ML20107G138
| ML20107G138 | |
| Person / Time | |
|---|---|
| Site: | Fort Saint Vrain  |
| Issue date: | 04/19/1996 |
| From: | Borst F PUBLIC SERVICE CO. OF COLORADO |
| To: | Weber M NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS) |
| References | |
| P-96031, NUDOCS 9604230117 | |
| Download: ML20107G138 (11) | |
Text
-. - _ _ _.
si
,k O rubiic service-
=e=-
16805 WCR 191/2; Platteville, Colorado 80651 April 19,1996 Fort St. Vrain P-96031 i
U.S. Nuclear Regulatory Commission 1
A'ITN: Document Control Desk Washington, D.C. 20555 A'ITN:
Mr. Michael F. Weber, Chief Decommissioning and Regulatory Issues Branch Docket No. 50-267
SUBJECT:
Rcsponse to NRC Questions Regarding PSCo's Proposed Revisions to Final Survey Plan For Survey of Piping Systems and Suspect Affected Survey Units 1
REFERENCES:
1.
NRC Letter, Pittiglio to Crawford, dated April 11, 1996 (G-96062) 2.
PSCo Letter, Fisher to Weber, dated October 12,1995 (P-95077) 3.
PSCo Letter, Borst to Weber, dated March 5,1996 (P-96009)
Dear Mr. Weber:
This letter submits Public Service Company of Colorado's (PSCo) responses to NRC comments provided in your April 11, 1996 letter (Reference 1), regarding proposed revisions to the Fort St. Vrain (FSV) Final Survey Plan. The propose changes were submitted in a October 12,1995 letter (Reference 2), and involved survey treatments for piping systems and suspect affected survey units. The NRC's comments requested clarification of proposed survey treatments for embedded piping, particularly regarding mformation PSCo had provided in a March 5,1996, letter (Reference 3).
The attachment to this letter provides PSCo's responses to the four comments in the referenced letter. Final surveys of affected FSV piping systems have begun, utilizing survey techniques described in the Reference 2 submittal. In order to minimize the 9604230117 960419 th I PDR ADOCK 05000267 W
PDH y
l
P-96031 April 19,1996 Page 2 amount of potential rework and avoid impacting the decommissioning schedule, PSCo requests NRC approval of the proposed treatment methodologies by May 24,1996.
If you have any questions regarding this information, please contact Mr. M. H. Holmes at (303) 620-1701.
Sincerely, hdfA<cL y
Frederic Jj Borst Decommirstoning Program Director FJB/SWC 4
Attachment l
cc:
with attachment Regional Admtnistrator, Region IV Mr. Robert M. Quillin, Director Radiation Control Division Colorado Department of Public Health arid Environment f
I 4
1 i
6
,9 P-96031 Attachment' i
Page 1 4
Attachment to P-96031 Public Service Company of Colorado (PSCo) and the Westinghouse Team (WT) submit the following response to NRC comments in their April 11,1996 letter, " Response to the Fort St. Vrain Proposed Revisions to Final Survey Plan for Site Release for Survey of 1
Piping Systems and Suspect Affected Survey Units":
NRC Comment No.1:
}
The TLDs appear to provide an acceptable estimate ofthe average contamination in the pipes. However, the uncertainty of measurements made by individual TLDs, and the ability ofindividual 7LD measurements to demonstrate compliance l
with maximum contamination limits, has not been supiciently characterized. The analysis provided in Attachment 1 to PSCo's March 5,1996, response to NRC February 12, 1995, Comment 10, provides only a panial estimate of this uncertainty.
Please provide a full characterization of the uncertainty of individual TLD measurements oflocalized contamination, i.e., small areas of contamination in geometries that diferfrom the calibration geometry.
Are TLDs suitedfor the quantitative assessment oflocalized contamination, or should TLDs be considered to quant {fy average contamination levels only, and to provide qualitative estimates oflocalized contamination?
PSCo/WT Response:
The WT developed the TLD string survey method for surveying small bore embedded piping with bends, because no other technology for this purpose currently exists. The intention was to provide additional information on contamination levels within long runs of embedded pipe in lieu of surveying only the piping ends with commercially available detectors, which is the method deemed adequate in Draft NUREG-5849. The technology has shown the ability to provide good estimates of the average contamination in piping and, provided that contamination is not highly localized, a reasonable representation of the contamination profile in the piping.
The TLD string method does not provide a high level of certainty for quantifying the maximum contamination level at a given localized area in a pipe (unless the location is specifically known and TLDs are positioned at that location).
For example, contamination may be at a location where TLDs are not located or in a different
.. =.
4 f
P-96031 Attachment Page 2 geometry (e.g., a small spot at the edge of a TLD) than assumed by the calibration process. Despite this limitation, using TLDs strings to survey embedded piping is considered appropriate for the following reasons:
2 1.
The bounding value of 100,000 dpm/100 cm for individual contamination measurements was proposed by PSCo as the point at which we would notify the NRC, not as an absolute limit on the maximum contamination level in piping to 2
be grouted. This level was included to provide an upper limit on individual measurements, which is especially appropriate in view of the fact that specialized detectors capable of surveying small bore embedded piping with bends would necessarily be less precise than standard detectors. Because the bounding dose i
calculation presented in the October 12, 1995, submittal assumed that average 2
contamination is at 100,000 dpm/100 cm, this level is appropriate as an average contamination limit.
l 2.
If embedded piping does not meet the SGLVs after completing an aggressive decontamination effort, it will be filled with grout, which fixes any residual 4
contamination in the piping. Therefore, reasonable exposure scenarios are dominated more by the total amount of contamination in the pipe (i.e., average) and less by the precise distribution of contamination within the pipe.
3.
Embedded piping is not accessible and poses a much lower radiological risk to personnel compared to localized contamination on accessible floors or walls.
Accessible ends of surveyed embedded pipes, which present the most likely exposure pathway, will be surveyed with gas flow pipe probes to the first elbow, so elevated contamination levels in these areas are more accurately characterized.
I 4.
Additional analysis indicates that localized contamination as high as 1,000,000 dpm/100 cm does not significantly increase the previously estimated exposures.
2 This 1,000,000 dpm/100 cm contamination levelis conservative for the following 2
reasons:
2 1,000,000 dpm/100 cm is a factor of 10 higher than the measurement 2
upper limit of 100,000 dpm/100 cm, which accounts for the TLD uncertainties identified in Section 5 below PSCo/WT's embedded pipe survey data for pipes to be surveyed with TLDs, taken before decontamination efforts, indicate that all 2
measurements were much less than 1,000,000 dpm/100 cm i
P-96031 Attachment '
Page 3 The exposure analysis includes the following conservative assumptions:
The previously evaluated dismantlement and disposal scenario is used Elevated contamination is assumed to remain in one-half of the dismantled 1-inch pipe sections; it is highly unlikely that this many areas of elevated contamination would go undetected In each of these pipe sections with elevated contamination, there are three 2
2 100 cm spots of contamination at 1,000,000 dpm/100 cm (.e., a total of 300 cm ).
S nce TLDs are spaced at 20" with each TLD calibrated 2
with a 100 cm source which is 5" of pipe length, the maximum 2
2 unmonitored area between TLDs is 15" or 300 cm. Therefore, the 2
exposure analysis also accounts for 1,000,000 dpm/100 cm og contamination over the area between two TLD locations.
j No credit is taken for the shielding effects of grout The rest of the pipes are contaminated to an average level of 100,000 2
dpm/100 cm The resulting dose is 3.1 mrem per year. This represents a modest increase l
above the 2.4 mrem dose previously evaluated; however, it is still small compared to the 10 mrem per year criteria provided by the NRC for soil and water pathways, as identified in the Final Survey Plan, and the limits being considered in the NRC Proposed Decommissioning Rulemaking.
2 5.
If individual measurements are greater than 50,000 dpm/100 cm, investigations 4
2 are performed with GM detector assemblies where possible. 50,000 dpm/100 cm is considered an appropriate investigation level for the following reasons:
i a.
Uncertainty analysis presented in Attachment 1 of PSCo's March 5,1996, response to NRC February 12, 1996, Comment 10, indicates that TLD uncertainty for a point source within the TLD's defined field of view is less than a factor of 6 (areas of larger size would be detected with lower uncertainty). Although this analysis did not include all possible sources 2
of uncertainty, investigations at 50,000 dpm/100 cm are a factor of 20 below the localized contamination level analyzed above, which provides reasonable conservatism.
b.
Uncertainty analysis presented in Attachment 2 of PSCo's March 5,1996, response to NRC February 12, 1996, Comment 10, indicates that TLD uncertainty for measuring maximum levels of actual contamination 2
distributions is less than 25 %. 50,000 dpm/100 cm allows uncertainty up to 100% prior to exceeding the individual measurement bounding value of 2
100,000 dpm/100 cm proposed by PSCo, providing added conse.rvatism.
4
P-96031 Attac'hment '
Page 4 c.
Investigation of highly localized contamination above 50% of the individual measurement bounding value is sufficiently conservative to ensure potential doses from the grooted piping are within the 3.1 mrem / year value identified above. Should an elevated area be larger in size (as indicated by multiple consecutive TLD readings), the TLD results can be expected to provide a reasonable estimation of contamination levels in the region. If there are many single TLD readings that are elevated, the average contamination in the pipe as determined by TLDs becomes more accurate.
6.
Even if investigations are not possible with GM detector assemblies, the large number of measurements taken when TLDs surveys are performed provide reasonable assurance that good estimates of average contamination are obtained which provides assurance that potential doses will not exceed those presented above. Also, additional decontamination may be performed at a reduced action level where investigations are not possible, as discussed in PSCo's response to Comment No. 3 below.
l l
I l
l
P-96031 Attachment Page 5 NRC Comment No. 2:
Based on the uncertainty analysis requested abow, please providejustifcationfor the investigation lewis provided in PSCo's March 5,1996, response to NRC February 12, 1996, Comment 1.
Because of the uncertainty in individual measurement results, it may be appropriate to base innstigation levels, and decisions to perfonn additional decontamination, on lower marimum measurements than proposed by PSCo.
PSCo/WT Response:
In the March 5,1996, response, PSCo proposed to investigate any individual 2
measurement in embedded pipes of 50,000 dpm/100 cm or greater, and any individual 2
pipe with average contamination greater than 25,000 dpm/100 cm, provided this individual pipe is significantly more contaminated than the rest of the pipes in the survey unit.
2 The proposed individual measurement investigation level of 50,000 dpm/100 cm ;s considered appropriate based on the following considerations:
1.
50,000 dpm/100 cm is 50% of the measurement bounding value proposed in the 2
2 submittal (i.e., 100,000 dpm/100 cm ) which is appropriate as an average contamination level limit. This is conservatively lower than the normal 75%
(
investigation action level associated with affected plant system survey units.
2.
The 50% investigation level is reasonable based on the analysis of past TLD survey data as presented in Attachment 2 to PSCo's March 5,1996, response to NRC's February 12, 1996, Comment 10. This data indicated that areas of elevated contamination are typically observed over several adjacent TLDs, so that 2
areas of contamination near the 100,000 dpm/100 cm level are highly likely to be detected at a level that would initiate an investigation, so that they can be adequately characterized.
3.
A highly localized " spot" of contamination that is measured below 50% of the measurement bounding value, such that an investigation is not required, would result in very small dose consequences, as discussed in PSCo's response to Comment No.1 above.
2 Investigation of pipes with an average contamination level of 25,000 dpm/100 cm and that are significantly more contaminated than the rest of the pipes in the survey unit is addressed in response to Comment No. 4, below.
==
P-96031 Attachment' Page 6 NRC Comment No. 3:
What percentage of the 1 inch pipes are accessible for surwy with the GM detectors if an innstigation is required? What actions are proposed if the GM detectors cannot be used? Will additional decontamination be performed?
PSCo/WT Response:
When initially proposed, investigation of individual TLD measurements by GM detector assemblies was thought to be possible for most 1" piping survey units. However, our I
subsequent experience has shown that, for some survey units, investigations by GM detector assemblies will be possible for only some of the pipes. Some 1" pipes are restricted to GM probes due to butt welded joints, or radical bends as piping appears to be bent around rebar. This condition precludes the use of GM detector assemblies for initial survey or investigations where use of TLDs may be possible.
The 1" piping survey units that will be surveyed with TLDs are listed below with an estimated percentage that is expected to be accessible by GM detector assemblies.
System 13 - Equipment Storage Well Vent 90 - 100% accessible by GMs and Drain Piping l
l System 14 - Fuel Storage Well Vent Piping 90 - 100% accessible by GMs Core Support Floor Column Piping 25% accessible by GMs PCRV Cooling Tubes 50% accessible by GMs For the Core Support Floor Column Piping and PCRV Cooling Tube survey units which have the most limited investigation potential, PSCo expects contamination levels well 2
belcw the 100,000 dpm/100 cm upper measurement limit. Survey measurements in both 2
of these survey units were less than 100,000 dpm/100 cm before decontamination efforts, and all pipes in these survey units have been aggressively decontaminated as described in our April 12, 1996, letter (P-96028).
For piping accessible by only TLD strings, additional investigation surveys will not be practical. Due to the long exposure periods required for TLD measurements (typically 2 to 3 months), a sigr.ificant schedule impact would occur if re-survey with TLD strings was required.
The additional minimal benefit provided by obtaining additional measurements does not justify the added survey costs and schedule delays.
l 1
l I
P-96031 Attachment' Page 7 For pipe sections that cannot be investigated, PSCo proposes to minimize associated dose by performing additional decontamination, followed by grouting. For pipes with readings 2
greater than 50,000 dpm/100 cm that cannot be investigated, additional decontamination will be performed for all pipes in the survey unit if the average contamination in the pipe 2
is greater than 50,000 dpm/100 cm. This is a factor of 2 less than the 100,000 dpm/100 cm upper measurement limit for pipes that can be investigated, and is conservative 2
because it would not be reasonable to expect that investigation surveys (if they could have been performed) would identify elevated contamination significant enough to raise the average contamination level by a factor of 2.
If individual contamination measurements are observed greater than 100,000 dpm/100 cm, PSCo/WT will perform an engineering evaluation of additional decontamination and 2
subsequent grouting actions.
If average contamination measurements are observed 2
greater than 100,000 dpm/100 cm or if individual measurements are greater than 200,000 dpm/100 cm, PSCo/WT will perform additional decontamination and 2
subsequent grouting. PSCo will notify the NRC of any individual measurement in excess 2
of 100,000 dpm/100 cm, prior to grouting.
l l
i
P-96031 Attachme'nt" Page 8 NRC Comment No. 4:
PSCo indicates that an investigation will be conducted if the average contamination in a glwn pipe exceeds 25,000 dpm/100 cm' and is twice the averagefor the surwy unit. This innstigation lent appears high. The objective of the comment (NRC February 12, 1996, Comment 2) was to ensure that the measurement resultsfrom selected pipes are representative of the entire surwy l
unit. The resultsfrom individualpipes could indicate a non-representative sample at awrage contamination lewis below 25,000 dpm/100 cm.
Please provide additionaljustifcationfor the selection ofthe 25,000 dpm/200 cm*
value at the "two times surwy unit average" value, and why investigation at lower values are not necessary to demonstrate that the survey results in a glwn pipe are representative of the entire surwy unit.
i i
PSCo/WT Response:
i i
Pipes selected for final survey are conservatively biased toward higher contamination potentials to the extent practical. For example, Core Support Floor Column tubes i
selected for survey include a higher percentage of vent and grout tubes than regular tubes. The vent and grout tubes are considered to possibly have a higher contamination l
potential due to possible exposure to primary coolant during plant operation.
4 Additionally, removable surface activity measurements are taken with each pipe (as the 4
fm' al step in decontamination); therefore, those pipes that contained higher removable surface activity (based on removable activity measurements or number of passes required i
to meet removable criteria) are selectively targeted for survey. By selecting biased pipes i
for survey, the survey approach is conservativmd will reasonably bound contamination i
remaining in the pipes.
(
25,000 dpm/100 cm was selected as an investigation level for the average contamination 2
in a given pipe because this corresponds to 25 % of the individual measurement bounding level. This investigation level (i.e.,25 %) is consistent with the lowest action levels used j
with final survey measurement results (i.e., the action level associated unaffected survey units). The factor of two above survey unit average was included in the event (although j
not expected) that a given survey unit's average pipe measurements are consistent but at 2
a level near or above 25,000 dpm/100 cm. In this example, a representative survey 2
could be performed with measurements above 25,000 dpm/100 cm but still significantly below the individual measurement bounding level and further investigation would provide very little benefit.
i l
P-96031 Attachinent' Page 9 These investigation criteria are considered to provide reasonable assurance that contamination levels are low enough to ensure potential doses from grouted piping are 1
well within the values identified above. These investigation levels will also help minimize unnecessary investigations which provide very little benefit in maintaining doses ALARA but require a high cost in additional survey time and schedule delay.
r 4
i.
J 1
4 4
n
~