ML20155F518
| ML20155F518 | |
| Person / Time | |
|---|---|
| Site: | Paducah Gaseous Diffusion Plant |
| Issue date: | 10/30/1998 |
| From: | Toelle S UNITED STATES ENRICHMENT CORP. (USEC) |
| To: | Paperiello C NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS) |
| References | |
| GDP-98-0224, GDP-98-224, NUDOCS 9811060055 | |
| Download: ML20155F518 (6) | |
Text
q.
]
9 u
j Go i
USEC cj.
A Global Energy Company g
1 October 30,1998 i
GDP 98-0224 a
i Dr. Carl J. Paperiello
- Director, Office ofNuclear Material Safety and Safeguards Attention: Document Control Desk U. S. Nuclear Regulatory Commission Washington, D.C. 20555-0001
~ Paducah Gaseous Diffusion Plant (PGDP)
Docket No. 70-7001 Safety Analysis Report Update-Testing of the Feed Autoclave High Pressure Isolation Valves
Dear Dr. Paperiello:
i By letter dated October 31,1997 (see the reference), USEC submitted a certificate amendment request containing the Safety Analysis Report Update (SARUP) required by Issue 2 of DOE /OR-2026, " Plan for Achieving Compliance with NRC Regulations at the Paducah Gaseous Diffusion Plant" (the Compliance Plan), for NRC review and approval.1":RC's October 31,1997 letter contained the following commitment:
The feed autoclave high pressure isolation system in C-333 A and C-337A contains isolation valves that utilize plant air to assist in closure. A valve closure time of 15 seconds is assumed in the SARUP accident analysis. The functional requirement in the DOE SAR Upgrade specifies that the system be capable of accomplishing the
/
required safety function independent of plant air; however, valve closure presently relies on plant air. The isolation valves fail in the safe position (closed) on loss of plant air, but valve closure times may be affected due to loss of the air assist. Current
/
valve testing relies on the air assist to determine if the system can isolate within the assumed time.
The feed autoclave isolation system valves will be tested without utilizing the plant air assist to determine the valve closure times. If the valves do not meet the required a
closure times without the air assist, a plant modification to modify the existing valves,
~
or replace the existing valves with faster operating valves, or other appropriate corrective actions will be prepared. The results of the valve testing, along with j
recommended corrective actions and an associated implementation schedule, will be submitted to the NRC by October 31,1998.
9811060055 981030 J
j PDR ADOCK 07007001 J l
C PDR
,a 6903 Rockledge Drive, Bethesda, MD 20817-1818 Telephone 301-564-3200 Fax 301564-3201 http://www.usec.com M02m
-a Omces in Livermore, CA Paducah, KY Portsmouth, OH Washington, DC
..-1..
m.
l Dr. Carl J. Paperiello October 30,1998 GDP 98-0224, Page 2 In accordance with this commitment. tests have been conducted to determine the valve closure times
'without utilizing the plant air assist and an engineering evaluation of the test results has been performed. The results of that eva[uation are summarized in Enclosure 1. Based on the results of the engineering evaluation, interim corrective actions have been implemented and modifications to eliminate the dependence of the feed autoclave high pressure isolation valves on plant air-assisted closure will be implemented by December 10, 2002, 1
Any questions regarding this matter should be directed to Steve Routh at (301) 564-3251.
a Commitments contained in this submittal are identified in Enclosure 2.
Sincerely,-
S. 4.
I Steven A. Toelle Nuclear Regulatory Assurance and Policy Manager
/
}
Reference:
Letter from James H. Miller (USEC) to Dr. Carl J. Paperiello (NRC), " Certificate Amendment Request - Update the Application Safety Analysis Report," Paducah Gaseous Diffusion Plant, Letter No. GDP 97-0188, October 31,1997.
Enclosures:
1.
Evaluation oftest Results for Feed Autoclave High Pressure Isolation Valves 2.
Commitments Contained in this Submittal Y
cc:
Mr. Robert C. Pierson, NRC-HQ NRC Region III Office NRC Resident Inspector - PGDP NRC Resident Inspector - PORTS a
4 1
4 E
.l
- ,1 -
4 J) e 9
1 vme re-m -
ne, --
u-~~
---n,-
x---
n-
,--in,---
O q
Enclosure I h
GDP 98-0224 Page1of3
[. -
Evaluation of Test Results for Feed Autoclave High Pressure Isolation Valves
. SARUP Commitment
,.j Item 11 of Table 1 of USEC Letter GDP 97-0188 dated October 31,1998 states:
y Testing of the Feed Autoclave High Pressure Isolation Valves 3
The feed autoclave high pressure isolation system in C-333 A and C-337A contains isolation valves that utilize plant air to assist in closure. A valve closure time of 15 seconds is assumed in the SARUP accident analysis. The functional requirement in the DOE SAR Upgrade specifies that the system be capable of accomplishing the required safety function independent of plant air; however, valve closure presently relies on plant air. The isolation valves fail in the safe position (closed) on loss of plant air, but valve closure times may be affected due to loss of the air assist. Current valve testing relies on the air assist to determine if the system can isolate within the assumed time.
The feed autoclave isolation system valves will be tested without utilizing the plant air assist to determine the valve closure times. If the valves do not meet the required closure times without the air assist, a plant modification to modify the existing valves, or replace the existing valves with faster operating valves, or other appropriate p
corrective actions will be prepared. The results of the valve testing, along with recommended corrective actions and an associated implementation schedule, will be submitted to the NRC by October 31,1998.
Test Results u
l A bench test was performed on two air-assisted feed autoclave isolation valves. One of the i
valve / actuator combinations tested was taken from an in-service autoclave while the other was taken from stores inventory. The valve / actuator combination that was removed from the in-service M
autoclave was in the XV-516 position. Historically, this valve position has the highest failure rate of 8
the four air-assist valve positions; thus, it would be the valve / actuator expected to be in the worst i
condition. No maintenance work was performed on this valve / actuator combination prior to
}-
beginning the tests.
The bench test conditions and configuration were established to approximate field conditions and d
configuration.- Tests of valve closure time were conducted with the air to both the valve open and lj valve close supplies at varymg pressures (80,60,40,20,10, 5 and 0 psig). A second set of tests was y
conducted with no air assist supply and the valve open supply at nominal plant air pressure of 80 psig.
Finally, the second set of tests was repeated with a quick exhaust valve installed in the valve open air supplyline to allow the air to exhaust directly to atmosphere upon valve closure. This configuration is similar to that provided for the high pressure isolation valves on the autoclaves in the C-360 facility.
L - - -
GDP 98-0224 Page 2 of 3 The tests consisted of 10 trials for each valve / actuator combination in each of the three test configurations.
For the valve / actuator combination removed from the in-service autoclave, at least 20 psig of air assist was required to meet the current Technical Safety Requirement (TSR) closure time of 10 seconds. At this pressure, about 10% margin (i.e., I second) was demonstrated. At 40 psig, about 40% margin was demonstrated. At less than 10 psig, the valve did not close. For the valve / actuator combination taken from stores, valve closure times were consistently around 3 seconds for 80, 60 and 40 psig. At 20 psig, the valve would not stay open, and at 10 psig or less, the valve would not open at all. These results show that the use of air-assist has masked potentially significant degradation of the spring-to-close mechanism in these valve / actuators while in service.
Prior to this test, a problem report had identified that the air assist valves may not close within 10 seconds without plant air and an operability evaluation was performed. The evaluation determined that the air assist valves were operable but nonconforming due to design inadequacy. The design inadequacy is the air assist valves (safety system components) rely upon the plant air system (a non-safety system) to perform the intended safety function. Based on the operability evaluation, a compensatory action to remove the autoclaves from service and declare the autoclave high pressure system inoperable upon loss of plant air has been implemented. This compensatory action ensures that the air assist valves will perform as required by the TSR, which requires a valve closure time of less than 10 seconds.
The tests conducted without air assist showed the valve / actuator removed from the in-ser ice autoclave would not close at all, with or without the quick exhaust valve installed. For the valve / actuator removed from stores, the closure times were less than 8 seconds without air-assist and less than 2.5 seconds without air-assist and with the quick exhaust valve installed.
Engineering Evaluation The following are the conclusions / recommendations of the engineering evaluation performed:
Based on the results of the bench test, the minimum air pressure required to close the air-assist valves is 20 pig.
The minimum air pressure supplied to the air-assist valves should be 40 psig to ensure that the valves close within 10 seconds and to provide a margin between the average valve closing time and the current TSR Surveillance Requirement closure time of 10 seconds.
Until the modification described below is completed, the Plant Air Failure procedure
+
should be revised to incorporate the 40 psig limit and include steps to take the feed autoclaves out of service and to declare the feed autoclave high pressure isolation system inoperable if plant air pressure decreases to below 40 psig.
L GDP 98-0224 l
Page 3 of 3 o(
- A modification should be installed to add a quick exhaust valve to each of the four a
high pressure isolation valves and to eliminate the air-assisted closure on each feed autoclave, similar tolhe modification completed on the C-360 autoclaves. This will p
allow the air that holds the valve open to exhaust straight to atmosphere rather than exiting to the air exhaust header. This modification is expected to dramatically i
improve closure times, based on the test results, and will eliminate the dependence upon plant air, as required by the SARUP accident analysis.
l Subsequent to completing this engineering evaluation, the Operations and Engineering organizations l:
determined that a field test method should be developed to test each of the feed autoclave air-assisted L
high pressure isolation valves to confirm that iney are capable of closing upon loss of plant air. This test would be conducted annually, in conju.ction with the other annual surveillance tests performed on the feed autoclaves. Quarterly TSR surveillance tests will continue to verify that the valves are l
capable of closing within 10 seconds with plant air available.
1 Plan of Action and Schedule The following actions have been or w.ll be taken based on the results of the engineering evaluation of the feed autoclave high pressure isclation valve test results:
1.
The Plant Air Failure procedure, CP4-CO-ON3003, has been revised to incorporate the 40
[
psig limit and include steps to take the feed autoclaves out of service and to declare the feed t
autoclave high pressure isolation system inoperable if plant air pressure decreases to below
,l 40 psig. This requirement will remain until the modification described below has been l'
completed for all feed autoclaves.
t.i 2.
A field test method will be developed to test each of the feed autoclave air-assisted high pressure isolation valves to confirm that they are capable of closing upon loss of plant air.
This test will be conducted annually, in conjunction with the other annual surveillance tests performed on the feed autoclaves. These tests will begin with the first scheduled feed autoclave annual surveillance occurring after January 31,1999, and will continue until the modification described below has been completed for all feed autoclaves.
T 3.
The four high pressure isolation valves on each feed autoclave will be modified to eliminate their dependence upon plant air to close within 10 seconds as specified by the current TSR Surveillance Requirement. These modifications will be completed by December 10,2002.
i-i, l
l 1
e,,
-.n,
,. -, - - ~ - - -
=. -. -..-.
s.
e GDP 98-0224 Page1of1 Commitments Contained in this Submittal
-rg a
1.
A field test method will be developed to test each of the feed autoclave air-assisted high pressure isolation valves to confirm that they are capable of closing upon loss of plant air.
This test will be conducted annually, in conjunction with the other annual surveillance tests performed on the feed autoclaves. These tests will begin with the first scheduled feed autoclave annual surveillance occurring aner January 31,1999, and will continue until the modification described below has been completed for all feed autoclaves.
2.
The four high pressure isolation valves on each feed autoclave will be modified to eliminate their dependence upon plant air to close within 10 seconds as specified by the current TSR Surveillance Requirement. These modifications will be completed by December 10,2002.
l k
i k
I b
3 i
i i
h' I
w<
-r-e.v--.-ye w--
w.v i-vv v
,--.w,s.,,mve-=v.
.-%i---
v w
,-e,