ML20235D063

From kanterella
Jump to navigation Jump to search
Ao:On 710422,during Analysis of Loss of Offsite Power Test Data,Initiation Setpoints of Four RCS Safety/Relief Valves Possibly in Excess of Tech Specs Limit Noted.Caused by Effect of Temp.Calibr Checks Done & Valve Settings Adjusted
ML20235D063
Person / Time
Site: Monticello, 05000000
Issue date: 04/30/1971
From: Duncanson R
NORTHERN STATES POWER CO.
To: Morris P
US ATOMIC ENERGY COMMISSION (AEC)
Shared Package
ML20235B311 List: ... further results
References
FOIA-87-111 2169, NUDOCS 8709250103
Download: ML20235D063 (3)
v • d • e


Text

,

l e

~

NSP NORTHERN 5TATES POWER COMPANY Minneapolis, Minnesota 55401 Regulatory File Cy.

' cv Ap ril 30, 1971 'b _

c}

d DOCKETG O S USAEC //

~~

3

~~

Dr. Peter A. Morris, Director 5 Ok Division of Reactor Licensing g gE g 6  % ."" #a44y ' /S/

}

United States Atomic Energy Commission DOCKET CLERK 6 g %7 *?a 7

Washington, D.C. 20545 g d l.C y ;7 a  %, s

Dear Dr. Morris:

4 QL, ., , g MONTICELLO NJCLEAR GENERATING PLANT Docket No. 50-263 License No. DPR-22 Safety / Relief Valve Settings Exceeding 1080 PSIG A condition occurred at the Monticello Nuclear Generating Plant which is reportable to your office in accordance with the provisions of Appendix A, Technical Specifications, of the Provisional Operating License DPR-22. The occurrence was classified as an Abnormal Occurrence as defined in Section I. A.1 of the Technical Specifications and requires reporting in accordance with Section 6.6.B.3 of the Specifications. The Region t il Compliance Of fice has been notified in accordance with the requirements of Section 6.6. A of the Technical Specifications.

Summarv Description of the Occurrence Analysis of reactor pressure data recorded during the " Loss-of-Offsite-Power" test performed on pA ril 22, 1971, indicated that the initiation setpoints of the four reactor coolant system safety / relief valves might be above the Technical Specification limit of 1080 psig. Calibration checks were formed on the pressure instrumentation and the settings of the safety /per-relief valves were tested under cold conditions, using nitrogen pressure.

These checks and tests indicated that the settings of three valves were above 1095 psig. The other valve was set at about 1080 psig.

All four valves were adjusted to initiate at less than 1080 psig.

De+ ailed Description and Analvsis l

As part of the startup test program, a " Loss-of-Offsite-Power" test was j performed at 25% of rated power. This test, as expected, resulted in a l reactor pressure transient. The wide range pressure recorder in the control room indicated that reactor pressure had increased to 1087 psig within approximately 24 seconds following the test initiation. At this time l safety / relief valve RV-2-71A opened, as indicated by a drop in pressure, a sharp increase in RV-2-71 A discharge temperature and indication from an accelerometer mounted on the valve. Pressure was reduced to 1042 psig

&92Q3870921 r f fj 25 f 3 !; 3 } [Gf MENZ87-111 PDR y~gg I Q

, z. . ,

... 'd

'within 8'secords and the valve closed. Duringthenext3{minutesthe pressure 1 increased slowly. to a peak value of 1100 psig, as indicated

. by the.wi de < range . recorder. The Disturbance Log, printed by the plant computer, .showed a peak pressure of 1104 psig at this time. Then RV-2-71 A opened a secord time. Pressure was reduced to' 1048 psig within 8 seconds.

Pressure then slowly increased to a peak of 1067 psig followed.by a steady decrease.

I The indicatior. that the seco$d initiation had o-ecurred at a pressure 13 psi above the first. initiation was puzzling. Also, since none of the remaining valves;had opened, it seemed that' the settings of the remaining ' valves were even higher.

The first initiation at.1087 psig would not have been considered unusual l by itself. TFe ASME code recpires a setpoint tolerance of one percent.

This means the valves could initiate as much as 11 psig above ' the nominal 1080 psig setting. Also, considering the various accuracy tolerances associated

~

with the pressure instrume.ntation, the recorded pressure could be in error by_ up te 19 psig. Additional errors can occur due to recorder chart variatiens. Similarly, the computer Disturbance Log could.be <

i in error by~ up to 16.5 psig. Even considering all of these' possible l errors, it appeared that at least three valves were set substantially above 1080 psig. Ft.rther investigation was therefore considered necessary.

A calibration check of the recorder _ and the computer point was performed.

At 1100 psig the recorder error was found to be + 5 psi and the computer error was fourd to be + 9 psi. Based on this calibration it appeared that the' first initiation of RV-2-71 A occurred at 1082 psig and the second j initiation at .1095 psig.

The four safety / relief valves' were inspected under the direction of the manufacturer's representative and the opening pressures of the first stage pilot valves were tested under ambient temperature conditions using nitrogen pressure. These tests and inspections showed the following:

RV-2-71A: The setting was essentially correct.

RV-2-718: The setting of the setpoint adjustment nut was found to be correct. However, an improper length dowel pin was found to have compressed the setpoint adjusting spring beyond the proper setting when the bonnet cap which covers the adjustment nut was bolted in

. place. This caused the actual initiation pressure to be at least 1110 psig.

RV-2-710: The setting was found to be approximately 1100 psig.

l RV-2-71 D: A slight leak in the firci stage pilot valve made it I

impossible to accuratel special test ecpipment.y determine However, the it can besetpoint inferred without that the setting was probably greater than 1095 psig.

i e l'

.? .

The observed 13 psi increase in .setpoint for the.second initiation- of RV-2-71 A is attributed tc the effect of temperature on the valve actuating system. When a valve is initiated, flow through the upper portions of the valve mechanism apparently results in sufficient additional heating.to affeet.a temporary setpoint increase. The manufacturers representative stated that an approximately is increase in setpoint would be expected as the valve is -

healed to normal operating temperature. This is _ confirmed by the results o_f the ambient temper'ature tests which showed the RV-2-71 A opening, point.

to be 1070 psig with nitrogen, whereas it had. initially opened during the loss of power test at about 1082 psig.

Correcti ve ' Ac+ iens The first stage pilot in RV-2-71D was repaired; the dowel pin in RV-2-71B was shortened to proper length; and all four valves were ' adjusted, at ambient temperature, to initiate at less than the equivalent of 1080 psig at normal operating temperature.

Conclus i ons The manufacturers certification of pressure setting had been obtained for all four valves prior to initial plant operation. However, only one of the four valves was found to be properly set. Two valves were apparently not properly adjusted and one valve was affected by an improper length dowel pin.

There is reasorable confidence that the valves are presently set properly.

The test gauge used to adjust the valve settings was compared with an accurate dead weight test set before and after making the adjustments.

During the remainder-of the startup test program particular effort will be made to monitor the safety / relief valve performance during transient tests.

Yours .very truly,

(*~ {) p-S%o<~.

P.E.

R.0.

Gen.Duncanson, Supt. of PowerJr. , Plants-Mechanical ,

Chairman Monticello Safety Audit Committee -!

ROD /MHC/caf l

i l

1

.1

?l a

Retrieved from "https://kanterella.com/w/index.php?title=ML20235D063&oldid=3192260"