ML20116M184
| ML20116M184 | |
| Person / Time | |
|---|---|
| Site: | Fort Saint Vrain  |
| Issue date: | 08/13/1996 |
| From: | Borst F Public Service Enterprise Group |
| To: | Weber M NRC, NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9608200062 | |
| Download: ML20116M184 (9) | |
Text
___ _ _ _ _ _ _ _. _ _ _ _ _ _ _.
S eubiic service' e= =-
16805 WCR 191/2; Platteville, Colorado 80651 August 13,1996 Fort St. Vraiu P-96066 U. S. Nuclear Regulatoy 43mmission ATTN: Document Control Desk Washington, D. C. 20555 ATIN:
Mr. Michael F. Weber, Chief Decommissioning and Regulatory Issues Branch Docim No. 50-267
SUBJECT:
Response to Comments Regarding Measurement of Exposure Rates with In-Situ Gamma Spectroscopy
REFERENCES:
1.
NRC Letter, Pittis io to Crawford, dated July 22,1996 t
(G-96127) 2.
PSCo Letter, Borst to Weber, dated May 17,1996 (P-96039)
Dear Mr. Weber:
Attached are Public Service Company of Colorado's (PSCo) responses to the NRC's comments in Reference 1, regarding use of in-situ gamma spectroscopy to measure exposure rates during the Fort St. Vrain final survey. PSCo had requested approval (Reference 2) to use an in-situ gamma spectroscopic instrument, the Microspec-2, to measure exposure rates directly in many areas of the facility.
If you have any questions regarding this information, please contact Mr. M. H. Holmes at (303) 620-1701.
qIi Sincerely, O
Mh Frederic Borst Decommt ioning Program Director f
9608200062 960813 PDR ADOCK 05000267 W
4 P-96066 i
August 13, 1996 Page 2 i
FJB/SWC Attachment t
(
l cc:
w/ attachment Regional Administrator, Region IV l
Mr. Robert M. Quillin, Director Radiation Control Dmsson Colorado Depaitment of Public Health and Environment t
h f
k l
i l
I l
i l
l l
I I
'l
l FORT ST. VRAIN NUCLEAR STATION USE OF IN-SITU GAMMA SPECTROSCOPY FOR EXPOSURE KATE MEASUREMENTS RESPONSE TO NRC COMMENTS PROVIDED IN JULY 22,1996 LETTER FROM C. L. PflTIGLIO ATTACHMENT TO P-96066
RESPONSE TO NRC COMMENTS ON PSCo SUBMITTAL P-96039 General:
1.
The document uses several terms to describe the output of the MICROSPEC-2 including effective dose equivalent rate, exposure rate, and tissue equivalent dose rate. Please clarify what units are used for the MICROSPEC-2 output.
Response
The MICROSPEC-2 provides output in units of effective dose equivalent and effective dose i
equivalent rate, with units of prem and prem/hr, respectively. Only the effective dose equivalent rate, in prem/hr, is used.
2.
Has a characterization been performed to determine which areas are potentially affected by gamma's originating in activated concrete that are not readily measurable using the MICROSPEC-2.
Response
Yes. Numerous measurements have been performed in and around the prestressed concrete reactor vessel (PCRV). The results indicate that europium (primarily Eu-152), a radionuclide in activated concrete, is the only radionuclide present that interferes with exposure rate measurements using the cobalt-60 region of interest (ROI) method. This interference is only encountered for measurements collected within the PCRV cylinder. PCRV internal surface concrete samples indicate that this is limited to the vertical concrete surfaces which were adjacent to the reactor core and above. Eu was not identified in concrete samples below the reactor core.
Eu is readily measurable with the MICROSPEC-2. However, Eu causes interference with the j
cobalt-60 ROI when using the application described in PSCo submittal P-96039 and FSV-FRS-TBD-202, Final Survey Erposure Rate Measurements Using the MICROSPEC-2*. An alternate method using the MICROSPEC-2 for performing measurements when Eu influences the exposure rate has been identified and submitted for review and approval in PSCo's July 19 Letter, (P-96058).
Page 1 of 6
Specific:
3.
Page 4, Paragraph 3 This section states that the use of the sample mean for background is acceptable in unaffected areas. Is the sample mean currently being used as background in unaffected areas? If so, please provide the reference approving this approach.
Response
Prior to collection of exposure rate measurements with conventional instrumentation, such as the Ludlum M2350 and 44-2 Nal detector, it was first necessary to collect background measurements using the approach described in the FSV Final Survey Plan. This approach for exposure rate background determination involved the collection of measurements from numerous onsite and offsite locations which were not influenced by licensed activities. The background measurements were then averaged, and the mean value was used as the most representative of the exposure rate background for those measurement conditions. At these background exposure rate measurement locations, PIC measurements were also collected and a PIC correction factor generated. For instances where the background exposure rate measurements were collected from onsite locations, in-situ spectral analysis was used to validate that the measurements were not affected by the presence of licensed material. PSCo and the WT/SEG consider that this approach is consistent with Section 4.4.3.f. of the FSV Final Survey Plan.
Final survey exposure rate measurements were initially collected using a Ludlum M2350 and 44-2 Nal detector and then corrected for response relative to the PIC. The PIC corrected background exposure rate was then subtracted from the final survey PIC corrected exposure rate measurement. For instances where the results of these background corrected exposure rate measurements exceeded the most restrictive action level (5 R/hr), or a significant bias in the data set was observed, an investigation was performed to determine the cause. Typical investigation included the collection of additional exposure rate measurements using the NaI(TI), and/or the collection of PIC measurements to define a location-specific correction factor for the Nal(TI).
In some instances in-situ spectral measurements were collected to determine iflicensed material was the cause of the anomalous results.
For instances where in-situ spectral measurements collected during investigation indicated that licensed material did not affect the exposure rate, it was concluded that the location met the criteria for an acceptable background measurement area. It can also be concluded from this information that the exposure rate at the measurement location is due to natural radionuclide constituents present in the materials of construction, which may have been further influenced by location geometry, and not the result of licensed material. Therefore, the mean of the PIC-corrected Nal(TI) measurements (without background correction) provided a representative background value that accounted for the location-specific construction materials and structural configuration. Due to the high and extremely variable background exposure rate at FSV, with Page 2 of 6
a variation typically greater than the exposure rate action level, this investigation approach provided adequate resolution of what initially appeared as elevated exposure rates.
4.
Page 8, Paragraph 2 What is the administrative action level? What is the maximum allowable MDA? Is the maximum allowable MDA value proceduralized?
Response
Administrative action levels are defined in Section 3.3.6 of the FSV Final Survey Plan. For exposure rate measurements, the action level is 5 R/hr due to licensed material.
The MICROSPEC-2 measurement system MDA was discussed in PSCo submittal P-96039 in response to NRC comment number 4. Section 5.5 of TBD-202 discusses the determination of
)
count time necessary to achieve a specific Minimum Detectable Dose Equivalent Rate (MDDER),
j which is equivalent to MDA for this measurement system. The ec.a ion to calculate MDDER l
is also provided in Section 5.5. A maximum MDDER for Co-6 0 ieen established at 50%
of the ROI limit, or 0.675 prem/hr. Note that rem /hr is considervu equivalent to pR/hr in this case. A second criterion requires the ability to detect Cs-137 at 0.15 prem/hr. This allows validation of the Co-60 exposure rate fraction (0.9). These measurement requirements are specified in TBD-202.
Typically, the required MDDER for Co-60 and Cs-137 can easily be achieved with a count time of 4 minutes. This was based on a background spectrum obtained from what is believed to be the worst case location (resulted in the highest background due to the naturally occurring radionuclide activity and area geometry). Even with the presence of a significant quantity of K-40 in the spectrum, the ROI can be adequately placed to achieve 25 % of the ROI limit (in a 220 channel spectrum, the 1460 kev K-40 peak may interfere with the 1332 kev Co-60 peak which imposes a 25% limitation).
FSV procedures require a minimum count time of 5 minutes for the MICROSPEC-2, which is based on the conclusions provided in TBD-202, a FSV controlled distribution document.
Page 3 of 6
5.
Page 8, Note Please describe the method for adjusting the pttr for a specific field measurement. Is the adjustment proceduralized? Will the adjustment to the pttr be made when it appears that pttr is too high, as well as when the pttr appears to be low? In addition, provide the method for adjusting the fraction of exposure rate assumed to originate from other nuclides? Will changes to the default values of 0.3 for the pttr and 0.9 for the nuclide fraction be documented as investigation in the survey packages and release records?
~
Response
Under normal conditions the pttr currently used (0.3) will not be adjusted and, therefore, the method is presently not proceduralized. Tables 5 through 8 in PSCo Submittal P-96039 show a change in pttr to demonstrate the conservatism in the choice of a pttr of 0.3.
For both measurement locations, the background was well characterized and then licensed material brought into the area for measurement. In this situation the pttr can be adjusted to better represent the source / scattering conditions. However, for field measurements using the MICROSPEC-2 the background is not known and the source may be present in any configuration, with a reasonably conservative configuration being that used to generate a pttr of 0.3. Although it is possible to unfold the contributors to the exposure rate measurement, i.e., background, photopeak(s) and scatter, this would be a considerable task. If a unique condition is identified and the pttr requires evaluation and possible adjustment (increased or decreased), the evaluation, method and conclusions will be documented in a technical basis document prior to implementation.
