ML20203H169

Summary of CRGR Meeting 92 on 860711 Re Proposed IE Bulletin, Unreliable Behavior of Static O Ring Differential Pressure Switches. Attendance List Encl

ML20203H169
Person / Time
Issue date:	07/23/1986
From:	Sniezek J Committee To Review Generic Requirements
To:	Stello V NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO)
References
NUDOCS 8608040323
Download: ML20203H169 (27)

Text

,7,.

JUL 2 31986 MEMORANDUM FOR:

Victor Stello, Jr.

Executive Director for Operations FROM:

James H. Sniezek, Chairman Committee to Review Generic Requirements

SUBJECT:

MINUTES OF CRGR MEETING NUMBER 92 The Committee to Review Generic Requirements (CRGR) met on Friday, July 11, 1986 from 10 a.m. to 12 N.

A list of attendees for this meeting is enclosed (Enclosure 1).

D. Allison (IE) presented for CRGR review, the proposed IE Bulletin,

" Unreliable Behavior of Static 'O' Ring Differential Pressure Switches." summarizes this matter (Category 1 item).

In accordance with the ED0's July 18, 1983 directive concerning " Feedback and Closure on CRGR Reviews," the item above requires written response from the cognizant office to report agreement or disagreement with CRGR recommendations in these minutes. The response, which is required within 5 working days after receipt of these meeting minutes, is to be forwarded to the CRGR Chairman and if there is disagreement with the CRGR recommendations, to the EDO for decision-making.

Questions concerning these meeting minutes should be referred to Walt Schwink (492-8639).

Original Signed by James H. Snie ek James H. Sniezek, Chairman Comittee to Review Generic Requirements

Enclosures:

As Stated cc: Commission (5)

Distribution:

SECY JHSniezek JRoe Office Directors RShumway WLittle Regional Administrators JClifford EFox

(

CRGR Members PTing RErickson i

G. Cunningham JPhilips FHebdon jh D. Allison JZwetzig DEDR0GR cf il PDR (NRG/CRGR)

ROGP Staff 860eo40323 e60723 knf&NRCCRGR PDR

,g 0FC :ROGR

ROGR/D

DEDROGR NAME :JConran

JFZerbe

' :J rezek M 86 DATE :7/21/86

7/g2/86

7/

/

LIST OF ATTENDEES CRGR MEETING NO. 92 July 11, 1986 CRGR MEMBERS J. H. Sniezek R. Bernero E. Jordan J. Scinto J. Heltemes R. Cunningham D. Ross OTHERS W. Schwink E. Baker J. Conran R. Woodruff T. Cox M. Srinivasan W. Hodges M. Virgilia G. Holahan J. Beard T. Collins D. Sullivan J. Zerbe D. Allison

,...,._____,--__...,____m._--_

1 to the Minutes of CRGR Meeting No. 92 Review of Proposed IE Bulletin on " Unreliable Behavior of Static 'O' Ring Differential Pressure Switches" D. Allison (IE) presented for CRGR review a proposed IE Bulletin addressing staff concerns regarding unreliable behavior observed at two operating plants of mechanical differential pressure (dp) switches employed in reactor vessel level and pump flow indication applications. The nature of the failures observed were such that staff concerns arose regarding the potential for common-mode failure of these dp switches and interference or defeat of critical safety functions. The level of staff concern that developed led the staff to designate the proposed bulletin as a Category I action item and to request expedited handling by CRGR. Copies of the briefing slides used by IE at this

- meeting are attached (see Attachment I to this Enclosure). A copy of the pro-posed bulletin is also attached (see Attachment 2 to this Enclosure). The package of documents submitted by IE for CRGR consideration et this meeting were transmitted by memo dated July 8, 1986, Jame-M. Taylor to J. H. Sniezek; that package included the following documents:

1.

Draft IE Bulletin No. 86-02, dated July 1986, entitled " Unreliable Behavior of Static 'O' Ring Differential Pressure Switches."

2.

"Information for CRGR Review of Proposed IE Bulletin 86-02," dated July 8, 1986.

Major points of discussion at this meeting regarding the proposed Bulletin were as follows:

1.

A question was raised regarding the need for any serious tor.cern over a level switch operating a few inches outside of speciitad water level, when the total amount of water in a reactor vessel das on the order of several hundred inches. A drop in water level of many feet (as contrasted to several inches) would be required to endanger the core.

In response, the staff first noted simply that the dp switch problems at issue are report-able as violations of water level limits / ranges established by Technical Specifications for the plants involved, aside from any immedtate or apparent safety concern. The staff acknowledged that the safety implica-tions of the dp switch failures that have been noted are not so serious in 1

the water level detection application. Of greatest concern to the staff is the potential common-mode dp switch problem possibly indicated by the experience to date, in particular the potential for common-mode failure of dp switches employed in the mini-flow recirc line application for protec-tion of pumps in critical ECCS/RHR functions.

2.

The Comittee inquired whether the dp switch vendor involved had provided guidance to licensees regarding the testing or calibration of the types of

4.

switches involved or regarding the causes and recommended corrective ac-i tions for the problems that have been experienced at Oyster Creek and LaSalle (e.g. comparable or equivalent to GE Service Information Letters).

The staff indicated general dissatisfaction with vendor actions to date in this matter, noting that no such generic guidance has been provided to licensees.

In fact, the vendor's view at this point seems to be that the erratic switch behavior that has been observed to date is completely /

uniquely switch dependent, therefore generic resolution is not feasible.

3.

The Committee inquired regarding licensee response to the IE Information Notice (86-47) that was issued in June immediately following the erratic behavior of dp switches reported at LaSalle on June 1.

