IR 05000344/1988007

Mgt Meeting Insp Rept 50-344/88-07 on 880201.Major Areas Discussed:Barton Pressure Transmitter Failures

ML20196J872
Person / Time
Site:	Trojan File:Portland General Electric icon.png
Issue date:	02/23/1988
From:	Rebecca Barr, Mendonca M NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V)
To:
Shared Package
ML20196J853	List: IR 05000344/1988007 ML20196J850
References
50-344-88-07, 50-344-88-7, NUDOCS 8803150084
Download: ML20196J872 (5)

Text

.- . .... . . - - _ . . . .. - -

e .'

<

.

U.S. NUCLEAR REGULATORY COMMISSION <

REGION V

Report N /88-07 Docket N ,

License N NPF-1-Licensee: Portland General Electric Company 121 Salmon Street Portland, Oregon 97204 Facility Name: Trojan Nuclear Plant Inspection at: Region V, Walnut Creek, California Report Prepared by: [ b U Rl C. Barr, Senior Resident Inspector Date Signed Approved by:

~

L/F-3 /#I M. M. Mendonca, Chief Date Signed Reactor Projects Section 1 i

Meeting Summary:

Meeting on February 1,1988 (Meeting Report No. 50-344/88-07)_

Scope: A licensee management meeting to review failures of Barton pressure transmitters at the Trojan nuclear facilit Results: A summary of the licensee's history of problems with Barton pressure transmitters, past actions and additional planned corrective actions was presented.

,

8903150084 880224 PDR ADOCK 05000344 O DCD

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

.. .'

.

.

DETAILS Meeting Attendees Licensee Attendees D. W. Cockfield, Vice President, Nuclear T. D. Walt, Manager, Nuclear Safety and Regulation A. N. Roller, Manager, Nuclear Plant Engineering D. W. Swan, Manager, Technical Services D. A. Regoli, Systems Engineer NRC Attendees J. B. Martin, Regional Administrator D. F. Kirsch, Director, Division of Reactor Safety and Projects R. P. Zimmerman, Chief, Reactor Projects Branch R. J. Pate, Chief, Reactor Safety Branch M. M. Mendonca, Chief, Reactor ProjectsSection I J. L. Crews, Senior Reactor Engineer R. C. Barr, Senior Resident Inspector Details of Management Meeting Mr. Martin opened the meeting stating the Region V staff has been closely following the problems the utility has been experiencing with Barton pressure transmitters and was interested in hearing further detail Mr. Walt introduced the licensee participants and stated the meeting objective Mr. Swan, Manager, Technical Services, explained the pressure transmitters design and operation by referring to a diagram and an actual Barton 763 pressure transmitte He also identified the differences between the Barton transmitter models 763A and 76 Mr. Pvan continued the presentation with an overview of plant applications and failure history of the Barton model 763A ar.d 763 pressure transmitter Mr. Swan stated that at Trojan, Barton 763 pressure transmitters were used to measure turbine first stage pressure, main steam line pressure, reactor coolant system (RCS) wide range pressure, accumulator pressure and post-accident sampling system pressure. He noted that only the turbine first stage pressure, main steam line pressure and RCS wide range pressure instruments provide input for sa'fety functions. Mr. Swan stated that all Barton model 763 pressure instruments with the exception of turbine first stage pressure, exhibited a drift that caused the instrument to be out of calibration low and, therefore, with respect to safety, drifted in a conservative manner with the exception of the RCS wide range instrument. He also noted that subsequent to identifying the drift concern, the instruments were recelibrated and have remained in calibratio He stated channel checks are being performed once every shift and one steam line pressure instrument is being calibrated each week to ensure early detection of any future instrument drift ..

., - - - .. . - - .- - -. --, __ -

. .

.

.

Mr. Swan stated significant effort h'ad been expended in attempting to identify the cause of the instrument drift. While the specific cause~of the drift had not yet been' determined, PGE had concluded that shortly after installation, the instrument experienced'an' apparent permanent set (drift). The component or components that cause the permanent set had not been identified. The vendor had been notified and did not necessarily agree with the licensee's conclusion. Mr. Cockfield noted a detailed assessment by the Performance Monitoring / Events Analysis (PM/EA)

organization determined the technicians followed the calibration procedure recommended by the vendors. He stated the only potential question was whether or not the instrument had been exercised (ranged full-scale three times) prior to.the initial calibration after installatio This uncertainty existed because the procedure did not require a sign-off for that step; however, a technician that performed the calibration of one of the instruments clearly recalls performing that ste Therefore, the Trojan staff concluded the instruments had been properly calibrate Mr. Swan stated PGE is developing a test program whose ptepose is to -

