ML20034H905
| ML20034H905 | |
| Person / Time | |
|---|---|
| Site: | 07000734 |
| Issue date: | 03/18/1993 |
| From: | Asmussen K GENERAL ATOMICS (FORMERLY GA TECHNOLOGIES, INC./GENER |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| 696-2044, NUDOCS 9303220291 | |
| Download: ML20034H905 (14) | |
Text
y i
& CENERAL ATDRNCE
]
l March 18,1993 l
696-2044 VIA OVERNIGHT EXPRESS MAIL
- a i
ATTN: Document Control Desk i
1 U.S. Nucicar Regulatory Commission I
Washington, D.C. 20555 t
l
Subject:
Docket No.70-734; License No. SNM-696: Reply to a Notice of Violation and Plan for Assuring Adequacy of Criticality Safety Analyses
Reference:
Scarano, Ross A., Letter to General Atomics, ATTN: Mr. R.N. Rademacher,
" Notice of Violation /NRC Inspection Report No. 70-734/93-01," ' dated I
February 18,1993 i
Gentlemen:
The enclosed Attachment 1 is General Atomics' (GA's) response to the Notice of
{
Violation issued on February 18,1993 (Reference). This response was prepared pursuant to
'l the provisions of 10 CFR 2.201.
j t
Also enclosed, as Attachment 2, is GA's plan (and schedule) describing the actions it l
]
has taken, and will be taking, to assure that all criticality safety analyses will be reviewed for their adequacy.
GA trusts you will find its corrective action measures to be appropriate and satisfactorf.
j If you should have any questions concerning this response, please contact me at (619) 455-s 2823.
Very truly yours, j
Kehh E. Asmussen, Director Licensing, Safety, and Nuclear Compliance l
KEA:shs Enclosures - as above ec:
Mr. C. A. Hooker, U.S. NRC Region V
'i Mr. John B. Martin, Regional Administrator, U.S. NRC Region V j
Mr. Ross A. Scarano, U.S. NRC Region V l
Mr. Robert E. Wilson, U.S. NRC Headquarters j
( DlJ) 9303220291-930318 a
l
.PDR..ADOCK 07000734-j C
PDR l
l-l 3550 CENERAL ATOMICS COURT. SAN DIEGa CA 921211194 PO BOX BL608, SAN DIEGQ CA 92106 9784 1619) Ab5 3000
{
' to General Atomics' Letter No. 696-2044 Dated March 18,1993 General Atomics' Response to the Notice of Violation / NRC Inspection Report No. 70-734/93-01 NRC Inspection Report No. 70-734/93-01 documents the results of a routine inspection of General Atomics' (GA's) NRC-Ilcensed activities which was conducted at the GA site on January I l-15,1993, and the inspection of GA materials in NRC's Region V offices on January 21-22, 1993.
The subject inspection report cited three violations of NRC l
requirements. Additionally, one non-cited violation was identified. GA's responses to the I
three cited violations are given below. Each violation is restated below, followed by the corresponding GA response.
VIOLATION A:
1 Condition 9 of License No. SNM-696 authorizes the use oflicensed materials in accordance with the statements, representations, and conditions contained in Part 11, " License Specifications," dated July 24,1981, and supplements dated March 16,1992 through June l
19,1992.
i Section 5.0, " Nuclear Safety - Technical Requirements," Part 11 of the license specifications, i
l states:
t "The continued nuclear safety of the licensee's operations shall be assured by limits and procedures documented in accordance with the specifications contained in the following sections-l
[
"5.2, Basic Assumptions "The basic assumptions that shall be utilized in arriving at particular criticality limits are as follows:
" Accident - At least two unlikely, accidental, concurrent, and independent events 1
must occur before a criticality hazard could exist.
t i
t t
. Atttchmint 1 to G:neral Atomics' Letter No. 696-2044 Dated 3/18/93 f
Page 2 of 12 "5.4.3.d.
The geometdcal safe limits shall apply to units meeting the following I
criteria:
3 11.
The unit shall be evaluated for all possible credible accident conditions I
to assure that the safety margin (difference between critical and safe i
Ilmits) is not exceeded.
"5.4.4.
"All geometries analyzed by hand calculations shall have a k,g of 0.90 or less."
Contrary to the above:
~
(a)
As of January 21,1993, Criticality Safety Analyses (CSAs) " Casting Mold" dated September 8,1976, and " Melting Fumace," dated December 27,1972, (for the j
TRIGA Fuel Fabrication Facility's induction fumace), were not adequate to determine that criticality safety criteda were satisfied, in that the CSAs failed to adequately
[
evaluate all credible accident conditions, e.g., flooding of the induction fumace and j
spilling of enriched uranium zirconium metal at the same time. On January 11, l
1993, such an event occurred when a water leak occurred in the induction fumace, j
resulting in molten enriched uranium-zirconium metal spilling from its favorable l
geometry crucible into the unfavorable geometry cavity of the fumace.
j i
(b)
As of January 21,1993, the CSAs did not provide the calculated K,g for geometries for the cylinders analyzed by hand calculations to demonstrate that they were critically j
safe vessels.
j This is a Severity Level IV violation (Supplement VI).
GA'S RESPONSE TO VIOLATION A:
l I
1.
Ileason for the Violation:
a)-
The original criticality safety analysis for the casting fumace did not evaluate the accident scenario involving both loss of favorable geometry and flooding j
i t
h
~
g.
Att:: chm:nt 1 to G n:ral Atomics
- i Letter No. 696-2044 Dated 3/18/93 Page 3 of 12 because, at the time, it was judged to be not credible. The reasoning being, first that the uranium metal can only lose its favorable geometry while it is molten and it is only molten for a brief time during the casting fumace operation and second, any moisture ingress would result in a loss of vacuum followed by a system shutdown. The more than 20 years of casting fumace operating history supported the original ludgement.
b)
The criticality safety analysis did not provide a specific k for each of the eg cylinders (e.g., crucibles and/or molds) analyzed by hand calculations because a buckling conversion (BC) method was used to evaluate the criticality safety of various geometries. The BC method does not yleid a calculated k,g, rather it provides only the assurance that the converted geometry is no more reactive than the " parent" geometry being used as the basis for the conversion. Please note, however, that the parent geometry of the mold was indeed analyzed, as described in Section 3.7.4.2.5 of GA's.SNM-696 Demonstration Volume, and produced a ken of 0.57. This value of k,g was used in nuclearInteraction calculations.
" Buckling Conversion" is a hand calculation method described in Critical Dimensions of Svstems Containina U-235. Pu-239 and U-233. tid-7028, 1964. The BC method is based on the observation that various geometries with the same geometdc buckling have the same neutron leakage and thus, can assumed to be equally reactive. Once a system is evaluated by calculations or experiment to be subcritical, a similar system with a different geometry, that is determined with the BC method to be " equivalent", is also subcritical to the same degree. Needless to say, this method is considered now i
as totally obsolete, but, however, that does not make the cdticality safety i
analysis based on this method invalid.
2.
Corrective Stens Taken and Results Achieved:
i i
A new criticality safety analysis for the TRIGA Fuel Fabrication Facility's casting (induction) furnace has been perfomled. This analysis evaluated the scenario wherein the molten enriched uranium-zirconium metal alloy spills (i.e., loss of geometry control) and the fumace is flooded. The results of this analysis indicate that the
)
I l
Attachm:nt 1 to Gen:rd Atomics
- Letter No. 696 2044 Dated 3/18/93 Page 4 of 12 l
fumace operation is safe from the standpoint of criticality under this scenario even if I
It is double batched; with a calculated k,g of <0.5. Clearly, the original criticality safety analysis based on geomeuy control was very conservative, and unnecessarily l
Implied a hazard when none existed.
The subject criticality safety analysis has been independently reviewed and undergone second level independent review by GA's Criticality and Radiation Safety Committee.
The analysis is contained in GA's Nuclear Safety File No. 530.1.
r The findings of the criticality safety analysis are based, in part, on the observation that there will be no intemal moderation introduced into the uranium-zirconium alloy due to the upset conditions. The expert opinion of metallurgists at GA and academia is that: a) the spillage of alloy into water does not result in the pulverization of the alloy, i.e., the creation of more or less uniform metal / water mixtures is not credible, and, b) the uptake of hydrogen (hydrolization), generated in the water / metal reaction f
by the alloy is negligible, i.e., much less than H/U = 1. The cleanup operation follow-ing the upset condition confirmed the expen opinions, i.e., there was no breakage or fonnation of the alloy into small panicles that could then fomi a slurry, and there was only a negligible amount of alloy lost due to oxidation, which releases hydrogen. As l
a matter of fact, the alloy survived the spillage without excessive degradation and therefore can be, and will be, used for recasting. In the absence of any intemal moderation, the alloy is very safe from criticality, even if it is fully reflected by water and carbon (from graphite components of the fumace).
j i
3.
Corrective Steps that will be Taken to Avoid Funher Violations:
Please see steps taken described in item 2 above. Additionally, the criticality safety analyses of all process stations at GA's TRIGA Fuel Fabrication Facility will be j
reviewed for adequacy by Nuclear Safety. Any analyses found to be not in compli-ance with current requirements w!Il be either revised or replaced with a new analysis.
t 4.
Date When Full Compliance will be Achieved:
GA is currently in compliance.
. to Genzrd Atomics
- Letter No. 696-2044 Dated 3/18/93 Page 5 of 12 VIOLATION B:
Section 3.2.2, " Compliance Functions," Part 11 of the Ilcense specifications, states:
"All functions responsible for assuring compliance with applicable Ilcense requirements and controlling the radiological and nuclear safety and safeguards oflicensed material are part of the Human Resources organization of General Atomics. Namely, these functions are: Nuclear Safety, Licensing, Safety and Nuclear Compliance, Nuclear Material Accountability, Statistics & Measure-ment Control, Security, and Health Physics."
"The Director of Human Resources, or his designee, will establish the necessary policies of operation, cause them to be published in company-wide guides and manuals, and coordinate related activities with operating groups to assure compliance with related polices, procedures, regulations and license conditions...."
Section 3.4.1, " Definitions," of the licensee's Nuclear Safety Guide, defines
" Parameter" as "A characteristic of a system which must be controlled for the purposes of criticality safety, e.g., mass, geometry, concentration and moderation."
Section 3.4.2, " Criteria for Reporting," of the licensee's Nuclear Safety Guide states, in part:
" Situations, events or items such as the following are to be reponed to management and Nuclear Safety for evaluation when [the] quantity of SNM is greater than an ever-safe mass:
1.
Procedure cannot be followed as written...."
Section 3.4.3, " Criteria for Action," of the licensee's Nuclear Safety Guide states, in pan.:
"Upon becoming aware of a situation, event, or item such as the above, Nuclear Safety, with the assistance from additional specialists as necessary, will
Attachm:nt 1 to Genarcl Atomics
- Letter No. 696-2044 Dsted 3/18/93 Page 6 of 12
?
evaluate and assess the safety significance of the occuaence. Nuclear Safety will also assess for adequacy, the conective actions taken, and/or planned.
i "Funher, Nuclear Safety will consult with and advise Licensing, Safety and Nuclear Compilance with regard to proper reponing, both intemally and extemally. Repons to the NRC Operations Center will also be reponed to the appropriate Regional Administrator.
"The criticality safety event reponing criteria and appropdate response actions are given in Table 3.4-1."
Table 3.4-1 requires, when there is substantial degradation or loss of a control parameter resulting in only one parameter remaining under control:
" b.
Notification of the NRC as son as possible within four (4) hours of i
discovery."
Contrary to the above, at about 1:10 pm on January 11,1993, while melting low enriched uranium-zirconium alloy of more than an ever-safe mass of U-235, an operator and the TFFF's QC inspector became aware of a significant process upset that resulted in termination of the operation in progress; shonly after the event, at about 1:30 pm, the TFFF's manager became aware of the event; the event resulted in an unusual condition that prevented completion of the process in accordance with the established operating procedure, " Induction Melting and Pouting of U/Zr and l
U/Zr/Er Alloy;" but this event was not reponed to the NRC until 1:04 pm on j
January 12,1993, a pedod exceeding four hours.
This is a Severity Level IV Violation (Supplement VI).
GA'S RESPONSE TO VIOLATION B:
1.
Reason for the Violation:
1 The Manager of GA's TRIGA Fuel Fabrication Facility knew that the quantity of U-235 (587 grams) being batchwise processed was less than the minimum critical mass i
a i
I
~
Attachmsnt 1 to General Atomics' Letter No. 696-2044 Dated 3/18/93
[
Page 7 of 12
+
t 4
c of U-235 (805 grams under optimum conditions of moderation and reflection). He knew he did not have an unsafe condition, and therefore, in his judgement, he did not believe the matter required reporting to Nuclear Safety. He also assumed that the station (the casting fumace) mass limit of 1600 grams U-235 had been established i
based on an accident analysis that included the simultaneous loss of geometry and i
flooding.
l It is to be noted that within four (4) hours of the arrival of the Health Physics technician on the scene (at about 9:10 a.m.), the Manager of Health Physics, the Director of Ucensing, Safety and Nuclear Compliance, the Manager of Nuclear Safety, the Vice President of Human Resources, and the NRC were notified of the situation.
(The NRC notification occurred at 1:04 p.m.).
GA believes the reasons for this violation to be the following:
I a)
Personnel not adhering to the requirements for intemal reporting of criticality 1
l safety events as stated in GA's Nuclear Safety Guide.
i t
b)
Failure to effectively and clearly communicate to all users of special nuclear material at GA the requirements for intemal and extemal reporting of criti-I l
cality safety situations, events or items and the significance and importance of j
adhering to those requirements.
2.
Corrective Steps Taken and Results Achieved:
i a)
The casting fumace station mass limit was immediately (on January 12,1993) reduced from 1600 grams U-235 to 457 grams U-235 and so posted by the Manager of Nuclear Safety. The reduced limit of 457 grams of U-235 is the ever-safe limit for 20% enriched uranium.
b)
The above action (Item a) had the effect of stopping production at the TRIGA Fuel Fabrication Facility until a new criticality safety analysis could be completed and independently reviewed (see below) -- i.e., until March 11, 1993.
4 J
Attachmsnt 1 to Gtn:rd Atomics
- I tetter No. 696-2044 Dated 3/18193 f
~
Page 8 of 12 j
c)
The Manager of Nuclear Safety conducted a special training session for all l
TRIGA Fuel Fabdcation Facility personnel on January 12,1993. This special training se:sion focused on the reporting requirements relative to NRC Bulletin j
91-01, with emphasis on internal reponing requirements.
l d)
The Manager of Nuclear Safety met with the TRIGA Fuel Fabrication Facility's upper management, i.e., the Director of the TRIGA Group and the Senior l
Vice President of Advanced Technologies, and discussed the role of Nuclear Safety and the significance and imponance of adhering to the intemal and l
extemal reporting requirements defined in GA's Nuclear Safety Guide relative i
j i
e)
On January 15,1993, the Director of 1.icensing, Safety and Nuclear Com-l pliance issued a memo to all SNM usen and all site management penonnel involved with activities using SNM. This memo stressed the importance of:
recognizing any failure or degradation of a criticality control, reporting such an occurrence to appropriate parties and management levels, and that it be done l
In a timely, expeditious manner. GA's response to NRC Bu!Ietin 91-01, l
l Including guidance /criteda for reporting, was attached to this memo.
1 f)
Guidance / criteria for intemal reporting of criticality safety events, situations, or items was posted at the TRIGA Fuel Fabdcation Facility. (Similarly, the 1
same information has been posted in GA's Building 39, aka SVB; where pre-parations are underway to conduct a project of limited duration involving the j
use of greater than 350 grams U-235 in the form of highly enriched uranium.
i g)
The content of GA's Nuclear Safety general training lecture has been exten-sively revised to place special emphasis on safety practices with examples from I
actual operating experience, and to stress the importance of adhering to GA's j
internal reporting requirements relative to NRC Bulletin 91-01. The exam l
given at the conclusion of the training session has also been improved by f
adding additional questions (including questions related to reporting require-ments).
I i
i i
- to General Atomics
- Letter No. 696-2044 Dated 3/18/93 l
Page 9 of 12 i
l l
h)
As discussed above in "GA's Response A," a new criticality safety analysis for the TRIGA Fuel Fabrication Facility's casting fumace has been perfonned.
l This analysis evaluated the scenario wherein the molten enriched uranium-zirconium metal alloy spills and the fumace is flooded. The results of this analysis demonstrate that the fumace is safe from criticality under this scenario even if it is double batched; with a calculated k of < 0.5.
The subject g
criticality safety analysis has been independently reviewed and undergone second level independent review by GA's Criticality and Radiation Safety Committu. The analysis is contained in GA's Nuclear Safety File No. 530.1.
Based on the above, the casting fumace's station mass limit was reinstated at 1600 grams U-235 on March 11,1993.
3.
_ Corrective Steys that will be Taken to Avoid Funher Violations:
i GA believes that the above described corrective steps taken will be adequate to avoid further violddons. However, additionally, the frequency at which Nuclear Safety inspects the TRIG A Fuel Fabrication Facility will be temporarily increased from at least once every year to at least quarterly, with ad hoc inspections and reviews conducted on an as needed basis. In twelve months time, the need for continuing inspections at this increased frequency will be reviewed and revised as appropriate.
i 4.
Date When Full Compliance will be Achieved:
GA is currently in compliance.
VIOLATION C:
j Condition No. I 1 of License No. SNM-696 states, in part:
r
" Records of all safety-related repons and analyses shall be retained as follows:
a.
Copies of criticality and radiation safety analyses shall be retained for at least l
2 years or 6 months after a project is terminated, whichever is longer."
i r
l
Att: chm:nt 1 to Grneral Atomics
- Lett:r No. 696-2044 D tid 3/18/93 Page 10 of 12 Section 3.7.5.3, Demonstration Volume I, stated in part that the cmcible used to melt enriched uranium-zirconium alloy in the TRIGA Fuel Fabrication Facility's induction fumace had been analyzed for criticality safety in a manner similar to that for the casting mold.
Contrary to the above, as of January 21,1993, the licensee had no record of a criticality safety analyses specific to the cmcible.
This is a Severity 1.evel V Violation (Supplement VI).
GA'S RESPONSE TO VIOLATION C:
1.
Reason for the Violation:
The criticality safety analysis for the cmcible was an integral pan of the criticality safety analysis of the casting fumace which was perfomled in the early 1970's. In Section 3.7.5.3 of SNM-696 Demonstration Volume 1, it is stated that "... the 5 inch diameter crucible has been analyzed according to a similar argument" [similar to the analysis for the mold]. The criticality safety analysis of the casting fumace, on which the presentation in Section 3.7.5.3 was based, was attached to a licensing submittal to the NRC; which in the early 1970's was an acceptable means of docu-ment control and retention because the analysis was both independently reviewed and subjected to a second level independent review by GA's Criticality and Radiation Safety Committee. Therefore, the retention of the analysis in the Licensing files rather than in the Nuclear Safety files should not be viewed as a failure to retain the criti-cality safety analysis. Funher, as discussed below, there was really no need to perform a specific or additional analysis to establish the criticality safety of the crucible since the analysis for molds is equally and totally app!! cable to the cmcible; as stated in Section 3.7.5.3 of SNM-696 Demonstration Volume 1.
In Section 3.7.5.2 of SNM-696 Demonstration Volume I it is stated that, "the proper amount of uranium and zirconium are weighed and charged into the induction furnace crucible having a 5 inch diameter and a volume of 3.2 liters." In Section 3.7.5.3 of the same volume it is stated that, "... the 5 inch diameter crucible has been analyzed according to a similar argument" [similar to the analysis for molds].
.. = =. _ _. _ _
L,:
i Attechmznt 1 to Gen:tal Attmics*-
.696-2044 Dated 3/18/93
- l Letter No.
Page 11 of.12
'l
/;
k The criticality safety analysis for molds, as described above in "GA Response A," was.
based on the use of data on favorable geometries extracted from manuals widely used by the nuclear industry, such as TID-7016, Nuclear Safetv Gulde, and 71D-7028, -
1 Critical Dimensions of Systems Containina U-235. Pu-239. and U-233;and buckling conversion calculations. The analyses for molds (cylinders) are equally.'
applicable to any other SNM process equipment, regardless ofits name, provided that j
lt has equivalent favorable' geometry. Therefore, no separate criticality safety analysis j
is needed for the crucible as long as its geometry was specified to be 5 inches (or less) l In diameter. Better, less ambiguous language in Section 3.7.5.3 might have been "...-
5 inch diameter cmcible can be justified as safe according to an argument similar to
[
that for molds."