Specifically, CRGR questioned whether there had been any action or response by any utility group relating to the problems identified in IEIN 86-47. The staff indi-cated general disappointment in licensee response to the earlier IEIN on j

the dp switches of concern; specifically there has been no generic actions or response by the utilities indicating that the developing problem was being addressed broadly and effectively. The staff felt that this sup-ported the need for the proposed Bulletin at this time.

i i

e 4.

The Committee inquired (in view of the staff's developing concern regard-ing a possible serious comon-mode failure problem, and the data indicat-i ing erratic / unreliable operation of the dp switches involved) why the I

staff simply did not specify replacement of the Static "0" Ring (SOR) dp switch with another component. The staff responded that they felt that the measured response being taken as reflected in the proposed Bulletin 4

was more appropriate.

In effect, the actions /information requested by the i

Bulletin amount to a notice to "show cause" why SOR switch should not be i

replaced. The reliability of that component is obviously seriously in question by the staff at this point; and if the testing data that will be generated does not clearly show that proper control of calibration / test methods can guarantee the necessary repeatability and reliability in dp i

switch operation, then the decision to replace it with another design would be fully warranted.

l 5.

The Committee commented that the circumstances and sequence of events in-

[

volved in the development of this problem raises serious questions not only about the licensees' programs for equipment qualification (or verifi-cation of equipment qualification), but also about the related internal NRC staff EQ review process. They questioned specifically whether any staff action had required replacement of previously approved level indicating equipment with SOR dp switches as part of the ongoing EQ verification process. The staff stated that this had not been required, but that some licensees had on their own initiative installed the SOR dp switches as a means of complying with harsh environment requirements. The l

Committee's questions then focused on whether the staff had adequately

~

i reviewed such changes, and whether the SOR dp switches in question had been placed on a list of NRC approved (EQ-rated) components, perhaps encouraJ ng other licensees to install the SOR switches. The staff felt i

that th" cheaper cost of the SOR switches had been a driving consideration for a number of licensees who had turned to them. The Comittee recem-i mended that the staff examine the EQ review / verification process in view of this experience to determine whether the process needs improvement.

l l

6.

The Comittee questioned whether the staff had fully considered the possi-ble adverse effects of the dp switch testing that is specified in the proposed Bulletin, in particular the testing of inputs to the scram i

function while the reactor is at power and the dropping of RV water level to test dp switch operation. The staff responded that they had not done complete analyses (probabilistic studies) that would provide incremental risk estimates in this regard; but they did try to consider bounding or worst-case type circumstances that might arise in testing dp switches in deciding to issue the proposed Bulletin.

Further, they have included as an action required of licensees a more complete analysis of the possible adverse effects of the proposed testing. As a final point, the staff also noted that the tests specified in the proposed Bulletin are similar to surveillance tests already specified in the Tech Specs of affected plants.

7.

In response to direct inquiries by CRGR, the staff acknowledged that some declarations of inoperable dp switches and resulting plant shutdown (or extension of current shutdown) can be anticipated as a result of issuing this Bulletin; but that is expected to occur in only one or two instances at most on the basis of the best information now available to the staff.

8.

The Comittee comented that is was important to indicate clearly to af-

~

fected licensees that additional regulatory action may have to be taken by the staff (e.g. a followup bulletin requiring additional testing) on the basis of the testing data and. scope-of application information requested in this bulletin. This could be an important consideration in licensees choosing imediately the SOR switch replacement option if, in their judge-ment, the switch testing /requalification process promises to be too long and drawn out a process.

9.

On the basis of the discussions at this meeting the following specific changes to the wording of the proposed Bulletin were recomended by the Comittee and agreed to by IE (changes keyed to Attachment 2 to this Enclosure).

a.

Page 1, unoer " Purpose" Change " safety-related systems" to read " systems subject to Technical Specifications."

b.

Page 2, Third Paragraph Delete reference to INP0 c.

Page 3, Second Paragraph Change " precursor" to "similar" d.

Page 4, Second Paragraph Change wording of last sentence to eliminate any suggestion that daily surveillance testing of dp switches is an appropriate resolution of the problem of unreliability/unrepeatability in dp switch operation.

e.

In" Action" Statements (throughoutBulletin)

Change "shall" to "should" Restate response time requirements as they are applicable to CP holders and/or plants now in extended shutdown (such as TVA plants).

Instead of 7-day, 30-day, 60-day response times, etc., indicate

" prior to critical operation" or "within one year" etc., as applicable for such instances.

f.

Page 5, Last Paragraph Delete reference to " applications in mild environment."

g.

Page 6, Top of Page, Action Statement 2 Change wording to indicate that the scope of survey of plant systems for 50R dp switch installation is the same as in Action Statement 1 (previous page). Scope of dp switch testing is " systems subject to Technical Specifications."

h.

Page 6, Under 7-day Actions Indicate that reactor operators on duty should be informed.....

i.

Page 6, Second Paragraph from Bottom Clarify that reporting will be in accordance with 10 CFR 50.72 and 10 CFR 50.73 and make clear that NRC staff has determined that SOR dp switch failures should be regarded as a generic, potentially coman-mode failure problem for reporting purposes.

j.

Page 6 (Sentence at Bottom)/Page 7 (First Paragraph)

Clarify staff intent with regard to criteria for declaring switches / channels operable or inoperable.

Delete requirement that restart of a plant shutdown because of inoperable SOR dp switches must be approved by NRC Regional Administrator.

k.

Page 6, Action Statement 2 Revise wording to indicate that repeat of testing required by bulletin nej end after two successive successful tests are completed.

1.

Page 7/Page 8, 60-day Action Statement 2 Revise generally to clarify staff intent.

On the basis of their review of the proposed Bulletin at this meeting, CRGR recommended the following:

1.

