ML20217D488
| ML20217D488 | |
| Person / Time | |
|---|---|
| Site: | Paducah Gaseous Diffusion Plant |
| Issue date: | 09/30/1997 |
| From: | Polston S UNITED STATES ENRICHMENT CORP. (USEC) |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| GDP-97-1033, NUDOCS 9710030243 | |
| Download: ML20217D488 (7) | |
Text
_ - _ - _ - _ _
Umted Stain Enrkhment Corsvatmn Paducah Site Offa P O ikin 1410 Paducah, KY 42001 id 502 44l 5803 Ik 502 44l *BOL September 30,1997 United States Nuclear Regulatory Commission SERIAL: GDP 97-1033 Attention: Document Control Desk Washington, DC 20555 001 Paducah Gaseous Dih61on Plant (PGDP). Docket No. 70 *,001 - Event Report ER-97-15 Pursuant to 10 CFR 76.120(c)(2), attached is the required 30-day written report for the steam leak at Autoclave 2 South in C-333-A. This was initially reported on September 1,1997 (NRC No. 32852). is a list of commitments made in this report.
Should you require further information on this subject, please contact Bill Sykes at (502) 441-6796.
Sincerely, L
1-J Steve Polston General Manager Paducah Gaseous Diffusion Plant SP:WES:JNH:mel Attachments (2)
/
cc:
NRC Region ill NRC Senior Resident Inspector, PGDP
/
9710030243 970930 ADOCK0700g1 PDR i
-j f)
.,q
Docket No. 70-7001
' Attach' ment 1 Page1of5 Event Report ER-97-15 BACKGROUND The autoclaves in the C-333 A facility are used to steam heat cylinders of UF. in order to feed the UF. to the cascade. The autoclave also provides containment around the cylinder in the event of a release of UF. from the cylinder or cylinder connections during the feeding of the cylinder. A section of the autoclave, the head, is mounted permanently, while the remaining section, the shell, is moved with a hydraulic ram on a track system. This movement is required to permit introduction and removal of the UF cylinders. The moveable shell of each autoclave is secured to the fixed head with a hydraulically operated flanged closure, which is sealed with an 0-ring gasket. The final sealing of the flanged closure is accomplished by a locking ring, which rotates and locks the autoclave shut.
A limit switch controls the hydraulic ram which brings the autoclave shell toward the head and stops the shell motion when the seal faces are in close proximity. Once the seal faces are brought together, this limit s. itch activates two additional hydraulic rams that work in concert to rotate the locking ring. The locking ring will rotate until a second limit switch deactivates the locking ring hydraulic rams, if either of these limit switches is out of adjustment, a tight seal between the head and shcIl may not be established.
DESCRIPTION OF EVENT 4
On August 31,1997, at 2245, aperators in C-333A Feed Facility observed a small quantity of steam leaking from the autoclave head-to-shell gasket area on Autoclave 2 South. The autoclave
- had been heating a feed cylinder for approximately 30 minutes (mode 5) when the leak was
- discovered. The autoclave was placed in mode 2 (autoclave open) at 2250 in accordance with Technical Safety Requirement (TSR) Limiting Condition for Operation (LCO) Action Step 2.2.3.1.C. Per TSR LCO Action Step 2.2.4.11.A, the operating cycle was terminated; heat was removed from the autoclave; and the cylinder was cooled for five days prior to moving it.
Additionally, as required by TSR LCO 2.2.4.11.A, the cylinder will not be reheated for an additional forty-day period. The cylinder was caution tagged and will be administratively controlled umil the forty-day period has been completed. Pursuant to 10 CFR 76.120(c), a notification of the event was made to NRC on September 1,1997, at 0928 (refer to NRC Event Worksheet No. 32852).
On September 3,1997, the system engineer and an operations manager conducted an inspection of Autoclave 2 South. The inspection revealed that limit switch ZSL540 for Autoclave 2 South
J Docket No. 70-7001
' Attachment 1 Page 2 of 5 i
l had a broken guide wheel. The break in the guide.vheel appeared to be fairly recent, based on L
the absence of dirt or grime on the surface of the break. This limit switch con'rols the amount L
- that the locking ring will rotate. If the limit switch is out of adjustment or cannot perform i
properly, the locking ring could stop rotating before an adequate seal is established. The observed break in the guide wheel could cause the limit switch to inader,uately rotate the locking ring, thereby allowing a steam leak at the autoclave head-to-shell gasket from an incomplete seal.
