IR 05000272/1987033
| ML18093A581 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 12/24/1987 |
| From: | Jang J, Pasciak W, Zibulsky H NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML18093A580 | List: |
| References | |
| 50-272-87-33, 50-311-87-34, NUDOCS 8801140362 | |
| Download: ML18093A581 (7) | |
Text
U. S. NUCLEAR REGULATORY COMMISSION
REGION I
Report No. 50-272/87-33 50-311/87-34 Docket N License N DPR-70 DPR-75 Priority --
Licensee:
Public Service Electric and Gas Company 80 Park Place Newark, New Jersey 07101 Category C
Facility Name:
Salem Nuclear Generating Station, Units 1 and 2 Inspection At:
Hancock's Bridge, New Jersey inspection Conducted:
November 16-20, 1987
~~~~--,..---:....._~~~~~~~~~-
Inspectors:
t "2--"2--"?--7S7 date
- ._-
- ;_1-?7 Approved by
pecia ist W. J. Pasciak,~<Effluents Radiation Protection Section, DRSS date Inspection Summary:
Inspection on November 16-20, 1987 (Combined Inspection Report Nos. 50-272/87-33 and 50-311/87-34)
Areas Inspected:
Routine, announced inspection of the nonradiological chemistry progra Areas reviewed included measurement control and analytical procedure evaluations *
Results:
No violations were identifie PDR ADOCK 05000272 Q
Details 1.0 Individuals Contacted
- J. Zupko, Jr., General Manger, Saiem Operations
- R. Dulee, Principal QA Engineer
- J. Trejo, Rad Protection/Chemistry Manager
- R. Dolan, Chemistry Engineer
- G. Slaby, Senior Chemistry Supervisor
- D. Zak, Technical Supervisor of Chemistry
- D. Vetal, Chemistry Consultant (NUS)
J. Riddle, Chemistry Consultant (NUS)
- Present at the exit intervie The inspectors also interviewed other licensee employees including members of the chemistry staf.0 Action on Previous Licensee Findings (Closed) 50-272/84-02-01 and 50-311/84-02-01 IF! (84725)
The same standard stock solution used for calibration was used to*
generate the control charts. Also, control charts were needed for the metal analyses and for the analytes analyzed with the ion chromatograp The licensee has generated and maintained control charts for the necessary analyte The licensee was using two independent bottles of commercial standard solutions but, in most cases, were of the same lo The licensee will use two different lots for independent cross check of the analytical quality of the calibration standard solution and the control standard solutio.0 Measurement Control Evaluation Verification of the licensee's measurement capabilities on actual plant waste samples is done by splitting samples with the licensee and Brookhaven National Laboratory (BNL).
The reactor coolant water was taken for anion analyses, the feedwater was taken for metal analyses and the aerator water was taken for hydrazine and ammonia analyse The reactor coolant sample was spiked with a standard solution of fluoride, chloride and sulfate and the feedwater sample was spiked with a standard solution of iron, copper, nickel and chromiu The standard spike solutions were prepared by BNL for the NR On completion of the analyses by BNL and the licensee, an evaluation will be made (Inspector Follow-up Item 50-272/87-33-01 and 50-311/87-34-01).
Two commercial standard stock solutions for calibration and measurement control were being used that were from the same lo For independence of the standard solutions, commercial standards should be of different lot The licensee will adopt this concept in their Procedure CH-3.9.017:
Chemistry Laboratory Quality Control Requirement ***
The licensee has generated control charts to include the analytes of interes The licensee utilizes control charts with acceptance criteria of +/-2 sigma and an unacceptable parameter of +/-3 sigm The licensee was not identifying systematic biases as shown on the char The inspectors told the licensee that seven consecutive data points on one side of the mean of the control chart constitutes a systematic bias and should be investigated even if it is within the +/-2 sigma paramete The inspectors told the licensee that the Oxford pipets and the Class A glass pipets should be calibrated and documente The inspectors reviewed the calibrations made on some laboratory measurement system The ion chromatograph #16 was calibrated with a single standar This single point calibration did not allow the licensee to identify calibration curve slope changes and nonlinearity which was the case of the sulfate analysis (see paragraph 4).