The Bulletin should be issued subject to changes agreed to by IE as re-flected in the preceding. The CRGR Chairman requested that the revised Bulletin be provided to him for final review prior to issuance.

2.

The staff should conduct a review of the Environmental Qualification review and verification process (both the licensee's methods for ensuring operability and qualification of plant equipment. and the related internal NRC staff review practices) in view of the circumstances that gave rise to this proposed Bulletin.

e

Atachment I BASIC PROBLEM SWITCHES BEHAVE DIFFERENTLY IN SERVICE THAN IN NORMAL CAllBRATION TESTS CAN BE A SIGNIFICANT PROBLEM DEPENDING ON APPLICATION ASKING FOR GOOD TESTS TO ASSURE WILL WORK IN' SERVICE 4

~

r-

.s t.

i 4

SUMMARY

OF REACTOR LEVEL DROP TEST DATA i

Test Rx. Pres.

Trip Setpoint (Rx. Level in Inches)

No.

Date Time (psig)

NO24A NO24B NO24C NO24D 1

6-2-86 1500 950 8.6 3.5 6.5 9.7 2

6-2-86 2100 950 10.3 4.6 8.4 7.9 3

6-3-86 1900 500 9.3 5.8 8.2 11.2 4

6-3-86 2100 500 12.0 10.7 11.2 10.3 5

6-4-86 2100 0~

10.7 11.7 11.4 11.6 6

6-5-86 1400 0

11.1 11.1 12.2 13.5 i

4 1

a l

l i

i i

t 1

4 g '.

/s

,t, 1,

lleocl 3 sde(.j l

I i

i 11 I

i i

~~

I I

I il I 11 I

I I

I i

4..

i i

i t

iI i

il t

• i l i

i l

+

S ric-u,-3ing,r...eremial,ressure l.-

Swi ca vpera)ili;y Tes;,Ja;a l

2B21-E24

~

60 =

E Test #1 (Static Pressure Effect) 1

_____ -+

-==

/D'hNb h15 9_A?y- - -

l d

g;C _a

.m

=

O d

i b2 -

Existing Allowable Tolerance l

' Lower Limit (b2.51) j-3 b3 g

[,,

5'r.

....* De s i y ed se t P o i n t ( 63. i 5 )

- Existing Allowable Tolerance l

64g

' Upper Limit (b3.78) N. T. S. P.

e i

~

C E

i opng

.- 65 1 4 e:

- o E

{.

o_

3 66

~

l V

W 67 E

.i i

i I

I I

I I

I I

I

)

68 l-1 2

3 4

5 6

7 8

9 18 Actuation l _.- c,. i. 7 c.... l - c,. i.. ci.,.. j _,_ c,.i.. c.,. i.i l _

c,. i. 1. ca.

.. l

[

c,. i. t c.... l -. c,.i. 2 ei. ri,i l _

c,.i. 2 c.,.i.i l

- c,.i.. ct

>,. l _., c,.i.. c.,.. j c,. i.. e...,.. 1 2

i I

I I

.i 4 eoql 7 sG(ek I

.I

,1 I

.i l

i 1.

S;a;ic-u,-Ring iferential.,)ressure -

Swi;ca vpera)ih. ;yTes;,Jaa 3

L 2321-X024B

{E. ~

60,

I Test #1 (Static Pressure Effect)

~

9 i

j '.

bi-i m

5 i

(_) b i Existing Allowable Tolerance

_ Lower Limit (62.51)

'3 5

63-p:

...e De s i r ed Se t Po i n t ( 63.15 )

, T

=

E

- Existine Allowable Tolerance 6 4 ? ___

' Upper Limit tb3.70) N. T. S. P.

=

+>

l

__4 O pWg C

E

. - s g =a

J o

. _a_

g

- -dd 'N

~

bbJ l

g_W 1_____;

/000 3Si}

~~

i

. D 67 2#

L 1

l-4 1

I I

I I

I 1

I I

il, gg:1 2

3 4

5 6

7 8

9 18 1:

F, Actuation l _

c,.i. 7 c.,.i., p-c,.i.. ci

.,i l _._ c,. i.. c.,. 6,i l _._ c,. i. i. cin.,...i l l _.i_ c,. i. t c.,.i,i l _._ c,. i. 2 c 1...,.i.i l _,_ c,. i.

c.,.i,i p-c. i.. c a...... p c,.i. s c.,. i.i H-c,. i. 6 ca..,..i l J

perg$ung poet ld P MOUNTrpC 50tEWS g

NOUS MG g

$N AP SwlTCH

{

PetE55UNE PORT o-n RG l

OPERATING ROD 6

ADJUST CAP N l

N OPERATMG ROO SU$NING 8

e

<(

OPER ATING ROD ADJUST SCREW

.i j

x o

t k

R h

AOJUST_

1,

/

I SwtTCN LEVER

~

.. ;..g,'

f pSWITCH HOUSING GA5p(ET & COVER drive SCREW 828 SPRING -

"' ga"'.

j E TAMER f

O NAMEPLATE

/

' (

N socy g.

/

g

~N

{g O-RINGS 123

$PRING

• SEAT SPmHg SP DIAPHRAM

'N' y

.. s, pr i

Nouswc

,RhG j

s l

l

=

i GASE T &

/

PISTON (2)

- COVER

/

,e PSTON RETAINER "G

l 4

LEVER

,PtsTON RE TAtNER SCREW 425 h

PtSTON LEVER l

=

cover PISTON OPERATING ROO

[NM -_

$2 l

SCREW CYLINDER DISC 1

wDr s

SPRING

$PRING BE ARING E

j NOUSING HOUSNG A('JU ST

~

.j

^

sgTg, scaEw r

- j' eEAR,NG 5

/ rA"

= 01 sE T sco.

sP s

/'

==-

a==

LEvfA.

