ML20202G412

Provides Event Rept 97-10,rev 3,for Event Involving UF6 Cylinder High Pressure Alarm Actuation at Portsmouth Gaseous Diffusion Plant

ML20202G412
Person / Time
Site:	Portsmouth Gaseous Diffusion Plant
Issue date:	02/12/1998
From:	Morgan J UNITED STATES ENRICHMENT CORP. (USEC)
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
GDP-98-2006, NUDOCS 9802200123
Download: ML20202G412 (6)

Text

p USEC

. A cion an.,my comp.ny a

Fcoruary 12,1998 United States Nuclear Regulatory Commission GDP 98-2006 Attention: Document Control Desk Washington,.9,C. 20555-0001 Portsmouth Gaseous Diffusion Plant (PORTS)- Docket No. 70-7002 - Event Report 97-10, Revision 3 Pursuant to Safety Analysis Report (SAR), Section 6.9, Table 6.9-1, J (2), F. closure I provides Event Report 97-10, Revision 3, for an event involving a UF. cylinder high pressure alarm actuation 4

I at the Portsmouth Gaseous DifTusion Plant. The revised event report is being submitted to provide a root cause and corrective actions Changes from the previous report are marked with a vertically dashed line in the right margin.

Should you require additional information regarding this event, please contact Scott Scholl at (614) 897-2373.

Sincerely, q&~

Jim Morgan Q Acting General Manager Portsmouth Gaseous Diffusion Plant JBM:SScholl:cw t

i fe,{

cc:

NRC Region III D. Ilartland, NRC Resident luspector, PORTS glg lll[]ll 9802200123 980212 PDR ADOCK 07007002' C

_PDR,

-" m.. x -

P.O. Ilox 800, Portsmouth, OH 45661 Telephone 614-897-2255 Fax 614 897-2M4 http://www.usec.com Offices in Livermore, CA Paducah, KY Portsmouth, OH Washington, DC

_O

Upited States Nuclear Regulatory Commission February 12,1998-

Page Two-Distribution bec:-

J. Adkins,HQ TJ. Anzelmo, PORTS -

R. Boelens, PORTS -

J. Bolling, PORTS

_ M. Boren, PGDP '

. S. Brawner, PGDP L. Cutlip, PORTS J. Dietrich, LMUS -

L. Fink, PORTS -

R. Gaston, PORTS

. M. Ilasty, PORTS -

l P. Hopkins, PORTS -

. L. Jackson; PGDP. -

J.- Labarraque, PGDP B. Lantz, PORTS-R. Lipfert, PORTS R. McDermott, PORTS.

J. Miller, HQ J. Mize, PODP

. J. Morgan, PORTS -

J. Oppy, PORTS G. Price, PORTS :

S. Polston, PGDP:

H. Pulley, PGDP A. Rebuck-Main, HQ R. Robinson, PORTS S. Routh, HQ S. Scholl, PORTS D. Thompson, PORTS

- D. Towne, PORTS

- M. Valentine, PORTS R. Wells, HQ Plant Shift Superintendents PORTS Record Management o:\\ users \\nra\\ letters.ntc\\ ports \\gdp98-2.006 l

4

Docket No. 70-7002 -

Enclosure 1 Page1of4 Event Report 97-10

- Revision 3 Description of Event On May 23,1997, at 0125 hours0.00145 days <br />0.0347 hours <br />2.066799e-4 weeks <br />4.75625e-5 months <br />, X-343 Autoclave (AC) #5 was in Mode II, heating 30 inch (2.5--

- ton) Russian Uranium Hexafluoride (UF.) cylinder #LUO705 when the audible alarm for the cylinder -

high pressure autoclave steam shutofTwas received. _ Operators responding to the alarm observed that the internal cylinder pressure was 115 psia, which was also the alarm setpoint. Steam supply valve

- FV-513 was verified closed as designed. Five minutes later at 0130 hours0.0015 days <br />0.0361 hours <br />2.149471e-4 weeks <br />4.9465e-5 months <br />, while in Mode VII and

~

with the steam supply valve still closed, A? #5 alarmed a second time when the cylinder pressure

' reached 130 psia.- The actuation of the UF. cylinder high pressure autoclave steam shutoff is reportable in accordance with the Safety Analysis Report (SAR), Table 6.9 1, J (2).

