05000389/LER-1986-007, Forwards Response to NRC 861001 Request for Addl Info Surrounding Event Reported in LER 86-007 Re Inoperable Pressurizer Safety Valves

Forwards Response to NRC 861001 Request for Addl Info Surrounding Event Reported in LER 86-007 Re Inoperable Pressurizer Safety Valves

ML20213D568
Person / Time
Site:	Saint Lucie
Issue date:	11/05/1986
From:	Woody C FLORIDA POWER & LIGHT CO.
To:	Thadani A Office of Nuclear Reactor Regulation
References
L-86-455, NUDOCS 8611120145
Download: ML20213D568 (11)

LER-2086-007, Forwards Response to NRC 861001 Request for Addl Info Surrounding Event Reported in LER 86-007 Re Inoperable Pressurizer Safety Valves

Event date:
Report date:
3892086007R00 - NRC Website

text

.-

• P. O. BOX 14000, JUNO BE ACH, FL 33408 FLORIDA POWER & LIGHT COMPANY NOVEMBER 5 1986 L-86-455 Office of Nuclear Reactor Regulation Attention: Mr. Ashok C. Thadoni, Director PWR Project Directorate #8 Division of PWR Licensing - B U. S. Nuclear Regulatory Commission Washington, D.C. 20555

Dear Mr. Thadani:

Re: St. Lucie Unit 2 Docket No. 50-389 Response to Request for Information on LER 86-07 Inoperable Pressurizer Safety Valves By letter dated October I,1986, you requested additional information surrounding the event reported in Licensee Event Report (LER) No. 86-07 for our St. Lucie Unit No. 2. The original LER was submitted by Florida Power and Light Company by letter L-86-212 dated May 19,1986.

The attached provides a complete response to your request.

Very truly yours, cl /

C. O. W y Group see President Nuclear Energy COW /GRM/gp cc: J. Nelson Grace, Region 11. USNRC Harold F. Reis, Esquire Attachment 8611120145 861105 9 GRM4/030/1 g DR ADOCK 0500 IE2.2 PEOPLE. . SERVING PEOPLE ilf j

QUESTION 1 EEC REQUEST:

The LER stated that valve V-1202 leaked too much to obtain a lift setpoint. What was the specific leakage quantity through the valve during the test? What is the flow capacity of the test facility used to setpoint test the valves? Was there any indication of valve leakage during Cycle 27 If there was indication of valve leakage during Cycle 2, what was done to justify continued operation? How long was this valve in a degraded condition? What was the root cause of the valve seat degradation? What was the corrective action in addition to rebuilding the valve?

FPL RESPONSE As per Section XI of the 1980 ASME Code, part IWV-3512, the pressurizer safety valves are bench tested with pneumatic equipment. The bench in use at the St. Lucie Plant employs an air compressor capable of producing 2.3 cfm, and the accumulator can reach a pressure of 4000 psig. When the test bench could not accumulate sufficient pressure under the disc seat, due to valve leakage, V-1202 was removed from the bench and the other safety valves were tested as stipulated in section IWV-3513. No leak testing was performed on V-1202 prior to rebuilding as per INV-35123 however, all three safety valves were rebuilt, and setpoints and absence of leakage subsequently verified as a part of the St. Lucie Plant's standard pressurizer code safety valve maintenance program.

Safety valve leakage was first indicated approximately mid-cycle during St. Lucie Unit 2's second cycle. As per Technical Specification 3.4.6.2.d and S,urveillance requirement 4.4.6.2.c, RCS inventory was balanced throughout the cycle at least once per seventy-two (72) hours. At no time was the total identified RCS leakage found to be greater than the Technical Specification limit. Thus, no justification was required for continued operation of Unit 2 due to safety valve leakage. Operability of the three safety valves had been assured prior to startup for Cycle 2 as per Surveillance Requirements 4.4.2.1 and 4.4.2.2. As per Technical Specifications, safety valve leakage is not considered indicative of a failure to perform the required safety functions; therefore, no justification for continued operation was required.

The valve seat was degraded due to steam cutting of the disk as the result of valve leakager the root cause of the leakage is not yet known, and no corrective action has bees identified at this time.

