IR 05000295/1988015

Insp Rept 50-295/88-15 on 880628-30 & 0803.Open Item Re Bench Testing of Crosby Valves Noted.Major Areas Inspected: LER 88-012 Re Test Methods of Main Steam Safety Valves Exceeding High Setpoint Tolerance When Tested by Steam

ML20151U353
Person / Time
Site:	Zion File:ZionSolutions icon.png
Issue date:	08/11/1988
From:	Danielson D, Huber M, James Smith NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
Shared Package
ML20151U337	List: IR 05000295/1988015 ML20151U331
References
50-295-88-15, NUDOCS 8808190049
Download: ML20151U353 (6)

Text

c- --

. '.

.

.

V. S. NUCLEAR REGULATORY COMMISSION

REGION III

Report No. 50-295/88015(DRS)

Docket No. 50-295 License No. OPR-39 Licensee: Commonwealth Edison Company P. O. Box 767 Chicago, IL 60690 Facility Name: Zion Station, Unit 1 Inspection At: Zion Site, Zion, Illinois Inspection Conducted: June 28-30

-

'

Inspectors:

M.P. Hub (eF A, and August ,/' 3, 1988

'

--

Date

/

'

-

d.' [.IIit 8-//-(C

/ Date ML sb~ p Approved by: D. H. Danielson, Chief Materials and Processes Date Section Inspectic.n Summary Inspection on June 28-30, and August 3, 1988 (Report No. 50-295/88015(DRS))

Areas Inspected: Unannounced special safety inspection to review the technical aspects of Zion LER 88-012 concerning the test methods of 20 Main Steam Safety Valves (MSSV) which exceeded the high setpoint tolerance (1%) when tested by steam at Crosby Valve Compan Results: Of the areas inspected, no violations and no deviations were identifie One open item was identifie * All 20 MSSV's were reset at Crosby Valve Company by bench testing with steam to achieve operation within 1% of nominal setpoint. The licensee has committed to investigate and test as necessary to determine the cause of the apparent setpoint drifting by December 20, 198 GS0819004C 88- t5 PDR ADOCK C' 295 Q .DC

-  ! -

.

.

I

.

DETAILS > Persons Contacted Comonwealth Edison Company (CEcol

• G. Plim1, Station Manager a * Paul J. Beinecke, Thermal Group Leader

• Richard Harwood, Technical Staff Engineer Ken Hayes, Thermal Group

• Robert C. Johnson, Assistant Superintendent, Maintenance Terry Printz, Assistant Technical Staff Supervisor Terry Rieck, Superintendent, Service

• Willie Stone, Regulatory Assurance Supervisor

' * Denotes those attending exit meeting cn August 3,198 . Follow-up on Licensee Event Report (LER) 88-012-00 s i

LER 88-012-00 identified a condition that resulted in the plant being '

, operated outsida the design basis. All 20 Main Steam Safety Valves l (MSSV) failed the 1% setpoint pressure tolerance testing performed 10 q accordance with Technical Specification Table 4.7- Background

. In accordance with ASME Boiler and Pressure Vessel Code Section XI

'

and the Zion Technical Specifications, Safety Reliuf valves were tested to detemine that the lift setpoints were within allowable limits. In this case the setpoint tolerance was defined as 1% in the Technical Specification. The valves were sent to Crosby Valve j Company to be tested using steam condition l When the testing was perfomed, the as-found setpoint pressures were greater than the allowable 1% deviation from the-required setpoint, i

on the high side. Subsequently, it was determined that the condition -

that resulted from the higher setpoints was beyond the design basi Specifically, the 20 MSSV's would not pass sufficient flow to

, maintain steam temperatures at 560*F or below when neede The MSSV's

were adjusted to ;he required pressure setpoints ano retested prior

.

to reinstallation and startup following the outage.

!

'

The NRC inspectors reviewed the technical aspects of the testing related to the valve failures through discussions with licensee personnel and a review of records, to determine that imediate 4 corrective action was accomplished, and corrective actions were i

taken to address the future testing of the valves in order to prevent recurrenc .

