ML20064L006

Responds to NRC Re Violation Noted in IE Insp Repts 50-111/82-02 & 50-297/82-02.Justification for Continued Use of Present Stack Gas Monitor Sys Procedures Provided

ML20064L006
Person / Time
Site:	North Carolina State University
Issue date:	01/13/1983
From:	Cockrell R North Carolina State University, RALEIGH, NC
To:	Julian C NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
Shared Package
ML20064K957	List: ML20064K988 ML20064L006
References
NRP-RGC-83-1, NUDOCS 8302150082
Download: ML20064L006 (3)

Text

.

• d v

NORTH CAROLINA STATE UNIVERSITY AT HALEIGH

,9 SCIIOOL OF ENGINEERING

} j ?!? I' L* 0 Di P AH1 MENT OF NUCLEAR I5NCINEFRING Nucuan REAcron PnocnAu nm se, zir 2n,so NRP-RGC-83-1 Mr. Caudie Julian Nuclear Regulutory Commission Region II 101 Marietta Street, Suite 3100 Atlanta, Georgia 30303 Docket No. 50-297

Dear Mr. Julian:

By letter c'ated 21 April 1982,

Subject:

Report Nos. 50-297/82-01 and 50-111/82-02, we were informed of the results of the PULSTAR inspection that occurred on 15 - 19 March, 1982. In the supporting section entitled " Details",

Section 13, Radiation Control (third paragraph), the inspector acknowledged the procedures that are followed in the calibration of the stack monitor sy-stem as fulfilling the requirements of our Technical Specifications. However, he did suggest that '?t is good practice to periodically perform a direct cali-bration by observing the monitor response to a kn"wn concentration of radio-active gas". In the following paragraphs, this suggestfun is discussed.

For the purpose of this discussion, our Stack Gas Monitor System will be used as an example of the operating system. This monitor is the one used to estimate the Ar-41 average concentration released during the reporting period.

The Stack Gas Monitor System includes:

1. G.M. detector in a special shield

2. Pre-amplifier

3. Cabling

4. Electronics Log Ratemeter including amplifier

5. Read-out meter having log system from 10 to 106 (5 decades;

6. Recorder, log scale to complement the read-out meter When the Stack Gas Mcnitor System is operating correctly and is in calibration, the following facts are known:

1. The activity of the radiation source, either sealed source or gaseous source, used in the calibration procedure.

2. The G.M. detector sensitivity to Ar-41.

3. The air flow in the exhaust system.

4. All electronic components are properly adjusted to the manu-facturer's specifications.

5. The System responds to a radiation source.

6. The recorder has been separately calibrated according to the manufacturer's specifications.

8302150082 830208

{DRADOCK 05000111 PDR

Mr. Caudie Julian Page 2

7. Auxiliary Ortec counting equipment does not electrically load the system.

8. The system read-out meter, the recorder, and the Ortec counting equipment are in agreement at all test points.

The calibration procedures used at NCSU require at least two (2) sealed sources of known activity (Ba-133 and Cs-137) as the known radiation input (Item 1 above). Items 4-8, inclusive, are individually checked, adjusted as necessary, and verified. A pulse generated by a calibrated pulse generator is applied to the system input in place of the detector and the system operation and stability are confirmed on all ranges before using the radiation sources. Finally, the radiation sources, individually and decayed to the day of the test, are used.

The three data outputs - read-out meter, the recorder, and the Ortec equipment -

must be in agreement. Further, the output data must agree within 10% of the input data for the calibration to be accepted.

, Independently, a detector shield very closely approximating the system I

operational shield has been constructed for the purpose of verifying the detector sensitivity. Item 2 above. The procedure for this determination includes:

l

1. Determining the activity of Ar-41 made by irradiation of air in a

! vial in the PULSTAR and actually transferred to the Test Shield.

! 2. Using Ortec counting equipment, record three (3), two (2)-minute counts.

~

3. Repeat 2 above for 4 or 5 sets of counts.

4. Compute all net counts to a common time; say, the time activity

! is transferred to the Test Chamber.

l 5. Knowing the activity transferred to the Test Chamber and the volume of the Test Chamber, compute the concentration in pCi/ml.

6. Divide the net counts, corrected (Step 4 above), by the concen-tration. The average of these values is the sensitivity of the G.M. detector.

7. A set of counts taken about two hours apart will permit a deter-mination of the half-life of the activity. A significant deviation from Tl/2 = 1.83 hours9.606481e-4 days <br />0.0231 hours <br />1.372354e-4 weeks <br />3.15815e-5 months <br /> is reason to question the validity of the test.

Attention is invited to the fact that as outlined above the sensitivity-has been determined while the radioactivity is in a " static" or non-flowing status. Further, the above procedures are not dependent upon knowing the flow rate of the exhaust from the reactor bay.

l A calibration procedure using the operating system (Direct Calibration) could satisfy either of two (2) objectives, namely:

1. Is the output data in agreement with the input data. Note: the i sensitivity of the detector must be known.

2. The determination of the G.M. detector sensitivity to Ar-41.

l Either of the above two procedures requires reasonably accurate knowledge of the activity input rate and the exhaust air flow rate. It should be noted that

9.

Mr. Caudie Julian Page 3 in these procedures th theexhaustsystem(10gradioactiveeffluentisflowing(dynamic)throughout cfm) and through the System monitor (10 cfm) where it resides within the detector shield a finite time.

s-The procedures used at NCSU do not require the activity input rate as a gas flow, but rather as a disintegration rate, and the exhaust air flow rate is not involved in any calculation.

Comparing the " Direct Calibration" procedure with that used at NCSU and re-ferring tothe eight (8) facts known about a properly operating system stated above, it is noted that:

1. Points 4 through 8 are common to both the " Direct" and "NCSU" ~

procedures.

2. Point 1 differs only in that the " Direct" procedure requires reasonably accurate knowledge of the activity flow rate; where-as, the "NCSU" procedures depend upon certified sealed sources and their disintegration rate. Note that an " activity flow rate" requires more test eeuipment; hence, a potential for problems.

3. Point 3, exhaust air flow rate is required by " Direct Calibra-tion" procedures, but not by the "NCSU" procedures. i These three points of comparison indicate that the "NCSU" procedure is, at least, as acceptable as the " Direct Calibration", if not a preferred pro-cedure. Therefore, we prefer to continue to use our established procedure, and respectfully request your concurrence with this position in the light of the information presented in this letter.

Very truly yours, Robert G. Cockrell RDC: Director, RGC:lsh Nuclear Reactor Program cc: Dr. Paul J. Turinsky Mr. Thomas C. Bray Mr. Robert D. Cross l

l

-,.- - .-- -- , _