At 0105 hours0.00122 days <br />0.0292 hours <br />1.736111e-4 weeks <br />3.99525e-5 months <br />, at the start of the cylinder heating process the cylinder cold pressure reading was 6.0 psia.1 At 0124 hours0.00144 days <br />0.0344 hours <br />2.050265e-4 weeks <br />4.7182e-5 months <br />, during the heating process the operator observed the cylinder pressure at 45.0 -

psia, which was higher than what is normally expected after 19 minutes of heating. In accordance with procedure XP4-TE-FD2701, "X-343 and X-342 Autoclave Operation," the operator prepared 1 to hot burp (vent to the cascade) enough UF to reduce cylinder pressure. Orse minute later, at 0125.

hours, before the operator could hot burp the cylinder the autoclave alarmed at i15 psia due to cylinder high pressure. The operator responded according to procedure and immediately hot burped the cylinder. The cylinder pressure following the hot burp was 9.0 psia. At 0130 hours0.0015 days <br />0.0361 hours <br />2.149471e-4 weeks <br />4.9465e-5 months <br />, although -

the steam supply was isolated from the autoclave, the operator noticed the cylinder pressure had -

increased to 70 psia and rising. While the operator was proceeding to hot burp the cylinder for the second time the autoclave alarmed again due to high cylinder pressure. The operator responded and

- immediately hot burped the cylinder reducing the pressure to approximately 50 psia. - The cylinder pressure then dropped to approximately 20.0 psia and stabilized. The highest pressure observed during this event was approximately 130 psia which is less than the 214.4 psia maximum allowable working pressure (MAWP) of 2.5 ton cylinders. Approximately thirty-thr.:e N. of material was removed from the cylinder during the two burps,

~'

The UF. cylinder high pressure autoclave steam shutoff system is provided to prevent internal cylinder over pressurization ' caused by normal heating of a UF. cylinder containing excessive

' amounts of" light" gases. In addition k the event an over filled cylinder is heated in an autoclave, the UF. cylinder high pmssure autoclave steam shutoff system prevents the intemal cylinder pressure from exceeding the hydrostatic test pressure of the cylinder.

Cause of Event The direct cause of the UF. cylinder high pressure alarm was the release of a pocket of trapped liquid i

surrounded by dense solid UF. along the cylinder wall. As heat was applied to the UF cylinder, i

6 liquid was formed and trapped inside the dense solid UF. and it vented to a void space above the i

- solid creating a high pressure excursion.

i

m i.' 4

< Dpcket No. 70-7002 J Enclosure 1 Page 2 of 4 L 2 Event Report 97-10

- Revision 3 Similar pressure increases have occurred previously while heating 2.5-ton cylinders. One event occurred September 9,1995, while heating two Commonwealth Edison cylinders filled in Russia.

LThe other occurred October 24,1996, while heating three cylinders containing blended UFifilled in Russia. 'Ihese pressure increases were similar to pressure increases experienced when heating 14--

ton Tails cylinders filled at PORTS and cylinders filled at PGDP during cold weather.

- Corrective actions resulting from these previous events included revising operating procedures to L give the operator the authority to hot burp a cylinder when cylinder pressure is rising rapidly or has exceeded 90 psia. However, in this event the pressure rise occurred more rapidly than expected, preventing the operator from relieving the cylinder pressure before the 115 psia set point was exceeded.-

a; Other factors that could have produced the type of pressure increase were eliminated by weight data and the analysis oflaboratory samples following the event. ; Weight data showed that the cylinder was filled to 4994 pourc',s,- which was below the 5020 pound maximum cylinder fill limit. The

. amount'of weight lost (pring the two hot burps indicates that liquid UF. was evacuated from the -

cylinder and not " light" gases.' The laboratory samples taken following hot burping indicate that no excessive " light" gases were present in the cylinder.

On May 23,1997, Customer Services and Product Scheduling put a hold on heating 2.5 ton Russian cylinders until an engineering evaluation could be completed to determine corrective actions io

~

prevent a recurrence of this event. On' May 29,1997, the cylinder hold was modified to allow --

heating of the Russian cylinders that have previously been through a heating cycle at PORTS.

