ML20212G713

Sor Differential Pressure Switch Diaphragm Failure Investigation Rept

ML20212G713
Person / Time
Site:	LaSalle
Issue date:	01/15/1987
From:	COMMONWEALTH EDISON CO.
To:
Shared Package
ML20212G701	List: ML20212G697 ML20212G713
References
2545K, NUDOCS 8701210145
Download: ML20212G713 (44)

Text

• • ,

ATTACHMENT A S.O.R. DIFFERENTIAL PRESSURE SWITCH DIAPHRAGM FAILURE INVESTIGATION REPORT i

COMMONWEALTH EDISON COMPANY LASALLE COUNTY STATION -- UNITS 1 & 2 JANUARY 15, 1987 4

i 8701210145 870115 KDR ADOCK 05000373 o PDR

SOR DIFFERENTIAL PRESSURE SWITCH DIAPHRAGM FAILURE INVESTIGATION REPORT TABLE OF CONTENTS Topic Page I. Executive Summary 1 II. Initiating Event 3 III. Examination of Diaphragms 4 IV. Comparison of Failure Rates with Those of Other D.P. Devices 8 V. Impact on Long-Term Actions 9 Appendices l

A. Summary of all SOR Differential Pressure Switches Removed from Service or Not Placed into Service A-0 B. CECO System Materials Analysis Dept.

Letter on SOR Differential Pressure Switch Diaphragm Inspection B-0 C. SOR, Inc. Letter on Switch Diaphragm Inspections C-0

.c A

SOR D.P. SWITCH DIAPHRAGM FAILURE INVESTIGATION I. Executive Summary During a December 9, 1986 functional test of a group of LaSalle 2 SOR, Inc. differential pressure (d.p.) switches, one switch failed to maintain a differential pressure.

A CECO maintenance engineer recalled that a similar failure .

occurred at least once before. An investigation was immediately launched to review previous failures.

The failed switch was also examined to determine the cause.

The investigation found that since the first SOR d.p. switch was installed in March, 1985, at LaSalle, a total of four switches had failed (inclusive of the December 9, 1986 ,

discovery) because differential pressure could not be maintained across its high and low pressure sensing ports.

In all four cases, subsequent inspection of the switch's Kapton diaphragm revealed holes or tears which were apparently largo enough to allow process fluid to pass from one side to the other. Additionally, subsequent inspections of other SOR differential pressure switches that had been removed from service for other reasons has revealed three other switch diaphragms with holes. However, for these three switches their initial inspections found no holes. The holes were discovered only upon re-inspection of these diaphragms after the switch parts had either sat unprotected in a bag in the LaSalle hot shop or had been discarded following their first inspections and retrieved from the trash bin at SOR, Inc. It is probable that the holes discovered on these latest three findings were developed during multiple handling of the diaphragms prior to this recent detailed inspection.

i Five diaphragms with holes, a new diaphragm intact in its i switch housing, and a diaphragm that had been pressure tested l by SOR, Inc. to its ultimate destruction were examined by the CECO System Materials Analysis Department (SMAD).

Microscopic analysis revealed that some holes appeared to have been caused by sharp-edged metallic and non-metallic l particles found deposited on the exposed diaphragm surfaces.

Similar debris was also found on the new diaphragm. The switch that had been pressure tested to destruction had a diaphragm rupture that appeared different from the punctures

and holes found on the other failed diaphragms. The results

. of SMAD's examination of these diaphragms and particles are l described in Appendix B of this report.

PAGE 1 l

l

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

~

.f.

y I. Executive Summary (continued)

Eight diaphragus from switches that had been removed from service or not placed into service for various other reasons were re-examined under magnification at SOR, Inc. Pinholes were observed in two of these switches. (These two findings represent two of the three subsequent inspection findings mentioned previously.) The results of SOR, Inc.'s examination are described in Appendix C of this report.

. To date, neither the source of the particles nor the mechanism for puncturing the diaphragm has been determined.

Analysis of the diaphragm perforations is ongoing.

The four diaphragm failures in approximately 171 switch-years of service since installation of SOR differential pressure switches at LaSalle represent a diaphragm failure rate of 2.7 X 10E-6 failures per switch operating-hour. This rate l compares favorably with failure rates of other manufacturer's i differential pressure sensing instruments in service at nuclear power plants throughout the U.S.

An analysis of the length of service for each failed diaphragm revealed that there is no common time interval between switch installation and failure discovery. No other common failure initiation has yet been determined and therefore, CECO considers these failures to be random in occurrence.

This type of failure is easily detectable during either the functional test or calibration. Since all SOR d.p. switches are tested each month at LaSalle, the likelihood of operating with a failed switch beyond one month is very low.

! The sixteen Main Steam Line Break Detection switches on each

] unit are not susceptible to diaphragm failures, since their differential pressure sensing component is a steel piston rather than a diaphragm.

The impact of these failures on the LaSalle long-term action i plan is minimal. However, SOR, Inc. is evaluating the use of .

I thicker diaphragms and is reviewing its manufacturing and materials handling procedures.

l i

I PAGE 2 i

.~ ----,%, ,4- - . . _ ,,.---_.--..,,..,w.---,---,,_-.,---..-..--.-,.-._m------..-,-mm .

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

. _ m . _ . _ . _ __ _ _ _ - .

II. Initiating Event During a surveillance test of the Unit 2 RHR/RCIC Line Break Detection switches on December 9, 1986, one switch, Tag No.

2E31-N013BB (S/N 85-1-2428) did not pass the functional test i because differential pressure between its high and low pressure ports could not be maintained. The switch was removed from service and replaced with a new one. The failure of this switch triggered the memory of a CECO maintenance engineer, who recalled that a previous similar t failure had occurred in July, 1986. The previous event had been treated as a normal random failure and was not thought to be significant during the LaSalle setpoint characterization program which had been centered around setpoint shift and potential common mode bearing failures.

All work requests since January 1, 1985 related to SOR differential pressure switches were then reviewed. A total of six previous switch failures were found. Three failures were related to the inability of the switch to maintain differential pressure. These three were: Tag No. 1E12-N010CA (S/N 85-1-2411), Tag No. 1E12-N010CA (S/N 85-6-3587),

and 1E12-N010CB (S/N 85-1-2413). (In addition, the search of work requests revealed work done on three other switches: Tag

, No. 2B21-NO37AA (S/N 85-1-2457)--microswitch problem, 2B21-NO38A (S/N 85-1-2376)--sluggish calibration response, and 2B21-NO38B (85-1-2378)--corroded bearing).

In addition, other station records such as Licensee Event i Reports (LER's), Deviation Reports (DVR's), Discrepancy Reports (DR's) and test data from the setpoint characterization program were surveyed. (Prior to the issuance of IE Bulletin 86-02, reporting requirements for

[ various types of failures were not as extensive.) Also, INPO l was asked to search the Nuclear Plant Reliability Database Survey (NPRDS) for all records submitted by LaSalle pertaining to SOR differential pressure switches. This I survey confirmed the earlier findings of a total of six previous failures. The results of this survey are summarized in Appendix A, which lists all of the SOR differential pressure switches that have been removed from service or found unacceptable for placement into service.

Since a total of four switch failures had been identified as relating to the inability of the switch to maintain differential pressure across its high and low ports, an investigation of such failures was immediately launched.

