ML20079J295

Updated Final Part 21 & Deficiency Rept SD-370/82-06 Re Evaluation of Itt Barton Class IE Transmitters Exhibiting Thermal Nonrepeatability.Safety Analyses Involving Protection Functions Using Transmitters Conservative

ML20079J295
Person / Time
Site:	Mcguire
Issue date:	01/11/1984
From:	Tucker H DUKE POWER CO.
To:	James O'Reilly NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
References
REF-PT21-84-244-000 PT21-84-244, PT21-84-244-000, SD-370-82-06, SD-370-82-6, NUDOCS 8401240282
Download: ML20079J295 (5)

Text

a u -

~ <

DUKE POWEd GOh1PANY-P.O. HOX 3318D GHAHLOTTE, N.C. 28242 HALD. TUCKER TER.EPHOME

= = = = = = 1984 (ro4) a7a-4sas

. - . . . . . - - , - f ojg ; 3 5 January,,l'g Mr. James P. O'Reilly, Regional Administrator U. S. Nuclear Regulatory Commission Region II 101 Marietta Street NW, Suite 2900 Atlanta, Georgia 30303 Re: McGuire Nuclear Station Units 1 and 2 Dociet Nos. 50-369 and 50-370 Significant Deficiency Report SD-370/82-06

Dear Mr. O'Reilly:

My December 10, 1982 letter submitted (pursuant to 10 CFR 50.55e) an interim Significant Deficiency report (SD-370/82-06) concerning ITT Barton Class 1E transmitters exhibiting thermal non-repeatability. "ince our last report dated April 28, 1983, further review by Westinghouse of the thermal non-repeatability problem associated with Westinghouse procured Barton transmitters has determined an unreviewed safety issue as defined under 10 CFR Part 50.59. The attached final update to SD-370/82-06 which has been expanded to also include McGuire Unit 1provides an evaluation of this safety issue with respect to McGuire Units 1 and 2, and concludes that the safety analyses involving those protection functions utilizing the Barton transmitters (low steamline pressure, low pressurizer pressure - S1 and reactor trip) have sufficient conservatism to include the additional errors. No corrective action appears necessary other than discussed in previous correspondence on this issue. Note that these problems were among the subjects of NRC/ ole Information Notice 83-72.

Very truly yours, b'

Hal B. Tucker f(/

PBN:jfw Attachment cc: Director Office of Inspection and Enforcement U. S. Nuclear Regulatory Commission Washington, D. C. 20555 Mr. W. T. Orders NRC Resident Inspector McGuire Nuclear Station I H h :1.y, n . .

hN ,

8401240282 840111 l I-PDR ADOCK 05000370 S PDR

DUKE POWER COMPANY MCGUIRE NUCLEAR STATION SIGNIFICANT DEFICIENCY REPORT NUMBER: SD-370/82-06 (Final Epdate)

REPORT DATE: January 11, 1984 FACILITY: McGuire Units 1 and 2 IDENTIFICATION: In a further review of the thermal non-repeatability problem associated with Westinghouse procured Barton transmitters, Westinghouse has determined an unreviewed safety issue as defined under 10CFR Part 50.59. This matter was initially discussed in Mr. E. P. Rahe's (W) letter to Mr. R. C.

DeYoung (NRC/0IE) dated October 13, 1983.

BACKGROUND AND DESCRIPTION: Based on a report of excessive errors at abnormal temperature conditions by one of their customers, Barton performed static tempera-ture calibration checks on several transmitters. As a result of this investigation, Barton discovered excessive errors at both abnormal and accident temperature condi-tions and determined two separate causes. Barton has previously reported this to the NRC via their October 29, 1982 and June 30, 1983 (Attachment A) telexs to Mr.

R. C. DeYoung.

t Barton's calibration technique for transmitter temperature compensation was found to result in previously undetected errors at both abnormal and accident temperatures.

As part of the compensation process the zero output of the transmitter was elevated in order to be able to observe negative errors. This procedure introduced false temperature errors which were then incorporated into the transmitter compensation.

The units were not checked at the elevated temperatures after the original zero was restored and therefore they were shipped with excessive temperature compensation.

The evaluation conducted by Barton showed that the resultant error would always be i positive. This compensation technique results in an overall change in the specified accuracy that was assumed for these transmitters. The accuracy deviation that results from this procedure potentially affects safety-related transmitters procured by Westinghouse from Barton Lots 1 thru 7.

During the investigation process, Barton also discovered an electrical leakage path thru the wiper arm and shaft of the zero and span calibration potentiometers to the instrument case. This path only creates significant errors at high temperatures and is only of concern during accident conditions. Transmitters procured by Westinghouse from Barton Lots 1, 2, 4, 5, and 7 are potentially affected.

Based on static calibration data received from Barton on a sample of approximately gggggg

. . . , . eighty transmitters representing Lots 2, 4, 5, and 7, Westinghouse has calculated expected error deviations and evaluated the effect of any additional deviation on

)p (;g g . ~. q functions performed by these transmitters. This evaluation has shown that there

.y'.5 e is a potential concern for the McGuire Units 1 & 2 pressurizer pressure control y ? l ,'. function. The maximum additional error under normal conditions for these elevated c_s 'y . zero transmitters is predicted to be a bias of +1.5% of transmitter span. Thie 7*83 6 positive bias will lead to a negative error in the control band of 12 psig (1.5%

lti x 800 psig = 12 psig) or a resultant control band of +30, -42 psig for the pressur-32.,f j izer pressure control function.

• • T

.!k.N p' *e__ -,

k .

s 1%~e

~

l l

Page 2 l

EVALUATION RESULTS: The additional errors introduced for those protection functions utilizing the affected Barton transmitters (low steamline pressure, low pressurizer pressure - S1 and reactor trip) have been evaluated. The safety analyses for which these functions were utilized were determined to have sufficient conservatism to

. include the additional errors.

The error '(bias) . introduced by the transmitters for pressurizer pressure resulted in a controller error of -42 psi. . The safety analysis assumption currently assumes -30 psi. Upon evaluation of the safety analysis the impact of this additional -12 psi introduced a DNB concern.

For both McGuire Unit 1 and Unit 2, standard thermal hydranc.ic methodologies were used in analyzing both McGuire 1 and McGuire 2 cycle 1 designs. Both units have measured. low reactor coolant flow and both units are equipped with pressure biased Barton transmitters. The following is a discussion of the DNBR margins and penalties

' associated with cycle 1 and justification for allowing operation with the pressure biased Barton transmitters.

Both McGuire Units have 9.1% generic DNBR margin associated with cycle 1 analyses.

This is due.to'conservatisms used in the analyses in comparison to conditions justified in Chapter 4 of the McGuire FSAR. The following is a sumuary of the conservatisms used in McGuire's cycle 1. analyses.

Design Limit DNBR of 1.30 vs. 1.28 Grid Spacing (Ks) of 0.046 vs. 0.059 Thermal Diffusion Coefficient of 0.038 vs. 0.051 DNBR Multiplier of 0.86 vs. 0.88 Pitch Reduction LThis generic margin is normally used to partially offset rod bow DNBR penalties.

-Due to low flow measurements in Unit 1 (1% flow) and Unit 2 (2% flow), a 1.6% and 3.2% DNBR penalty was applied respectively. McGuire end of cycle 1 rod bow DNBR penalty using current technical specifications methodology is 6.8%. The 9.1%

generic' margin is enough to offset the flow penalty and rod bow penalty for Unit 1 but is not enough for Unit 2. As a result, a more restrictive rod bow F A H penalty curve had been provided to Duke Power for Unit 2's technical specifications. A primary system pressure bias of.12 psi (applicable to the Barton transmitters used in both McGuire units) corresponds to a DNBR penalty of 1.2%. With this additional penalty, the generic margin would not be enough to offset the flow, rod bow, and pressure penalties for either unit. Therefore, new rod bow F AH penalty curves would have to be supplied for McGuire's technical specifications.

