ML20141B268
| ML20141B268 | |
| Person / Time | |
|---|---|
| Site: | Paducah Gaseous Diffusion Plant |
| Issue date: | 05/09/1997 |
| From: | Polston S UNITED STATES ENRICHMENT CORP. (USEC) |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| GDP-97-1008, NUDOCS 9705150214 | |
| Download: ML20141B268 (5) | |
Text
United States C
Enrichment Corporrtion ESEC EE Tet 502 4415803 Ias: 502 441.$801 May 9,1997 United States Nuclear Regulatory Commission SERIAL: GDP 97-1008 -
Attention: Document Control Desk Washington, DC 20555-001 Paducah Gaseous Diffusion Plant (PGDP)- Docket No. 70-7001 - Event Report ER-97-07 Pursuant to Safety Analysis Report (SAR), Section 6.9, Table 6.9-1,J(2), enclosed is the required 30-day written report for two water inventory control system actuations (WICS) at PGDP which were reported on April 10,1997 (NRC No. 32128), and April 23,1997 (NRC No. 32212). Investigation activities are continuing with a final report targeted for June 15,1997.
Should you require further information on this subject, please contact Bill Sykes at (502) 441-6796.
Sincerely,
>J U
Steve Polston General Manager Paducah Gaseous Diffusion Plant SP:WES:MLB:mic Enclosure cc.:
NRC Regionlli ll NRC Senior Resident inspector, PGDP l \\,
f 1400%
gp gy 9705150214 970509 PDR ADOCK 07007001 C
PM s
F i
Dnited States Nuclear Regulatory Commission Page Two.
May 9,1997 Distribution 4
Robert L. Woolley i
. bec:
J. Adkins, USEC-IlQ D. Allen. PORTS M. Boren.
S. Brawner
- J. Dietrich, LMUS 11Q
- L. Fink, PORTS
= R. Gaston, PORTS
-'C. Ilicks i
J. Labarraque B. Lantz, PORTS R. Lipfert, PORTS J. Miller, USEC-IlQ J. Mize l
S. Morales l
S. Penrod l
- 11. Pulley v'
A. Rebuck-Main, USEC IlQ S. Routh, USEC-IlQ J. Thomas S. Scholl, PORTS 3
B.Sykes i
R. Wells, USEC-IlQ File: 97-890-165 o:\\ users \\nra\\pgdp\\97-07-er
]
. 1
[.5
)
1
l
)
s
! Docket No.'70-7001 3
Page1of3 EVENT REPORT ER-97-07 l
BACKGROUND The C-360 autoclave Water Inventory Control F stem (WICS) gevides the means to limit the water I
. inside an autoclave such that upon a UF. release from a 15. cylinder, over pressurization of the autoclave or a nuclear criticality cannot occur. His is w mplished by two electrically powered sonic probes located slightly below the autoclave in the 3-ir.ch drain line. The probes actuate upon the presence of excessive condensate in the autoclave and send a signal through electrical interlocks to remove power from the solenoid valves which remove air from two steam isolation valves and the vent steam valve. These valves are air powered and fail closed upon loss of air (thus fail closed). The system
. function is to isolate the sources of steam from the autoclave upon detecting a level of water accumulating in the condensate drain line. The WICS is designed as a "Q" safety system and required to be operable by Technical Safety Requirement (TSR) 2.1.4 when in Mode 5 (heating).
In C-360, unlike C-337 A and C-333-A Feed Vaporization Facilities, autoclaves Nos. I and 2 share common steam supply and condensate drain systems as do autoclaves Nos. 3 and 4.
This report provides the information presently available on two WICS actuation events which occurred at l
the C-360 toll transfer and sampling facility on April 9,1997, and April 23,1997, respectively.
l DESCRIPTION OF EVENT (Reference NRC No. 32128)
On April 9,1997, C-360 UF. sampling and transfer facility operators prepared autoclaves Nos. I and 2 for the heating and sampling of UF. cylinders.
At 1955, steam was simultaneously valved to both autoclaves. At 2030, a WICS actuation occurred on autoclave No. I which automatically isolated the steam supply to the autoclave. All s&4y system components functioned as designed with the exception of one of the condensate level probes which did not show alarm status. Facility operators responded by entering alarm response precedure (ARP) CP4-CO-AR8360-15. The front line manager (FLM) and Plant Shla Superintendent (PSS) were notified.
The operators reported that only one of the two redundant condensate level probes was in alarm at the
~,
time of the actuation and that after jetting the steam and opening the autoclave they did not see any water standing in the bottom of the autoclave. They concluded that, based on this, the actuation was caused by a spurious invalid signal. The PSS agreed and the autoclave was returned to service at 2130, April 9, 1997.
On April 10,1997, the responsible system engineers reviewed the facts known about the actuation and concluded that the WICS actuation was indeed valid. The PSS was notified of this conclusion.' The NRC-IlQ Operations Office was notified at 1755 (CDT) on April 10,1997, of the Safety System Actuation per SAR 6.9 Table 1, Criteria J.2 (reference NRC ERNo. 32128).