PACE 3 t

Fe III. Examination of Diaphragas Prior to the December 9, 1986 switch failure, holes had been observed in the diaphragm of only one switch (1E12-N010CA, S/N 85-6-3587). This switch had failed to maintain differential pressure during a March 31, 1986 calibration, and had been disassembled at LaSalle by a maintenance engineer in May, 1986. A "U"-shaped rip with a. dimple at the bottom was observed at that time, and the switch was then packaged and shipped to SOR for subsequent inspection. Two other switches, (1E12-N010CA, S/N 85-1-2411, and 1E12-N010CB, S/N 85-1-2413) which had similarly failed surveillance testing, had been disassembled and visually inspected in October, 1985 and July, 1986, respectively, resulting in no observances of unusual diaphragm conditions. Switch 1E12-N010CA, S/N 85-1-2411 was contaminated, so it was packaged in a disassembled condition and placed in the LaSalle " hot shop." Switch 1E12-N010CB, S/N 85-1-2413 was packaged in a disassembled condition and sent to SOR, Inc. for subsequent examination. SOR, Inc.-later inspected the switch in August, but did not notify CECO of the results of this inspection as they were at that time concentrating on supporting the LaSalle setpoint characterization testing program, static shift phenomenon, and rusted bearing investigation. This failure was also treated !>y CECO as a random event, typical of differential prassure sensing components.

After December 9, 1986, the diaphragms of the three above described switches were re-examined under magnification.

This re-examination resulted in the observance of holes in all three diaphragms. In addition, holes were discovered in the diaphragm of a switch (Tag No. 2B21-NO38A, S/N 85 2376) that had been removed from service during the setpoint characterization test program because of sluggish calibration response and unusual trip behavior (however, this switch had

always been observed to be able to maintain differential pressure.) This switch, described in the August 1, 1986

" Report of Investigation of Static-O-Ring Differential Pressure Switches", had been visually inspected on July 30, i

1986, and no damage of the diaphragm was noted at that time.

Since the switch was contaminated, it was placed disassembled t

in a bag, and stored in the LaSalle hot shop. It was noted i following the re-inspection of this diaphragm in December, 1986, that the holes could have been inflicted on the diaphragm during the multiple handling it received while packaged in the same bag with the remaining switch parts following the initial disassembly.

PAGE 4 l

III. Examination of Diaphragms (continued)

Following the December 9, 1986 failure of switch 2E31-N013BB, S/N 85-1-2428, it was carried by a CECO engineer to SOR, Inc.

where it was disassembled and inspected. A tear was found in its diaphragm. SOR, Inc. then pressurized the diaphragm of another switch to 2200 psig until it ruptured, to compare the shapes of the tear holes. There was a definite difference in the size and shape of the tears, and it was believed that the failure modes for both tears were different for each case.

From this initial investigation five switches were observed to have holes in their diaphragms. These are:

Tag No. Ser. No. Application Installed Failure Date Discovery 1E12-N010CA 85-1-2411 LPCI Loop C 03/31/85 09/25/85 Min. Flow Valve Control .

1E12-N010CA 85-6-3587 LPCI Loop C 10/01/85 04/01/86 Min. Flow Valve Control 1E12-N010CB 85-1-2413 LPCI Loop C 03/31/85 07/04/86 Sys. Flow Alarm 2E31-N013BB 85-1-2428 RHR/RCIC Stm 04/17/85 12/09/86 Line High Flow 2B21-NO38A 85-1-2376 Rx Lev. 3 ADS 05/16/85 07/03/86 Confirmatory The diaphragms of these five switches were carefully packaged and sent to the CECO System Materials Analysis Department (SMAD) for further analysir,. In addition, SMAD was given an intact new switch (serial number 86-9-2674) to disassemble and compare its diaphragm with the failed ones. This new switch was from the LaSalle instrument storeroom, and had never been put into service. Finally, SMAD was also sent the

, diaphragm which had been pressure tested to destruction by i SOR, Inc, for comparison purposes.

PAGE 5

III. Examination of Diaphragms (continued)

The results of the SMAD analysis are contained in Appendix B.

The switch that had been pressure-tested to destruction had a diaphragm rupture that appeared completely different from the small punctures and tears found on the diaphragms of the failed switches. It appeared as a clean rip of the material at the burst point of the diaphragm, with no jagged edges at the rip locations. The new switch was found to have several small non-metallic particles imbedded in its surface. Two of the five diaphragms with punctures, tears, or dents were also found to have small metallic particles on their surfaces.

At the location of the punctures, a distinct jagged edge was apparent. All locations, sizes and shapes of the perforations are detailed in the Appendix B. The SMAD group was able to create similar holes in a sample of the diaphragm material by squeezing the diaphragm sample and metal fines from their laboratory between two glass plates using finger pressure.

SOR, Inc. had been asked to re-examine the diaphragms of eight switches which had previously been sent to them because they were found unacceptable for use at LaSalle during the setpoint characterization program. (These eight diaphragms had been examined in November, 1986, and then were discarded into the trash bin.) A summary of their re-examination is contained in Appendix C. This re-inspection used a 20X microscope. Two switch diaphragms were observed to have a pinholes--see Appendix C. However, these holes were probably inflicted during handling following the previous inspection. Also, testing of all eight of the switches last November showed that all switches were able to maintain differential pressure. These diaphragms will be l examined further by SMAD.

The source of the various metallic and non-metallic fragments has not yet been determined.

The ages and service applications of the switches found to have failed in service were also examined. This examination showed that the failure intervals for switches with failed diaphragms ranged from six months to 20 months. CECO considers these failures to be random in occurrence.

Although three failures were on switches in identical service applications, (LPCI Loop C flow) it was determined that these failures were coincidental. Since no failures have been l PAGE 6 l

III. Examination of Diaphragms (continued) experienced in any of the switches in LPCI Loops A and B, and since they were installed at the same time, it was also concluded that service time is not a factor in determining the failure initiation. Loop C of the LPCI system is the

'least used of the three LPCI loops at LaSalle. Loop C is run perhaps 1/4 of the time that Loops A and B are run. The most process disturbances that the LPCI Loop C switches see is during monthly surveillances.

When the LPCI Loops are running, the applied differential pressure across the switch is far in excess of the switch's measuring range, however it is well below the switch's rated burst or proof pressure. When the RHR/RCIC line is in service, the differential pressure across the switch is well below the switch's trip point. Therefore it appears that failures are not related to the frequency of operation of the system served by the switches.

1 i

i l

PAGE 7 i

?

De 4

IV. Comparison of Failure Rates with Other D.P. Devices An analysis has been performed of.the approximate failure rate of SOR d.p. switches using the rather large data base of switches that have been in service or scheduled for service at LaSalle. The focus of this analysis was to determine whether the few diaphragm failures found at LaSalle were comparable to'the number of failures of differential pressure sensing components of other manufacturer's instruments.

i Since there have been a total of four diaphragm failures found on switches in service at LaSalle, and plant modification records indicate that there have been approximately 1,500,000 switch hours of operation between the first switch installation in March, 1985 and January 3, 1987, a diaphragm failure rate of about 2.7 X 10E-6 failures per

+

operating hour was calculated.

A study of IaSalle NPRDS data relating to SOR dp switches which have been removed from service for any reason reveals a

4. total of six NPRDS reported failures over this same period of time, yielding a reported failure rate (due to any cause) of

. 4.0 X 10E-6 failures per switch operating hour.