specifically identify the cause of the transmitter drif The program will be perforded by an independent laborator Mr. Swan next discussed the Trojan plant experience with Barton Model 763A pressure transmitters. He noted these transmitters have been used to sense only pressurizer pressure. He stated Trojan had experienced five transmitter failures. The failures were characterized by a rapid (2-3 hour), ramped (linear), high off-scale drif He stated that between September 12 and November 22, 1987, pressurizer pressure transmitter PT-455 had failed four times and that on January 16, 1988, pressurizer pressure transmitter PT-456 faile M- Swan next discussed actions taken while attempting to identify the cause of the failures. He stated the pressurizer pressure instrument sensing line configuration had been evaluate The assessment included a change analysis (performed by PM/EA), a walkdown of the sensing lines, replacement of the PT-455 instrument line root valve internals and blowdown of all pressurizer instrument sensing line However, the evaluation identified no conclusive contributing causes for the failure Mr. Swan stated during the blowdown of the PT-455 instrument sensing lines a minor blockage of corrosion products was cleared using 100 psig of air pressur No other sensing lines exhibited corrosion product discharge or blockage. Mr. Kirsch questioned whether the blockage could

-

have contributed to the failur Mr. Cockfield stated that the Trojan staff concluded it was unlikely, because computerized trending of pressurizer pressure during the period of failure appeared to indicate continuous fluid communication from the pressurizer to the transmitte Mr. Swan stated that an analysis by an independent laboratory of a failed transmitter indicated the instrument's Bourdon pressure sensing element had been overstresse The vendor told PGE that the Bourdon tube had failed as a result of overpressurization which required a pressure of approximately 6500 psig. Mr. Swan stated that in an effort to determine if the sensing lines were experiencing random pressure pulses, a Rosemount pressure transmitter and a recorder had been installed in the

,

I

. . . __ . _ _ ___._ _

.. .

.

.

PT-455 instrument sensing line. No pressure pulses were detecte Because the recorder being used was a RMS (root means square) device, only pressure pulses in the range of millisecond could be detecte A recorder that can measure pulses in microseconds is being procured and will be installed to ascertain if high pressure pulses of very short duration occu Mr. Swan stated that outside sources such as the Nuclear Plant Reliability Data System (NPROS) and the Institute of Nuclear Power Operation's "Note Pad" had been accessed to obtain industry history with Barton 763A transmitters. Only Palo Verde and Calvert Cliffs were identified as users of the instrument. Neither facility had experienced failures similar to Trojan transmitter failures, although some cases of transmitter drift had occurre Mr. Swan stated that there had been significant corporate involvement with the investigation of the Barton pressure transmitter He noted the Nuclear Safety and Regulation Department (NSRD) had evaluated the impact on safety for both the Barton 763 instrument drif t and the Barton 763A instrument failure He also noted that the corporate engineering organization had assisted in evaluating the failure Mr. Swan concluded the presentation on the failure of the pressurizer pressure transmitters by stating that there was no impact on safety because the failure were gradual, detectable and allowed operator response prior to an automatic response by the plant protective syste Additionally, he noted the output of the pressurizer pressure instruments was being continuously trended, and that the plant computer performing the trending had its high pressure setpoints lowered from 2300 psig to 2275 psi Mr. Cockfield stated that as a near term solution PGE olans to replace the Barton 763A pressurizer pressure transmitters with 6arton Model 763 pressure transmitter He noted that the 763 transmitters were used previous to the 763A for measuring pressurizer pressure. Mr. Cockfield stated the reason 763's were replaced was because the 763's required calibration at a frequency greater than the vendor specifications. He stated that an acceptable calibration frequency would be determined. He also stated as a long term solution PGE would re-evaluate commercially available transmitters.

.

Mr. Swan stated that two management issues resulted from PGE's

'

investigation of Barton pressure transmitters: the need to perform more timely reviews, and corrective action implementation of instruments found out-of-calibration; and the need to improve parameter verification (i.e.,

checking actual parameter value against expected value) during power  !

escalation. Mr. Cockfield noted actions had been taken or were planned I to resolve these issue Mr. Barr pointed out that the licensee's action plan could have been improved in content and use as well as effectiveness of corrective actions. Mr. Barr also pointed out that licensee management should assure that the recommendations of PM/EA and QA are appropriately considered in the licensee's problem resolution program .-, _ . _ _ _ - , _ - - - - , , - -.-n .,- -- , -

- - - - - - + ~ - - - -

~ -<- ~ -

.. ..

4 j

'

'

..

Mr. Martin concluded the meeting by stating it was important to determine .

the cause of the failure of the pressurizer pressure transmitters and the !

drifting of the Barton 763 pressure transmitter He stated that the Region V staf f will continue to follow these problem l l

l l

l I

i l