, >,U v / ;' As,N.

f

- =.

> \\\\.s_A'd s

s

" " " " '," " E 2V bNs.?

. 6 CRos,s w

~

rm,

.a sNA T

.;,s,,,

l

__. _ _.. _ _ m

~~

SET SCREW

~._.:::::

.l

- in ;

v_.

g,y c.

13

'N:

'<:xdpatE(

= 11;; _:-

2

- x.

.,. e maan-=- g

\\\\

f N'

go36 p

).

., N g 1

/

COMPONENTS LOCATW o ging

\\

)

\\

swlTCN unusyYgg

~

SPRMG OUTER (2)

CEN TER (7)

INNER IOU $ LNG SLEEvt steeve SLEEVE SCRE,

lM.)

uosne y ser. vstt w i2) en en

.it

., t. cosz$ NKL. MtblUN OH PU4 tj lt D

1.Q...

~

• _l A.a: q,..

d @

SteG g>g.<a

..cp.gwttehe,= _

. -.. iw. -. 4...t A.,J

'1i t, XW

..f..)

.1j No, Switch No.

Description Range Set Pt.

Model 9

1,2 E31-N007AA,AS RHR/RCIC Stm Hi 28-299"

+123"(AA)' 193-B283 Flow

-123"(AB) 3,4 E31-N007BA,BB RHR/RCIC Stm Hi 29-200"

+B6" (BA) 103-B203 Flow

-B6" (BB) 5 to B E31-N000A,B,C, MS Hi Flow A 100-500 111 PSID 102-B305 D

PSID 9 to 12 E31-N009A,B,C, B

D 13 to 16 E31-N10A,B,C, C

D 17 to 20 E31-N011A,B,C, D

D 21 to 22 E31-N012AA,BA RHR Wtr Hi Flow 20-200" 121.5" 103-B203 23,24 E31-N012AB,BB

-121.5" 25 to 20 B21-N024A,B,C, RPV Level 3 7-100" 63.78" 103-B212 D

29'to 32 B21-NO31A,B,C, RPV Level 2 20-200" 145.63" 103-B203 D

./

33 to 34 B21-NO37AA,BA RPV Level 1 40-300" 202.16" 103-B3205

/

35,36 B21-NO37AB,BD RPV Level 2 20-200" 145.63" 103-B203 37,38 B21-NO376A,DA RPV Level 1 40-300" 202.16" 103-BB205 39,40 B21-NO37CD,DD RPV Level 2 20-200" 145.63" 103-B203 41,42 B21-NO38A,B RPV Level 3 Interi.

7-100" 63.78" 103-B212 43,44 E12-N010AA,BA RHR Flow DP 5-35" 5.9" 103-B202 45,46 E12-N010AB,BB RHR Flow DP 5-35" 15.3" 103-B202 47 E12-N010CA RHR Flow DP 5-35" 5.9" 103-0202 48 E12-N010CB RHR Flow DP 5-35" 15.3" 103-B202 49 E21-N004 LPCS Flow 5-35" 8.6" 103-B202 50 E22-N006 HPCS F1ow 5-35" 10.1" 103-BB202 51,52 B21-N026,AB,BB RPV Level 2 30-200" 145.63" 103-DB203 53,54 B21-N026CB DD RPV Level 2 30-200" 145.63" 103-BB203 55,56 E31-N013AA,DA RCIC Stm Flow 20-200"

+130" 103-B203 57,58 E31-N013AB,BB RCIC Stm Flow 20-200"

-130" 103-D203 59,60 B21-N100A,9 RPV Level B 7-100" 33.27" 103-B212 61,62 B21.-N101A,B RPV Level O 7-100" 33.27" 103-B212 l

6 6

l l

G j.

m+-

OTHER APPLICATIONS 4

WNP-2 VESSEL LOW LEVEL l

MILLSTONE 1 TORUS TO REACTOR BUlLDING VACUUM BREAKER DUANE ARNOLD VESSEL DOUBLE LOW LEVEL OYSTER CREEK VESSEL. LOW LEVEL VESSEL DOUBLE LOW LEVEL

,_ SEQUOYAH UPPER HEAD INJECTION LOW LEVEL ISOLATION BROWNS FERRY RHR MINI-FLOW CONTROL 4

e e

~ - - -,.

. _ - ~

_~

ACTIONS SIGNIFICANT TO SAFETY 7 DAYS REPORT SAFETY RELATED

,i 7 DAYS - BRIEF OPERATORS 1

30 DAYS - TEST 4

• PREFER SYSTEM TEST 1

b

• DON'T JUST READJUST

• DON'T JUST REPLACE NRC APPROVE RESTART

~

REPEAT TEST CONSIDER COMPLICATIONS 60 DAYS - LIST RESULTS CONTINUING DEMONSTRATION ANALYSIS i

i t

4 e

9

..r.n

r.

BENEFITS AND COSTS J

SAFETY SIGNIFICANCE DEPENDS ON APPLICATION COMMON MODE FAILURE PROBABILITY I

COSTS INDUSTRY

$1,500,000 NRC 87,000 l

l 1

1 r

h 4

i

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

Attcchment 2 SSINS No.:

6820 OMB No.:

3150-0012 IEB 86-02 UNITED STATES NUCLEAR REGULATORY COPNISSION OFFICE OF INSPECTION AND ENFORCEMENT WASHINGTON, DC 20555 July XX, 1986 IE BULLETIN N0. 86-02:

UNRELIABLE BEHAVIOR OF STATIC "0" RING DIFFERENTIAL PRESSURE SWITCHES Addresses:

All power reactor facilities holding an operating license (0L) or a construction permit (CP).