..-. - - .. . - . . - - . . . ..~ ._ . .. - -. -.-_ _ _ - .. -.

l OUESTION 2 l une annumsT:

. The LER stated that valve V-1201 lif t'ed at 2893 psi. When V-1201 was

disassembled, it was found that the bellows was ruptured. This allowed boric acid in the RCS to get at the valve internals. The boric acid caused extensive corrosion of some of the carbon steel components.

i~

The corrosion caused excessive binding that redulted in the lift setpoint being too high. What was the failure mechanism of the failed bellows?

Was leakage and/or boric acid corrosion a factor? Was this a maintenance related failure? How long was this valve in a degraded condition?

4 What was the corrective action in addition to rebuilding the valve?

, FFL RESPONSE:

4 An investigation into the cause of the valve leakage and bellows failure on valve V-1201 is currently underway by the Florida Power and Light engineering department at the request of the St. Lucie Plant.

l Conversations with the vendor (Crosby Valve and Gage Co. ) have revealed that this type of bellows failure is the first in the experience of

the manufacturer. A search conducted via the Nuclear Plant Reliability Data Service (NPRDS) confirms the vendor's comments. As the cause of the failure is not known, and there exists no related prior experience to draw upon, it is not possible to answer the NRC staff's questions regarding the failure -mechanisms, possible contributing factors and potential corrective actions at this time.

t j As stated in the response to question 1, pressuriser code safety valves are determined to be operable as per Surveillance Requirements 4.4.2.1 and 4.4. 2.2. The ASME Code referenced by these specifications provides

a surveillance time period and a test method, both of which the St.

l Lucie Plant complied with prior to beginning Cycle 2 and during the

! refueling outage at the end of the run. It is not possible to provide a concise date for the failure of the bellows and the subsequent binding l of the valve internals other than to state that it was within the bounds

set by the surveillance requirements for valve operability. However, j it should be noted that the failure, when found, was reported to the

} NRC staff and the valve refurbished. ,

4 4

i '

OUESTION 3 BC REQUWT The -most currently adopted industry standard for safety valve setpoint testing is ASME OM-1-1981. This standard requires setpoint testing using the operating fluid medium (2500 psia saturated steam) unless i an accurate correlation can be made using an alternate test fluid (such as cold air). Describe the current plant testing procedures which assure that the pressurizer code safety valves are within 14 of the setpoint specified by the plant's Technical Specifications.

, FPL R 5 705BE There are three procedures which govern the testing of pressurizer code safety valves at the St. Lucie Plant. Plant procedures are verified

, to be in agreement with ASME PTC 25.3-1976, as specified in IWV-3512 of the 1980 ASME Code.

4

! 1. Quality Instruction QI 11-PR/PSL-7, " Control of Safety and 1

Relief Valve Testing" -

delineates methods to be used and departmental responsibilities for implementing periodic testing to verify the operational readiness of code safety and relief

, valves.

2. General Maintenance Procedure M-0810, " Bench Testing of Safety Relief Valves" - provides instruction for the setpoint testing

, of safety and relief valves using bench testing apparatus.

. 3. General Maintenance Procedure M-0017, " Pressurizer Safety Valve Maintenance" -

delineates instructions necessary for

-the maintenance of the pressurizer safety valves.

{ The valves are tested on a test bench under cold conditions. As per the manufacturer's instructions and subsequent analysis by the FP&L Engineering Department, the valves are set to lift at a cold setpoint of 2530 psia + 14, and then leak tested at 90% of lif t pressure. The i response to question 1 described the flow capacity of the test bench

!~

at ambient conditions. Calibrated pressure test gauges are used to

! determine the lift pressure. The procedures for bench testing and maintenance of safety relief valves specify that a successful lift test must be repeated; a minimum of two openings at the same setpoint

, adjustment are required to demonstrate the valve's ability to open more than once within the required setpoint tolerance.

Leak testing is done by installing a blind flange with an outlet tube

+ on ~the discharge flange of the valves the outlet tube is immersed in a container of water. The pressure to the valve is raised to nominally 90% of setpoint pressure or 2265 psig (2280 psia) . Valve leakage is measured by recording the number of air bubbles generated per minute through the outlet tube in the container of water over a minimum five i minute period. A leakrate of less than eight air bubbles per minute is acceptable with the Maintenance Supervisor's concurrence.

I

OUESTION 3 Page 2 Both the lift setpoint and leak rate procedures must be witnessed and verified by a Quality Control Inspector. Test failures are recorded via Noncomformance Reports and must be dispositioned by the Engineering Department prior to startup.

i-4 I

i

+-7-n---- -- ,, , , - ,---,.--,----..n.,,,.--,-,-,_.-. - . - , - . - _ , - , ,.-w--. , _ , , , , - -

..,,4 e_, , , - - - - , , --

~ , .._ - ._._ , ._ .

OUESTION 4 ,

WBW:

In regard to all three pressuriser code safety valves, what is each 4

valve's maintenance / rebuilding history and what is the current J!

operability of each valve? Is any pressuriser code-safety valve leaking at this time and, if so, what is the leak rate?

I FFL RESP M E:

l l A summary of the maintenance and rebuilding history for all three pressuriser code safety valves is provided by Table 1, " summary of Maintenance History -

Unit 2 Pressuriser Code Safety Valves." The table shows that all pressuriser code safety valves have either been rebuict or, in the case of V-1202, replaced during each refueling outage.

3 Prior to reinstallation, the setpoint of each valve is tested twice for verification of repeatability; both' lifts are independently verified

by a plant Quality control Inspector. .This testing had been done according to plant procedures, and the valves subsequently reinstalled, prior to Cycle 3 startup. Therefore, all pressuriser code safety valves are . now operable as per the requirements of Technical Specification surveillance 4.4.2.2.

.At this time, evidence of minor intermittent leakage of the safety valves has been identified on Unit 2; however, no valve is currently l exhibiting evidence of sustained leakage.

i a

4 l

9 1

I I

1984 DEPUELING OUTAGE V-1200 V-12@l V-1202 SETPOINT DATA (PSIG,

$ COLD):

3 As Found 2300 2475 2600

• As Left 2515 2510 2515

$ Leakage Rate After j Rebuild Satisfactory Satisfactory Satisfactory

" (Bubbles / Min.) 4 Bubbles / Min. Neglible (< 1 Bubbles / Min.) (Data Not Available) fu Ca== ants Valve was rebuilt prior to reinstallation; valve was rebuilt prior to reinstallation; Leakrate test results recorded only as

• Setpoint verified by Setpoint verified by satisfactory; valve was t two lifts two lifts rebuilt and setpoint a verified by two lifts g

1986 REFUELING OUTAGE

• n N

N SETPOINT DATA (PSIG, 5 D COLE):

' As Found 2554 2893 Not Determined

[ As Left 2525 2510 2500 0*

Leakage Rate Satisfactory Satisfactory Satisfactory

• (Bubbles / Min.) O Bubbles / Min. Negligible Leakage Negligible Leakage 8

$ Ca== ants As found Setpoint As found Setpoint New valve was installed

$ determined by Furmanite as determined using prior to Cycle 3 heatup.

p Trevitest Equipment Furmanite Trevitest Setpoint of valve was and test procedurer bellows rupture & was determined for two E acceptance criteria internal corrosion lifts prior to y did not change. noted. Valve-was _ installation.

> Subsequent test after rebuilt, nozzle was k

rebuild done on plant remachined, and setpoint test bench. Nozzle verified to two lifts

, was remachined due to prior to reinstallation, steam cutting. Setpoint verified to two lifts.

O

OUESTION 5 KEM:

What plant . features or operating procedures are in ' place which assure i that valve leakage is detected? If valve leakage is found, what criterion / criteria is used to declare a valve inoperable?

FFL RESPOWBE St. Lucie Plant Unit 2 has two different types of instrumentation which monitor for evidence of pressurizer code safety valve leakage.

1. Each safety valve tail pipe is equipped with a safety grade RTD temperature detector; each detector has its own indicator-on the control room board. A high temperature alarm will annunciate if the temperature of the associated piping reaches

, 280*F.

2. Each safety valve tail pipe is equipped with an acoustic flow monitor which indirectly monitors for evidence of flow through the discharge line by monitoring the acoustic activity in the line. The acoustic signal is processed and indicates relative flow for each tail pipe on a LED bar graph display on the plant auxiliary panel in the control room. A common alarm is actuated if flow on any safety valve or PORV line

, exceeds approximately 40 gpm.