.l l
t r
a Docket No. 70-7001 1
Page 2 of 3 The failure of the facility operators and the PSS to recognize the WICS actuation as being valid and subsequently returning the autoclave to service without a true understanding of the actuations cause is a violation of ARP step No. 9. This step states,"When any condensate is drained and cause of alarm has 1
been determined and corrected, go to CP4-CO-ON3039, "C-360 Interrupted lleat Cycle," to return autoclave to service. The operators felt that they had determined the actuation to be spurious and invalid, based on the fact that no free standing water was found in the bottom of the autoclave when opened and only one of the two probes actuated.
Past experience has shown that this is not always an indication of a false alann. Even if the actuation was determined to be false, returning the autoclave without determining why the false alann occurred is inconsistent with the ARP requirement (step No. 9), and the conduct of operations philosophy of treating alarms as real and valid until a definitive cause for the alarm is determined. As a compensatory measure, a Long Tenn Order (LTO) was issued that requires the involvement of the shift or system engineer when determining the validity of all safety system actuations.
On Anril 11,1997, Instrument Maintenance performed the WICS functional test and found one of the two condensate ultra-sonic probes to be inoperative due to a coating of steam contaminates. Both probes were cleaned, the system functionally tested and operability restored.
DIRECT CAUSE (NRC ER No. 32128)
The direct cause of the WICS actuation was initially attributed to the C-360 autoclave loading and heating procedure which did not preclude the simultaneous initiation of the heat cycle for a companion pair of autoclaves, in this case autoclaves Nos.1 & 2. This caused the flow of condensate from autoclave No. I to be restricted by the condensate flow from autoclave No. 2 which, due to the design of I
the system, has the advantage over autoclave No. I for both steam supply and condensate drainage. The condensate from autoclave No. I backed up and contacted the condensate level probes which actuated the WICS. Procedure CP4-CO-CN2051a,"C-360 Autoclave Loading and lleating," Rev.1, was revised on April 18,1996, to establish a one-hour minimum stagger of the initiation of the heat cycle for a companion pair of autoclaves. This provides a compensatory measure until the event investigation is i
completed and the root cause determined. Autoclave No. I was tested, declared operable, and returned to service under the revised procedure on April 18,1997.
DESCRIPTION OF EVENT (Reference NRC ER No. 32212)
Prio; 1 determination of the root cause of the event on April 9,1997, on April 23,1997, at 1542 CDT, whue heating a cylinder in preparation for liquid sampling, a second WICS actuation occurred on autoclave No, l. Facility operations responded per ARP CP4-CO-AR8360-7 and subsequently removed the autoclave from service. The NRC-IlQ Operations Ollice was notified of the event at 2300 CDT on April 23,1997 (reference NRC No. 32212).
DIRECT CAUSE Investigation by the System Engineer revealed that the direct cause of the actuation was related to the failure of the pressure to current transducer (PY' 1 %) in the autoclave temperature / pressure control loop
- =
11'
..s:
~
Docket No. 70-7001 ~
Page 3 of 3 -
1 instrumentation.' This failed transducer receives pressure input from the autoclave pressure transmitter
)
. (PTl ISA) and translates this signal from pressure input to an electrical current output to the j
. temperature / pressure control loop. This causes the autoclave steam control valve to open or close ta -
maintain the autoclave temperature and pressure at the desired set-points.
i When the transducer failed,'it read a pressure lower than actual and signaled the steam control valve to -
open. This, in turn, caased more steam to be admitted to the autoclave than necessary, This situation ~
l
. when combined with autoclave No. 2 in mode 5 (heating), caused an excessive amount of condensate in the drain which eventually contacted the condensate probes resulting in the WICS actuation. This failed transducer was subsequently replaced and the autoclave passed post-maintenance functional testing.
ROOT CAUSE j
The engineering analysis to determine root cause of both the April 9,1997 (reference NRC No. 32128),'
and the April 23,1997 (reference NRC No. 32212), WICS actuations has not been completed.
j Additionally, on May 2,1997, with autoclave No. I out-of-service and open, and with autoclave No. 2 m mode 5 (heating), a WICS actuation occurred on Autoclave No.1. System Engineering reviewed the fam of; Sis al - rnd concluded that condensate from autoclave No. 2 had backed up in the common x act with the sonic condensate probe on autoclave No.1. Although, not reportable, dra A
lthis insident has caused a reevaluation of the root cause conclusions reached thus far.
The operation of the C-360 autoclaves has been halted pending additional evaluation. By June 15,1997, l
this report will be supplemented and issued as a rn ised repcat to summarize the completion of the i
ongoing engineering analysis to detennine the root cause and corrective actions to prevent recurrence of
]
these events.
~ CORRECTIVE ACTIONS COMPLETED
~
5 L
On April I 8,1997, the C-360 autoclave lo.ading and heating Procedure No. CP4-CN2051a was revised to exclude the simultaneous initiation of companion autoclave heat cycles.
2.
On April 14,1997, a LTO (300-97-003) was issued which required the involvement of a shift or system engineering in the evaluation of safety systems actuations.
PLANNED CORRECTIVE ACTION To be determined on root cause analysis 1
. EXTENT OF EXPOSURE OF INDIVIDUALS TO RADIATION OR RADICACTIVE M ATERIALS l
c There were no releases of radioactive materiab as a result of these events.
i LESSONS LEARNED j
'To be determined
)
1
- '