ML20198E161
| ML20198E161 | |
| Person / Time | |
|---|---|
| Site: | Framatome ANP Richland |
| Issue date: | 12/01/1992 |
| From: | Maas L SIEMENS POWER CORP. (FORMERLY SIEMENS NUCLEAR POWER |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| IEB-92-001, IEB-92-1, NUDOCS 9212040246 | |
| Download: ML20198E161 (4) | |
Text
.
SIEMENS t
December 1,1992 U.S. Nuclear Regulatory Commission Attn: Document Control Desk Washington DC 20555 Gentlemen:
Re:
Follow-up to NRC Bulletin 91-01 Report No. 24416 - Tank 10 Hydrolysis Overflow.
On 10/13/92 Siemens Power Corporation (SPC) reported a criticality safety incident to the NRC Operations Office per NRC Bulletin 91-01. SPC Internal procedures require a written follow-up report of the initial telephone report. This letter fulfills this requirement.
Backaround The UF to UO powder conversion process used by SPC includes a hydrolysis step to 6
convert UF to O F. In the une 1 vaporization room there are two vaporization chests that 6
p2 are used to heat UF cylinders. The UF gas that evolves is piped to a hydrolysis tank and s
6 comb!ned with deionized water. This process, under normal process conditions, produces UO F in the 80 to 150 gU/ liter range and HF gas as a byproduct. The process specification p2 limits the concentration to 250 gU/ liter to limM the amount of heat generated by this reaction.
The maximum concentration possible in this system due to the physical properties of the hydrolysis process is 361 gU/ liter.
Description On 10/13/92 at 0230 hours0.00266 days <br />0.0639 hours <br />3.80291e-4 weeks <br />8.7515e-5 months <br /> a conversion area chemical technician toured the Une 1 conversion area in order to update the shift log. He noticed what appeared to be UO F 22 solution running out from under the Une 1 vaporization room door. He immediately went to the control room and shut off the UF gas flow and contacted his Shift Supervisor.
6 Entry into the vaporization room by a Chemical Technician Specialist revealed UO F solution 22 leaking from the flex hoses connecting the process offgas (POG) duct to the vaporization chests. Source of the UO F solution was believed to be Tank 10, the hydrolysis tank. By 22 0250 hours0.00289 days <br />0.0694 hours <br />4.133598e-4 weeks <br />9.5125e-5 months <br /> the delonized water flow to Tank 10 was secured, the lower header nitrogen purge was stopped, and Tank 10 was pumped down. Cleanup of the spill, estimated at approximately 40 gallons, was initiated at 0300 hours0.00347 days <br />0.0833 hours <br />4.960317e-4 weeks <br />1.1415e-4 months <br />.
At 0330 hours0.00382 days <br />0.0917 hours <br />5.456349e-4 weeks <br />1.25565e-4 months <br /> the Chemical Technical Specianst cGed an instrument technician to check the level control transmitter on Tk-10. The indicator and the alarm calibrated properly. Since the Siemens Power Corporation
(,4QO1,3
[
Nuclear Division - Engineering and Manufacturing Facility
[
2101 Horn Rapids Road. PO Box 130 - Richland, WA 993524130 im. (509) 375-8100 Fax: (509) 37i>-8402 9212040246 921201 PDR ADOCK 07001257 C
Document Control Desk Page 2 December 1,1992 instrument technician could not find anything wrong with the transmitter / alarm instrumentation, he suspected a mechanical problem with the dip tube assembly in the tank which could cause an erroneous input to the tank level transmitter. At 0500 a pipetitter and the Chemical Technical Specia;st removed the dip tube from the tank and did a quick air pressure check for leaks or plugs. They did not find any readily apparent problems, however a complete pressure check of the tube was not performed. The instrument technician did another calibration test on the transmitter. A conversion area chemical technician filled 410 one-half full with dolonized water (DlW) to check the level control transmitter. It read correctly. The shift supervisor then filled Tk 10 to 70 inches with Dr". After running DlW flows for 20 minutes with no overflow, the shift supervisor thought the problem had been corrected and at 0600 restarte-I UF flows. At 0650 the Chemical Technical Specialist went 6
back into the Line 1 vaporization room to verify there were no more leaks and found none.
The shift supervisor nottled his General Supervisor at 0610 hours0.00706 days <br />0.169 hours <br />0.00101 weeks <br />2.32105e-4 months <br /> and the Plant Operations Manager at 0700 hours0.0081 days <br />0.194 hours <br />0.00116 weeks <br />2.6635e-4 months <br />.
At 0800 the day shift conversion area chemical technician entered the Line 1 vaporization room to remove the UF cylinder from Chest 2. He closed the upper doors and prepared to 6
unhook the cylinder. After he unhooked the cylinder and reopened the upper doors, he heard liquid flowing into the chest and noticed a liquid solution in the bottom of vaporization i
chest 2. The UF process was again shut down. Operations personnel started cleanup by e
removing the liquid from the vaporization chest to the Miscellaneous Uranium Recovery System (MURS). No liquid was found in chest 1. At 1000 hours0.0116 days <br />0.278 hours <br />0.00165 weeks <br />3.805e-4 months <br /> operators began opening drcin lines in the duct work. Approximately 12 gallons of liquid (presumed to be UO F ) was 22 discovered and drained. At 1100 hours0.0127 days <br />0.306 hours <br />0.00182 weeks <br />4.1855e-4 months <br /> the vaporization chest 2 heaters ware lifted from the bottom of the chest so cleaning could be done. A green stain line on the sides of the chest indicated there had been up to four inches of UO F solution in the bottom of the chest.