Purpose:

. The purpose of this bulletin is to request that boiling water reactor (BWR) and pressurized water reactor (PWR) licensees determine whether or not they have Series 102 or 103 differential pressure switches supplied by SOR, Incorporated (formerly Static "0" Ring Pressure Switch Company), installed in systems impor-tant to safety, which includes safety related systems and certain non-safety-related systems.

Those licensees that have SOR Series 102 or 103 differential pressure switches in safety related systems are requested to take certain actions to assure that system operation is reliable.

Description of Circumstances:

SOR Series 103 differential pressure switches were installed in LaSalle 2 in mid 1985 as part of an environmental qualification modification which was performed after initial operation of the unit.

Identical switches were also installed in LaSalle 1.

LaSalle 1 and 2 each have about 60 of these switches in various systems, including the reactor protection system and the emergency core cooling system.

On June 1, 1986, LaSalle 2 experienced a feedwater transient that resulted in low water level in the reactor vessel.

One of four low level trip channels actuated, resulting in a half scram. The operator recovered level and operation was continued.

However, subsequent reviews by some of the Licensee's personnel raised concerns that the level apparently had gone below the scram setpoint and that a malfunction of the reactor scram system may have occurred.

Based on this concern, the Licensee declared an " Alert," shut the plant down, notified the NRC, and subsequently informed SOR of possible switch malfunctions.

NRC dispatched an augmented, inspection team to the site on June 2 to review the root cause and significance of the feedwater transient, the performance of the differential pressure switches in the low level trip channels, the response of the reactor protection system, and related matters.

IEB 86-02 July XX, 1986 Page 2 of 8 After recalibrating the level switches on June 1, the Licensee tested the performance of the level switches by lowering water level (drop test) in the reactor and reading the levels indicated on each of four level transmitters when each of the four level switches tripped.

The results were erratic with the switches tripping at levels between 2.4 inches plus or minus 1.5 inches and 12.2 plus or minus 1.7 inches depending on the transmitter read.

These measurements are relative to instrument zero which is at 161.5 inches above the top of active fuel.

The technical specifications require that level channels be declared inoperable if the setpoint drifts below 11.0 inches and that the switch be recalibrated to trip at 12.5 inches.

4 As of June 9, the Licensee had tested differential pressure switches in the residual heat removal systems and the high pressure core spray system.

These switches open valves in minimum flow recirculation lines so that adequate cooling to pump seals and bearings is provided when system flow is low.

One of the switches actuated within the range permitted by technical specifications; the others did not.

The switch for the high pressure core spray system was calibrated to actuate at 1300 gpm but did not actuate until flow decreased to 530 gpm.

The switches for the two residual heat removal systems should have actuated at 1000 gpm but did not actuate until flow decreased to the 480 to

-~_ 800 gpm range.

On the basis of these results, the Licensee declared all emergency core cooling systems for Units 1 and 2 to be inoperable.

Both units remain in cold shutdown.

Information Notice 86-47 was issued by the Office of Inspection and Enforcement on June 10 and informed licensees of the erratic behavior of 50R differential pressure switches during the event at LaSalle 2 on June 1 and during subsequent testing.

An attachment to the information notice listed plants to which SOR had supplied Series 103 differential pressure switches.

That list has been updated (Attachment 1) to include Series 102 differential pressure switches which have important similarities to Series 103 switches.

The information notice also announced a meeting of representatives from NRC, General Electric Company, 50R, and interested licensees to discuss the application and performance of Series 102 and 103 switches in safety related systems.

The meeting was held on June 12 and was also attended by a representative from the Institute of Nuclear Power Operations.

The Licensee has done a significant amount of drop testing and static pressure testing and has initiated a short range test program to support restart of Lasalle Units 1 and 2.

SOR has initiated short and long range test programs to support continued use of the Series 102 and 103 switches in safety related l

systems.

To demonstrate reliability of the switches, SOR testing will continue for 18 months.

During the vessel water level drop tests at LaSalle 1 on June 2, one of two Series 103 switch used to provide a confirmatory water level input signal to the j

automatic depressurization system failed to function.

Subsequent investigation l

revealed that the switch had not actuated.

On June 17, testing showed that the setpoint had shifted nonconservatively by 25 inches.

In this application, the relative locations of the instrument taps are such that the system could not l

l

IEB 86-02 July XX, 1986 Page 3 of 8 produce sufficient differential pressure to actuate the switch.

Therefore, this amount of shift constitutes a functional failure. On June 25, the switch was disassembled and inspected.

Rust (severe corrosion) was found inside the switch assembly and probably caused a cross shaft bearing, which is outboard of the 0 rings, to seize.

A precursor event (Licensee Event Report 86-001-00) occurred at Oyster Creek 1 on January 17, 1986, during monthly surveillance of four SOR differential pressure switches which detect low water level in the reactor vessel.

The "as-found" setpoints for three of the switches had dri_fted downward as much as 6 inches.

During the subsequent 11 weeks, the level switches continued to perform erratically, each switch was replaced one or more times, and modified switches were installed.

On April 7, after a modified switch had nonconservative setpoint drift, the Licensee elected to perform daily surveillance until about April 12 when the reactor was shutdown for a.six month outage.

Earlier concern for mechanical level indication equipment was expressed in NRC Generic Letter No. 84-23 which addressed water level instrumentation for BWR reactor vessels. The generic letter was based on NRC's evaluation of a report

-, by S. Levy, Incorporated, which had been commissioned by a BWR Owner's Group.

The generic letter included a requirement for BWR licensees to review plant experience related to mechanical level indication equipment, indicated that analog trip units have better reliability and greater accuracy than mechanical level indication equipment, and stated that BWR licensees should replace such equipment with analog transmitters unless operating experience indicate's otherwise.