4 A similar review of the NPRDS database for reported failures of another manufacturer's model of differential pressure transmitters was conducted last October. The NPRDS database i indicated that the these transmitters have an average failure

! rate of 15.8 X 10E-6 failures per operating hour. This failure rate is based on 10,035,214 transmitter hours of operation. Failures include physical faults, out of specifications, demand faults, leakage, and abnormal characteristics.

i Finally, it was attempted to compare the above data with

, failure rates for Barton differential pressure switches and l

transmitters. A search of the "Cause of Failure" column for the key words "Barton" and " Bellows" revealed 42 reported failures of Barton devices from 1974 through 1986, 23 of which were outright failures of the bellows assembly. These failures included leakage of the fill fluid, kinking of the bellows to prevent linear motion, and rupture of the bellows.

The total operating hours over which these failures occurred I

could not be determined, however, i

PAGE 8 l

l i

V. Impact on Long-Term Actions On December 22, 1986, CECO submitted a report detailing the long-term action plans regarding the further operation of LaSalle using SOR differential pressure switches. In this report CECO described its plans for maintaining an increased surveillance schedule on all installed switches.

It has been shown that diaphragm failures (in which differential pressure fails to be maintained across the switch's high and low pressure ports) are readily detectable during functional or calibration testing. Since each switch is tested at least once each month, the likelihood of operating with a failed switch for a period longer than one month is quite small. Also, the diaphragm failure rate at LaSalle is comparable to or lower than failure rates of other differential pressure devices in use in nuclear service applications. Since these failures have not been shown to be related to the frequency of operation of the system, or to time in service, these failures are considered to be random in occurrence, and no further changes in the action plan are warranted.

The December 22nd report also described the planned replacement of all reactor water level sensing switches with analog transmitter / trip systems for economic reasons, while keeping the SOR switches in service for the long-term line break detection for the RCIC, RHR/RCIC, RHR Shutdown Cooling, and Main Steam lines, as well as for the ECCS minimum flow valve control functions. The Main Steam Line break detection switches are not affected by the diaphragm failure phenomenon because they make use of a piston rather than a diaphragm design.

I

! As a precautionary measure, however, the methods of handling SOR differential pressure switches between the time they are received at LaSalle and installed will be reviewed with regard to preventing accidental contamination of the switch.

SOR, Inc. informed CECO that they have begun to test prototype switches with diaphragms of different design design characteristics, such as thicker materials. They have also begun evaluating what additional steps, if any, could be taken during the manufacturing process to prevent contamination of the diaphragms with particles.

I h

l PAGE 9

J u 4 e m . m 4 .w. #.- _a A s.-

APPENDIX A 4

SUMMARY

OF ALL SOR DIFFERENTIAL PRESSURE SWITCHES REMOVED FROM SERVICE OR NOT PLACED INTO SERVICE i

s A-0 r

1 i

List of SOR DP Switches Removed from Service or Not Placed into Service LaSalle County Station - Units 1 and 2 l

I REASON FOR REMOVAL IN SERVICE DISCOVERY REPORTING REPORTING TAG # SERIAL # OR REJECTION DATE 3 DATE DATE MEDIUM 1B21-N024B 85-1-2369 Rejected during character- N/A 7/22/86 12/22/86 Letter 2 ization program due to offset I

1B21-N024D 85-1-2373 Rejected during character- N/A 7/24/86 12/22/86 Letter 2 ization program due to offset 1B21-NO38A 85-1-2375 Rejected during character- 3/23/85 7/30/86 12/22/86 Letter 2 ization program due to offset SPARE 1 86-6-2508 Rejected during character- N/A 8/27/86 12/22/86 Letter 2 ization program due to offset SPARE 2 86-6-2513 SOR Factory Test exceeded N/A 8/27/86 12/22/86 Letter 2 repeatability criteria IE31-N007AA 85-4-260 Rejected during characteri- N/A 8/11/86 12/22/86 Letter 2 zation program due to repeatability 1E31-N013BB 85-1-2427 Rejected during characteri- N/A 8/16/86 12/22/86 Letter 1 zation program due to 12/22/86 NPRDS repeatability 1E31-N012BB 85-1-2435 Rejected during characteri- 3/29/85 8/20/86 12/22/86 Letter 2 zation program due to drift l

! 1E12-N010CA 85-1-2411 Switch would not trip on 3/31/85 9/25/85 2/10/86 NPRDS

station functional test 12/22/86 Letter 2 l 1E12-N010CA 85-6-3587 Pressure would not equalize 10/1/85 4/1/86 6/23/86 NPRDs 4 on station calibration 12/22/86 Letter 2 3 5/28/86 DVR 1-1-86-039 Page A-1 7061L

T J

List of SOR DP Switche3 Removed from SIrvica or Not Pltc2d into Servics ,

LaSalle County Station - Units 1 and 2 REASON FOR REMOVAL IN SERVICE DISCOVERY REPORTING REPORTING TAG'# SERIAL 4 OR REJECTION DATE 3 DATE DATE MEDIUM IE12-N010CB 85-1-2413 Pressure would not equalize 3/31/85 7/04/86 10/23/86 NPRDs in Station Calibration 12/22/86 Letter 2 Spare 3 86-6-2511 Rejected during charactor- N/A 9/02/86 Not reported ization program 2B21-N024A 85-1-2368 Rejected during character- 4/06/85 7/7/86 8/1/86 Report 1 ization program due to offset 12/22/86 Letter 2 and erratic microswitch behavior 2B21-N024B 85-1-2370 Rejected during character- 4/25/85 6/23/86 8/1/86 Report 1 ization program due to offset 12/22/86 Letter 2 and erratic microswitch behavior 7/01/86 DVR l-2-86-037 2B21-N024C 85-1-2372 Rejected during character- 5/03/85 7/06/86 8/1/86 Report 1 ization program due to offset 12/22/86 Letter 2

2B21-NO38A 85-1-2376 Rejected during character- 5/16/85 7/03/86 8/1/86 Report 1 ization program due to 12/22/86 Letter '

unusual cal. data & pressure 8/21/86 NPRDs relax behavior 2B21-N038A 85-1-2395 Exceeded reject limit in 2- 7/30/86 11/24/86 12/22/86 Letter 2 mo. station calibration 1/15/P.7 LER 374-86-019-00 1/15/87 DVR l-2-86-069 2B21-NO38A 85-1-2398 Performance in level drop 11/26/86 1/3/87 1/3/87 Courtesy call test was outside of expected to NRC range Page A-2 7061L

s List of SOR DP Switches Removed from Service or Not Placed into Service LaSalle County Station - Units 1 and 2 REASON FOR REMOVAL IK SERVICE DISCOVERY REPORTING REPORTING TAG # SERIAL # OR REJECTION DATE 3 DATE DATE MEDIUM 2021-NO38B 85-1-2378 Excessively hi trip value in 4/18/85 6/17/86 6/86 Call to NRC calibration check 8/1/86 Report 1 j

d 12/22/86 Letter 2 7/1/86 LER 374-86-011-00

7/1/86 DVR 1-2-86-037 11/21/86 LER 374-86-011-01 i 8/19/86 NPRDS i

2B21-NO38B (Not available) Rejected during character- N/A 7/12/86 Not reported ization program due to offset 2E31-N013BB 85-1-2428 Would not develop DP in 4/17/85 12/09/86 12/22/86 Letter 2 functional test 1/08/87 LER-374-86-018-00 1/08/87 DVR-1-2-86-073 2B21-NO37AA 85-1-2457 Microswitch had wide dead 5/21/85 5/30/85 12/22/86 Letter 2 band 4

NOTE 1 " Report of Investigation of Static-O-Ring Differential Pressure Switches" submitted to NRC August 1, 1986.