22 Operations notified Safety, Security, and Licensing (SS&L) of the incident at 1215 hours0.0141 days <br />0.338 hours <br />0.00201 weeks <br />4.623075e-4 months <br />.
j SS&L called an Incident Investigation Board (118) which first met at 1354 hours0.0157 days <br />0.376 hours <br />0.00224 weeks <br />5.15197e-4 months <br /> on 10/13/92.
l The llB concluded that the likely source of the UO F was an overflow from Tk-10 up the tank 22 vent into the K10 POG exhaust system located on the second level and back down the exhaust system to the vaporization chests in the Une 1 vaporization room located on the lower level. Laboratory analyses confirmed that the solution in question was indeed UO F at 22 a U concentration (70ams U/ liter) and enrichment that matched the material being processed in Line 1 conversion. The lib recommended Operations take the following actions:
1.
Put Une 2 conversion on hot standby as a conservative precaution even though UF vaporization and hydrolysis was not being conducted.
e
- 2. Maintain Line 1 conversion in a shut down mode as it had been since 0830 hours0.00961 days <br />0.231 hours <br />0.00137 weeks <br />3.15815e-4 months <br />.
Rocheck the dip tube and level control indicators and attempt to explain how the level indicator in Tk-10 could be working correctly, the high level alarm functional,'and the tank overflowing.
Document Control Desk Page 3 December 1,1992 The 118 recessed until 0830 hours0.00961 days <br />0.231 hours <br />0.00137 weeks <br />3.15815e-4 months <br /> on 10/14/92 at whici time Operations and Engineering reported the following:
1 The Tk 10 dip tube was cracked near the top threads. This would give the leve!
control indicator an incorrect input and would indicate that the level in the tank was significantly lower than it really was.
- 2. The delonized water flow into Tk 10 was approximatoly 250 gallons per hour. Tk-10 did not have an overflow line. Therefore, as the liquid solution rose in the tank it overflowed into the vent line and entered the POG exhaust duct. Some of the liquid flowed through the POG duct into the Line 1 vaporization chest 2.
Cause The overflow of the UO F solution from Tk-10 to vaporization chest 2 in the Line 1 22 vaporization room was caused by a malfunction of the tank level control system. Because TR 10 had not been equipped with an overflow and all credible pathways for liquid flow in the POG ducting had not been adequately analyzed, a mechanism existed for fissile material-bearing solution to enter the vaporization chest. The vaporization chests had not been analyzed for containing solutions of fissile materials.
Corrective Actions In addition to the required repairs to the Tank 10 level indicating system, the llB recommended a namber of corrective actions to preclude similar events in the conversion lines and related systems, as follows:
- 1. All tanks in the Line 1 and Line 2 conversion areas that did not have overflows were equipped with overflow lines capable of handling the maximum credible overflow rate.
- 2. Those that were equipped with overflows were verified to be adequately sized to handle the maximum credible overflow rate.
- 3. The tank overflow lines were re-routed to favorable geometry drop legs in the POG system.
- 4. The favorable geometry drop legs were equipped with trapped overflow lines that were adequately sized to handle credible overflow conditions.
- 5. Criticality Sefety Instruction Cards were added at each of the overflow lines stating that the overflow must remain unobstructed and free flowing.
- 6. Each of the overflow lines was placed on a PM system to periodically verify the system remains unobstructed and free flowing.
. ~ - _. -
4 g.
4 I
Dwument Control Desk Page 4 Dacember 1,1992
- 7. Other areas of the plant were reviewed for similar conditions by a separate Independent team of personnel from Engineering and Safety, Security and Licensing.
l Those areas with similar deficiencies were modified as appropriate to eliminate similar -
unanalyzed flow pathways of uranium-bearing solutions via POG systems.
1 2
in response to this event, NRC Region V issued a Confirmatory Action Letter (CAL) on i
10/14/92 placing certain operational restrictions on Siemens until the cause of the incident -
1 was determined and all corrective actions had been completed. A requirement for NRC concurrenen of the restart of Line 1 and processing of UF in Line 2 was included.
8 To provide effective review of the performance / completion of all corrective actions'and to assure a safe startup of the conversion lines, SPC convened a Startup Council. Startup i
approval from the Council was granted on October 22. NRC concurrence, i.e notification that requirements of the CAL had been met, was received on October 23. UF processing on o
Line 1 resumed on the afternoon of October 23.
l Questions regarding SPC actions in response to this situation can be directed to me on (509) 375-8537.
Very,truly yours W
N'M Loren J. Maas, Manager Regulatory Compliance LJM:pm l
cc:
J. W. N. Hickey, NRC J. B. Martin, NRC l
1 i
l i
i r
.a
=.
w r-,,
~.. -
yv
-i-ew.,vi-.w,.v g y a
,v
-w e,.7.+
.v
-