IR 05000295/1980022
| ML19340E778 | |
| Person / Time | |
|---|---|
| Site: | Zion File:ZionSolutions icon.png |
| Issue date: | 12/03/1980 |
| From: | Junuska A, Paperiello C, Phillips M NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML19340E770 | List: |
| References | |
| 50-295-80-22, 50-304-80-24, NUDOCS 8101150640 | |
| Download: ML19340E778 (8) | |
Text
..
-..-_- -
.
_-
_. - - - - - -.
- -. - _ -
_..
. _ _ -.
-
..-
-_.
l
.
l
'
O i
U.S. NUCLEAR REGULATORY COMMISSION
'
0FFICE OF INSPECTION AND ENFORCEMENT
REGION III
i Reports No. 50-295/80-22; 50-304/80-24 Docket Nos.
50-295; 50-304 Licenses No.
Licensee:
Commonwealth Edison Company P. O. Box 767 Chicago, IL 60690
<
Facility Name: Zion Nuclear Power Station, Units 1 and 2 Inspection At:
Zion Site, Zion, IL l
Inspection Conducted:
November 17-19, 1980 Inspector hi
/Z C
h.
.
i 4 t. w N L P A. G.
anuska
- ay &
8.}.Paperiello,ActingChief i;
l e,
i Approved By:
C.
l Environmental and Special
Projects Section Inspection Summary
,
I~
Inspection on November 17-19, 1980 (Report No. 50-295/80-22; 50-304/80-24)
Areas Inspected: Routine, announced inspection of (1) corrective action indicated in Reportable Occurrence Reports No. 50-295/80-45 and 50-295/80-47 and (2) Confirmatory Measurements including program for quality assurance and quality control of analytical measurements; and collection of samples, analysis onsite with the RIII Measurements Van, and discussion of results.
l The inspection involved 46 inspector-hours onsite by two NRC inspectors.
Results: Of the three areas inspected, no apparent violations were identified in two areas; one apparent violation was identified in one area (Severity VI -
'
failure to accurately report the total curies for each of the radionuclides released in gaseous form - Paragraph 3.a.1).
!
.
81011so p
,
..
.
.
.
.m.
.
.
..
.
.
.
.
.m
.
DETAILS 1.
Persons Contacted
- K. Graesser, Station Superintendent
- G. Plim1, Assistant Station Superintendent
- D. Howard, Rad / Chem Supervisor
- P.
Zwilling, Chemist
- J.
Zirka, Chemist AB. Schramer, Chemist
- T. Lukans, Quality Control Inspector
,
- P. Kuhner, Quality Assurance Inspector S. Gurunathan, Engineer
- Denotes those present at the exit interview.
2.
Review of Reportable Occurrences (0 pen) Reportable Occurrences 50-295/80-45 and 50-295/80-47: pH outside range of T/S Section 1.3.C.
The inspectors discussed the reported modifi-cation. As the corrective action is not complete, this item will be reviewed and closed out during a subsequent inspection.
3.
Results of Comparative Analyses Results of comparative analyses performed on effluent samples split and analyzed in the RIII Measurements Van onsite during this inspection are shown in Table I.
The criteria for comparing measurement results are given in Attachment 1.
a.
Results of the November 1980 Split Samples split during this inspection were analyzed by the licensee
,
and onsite by the RIII Measurements Van.
For 27 sample comparisons, the licensee's results yielded 12 agreements or possible agreements.
(1) Gas Sample Split The licensee did not accurately quantify Xc-133 in a contain-ment building gas sample, but was a factor of two below the NRC result.
In an earlier gas sample split, not listed in Table I, the licensee had been low by a factor of five which was caused by a bad sample split. For the second set of samples, NRC sample containers were placed in the sample flow path before and after the licensee's sample container. A count of both NRC sample containers showed that the Xe-133 concentration was approximately equal indicating a good split.
Based on these results, the licensee has been underreporting the discharges from the facility. This constitutes an ap-
_
-2-
.
%
.
-
.
parent violation of Technical Specifications Section 6.6.3.a.1.a.3.
The licensee recalibrated his counting equipment during April 1980 for all geometries when all calibration data had been inadvertently lost.
In a memo issued by the licensee on November 26, 1980, the licensee applied the factor of two determined in the NRC analysis, to the effluent results since April, 1980, and found that the amount of radioactivity released to the environment at all times was less than one percent of the Technical Speci-fication Limits.
(2) Charcoal Cartridge Sample Split The licensee did not accurately quantify I-131 or I-133 in a containment charcoal sample, but was a factor of two be-low the NRC result. A previous comparison of an effluent charcoal sample indicated an agreement for I-131, with the licensee's results being higher than the NRC's.