The nuclide fraction is the easiest of the two values to evaluate. This may be adjusted using the same protocol which established the current value (0.9). Again, any change will be documented prior to implementation. Any change to a measurement result following initial collection will be documented as an investigation in the survey package and release record.
Page 4 of 6
1 6.
. Page 19 and 20, Table 1 and 2 The ratio of the MICROSPEC to PIC background measurements in Tables 1 and 2 are 0.3 and 0.85, respectively. The staff assumes that the PIC exposure rate measurements are relatively independent of gamma energy. This was confirmed by a review of the instrument specification sheet for the Reuter-Stokes Plc which shows an instrument response of +/- 35% from 70 kev to 10 MeV. Since the exposure rate reported by a PIC is based on the entire spectrum, exposure rate measurements will vary much less than 35%. Therefore, it appears that the MICROSPEC is underestimating the exposure rate for some spectrums. However, the staff acknowledges that this under response does not necessarily affect the conclusions regarding the effectiveness of the MICROSPEC when the Co-60 region of interest is used to calculate exposure rate. Please explain the difference between the background measurements in Tables 1 and 2, and the potential effect of this difference on exposure rate measurements that are based on the Co-60 region of interest.
Response
Tables 1 through 4 in PSCo Submittal P-96039 were generated to compare net exposure rates obtained with both the PIC and MICROSPEC using the conventional method, i.e., subtracting background from gross measurement results. The measurement location represented by Tables 1 and 2 was an area with metal grate floor (concrete floor approximately 15 feet below grate) and ceiling approximately 20 feet high. The measurements were performed between two large empty steel tanks separated by approximately 10 feet.
Several evaluations were performed at this location to determine the cause of the difference in background between the MICROSPEC and PIC. Although the results of these evaluations were inconclusive, a couple of possibilities exist. First, when evaluating the spectrum obtained with the MICROSPEC, all counts were obtained in the low end of the spectrum, below 250 kev, where the PIC demonstrates the greatest over response. This was likely due to the conditions at the measurement location (large metallic components and the absence of concrete within the immediate measurement location). Another possibility for the difference in the response of MICROSPEC and the PIC is the difference in the cosmic radiation component measured by the PIC (higher) and the MICROSPEC.
However, the most important outcome of the comparison is the accuracy of the two instruments when measuring the exposure rate due to licensed material. In all cases, the MICROSPEC measurement of licensed material exposure rate resulted in less than 20% error when compared to the' expected or calculated value. At the investigation level, 5 pR/hr, the MICROSPEC measurement was within 10%. The PIC, however, underestimated the exposure rate due to licensed material significantly. It was not until the licensed material exposure rate determined with the PIC approached approximately 20 R/hr that the error was less than 20%.
Page 5 of 6
i i
r Tables 5 through 8 demonstrate results when using the proposed ROI approach. Since the ROI e
j approach does not consider background, the difference between the PIC and MICROSPEC background measurement previously discussed has no effect on the results. Exposure rate f
measurements at FSV, excluding portions of the internal surface of the PCRV, will be performed using the ROI method described in PSCo submittal P-96039. Those portions of the PCRV l
l excluded are the subject of a separate submittal. (P-96058, dated July 19, 1996) i 7.
Page 8, Table 1
)
How were the values in the Calculated (pR/hr) column derived.
-l i
I i
Response
]l 1
Expected exposure rates " calculated" in all tables were generated using MicroShield Version j
4.10, taking into consideration the radionuclide, source strength (decayed), and measurement j
distances.
All measurements and calculations were performed using a Co-60 point source. Calculated i
results include the build-up in air based on the measured source-to-detector air gap. The source l
strength used in the MicroShield calculations was determined by decay correcting the source manufacturer's reported activity for the period of time between manufacture and measurement.
t 4
To minimize backscatter, which would not be accounted for in the MicroShield calculations, i
measurements were performed with the source approximately 3 feet above the floor surface, and j.
a minimum of 6 feet from any other solid surface. The source was also positioned to minimize l
the scatter from the source holder. This was accomplished by suspending the source at the measurement location. The source / detector configuration was the same for both the PIC and the MICROSPEC-2.
8.
Page 8, Table 2 l
Were the PIC measurements made in the rate meter mode or the integrating mode?
i
Response
i PIC measurements were obtained in the rate meter mode. For all measurements, following relocation of the source or instrument, a minimum of 12 measurements were performed. Since the Reuter-Stokes PIC does not reset once acquisition is stopped, during the following acquisition after relocation, the first 2 measurements were considered suspect and dropped. The remaining 10 measurements were then averaged to provide the results presented in the Table 2.
Page 6 of 6