ML20133H271
| ML20133H271 | |
| Person / Time | |
|---|---|
| Site: | 07001113 |
| Issue date: | 07/26/1984 |
| From: | Richards B NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| Shared Package | |
| ML20133H069 | List: |
| References | |
| FOIA-85-158 NUDOCS 8508090269 | |
| Download: ML20133H271 (8) | |
Text
.'
s ENCLOSURE MEETING
SUMMARY
~
Licensee:
General Electric Company Facility:
Wilmington Manufacturing Department License No.:
SNM-1097 Docket No.:
70-1113
SUBJECT:
STATISTICAL METHODS FOR CONTROL CHART LIMITS On July 10, 1984, representatives of the General Electric Company met with NRC Region II personnel to discuss the denial by General Electric of a violation concerning statistical methods used to calculate measurement control chart limits.
The attendance list is attached as Appendix A to this Summary.
The meeting was held at the request of the General Electric Company.
NRCrepresentativessummArizedtheinspectionfindings,presentedareviewofthe applicable requirements associated with statistical control systems, reviewed appropriate statistical techniques described in recognized quality control textbooks, and summarized the General Electric statistical methods.
NRC personnel also provided General Electric representatives with a verbal explanation of the review criteria which was used to approve all Fundamental Nuclear Material Control (FNMC) plans. A review of the General Electric FNMC plar. revealed that the commitments contained in the plan were consistent with the NRC review criteria, but inconsistent with current General Electric practices.
General Electric representatives presented their justification for using their
-statistical control methods. A General Electric representative indicated that the commitments in the FNMC plan were not consistent with what General Electric
, had intended to convey, but that it was clear that General Electric's practices were not consistent with the FNMC Plan approved by the NRC. The NRC stated that if GE wanted to utilize such a technique, they should provide appropriate background material to the NRC for proper review.
Copies of the visual aids used by General Electric and NRC in th.eir July 10, 1984 presentation are attached as appendices B and C, respectively.
d[W O L~
7/26/ 4 i
Briah L. Richards #
Date Statistician Attachments:
A - Attendance List B - General Electric Visual Adis C - NRC Visual Aid 8500090269 e50621
%NE
-158 PDR
n D
ATTACHM'ENT A
~
ATTENDANCE LIST
~
Attendance at the General Electric - NRC meeting of July 10, 1984 at the NRC Region II Office was as follows:
General Electric Company C. M. Vaughan,. Manager, Licensing & Nuclear Materials Management R. I. Parnell, Control Programs Engineer W. Tucker, Statistician, Corporate R&D Center Nuclear Regulatory Commission K. P. Barr, Chief, Nuclear Materials Safety and Safeguards Branch, Division of Radiation Safety and Safeguards, Region II E. J. McAlpine, Chief, Material Control and Accountability Section, NMSS Branch C. N. Smith, Chief, Nuclear Materials Control Licensing Section, Fuel Facility Safeguards Licensing Branch, Division of Safeguards, NMSS
- b. W. Jones, Statistician, Material Control and Accountability Section B.
L.
Richards, Statistician, Material Control and Accountability Section e
s'..
I O
ATTACHMENT B
~
g d.,
// (p s.72
- [
INSPECTION REPORT 70-113/83-28 Copy A p,.,1; recarer.. is r.c.t to be rcrrcCJcedete mtspeciU 0 l
appro cal of II:ll NOTICE OF VIOLATION B:
10 CFR 70.57(b)(5) REQUIRES THE LICENSEE TO REVIEW AND APPROVE, BEFORE USE, WRITTEN PROCEDURES FOR ANALYZING PROGRAM DATA AND CONTROLLING MEASUREMENT PERFORMANCE.
IN ADDITION, 70.57(B)(10) REQUIRES THE LICENSEE TO EVALUATE WITH APPROPRIATE STATISTICAL METHODS ALL PROGRAM DATA AND INFORMATION USED TO ESTABLISH PARAMETERS PERTAINING TO SPECIAL NUCLEAR MATERIALS CONTROL AND ACCOUNTING.
CONTRARY 'IO THE ABOVE, THE LICENSEE REVIEWED AND APPROVED PROCEDURE COI 1002, DATED MAY 6, 1983, WHICH DESCRIBES THE METHOD 3 USED TO CALCULATE CONTROL CHART LIMITS FOR CONTROLLING MEASUREMENT PERFORMANCE, AND FAILED TO DETECT THAT THE PROCEDURE CONTAINED THE USE OF 1
l INAPPROPRIATE STATISTICAL TECHNIQUES.
THE SUBSEQUENT USE OF THIS PROCEDURE CAUSED THE LICENSEE TO CONTROL -
MEASUREMENT PERFORMANCE AT A LEVEL WHICH WAS INCONSISTENT WITH THE REQUIREMENTS OF 10 CPR 70.57(b)(11).
SEVERITY LEVEL: IV CMV 7/10/84 m
REGULATORY REFERENCES s
10 CFR 70.57(5)(5):
(5' 7"E PROGRAM SkALL INCLUDE PROVISIONS FOR THE REVIEW AND APPROVAL, BEFORE USE, OF WRITTEN PROCEDURES FOR:
(i) PREPARING OR ACQUIRING, MAINTAINING, STORING AND USING REFERENCE STANDARDS,
( ii) CALIBRATING MEASUREMENT SYSTEMS, PERFORMING BULK MEASUREMENTS, OBTAINING SAMPLES, AND PERFORMING COMPOSITIONAL ANALYSES, (iii) RECORDING, ANALYZING AND REPORTING THE PROGRAM DATA AND INFORMATION, AND (iv) CONTROLLING MEASUREMENT PERFORMANCE.
10 CPR 70.57(b)(10):
(10) THE LICENSEE SHALL EVALUA7E WITH APPROPRIATE STATISTICAL METHODS ALL PROGRAM DATA AND INFORMATION, AND RELEVANT PROCESS DATA USED TO ESTABLISH BIAS CORRECTIONS AND THEIR ASSOCIATED UNCERTAINTIES, RANDOM ERROR VARIANCES, LIMITS FOR SYSTEMATIC ERROR, AND OTHER PARAMETERS PERTAINING TO SPECIAL NUCLEAR MATERIALS CONTROL AND ACCOUNTING MEASUREMENTS, AND TO CONTROL MEASUREMENT PERFORMANCE PURSUANT TO S 70.58 ( f).
BIAS CORRECTIONS SHALL BE MADE BY AN APPROPRIATE STATISTICAL
~
PROCEDURE.
10 CPR 70.57(b)(11):
(11) THE LICENSEE SHALL ESTABLISH AND MAINTAIN A 1
STATISTICAL CONTROL SYSTEM, INCLUDING CONTROL CHARTS AND FORMAL STATISTICAL PROCEDURES, DESIGNED TO MONITOR THE o
QUALITY OF EACH TYPE OF PROGRAM MEASUREMENT.
CONTROL CHART LIMITS SHALL BE ESTABLISHED TO BE EQUIVALENT TO LEVELS OF SIGNIFICANCE OP 0.05 AND 0.001.
WHENEVER CONTROL DATA EXCEED THE 0.05 CONTROL LIMITS, THE LICENSEE SHALL INVESTIGATE THE CONDITION AND TAKE CORRECTIVE ACTION IN A TIMELY MANNER.
THE RESULTS OF THESE INVESTIGATIONS AND ACTIONS SHALL BE RECORDED.
WHENEVER THE CONTROL DATA EXCEED THE 0.001 CONTROL LIMITS, THE MEASUREMENT SYSTEM WHICH GENERATED THE DATA SHALL NOT BE USED FOR MATERIAL CONTROL AND ACCOUNTING PURPOSES UNTIL THE DEFICIENCY HAS BEEN BROUGHT INTO CONTROL AT THE 0.05 CONTROL LEVEL.
CMV 7/10/84
6 3
GENERAL ELECTRIC'S POSITION
~
COI #002 REVIEWED BY APPROPRIATE & COMPETENT PERSONNEL COI #002 SUBSEQUENTLY REVIEWED INTERNALLY MOST RECENTLY REVIEWED BY GE-CR&D 10 CPR 70.57(b)(10)
SIMULTANEOUS STATISTICAL INFERENCE SIMPLE APPLICATION DERIVED FROM PROBABILITY THEORY RECOGNIZED IN LITERATURE 10 CFR 70.57(b)(11)
PPOGRAM MEASUREMENTS CCNTROLLED AT 0.05 AND 0.001
~
S1MULTANEOUS STATISTICAL INFERENCE PROBLEM TECHNICAL MISUNDERSTANDING PROCEDURAL CLARITY CMV 7/10/84
4 SIMPLE FORM SIMULTANEOUS STATISTICAL INFERENCE CONTROL REQUIREMENTS MEASUREMENT QUALITY CONTROLLED SO THAT PROGRAM MEASUREMENTS ARE REPORTED AT THE 0.05 AND 0.001 SIGNIFICANCE LEVEL MEASUREMENT SEQUENCE SINGLE DETERMINATION OF S'TANDARD SEQUENCE OF PROGRAM MEASUREMENTS SINGLE DETERMINATION OF STANDARD s.
DERIVATION OF CONTROL LIMITS Pr(PASS) = Pr(PAS 3 1st TEST & PASS 2nd TEST)
= Pr(PASS 1st TEST) Pr(PASS 2nd TEST)
~
.95, ALARM TWO SIDED LIMITS
=
.999, OUT-OF-CONTROL TWO SIDED LIMITS IF.95 = X2 THEN X =.9746794
. EACH TAIL Pr = 0.0126603 i
/ 0/84
_a m
Am
.su.
e a.s.s,w mS--.
m h
e e
5 DICHROMATE TITRATION SIMPLE FORM MEASUREMENT SEQUENCE SINGLE DETERMINATION OF STANDARD SEQUENCE OF PROGRAM MEASURESENTS SINGLE DETERMINATION OF STANDARD DICHTOMATE MEASUREMENT SEQUENCE SINGLE DETERMINATION OF NBS 950b SRM PROGRAM MEASUREMENTS (MAX 15 UNKNOWNS)
SINGLE DETERMINATION OF NBS 950b SRM
-=
~
DICHROMATE MEASUREMENT SEQUENCE IS THE SAME AS THE SIMPLE FORM EXAMPLE OF SIMULTANEOUS STATISTICAL INFERENCE WTT 7/10/84
g ATTACHMENT C CONTROL LIM.IS FOR PPM IN RAD WASTE Se IMR.fal G
CmTLIMIT c
48 G
IITER.B5 C0NTLIMIT i
3g LOVER.85
~
CONT LIMIT
, L[ =
25 ppm S=5.65
~ ~ ~ ~ ~ ~
28 r
L0ER.E91
~
Emi LIMIT C
g 18 MB UIE c
~_.__._.___..___
(= }{ }
3 l
g G = General Electric meth'od C = classical method LIMITS IN EFFECT-SEPT '93
s 1
Y
- s G EYE R'At h E LE CTRI C WILM,pGTONy
- QRING DEPARTMENT GENERAL ELECThC'CMvMY FC) Boxu a. wlLMWGTON. NORT]99OLI,N,(~?q, 7 (919) 343-5656 June 7, 1984 Mr. J. Philip Stohr, Director Division of Emergency Preparedness &
Materials Safety Programs U. S. Nuclear Regulatory Commission, RII Post Office Box 2203 Atlanta, Georgia 30301
Dear Mr. Stohr:
References:
(1) NRC License SNM-1097, Docket 70-1113 (2) NRC Inspection Report 70-1113/83-28, 10/31/83 (3) Letter, C. M. Vaughan to J. P. Stohr, 11/29/83 (4) Letter, J. P. Stohr to J. A. Long, 1/5/84 (5) Letter, C. M. Vaughan to J. P. Stohr, 2/3/84 (6) Letter, J.
P. O'Reilly to J. A. Long, 5/11/84 Thank you very much for your letter reporting your review of General Electric's responses to the findings of the referenced inspection.
General Electric replies pertaining to the items of apparent noncomoliance are alven in the attachment to this letter.
We welcome further di5cussion with your staff on these items or to our related replies for further clarificacion.
Pursuant to 10 CFR 2.790(d), General Electric Company reguests
.that the attachment to this letter be withheld from cublic disclosure since this attachment identifies details of General
. Electric's control and accounting procedures for safeguarding licensed special nuclear material.
Very truly yours, GENERAL ELECTRIC COMPANY
' h.
'8
A Charles M. Vaughan, Manager Licensing & Nuclear Materials Managemer.t M/C J26 CMV:cd SGD-I Attachment
--G407140t95-940622 hff m
- DR ADOCK 07COiT13 g
3/
C.
FDR
(
GENER AL h ELECTRIC Mr. J. Philip Stohr June 7, 1984 ATTACBMENT The following violation was issued in the May 11, 1984 letter from J. P. O'Reilly as a new Notice of Violation and identified as.
Violation A from inspection report 70-1113/83-28 was withdrawn.
10 CFR 70.51(c) requires the licensee to establish written material control and accounting procedures.
In addition, 70.57(b)(12) requires the licensee to provide a recod system which retains all data, information, reports, and documents generated by the measurement control program.
Contrary to the above, the licensee did not establish a written procedure to describe the methods and criteria to be used for evaluating the applicability of current control chart limits.
In addition, the licensee's lack of a written procedure for this measurement control function resulted in the licensee failing to retain the data used in the control chart limit evaluation performed on July 12-15, 1983, associated with the NBL 97 standard (used to monitor uranium concentration' measurements of uranium dioxide powder.)
1 This is a Severity Level V violation (Supplement III).
General Electric admits the violation as stated in the revised NRC report and letter of 5/11/84.
This position is consistent with our response of 11/29/83.
The violation occurred because the procedures thar WMD was operating to during the inspection did not require changing control limits calculated during the current period when there were no significant dif ferences from those used during the prior period.
Therefore, test records for these calculations were not required to be retained under the procedures that existed.
l.,
s
~
GENER AL h ELECTRIC Mr. J. Philip Stohr June 7, 1984 Attachment - Page 2 Procedure COI 002 has been revised to stipulate that all control limits will be recalculated and used whether or not significant j
differences exist with prior period limits.
This action requires i
the generation of a record which will be retained.
Records are retained in accordance with procedure P/P 140-13.
The control limits in question have been recalculated and updated.
General Electric is currently in compliance and no further action is planned.
The following was received in the May 11, 1984 letter from J. P. O'Reilly as Enclosure 2 and addressed General Electric's 2/2/84 response to inspection report 70-1113/83-28, Violation B:
We have reviewed your response to Violation B and still consider it to be a violation.
We have consulted with the Fuel Facilities Safeguards Licensing Branch (Office of Nuclear Material Safety and Saf eguards) and have determined that the statistical methods described in Introduction to Statistical Analysis, Dixon and Massey, McGraw-Hill, 1957, (pages 332-338) are not applicable in determining control limits associated with measurement control.
heneral Electric continues to deny the violatien.
10 CFR 70.57(b)(10) requires evaluation of all program data and information with " appropriate statistical methods" and we firmly believe that the analytical techn! ques used by General Electric are appropriate to the application.
The analytical techniques used at GE-WMD were selected in 1981 by expert statistical technicians who judged them to be fully compliant with the requirements of the GE measurement control program and 10 CFR 70.57(b)(11).
They have proven to be fully satisf actory from an operational standpoint and have been audited by internal and external functions several times without adverse criticism.
m
cj s
GENER AL h ELECTRIC s
Mr. J. Philip Stohr June 7, 1984 Attachment - Page 3 The techniques are well-founded in standard technical publications.
In the interim since the issue surf aced, we also requested and received an evaluation of our methods by authorities in the GE Corporate Research and Development Information Technology Branch, who determined that the WMD analytical methods were a correct application of simultaneous statistical inference to satisfy the regulatory requirements.
It is apparent that there continues to be a lack of understanding between GE-WMD and the NRC which has led to contrary conclusions as to whether the GE-WMD measurement control techniques are correctly applied.
We believe that a meeting of the technical people is warranted to resolve the misunderstanding and discuss the issues.
We suggest that this meeting be held in Washington and attended by members of our Corporate Research and Development Center, members of the Fuel Facility Safeguards Licensing Branch, as well as representatives of Region II and GE-WMD.
u.
6 CMV/REDF
~
/cd
% M N,,,
'o UNITED STATES
(.
g NUCLEAR REGULATORY COMMISSION
, [, ",
g 4,
REGION 11
/ *z 101 MARIETTA ST., N.W.. SulTE 310o O
E ~ ~ -*6 o ATLANTA, GEORGIA 303o3 ArR 0 81982 General Electric Company ATTN: Mr. J. A. Long, General Manager Wilmington Manufacturing Department P. O. Box 780
)
Wilmington, NC 28402 Gentlemen:
Subject:
Report No. 70-1113/82-07 This refers to the routine safeguards inspection conducted by Mr. c. W. Bates of this office on March 15-18, 1982, of activities authorized by NRC License No. SNM-1097 at General Electric facility and to the discussions of our findings held with Mr. C. M. Vaughan at the conclusion of the inspection.
Areas examined during this inspection included your program for nuclear material control and accountability under the applicable provisions of Title 10, Code of Federal Regulations, Part 70, "Special Nuclear Material," and specific license conditions. Within these areas, the inspection consisted of selective examina-tions of procedures and representative records, interviews with personnel, performance tests, and observations by the inspector.
Within the scope of this inspection, no violations or deviations were disclosed.
We have examined actions you have taken with regard to previously identified enforcement matters. These are discussed in the enclosed inspection report.
One new unresolved item resulted from this inspection and is discussed in the enclosed report. This item will be examined during subsequent inspections.
In accordance with Section 2.790(d) of the NRC's " Rules of Practice", Part'2,
., Title 10, Code of Federal Regulations, activities involving safeguards and security measures are exempt from public disclosure; therefore, the enclosed report with the exception of the cover page, which is an inspection summary, will not be placed in the Public Document Room.
Should you have any questions concerning this letter, we wi'l be glad to discuss them with you.
Sincerelys l
\\
)
^
A. F. Gibson, Chief Technical Inspection Branch Division of Engineering and Technical Programs Enclosures (See Page 2)
@ 2 h 2 3 t,/r3 7 f 4 y
l
General Electric Company 2
APR 0 81982
Enclosures:
1.
Inspection Report No. 70-113/82-07 (Exempt from Disclosure) 2.
Inspection Summary (Not Exempt) cc w/ enc 1:
C. M. Vaughan, Manager Licensing and Compliance Audits 1
8
'o, UNITED STATES 8
NUCLEAR REGULATORY COMMISSION
- U\\"
REGION il i
101 MARIETTA ST.. N.W.. sulTE 3100 8
ATLANTA, GEORGIA 30303 g,.*.Q... j At R 0 81882 3
Report No. 70-1114/82-07 Docket No. 70-1113 Safeguard Group No. III i
Licensee:
General Electric Company Wilmington, NC 28401 l
Date of Inspection:
March 15-18, 1982 Type of Inspection:
Unannounced Material Cortrol and Accountability Inspector
- b. W <M-n.d.44 /
4h/f L J. W. Batds, Senior Safeguarcs Chemi:t Datd St'gned 1
Approved by:
0 i ( E *
.s 2
U E. J. McAlpine. Chief,\\ Material Control and Date Signed j
4: car.taci l ':, Sect' or.
Tecnnical Inspection Branch Division of Engineering and Technical Programs l
l Inspection Summary l
Areas Inspected:
Measurements; Measurement Control Program; and Followup on Previous Inspection Findings.
l The inspection involved 31 inspector hours on site by one NRC inspector and was cegun curing the regular hours.
l Results: The licensee was found to te in compliance with NRC requirements in the
, three areas examined during the inspection.
l l
s-;
.PJ 6 S.
Copy
/
of M Copies 7
Pages This Document is not to be reproduced without specific approval of IE II b
' N 3 b.5T4f
?
REPORT DETAILS Report No. 70-1113/82-07 l
1.
Key Persons Contacted H. Stern, Acting Mar ager, MT&E0 W. J. Hendry, Manager, Regulatory Compliance C. M. Vaughan, Manager. Licensing and Cortpliance Audits J. R. Nettles, Acting Manager, P&P R. C. Hawk, Acting Manager, Materials Operations R. C. Stein, Acting Manager, NMM B. F. Bentley, Acting Manage, Fuel Manu'acturing Operations R. I. Parnell, Control Program Engineer R. L. Bruce, NMM Specialist The inspector also interviewed several other licensee employees.