Further investigation revealed that the nozzle on the articulating arm of the capture ventilation system for Autoclave 2 South had a crinkle in the metal which matched the break on the guide
-wheel. The nozzle and associated hose is suspended above the autoclave head by the articulating L
arm. The arm allows movement of the hose and its aluminum nozzle to capture small residual amounts of UF left in the pigtail. The nozzle appears to have been inadvertently caught 6
between the locking ring limit switch block and the limit switch itself, causing the guide wheel to break.
- An additional problem was observed that could impact the alignment of the autoclave and/or the compression of the 0-ring. The hydraulic mechanisni which closes the shell is designed to slow -
the shell movement as the shell comes in close proximity to the head and then stops the autoctave movement. Observation of the 2 South autoclave closing operation indicated that the hydraulic mechanism was not consistently operating as designed to slow the shell as it closed the autoclave. The hydraulic mechanism was allowing the shell to hit the head with an excessive force.
CAUSES OF EVENT A.
Direct Cause One of the direct causes of the steam leak was that the locking ring was likely not rotated far enough to establish an adequate seal. This was due to the broken limit switch which controlled the amour 2 oflocking ring rotation. The limit switch was replaced and an aligninent test was conducted according to procedure CP4-GP-MM4159," Alignment Check of UF. Autoclave Head-to-Shell and 0-ring Replacement." The gap between the -
head and shell had changed due to the broken and maladjusted limit switch. Replacement and adjustment of the limit switch, under CP4-GP-MM4159, corrected the gap measurement within tolerances. System Engineering inspected the limit switches on other autoclaves and determined that no other limit switches were similarly damaged.
A second direct cause of the steam leak is the failure of the hydraulic mechanism to slow the movement of the shell as the shcIl approached the head. The resulting excessive force of the shell against the head could accelerate the compression of the 0-ring and can also
Docket No. 70 7001
' Attacliment 1 Page 3 of 5 i
cause the head-to-shell interface to become misaligned. Misalignment and 0-ring i
compression are both factors which can contribute to an inadequate seal. This condition was corrected by cleaning the hydraulic switch which controls the process of slowing the shell movement.' Inspection of the 0-ring showed only ordinary degradation due to
- normal operations. Based on a review of other operating autoclaves by Operations, no other operating autoclaves are exhibiting this same problem. Corrective Action No. 5 will revise procedure CP4-CO-CN2045a," Operation of the C-333A and C-337A
- Vaporizer Facilities," to add an action step to require operators to monitor the slowdown j
speed of the autoclave during cylinder changes and take corrective actions if the i
autoclave is closing abnormally.
B.
Root Cause j-l-
Root cause of the broken limit switch relates to the design of the artictJating arms of the capture ventilation system, which was installed several menths ago. The design of the articulating arm does not prohibit the nozzle from inadvertently coming in contact with and damaging the limit switch because it is necessary to place the nozzle as close as possible to the pigtail connection to minimize release of residual UF.. ' However, the arm frequently gets out of adjustment and, as a result, will not hold a suspended position and falls to rest whenever released. A similar concern which is related to the design of the articulating arm was submitted in May of 1997 via Problem Report PR-CO-97-2685.
This documents the destruction of an articulating arm end which was closed in an--
autoclave. The response to that problem report was to repair the hose erm and to discuss i
" lack of attention to oetail" with personnel. The design of the arm was not addressed
- through that problem report. Under Corrective Action No. 2, Engineering will evaluate the current design of the articulating arms in C-333 A and C-337A and develop an action plan as appropriate for improvements or substitutions that could alleviate problems -
- similar to this event.
C.