NOTE 2 Letter addressed to H.R. Denton (NRC) from B.L. Thomas (CECO) dated 12/22/86 NOTE 3 In-Service date refers to date the switch was first put in contact with process fluid following installation.

I Page A-3 4 7061L

4 1 ___J,.

2__ - _m a 1-APPENDIX B CECO SYSTEM MATERIALS ANALYSIS DEPARTMENT LETTER ON SOR DIFFERENTIAL PRESSURE SWITCH KAPTON DIAPHRAGM INSPECTION 4

l B-0 I

i l

January 14, 1997 To: J. s. Abel subject: Kapton Diaphragas - Lasalle co. station seven (7) som switch diaphragas from Lasalle Co. station were ox00ined in the laboratory at Tech. Center. Four diaphragas had failed in service. One diaphragm was removed from a switch prcviously in service, after the switch was found unacceptable due to it exceeding the switch static shift criteria. one diaphragm had boca pressure tested to failure at son. One diaphragm was removed fcco a new (unused) switch (received from LaSalle Co. Station) which was disassembled and examined in the laboratory.

Optical examination at up to 50 diameters magnification revealed '

nunorous defects and debris on the five diaphragas removed from corvice. The five diaphragas which had seen service contained dofcots classified as punctures, tears, or dents. These defects varied in size from approximately 0.23 inch to less than 0.005 itch. The diaphraga that had been pressure tested to failure ocntained a tear approtiaately 0.67 inch in length. The shapes and siscs of the defects are listed in the attached table. Photographs of typical defects and debris are shown in Figures 1 through 14.

For reference, a photograph of a scale at 13X magnification is also included.

The debris found on two of the Kapton diaphragas which had seen service consisted of loose (foreign) particles on the surface. Many of the loose particles observed exhibited a metallic luster. Bone looce particles were silver in color with hues of blue, similar in app;arance to machining chips. Other loose particles were copper colored. The metallic appearing particles were generally sharp and cEgular in appearance. Their size varied from less than 0.005 inch to at least 0.030 inch. Loose, amorphous, globular debris, probably chtvings from the o-ring material were observed along with non-metallic particles imbedded into the new diaphraga. A fine, bicek deposit was observed on one diaphragm which had seen service.

A new switch was disassembled in the laboratory, the internal surfaces and diaphragm rinsed with alcohol and the run off was ccptured on filter Japer. Very small metallic app 9aring and non-tetallic particles were found upon optical examination of the filtor paper.

ntw diaphragm.

In addition, a light oil or grease was found on the optical examination of the parts leading to the dicphraga chamber revealed small metal fines adhering to some throces and the thru hole.

-oo Several of the metallic and amorphous particles were gun 11tatively analysed with the energy dispersive x-ray spectrometer (EDs) attachment to the scanning electron microscope (SEN). The s nalysis showed the silver colored particles collected on filter popar to be similar in composition to an austenitic stainless steel.

probably type 316. A copper colored particle analysed revealed it l

to be virtually pure copper with traces of silicon, potassium and irca. The amorphous material consists of approximately 50s silicon, 24% iron, 44 chlorine, 15% chromium, and at potassium. The -

particles found imbedded into the diaphragm from the new (unused) switch were primarily silicon.

A test was performed with one Kapton diaphragm. The Kapton diaphran was placed on a glass slide. Very fine metallic machining dobris was placed on the edge of the Kapton diaphragm and cavered i

cith a second glass slide. A pressure was then applied by squeezing tho glass slides between fingers. This resulted in punctures of the i

K0pton material which 49peared very similar to those observed in the failed diaphrama.

Based on the examination of the five failed diaphragas and one ncw diaphraga, we believe that punctures in the Kapton diaphraga material were initiated by very small metallic and/or glassy (silicon) type particles. The particles may have come from the switch body during either assembly of the switch or its testing / installation. Ultrasonic and solvent cleaning of all the

, parts and careful examination of the diaphragm assembly under a low pcwer microscope for presence of metallic particles may help in reducing the foreign object damage to the diaphraga material.

Prepared by: - M v ,.g. ..rs.n Approved byt .

7' i @. s**

M r M. Gaitonde C52/rp 00: K. L. Graesser/E. E. Spitanet J. C. Renwick 0262P r

SUIstARY OF FLAME OSSBaVED DURIsc LAa0RATORY EIAJIIstAi105

< OF IAFT05 DIAPHRAcaIS Fa0M LASALLE CO. STATION serial # Location

• surfacel Flaw Flaw Flast Flaw

& circled Type 2 sise3 Orientation 4 shape 5 comuments Area F 5 D F S N L C R DLR0CT 85-1-2376 1 3 I I I I I 2 3 I I I I I 3 3 I I I I I 85-1-2413 1 3 I I I - - - I i 1 3 I I I I I 3 3/4 I I I I I 2 Flawa 4 3 I I I I I 5 3 I I I I I 6 3 I I I - - - I

$5-1-2428 1 4/3 I I I I I I I

$5-6-3587 1 4/3 - - I I I I I

2 4/3 - - I I I I l 3 3 I - - I - - - I 4-11 4/3 - - I I I I 85-1-2411 4/3 I I I I I f 1 I I i

2 4/3 I I I I

! 3 4/3 I I I - - - - - - - -

4 4/3 I I I - - - I 2 Notee <

5 4 - - I I - - - I 2 Holes 6 3 - - I I - - - I 2 Moles 7 3 - - I I - - - I 4 Holes 8 3 - - I I - - - 3 1 Hole 9 3 - - I I - - - 1 2 Holes l 1 Hole 10 4/3 - - I I - - - I 84-9-2674 (unused) 1 4/3 I .. - - - - - - - - - - - - 0-ring material on surface and non-metallic in-bedded in surface 2 3 I - - - - - - - - - - - - - non-metallie in-bedded in surface no serial number (pressure l tested ct son) 1 4/3 3 - - - - I I I 02577

/

KEY Location * (radial location of observed flaws) 1 = bolt hole area 2 - O-ring seal area 3 = flexure area 4 = adjacent to edge of piston plate 5 - captured between piston plate Surfacel F = front (labelled)

B = back Type 2 D = dent P = puncture Size 3 S = small M = medium L = large Orientation 4 i C = circumferential R = radial D = diagonal Shapes L= linear R= round O = oval C = crescent T= tongue l

l l

l I

1 r

i

/

• e LASALLE COUNTY STATION SOR SWITCH DIAPHRAGMS Switch No. lE12-N0lOCB (Writ L59561)

Serial No. 85-1-2413 j g :, y. [g-tv fy* ? Uhl) .

..w;y

~

[!.a o

-i . . ;; , c

-y .-  ;,

2 .

_].yz;);y 'h yl ,h{-

1 -

_e i. I

~t: M

• 3 u

3;ty '

t )j.

Sr: s

., y. .

- n~

2 ' .

,s:  : i .'

s

~3 %y

?

D

<ti.

T 3m;, . e

• -#s .'.

i

! FIGURE 1 i

l Overview of Kapton diaphragm. Circled areas highlight defects observed during laboratory examinations.

i

4 LASALLE COUNTY STATION SOR SWITCH DIAPHRAGMS Switch No. lE12-NO10CB (WR# L59561)

Serial No. 85-1-2413

~.7-s- r ;; .- ,,

_rg ump!y.-- -  ; ; - .- - .

.re. e4 .

. s sa ' ' 4 *:

' }* Ng *

, j'e4. lE.' - -.,f'W:.y*Q..'.,p,:*)s'e'\e

., s? h '

_ :.. .,e . . . .

.s....- .. r .

y, , ,: p ,- a.

3g

. < * ;*\.;,,... g .' .s

• ' :.g ~ W;i,C- i ; E ,: . ::/

.. g 77:. w}q , r;. p. y.k %,'q:.

.?

p..t -1 '.

u .

. . . ,.- C ,. -

l,':y,*. ' *: .

['

lih 5&. 'f,

....r.. ' . -:. . ' _,' $ l~',h LY:; 5 .. . . . ,

g. . .i.. ***. % ',', Q.W. ;. . * ':4.6 %:q:.{.\.g:. .

V". ,.;,<

+ .. - .* s. x.: .. ..;.,t

$n.p

s : y- ,s .w,..v .

i^  :. :Y ' ;, , > . s.] - f h. .' $ng, .U, 'Y .. er ,A f

, Le t

.x . . .

. y . . qa 1:. y ,q .

'Y m y?f: w*h y .; .#* ..,

Vfi..

,&......n 9, s$s.x .

m . W-

,.s., m .:

. c.

. . fw:- _

'..>.? S.Q u- .. ..( y3;., W. 7 I. jw .p i.,.y .

9f -i W p y,

?y q -

.: . .e . .y ,,

..i*. 't ~ ..? ,,

.a , ,

..g',~.s

.. f&.

'; s. .,. r.,,J,,

i?i d...g V't, , ....f. ..

• .&.~,'x gz'[6f'.4.f

..-  ?$

, ,,i " , ,. .s. . ~ ,, 4

. . e\ ,1q.f* ch., . .

y

,. s ,;-

n. . g.

Figure 2 magnification 13x Area 3 from Figure 1 showing two tongue shaped tears in the Kapton diaphragm. These flaws are located adjacent to the outer edge of the piston plate.

1

(

l I

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

1 1

s l

• j LAGALLE COUNTY STATION SOR-SWITCH DIAPHRAGMG Switch No. lE12-N010CB (WRM L59561)

Serial No. 85-1-2413 1

l r..

p >,; ,  ;

u + ;s.a .

%. . 4

~ u .

e,..,-..,..<- '

3 2,, ; ,, . f .g .-w ..- ,- - .

.?  :- .:.

./ }, , .. . . . ,.s, . ;-...h.. _.

. . -s

.- ,, .4 .s 3y

.s,

.f .. .

.=s /..:.

. ,,. . .?.. .. . . , , .. . .

y... .. A .; .<,q . . - e.. ,.

'.k;,n*p, e:%:.N .* :., . .'. .:.i ..a N .o L

~v p i ,. <$,*'.

.sn'-

..A ,. . . .

d.  ::. '.: . .%'.8$ . :s,Ef. C.

• ~

~ '

.w~yy. . .

.,e $<: '. * '

. , ,::. . , +;*;'3'? f"y.75 ,h,0{&p 2. : .::MM i)g $yr;, ".

' *. .n

_ , ?, .. .c y.-< .. .-

4. ;,,.,.:..,,m.*+ .y,y p .c # *j

, . .% .4.. e . w ; .,a

..~4.-

q h , . t.,, . g, a. ,.

. =g . . , .

>g c.

wm.,e. ,9;, w,}<

,4.- @. . 7. n , s;+%.,4%.+._ y*

tt .. . . t. .

W .-

g% tNg: p 4W'

.. .g :~.M%&q

_c .

+ . . . -

fit swy

-~

, .. ds 4,  : W .sp.tgf

, lt Y

...-T.

~

2

. .1;,

~ .. n,j v.:

q.; 4 ; .z.h 1 v.

.: . ,&p2I:yj@,k.s:"!.1%p;lG:

. .; >s,.w.

-%.w J ~+w mW q z

-a .

-m .

ay ' A,

..- m .: yi.'.'bi!Wj$$w'5 m s ,,L h.iE?ufIf<e'$e$ Q e m . tY f'Tff:

Figure 3 magnification 13x l

Area 5 from Figure 1 showing a puncture type perforation of the Kapton diaphragm. The location of this flaw is similar to that described in Figure 2.

I LASALLE COUNTY STATION SOR SWITCH DIAPHRAGMS Switch No. lE12-N010CA (WRM L57245)

Serial No. 85-6-3587 r

M,y k.sh. , ,

• q P~ , .I ny* WY-

$._Q sg,ygd.

(N h;,

Uf, l{K f$

,f "g

= 'n:

32' y.[-$ **k

~

2

[

i; . ?l,';:<. . l%

51% _

' ;/ @ '

W

, h, . F h'* *'

f

f

4 g y. ?

3

,u, n$e;,

h h(>V w= w  %  : s

..y

. . -g : 3 .;: c #

~

$1 ..,f'W- " ' '

%-:: ) ~ d.. w,- 3 , / -

Figure 4 l

Overview of Kapton diaphragm. Circled areas highlight defects observed during laboratory examinations.

l l

(

l 1

I i

I l

l t

J LASALLE COUNTY STATION SOR SWITCH DIAPHRAGMS Switch No. lE12-N010CA (WRM L57245)

Serial No. 85-6-3587 y . ...

n ., . ., .

'I . y .. -'

t .

,-i ) '; _

.:.m'

. l-

,,09 ' p-4.:&. , .s-I). -

e.,,.

s .

. . . ~, ..: . .

,h,

. - f ,"l , _ .  ? .N . ,$

gc ' l.ql,ll$ 'N

. , ,; , i . .

,;~..

v o-

.s

,.% M yh ? ty* * ' W '* *L - WWWOW-i Figure 5 magnification 13x Area 1 from Figure 4 showing a tongue shaped tear in the Kapton diaphragm. The tear appears to have initiated from an area of mechanical damage (arrow) which locally gouged and thinned the diaphragm. This flaw is located adjacent to the outer edge of l the piston plate.

e LASALLE COUNTY STATION SOR SWITCH DIAPHRAGM Switch No. lE12-N010CA (WRML52226)

Serial No. 85-1-2411

, +

rx

• - hi .

f .j, i: ' '

p!L -- . . , --- -' '

M ,;e, p.g),s;y

&[e'. 2.:

• . Q:: ' a-

,,;Yp{_-.

a-

v;--

g .,. ;

l' - f.,7X . .}f O,.- 4-l'.g f _--

<<*.*e -y <

y _p,i: t;;p; n 9 Q. .5~

khhbf

~

• -m

.v q ,. - ut

.h(::n

. ;k g, .;.g g,y s

ap:1{_ -

lj,_gr{k?

v

-lw);, g** Q

..z.., ).{y.

?y- _

1 '1n;ylgC:inR g j Q bn. ' ,,. w: :;, 3 : ,0 & ~

' : " ' 'f {,ly-Q ..% s '?p -  ?

~ ww . , , .c'4- G,-g,

s. .

,;q -

n -

g- r r-f 4 , 4  ;

' i t p, ,-

8f.y  ;

, ,j-

,, s Figure 6 overview of Kapton diaphragm. Circled areas highlight defects observed during laboratory examinations.

r LASALLE COUNTY STATION SOR SWITCH DIAPHRAGM Switch No. lE12-N010CA (Writ L52226) t I

i Serial No. 85-1-2411 i

I L

l a +:s ,.;,lpt. o2..

