PLA-1872, Second Interim Deficiency Rept Re GE Reactor Mode Switch. Initially Reported on 830407.Mode Switch Redesigned.Sample Will Be Independently Tested.Also Reported Per Part 21.Final Rept Will Be Submitted by Dec 1983

Second Interim Deficiency Rept Re GE Reactor Mode Switch. Initially Reported on 830407.Mode Switch Redesigned.Sample Will Be Independently Tested.Also Reported Per Part 21.Final Rept Will Be Submitted by Dec 1983

ML20090H327
Person / Time
Site:	Susquehanna, North Anna
Issue date:	09/30/1983
From:	Curtis N PENNSYLVANIA POWER & LIGHT CO.
To:	Murley T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
References
REF-PT21-83 10CFR-050.55E, 10CFR-50.55E, PLA-1872, NUDOCS 8310280065
Download: ML20090H327 (6)

Text

-

,J.~ ,

~

@ Pennsylvania Power & Light Company

- Two North Ninth Street

• Allentown, PA 18101 + 215 / 770-5151 Norman W. Curtis Vice President-Engineering & Construction-Nuclear

'215/770-7501-September 30, 1983 Dr. Thmas E. Murley Regional Administrator, Region I U.S. Nuclear Regulatory Ccmaission 631 Park Avenue 4~

King of Prussia, PA 19406 SUSQUEHANNA STEAM ELECTRIC STATION SECOND INPERIM REPORP OF A DEFICIENCY INVOLVING GE REACIOR MODE SWITCH ER 100508 FILE 821-10 PIA-1872 Docket No. 50-388

Reference:

PIA-1652 dated May 6, 1983

Dear Dr. Murley:

This letter serves to provide the Ccmuission with an interim report on a deficiency involving the Unit 2 Reactor Mode Switch. This deficiency was originally reported by telephone to Mr. D. Johnson of NRC Region I on April 7, 1983, by Mr. J. Saranga of PP&L as potentially reportable under the provisions of 10CFR50.55(e) for Unit 2. The referenced PIA-1652 provided the Ccmnission with an interim report which included the results of the tests and analyses performed by Franklin Institute Research Laboratory on the mode switch which failed in Unit 1 (Unit 1 failure was reported as LER #83-043) .

The attachment to this letter contains a description of the deficiency, its safety impact, and the corrective action plan. PP&L anticipates providing the Ca mission with a final report in December, 1983.

Since the details of this report provide information relevant to the reporting requirements of 10CFR21 for Unit 2, this correspondence is considered to also discharge any formal responsibility PP&L may have in ccmpliance thereto.

1 8310280065 030930 5 7 DRADOCK05000g

p -g * .

.i .

fSeptember 30,-1983 ' SSES PIA-1872 ER 100508 File 821-10 We trust the Camission will~ find this report to be satisfactory.

Very truly yours,.

N. . Curtis Vice President-Engineering & Construction-Nuclear-JS:sab Attachment js/lt/1/145/a

(

.. . .: l

, I

September.30,-1983 - 3'- .SSES . PIA-1872

.ER 100508: File 821-10 Copy to:. _ .

Mr. Richard C. DeYoung (15)

Director-Office of Inspection & Enforcement U.S. Nuclear Regulatory Camission

, Washington,- D.C. 20555

.Mr. G. Mcdonald, Director Office of Management Information & Program Control

'U.S. Nuclear Regulatory Camission Washington, D.C. 20555 Mr. Cary Rhoads U.S. Nuclear Regulatory Cm mission P.O. Box 52 Shickshinny, PA 18655 Records Center Institute of Nuclear Power Operations 1100 Circle 75 Parkway, Suite 1500 Atlanta, GA 30339

(

_ a

y. .,: , r w _ _., _

5% u '< " /

3L, ~ -

f- .

~

. "!Atthchment ' SSES ' PIA-1872 '

-Page 1.of'3 ER 100508 File 821-10~

h; ,

+

A SECCND INTERIM REPORP ns.

SUBJECT Deficiencies detected in the Unit 1 Reactor Mode Switch when it failed-to

, actuate a channel of-the RPS as designed.

DESCRIPTICN/EXTINf CF DEFICIENCY

~

The original Unit.l_ Reactor Mode Switch failed.to actuate RPS Channel B when moved frcm STARIUP to SHl1IDG4N. A bench check at SSES of a replacement switch Lindicated intermittent operation of this second switch. After bench testing at GE four additional switches were received at the site and-successfully-bench tested. One of these was installed in Unit-1. The original switch was then examined by Franklin Institute Research Laboratory. 'Ihis examination

. indicated a ntuber of design and manufacturing deficiencies. 'Ihese included:

1) Cams were made of softer material than the cam follower, leading to possible wear concerns.

2) Cams that performed identical functions did not have identical shapes.

3) In the transition frm metal key shaft to plastic cam shaft, there was not a tight fit, resulting in wear on the plastic transition piece, and switch handle movement without identical plastic cam shaft 4

novement.

i 4)- 'Ihe cam shaft was made of sey=-dwd plastic pieces that fit together.

3 As a result of tolerances this allowed a difference in cam shaft rotation frm front to back.

L

~

!. 5) The cam followers rode in a relatively wide set of guides. This can L result in the cam follower cocking. The contact bar is attached directly to the cam follower. The cocking of the cam follower can

, result in a cocking of the contact bar and can result in incorrect contact position.

V i

Attachment- SSES PIA-1872 Page~2-of 3 ER 100508 File 821-10

6) 'Ihe wire tenninal point is a direct' extension of the stationary contact. . bbvement of the wires can, if the contact is held in place loosely, .(due to tolerance build-up) cause the stationary contact to move.

A mode switch of the same design as the original Unit 1 mode switch was intended for use in Unit 2..

SAFEIT DIAIDATIN This-failure represents a significant deficiency in the design and manufacturing of the' Reactor Mode Switch which, if it is not corrected or otherwise accounted for, could adversely affect the safety of the plant by either failing to scram the plan manually when the switch is moved to SHUPDOWN or.by failing to clear a bypass when switch position is changed.

CORRECPIVE ACTIWS (UNIT 2)

Initially GE partially redesigned the mode switch. Testing based on the switch' specification, the test laboratory findings and previous testing indicated that the redesign failed to meet a design requirement (i.e. contact make before break). As a result PP&L has undertaken two parallel courses of action.

(a) GE Switch Redesign GE, in cooperation with Gould Rundel, has redesigned the mode switch to correct deficiencies 1 through 5 identified above.

Deficiency 6 has been judged to be minor by GE. The positive benefit (ie. a slight wiping of the contacts to help maintain good surface to surface contact) outweighs any perceived negative impact. Prototypes of the new design have been successfully tested by GE (witnessed by PP&L) .

It is PP&L's intention to take a sanple frcan the next production lot (delivery tentatively scheduled for 10/7/83) and have this tested by an independent testing laboratory (completion scheduled for 12/1/83) to dcmonstrate: correction of previous deficiencies,- lack of new problems, and adequate switch life.

y o., ?

.2

.3.

, f' '

'__Y s ' Attaclument : -

y :SSES PIA-1872.

w ' ER 100508 File 821-10

~Page 3 of 3

- (b). Alternate Switch Design.

~

A- . _

PP&L has-sent out for' bids a specification based on'the original -1 mode switch specification, the specific conditions of operation-at Susquehanna SES and standard switch requirements. Due to the requirements for testing and doctznentation-for a new class IE

. design,.present schedules call for availability no earlier than 12/15/83.: