ML20132E054

Discusses Event Which Occurred on 961203 Re Enriched Matl Observed Between Line 3 Calciner Inner Tube & Outer Heat Shield.Active Engineered Interlock on Tube Rotation Failed to Minimize Accumulation.Revised Change Control Process

ML20132E054
Person / Time
Site:	07001113
Issue date:	12/11/1996
From:	Kipp C GENERAL ELECTRIC CO.
To:	Mallett B NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
References
NUDOCS 9612200298
Download: ML20132E054 (8)

Text

gtACT}M/

~

GENuclearEnergy

)0 ~ 00 Craig P. Kpp Nuc>est Energy Production GeneralManager GeneralElectnc Company P O Bos 780. Wultmngton. NC 18402 0180 M!C A20 910 615-5666 fu 910 675-66ES December 11,1996 Dr. Bruce S. Mallett, Director Division of Nuclear Materials Safety US Nuclear Regulatory Commission, Region II 101 Marietta St., NW, Suite 2900 Atlanta, Georgia 30323

Subject:

Summary of Actions for Line 3 Calciner Tube Reportable Event

Reference:

NRC License SNM-1097, Docket # 70-1113

Dear Bruce,

Pursuant to our telephone conversations, the following information is provided regarding the Line 3 Calciner event. This event was called into the NRC Operations Center in Washington, DC, shortly after noon on December 3,1996, stating:

At approximately 0900 a.m. on December 3,1996 38. 77 kgs of 4.90% enriched material was observed between the Line 3 Calciner inner tube and the outer heat shield. The material waspromptlyplaced into safe geometry 3-gallon containers.

The active engineered interlock on tube rotationfailed to minimi:e the accumulation.

We are reporting this condition pursuant to NRC bulletin 91-01. This report is required within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> since the active engineered control minimizing accumulation outside ofthe calciner tube was lost and the time requiredfor calciner tube replacement is greater than 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />. The Line 3 Calciner operation has been shutdown pending alloy tube replacement and implementation of corrective actions.

'Ihe remaining calciners were shutdown later in the afternoon of December 3rd (see the timeline in Attachment I). Separate GENE Root Cause and Corrective Action teams were immediately chartered to investigate this incident. (See Attachment 11 for the Tap Root Summary and Attachment III for the Corrective Action Matrix).

'Y P

9612200298 961211 PDR ADOCK 07001113 C

ppg l

200030 l

-_m l

4 t

Dr. B.S. Mallett i

December 11,1996 j

Page 2 of 8 i

On December 4,1996, Mr. Ed McAlpine, Chief-Fuel Facilities Branch, arrived to j

j lead the NRC's "GE Special Inspection Team", which included Messrs. David Ayres (Region II), Garrett Smith (NMSS), Donald Stout (NMSS), and Chris Tripp (NMSS).

At the entrance meeting with the Special Inspection Team, the ' margin of safety' of the 'as found' condition and the ' worst credible' condition were discussed. Calculations j

show that a wide margin of safety existed, and the system was ' deeply suberitical' (see Attachment IV).

l On Friday December 6,1996, GE and NRC personnel in Wilmington participated in a j

conference call with NRC Region II and NRC Headquarters to review the root causes and corrective actions for this incident. During this discussion GE detailed the root causes and i

the NRC Special Inspection Team concurred Also during this discussion, it was determined that GE would work to develop an agreed upon restart plan detailing short-term corrective actions which would specifically address the root causes. 'Ihe checklist of short term corrective actions was finalized on the aftemoon of December 6th (see Attachment V). All short-term corrective actions are being applied to each process line prior to re-start.

l All short-term corrective actions for Line 3 were implemented by Saturday evening (12/7/96), with inspection verification performed by on-site NRC personnel. 'Ihese l

corrective actions are summarized in the following main categories:

- Tighten the tolerance of the tube rotation timer AEC i

- Enhance the isolation of ADU to the calciner

- Correct stack monitoring system

- Provox Operator Action Request for calciner stack data l

- Formalize calciner tube Preventative Maintenance programs i

On Saturday night (12/7/96) I spoke to you over the telephone about completion of the checklist items. Following our telephone conversation, I gave the approval to restart Lme 3 Calciner, when ready. Feed to Line 3 Calciner was introduced at ~10:50 AM on l

Sunday December 8,1996.

l*

n Dr. B.S. Mallett December 11,1996 l

Page 3 of 8 i

Long term corrective actions have also been developed (see Attachment III) and are i

summanzed as follows:

I

- Revise Criticality Safety Analysis to improve the accident l

condition model l

~

- Revise design specifications for the calciner tubes i

- Revise change control process l

- Evaluate measurement methodology for the detection of material accumulations a

I hope that this letter provides you with the information you require and that our response to the event demonstrates our commitment to safety and open communications.

Please contact me at (910) 675-5666 if you have any questions or would like to discuss this matter further.

Sincerely,

]

t (ex -

C 76 Craig P. Kipp to General Manager Nuclear Energy Production l

cc: M.A. Larnastra - Washington, DC P. Ting - Washington, DC M.F. Weber - Washington, DC l

l I

4 P

Dr. B.S. Mallett l

. December 11,1996

[

Page 4 of 8 i

Attachment I l

TIMELINE 11/30/96 -

1150 Smoke observed in Controlled Area 4

j-

-Investigation Began

[

1200 Squirt Tube pulled Line 3 1300 Line 3 shut down. Stack sampler changed.

l i

12/1/96 Calciner-Cool Down t

1200 Experienced contractor crew organized I

for Calciner disassembly i

12/2/96 0700 Front and rear end Calciner removed 12/3/96 0900 Removed calciner heat shield top.

l Accumulation observed 1250 NRC notified (Ops Center) 1745.

All ADU lines shut down. Investiga'. ion Teams initiated.

i 12/4/96 0830 NRC Special Investigation Team onsite

-12/6/96 0930 Root Causes and Correctiw Actions j

Identified i

t 1300 Restart Checklist Developed l

r i

1

'12/7/96 1800 Checklist items Completed l

t 2000 Restart initiated - Line 3 12/8/96 1050 Wet Feed to Calciner-Line 'l

\\

-.-. _. -. - - -. _ - - -.. -. - -.. ~. - = -. ~.. _

1

~

Dr. B.S. Mallett j

December 11,1996 j

Page 5 of 8 Attachment II 4

i Taproot investigation Excessive accumulation of UO2 outside #3 defluorinator tube l

i j

Causal Factor #1 : Failure of Active Engineered Control (AEC) to prevent l

UO2 accumulation l

j-Changes to tube material and design indirectly impacted integrity of AEC j

f Eauinment re -hility Less Than Adeausta (LTA) : Problem not ankinated u

i Causal Factor #2 : Broken Tube / Design Control l

A. Final anneal impact on rolled tube lifetime was not recognized, j

nor 360 degree welding of flight sections; inadequate design control i

l Eauioment R "-hility LTA : Snees LTA Eauioment Re mhility : Problem not anticioated u

B. The Preventative Maintenance (PM) program for newly designed, rolled l

alloy tubes was inadequate l

PM for Eauioment LTA

]

i i

l Causal Factor #3 : Delayed notification of stack results 1

l l

A. The design of the stack reporting software did not flag data out-of-spec j

activity levels Eauinment Reli=hility LTA: Soecs LTA B. Personnel changes, Thanksgiving Holiday, computer network problems interfered with timely delivery of stack results i

i Management System: Administrative controls not strict enough (Note: sta:k reporting system was intended for emissions reporting, not

{

for safet/ control or tube integrity oversight) 4 i

I

,,----y

_y c

3

-7.

-m

_y

Dr. B.S. Mallett December I1,1996

. Page 6 of 8 AttachmentIII I

se m

m CORRECTIVE ACTIONS RIAN DGACMME 6

TERM

• TERM CAUSE X

X X

1-

1) Tighten tolerance on tube rotation limit switch

• 30 seconds reduced to 19 seconds (tube rotases once every 17 seconds)

• Educase all operations personnel on lessons learned from this rolled alloy tube faihne.

X X

1

2) Enhance the isolation of ADU to the Calciner.

3) Correct weaknesses in stack monitoring data collection / reporting X

X X

3

4) Modify the Provox control system such that the weekly stack resuk MUST X

X X

X 3b be entered as a Operdor Action Request (OAR) for each calciner. Failure to input cil! stop !!UF flow.

5) Evaluate and formalize Preventative Mamtenance (PM) program for cast and X

X 2b rolled calciner tubes (Maintenance to own). Establish maximum throughput for both tube ' flip' and tube replacement.