Contributing Causes i A contributing cause is failure to ensure there are no obstructions near the limit switches
. prior to closing the autoclave. Corrective Actions No. 3 and 4 will address the generic issue of obstructions damaging the autoclave by adding an action step to applicable -
- procedures to check for obstructions prior to closing the autoclave.
Another possible contributing cause is related to the locking ring wedges. Event Report ER-96-66, issued under DOE regulation, identified a problem with normal wear of locking ring wedges which are used in conjunction with a knife switch to limit the slippage of the autoclave locking ring and subsequent relaxation of the 0-ring. The
a'
- j s
- j i
Docket No,70-7001
[
Attacliment 1 Page 4 of 5
)
- report included a corrective action to relocate or change the wedges to ensure appropriate
)
_ anti-rotation limits are met. Per this action, autoclaves are to be refurbished, as required, i
by October 30,1997. Refurbishment of Autoclave 2 South has not'yet been completed.
l This may have contributed to the event by allowing the knife switch to be engaged even L
though the autoclave shell and head was not sufficiently closed.
Corrective Action No. I will review this event with C-333A, C-337A, and C-360 qualified operators through the required reading program.
CORRECTIVE ACTIONS A.
Completed Corrective Actions 1.
On September 6,1997, Maintenance replaced limit switch ZSL540 on Autoclave 7 South in C-333A.
- =
l 2.
On September 8,1997, Maintenance repaired the shell closing problem on Autoclave 2 South in C-333A by cleaning the 1.ydraulic switch, t-f
'3.
On September 26,1997, Operations issued Long Term Orders (LTO) to require -
operators to monitor slowdown speed of autoclaves during cylinder changes and-log if the autoclave closes abnormally so that repairs can be made.- The LTO also -
requires a check for material deficiencies and obstructions prior to closing th:
. autoclave.
B.
Corrective Actions Planned 1.
By November 21,1997, Operations will review the event report with C-333A, C-337A, and C-360 qualified operators through the required reading program.
2.
_ By December 15,1997, Engineering will evalufe the current design of the articulating ann for the NAM in C-333A and C-337A and develop an action plan as appropriate for improvements that alleviate problems similar to this event (ER-97-15).
3.
By January 23, '1998, Operations will identify procedures which require them to close the autoclave and will revise these procedures to include an action step to check for obstructions prior to closing.
,p p,-
-,,-q_e3
=r4m- + - - +
44 64
- Docket No. 70-7001 AttacIlment 1 Page 5 of 5 4.~
By January 23,1998, Maintenance will identify procedures which require them to close the autoclave and will revise these procedures to include an action step to check for obstructions prior to closing.
5.
By January 23,1998, Operations will revise CP4-CO-CN2045a to monitor the slow down speed of autoclave closing during cylinder changes and to take corrective actions if the autoclave closes abnormally.
EXTENT OF EXPOSURE OF INDIVIDUALS TO RADIATION OR TO RADIOACTIVE
-MATERIALS There was no exposure ofindividuals to radiation or to radioactive materials as a result of the steam leak.
- LESSONS I RARNED (1) Equipment design should consider the operating environment and include, where possible, features that prevent equipment damage. (2) Prior to closing the autoclave, personnel should i
ensure that there are no obstructions to closure, s
._ __. ~.._. - - - _ _ _ _. _. _... _,.
1
- Docket No. 70-7001 Page1ofI 1.
ER-97-15 Corrective Actions Planned 1.
By November 21,1997, Operations will review the event report with C-333 A, C-337A, and C-360 qualified operators through the required reading program.
2.
By December 15,1997, Engineering will evaluate the current design of the articulating arm for the NAM in C-333A and C-337A and develop an action plan as appropriate for improvements that alleviate problems similar to this event (ER-97-15).
3, By January 23,1998, Operations will identify procedures which require them to close the autoclave and will revise these procedures to include an action step to check for obstructions prior to closing.
4.
By January 23,1998, Maintenance will identify procedures which require them to close the autoclave and will revise these procedures to include an action step to check for obstructions prior to closing.
5.
By January 23,1998, Operations will revise CP4-CO-CN2045a to monitor the slow down speed of autoclave closing during cylinder changes and to take corrective actions if the autoclave closes abnormally, h