It was also noted that the calibration was over a large area with very few data point The ammonia curve was from 50 ppb to 5 ppm using six data point The area of interest was around 50 ppb to 1 pp The hydrazine and silica calibration curves also covered too large a concentration range. The plot of data points obscured the nonlinearity in the areas of interes The calibration curves were not statistically fit but were graphically approximate Error in the low concentration range could amount to, as much as, 20 per cen The licensee had a linear regression program for their computer and will generate new statistically fitted calibration curve.0 Analytical Procedure Evaluation During the inspection, standard chemical solutions were submitted to the licensee for analysi The standard solutions were prepared by BNL for the NRC, and the standards were analyzed by the licensee using normal methods and equipmen The analysis of standards is used to verify the licensee's capability to monitor chemical parameters in various plant systems with respect to Technical Specificat1on requirements, vendor and fuel warranty requirement In addition, the analysis of standards is used to evaluate the licensee's analytical procedures with respect to accuracy and precisio The results of the standard measurements comparison indicated that twelve out of forty-five comparisons were in disagreement under the criteria used for comparing results (see Attachment 1).
The results of the comparisons are listed in Table The anion disagreements identified with both ion chromatographs were due to a systematic bias and a single point calibratio At least three data points must be used for calibration or slope changes or biases cannot be determine Originally, the sulfate calibration was made with a 600 ppb standard, at least ten times the licensee's sample value The true calibration was not attained because the calibration curve was given more linearity than was the cas The calibration curves for fluoride and
~.
hydrazine were not statistically fit. If necessary, more than one calibration curve should be generated over each concentration range and statistically fit using linear regressio The ammonia. disagreements were due to a procedural misinterpretatio The analyst took a mV reading at 5 minutes instead of waiting for stabilization of the electrod The licensee will wait for electrode equilibrium and stabilization rather than specific time cut-of The poor calibration procedures identified by the inspectors were from advice given by the licensee's outside contracto Improvements had been noticed by the inspectors of the licensee's measurement progra Since the evaluation in January 1986, the ion chromatograph has been introduced which has given the licensee analytical capabilities to determine chloride and sulfate to less than 5 pp With the recommended calibration procedures, the licensee should achieve a good measurement progra.0 Exit Interview The inspectors met with the licensee representatives (denoted in paragraph 1) at the conclusion of the inspection on November 20, 1987, and summarized the scope and findings of the inspectio At no time during this inspection was written material provided to the licensee by the inspector I