Responses to Generic Letter No. 84-23 show that 80% of BWR licensees have replaced or plan to replace their mechanical level instrumentation with analog level transmitters.

Recipients of this bulletin should recognize that while this bulletin focuses on more immediate problems with two kinds of mechanical differential pressure switches manufactured by 50R, Incorporated, the reliability of other mechanical instrumentation is also in question because it may be vulnerable to similar problems.

Discussion:

DESCRIPTION OF SERIES 102 AND 103 PRESSURE SWITCHES The Series 102 and 103 differential pressure switches consist of a piston (Series 102) or a diaphragm (Series 103) which moves a lever that rotates a cross shaft.

These components are contained in a steel case designed to withstand system pressure.

Both ends of the cross shaft extend out of the wetted volume and 0-ring seals prevent leakage of water along the cross shaft.

The condition of these surfaces and the condition of the 0-rings will determine the extent to which frictional forces cause a torque which opposes rotation of the cross shaft.

A lever is attached to each end of the cross shaft.

When the cros; shaft rotates, one lever moves to actuate a microswitch. The other lever bears on a helical spring.

An adjusting screw is used to change the compres*

sion of the spring and thus change the setpoint of the differential pressure switch.

IEB 86-02 July XX, 1986 Page 4 of 8 The case contains two ports on either side of the 5,iston or diaphragm.

The lower port on one side is connected to the system reference Icg, and the lower port on the other side is connected to the lower instrument tap (i.e. variable leg). The upper ports on both sides are connected to vents which are valved closed when the switch is in service.

The design of the cavity containing the diaphragm (or piston) is such that motion of the diaphragm is limited to 0.015 inch.

Most of the time, the diaphragm is against one or the other of the mechanical stops which limit motion of the diaphragm. Thus the sum of the unbalanced hydraulic forces across the diaphragm is supported by one stop or the other except when the microswitch is forced to change position.

This occurs when the absolute value of the torque caused by the unbalanced hydraulic forces changes from a value less than to a value greater than the torque caused by the helical spring.

This movement causes the cross shaft and the levers to rotate 1.8 degrees.

The differential pressure switches are often calibrated without removing them from the system.

A test rig consisting essentially of two bottles each containing water and air or nitrogen are connected to the differential pressure switch with one bottle on either side of the diaphragm.

The differential pressure for calibration is established by adjusting the gas pressures in the bottles.

Often, the lower pressure is at or near atmospheric pressure. When the differential pressure switch is in service in BWR water level applications, pressure is approximately 1000 psi.

PROBLEM AREAS Calibration of Series 103 differential pressure switches at atmospheric pressure sometimes results in an offset of the setpoint in the nonconservative direction when the switch is returned to service at BWR system pressure.

It has been postulated that this may be caused by deformation of the 0 rings on the cross shaft when system pressure is applied.

For water level applications and depending on the location of the lower instrument tap relative to the required setpoint, offset may be so large that the switch will not actuate before the level drops below the tap.

In this case, the switch would not actuate no matter how low the level dropped.

In other cases, the offset of setpoint due to calibration at atmospheric pressure has been found to be in the opposite direction.

The manufacturer has stated that each switch has unique offset characteristics and that switches with the same model number do not behave in the same way.

For minimum flow applications where it is necessary to open a valve in a recirculation line to protect a pump in an emergency core cooling system, assurance is needed that offset will not delay that action and result in pump damage.

However, it has been the practice at LaSalle to calibrate at atmospheric pressure without compensating for errors due to pressure offset.

During the week of June 24, 1986, testing of a Series 103 differential pressure switch at LaSalle showed that application of a static pressure to the switch for a period of time resulted in a significant shift in the setpoint of the switch.

A After being calibrated at 1000 psig, the static pressure was maintained.

TEB 86-02 July XX, 1986 Page 5 of 8 recheck of the setpoint at the end of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> showed that the setpoint had shifted nonconservatively by about 10 inches.

Subsequent rechecks continued to show shifting but in lesser amounts.

To be valid, it appears that calibration and tests would need to be rechecked after static pressure has been maintained for at least 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />.

Recent testing at LaSalle has shown that the point at which trip occurs depends on whether the switch setpoint is being approached from low differential pressure or high differential pressure. This is particularly important for automated blocking valves in the recirculation lines which protect emergency core cooling pumps from damage when system flow is low. When flow dec.3ases to a value below the setpoint, the switches should actuate to open the valves.

Conversely, when flow increases, the switches should deactuate to provide maximum flow to the core.

In addition to showing offset problems, some of the Series 103 switches evidence sticky behavior, i.e. a larger change in differential pressure is required to actuate the switch on the first test than on subsequent tests and on subsequent tests actuation may be erratic.

It is believed that starting friction and the condition of the cross shaft surfaces may cause these problems.

If the 0 rings e

stick, then the torque that they apply is added to the torque applied by the calibration spring.

50R is conducting a long range test with switches that have polished finishes on those parts of the cross shafts that are in contact with 0 rings.

A group of control switches having cross shafts with standard finishes is also being tested.

It has been the practice at LaSalle to actuate the switches several times and to record the differential pressures required for actuation before adjusting their setpoints.

It appears, however, that the Licensee has not emphasized that the "as-found" condition of the switch is the value of differential pressure required to actuate the switch during the first pressure cycle after it has been taken out of service.

It is this value that must be used to determine whether the switch and its safety related system would have performed their intended functions if called upon to do so.

The life of Series 103 switches has been said to be 20 to 40 years.

However, the shelf life of the elastomeric material used in the 0 rings is considerably less than 40 years. The 0 rings may need to be changed several times during the life of the plant.

