ML18095A232

Part 21 Rept Re Discrepancies Identified in ECCS Flow Orifice K Factors Supplied by Westinghouse.Initially Reported on 900518.Util Fabricated New Orifice Plates to Original Specs

ML18095A232
Person / Time
Site:	Salem
Issue date:	05/24/1990
From:	Crimmins T Public Service Enterprise Group
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
REF-PT21-90 NLR-N90111, NUDOCS 9006040212
Download: ML18095A232 (4)

Text

)*

,. - f ...

Public Service Electric and Gas Company Thomas M. Crimmins, Jr. Public Service Electric and Gas Company P.O. Box 236, Hancocks Bridge, NJ 08038 609-339-4700 Vice President - Nuclear Engineering MAY 2 4 1990

..\ NLR-N90111 United States Nuclear Regulatory Commission Document Control Desk Washington, D.C. 20555

-Gentlemen:

NOTIFICATION OF NONCOMPLIANCE; ECCS FLOW ORIFICE K FACTORS SALEM GENERATING STATION UNIT NOS. 1 AND 2 DOCKET NOS. 50-272 AND 50-311 In accordance with the requirements of 10CFR21, Public Service Electric and Gas Company (PSE&G) forwards the attached report, regarding discrepancies identified in vendor supplied orifice K factors. Incorrect K factors can result in actual pump flow exceeding indicated pump flow. The potential exists for Technical Specification violation and operation in an unanalyzed condition during accidents (i.e., pump runout, inadequate NPSH, or inadequate core injection flow).

The attached report provides detailed information about the circumstances that led to the reportability determination. PSE&G initially reported this concern on May 18, 1990, in a conference call between R. Drewnowski, H. Berrick, c. Williamson, W. Grau, R. Brown, of PSE&G, and T. Johnson, NRC Sr. Resident Inspector.

Should you have any questions regarding this matter, please do not hesitate to contact us.

Sincerely,

~Ju~

Attachment

~ 90060.£1.0212 906~s24-~--~.

PDR ADOCK 05000272 S PDC

Document Control Desk 2 NLR-N90111 c Mr. J. c. Stone Licensing Project Manager - Salem Mr. T. Johnson Senior Resident Inspector Mr. T. Martin, Administrator Region I Mr. Kent Tosch, Chief New Jersey Department of Environmental Protection Division of Environmental Quality Bureau of Nuclear Engineering CN 415 Trenton, NJ 08625

NLR-N90111 Attachment EVALUATION OF NONCONFORMANCE ON ORIFICE K FACTORS Introduction On May 18, 1990, PSE&G reported a nonconformance concerning flow measuring orifice K factors supplied by Westinghouse.

Improper K factor usage can result in actual flow deviating from indicated flow in the nonconservative direction.

The term K factor represents a relationship between volumetric flow rate and measured differential head, as a result of the change in velocity across an orifice. The K factor is the ratio used to convert the measured differential head to a flow rate. K factor is derived mathematically as a function of many variables:

internal pipe diameter, orifice diameter, fluid temperature, density, viscosity, flow rate and pressure tap locations.

Chronology of Events December 19, 1989, PSE&G received a letter from Westinghouse entitled "ECCS Flow Inconsistencies". PSE&G Engineering began a review to determine the impact at Salem. An inconsistency was identified in the K factors used for the 1.5 inch cold leg injection orifices, within the Charging/Safety Injection system.

Westinghouse originally specified a K factor of 17.7. PSE&G Engineering recalculated a value of 18.1 in 1982. PSE&G had been using the 18.1 K factor since 1982.

On April 9, 1990, Salem Unit 1 tripped off line. The Technical Specification surveillance on ECCS flow was required prior to Unit Restart. Engineering determined that the 18.1 K factor would yield acceptable results, but 17.7 would be marginal or unacceptable. Engineering worked to determine the correct K factor. An independent calculation was performed using the primary ASME methodology for flange tap applications. The result agreed with the original Westinghouse K factor of 17.7. This value was assumed to be correct.

The Technical Specification surveillance was completed on Unit l's number 12 Charging/Safety Injection pump and evaluated as unacceptable.

A decision was made to test the existing orifices in their as-found condition, to resolve any doubt concerning the calculated K factors. An independent Laboratory, approved by the National Bureau of standards, tested the orifice plates. The test results indicated an average K factor of 19.4. PSE&G Engineering reviewed the test practices to identify any procedural errors. No deficiencies were noted. The test data was repeatable and considered valid. This raised questions concerning potential degradation of the orifices (they were

II',

,~/'

NLR-N90111 slightly larger in diameter then specified by Westinghouse) and the applicability of the ASME equation used.

While repair was in progress on number 12 Charging/Safety Injection pump, PSE&G decided to replace all metering orifices in this system with new orifice plates fabricated to the original specifications. One orifice of each type was tested to verify calculated values.

Laboratory testing of the new 1.5 inch orifice plates yielded an average K factor of 18.9. This value was close to the tested value of the original orifice plates, allowing for the differences in orifice diameter.

Further investigation by PSE&G Engineering uncovered the reason for the discrepancy between the calculated and tested K factor.

The ASME standard, although not disqualifying the primary equation for small bore applications, does provide a separate equation for pipe in the 1-1.5 inch internal diameter range, as a stand alone supplement. However, the user is not explicitly directed to this alternate method for small bore piping.

Recalculating the K factor for the 1.5 inch orifices using the alternate method resulted in a K factor of 19.4.

Conclusions It appears that Westinghouse and PSE&G failed to detect the limitations of the ASME primary equation for small bore piping.

Nonconformance identification occurred as a result of PSE&G's decision to test the new orifice plates. Low K factors can result in actual pump flow exceeding indicated pump flow. The potential exists for Technical Specification violation and operation in an unanalyzed condition during accidents (i.e., pump runout, inadequate NPSH, or inadequate core injection flow).

Corrective Actions PSE&G fabricated new orifice plates to the original specifications. One orifice of each type was tested by a National Bureau of standards approved laboratory, to determine the K factor. Calculated K factors using the alternate equation compared favorably to the tested values. Subsequent pump surveillance testing met the Technical Specification requirements.

Because the ECCS Technical Specification acceptance band at Salem is very tight, PSE&G plans to use the tested K factors in this particular application, even though a favorable correlation exists between the calculated and tested values.