ML20072A787

Initial Part 21 Rept 038 Re Failure of Gimpel Trip Throttle Valve to Open During Testing of AFW Pump Turbine at Toledo Edison.Subj Valve Stems Should Be Tagged W/Instructions Defining Proper Procedures for Securing Stem to Coupling

ML20072A787
Person / Time
Site:	Davis Besse
Issue date:	08/05/1994
From:	Loughlin R DRESSER-RAND CO.
To:	NRC
References
REF-PT21-94 038, 38, NUDOCS 9408150215
Download: ML20072A787 (3)

Text

i AJG 05 '94 10:49AM ENGINEERING DEPT. P.1 a-FAX TRANSMITTAL 4

DRESSER-RAND STEAM TURBINE, MOTOR & GENERATOR DIVISION WELLSVILLE, NEW YORK Robert M. Loughlin Tel. 716-596-3343 Fax. 716-596-3753 Date: August 5, 1994 To: NRC Operations Center Fax No. 301-816-5151

' ~

From: Robert M. Loughlin

Subject:

- " Verbal" Notification by facsimile of a Potential '~~

Safety Hazard - 10CFR Part 21 Initial Report No. 038 A copy of approved 10CFR Part 21 Initial Report No. 038 is provided as pages 2 and 3 of this facsimile transmittal (complete transmittal is 3 pages).

Dresser-Rand Steam Turbine, Motor & Generator Division will submit a written Final Report no later than September 2, 1994.

Questions relating to this 10CFR Part 21 Report may be directed to the undersigned.

Robert . Loughlin Manager - Nuclear and Navy Product Engineering i

a **

w s w' v L- ..y 9408150215 940805 PDR ADOCK 05000346 S PDR g\'

AUG 05 '94 10

50Ar1 Et4GItEERIt4G DEPT. P.2 NNSOP 1.2.001, Rev. 2, Page 7 of 9 Fenn EF 065 REPORT NO. 038 Riv B

, Nuc!Iat NNSO'P 1.2.001 INITIAL REPORT 10CFR PART 21 '

y REPORT OF A POTENTIAL SAFETY HAZARD .

Date: July 26, 1994

1. lSSUED 70: y 97aggA/p Mbert Irughlin Mgr. Navy /Nucicar Eng. Fl!e No.:

_ Richard KlinR Nuc! car Product Eng. Serial No.: T37H86A/B Btrold Paris QA Manager Type: GS2N

'V ^

P. J . Meann VP Engineering R M"-

John Gems -- VP Engineered Products SBU N/k Part No.:

Peter Salvatore VP Parts and Servico Business Unit. DRSD Dwg. No./ Gin 9el P-2989 Rev. Level: _

P.1 chard Klint PREPARED BY: .-

TITLE: Supv. N/N Prod. Eng - i PART NAME: Gimoel - Trip fc Thrcttle Valve ,

DESCRIPTION OF DEFECT:

'Daring the testing of the aux. feed ptr4: turbine (AFPT) No. 1 at Toledo Edison the Gimpel_ trip throttle _ valve would not onen. The vnive indicated 28 percent open in the trioned cosition, but in reality the valve was__ closed _

0 rercent ot>2n_ and the valve stem had unthreaded itsal f fw tha cruol i ner.

l This prevented the valve frern opening.

2. ENGINEERING EVALUATION & RECOMMENDATION: l See attached page.

Evaluation /Recommendatlon Prepared By:

Title:

Date: .

. Approved by Mgr. Navy / Nuclear Product Engineering: _Date:

3. DISPO ITION, CHECK CNE ,

V Yes, this constitutes a safety hazard and requires that a report be prepared (Form NN-0002)

No, this does not constitute a safety hazard and does et require any further reporting Disposition by D.R President:

/

w Dato: 5 b94 o n-RETURN TO MOR NAVY & NUC1. EAR PRODUCT ENGINEERING

7 AUG 05 '94 10:50Af1 ENGINEERING DEPT. P.3 l

2. ENGINEERING EVALUATION & RECOMMENDATION:

.To 'our knowledge this experience at Toledo Edison represents the first time that this defect has been encountered, and it seems probable that it is an isolated incident. However, as the incident illustrated, if a valve stem coupling is not adequately secured to the trip throttle valve stem, the valve stem can unthread from the coupling and prevent the valve from opening either completely or partially. In this condition the turbi.no is unable to operate as required to perform its intended safety function.

By design secure attachment of the coupling to the stem is assured by proper installation of a setscrew which is inserted through a threaded hole in the coupling and seated firmly against a drilled indentation

(.25 diameter x .125 deep) or machined flat (.25 wide x .125 deep) on the stem. Since proper assembly requires adjustment of.the stam position within the coupling, the indentation or flat can be machined on the stem only after proper location is determined during assembly.

In this case at Toledo Edison there was no evidence that any indentation or flat had been machined during the assembly of the coupling and stem. Failure to include this step"in' the ass'aintily'is believed to be the reason, that the stem was able-to' aventually.~r-r- ~m - --

unthread from the coupling, i

Based on the information relating to the Toledo Edison incident that has boon mado available to us and on initial discussions with,Gi.gpel_ __._ ___. _

"(the~ trip valve' manufacturer) and with Dresser-Rand Quality Assurance personnel, Product Engineering has the following recensendations.

1. This defect does constitute a safety hazard and requires that a report be prepared.

2. All trip throttle valve stems supplied in the future by Dresser-Rand should be tagged with instructions defining proper proceduras for securing the valve stem to the coupling.

3. All trip throttle valve stem couplings supplied in the future by Dresser-Rand should also be tagged per item 2 above.

4 The dedication procedure for these couplings should be enhanced to include confirmation that the setscrew hole is properly tapped fully through.

5. Dresser-Rand Quality Assurance should implement a dedication practice for any complete valve assemblies purchased from Gimpel to provide objective confirmation that the stem and coupling assembly procedure is fully and properly carried out.

6. Nuclear customers having these Gispel trip valves should be made aware of this potential defect and advised to confirm appropriate assembly at first convenient opportunity.

Evaluation /Recounendation Prepared by: Titles b Dates Approved by Nor.-Navy / Nuclear Product Engineering -

/ -

g-

- Date: < j

!b j