. _ _ _ - _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ ___

.

. ;

,

-

.

b. Design Basis of MSSV's MSSV's' serve two functions

1 to provide Main Steam System overpressure protection and 2))as a. heat sink for reactor cooldow In the event of load losses without availability of the steam dump system, the MSSV's are sized to protect the reactor coolant system (RCS) and the steam generators against overpressur For the loss of load transients where MSSV's provide the heat sink, it was assumed in the FSAR that the 20 MSSV's would relieve in order to maintain the saturated steam generator pressures _ low enough to keep the steam temperature below 560' The combined capacity of all 20 MSSV's would relieve the total steam ,

flow of the unit. With the loss of MSSV capacity, the reactor trip setpoints were restricted by-Technical Specifications to correspond to the maximum relieving capacity of the most restrictive condition In the FSAR the unit was evaluated for a complete loss of load from 102% of full power without a direct reactor tri During the worst case (loss of load at beginning of core life) p. transient, the MS$Y setpoint was reached at 16 seconds and was required to relieve a peak flow of 56% of the flow at rated full power at 18 second In the case of extensive setpoint drift, the higher MSSV setpoints resulted in a condition that resulted in operation in an unanalyzed condition. Using the "as-found" setpoint data, the maximum steam conditions would have been approximately 1143 psia and 561* The significance is discussed belc ! c. History of Main Steam Safety Valve Setting at Zion Until the outage of 1936-87, all MSSV's were set and tested using steam as a medium. The valves were tested for as-found setpoint .:.d this data was recorded. They were then cleaned, refurbished as ,

necessary, and reset to within a specified range of the nominal setpoints. The as-left setpoints were then teste Valves with tests falling outside of requirements were reset and retested i as necessary to produce acceptable results. As-left test values were then recorde The licensee's Technical Specification and ANSI /ASME PTC 25.3-1976 a'

both permit the use of a calibrated auxiliary device or bench ,

testing using compressed gas as the test medium. Neither of these i documents mandates the establishment of a correlation between test

,

results obtained with steam and those obtained by bench testing i with other media. When the use of nitrogen testing was applied in the 1986-87 outage, it was used as a direct replacement for steam J testin :

3

.

. ;

-

.

In the 1988 outage, when the valves were sent to Crosby Valve Company for testing on steam the same philosophy prevailed. The steam test results were used for direct comparison with those of the nitrogen test, by which the valves were previously set and tested. At this point the violation of the setpoint tolerance range was identifie All 20 MSSV's exceeded the i 1% setpoint tolerance on the high side, d. Discussion A review of the licensee's activities in the area cf setpoint setting and testing disclosed no new violations of pertinent codes nor of the Technical Specification. The licensee perfomed all work in confomance with procedures existing at the time and according to methods pemitted by the Technical Specificatio In order to determine the effect of the high setpoints on the operatien of the plant, the licensee's Nuclear Fuel Services Group has undertaken the analysis of the worst case transient where MSSV's open at a higher pressure than required. The investigation will be documented in NFS Report "RSA Calc Note Z88-03". The licensee's preliminary results c,f that investigation conclude that the higher setpoints reported by Crosby Valve Company do not constitute a significant safety concer The licensee has committed to investigate and test as deemed ,

necessary, to detemine the cause of the apparent setpoint ;

drif ti ng . This investigation will include a correlation of the affect of various foms of setpoint testing on the resulting setpoints when tested using stea The methods to be investigated include setpoint testing with nitrogen and setpoint testing with steam and an auxiliary calibrated lifting device. This comitment is numbered 304-200-88-01800 and has a scheduled completion of December 20, 1988. Although the results of the licensee's program are not available to provide firm evidence on which to base a conclusion, several observations suggest that drifting was not the sole cause of the out-of-tolerance setpoints. These include the following points:

,

EPRI NP-4235 dated September 1985, "Setpoint Testing of Safety

. Valves Using Alternative Test Methods," states that the steam

.

test showed an overall decrease in setpoint from the nitrogen l

'

test data. This means that valves set with nitrogen show a ;

higher setpoint than those set in steam when both are subsequently tested in the same medium. Thus a valve set using nitrogen would indicate a high setpoint when tested on ,

steam. Conversely, a valve set on steam would indicate a low '

setpoint when tested on nitrogen. The lower setpoint observed when using steam as the test medium is attributed to the l elevated temperature of the valve caused by the elevated l ambient temperature and the high pressure steam at the valve j inle l

.

l

.

,

.'

-

.

ANSI /ASME OM-1,1981, "Requirements _for Inservice Testing of Nuclear Power Plant Pressure Relief Devices," is not applicable to the Zion Station. This standard is of interest only in that it identifies that differences are likely between tests involving different test media. It states, in 4.3, "Alternate Test Media." that pressure relief devices may be subjected to set pressure tests using a medium other than that for which they are designed, provided correlation of pressure relief device operation is established between operation in the alternative medium and operation in the normal operating medium. The results must be certified and performed in accordance with a written procedur *

Until the 1986-87 outage, all safety valves were set using steam and no outstanding problems in setpoint drifting were observe *

During the 1986-87 outage, when the valves which were previously set on steam were tested using nitrogen, all valves were found to be below the nominal setpoint *

During the 1986-87 ottage, the setpoint procedure in use was P/M003-5N, "Bench Testing Section XI Safety Relief Valves (For Use on Vapor / Gases)," which was approved September 8, 198 This procedure was cited for a violation in NRC Inspection Report No. 295/88012(ORP) dated June 24, 198 The violation was that the procedure allowed a setpoint tolerance of i 2% although the Technical Specification allowed only 21%. This condition was reflected in the as-left settings of the MSSV's set with nitrogen which averagcd 1.3%

over the nominal setpoint *

The as-found settings of a sample of the valves tested by steam at Crosby Valve Company averaged 4% over nominal setting Subtracting the 1.3% that the values were high when set in nitrogen, this leaves a difference of 2.7% between the values for valves set by nitrogen and those observed for steam testin A preliminary review of the above available evidence suggests that the direct change from steam testing to nitrogen testing and then back to ste0m testirg could either contribute to or be primarily responsible for exceeding the i 1% tolerance on MSSV setpo%ts as ;

found at Crosby Valve Company, i

!

The valves were set at Crosby using steam, and steam testing is ;

generally acknowledged to be the best available method. They are I all within the 1% tolerance required by the current procedure and ,

Technical Specification. Future testing will be performed using :

steam plus a calibrated auxiliary lifting device. The presence of j l

l

'

l

'

- a

.

.

steam is expected to more closely approximate the conditions of service because of the MSSV's resulting higher temperatures. The commitment to investigate the cause of the setpoint drifting is expected to also- -

provide a correlation between the results of valves set in steam and valves set with steam and auxiliary force. Such correlation will provide additional confidence that the setting will meet the setpaint telerance requirement With the currently installed valves being set in accordance with optimum procedures and future methods given the benefit of the results of Commitment No. 304-200-88-01800, it appears that all appropriate measures to resolve this matter with regards to the testing have been initiated. The completion of this commitment by December 20, 1988, isconsideredanopenitem(50-295/88015(ORS).

3. Open Items Open items are matters which have been discussed with the licensee, which will be reviewed further.by the inspector, and which involve some action on the part of the NRC, the licensee, or both. An open item is discussed in Paragraph ,

4. Exit Interview

'

The NRC inspectors met with licensee representatives (denoted in Paragraph 1) on August 3,1988, to discuss the scope and findings of the l inspection including the one open item. The inspectors also discussed the likely informational content of the report and summarized the results. The licensee acknowledged the statements made by the inspectors a with respect to the content of the report. The licensee did not identify any such documents of processes proprietar .

-

f

'

6