Additional evidence of safety valve leakage is provided by the quench tank level and temperature indications.

, In addition, two periodic safety valve checks are performed e

1. The Reactor Control Operator Log requires that the temperature of each code safety discharge line to taken on a twice-shiftly basis; i

2. a RCS inventory balance is calculated every 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />.

i The existing criteria for verification of pressurizer code safety valve

operability are Surveillance Requirements 4.4.2.1 and 4.4.2.2, which

! are met prior to cycle startup after each refueling outage. The

! criterion for determining acceptable RCS leakage rate is Specification 3/4.4.6.2.

t l

[

, - . , - - ,...-,,,..,,-,e~n,,.--._nn,ennn.,,_,,-nn,, , ,...,.n....._,.._,n..,,~m.,e.,.n.,_,-,.--.,, , , - -,,-----n.--w ~ , - - , - ,

OUESTION 6 W REQUEBT The Technical Specifications require each power operated relief valve block valve to be operable and no more than one block valve to be open at any one time while in modes 1, 2 and 3 (TS 3.4.4). Were both block valves operable during Cycle 2 while in modes 1, 2 and 37 Although not a TS requirement, were both PORV's operable during Cycle 2 while in modes 1, 2 and 37 Was the power removed at any time from any block valves during cycle 2 while in modes 1, 2 and 3? "are both block valves closed at any time during Cycle 2 while in modes 1, 2 and 37 FPL REBPOMBE Research of plant files and interviews with plant personnel indicate that both PORV block valves were operable during Cycle 2 while in modes 1, 2 and 3. Plant records indicate that power was not removed from either block valve during Cycle 2 while in modes 1, 2 and 3, nor was there any time during the cycle when both block valves were concurrently closed.

The determination of the operability of the PORVs is performed according to Surveillance Requirement 4.4.9.3.1. As per this requirement, both PROVs are operable.

OUESTION 7 ERC REQUEST:

Demonstrate that reactor coolant system overpressure protection was not lost at any time during Cycle 2 while in modes 1, 2 and 3.

FPL RESPOWBE Appendix 5.2A of the Unit 2 FSAR documents the valve sizing procedure for the overpressure protection of the RCS. The design basis for safety valve capacity is to maintain the RCS below 110% of design pressure during the worst case transient, loss of turbine load with delayed reactor trip. Figure 5.2A graphically illustrates the relationship of safety valve capacity to peak RCS pressure during this event. Using additional data provided by Combustion Engineering to extrapolate this curve, the peak RCS pressure for 33% of design valve capacity is found to be approximately 1084. When the elevated lif t point of valve V-1200 (14 psi, or 1/2%) is considered, the peak RCS pressure reached is 108.5%

of design pressure. It should also be noted that the PSAR analysis assumes the worst case initial conditions and nuclear parameters and assumes no PORV participation in the event.

Based upon the above, FP&L is the opinion that the observed as found condition of the pressurizer code safety valves would have resulted in a DBA peak RCS pressure of less than 110%.

SINEIARY

1. During the 1984 Unit 2 Refueling Outage, all three pressurizer code safety valves were rebuilt and tested prior to reinstallation as per the requirements of Section XI of the 1980 ASME Code, part IWV-3500. The valves were verified operable as per the requirements of the code and St. Lucie Unit 2 Technical Specificiations 4.4.2.1 and 4.4.2.2. -

2. Approximately mid-cycle of Cycle 2, leakage was detected through the code safety valves. Leakage was monitored as per Surveillance Requirement 4.4.6.2.1; at no time were Technical Specifications violated.

3. During the 1986 Cycle 3 refueling outage, the code safety valves were again tested for operability as per Surveillance Requirements 4.4.2.1 and 4.4.2.2. The valves failed, and a Licensee Event Report was subsequently issued. Two of the three code safety valves were rebuilt and reinstalled; a third was replaced. Prior to reinstallation, the surveillance requirements were met and all three valves declared operable.

4. At the present time, the valves have exhibited signs of minor intermittent leakager no . evidence of continued leakage of any of the pressurizer code safety valves currently exists.

5. FPL is conducting an investigation into the causes of the safety valve leakage and the bellows failure experienced by one valver however, this reasearch is currently underway and no results are available at this time.

1

- _ _ - _ _ ___ . _ _