-:? ^.~.. . .;:.r *

.' S -

'(;df$.y.f. . g[t'f G @-i * ~ ?

4..+.. ..

' , . _ . '** j { ,%* s; ,_* . _ , ,..,

<t.' , ,

i

.- O, .a, . ',: u.if{' .~2y-? :-~ :.L.;76 y . g Y'

'~

A; t Y &t . ,.#.K': .7 ::y, w,

• G.N

& ~i .l p ;,I**= 'j;;.h 'Y...>.n, w:-> :- fAyR\ Tf-Q

-. '.~.,sc *,5[ * .wi b. p'.

?sor l p .v :  : ~g' . ;, '.rT U -

l <. - ui l

4 1

• on.d ,. , : . . -

• F.' pans'

.: di y ,c

• $5?n.Yz,!.

k :,J') q;h ,k;; & .f] ??' i a _ -. - ' >.

Figure 7 magnification 13x Area 1 from Figure 6 showing a crescent shaped puncture in the i Kapton diaphragm. This flaw is located adjacent to the outer edge of the piston plate.

i l

l l

l l

l 1

I

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

. r i

LASALLE COUNTY STATION SOR SWITCH DIAPHRI.UM i

Switch No. lE12-N010CA (WRM L52226)

Serial No. 85-1-2411 i

4

..7 ' , l

'u

• 9. 33 ..; '. . , ,

. t

} *

., ) # .

9=I

• I '

' , ,7 . .. .

• E

, '~$hW'O

.\ , yh , - f, -

Y."'~*

p, . . . ..'

i . . _ M *., $ , ';

j ,,',.',.', b .-

I,* l .

',n )( !j ' ' '

.J3 ,, t

. .s4

, t

_,..,a* -

Figure 8 magnification 13x Area 5 from Figure 6 showing several punctures in the Kapton diaphragm. These flaws are located adjacent to the outer edge of the piston plato.

i I

o LASALLE COUNTY STATION SOR SWITCH DIAPHRAGM Switch No. lE12-N010CA (WRM L52226)

Serial No. 85-1-2411 4

i

.I 4

,..<. " ~ - .,- g :. 'i **.g.: :'. .u - .

j .

s.

.. -:fs:..,,; . < . . , .

p , ' .. ...- ~ .

.,..s s.

i

.- ,.....g . . . -: .g -.,.,4-.

. . ~- -

/

_ ."2,e,w.

, g . .g* *s r. y.*

. 's

. . ,g, ,*;** ,

/ , +f,e* . -

i . r c. W 7 . , . ,- '1, . % ,~' .T ').s + ';: Jt

, . ,*.,.,' i wr

-;f I ,

%e

,....-.a,

.

• w!q', qJeyp* .!' g

.k. * . .**'. 3.9 s

...e.

\

.y yys.o ,. >.-. . .

~, ,-,

., ,- > c. t , ,- .

g,

. . . .e . , .

.... .A.

.....,.r.,..

9 +. u

'g'

,. e .% . T*g . 4 e' g.,

. .Q'V V, ' ..*, af' ' e - ?'fY.

' =*

n ,  :, 3

'hsa .'( ;..y#'Qr * .;x;

~ -

l ,, . Le*. . = , , -' ' . ,.

u . c{ .,e

, <Y

.. - j. . _ _j i

C'

,, . .._ ~ y/c.4. ?.. . ./ l

,.yd' A '

. ::O_'

., .. a.?it%;f 4a .' 's f$

Figure 9 magnification 13x Area 7 from Figure 6 showing numerous small and one larger puncture in the Kapton diaphragm. These flaws are located between the piston plate and the 0-ring.

. LASALLE COUNTY STATION SOR SWITCH DIAPHRAGM Switch No. (NEW) (WRM None)

Serial No. 86-9-2674 P.O. #307922 - Item 4, CE Item 767D03 n ~ ; .x :.ng , ,'Ay,n'#::

. jp. ,

i ". v .q, a . ' - ,

{

4 /**

,; RT.. , ,

4

.: , - ,;. n; & .,

, ' L=

,.v' ,~ :; . . , .

4  %/ .;f 4'1 i G:S 0; $lf&kel~~'

, - $f," h l

l

( ." .

d

.Mg

%@p[.1 g%t - wc 9

i vuf.

]GL'; j~

n;

' ', .( 2

';: d, Figure 10 Overview of partially disassembled switch showing the switch body, piston plates (N-1-B, N-2-A), and Kapton diaphragm (labelled NEW). The circled areas on the NEW diaphragm highlight debris. .

1-LASALLE COUNTY STATION SOR SWITCH DIAPHRAGM Switch No. (NEW) (WRW None) -

Serial No. 86-9-2674, P.O. #307922 - Ite' 4, CE Item 767DO3

-- w ..,

^

[i ji[I':s ci; V

,[, R{.,M_

,4 4;y 7;y l am I >

Q _T!;'4-ih r:-

'ge'.

' k~.

~_

p 2;3lM t>gkyn (1 ( '_3 A

3

, l.

~

> W;M.,

. Oc

,,Wl;c.c }jf r

- -  ! - o A

. - .t ?q',cp a .

.. qQyl [{pj

=

V li:f. ,,: ' f.w 6

>2

hln); ,. 1

, s ea.

3 -

+r, n ;. . %-;&p w;

-i '

'N .  %% _

l )V

= * -

jl. t..{. %$$k @,y '

_T s;Q;i sq*'D Figure 11 Overview of Kapton diaphragm. Circled areas highlight debris observed during laboratory examination following switch disassembly.

i i

l i

LASALLE COUNTY STATION SOR SWITCH DIAPHRAGM I

j Switch No. (NEW) (WP.M None) i l Serial No. 86-9-2674, P.O. #307922 - Item 4, CE Item 767D03 i l

l

! r i

+

j 1 .

e p' -

.l b :. ; g

> e c '., -

. =

g i

. . . . . ~ . ~.- .*

g .

, , . .s .

l

'~

M,.

~

l . l

.,.y, .  ;

I l

i i

Figure 12 magnification 48x Area 1 from Figure 11 showing transfer of the O-ring material (gray globular) to the Kapton diaphragm and two metallic appearing i particles imbedded into the diaphragm (arrows). The debris is located adjacent to the O-ring groove.

l I

l

LASALLE COUNTY STATION SOR SWITCH DIAPHRAGMS Switch No. (NEW) (WRM None)

Serial No. 86-9-2674, P.O. #307922 - Item 4, C.E. Item 767D03

(

p - @

, S.,-rfS 4 -

- - . (

-Ql$',lA'~~ h ~. $ '

'j

.o

.~

g  ;

I

.j Figure 13 magnification 13x Area 2 from Figure 11 showing a metallic appearing particle imbedded into the NEW Kapton diaphragm. This particle is located adjacent to the 0-ring groove.

a

.e LASALLE COUNTY STATION SOR SWITCH DIAPHRAGMS Switch No. (NEW) (WRM None)

Serial No. 86-9-2674, P.O. M307922 - Item 4 CE Item 767D03

. - - ,~.7. , . 7 ,; , . .q ,. .- .

m ._ y .