Current technical specifications contain a rod bow F AH penalty curve based upon the methodology'used before NRC approval of WCAP-8691 Revision 1, " Fuel Rod Bow Evalua-tion", July 1979 (Skaritka, J.,.(Ed)). McGuire Cycle 2 DNBR analyses, which have recently been submitted to the NRC (ref. Mr. H. B. Tucker's December 12, 1983 letter to Mr. H. R. Denton), performed for the implementation of the OFA design, have shown that new methodologies (Improved Thermal Design Procedure /WRB-1 DNB correlation and new rod bow methods) increase the DNBR margin (i.e., margin between design limit DNBR'and safety limit DNBR). By using new methodologies and accounting for reduced

• ^

r Page-3 flow specifically in the analyses, the total DNBR penalty at the end of cycle 1 would be 2% for both units for rod bow and pressure bias effects. The 10%

available margin between the design limit DNBR and the safety limit DNBR would be more than enough to offset this penalty.

- Westinghouse recommends that McGuire's technical specifications not be changed to include the effect of the. pressure biased Barton transmitters since analyses have shown that with the use of new methodologies, approved by the NRC in connection with other licensee submittals, operation under the current technical specifications in fact results in sufficient margin through cycle 1. Therefore, the existing technical specifications are deemed adequate for the remainder of McGuire cycle 1 operation.

CONCLUSION: Based on the evaluation discussed above, no corrective action appears necessary other than discussed in previous correspondence on this issue.

Attachment A June 30, 1983 TELEX NUMBER 908142 TELEX TO: NUCLEAR REGULATORY CO!: MISSION ATTN: MR. R. C. DEYOUNG, OFC. OF INSPECTION AND ENFORCE:ENT ITT BARTON IS A SUPPLIER OF BASIC COMPONENTS TO THE 11UCLEAR POWER INDUSTRY. THE SPECIFIC COMPONENTS BEING REPORTED ARE OUR

- SUPPRESSED ZERO MODEL 763 STATIC PRESSURE TRANSMITTERS. .

THE DEFECT EXHIBITS ITSELF AS A NEGATIVE SHIFT IN OUTPUT DURING INITIAL EXPOSURE TO OPERATING PRESSURE. THE IJiOUNT OF THE SHIFT

'IS A ' FUNCTION OF PROCESS PRESSURE AND THE CALIBRATED SPAN OF THE INSTRU:iENT. THIS INFOP3.ATION WAS !EDE AVAILABLE TO ME ON 29 JUNE 1983 AND USERS WILL BE NOTIFIED BY 8 JULY 1983.

ALL SUPPRESSED ZERO MODEL 763 TRANSMI;TERS SHIPPED PRIOR TO THIS DAY tiAY NOT PERFORM TO BARTON'S SPECIFICATIONS. THE CUSTOMER NAME AND QUANTITY OF ALL POTENTIALLY AFFECTED INSTRUMENTS WILL BE IDENTIFIED AND PROVIDED TO YOU BY 8 JULY 1983.

TESTING HAS BEEN INITIATED TO DETERMINE THE SPECIFIC CAUSE OF A THE DEFECT AND IS EXPECTED TO BE CC:1PLETED BY 15 AUGUST THE 1983.

INDIVIDUALS REPORT WILL BE SUB:!ITTED TO YOU AT THAT TII.E.

RESPONSIBLE FOR THE TESTING AND ANALYSIS ARE EDGAR ROMO, MANAGER, INSTRUMENT DE7ELOP!1ENT AND GERALD WELT, DIRECTOR, QUALITY ASSURANCE.

REGARDS, THO:tAS J. SHIDELER GENERAL l'.ANAGER ITT EARTON INSTRUMENTS COMPANY 1.vg 0020SL ,

-_