Further, there is some concern for the effect of reactor water on the 0 rings, cross shaft surfaces bearing on the 0 rings, and on the l

diaphragm material.

Actions Required for All Licensees:

[,

PROMPT ACTION j

1.

All licensees shall promptly determine whether or not your facilities have 50R Series 102 or 103 differential pressure switches installed in, or have i

plans to install such switches in systems important to safety, as defined in 10 CFR 50.49(b), but also including applications in mild environment.

i IEB 85-02 July XX, 1986 Page 6 of 8 2.

Licensees shall submit a report within 7 days on action item 1 above.

For each switch installed in systems important to safety, the report shall i

identify: the model number, the system in which it is installed (e.g., low pressure safety injection system), its application (e.g., flow measurement),

and its function (e.g., automatic control of valves).

If such switches are not installed in safety related systems and if such i

installations are not planned, licensees are expected to review the information in the Bulletin and consider actions, if appropriate, to preclude similar problems from occurring.

iio specific action or reporting is required at this time.

Actions described below are Required for All Licensees with SOR Series 102 or 103 Switches in Safety Related Systems.

These actions are not required for such switches in non-safety-related systems.

ACTIONS REQUIRED WITHIN 7 DAYS Within 7 days of receipt of this bulletin, advise each licensed reactor operator of this potential problem, including: (a) a description of the LaSalle incident, l

(b) the uses of the same type of switche:: at your plant, (c) methods to detect the occurrence of a similar problem at your plant, and (d) actions the operator should take in the event that such a problem should occur at your plant.

ACTIONS REQUIRED WITHIN 30 DAYS Within 30 days of receipt of this bulletin:

1.

Perform a functional test of each switch in order to determine if the switches function properly or if short term corrective actions are necessary.

Each functional test should be performed at least 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> after the switch is continuously subje::ted to actual conditions which are as close as possible to those experienced during system operation including temperature, static pressure, and increasing or decreasing differential pressure.

If the tests cannot be reasonably performed under actual system operating conditions (e.g. while actually lowering water level in the reactor vessel), then a test rig that simulates system operating pressures may be used.

The functional test of each switch shall determine the "as-found" value of its setpoint which (.ccurs on first actu-ation of the switch while the switch is at system operating conditions.

If a test rig is connected to the switch and if the licensee intends to take credit for system conditions during the 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> prior to connection of the test rig, then care must be taken to preclude inadvertent actuation

~

of the switch without obtaining a measurement of the differential pressure which actually caused actuation.

)

The regulations require that test results which do not comply with limiting conditions for operation in technical specifications be reported to NRC in i

accordance with 10 CFR 50.72 and 50.73.

If an "as-found" value does not comply, then the test must be repeated several times before the switch is recalibrated in order to determine the

~

IEB 86-02 July XX, 1986 Page 7 of 8 effect of time on offset and the effect of sticky operation.

All of these results shall be recordea and reported, If one channel of a system of redundant channels is found to have an actual setpoint which is cutside the allowable values in the technical specifi-cations, then the other redundant channels shall be tested as socn as possible.

Channels that are found t.o be inoperable will not be declared operable on the basis of setpoint readjustment alone.

If the minimum num-ber of channels are not operable, then to exit from the action statement, additional actions must be taken to assure that the system will perform its intended function if called upon to do so.

If a plant shutdown is required, restart shall be approved by the appropriate NRC Regional Administrator.

2.*

Unless more frequent surveillance is required by technical specifications, initiate monthly surveillance testing.

In either case, use the technique described in the action item above.

This surveillance shall ccntinue on a nionthly basis until corrective action identified in Item 2 of Actions Required Within 60 Days is fully implemented and demonstrated to be effective.

If other mechanical switches are installed, the licensee must

~.

provide assurance that similar or other problems (e.g. equipment qualifi-cation) are not introduced.

3.

The licensee is to develop and implement a testing method for the two testing requirements above (i.e., the initial test that is to be completed within the target goal of 30 days, and the monthly continua-tion of such testing) that will minimize the potential adverse affects of the testing itself.

Specifically, the licensee shall consider the risks associated with operation of the plant potentially for a signifi-cant fractior. of each month with certain systems important to safety in t

either a half-tripped condition or a bypassed condition (in that their connection to the system is isolated).

ACTIONS REQUIRED WITHIN 60 DAYS Within 60 days of receipt of this bulletin, submit a written report which includes the following:

1.

A list of all 50R Series 102 or 103 differential pressure switches in safety related systems which includes, ir, addition to information previously submitted, the manufacturer's specified range for the switch, the nominal and allowable values of the setpoint as required by the technical specifi-cations and in the same units as given for the manufacturer's specified range, the elevation of the instrument taps for water level applications, known sources of systematic errors (e.g., differences in static head due to different elevations of condensing chamber water levels), and "as-found" and subsequent data which do not conform to requirements in the technical specifications.

2.

A description and the results of an analysis to assess the impact of the worst setpoint discrepancy observed at the licensee's plant for each

• Note: This requirement (repeated montly tests) in being considered further at this time.

IEB 86-02 July XX, 1986 Page 8 of 8 9

general type of switch application (e.g. water level, flow).

The report shall specifically address each switch application that is intended to preserve (a) a safety limit, (b) an acceptance criterion for an accident or transient analysis, or (c) protection for safety related equipment.

The intent of the analysis is to determine if improvements in calibration and testing methods, improvements in setpoint methodology, changes to tech-nical specifications, repair, modification, or replacement, or other improvements are needed in order to meet existing regulatory requirements (e.g. General Design Criterion 21).

The worst observed discrepancy for each type application shall be used to determine the corrective actions for all instruments of that type.