- Denotes these cresent at the ex t interview i
2.
Licensee Action on P*evious Inspection Findings (Closed) Violation ($1-05-01) - Failure to investigate anc cocument all cut-of-control situations. This war determined not to be an out-of-control condition, but resultad from a human error in reporting data. The condition has been corrected.
(Closed) Violation (81-05-02) - Failure to certify all personnel who perform ma+erial control and accounting measurements.
This resulted from an oversight of the training program.
The two inspector identified employees were certified for pellet sampling on November 4,1981.
(Closed) Violation (81-05-03) - Failure te establish, maintain and follow a system to measure all SNM discarded.
This resulted from a transposition error in reporting data.
The final reports are manually overchecked by a
. person independent of the inittai reporting.
I (0 pen) Violation (81-05-04) - Failure to verify the assigned value of standards by two methods or 1.iboratories. An exception to this requirement will be sought by the licensee.
l (Closed) Violation (81-01-01) - Failure to maintain a sufficient procedure.
The procedure entitled Nuclear Material Receipts, NMMP 109 has been revised to include instructions for determining quantities of nonvolatile impurities.
3.
MC 85206B - Measurements and Statistical Controls The program for monitoring and controlling the performance of standard measurements for the more important analytical methods used for
2 accountability purooses was examined.
The analytical methocs for uranium and U-235 have limit; determined by statistical tecnniques and based upon the historical performan:e of the method.
Each measurement system has a warning 1;mit and an out-of-control limit that are re-evaluated at least once each material balance perioc.
The measurement methoc, the frecuency of out-of-control situations, the periods examined, and the number of stancard
{
measurements performed during the period are presqnted in Table I.
Any time that the alarm or contr31 limits are exceeced. the investigation, the procable cause of the occurrence, and the corrective actions taken are l
documented ts shown in Table I, two points exceeded the out of-control I
limits f* the periods examined, one of which was the 25 ppm uranium standard for the 'iquid uranium measurement (LUM) system and the I
spectrographic metallic imourity standards. The LUM standard exceeded the upper control limi on N:vemcer 30, 1951 anc tne test -as shut down.
TES cleaned and changed the mylar protective covering on the detector.
The measurement system was then recalibrated by Chemet laboratory personnel and the system was placed back in service.
In the other case, the control
'im ts
'e-the eta'
' ruri ts standa-c (95-4) was ; ce-stated on d
W e.ce
..:. '.95'.
e -
e-
- ~ ;- e oas - ac.e
.e-i., :- - ec ' := ne calibration. A new total for nonvolatile metal imou-ities was cetermined and five acc'ticca' stanca-ds un.
No significant c'#fe en:es were n:ted.
One question arose concern:ng the licensee's practice of responding to out-of-control conditions and is discussed in paragraph 5 of this report.
The documentation suoporting UF6 measurements and scale calibration was evaluated during the inspection.
UF6 receipts vary from one to five cylinders per shipment. The UF6 scale is adjusted to :ero and a checkweight (CW) cylinder (6357.0 pounds) applied to the scale prior to weighing a cylinder. The weight of the CW cylinder must agree within 22 po ads of the stated value. The CW cylinder is then removed and the cero point verified.
An unknown cylinder is gross weighed, removed and tne CW cylinder reweighed.
The CW cylinder is weighed before and af ter the weight of each unknown cylinder.
The UF6 scale prints to the nearest pound but can be read from the cigital readout of the scale to the nearest 0.1 pound. The weight of each unknown cylinder is corrected for scale bias.
The bias, determined from the weighing of the CW cylinder before and after the weigning of each unknown cylinder is averaged anc applied to the digital reacout weight accordingly.
The cylincer weight is tnen normally rounded to the nearest f;ll pourd and entered into MICS for accountability purposes.
A cumulative rouncing bias is maintained to hep the cumulative rounding bias near zero. The checkweight before and af ter the actual weight of the unknown for full cylinders is input to a computer program to determine bias.
1 The weighing of emoty UF6 cylinders is tested in the same manr.er but 'the l
bias is determined manua'ly.
l l
l
.J
3 An examination was made of the Chemet laboratory training and
-equalification program. Tecnnicians are initially trainec and recualified r,ary two years using 5.0.P.s as the training criteria.
In order to retain certi#ication, a technician must perform :ne test at least every six months.
A Certification Evaluation Statement is documented for each tecnnician certified. T7e certification identifies the initial certification date and the date the technician is to be recertified. A Quality Assurance Admin-istrative Routine No. 32C-SO.4 identifies the frequency for recertification.
The training records for seven technicians certified to perform 0/U ratios in UO2 powder and pellets were examined during.he inspection.
Table II identifies the te:hricials and the training received.
Dresently, tne Chemet lateratory is Tease 'ng ma:-o amounts of uran'um in a wide range of scra; materials using a volumetric based dichromate titration.
The dichromate titration was the major error contributor to LEID during the August 1981 inventory.
A project currently underway is to alter the is :ssed en grav metr'c cete--'nat ons cf e
d dichromate tit-at'en method sc 't ' +.
- '- "att..'.++--:-5:5
- +
.5.
. :- + :
4.sc.
- + Ia :
to tencerature anc mase for ce::er
.crovements in ne cresent analyti:al
- acabit es.
i 4
Incependent Inspection Ef fort An examination was made of the gross weights of all 2.79 ard 3.95 percent enriched UF6 cylinders on March 15, 1982 to determine if the licensee is making gross weight measurements within the calibrated range.
In a few dattPces 9e unknown exceeds the upper controi artif act weight of 6357 pounds.
In such instances, a series of 50 pound standard weights are applied as needed to cover the applicable range.
On occasion, UF6 heel cylinder weights are determined to be less than the lower control artifact weight of 1396 pounds.
The licensee contends this practice satisfies the intent of a point calibration.
Inside the range for a point calioratior needs to be defined.
This was identified as an unresolved item in I&E Report No. 70-1113/80-07.
5.
Unresolved Iten Section 4.5.3 of the licensee's approved FNMC plan states that whenever two or more measured values of a standarc or the result: of a cuplicate fall outside the warning limits, an investigation of the condition is initiated and a need for corrective action is determined.
The licensee's FNMC plan should be revised to reflect the current practice of responding to each out-of-control situations, or the licensee should seek an exemption for the,)rovisions in the plan (82-07-01).
7 4
6.
Exit Interview The inspection scope anc findings were summari:ec c1 March 18,1952, with those persons tridicated in paragraph 1 above.
I e
i l
l l
l l
l e
=
1 f
1,
~
~
e TABLE I MEASUREMENT CONTROL LIMITS CONTROL LIMITS NO. OF SYSTEV EXCEECEO (.05)
EXCEECED (.001)
STOS. RUN 10/15/81 - 12/30/81 0/U 0
0 207 1/4/82 - 2/29/82
- U 2
'. 9 6 10/15/81 - 2/15/82
+-
- If Engr. Analyzer 0
0 55 NBL 97 10/22/81 - 3/3/S2 Engr. Analyzer 0
0 44 RIP 003
~.
10/15/ 81 - 3/15c 52 Metal Imp.
5 1
322 (Direct Reader) i l
10e 15/81 - 12 '3e S1 Liqui: Uranium Measurements 1
0 95 25 com U 1/5/82 - 2/14/82 25 ppm V 4
1 98 i
r l
2 MEAS REMENT CONTROL LIMITS CCNTROL LIMITS NC. OF SYSTEM EXCEEDED (.05)
EXCEEDED (.001)
STDS. RUN I
10/15/81 - 12. 21. 51 Licuic Uranium Measurements 1
0 115 00 DDM U 1/5/82 - 2/23/82 100 nm V 2
0 98 i
i i
l l
I I
l l
l l
l
_ ~ -
1 i
i
}
TABLE II d
i Name Bac;e No.
t R. Carpenter 13821 i
R. W. Fucson 16455 l
W. Lacewell 16612 W. H. Mathewson 1642S A. Neil 15350 i
j B. Pierce 13315 i
l
--. y 4
I i
4 I
e I
I I
4 l
i i
-