,,y .ys. , <;;

f __ _

t ;

et>, ..t, <u v.

l e !J .$.,s, *'e' -V Q 6- #6*

} j ~*g _ k k k,g, .i.<M' , . ;;j :;lc:'.-(.< N ;;; j ' y

.p.iplch;g

• 4Ng df/fg g f_,

,. -; L oJ4p n_ c ;@'!!J,

. g.:1 , y3 '._ ): % )rf: *;, ~, 'h;:r GP)

. ,<. , L y]h;c yy*.;A.&vV ;r,'

Q 3%y4 4 L z f

: f K k v ' ,,

. ' g ;f . T;e cet; . .ta y 3- <r !y

t

i <b 1 ? y?.h;. a ,.4 V .' ' ,.1% ).. :-* i: p ,

&t < , 'f,y : ;{l

};m!-lf d: e , j 2f<  ; [, . !.;,.' . . Q i, .e y G l

- r.  ;:s : a

n.  ;;u p.

gg . g

. _v.

~. JV' %f My.& m

.yuw m }::Q i % , ,. , 7 u e O A;j , ' - V::ppgnsy;p v Mt. 7 s 1:,  %'y,uy , .

s.>

Y lllRtQ.?&?}&Qj?1kll;;%g.;;pe $; ~ ' ' :;

• Q4 p,gg t .Wo -Lm:t.",.. A y ; /d[#V :M gLr TF j

~

M V sn v~

, f* .tb

• }d'^,'t.Ryt jf Q'y'

• W4 ,

1r"t

+Ff',b.M r p,3;6.n; 9m- 10% p g, a

f~s .; ja

,[/N$,

',, u .:.N,h86x m d y %.'fg ..

A s

J, .* J,f

~_ : , m :M }x m, vMer

.,9 'r G'Y f W eL:nr nj, sp  ;

j$n 4 e

q: ..

t- ..g a 2:.. ! .9 ( ,, &s.r:tw a M. t m n:gg .

, r. ry t

y;gy .ggg g .

gia Figure 14 magnification 30x Typical debris collected from the diaphragm / piston plate assembly of the NEW switch. The debris was collected by rinsing the assembly with alcohol and filtering the run-off. The particle

~

l labelled by arrow A appeared copper colored, arrow B marks a particle which appeared non-metallic and arrow C marks a metallic (silver) appearing particle partially covered by a non-metallic globular substance. The two parti.cles between arrows A and B also appeared metallic (silver).

. g.

~( )7777

~

. )$X 3 'h,,

gi '

I DIV = 0005' I!!!llllllllllllllll!!!!fllllll lllllhllllllllll!!Ilf!IllfllIlllll

.3

.4 -

.2

[

I DIV.=0lmm 7 8 9 5 6

' J, 4 , _ .

i i

~_

e l

1 1

APPENDIX C

' INC. LETTER ON SWITCH DIAPHRAGM INSPECTIONS C-O s

I

-. .. . = .

r ,

y,

~

(

A a con. 'NC January 14, 1987 Mr. Everett Spitzner Maintenance Staff Assistant COMMONWEALTH EDISON COMPANY PO Box 767, Room 1230 Edison Chicago, Illinois 60690

Dear Mr. Spitznce:

Wo have reviewed the report from the Commonwealth Edison Technical Center of 12/31/86 and have the following comments:

Switch No.1E12-N010CB S/N 85-1-2413 Fig. 2 The tongue-shaped tear at the right of the photograph is similar to a tear noted in a diaphragm cycled to failure on 12/7/82. The diaphragm was cycled with water pressure from zero to 1500 psi on the hi-pressure side with a constant air pressure of approximately 100 psi on the lo-pressure side. The diaphragm failed at 1560 cycles.

The tongue shaped tear at the left of Figure 2 appears to have been started by a puncture.

Switch,tjo.1E12-N010CB S/N 85-1-2413 Fig. 3 We agree, this appedrb to be a puncture.

Switch No.1E12-N010CA S/N 85-6-3587 Fig. 5 We agree, the tear seems to have been started by mechanical damage.

Switch No.1E12-N010CA S/N 85-1-2411 Fig. 7 Because of its crescent shape, this puncture may have been caused by a metal chip.

Switch No.1 E12-N010C A S/ N 85 241_1_ Fig. 8 These two punctures appear to have been made by round objects.

Switch No.1E12-N010CA S/N 85 _1-2411 Fig. 9 These punctures also appear to have been made by round objects. '

Ro. BJ 591,1170L BLAC,KDUD ROAD. CLATHL, AANSAS 60061,913J04-2530. THE GREATW M ANSAS CITW ASEA CABLE SOR INC. TELEX 47 8133 0*.* ter 9p e," ' nus CAY AC4. OEVsCrS LNTHWAT:DNAL TILU NUMSG 499 2783 SD51 Ut

m.

se New Switch S/N 86-9-2674 Fig. 12 Some transfer of o-ring material is not uncommon. The origin of the metallic particles is unknown.

New Switch S/N 86-9-2674 l'ig. 13 ,

Again, the origin of the metallic particle is unknown.

New Switch S/N 86-9-2674 Fig. 14 The origin of the debris is unknown.

Per your request, we have examined the diaphragms removed from the  ;

following returned switches in September,1986:

Serial Number Tag Number Model Number _

86-6-2513 Spare 103AS-B212 85-1-2369 1821-N024B 103AS-B212 85-1-2373 1821-N024D 103AS-8212 85-1-2375 1 B21-NO38A 103AS-B212 85-4-260 1E31-N007AA 103AS-D203 85-1-2427 1 E31-N013BB 103AS-B203 85-1-2435 1 E31-N012BD 103AS-8203 86-6-2508 Spare 103 AS-8212 The diaphragms were inspected with a 20 p1 wor stereo microscope, utilizing the base light to accentuate perforations.

One small perforation was found in two of the diaphragms. The perforations measured approximately .0025 and .000 inches in diameter and were located over halves.

the large o-rings that seal the hi-pressure and io-pressure body Evidently, the o-rings were sealing the perforations since the switches did not leak when received. The above diaphragm with the .0025 diameter perforation also had an oval-shaped perforation (approximately

.002 x .005) in an area protected by the piston plates. Since the eight diaphragms were retrieved from the scrap bin, it is also possible that diaphragms were damaged after the switches were disassembled.

We are investigating the possibility of replacing the 2-mil Kapton used in current production with 5-mil Kepton. This investigation has begun and will include overpressure cycling, dead band, hysteresis, static influence and -

24-hour stability testing.

I We are reviewinging our manufacturing processes for possible improvements

and would welcome any suggestions from Commonwealth Edison.

[

e

---

e - -- -- -

- - . . . . . . - , , . - . . . . , . , ...,,,,,a, , , - - , _ . . ,-..,-e-. , . . . , - - - - - - - - - - - , - - . - - - - ,

e.

Since all internal wetted metal parts of the SOR Model 103 Delta-P Switch are ultrasonically cleaned (both commercial and nuclear), we are at a loss to explain the presence of the metallic particles. ..

Regards, Way Priest, Chief Engineer New Product Design -

WP:es 010987-01 i

f i

I I

1 l

l l

l 1 -

1 l

t

e g ATTACHMENT B i

DIAPHRMRI INVESTIGATION CHROWO 5/31/85

• Work completed on replacement of 2B21 NO37AA. Ser# 85-1-2457

- Microswitch had wide dead band during operation

- WR# L48894

• No evidence of diaphragm damage found during later inspection at LSCS by Ceco personnel on 12/18/86

- no dents, gouges or observable damage to diaphragm l

10/01/85

• Work completed on replacement of IE12 N010CA. Ser# 85-1-2411

- WR# L52226

• Switch was contaminated. It was placed, disassembled, in bag and left in LSCS " hot shop"

- Originally thought switch failed as a typical failure of a mechaniemi device. The switch had been in service only for a short time

• Reported in NPRDS

• Switch was subsequently inspected at LSCS on 12/18/86.

