ML19317D382

Responds to NRC 750321 Ltr Re Violation Noted in IE Insp Repts 50-269/75-03,50-270/75-03 & 50-287/75-03.Corrective Actions:Effluent Monitors Periodically Decontaminated,Sample Collecting & Counting Revised & Normalization Factor in Use

ML19317D382
Person / Time
Site:	Oconee
Issue date:	04/11/1975
From:	Thies A DUKE POWER CO.
To:	Moseley N NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
Shared Package
ML19317D375	List: ML19317D382
References
NUDOCS 7911270589
Download: ML19317D382 (6)

Text

. - . . _ _ _

- - - - ~ ~ - __

~

' i -

- DUKE PowEn Co> tem

- Powra Burtorno 422 Sourn Cnunca Srazer, CnAarortz,N.C.seso A c. THiss @ P. O. Box 2ns semos v,ce paesicant Paoowctiog AND Tashsuission 1

I April 11, 1975 i

7 Mr. Norman C. Moseley, Director U. S. Nuclear Regulatory Comnission Suite 818 230 Peachtree Street, Northwest Atlanta, Georgia 30303 Re: IE:II:TNE 50-269/75-3 i

50-270/75-3 50-287/75-3

Dear Mr. Moseley:

In reference to your letter of March 21, 1975 which transmitted IE '

inspection Report 50-269, -270, and -287/75-3, Duke Power Cocpany

' does not consider any information contained in this report to be proprietary.

Please find attached our response to Items,I.A.1 and Unresolved Item III.75-3/1.

1 Very' truly yours, A. C. Thies 4l ACT:vr.

l Attachment

• 1 4

e

~

i.

7911270 N .

m.. _.__-.-..._...-______-_m _ _ _ _ _ _ _ . _ s. .s - - .. _ _ _ _ _ _ _ - . _ . . . _ . _ _ _

. l i

.m l DUKE POWER C0MPANY OCONEE NUCLEAR STATION RESPONSES TO ITEMS I.A.1 ASD III. 75-3/1 0F INSPECTION REPORT 50-269/75-3, 50-270/75-3, 50-287/75-3 April 11, 1975

,. M

~

\

I.A.1 Contrary to Technical Specification 3.9.7, tha effluent control monitor was not set to alarm and automatically close the waste discharge valve so that-the appropriate requirements of the specification are met.

RESPONSE

The procedure used at the time of the. inspection required setting the effluent monitor at one-half decade above detector background activity or tank activity (whichever is higher). Due to the high background during this period, the former setting was utilized. This would have prevented the release of liquid effluents which had a substantially higher activity j than anticipated.

The liquid waste process monitors have been acved to an area of lower background on the turbine floor. New dual monitors have been installed which vill permit decontamination of one while the other is la service.

On March 20, 1975, a procedure change was made which provides specific setpoints (numerical values) for each combination of discharge conditions (example: 50 gpm, 40 cfs dilution). The setpoints are such that an alarm would be sounded and the release terminated preventing any possibility of exceeding values listed in 10 CFR 20, Appendix B, Table II, Column 2.

The initial readings of the effluent monitor were well within expected values. However, after a few discharges through the new monitor chamber, the bachground level has increased. Periodic decontamination of the chamber 1 internal surfaces has been necessary to reduce this background. Although '

3 efforts are being made to determine the best method of operation of the effluent monitors, periodic decontamination assures that the intent of Technical Specification 3.9.7 is met.

j III.75-3/1 ANALYSIS OF LIQUID WASTE SA'!PLES l ITEM:

i The licensee is required to measure the quantities and concentrations of  ;

I radioactive material in effluents from his facility to assure that they I are within the limits specified in his license and the NRC regulations. ,

The inspection consisted of testing the licensce's measurements of ra-  !

-dioactivity in samples of his effluents and prepared test standards by- l comparing his measurements with those of the NRC's reference laboratory.

The measurements made by the NRC laboratory are referenced to the National Bureau of Standards radioactivity measurements system by laboratory intercomparisons.

i

The test results from previous split sample measurements in June and

.se September 1974 showed the licensee's measurements to have several dis- l l

1 l

1

, +-

I w ,

m

~

• crepancies. Capability test srcadsrds were sent to the licensee in June and October 1974 to resolve the discrepancies. See Table 1-4. -

The dis-crepancies were partially resolved by the analysis of the test standards.

