ML20137J886
| ML20137J886 | |
| Person / Time | |
|---|---|
| Site: | Hatch  |
| Issue date: | 01/13/1986 |
| From: | Gucwa L GEORGIA POWER CO. |
| To: | NRC OFFICE OF ADMINISTRATION (ADM) |
| References | |
| 0244C, 244C, SL-220, TAC-61074, NUDOCS 8601230267 | |
| Download: ML20137J886 (3) | |
Text
m_
Georg>a Power Company 3;.s P edmer11 Avenue At: acta Georg a 33v's Teep %1e 404 526 CS26 Mad rg Aca ess Pat Ottce Dex 4545 At: art t. Gecrga 30302' b
Georgia Power
' "" ' fM
- N W "'
L. T. Gucwa Mn;p N;ciear Lng reer. g ed Cr, cf NucMar Enreer SL-220 0244C January 13, 1986 U. S. Nuclear Regulatory Commission Document Control Desk Washington, D. C.
20555 Attached is a report on the failures of the Main Steam Isolation Valve (MSIV) stems which took place on July 24, 1985 and December 9, 1985 at Plant Hatch Unit 1.
This submittal is for information only since the subject event is not reportable per 10 CFR 50.72, 50.73 or the plant i
Technical Specifications.
Very truly yours, LT Cm U L. T. Gucwa CBS/lc Attachment c:
Mr. J. T. Beckham, Jr.
Mr. H. C. Nix, Jr.
NRC-Region II G0-NORMS 8601230267 860113 PDR ADOCK 05000321 s
s
- PDR, 1 %
\\
Plant Hatch Unit 1 Failures of MSIV Stems (Page 1 of 2)
On 07/24/85, at approximately 1328 CDT, Plant Hatch Unit 1 was operating at 100% power when a reactor scram occured. As described in Licensee Event Report (LER) 50-321/1985-027, the cause of this scram was clomre of the "A"
Inboard Main Steam Isolation Valve (MSIV)
(plant identification nlmber B21-F022A) following a failure of its stem.
Initial investigation of this event determined that the stem had coupletely severed due to fatigue from excess cyclical vibration.
The EIV vendor (Atwood and Morrill Company) attrituted these excessive vibratory stresses to improper backseating of the valve when it was in the open position.
As a remit of these findings the plant procedure used for repacking and adjustment of the MSIVs (5204-bNE-001-0) was schedaled for revision to include the vendors instructions for properly setting the backseating on these valves.
Prior to restart of Unit 1, following the occurrence of the subject scram, the failed stem in B21-F022A was replaced, and each of the other seven MSIVs were inspected for indications of similar problems.
All eight MSIVs were then properly backseated (under mpervision of a
vendor representative) and tested prior to their return to service.
On 08/29/85, plant management approved a redesign of the stems on all Unit 1 EIVs following meetings with vendor engineering personnel to discuss the mbject event.
%e new design included an increase in valve stem diameter (from 1.75 to 2.00 inches) as well as modifications to the backseat undercut.
Both vendor and utility personnel had determined that these modifications wo2ld reduce the localized alternating stresses in the MSIV stems, thereby increasing their design lives.
These modifications were sched21ed for irplementation daring the Unit 1 refueling oatage which was planned to coamence on 11/30/85.
%e inportance of proper setting of the backseating by utility personnel was reemphasized daring those meetings.
On 12/09/85, with Unit 1 in cold shutdown for a sched21ed refueling oatage, maintenance personnel were disassembling MSIV B21-F022A when the stem again broke.
Subseolent irspection revealed that the severance of the value stem was identical to the failu re which occurred on 07/24/85.
Inspections of the seven other Unit 1 MSIVs revealed a similar nearly complete stem failure on valve B21-F022B as well as minor dye-penetrant test indications on valves B21-F022D and B21-F028D.
Onsite engineering initiated another evaluation of the ca2ses of these MSIV failures and indications.
%is evaluation determined that the setting of tne backseating had apparently been 01/13/86 0942y
Plant Hatch Unit 1 Failures of MSIV Stems (Page 2 of 2) performed properly on these valves since the time of occurrence of the first event, and was therefore probably not the ca2 se.
The investigation did identify severe wearing of the guide ribs inside the valve bodies of both B21-F022A and B.
'Ihese ribs guide the valve poppet (seat) to its proper position on the seating ring & ring an MSIV closure.
The evaluation concluded that this guide rib wearing was ultimately responsible for the degradation and failure of the MSIV stems, since this condition wo21d allow excess vibration of the poppet to occur during normal operations.
It was also concluded that inplementation of the above referenced valve stem redesign is neccessary to reduce fatigue stress to acceptable levels.
A modification of the MSIVs which was performed in 1984 to re&ce leakage by preventing poppet rotation was also identified as a potential cause, however the valve vendor has stated that its affect on stem stress is minimal if proper backseating techninaes are in use.
To prevent future recurrence of these types of events site management has initiated in-depth evaluations of the causes for these EIV stem problems by independent offsite organizations with expertise in failure and stress analysis.
Battelle Inc., Columbus Division is conicting a detailed engineering evaluation to determine if there are I
any further unidentified root causes for the subject occurrences.
Structural Integrity, Inc. is modeling the detailed sealence of events which is believed to remlt in ultimate valve stem failure.
Plant Hatch engineering will review the remits of these offsite evaluations and factor their results into the final long term corrective action plan for the s2bject occurrences.
Prior to startup from the Unit 1 refueling outage currently in progress, the above referenced replacement of the valve stems on all eight EIVs will be completed.
In addition, the severely worn guide ribs on B21-F022A and B will be reconditioned to bring those 4
components back within the vendor's specified tolerances prior to plant restart.
Inplementation of the above noted revisions to proce& re 52CM-ME-001-0 will also be conpleted within that time frame.
'Itese corrective actions should be m fficient to at least i
prevent stem failure recurrence & ring the next cycle of Unit 1 operation.
J It has been determined by onsite engineering that the subject types of I
MSIV stem failures would not prevent closure of these valves on demand, and that the only unexpected change in valve position &e to mch a failure wo21d be in the conservative (i.e. closed) position.
Therefore, these failures have not had a detrimental effect on plant safety or the health and safety of the general p2blic.
01/13/86 0942y
. _.