The licensee felt that the disagreement was probably caused by counting the charcoal sample upside down. A subsequent count of another charcoal sample which was verified to be counted correctly resulted in agreements for I-131 and I-133 with the licensee's results again being higher than the NRC's.
To confirm this hypothesis, the licensee was given a face loaded charcoal sample that had been spiked with NBS trace-able amounta of several nuclides.
For all * radionuclides identified by the licensee in this spiked sample, the results were approximat ely a factor of two below the actual activities in the spiked sample. This disagreement was verified in a subsequent count by the licensee. Two charcoal samples have been sent to the NRC's reference laboratcry as an independent check on these results.
If the licensee is found to be in disagreement and lower than the NRC's results, the licensee has been underreporting the discharges from the facility.
This is considered to be an unresolved item.
The inspection revealed that the licensee had calibrated this geometry using a point source located one centimeter from the detector, instead of an extended source. This results in the calculation of a higher efficiency upon calibration and the subsequent lowering of actual sample results. This was discussed at the exit interview.
In a memo issued by the licensee on November 26, 1980, the licensee applied the factor of two, determined in the NRC analysis, to the effluent results since April, 1980, and found that the amount of iodine activity released to the environment at all times was less than one tenth of one per-cent of the Technical Specification limits.
-3-
' dim
- -
..
-
. - - -
-
.
.
(3)
Particulate Filter Sample Split The licensee failed to quantify Co-58, Na-24, and Cs-136 on the two particulate filters counted.
In the case of Cs-136, the licensee stated that his counting library did not include this isotope, although the spectrum result did contain peaks at both major energy lines (0818 MeV and 1.05 MeV).
Based on the NRC results, the concentration of this nuclide was the fourth highest in the sample (2.4 E-11 uCi/cc), and approxi-mately equal to the Technical Specification requirement for the sensitivity of individual gamma emitters on particulate filters (3.0 E-11 uCi/cc).
In this case, the sensitivity 1cvel had not been exceeded for this isotope; however, the licensee could not guarantee that this isotope was not present in reportable quantities in past samples. The inspectors emphasized the need for verifying that all peaks present in the sample are accounted for and reported in the results, irregardless of whether or not they are listed in the particular library used.
The Co-58 and Na-24 activitics found by the NRC were well below the minimum sensitivity required by the Technical Specifications; however, the licensee's result form did show that the 2.754 MeV peak for Na-24 was present in the sample.
This problem of failure to identify peaks that are present is discussed in more detail in Paragraph 3.b below.
The Cs-137 result enclosed in parentheses was not technically identified by the licensee. The summary report printout from the particulate analysis did give a result, which had been lined out because the licensee routinely disregards all isotope results where the 20* error result is greater than 25 percent.
In this case, the error was 26.7 percent. A 2'~ error value specifies the range of values inside which a particular recount has a 95 percent chance of falling. This means that for error values of less than 100 percent, 95 percent of recount results will always have a finite positive value and be within the error band. Of the re-maining five percent half should be greater and half less than the original result.
In other words, if the error value is less than 100 percent, then there is a 97.5 percent chance that the sample does contain that isotope, and therefore should not be ignored or deleted.
Since the counting program that determines the Minimum Detectable Activity (MDA) for each isotope uses an error level of 100 percent, the licensee has in effect raised his MDA by an undetermined amount for each isotope, and as a result may not be able to meet the sensitivity requirements specified in Technical Specification, Table 4.12-2 Section D.
The inspectors emphasized that the use of an acceptable error criteria of less than 100 percent must guarantee that the sensitivity requirements of the Technical Specifications are still met.
This was discussed at the exit interview.
-4-
.
"-
.
.
-
.
Although the remaining radionuclides identified were in agreement, the licensee's results were consistently lower than the NRC's.
The inspection revealed that the licensee had calibrated this geometry using a point source located one centimeter from the detector, instead of an extended source. This results in the calculation of a higher counting efficiency upon calibration, and a subsequent lowering of actual sample results.
(4)
Liquid Sample Split The licensee failed to quantify Xe-133 and Sr-92 in the lake discharge tank sample.
In the case of Sr-92, the licensee stated that his counting library did not include this isotope.
Sr-92 is a fission product and should be expected to be en-countered under normal operations.
Although the energy peak for Xc-133 was present in the spectrum collected by the licensee at an error level of less than 10 percent, the licensee failed to quantify this isotope.
Under the procedures in effect at the time of this inspection, only the " Summary of Nuclides in Sample" is printed as a final report. After the licensee was informed of the major disagree-ments, he printed a listing of all the peaks identified in the spectrum and found Xe-133.
Because the licensee does not cur-rently participate in any crosscheck program, He was unaware
-
that various radionuclides actually present in significant quantities were being omitted from his sample results until this inspection. This is partially due to an inadequate count-ing procedure, which does not provide for verification that all peaks in the spectrum are identified and accounted for in the radionuclides listed in the " Summary of Nuclides in Sample."
b.
Procedure Review The inspectors reviewed selected procedures related to calibration, sample collection, and sample analyses. The procedures were found to be too general in some cases and did not contain all necessary information.
Procedures ZCP-118, ZCP-119, ZCP-402, and ZCP-409 do not address the problem of counting a mixed alpha-beta sample for beta using the beta voltage plateau.
In this case, the licensee would overestimate the amount of beta radionuclides present, as in fact he would be counting both alpha and beta disintegrations and reporting them as beta.
ZCP-16 and ZCP-407 calibration procedures do not describe how to prepare the standard.
ZCP-407 specifies using a point source for determination of the efficiency for the particulate filter and charcoal cartridge geometry. As discussed in Sections 3.a.2 and 3.a.3 above, this results in an overcalcula-tion of the efficiency and helped contribute to the poor percentage of agreements on these analyses. None of the procedures reviewed have provisions for supervisory review of analysis results.
ZCP-413-5-
,
.
"e
--
,
.
.
does not specify that all peaks seen in the spectrum have been ident-ified as peaks by the system, and further, that these peaks can all be accounted for by the isotopes present in the sample. This pro-cedural shortfall is directly responsible for the failure of the licensee to identify Xe-133 in the liquid sample. ZCP-413 further specifies that an isotope is not to be identified if the counting error is greater than 25 percent. As is discussed in Paragraph
'
3.a.3 and 3.a.4 above, this effectively ra'ises the MDA for sample analysis. ZCP-416 has been revised to verify that the Cs-137 and Co-60 peaks are within one channel of their original calibration
location, but no check is conducted at the very low energies of Xe-133 (e.g., approximately 80 kev). Although the 500 KcV to 1,500 KcV energy calibration may be accurate, energy peaks for radionuclides outside of this range may have shifted in excess of one channel and may subsequently be either misidentified or not quantified by the counting system.
4.
Unresolved Items Unresolved items are matters about which more information is required in order to ascertain whether they are acceptable items,. violations, or deviations. An unresolved item disclosed during the inspection is discussed in Paragraph 3.a.2.
5.
Exit Interview
'
The inspectors met with licensee representatives (denoted in Paragraph 1) at the conclusion of the inspection on November 19, 1980. The inspectors summarized the scope and findings of the inspection. The licensee made the following remarks in response to certain of the items discussed by the inspectors:
a.
Acknowledged statements by the inspectors with respect to the violation (Paragraph 3.a.1).
b.
Agreed to obtain appropriate standards and recalibrate the gamma spectrometry system (s) for all geometries. Tnese will be completed within one month of receipt of the standards.
c.
Agreed to revise procedure ZCP-413 regarding either counting time or maximum acceptable percent error to ensure that the-minimum sensitivity of detection specified by the Technical Specifications would always be achieved.
Attachments:
1.
Attachment 1, Criteria for Comparing Analytical Measurements 2.
Table I, Confirmatory. Measurements Program Results 6-
-
.
- e
,.y
,,.---a w-r-
-e-e+*#
rs--.--
--
&
y
- *,,
-e~
w-y5 w--v,
-
.-
M e
---e 4m-e
_....
_..
_
--.. _ - _.
- -
.-.. - -
-. _. _.. -_-
-
_
_
,
- .
ATTACINENT 1 CRITERIA FOR COMPARING ANALYTICAL MEASUREMENTS
.
t'
This attachment provides criteria for comparing results of capability tests and verification measurements. The criteria are based on an
-
empirical relationship which combines prior experience and the accuracy needs of this program.
a
~
}
In these criteria, the judgment limits are variable in relation to the
'
comparison of the NRC Reference Laboratory's value to its associated one sigma uncertainty. As that ratio, referred to in this program as
" Resolution", increases, the acceptability of a licensee's measurement i
.i should be more selective.
Conversely, poorer agreement should be con-sidered acceptabic as the resolution decreases. The values in the ratio
criteria may be rounded to fewer significant figures to maintain statistical consistency with the number of significant figures reported
-
by the NRC Reference Laboratory, unicss such rounding will result in a narrowed category of acceptance.
The acceptance category reported will be the narrowest into which the ratio fits for the resolution being used.
i RESOLUTION RATIO = LICENSEE VALUE/NRC REFERENCE VALUE
Possible Possibic
,
Agreement Agreement "A" Agreeabic "B"
!
<3 No Comparison No Comparison No Comparison
>3 and <4 d.4 - 2.5 0.3 3.0 No Comparison
-
>4 and <8 0.5
- 2.0 0.4
- 2.5 0.3
- 3.0 I8 and <16 0.6 - '1.67 0.5 2.0 0.4 2.5
-
-
I
>16 and <51 0-75 - 1.33 0.6
- 1.67 0.5
- 2.0
!
551 and <200 0.80 - 1.25 0.75 - 1.33 0.6 1.67
-
[200 0.85 - 1.18 0.80 1.25 0.75 - 1.33
-
,
"A" criteria are applied to the following analyses:
Gamma spectrometry, where principal gamma ener;y used for identifi-cation is greater than 250 kev.
Tritium analyses of liquid sampics.
"B" criteria are applied to the following' analyses:
~
Gamma spectrometry, where-principal gamma energy used for identifi-cation in less than 250 kcV.
&
Sr,-89 and Sr-90 determinations.
Gross. beta, where samples are counted on the came date using the same reference nuclide.
.
.
e
- eo
f (
.
(
TABLE I (
U S hUCLEAR REGULATORY CCMMISSION OFFICE OF IN SF E C TI ON AhD ENFORCEMENT
,
CONFIRM LTORY 11E ASUREM ENTS PROG 9 AM FACILITY: ZION (
FOR THE 4 OVARTER OF 19R0 (
-
NRC-------
- -- L I C E.N S E E -- -- -
--NRC: LICENSEE----
SAMPLE I S O T CPE RESULT ERROR RESULT ERROR RATIO RES T
(
OFF GA9 XE 133 1 4 E-0 4 2 2E-06 6 8E-0 5 1.3E-06 4 9E-01 6.4E+01 D
( L VASTE CO 60 2 3E-05 f.4E-07 2 2E-05 9 6F-07 9 6E-01 4 3E+01 A
XE 133 4.1E-05 3 0E-07
0. 0
5.1E+01 D
SR 92 1.4E-06 2 5E-07
0.0
5.6E+00 D
P FILTER NA 24 3.2E-02 3.0E-04 2 7E-02 B.1E-04 6 4E-01 1.1E+02 A
CO 58 3.4E-04 4 5E-05
0.0
7.6E+00 D
(
I 131 5 6E-03 7 4E-05 4 5E-03 2 3E-04 o.0E-01 7.6E+01 A
I 133 7.6E-03' F.3F-05 2 1 E-03 3 6E-04 8 1E-01 3 1E+01 A-CS 134 1.5E-02 1 3C-04 1 3E-02 3 5E-04 S.7E-01 1.2E+02 A
(
0.0
2.7E+01 D
CS 137 1.6E-32 1.4E-04 1 3E-02 4.0E-04 8 1E-01 1.1E+02 A
NA 24 2 6E-04 3. 9F- 0 5
0.0
6.7E+00
(
CS 134 1 5E-0 4 1 3E-05 1.oE-04 4 5 E-0 5 1 3E+00 1.2E + 01 A
'
CS 137 2 5E-0 4 2 3E-05 9 2E-04) (5.SE-0 $
G. 8 E -01)
1.1E+01 (A)
(
C FILTER I 131 1 6E-03 4 0E-05 2 0E-0?
1.3E-04 1 3E+00 4 0E +01 A
I 131 1 0E-01 3.1E-04 6 0E-02 7.2E-04 6 0E-01 3.2E+02 D
I 133 4 0E-02 2 7E-04 2.?E-02 6.0E-04 5 7E-01 1.5E*02
(
I 135 4.3E-03 4 6F-04 3 1 E-03 7 4E-04 7 2E-01 9 3E+00 A
I 131 3 4E-02 1.6E-04 3.6E-02 5.4E-04 1.1E+00 1.9E+02 A
I 133 6 5E-0 3 1.6F-04
.6 9E-03 4.cE-04 1.1E+00 4 1E +01 A
(
C SPIKED C0 57 5 8E-03 2.6E-04 2 0E-n3 1 3E-04 3 4E-01 7.2E+01
CO 60 4.4E-02 1.8F-03 2 1E-02 5 0E-04 4.8E-01 2 4E+01 n
(
00
2.4E+01 D
CS 137 3.3E-02 1 3E-03 1 4E-02 4.8E-04 4 2 E-01 2.5E+01
SN 113 1 5E-02
/.3F-04 6 4E-03 9 1E-04 4 3E-01 2.4E +01 D
L SR 85 2.0E-02 7.LE-04 9 4E-03 2 6E-03 4 7E-01 2 6E + 01 D
Y
7 5E-02 3.2c-03 3. 7 E-0 2 2.4E-03 4 9E-01 2 3E+01
l T TEST RESUL TS:
A = A G RE EMEN T l D=DISAGREEMFNT P=POSSIBLE A GREEMENT N=NO COMPARISON L
L
.
.
-