6) Revise Criticality Safety Analysis (CSA)- Improve on current neutronic model X

X 1

of the postulated ' tube break' accident condition.

a) Explicitly model material accumulation external to calciner tube within refractory shell annulus (above and beyond 25 kg hemisphere now presently in the criticality safety basis).

b) Evaluate addition of new independent mass / moderation parameter control to existing basis for safety. If new control adopted, derme all required

' process controls' needed,

7) Revise vendor specification for inconnel-600 (wrought) rolled-alloy calciner X

X X

X 2

tubes, with emphasis on annealing of the rolled product. Ensure vender quality assurance certification meets new design specification.

8) Revise change control process with emphasis on the impacts of material X

X X

2 property changes of equipment / processes. Conduct sensitivky training for all fuel manufacturing process engineers.

9) Evaluate new technology (e.g., Quantran Sensor) for applicability to X

X X

l gamma-scan detection of uranium accumulation within refractory annulus.

j i

• To be completedprior to resentting calciners v

5 m

- ~...

l Dr. B.S. Mallen December i1,1996 Page 7 of 8 Attachment IV 1

3 MARGIN OF SAFETY When wet ADU is discharged from the end of a squin tube, the material enters a i

rotary kiln operating at an atmosphere of 700 - 765 degrees C, thus the moisture j

immediately flashes to steam and is driven off with the offgas flow. As expected, the material removed from beneath the calciner tube contained less than 1000 ppm equivalent H20. An analysis was performed with a half full tube containing optimally moderated l

U308, with a half full refractory annulus containing 50,000 ppm equivalent H20. The j

i resulting Keff(worst credible case) was determined to be 0.92 for this bare system.

d Another run was made in which a half full tube containing moderated U308 with 50,000 i

ppm equivalent H20, with half full refractory annulus containing 1,000 ppm equivalent i

H20. The resulting Keff(conservative 'as found' condition) was determined to be 0.67 for this bare system.

i 4

i As discussed with the NRC Special Inspection Team, the existing analysis assumes a completely full tube, optimally moderated U(5)O2, with a 25 kg hemiaphere buildup beneath the tube (also optimally moderated U(5)O2). The ' accident condition' contained a mass in excess of 700 kgs UO2. The available feed mass in the event of the tube rotation interlock failure was conservatively estimated to be only 80 kgs UO2. The mass in the existing criticality safety analysis is therefore 'very conservative' model of an i

j accident condition (even if one accounts for the residence time of material throughput), as i

no credit is taken for moderation control. Even if the optimally moderated material n

modeled redistributed from inside the tube to outside within the refractory annulus, the i

surface area would increase which would increase neutron leakage, thereby decreasing the effective multiplication of the system.

I i

i 3

).

1 t

4 4

.~.

,,,.,v

Dr.B.S. M Attachment V December 11,1996 Page 8 of 8 POST LINE 3 TUBE BREAK-RESTART CHECKLIST]

l

[_l l

l TO BE SIGNED OFF BY AREA COORDutATEML AMA AAANAGER. AND PRODUCT LSE WutMER PRIOR TO EQUIPMENT RELEASE FOR START UP This checMeet shoued be appensied to the startup cW for each line (14)

{_ ]

l sK1N OFF4NTERINITIALS AC l __

DATE M

CO M AREAa*

^"M lChPL MANAGER I

~This Line to Asesowed For Restert ll l

1 Tighten Tolerance Tusse Rotation Timor 1e Reduce trner to 19 escxmde~

17 Change inihetson RatM Completed _

1e2 Fil Completed te3 Trner seting changed on floor ted Uncoln 5 Recuc automated te5 Functonal test completed i

1TAEC list modded 1_e7 NSR/R Modmod 1s8 O P.Modmod L

1b opereeng Personnet Trained ll 1b1 Trainng Plan WNtten l

tb2 Operatore Trained r

-2~Conect stock aponstering Program ll 2 _eAutomated ches sor tbr rossiebove ocean invet

~

2b Data estribumon coverd in pnxsdure 2c Envronmental Tech retramed - software procedure j

-~cnongos, sensevny to reousie

~"Provos Operator Action Fleepseet (OAR) for Calciner stecte ll 3

3e Change Iniboson Request (CIR) Completed 3 _bSofhmore Services Request (SSR) Completed 3c Fundsonal Test instrucson (FTI) Completed 3.-Moddy ProVor and Perform Funcbonea Test 3d essify Operetng Procedures Fonnellus Calciner Tulse NProgrome

]!

4e Estettoh Tube Flip and Repeicament critene sor 4ei Cent tubee

~-

ERolled tubes de3_, Rolled and Annesied tubes 4b implemented into MPAC I

4c-Tube has been evolueted egemet PM Critane l

Page1 r

-