prog' ram proposed to replace the Stellite pilot valve discs with a new disc made of PH13-8Mo stainless steel. The Stellite material was to be retained for the seat. It was expected that this new material combination, PH13-8Mo disc and Ste111;e seat, would alleviate sticking of discs in their seats. It had been determined in earlier investiga-tions that the predominant cause of upward setpoint drift was due to sticking of Stellite discs in their seat as a result of corrosion buildup between the disc and seat.

Since mid-1986, several BWR units (see attachment) have placed the new PH13-8Mo discs in service in a number of their valves. Before perma-nently changing over to PH13-8Mo discs in all the valves, the in-service performance of the new material was to be evaluated over the next couple of years until it was proven acceptable. Recently, the BWROG undertook an accelerated corrosion testing of PH13-8Mo and Stellite disc samples in a laboratory to gain early confidence in the proposed fix.

CURRENT STATUS The BWR Owners' Group SRV Comittee met on February 4 and 5, .1987 to review the in-plant performance of PH13-8Ho discs and the accelerated corrosion tests.

8709300116 70925 PDR ADOCK 05000333 0 0 PDR

I I

.e e BWROG-8714 The in-plant performance of the PH13-BMo material discs to date (see attachment) has been satisfactory. Five events have challenged automat- ,

ic opening of SRV's (three at Hatch and two at Brunswick). There was no l apparent evidence of any PH13-8Mo discs sticking in their seats and {

failing to open.

Preliminary results from the limited laboratory corrosion tests indicat-ed presence of some sticking between PH13-8Mo and Stellite seats. A review of the test setup indicated that Stellite / Stellite samples were galvanically protected from experiencing higher corrosion rates. There-fore, no real comparison could be made between the performances of PH13-8Mo and Stellite disc material. Further revised testing is now in progress to enable relative comparison of performance of the two l material s. I NEAR-TERM ACTIONS PLANNED It is expected that in May 1987, Plant Hatch 1 valves will be availtble for examinatica at Wyle Labs after an in-service use of approximately l one year (since May 1986). Also, the modified corrosion testing results should be available at the same time. A detailed assessment of PH13-8Mo performance will be done at that time.

The BWR Owners Group SRV Committee considers resolution of SRV setpoint drift concern as a high priority issue. Evaluation of PH13-SMo as a replacement disc material is being aggressively pursued. I will keep you informed of any significant developments. In the meantime if you have any questions regarding this, feel free to call me.

This evaluation has been endorsed by a substantial number of the members of the BWR Owners' Group; however, it should not be interpreted as a comitment of any individual member to a specific course of action.

Each must formally endorse the BWROG position in order for that position to become the member's position.

Very truly yours,

/ @

T. A. Pickens, Chairman BWR Owners' Group jmh cc: R. F. Janecek, BWROG Vice-Chairman C. Wargo, SRY Setpoint Drift Comittee Cheirman ..

L. A. England, Standing Functional Comittee Chairman D. R. Helwig, RRG Chairman SRV Setpoint Drift Comittee Members Primary Representatives of Utilities using Two-Stage Target Rock SRV's J. W. Power, EPRI W. S. Green, INPO C. L. Tully, AIF

1 i

o J

\

i

1 ATTACHMENT l

l l

IN-SERVICE EXPERIENCE OF PH13-8MO (TO JANUARY 1987)

NO. OF NO. OF SRV IN-SERVICE CHALLENGES TOTAL SINCE TO DATE REMARKS

' PLANT OF VALVES -W[PH13-8M0 May 1986 2) a Satisfactory Id) 11 5 HATCH 1 operation l I

b Satisfactory (d) 11 5 Dec 1986 l)

HATCH 2 operation June 1986 2) c Satisfactory (d) 11 6-

-BRUNSWICK 2-' operation 8 Aug 1986 None FEPHI 2 15 1 Aug 1986 None HOPE CREEK 14

-Notes:

a) January 1987, all eleven valves lifted. (Peak pressure within 3% of highest setpoint.)

b) January 1987,10 of 11 lifted (one Stellite valve did not lift--peak pressure within 1% of highest setpoint).

c) August 1986: 7 of 11 valves lifted including 4 with PH13-Mo disc. (Peak pressure within 1% of highest setpoint). January 1985, 5 of 11 lifted

' including 3 with PH13-8 Mo disc (peak pressure within 2% of highest setpoint),

d) All valves expected to lift (i.e. with setpoints below the recorded

_ peak reactor pressure) had indeed lifted to control vessel pressure to within the ASME allowable limit (1375 psig).

I S

  • q

/

p .