The analysis should demonstrate that the long term corrective action will provide an adequate safety margin to assure high functional reliability.

If such improvements are necessary, describe the actions which you intend to take so that your plant is in compliance and the implementation schedule.

Should a licensee determine that any action requested by this bulletin jeopardizes

-. overall plant safety, the NRC should be notified of that fact and provided with appropriate justification for not implementing the requested action.

Such notification shall be made within 7 days of receipt of this bulletin.

The written reports shall be submitted to the appropriate Regional Administrator under oath or affirmation under the provisions of Section 182a of the Atomic Energy Act of 1954, as amended.

Also, the original copy of the cover letters and a copy of the reports shall be transmitted to the U.S. Nuclear Regulatory Commission, Document Control Desk, Washington, DC, 20555 for reproduction and distribution.

The request for information was approved by the Office of Management and Budget under blanket clearance number 3150-0012.

Comments on burden and duplication may be directed to the Office of Management and Budget, Reports Management, Room 3208, New Executive Office Building, Washington, DC, 20503.

If you have any questions regarding this matter, please contact the Regional Administrator of the appropriate NRC Regional Office or one of the technical contacts listed below.

James M. Taylor, Director Office of Inspection and Enforcement Attachments:

1.

Plants with Similar 50R Switches 2.

List of Recently IE Bulletins Technical Contacts:

J. T. Beard, NRR (301) 492-4415 Roger W. Woodruff, IE (301) 492-7205

IEB 86-02 July XX,1986 Page 8 of 8 general type.of switch application (e.g. water level, flow).

The report i

shall specifically address each switch application that is intended to preserve (a) a safety limit, (b) an acceptance criterion for an accident or transient analysis, or (c) protection for safety related equipment.

The intent of the analysis is to determine if improvements in calibration and testing methods, improvements in setpoint methodology, changes to tech-nical specifications, repair, modification, or replacement, or other improvements are needed in order to meet existing regulatory requirements (e.g. General Design Criterion 21). The worst observed discrapancy for each type application shall be used to determine the corrective actions for all instruments of that type.

The analysis should demonstrate that the long term corrective action will provide an adequate safety margin to assure high functional reliability.

If such improvements are necessary, describe the actions which you intend to take so that your plant is in compliance and the implementation schedule.

Should a licensee determine that any action requested by this bulletin jeopardizes overall plant safety, the NRC should be notified of that fact and provided with appropriate justification for not implementing the requested action.

Such notification shall be made within 7 days of receipt of this bulletin.

The written reports shall be submitted to the appropriate Regional Administrator under oath or affirmation under the provisions of Section 182a of the Atomic Energy Act of 1954, as amended.

Also, the original copy of the cover letters and a copy of the reports shall be transmitted to the U.S. Nuclear Regulatory Commission, Document Control Desk, Washington, DC, 20555 for reproduction and distribution.

The request for information was approved by the Office of Management and Budget under blanket clearance number 3150-0012.

Comments on burden and duplication may be directed to the Office of Management and Budget, Reports Management, Room 3208, New Executive Office Building, Washington, DC, 20503.

If you have any questions regarding this matter, please contact the Regional Administrator of the appropriate NRC Regional Office or one of the technical contacts listed below.

James M. Taylor, Director Office of Inspection and Enforcement Attachments:

1.

Plants with Similar SOR Switches 2.

List of Recently IE Bulletins Technical Contacts:

J. T. Beard, NRR (301) 492-4415 Roger W. Woodruff, IE (301) 492-7205 D:DEPER:IE D:IE ELJordan JMTaylor 7/ /86 7/ /86 DEPER:IE NRR DEPER:IE DEPER:IE RII DD:DEPER:IE RWWoodruff JTBeard RLBaer DPAllison CENorelius SASchwartz 7/8/86:jr 7/ /86 7/ /86 7/ /86 7/ /86 7/ /86

Attachm:nt 1 IEB 86-02 July XX, 1986 9

PLANTS WITH SERIES 102 OR 103 DIFFERENTIAL PRESSURE SWITCHES Series 102:

Florida Power and Light Series 103:

Commonwealth Edison General Public Utilities - Nuclear Corporation Houston Lighting & Power Company Northeast Utilities Pennsylvania Power & Light Southern California Edison Tennessee Valley Authority Washington Public Power Supply System 1

b I

a IEB 86-02 July XX, 1986 LIST 0) RECENTLY ISSUED IE BULLETINS Bulletin Date of No.

Subject Issue Issued to 86-01 Minimum Flow Logic Problems 5/23/86 All GE BWR facilities That Could Disable RHR Pumps holding an OL or CP 85-03 Motor-0perated Valve Common - 11/15/85 All power reactor Mode Failures During Plant facilities holding Transients Due To Improper an OL or CP for Switch Settings action 85-02 Undervoltage Trip 11/5/85 All power reactor Attachments Of Westinghouse licensees and DB-50 Type Reactor Trip applicants Breakers

', 85-01 Steam Binding Of Auxiliary 10/29/85 Nuclear power facil-Feedwater Pumps ities and cps listed in attachment 1 for action; all other nuclear power facil-ities for information 84-03 Refueling Cavity Water Seal 9/24/84 All power reactor facilities holding an OL or CP except Ft. St. Vrain 84-02 Failures Of General Electirc 3/12/84 All power reactor Type HFA Relays In Use In facilities holding Class 1E Safety Systems an OL or CP 84-01 Cracks In Boiling Water 2/3/84 All BWR facilities Reactor Mark I Containment with Mark I contain-Vent Headers ment and currently in cold shutdown with an OL for action and all other BWRs with an OL or CP for information i

l l

OL = Operating License CP = Construction Permit

__.______