Today, the justification for the criticality safety of the casting fumace cmcible would be based either on Standard Limit Type B (3.6 kg of U-235 in a 3.6 liter container) j
- or on Standard Limit Type C (10 kg of U 235 in 5 inch ID cylinder)'or both. The
{
~
standard limit types are defined in SNM-696 Demonstration Volume 11. The nuclear criticality safety evaluation of the casting fumace was, however, carried out in the early 1970's and pre-dates the adoption of Standard Limit Types defined in SNM-
.l I
696 Demonstration Volume 11.
i l
2.
Corrective Steps Taken and Results Achieved:
iy A new cdticality safety analysis for the casting fumace has been performed. 'This l
analysis has been independently reviewed and has undergone second level independent
{
review by GA's Criticality and Radiation Safety Committee. The analysis is contained In Nuclear Safety File No. 530.1. ' The results of this analysis demonstrate that' l
geometry control by the cmcible is not necessary for criticality safety control during operation of the casting fumace. The casting fumace is safe from criticality (with a-l I
calculated kg <0.5) even with double batching, flooding and loss of geometry-control (i.e., SNM spilled from cmcible and/or mold).
3.
Corrective Steps that will be Taken to Avoid Further Violations:
.j 1
GA believes that the above described " Corrective Steps Taken" will be adequate to avoid funher violations. Additionally, the criticality safety analyses for all process jj i
j
Attachm:nt 1 ts Grneral At:mics' Letter No. 696-2044 Dated 3/18/93 Page 12 of 12 stations at GA's TRIGA Fuel Fabrication' Facility will be reviewed for adequacy by-Nuclear Safety.
4.
Date When Full Compilance will be Achieved:
GA is currently in compliance.
[
i
)
i l
i i
i
. 13 General Atomics
- Letter No. 696-2044 l
Dated March 18,1993 l
l l
l General Atomics' Plan and Schedule for Reviewing Criticality Safety Analyses for Adequacy i
The cridcality safety analysis of each process station stIII in operadon at General Atomics will be reviewed by Nuclear Safety to verify its adequacy. The order in which the j
criticality safety analyses will be reviewed will be established by Nuclear Safety based on an assessment of the relative probabilldes that the stations may experience a degradation of a i
cridcality control.
l If, as a result of the verification process, a criticality safety analysis for a panicular
{
stadon is found to be incomplete, inadequate, or erroneous, the analysis will be revised, l
corrected or re-done, as appropriate, in accordance with the requirements and specifications contained in GA's " Nuclear Safety Evaluation Guide" and " Nuclear Safety Guide." The corresponding Nuclear Safety files will be updated accordingly.
Fun.her, the verification exercise may identify a need for corrective actior.s at the.
e operating level, for example, a posted stadon Ilmit may need to be revised and/or for additional (perhaps temporary) nuclear safety related restdctions may need to be imposed l
on certain activities involving the use of SNM. If a new limit (s) and/or work restrictions are l
Imposed, then cenain personnel working at the affected facility will be given appropdate additional ad hoc training by Nuclear Safety.
l t
it is difficult to accurately estimate the time needed to complete the review of the I
criticality safety analyses for all process stadons.
However, assuming no substantive i
deficiencies will be discovered, GA has established a target date of June 20,1993 for
.l completing the for verification of adequacy.
If substantive deficiencies are identified, additional time will, or course, be required l
4 to revise, correct or re-generate the conesponding cridcality safety analyses. Not knowing l
4 how many, or if any, deficiencies might be discovered, it is not possible to predict how much additional time might be required. GA is, however, committed to completing this effort i
4 expedidously. Accordingly, GA will on a "best effort" basis attempt to complete any and all necessary remedial acdons within an additional three (3) months. If this can be achieved, i
the entire review and remediation effort will be completed on or before a target date of September 30,1993.
l f
I I
f I
,