.j L-Attachment 2 i

l FITZPNERICK RECENT TEST HISTORY

.1

[

(Reference Wyle Test Reports 46602, 46861, 47627, Land 48669)

%'ofLNominal'Setpoint, First' 1 Topworks Test Run (Satisfactory meane- 'I

. Year Serial Number within +/-'one percent ofsnominal) {

1987. 1088 Satisfactory 1053 ,98%

1080 .104%

1056' 112%

1012 Satisfactory 1047 Satisfactory;

?

1985 -1045 106%. I 1050 101.5%

1051 106%

1 1032 106%

1053 Satisfactory 1056 102.7% , .,

1080 Satisfactory 'd 1097 Satisfactory 1983 1012 97.4%

1013 103%

1062 '101%

1080 103.6%

1088 ' Test Stopped 1056 102.3%

1087 Satisfactory I

t l

1 i

I i

Attachment 3 GENER AL O etecTaic 1

Detectability. History. and Impact of SRV Drift March 18, 1987 l DRF A00-02948

Background:

In March of 1987, New York Power Authority (NYPA) had six of the eleven two-stage Target Rock Safety Relief Vr.1ves (SRVs) from the James A.

FitzPatrick Nuclear Plant (FitzPatrick) tested at the Wyle Laboratory. j The "as received" test data (Table 1) indicates that three of the six 1 valves tested had experienced a setpoint drift outside the acceptable limits of i 1% as specified in the FitzPatrick Technical Specifications. s The primary function of the SRVs is to prevent vessel overpressurization 4 during transient events in which the reactor vessel becomes isolated from the main condenser. The overpressure criteria for these events is .

specified in Section III of the ASME Pressure Vessel Code, and )

corresponds to a peak vessel pressure of less than 110% of the vessel l design pressure or 1375 psig. The limiting pressurization transient  !

event for FitzPatrick is the simultaneous closure of all MSIVs with a coincident failure of the MSIV position scram signal. A high neutron  ;

flux signal is assumed to initiate the scram. The high flux signal

,I would actually occur after the MSIV position scram signal, consequently a conservative peak pressure is obtained by evaluating the MSIV closure with flux scram event.

l The purpose of this paper is:

1) to provide some historical data regarding SRV setpoint drift and its {

potential impact on vessel overpressure protection based on evaluations I which have been performed for other BWRs, and )

2) to document information regarding the potential for accurately detecting SRV setpoint drift from plant data ~ recorded during transient events.

History and Imoact of SRV Setooint Drift  ;

In 1982, under the sponsorship of the SRV Setpoint Drift Committee of the BWR Owners Group, General Electric Company embarked upon a program to investigate the cause of SRV setpoint drift (Reference 1). As a part i of this program, the impact of SRV setpoint drift was assessed for the j limiting pressurization transient. The conclusion reached from this l assessment was:

}

"All BWR plants equipped with the TRC (Target Rock) two-stage SRVs j have sufficient margin for overpressure protection and can tolerate i gross setpoint drift up to 10%."

l

. .c 1 Attachment 3 l

SRV setpoint drift concerns have been previously evaluated for several BWRs. Specifically, evaluations performed in 1982 and 1986 for two I BWR 4 plants which are very similar to FitzPatrick are discur, sed below. )

~

1982 Evaluation l l

Sensitivity studies were performed to determine the .'.mpact on reactor l overpressure protection if all SRVs drifted from 0 to over 200 psi (approximately 20%) above nominal se point pressurc. Based on these studies the peak vessel pressure would remain below the upset limit of 1375 psig for the limiting postulated hypothetical event even if all SRVs experience a'setpoint drift of about 180 psi or about 16% drift above setpoint (nominally around 1100 psi). In fact, an extrapolation of these studies (See Figure 1) suggests that the peak vessel pressure would remain below the emergency pressure limit of 1500 psi even if all SRVs drif ted by about 30% above setpoint. Hence, these evaluations demonstrated that there was no safety concern even with significant SRV setpoint drift (up to about 16%).

1986 Evaluation "As-received" tests were performed in 1986 on all eleven SRVs from a BWR 4 plant. These tests indicated that significant setpoint drift had occurred, in excess of 20% for some valves. The average drift was )

about 13% above nominal setpoint pressure. Analyses demonstrated that if all the valves had actually drifted by the average amount of 13%,

that the plant would still meet the ASME upset limit of 1375 psig for the most limiting hypothetical event. When the actual "as-received" setpoints were utilized to evaluate the most limiting event with realistic valve performance characteristics the peak vessel pressure was well below the ASME limit of 1375 psig.

Detectability of SRV Setooint Drift:

Two-Stage Target Rock SRVs are designed to open when the pressure in the steam lines exceeds the spring force imposed upon the pilot disk of the SRV. During plant operation an SRV actuation is generally indicated by either a high temperature or pressure in the SRV discharge line.

Although these methods of monitoring provide a reliable indication that an SRV has opened, it makes the determination of the exact time and pressure at which the SRV started to open rather difficult.

The instrumentation associated with registering and recording the pressure in the reactor is generally limited to indicating the pressure in the reactor vessel dome region. By virtue of its physical location, the vessel dome pressure instrumentation is incapable of capturing the dynamic pressure effects which occur in the steam lines during rapid transient events that involve vessel pressurization sufficient to actuate the SRVs.

l I

, .. \

Attachment 3 I f i I

An example of such an event is a turbine trip. This event is initiated I by a rapid closure of the turbine stop valves, which sends a traveling )

pressure wave from the turbine stop valve through the steam, lines and into the reactor vessel. Once the wave reaches the reactor vessel dome it is attenuated due to the relatively large steam volume in the vessel dome region. A closure of the turbine stop valves during power ]

operation will generally induce a vessel pressurization of sufficient i magnitude to cause SRV actuations. Each SRV actuation will generate l additional pressure waves which travel down the steam line in both {

directions until they are either reflected or attenuated. Other factors I such as, subsequent MSIV closures, differences in the configuration of l the four steam lines, and the potential for small differences in the j initiation signals for closure and the closure rates of the turbine stop j valves or MSIVs, can contribute to the already complicated dynamics i which occur in the steam lines during a transient event. j l

Turbine trip tests have been conducted at an operating BWR which is very similar to FitzPatrick. As a part of the test procedure the steam lines at the plant were instrumented in order to capture some of these dynamic effects. The test revealed that variations in the pressure between the i steam line and the vessel dome of approximately 50 psi were experienced (Reference 2).

Based on the information presented above, it is concluded that for a rapid pressurization event it is very difficult to ascertain from the recorded vessel dome pressure whether or not an SRV actuation occurred l within the Technical Specification limit of 1% (or 11 psi).

References:

1

1) "Setpoint Drift Investigation of Target Rock Two-Stage Safety

]

Relief Valve (Final Report)", General Electric Company, NEDE-30476, j February 1984. l l

2) "BWR Refill-Reflood Pro' gram Task 4.8 - TRAC-BVR Model Qualification i for BWR Safety Analysis; Final Report", NUREG/CR-2571, EPRI I NP 2377,GEAP-22049, October 1983.  !

Prepared by: b;h[wn//! j K.F. Cornwell, Engineer l Application Engineering Services l I

I 1

I Approved by:

1.L. Sozzi,Maila(er Application Engineering Services 3-1

I

. I

,^ f.

Attachment 3 1.

1 Table 1 {

- 4 SRV Test Data for FitzPatrick l l

I i

1 Nameplate As Tested Opening M Setooint (nsin) Setooint (nsin) t Drift i

A 1140 * -

B 1140 1144 0,4 C 1140 1120 -1.8 D 1105 1145 3,6 E 1105 1243 12,5 F 1140 * -

G 1140 1131 ,-0,8  ;

H 1140 * -

J 1140 * -

K 1090 1083 -0.6 L 1090 * -

J

  • Valves not tested during this outage i

1 4

~

Attachment 3 1500 _ SERVICE

_ _ _ LEVEL

_ ._ C_LIMIT( 00)15 l

l l

1 I

J

_ j I

, 1400 _

o to (1395) j 6 ,,,

UPSET LIMIT (1375) l

{

5 l in g _

I, a:

a. (1354) d m

D g (1313) a.

1300 --

1 1

1

.- i (1226) 1200 ' ' ' ' ' ' ' '  !

O 100 200 SRV SETPOINT (PSI ABOVE UPPER NAMEPLATE LIMIT) l Figure 1: Effect of SRV Setpoint for an MSIV Closure with Flux Scram Event for a typical BWR 4.

l

,Attaphment 4-1400 UPSST LIMIT pgLGMid W 1300 l

w I

E

/

i i epce . ..

(. , ' E 4 EVENT SASIS.

l MSIV - FS O ALL VALYSS ASSUMED I

TO DMIFT

= l 1300 5

f I I t I

4 6 5 10 .)

0 2 I

  • SETPOINT OMIFT (% above nomined

( Figure 6-1. Sensitivity to Setpoint Drif t on All SRVs 6-6 f

l l l

t l