L-PI-11-019, Supplement to License Amendment Request to Exclude Dynamic Effects Associated with Certain Postulated Pipe Ruptures from Licensing Basis Based Upon Application of Leak-Before-Break Methodology - Response to Requests for Clarification

Supplement to License Amendment Request to Exclude Dynamic Effects Associated with Certain Postulated Pipe Ruptures from Licensing Basis Based Upon Application of Leak-Before-Break Methodology - Response to Requests for Clarification

ML110550582
Person / Time
Site:	Prairie Island
Issue date:	02/23/2011
From:	Schimmel M Northern States Power Co, Xcel Energy
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
L-PI-11-019, TAC ME2976, TAC ME2977
Download: ML110550582 (9)

Text

L-PI-11-019 I0 CFR 50.90 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555-0001 Prairie Island Nuclear Generating Plant Units I and 2 Dockets 50-282 and 50-306 License Nos. DPR-42 and DPR-60 Supplement to License Amendment Request to Exclude the Dynamic Effects Associated with Certain Postulated Pipe Ruptures From the Licensing Basis Based Upon Application of Leak-Before-Break Methodoloav - Response to Requests for Clarification (TAC Nos. ME2976 and ME29771

References:

1. Letter from M.

A. Schimmel, Northern States Power Company, a Minnesota corporation (NSPM), to Document Control Desk (Nuclear Regulatory Commission, NRC), License Amendment Request to Exclude the Dynamic Effects Associated with Certain Postulated Pipe Ruptures From the Licensing Basis Based Upon Application of Leak-Before-Break Methodology, L-PI-09-134, dated December 22, 2009, ADAMS Accession Number ML100200129.

2. Letter from M. A. Schimmel (NSPM) to Document Control Desk (NRC), Supplement to License Amendment Request to Exclude the Dynamic Effects Associated with Certain Postulated Pipe Ruptures From the Licensing Basis Based Upon Application of Leak-Before-Break Methodology - Response to Request for Additional Information (TAC Nos. ME2976 and ME2977), L-PI- 11-006, dated January 14, 201 1, ADAMS Accession Number MLI 10140367.

In Reference 1, Northern States Power Company, a Minnesota corporation (NSPM), doing business as Xcel Energy, submitted a License Amendment Request (LAR) to apply leak-before-break (LBB) methodology to certain piping systems at the Prairie Island Nuclear Generating Plant (PINGP). As part of the review effort for this LAR, NSPM submitted additional information regarding PINGP Reactor Coolant System (RCS) leak detection capabilities in Reference

2. 171 7 Wakonade Drive East Welch, Minnesota 55089-9642 Telephone:

651.388.1 121 Document Control Desk Page 2 In e-mails dated January 18 and 263, 201 I, and baeed on an informal dissussion on January 26, 201 1, the NRC staff requested clarification of the information provided in Reference

2. The enclosure to this leBer provides the requested clarification of PINGP RCS leakage detection capabilities, NSPM submits this supplement in accordance with 10 CFR 50.90. The supplemental information provided in this ieaer does not impad the conclusions of the Determination of No Significant Hazards Consideration or Environmental Assessment presented in the Reference 1 submittal, In accordance with 10 CFR 50.91, NSPM is notiving the State of Minnesota of this LAR supplement by transmitting a copy of this letter to the designated State Official.

If there are any questions or if additional information is needed, please contact Sam Chesnutt at 651-267-7546.

Summaw of Commitments This letter contains no new commitments and no revisions to existing commitments.

I 1 I declare under penalty of perjury that the foregoing is true and correct.

i i Executed on FEB 2 3 2011 i I I Mark A. Schimmel Site Vice President, Prairie Island Nuclear Generating Plant Northern States Power Company - Minnesota Enclosure cc: Administrator, Region Ill, USNRC Project Manager, PINGP, USNRC Resident Inspector, PINGP, USNRC State of Minnesota This enclosure includes responses from the Northern States Power Company, a Minnesota corporation (NSPM), to requests for clarification of information regarding Reactor Coolant System (RCS) leakage detection capabilities at the Prairie Island Nuclear Generating Plant (PINGP). These requests were transmitted in e-mails dated January 18 and 28, 201 1 (References 6 and 7), and addressed information previously submitted January 14,201 I (Reference 5). The information provided in this enclosure is associated with NSPMk License Amendment Request (MR) submitted December 22,2009 (Reference

1) regarding the use of Leak-Before-Break (LBB) methodology.

To support review of Reference 1, the Nuclear Regulatory Commission (NRC) issued an RAI regarding, in part, RCS leakage detection capabilities (Reference 2). NSPM responded to the Reference 2 RAls regarding RCS leakage detection in Reference

3. The NRC requested additional information in Reference 4, and the NSPM response was submitted in a letter dated January 14, 201 1 (Reference 5). This Enclosure quotes each request for clarification in italics and each question is followed by the NSPM response. Referenced documents are identified at the end of this Enclosure.

NRC Clarification Question 1 (from January 18, 201 1 e-mail): In the response to question 3.A., the licensee stated: "Additional moisture will then flow to the sump.

Thus, a 0.2 gpm leak will be detectable by the containment sump run time monitors despite the effects of evaporation. " Is NSPM now crediting the containment sump run time monitor for LBB? If so, what is the expected response time once a 0.2 gpm flow reaches the sump? NSPM Response to Clarification Question I: No. The containment sump run time monitor was described as one of the eight diverse monitoring methods that are available to detect RCS leakage at PINGP, but it is not being credited for LBB. Page 1 of 7 Enclosure Clarifications Regarding RCS Leakage Detadion NSPM NRC Clarification Question 2 (from January la8, 2011 ammail): In the response to question I.A., the licansee stated: "As part of the OBN rescllution effort, NSPM performed a more detailed calculation of the msponse time capabilities of the containment particulate monitors, If?-I I and 2R-7 I, The new calcul~tion accounted for the additional activity contributc43.d by the daughter products of noble gas decw which were not previously included, while continuing to assums conssntativdy low circulating activity levels consistent with curmnt nomal plant operations.

The results of this calculation show that the R-I I monitors in both Units I and 2 are capable of detecting a I gpm leak within I hour." Daughter products of noble gas decay do not typically remain airborne indefinitely, and the credit of the daughter products appears to provide greater than an order sf magnitude improvement in the responsiveness of the detector. Explain how the accumulation of daughter product activity was modeled in the detector response time determination.

Also, describe any benchmarking or other testing that supports the results of the response time determination.

NSPM Response to Clarification Question 2: The accumulation of daughter product activity in the containment atmosphere is modeled as follows: The accumulation of the parent noble gas is modeled using equations described in ISA 67.03-1982, "Standard for Light Water Reactor Coolant Pressure Boundary Leak Detection." The production and depletion of daughter products is modeled to reflect their radioactive decay properties.

The behavior of particulate daughters is modeled using removal coefficients in NUREGICR-6189, "A Simplified Model of Aerosol Removal by Natural Processes in Reactor Containments." Noble gas accumulation The accumulation of parent noble gas in containment due to RCS leakage is calculated using equation B-27 contained in ISA 67.03-1 982, Section B.3.1, "Airborne radiation monitoring coolant leakage measurement." This equation computes the concentration of the parent nuclide in containment at any time following the onset of an RCS leak, with consideration of depletion of the nuclide due to decay. Radioactive decav process The production rate of the daughter particle at any instant in time is a function of the parent activity in containment at that instant and the decay constant of the parent. The depletion rate of the daughter due to decay at any instant is a function of the daughter activity at that instant and the daughter's decay constant. The calculation of daughter activity at any time requires integration of the overall expression over the time interval of interest.

This calculation was performed by direct solution of the differential equation.

Page 2 of 7 1 Enclosure Clarifications Regarding RCS Leakage D@ts&ion NSPM Particulate behavior Once generated in the containment atmosphere, the behavior of the particulate daughters of noble gas decay is modeled using the removal coefficients from NUREGICR-6189.

While the removal coeRcients in the NUREG are developed to model the deposition processes in effect in an accident environment, it is consenrative to apply these factors to normal at-power reactor containment conditions because the aerosol removal mechanisms are much stronger in a post-accident environment. The processes of agglomeration, settling, and plateout of aerosols are time-dependent and are accounted for in the time dependent removal coefficient used in the model. Prairie Island maintains the airflow through the detector tubing at very low velocities to maintain laminar conditions, as recommended in HPSlANSI N13.1-1999 for minimizing iner(ial impaction losses. Therefore, plateout in the tubing is considered negligible. The filter paper has a collection efficiency of 99.99% of particles greater than 0.3 micron size, so filter pass-through is considered inconsequential.

Particulates originating from the leaking coolant are modeled with an instantaneous 99.9% plateout at the source of the leak in accordance with ISA 67.03-1 982. Noble gases are not subject to plateout at the site of the leak, and the daughter products of the noble gas decay are born in the well mixed containment air before being subject to removal by the time-dependent agglomeration and settling processes. Source term The coolant source terms used in the calculation of the R-I I response are obtained from recent plant data.

For particulate in the coolant, values were averaged from radiochemistry samples performed in early 201

0. The time period represented by these samples is one of excellent fuel performance and no known leaks. Due to the 99.9% plateout factor applied, particulate in the coolant was not a significant contributor to the results. For the noble gases, which generate the decay daughters that dominate the R- 11 response, the source term was generated by drawing a conservative lower bound beneath all of the radiochemistry data for power conditions. Because the noble gases are a result of fissions of tramp uranium in the fuel clad rather than being due to fuel defects, it is anticipated that the lower bound noble gas activity source term will remain valid throughout future operating conditions.

Fluorine-1 8 (F-18), an activation product that exists in high concentrations in Pressurized Water Reactor (PWR) coolant and which is known to generate particulate activity, has not been considered in Prairie Island's calculation of the R-I I response due to the difficulty in quantitative radiochemical analysis for F-18, and this adds conservatism to the results. Detector efficiencv The detector efficiency used in the calculation of the R-I I response is conservative. The vendor supplied calibration data for detector efficiencies for three radionuclides:

Page 3 of 7 Cesium (&)I37 Technetium (Tc) 99 Strontium

/ Yttrium (SrJY) 90 2.06E+5 cpm/pCi 1.08E+5 cpm/pCi 4.93E+5 cpm/pCi Enclosure I Clarifications Regarding RCS Leakage Detection NSPM The SrN-90 source used in the calibration emits Zwo betas, one with an average energy of 196 keV and the other with an average energy of 934 keV, while the Cs-137 and Tc- 99 sources have lower energy betas. The two dominant daughter products of noble gas decay considered in the R-I I response calculation are Rubidium (Rb)-88 and Cs-138. Rb-88 has a mean beta energy of 2051 keV and 6s-138 has a mean bet8 energy of 1240 keV. Since the nuclides of principal interest to this calculation have much more energetic betas than the SrN-90 source, it can be concluded that the d-ctor emciency for Rb-88 and Cs-I38 should be at least as high as for SrlY-90. However, for conservatism, and in order to allow the use of a single efficiency to represent the response of all the radionuclides present on the filter paper, a vendor-supplied nominal detector efficiency value of 3.50*+5 cpm/pCi was applied, Detection The instrument sensitivity and detection time for the postulated RCS leak is based on the criteria in ISA 67.03-1982 (included in Bibliography of Regulatory Guide (RG) 1.45, Revision I), which specifies that the change in the process variable is statistically detectable at a 99% confidence if it exceeds 2.56 standard deviations of the background count rate distribution.

Benchmarking The calculated response time of the newly installed R-I I instrumentation has not been tested or benchmarked to actual plant leakage events because an active leak of sufficient magnitude and duration has not been experienced. NRC Clarification Question 3 (from January 28, 201 1 e-mail): As discussed during our January 26, 201 1, telephone conversation, following is a supplemental request for clarification of your January 14, 201 1, RAI response concerning Prairie Island's Leak-Before-Break license amendment request: In the Response to NRC Question

?(a), which was provided in Enclosure I to the letter dated January 14, 201 1, the R-I I monitor was described as having the ability to detect a I gallon per minute (gpm) leak within I hour, consistent with Regulatory Guide (RG) 1.45, Revision I, "Guidance on Monitoring and Responding to Reactor Coolant System Leakage." In addition, the response described that a new calculation showed that the R-I I monitor is capable of detecting a 0.2 gpm leak within approximately 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />. However, Regulatory Positions 2 and 3 of RG 1.45, Rev.

1, state that:

leakage detection systems should have a response time of no greater than 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> for a leakage rate of I gpm leakage detection systems should provide output and alarms in the main control room Page 4 of 7 Enclosure Clarifications Regarding RCS Leakage Detection NSPM procedures to convert the instrument output to leakage rste should be available to operators plant procedures should specij/ opembr actions in response to leakage rates less than the limits set fodh in the t8chnical spscifications Please clarify the meaning of the tern "datecf" in the response to NRC Question l(a). The Glossary provided with RE 1.45, Rev. 1, may be helpful. In addition, based on plant operating and suweillance pmcedures, speciy the expected maximum time for operators to determine that the 0.2 gpm leak rate has been exceeded using the R-7 7 instrument output, the leakage rate expected to correlate with the setpoint of any alarms associated with the R- I I instrument, and the opera for actions speciflied for unidentified leakage exceeding 0.2 gpm based on the R-I I instrument output.

NSPM Response to Clarification Question 3: Use of the containment radiological particulate monitor R-1 I to detect and initiate effective responses to RCS leakage is clarified as follows. Meaning of "detect" I The statements in the January 14, 201 1 response to RAI Question 1(a) regarding the ; I ability of the R-I I monitor to "detect" RCS leakage are based on the detection definition in ISA 67.03-1982, as discussed in the response to Clarification Question 2 above. This I I I standard is included in the Bibliography section of RG 1.45, Revision 1. ISA 67.03-1 982 I specifies that, for a particulate radiation monitor, the "minimum detectable concentration" of a radioactive particulate aerosol is that which generates a net counting rate greater than 2.56 times the standard deviation of the background counting rate, at 99% statistical confidence.

It follows that the increase in containment activity resulting from a leak is detectable once it has increased above the minimum detectable concentration of the instrument.

Therefore, the response time for the R-I I instrument to detect a leak of a given size is the calculated time from the onset of the leak to the time when the containment activity results in an increase in the count rate greater than 2.56 times the standard deviation of the background count rate. Such an increase would be discernable from background with 99% statistical confidence.

Time for operators to identifv a 0.2 npm leak Plant operators review R-I I data every hour as part of the Emergency Response Computer System (ERCS) log check, and this would be an opportunity for an elevated count rate to be observed. When the ERCS computer is not available, the ERCS Computer Out of Service Log procedure is performed which requires R-I I data to be checked every 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. The Radiation Monitor Panels are channel checked for operability every six hours, and this would be another opportunity to observe elevated count rates. In addition, operators trend and average R-I I data every 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. Page 5 of 7 Enclosure Clarifications Regarding RCS Leakage Detection It is important to point out that plant operators use a number of indicators to make decisions regarding RCS leakage, although the NRG" question is specific to use of R-I I instrument output. It is expeded that when an elevated R-I I count rate is observed, operators would look at other indications such as containment humidity, charging pump operation (an RCS leak would cause the charging pumps running with speed control in automatic to speed up), volume control tank levels, and pressurizer levels. Increased radiation indicatian on the R-1 I monitor is an entry condition for the Reactor Coolant Leak Abnormal Operating Procedure.

This procedure describes the symptoms associated with small reactor coolant leakage, the methodology for determining the path of such leakage, and the necessary corrective action. Also, an RCS inventory balance is performed every 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> as previously described in the Reference 3 response to Question E2-2, and this inventory balance can be performed at any time an operator suspects unidentified leakage.

Based on the R-I I instrument detection time provided in Reference 5, and on the procedure requirements and operator practices described above, the maximum time that it would take plant operators to determine that the 0.2 gpm leak rate has been exceeded is expected to be 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. Leakage rate associated with alarms The alarm point established for the PlNGP R-I I containment particulate monitor is a high radiation monitor alarm that is part of the Emergency Action Levels in the Emergency Plan.

This alarm setpoint is not correlated to any specific RCS leakage rate. I Specified Operator actions Current procedures require a trend analysis of R-I I indications once per 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, and if R-I I activity is observed to increase by at least a factor of 3 above the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> average and the source cannot be determined, then a Reactor Coolant Leakage lnvestigation is initiated.

Operators use a number of indicators to make decisions regarding RCS leakage, although the NRC question addressed the use of R-I I instrument output. Increasing R-I I instrument output is a direct entry into the Reactor Coolant Leak Abnormal Operating Procedure as described above. The RCS leakage test surveillance procedure includes a requirement that if the unidentified leakage rate exceeds 0.2 gpm, a Reactor Coolant Leakage lnvestigation is initiated.

If this investigation determines that leakage inside containment exceeds 0.1 gpm, specified response actions include issuing an Action Request within the PlNGP Corrective Action Program, performing the 8 Reactor Coolant Leak Abnormal Operating Procedure if necessary, considering a containment entry to identify the source of the leakage, and taking further actions as required by the Technical Specifications for unidentified leakage. Page 6 of 7 Enclosure Clarifications Regarding RCS Leakage Detedian NSPM References

1. Letter from M. A. Schimmel (NSPM) to Document Control Desk (NRC), "License Amendment Request to Exclude the Dynamic Effects Associated with Certain Postulated Pipe Ruptures From the Limnsing Basis Based Upon Application of Leak-Before-Break Methodology," L-Pl-09-134, dated December 22, 2009, ADAMS Accession Number MLI 000200129, 2. Letter from T. J. Wengert (NRC) to M. A. Schimmei (NSPM), "Prairie lsland Nuclear Generating Plant, Units 1 and 2 - Request for Additional lnformation Related to License Amendment Request to Exclude the Dynamic Effects Associated with Certain Postulated Pipe Ruptures From the Licensing Basis Based Upon Application of Leak-Before-Break Methodology (TAG Nos. ME2976 and ME2977)," dated June 10,201 0, ADAMS Accession Number MLI 01 550668. 3. Letter from M. A. Schimmel (NSPM) to Document Control Desk (NRC), "Supplement to License Amendment Request to Exclude the Dynamic Effects Associated with Certain Postulated Pipe Ruptures From the Licensing Basis Based Upon Application of Leak-Before-Break Methodology - Response to Request for Additional lnformation (TAC Nos. ME2976 and ME2977),11 L-PI 094, dated October 8,201 0, ADAMS Accession Number ML102810518.

4. Letter from T. J. Wengert (NRC) to M. A. Schimmel (NSPM), "Prairie lsland Nuclear Generating Plant, Units 1 and 2 - Request for Additional lnformation Related to Request to Exclude the Dynamic Effects Associated with Certain Postulated Pipe Ruptures From the Licensing Basis Based Upon Application of Leak-Before-Break Methodology (TAC Nos. ME2976 and ME2977)," dated December 14,2010, ADAMS Accession Number ML103280398.

5. Letter from M. A. Schimmel (NSPM) to Document Control Desk (NRC), "Supplement to License Amendment Request to Exclude the Dynamic Effects Associated with Certain Postulated Pipe Ruptures From the Licensing Basis Based Upon Application of Leak-Before-Break Methodology - Response to Request for Additional lnformation (TAC Nos. ME2976 and ME2977)," L-PI-I 1-006, dated January 14, 201 1, ADAMS Accession Number MLI 10140367. 6. E-mail from T. Wengert (NRC) to S. Chesnutt (Xcel), "Prairie lsland - Request for Clarification of January 14, 201 1 Response Concerning Prairie lsland Leak-Before-Break LAR," January 18, 201 1. 7. E-mail from T. Wengert (NRC) to S. Chesnutt (Xcel), "Prairie lsland - Supplemental Request for Clarification of January 14, 201 1 Response Concerning Leak-Before- Break LAR," January 28, 201 1. Page 7 of 7