Except for strontium, most remaining discrepancies are in the conservative direction. The strontium procedures are currently being eval.uated by our reference laboratory to resolve the disagreement in measurements. On February 19, a particulate filter test standard with a complex ga=ma I spectrum was brought to the licensee to analyze. The purpose was to further resolve the licensee's large dif ferences on previous split sample measurements and his failure to identify all the isotopes present. It was determined that previously used efficiency curves were not good in certain

! energy ranges, that the computer program was inadequate to identify many

• isotopes found in their liquid waste, that isotope spectrums were not being i fully analyzed by the computer or personnel, and that the computer programs contained avne countants which were wrong. The computer program is now being updated and licensee management stated that this would be completed i by April 1, 1975. New calibrations have been and are being perforted.

The results of the charcoal adsorber and particulate filter standards are now in agreement. See Table 5. The criteria used for comparisons are attached. Apparently, no specific isotopic release limits have been exceeded as a result of the above findings.

4

RESPONSE

1 l Although Duke Power Company acknowledges responsibility for the results of strontium analyses, these analyses are performed by an outside vendor. The vendor has expressed his cooperation in resolving the discrepancies and has submitted strontium procedures for evaluation by NRC's reference laboratory and has also requested additional samples to be analyzed for strontium. We are confident that this discrepancy between the vendor and NRC's reference laboratory will be resolved in the near futures The particulate filter test standard was' analyzed in the presence of the IE inspector on February 19, 1975. The test standard was placed in the geometry used for normal station particulate samples and analyzed. This-analysis identified isotopes not present in the sample and did not identify two of those present in the sample; the Assistant Health Physics Supervisor, responsible for the counting laboratory, stated that the results were invalid t due to the high activ_" of the sample. Normal particulate sa=ples contain several (usually <10) isotopes in concentrations of 10-7 to 10-11 Ci/ml; theteststandardcontainedfourisotopes'inconcentrationsof10g pCi/nl.

The analysis of. this standard, in the normal counting geometry, resulted in over-ranging of the counting equipment. Thus, energy peaks which normally l

cover 4 to 7 channels were spread over 17-30 channels. Each of these channels represents increce. ts of energy corresponding to 0.5 kev, therefore, rather than indicating peaks of widths of 2.0 to 3.5 kev, the results of the sample indicated peaks of widths 8.5 to 15.0 kev. -The wider the energy peaks,.the more pcssibilities that exist for the identification of isotopes. This is the reason several isotopes not present were identified by the computer as s- being present. .The two isotopes not identified as present were Sb-125 and Na-22.

?~~ .

~

Both of these isotopes are not seen in routine station particulate samples and therefore had not been included in the computer library of possible isotopes. When the satple was re-analyzed in a geometry such that the peaks covered only 4-7 channels, the results were in agreement with results obtained by the NRC reference laboratory.

We are in the process of procuring calibrated gamma sources with energies abovc 133 MeV to correct efficiency curves for these ene gies. This energy is, however, greater than that necessary for routine sample analyses.

Although the computer progran is adequate to identify the isotopes routinely present in the various station samples, the computer program libraries have been expanded to include those isotopes occasionally present so that isotope spectra are fully analyzed by the computer and therefore require less technician analysis when the computer indicates a peak present but cannot identify it.

The constant incorrectly contained in the computer program was the half-life for Ag-100a. These estrections to the cc=puter pr eram were completed prior to the April 1, .975 target date.

The differences between the licensee's and the NRC reference laboratory's results on split samples are due to several variables in the analysis of the samples:

. 1. Sample Geometries: Split samples of gas have been collected in containers of varying sizes; the containers have been in geometries not used routinely at the station. Special calibration has been performed so that these split samples could be analyzed.

2. Countine Times: In instances where split samples are not possible, such as particulate samples, the station has analvred the sample incediately so that the NRC reference laboratory may analyze the same sample. In such cases, short-lived isotopes and adsorbed gases will produce activity seen by the station but not seen by the reference laboratory at the time it analyzes the sample. Also, daughter produce ingrowth is not consider)d in analysis of the sample by the reference laboratory.

If these corrective reasures for collecting and counting samples are instituted, we are confident that the results achieved by our laboratory and yours will be consistent.

1

,lTEM: 1 Liquid waste discharge records were exacined. It was found that several l releases a day were being made based on the analysis of degassed gross beta samples. One release per day was being analyzed by ga=ma spectrometry. ,

This analysis was used to quantify releases and was consistently significantly I higher than the gross beta measurecent. Gross beta analysis had not been l normalized against results of the specific isotopic analysis. Apparently, no specific isotopic release limits heve been exceeded. An inspector i stated that future releases by gross measurerents should be normalized to

t -

• i . -)

the isotopic analysis, and that this normalization factor should be verified periodically for char.ges in the isotopic spectrum. Licensee management stated that a normalization factor would be applied beginning March 10, 1975. This item is carried as an unresolved item.

RESPONSE

The normalization has been accomplished and a normalization factor has been in use since March 10, 1975.

\

-