On July 9,1997, heating of Russian cylinders was resumed following implementation of temporary

- procedural controls. The procedural controls included an enhanced valve clarity check to identify whether unequal distribution of UF. existed behind the cylinder valve. - The new controls also provided a modified heating process that reduced the rate that heat was added to the cylinders and an attemate valving configuration that reduced the time needed to burp cylinders to avoid cylinder high pressure alarms. Additional cylinder void space gas samples were also taken after modified i heating to allow analysis of trapped gases. These actions wers taken to reduce the probability of a Lrecurrence of this event and allow additional data to be gathered to determine the root cause of the pressure spikes.

Following the implementation of the temporary controls described above, the analyses of the gas i

samples of approximately 47 cylinders were completed. The samples indicated that the gases i

released were typical of what is normally emitted from solid UF upon heating. As a result it was i

concluded that excessive light gases were not the cause of the pressure spikes.

i On August 23,1997, two cylinders experienced pressure spikes within the first 15-minute heating 3

r.

R e

Docket No. 70-7002 Page 3 of 4 Event Report 97-10 ;

Revision 3 cycle. The peak pressures were well below the ' alarm setpoint of the autoclaves. Enhanced valve clarity checks showed no indications of UF. crusting behind either cylinder valve. - Gas samples L obtained on the trapped pressure indicated typical composition results. The substantial weight of

'I the material collected _ in the' sample containers for these two cylinders indicated that the high pressure spikes were caused by the formation of liquid UF. in the pigtailD Upon resumption of -

heating on both cylinders, pressure spikes again occurred after approximately five minutes and required hot burping to prevent alarm activaticn.

Engineering has determined that the root cause of the UF. cylinder high pressure alarm was the rapid i

- cylinder cooling condition tha' took place in the cold emironment of Russia.' This is similar to what

' has been.previously experienced at _ PORTS and PGDP-with tails cylinders. Cylinders cooled i

outdoors during winter months in Ohio and Kentucky have produced pressure spikes when heated.

i

- Laboratory tests conducted in 1970 have shown that liquid UF. cylinders cooled rapidly in a winter i

environment form a solid that is dense and hard with large voids. Transversely, liquid UF cylinders i.

cooled slowly in a summer environment form a solid that is flaky with small voids. The strength i

of the solid varies in each cylinder and explains why the pressure spikes occur randomly.

I Elimination of the pressure spikes has not been achievable due to the cold ambient environment in _ i

- Russia where the cylinders are cooled. However, the controlled hectir.g process currently being used - i

- has been effective in reducing the resultant pressure peaks to levels well below the UF. cylinder high - i

. pressure alarm set point.

i Corrective Actions:

1; On July 8,1997, temporary procedure XP4-TE-FD2791-TMP was implemented to provide;

1) a new method of identifying the existence of unequal distribution of UF. behind the cylinder valves,2) a modification to the cylinder heating process, reducing the rate of heat applied to the cylinders,3) an alternate valving configuration reducing the time required to burp cylinders to an evacuation header and 4) collection of cylinder void space gas samples after modified heating for analysis of trapped gases.

2.

~ Effective September 17,1997, temporary procedure XP4-TE-FD2791-TMP was revised to; i

1) delete further requirement for gas sampling after initial controlled heating,2) change controlled heating conditions.

3.

Effective January 9,1998, the temporary procedure provisions of September 17,1997, were i

incorporated into procedure, XP4-TE-FD2701, Revision 5, "X-343 and X-342 Autoclave i

Operation," to provide; 1) an enhanced valve clarity check. 2) a modification to the cylinder i

heating process, reducing the rate of heat applied to the cylinders,3) an alternate valving i

configuration reducing the time required to burp cylinders to an evacuation header.

i

Docket No. 70-7002 Page 4 of 4 Event Report 97-10 Revision 3

^

Extent of Exposure ofIndividuals to Radiation or Radioactive Materials There were no exposures to individuals from this incident to radiation or radioactive materials.

Lessons Learned The event demonstrated that pressure rises experienced while heating Russian cylinders can occur i

more rapidly than previously believed Controlling the cylinder heating process and aligning the i

evacuatien route before heating the cylinder has strengthened the operator's ability to prevent the i

actuation of the UF. cylinder high pressure alarm-i

_ _ _ _ - _ _ _ _ _.