-
Chemical Parameter Chloride Chloride Sulfate Sulfate Sodium
- Fluoride Hydrazine Ammonia Silica Table 1 Capability Test Results Salem Nuclear Generating Station, Units 1 and 2 Results in parts per billion (ppb)
Analytical Procedure Ion Chromato-graph 2020i Ion Chromato-graph 16 Ion Chromato-graph 2020i Ion Chromato-graph 16 Ion Chromato-graph 2020i Sp. Ion Electrode Spectrophoto-metric Sp. Ion Electrode Spectrophoto metric NRG Value 1. 9+/-0. 2 3.0+/-.2+/-.1+/-.5+/-.1+/-1+/-1 1. 6+/-.3+/-.2+/-.0+/-.5.+/-. 4+/-1. 5 4.6+/-.6+/-.6+/-.1+/-. 5+/-1. 9 83.5+/-.3+/-.9+/-. 0+/-1. 0 87.6+/-.0+/-13 469+/-43 54.3+/-. 109+/-7 160+/-5 Lie Value 1. 5+/-0. 3 2.3+/-.7+/- Ratio (Lic./NRC)
0.79+/-0.18 0.77+/-0.04 0.84+/-0.03 10.9+/-0.8 0.90+/-0.14 20.1+/-0.5 1.02+/-0.06 37.9+/-0.5 0.95+/-0.03 1. 6+/-.0+/-.9+/-.0 0.91+/-0.06 0.91+/-0.04 20.0+/-2.4 2.00+/-0.26 26.6+/-0.9 1.30+/-0.09 39.9+/-2.0 0-99+/-0.06 5.3+/-.1+/-.9+/-.15+/-0.15 1.11+/-0.13 1.08+/-0. 07 21.9+/-0.1 0.95+/-0.02 37.7+/- ~87+/-0.05 84.7+/-1.2 1.01+/-0.04 19.7+/-0.6 0.88+/-0.06 51.2+/-1.0 0.90+/-0.02 101.0+/-1.0 0.97+/-0.01 70.0+/-3.5 0.80+/-0.06 133+/-1 0.85+/-0.07 336+/-13 0.72+/-0.07 54.3+/- +/-0 i65+/-.0 1. 0 1. 03+/-0. 04 Comparison Agreement Disagreement Disagreement Agreement Agreement Agreement Agreement Agreement Disagreement Disagreement Disagreement Agreement Agreement Agreement Agreement Disagreement Disageement Agreement Agreement Disagreement Disagreement Disagreement Disagrement Disgreement Agreement Agreement Agreement
Results in parts per million(ppm)
Boron Titrimetric 500+/-5 508+/-10 1. 02+/-0. 02 Agreement 1008+/-15
'1006+/-0 1. 0 Agreement 989+/-12 992+/-6 Agreement Sodium AA 0.92+/-0.10 0.97+/-0.01 1.05+/-0.11 Agreement 1.85+/-0.16 1.99+/-0.02 1.08+/-0.09 Agreement 2.88+/-0.16 2.98+/-0.01 1.03+/-0.06 Agreement Iron AA 0.98+/-0.07 1.01+/-0.01 1.03+/-0.07 Agreement 1.91+/-0.07 1.99+/-0.02 1.04+/-0.04 Agreement 2.94+/-0.08 2.93+/-0.01 Agreement Copper AA 0.94+/-0.05 0.95+/-0.01 1.01+/-0.05 Agreement 1.93+/-0.10 1.97+/-0.02 1.02+/-0.05 Agreement 2.90+/-0.12 2.94+/-0.01+/-1.01+/-0.04 Agreement Nickel AA 1.02+/-0.05 1.03+/-0.01 1.01+/-0.05 Agreement 2.04+/-0.06 2.00+/-0.01 0.98+/-0.03 Agreement 3.06+/-0.08 2.97+/-0.03 D.97+/-0.03 Agreement Chromium AA 1.02+/-0.06 0.99+/-0.04 0.97+/-0.07 Agreement 1.88+/-0.06 1.97+/-0.06 1.05+/-0.05 Agreement 2.86+/-0.16 2.98+/-0.02 1.04+/-0.06 Agreement
- ATTACHMENT l CRITERIA FOR COMPARING ANALYTICAL MEASUREMENTS This itt*chment provides cr1teria for cOllpiring results of Cipibility test In these criterii the judgement limits ire based on the uncertainty of the ritio of the licensee's vilue to the NRC valu The following sups ire pe*rformed:
(1)
the rit1o of the licensee's value to the NRC vilue 1s c0111puted Licensee Value (ratio = NRC Vilue
);
(2)
the uncertainty of the ritio is propagated. 1 If the absolute value of one minus the ratio is less than or equal to twice the ratio uncertainty, the results are in agreemen Cll-rat1ol s Z uncertainty)
1 Z= x then s2 s2 s2
-*
z
=
x
+.. y T
T l
1(From: Bevington, P. R., Data Reduction and Error Analysis for the Physical Sciences, McGraw-Hill, New York, 1969)