- Numerous punctures were found in the diaphragm and black

, residue found in cavity

[

4/12/86

• Work completed on replacement of IE12 N010CA. Ser #85-6-3587

- WR# L57245

• Switch was disassembled at LSCS between April 11 and May 19, 1986

- tear was identified as failure mode

- "U" shaped rip with a dimple at bottom

• Reported in NPRDS '

• Switch was shipped to SOP Inc to determine why switch failed in

- that application (second failure of switch in that location)

• A Ceco engineer saw failed diaphragm at SOR Inc. on 12/16/86, but believed it to be the diaphragm of IE12 N010CB (which he was aware had failed in July 1986) l 6/01/86 Level transient with failure to trip at proper level.

t 4

++--m- - me, am m e,-w- , - - - - -e,,+--- x , ms -em ,e-,,e- e, , - , ~ + , - +~~---m-ww-r-.e,,,- nr-----, - - ,---e+ gey-ym, ~ , + - g --ew w--vv,-

e o

7/17/86

• Work complete on replacement of 2B21 NO38A. Ser# 85-1-2376

- WR# L59802

• Switch disassembled and inspected at LSCS, July 30, 1986

- no damage noted in diaphragm

- switch was displaying relaxation phenomena and calibration was very sluggish and erratic, e.g., setpoint did not follow adjustment

- switch report contained in August 1, 1986, " Report of Investigation of SOR Differential Pressure Switches, LaSalle County Station "

• Reported in NPRDS

• Disassembled switch was placed in bag and left stored in LSCS

" hot shop"

• Pinholes discovered 12/18/86 upon reinspection of the switch at LSCS

- pinholes could have been caused in storage after disassembly 7/18/86

• Work complete on replacement of 1E12 N010CB. Ser# 85-1-2413

- WR# L59561

- IM's could not maintain differential pressure across switch while attempting to calibrate

• Switch operated in service from 3/85 to 7/86

• Switch failure was prior to start of Unit 1 switch characterization tests

- was not included in August 1, 1986 report

- CECO was not notified of diaphragm failure by SOR Inc. until 12/18/86

• Switch disassembled at LSCS in July prior to shipment to SOR Inc. and no problems noticed in diaphragm

- Body bolts in switch were believed to be insufficiently torqued

- switch was shipped to SOR Inc. for inspection as potential manufacturing or qualltf assurance problems

• Reported in NPRDS

• SOR inspected switch in August but did not notify CECO of diaphragm failure as they were at that time concentrating on

" static shift" and " rusted bearing" problems 12/09/86

• Switch 2E21 N013BB failed functional test

- WR# L33908

- could not get pressure across switch

• Red phone call made 12/09/86 to NRC

• Switch sent to SOR Inc. for disassembly and inspection

• Switch disassembled and inspected at SOR with CECO representative on 12/16/86

h 12/10/86

• Work completed on replacement 2E21 N013BB. Ser# 85-1-2428 i

• Failure of 2E31 N013BB triggered memory of CECO engineer  !

- recalled previous failure in July (1E12 N010CB Ser# 85-1-2413)

- discussed with LaSalle engineers who remembered other l previous failures

• Station started ausembling data to identify all SOR differential pressure switch failures

• NRR Project Manager given verbal description of failure

- suspected diaphragm failure, not unusual for mechanical switch

- switch will be sent to SOR Inc. for disassembly and inspection

- this is not only diaphragm failure 12/16/86

• Switch 2E31 N013BB Ser# 85-1-2428 disassembled and inspected by SOR Inc. with Ceco engineer at SOR Inc.

- tear found in diaphragm

- tear postulated by SOR Inc. to be from fatigue failure of diaphragm material

• SOR Inc. notified CECO engineer that earlier switch had puncture failure and showed him ' ie diaphragm from switch Ser# 85-6-3587 which failed in April 1986

- CECO engineer believed that to have been 1E12 N010CB Ser#

85-1-2413 which had failed in July 1986

• SOR pressurized diaphragm on another switch up to 2200 psig to rupture it for comparison of failure modes

- the appearance of the tears from each of above appear different therefore failure modes believed to be different for each case.

• Telecon with NRR regarding Long Term Action Plan and Justified Continued Operation report which was due 12/22/86

- commitment made to provide information about 2E31 N013BB (Ser# 85-1-2428) switch failure to NRR when available 12/18/86

• CECO engineer notified that switch diaphragm he saw at SOR Inc.

on 12/16/86 was 1E12 N010CA Ser# 85-6-3587

• Discovered that there were two IE12 N012CA switches replaced 1

(Ser# 85-1-2411 and 85-6-3587) '

- 1E12 N010CA Ser# 85-6-3587 was at SOR Inc. and had holes in diaphragm i

- IE12 N010CA Sor #85-1-2411 was at LSCS in " hot shop" l

• 1E12 N010CA SER# 85-1-2411 was disassembled and inspected at  !

LSCS by CECO personnel and pinholes were found in diaphragm i

• 1E12 NO38A Ser# 85-1-2376 was located in " hot shop" and l disassembled

- switch diaphragm was inspected and small pinholes found

- appearance is different from other holes previously found

• Rough writeup of diaphragm history was given to Region III Resident Inspector

• CECO engineer notified that switch IE12 N010CB Ser# 85-1-2413 had been disassembled and inspected by SOR Inc. and holes were found in diaphragm ,

2545K

. .o i

12/23/86 Telecon with upper CECO management to assign responsibilities for investigation of switch diaphgragm performance

• Ceco's System Material Analysis Department (SMAD) assigned to perform investigation of diaphragms from disassembled switches

• LaSalle Station assigned to research work history to insure that all removed switches had been identified

• Ceco's Station Nuclear Engineering Department (SNED) assigned responsibility to write report of investigation 1/08/87 SMAD Report

• Examination using 50K magnification revealed

- additional perforations

- foreign material embedded in diaphragm

• Additional switch which had never seen service was examined and found to have foreign material on diaphragm 1/14/87 Static "O" Ring Report

• Results of SOR investigation of 8 diaphragms from previously disassembled switches found two additional diaphragms with small perforations

• Provided their interpretation of 50K pictures of diaphragms examined by CECO SMAD 2545K

s 4

12/23/86 Telecoin with upper CECO management to assign responsibilities for investigation of switch diaphgragm performance e Ceco's System Material Analysis Department (SMAD) assigned to perform investigation of diaphragms from disassembled switches e LaSalle Station assigned to research work history to insure that all removed switches had been identified e Ceco's Station Nuclear Engineering Department (SNED) assigned responsibility to write report of investigation 1/07/87 SHAD Report

• Examination using 50K magnification revealed

- additional perforations

- foreign material embedded in diaphragm e Additional switch which had never seen service was examined and found to have foreign material on diaphragm and internal surfaces 1/14/87 Static "O" Ring Report

• Results of SOR investigation of 8 diaphragms from previously disassembled switches found two additional diaphragms with small perforations e Provided their interpretation of SOK pictures of diaphragas examined by Ceco SMAD 2545K