GNRO-2012/00117, Response to Request for Additional Information (RAI) on Severe Accident Mitigation Alternatives Dated August 23, 2012

Response to Request for Additional Information (RAI) on Severe Accident Mitigation Alternatives Dated August 23, 2012

ML12277A082
Person / Time
Site:	Grand Gulf
Issue date:	10/02/2012
From:	Mike Perito Entergy Operations
To:	Office of Nuclear Reactor Regulation, Document Control Desk
References
GNRO-2012/00117
Download: ML12277A082 (17)

Text

• ~*Entergy Entergy Operations, Inc.

P. O. Box 756 Port Gibson, MS 39150 Michael Perito Vice President, Operations Grand Gulf Nuclear Station Tel. (601) 437-6409 GNRO-2012/00117 October 2, 2012 U.S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555-0001

SUBJECT:

Response to Request for Additional Information (RAI) on Severe Accident Mitigation Alternatives dated August 23,2012 Grand Gulf Nuclear Station, Unit 1 Docket No. 50-416 License No. NPF-29

REFERENCE:

NRC Letter, "Requests for Additional Information on Severe Accident Mitigation Alternatives for the Review of the Grand Gulf Nuclear Station, License Renewal Application," dated August 23,2012 (GNRI-2012/00187) (ML12227A735)

Dear Sir or Madam:

Entergy Operations, Inc is providing, in Attachment 1, the response to the referenced Request for Additional Information (RAI). Attachment 2 contains the new commitment required in response to the letter.

If you have any questions or require additional information, please contact Christina L. Perino at 601-437-6299.

I declare under penalty of perjury that the foregoing is true and correct. Executed on the 2nd day of October, 2012.

Attachments: 1. Response to Requests for Additional Information (RAI)

2. List of Regulatory Commitments cc: (see next page)

GNRO-2012100117 Page 2 of 2 cc: with Attachments Mr. John P. Boska, Project Manager Plant Licensing Branch 1-1 Division of Operating Reactor Licensing Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Mail Stop 0-8-C2 Washington, DC 20555 cc: without Attachment Mr. Elmo E. Collins, Jr.

Regional Administratior, Region IV U. S. Nuclear Regulatory Commission 1600 East Lamar Boulevard Arlington, TX 76011-4125 U.S. Nuclear Regulatory Commission ATIN: Mr. A. Wang, NRRlDORL Mail Stop OWFN/8 G14 11555 Rockville Pike Rockville, MD 20852-2378 U. S. Nuclear Regulatory Commission ATIN: Mr. Nathaniel Ferrer NRRlDLR Mail Stop OWFNI 11 F1 11555 Rockville Pike Rockville, MD 20852-2378 NRC Senior Resident Inspector Grand Gulf Nuclear Station Port Gibson, MS 39150

Attach ment 1 to GNRO-2012/00117 Respo nse to Requests for Additi onal Inform ation to GNRO-2012/00117 Page 1 of 12

1. RA11.c The response to this request for additional information (RAI) indicates that there are no unresolved equipment reliability or plant data issues that would impact the severe accident mitigation alternatives (SAMA) analysis. Clarify what is meant by unresolved and indicate if there are any resolved issues that could impact Grand Gulf Nuclear Station (GGNS) plant specific data involving risk significant components or systems that might significantly impact the SAMA analysis results.

Response to clarification RAI 1 related to RAI 1.c:

The word "unresolved" was used in two places in the response to RAI 1.c. Clarification is provided below for both instances.

1. The maintenance rule system health reports indicate that there have been a few recent functional failures. Actions have been taken to correct these failures and preclude repeat occurrence. Therefore, the issues are considered resolved. These recent functional failures do not have an adverse impact on the SAMA analysis because either generic data was used in the failure analysis for the component or the recent failure would replace an old failure in the calculation, keeping the failure rate constant.

The maintenance rule system health reports also indicate that the unavailability of two systems (high pressure core spray (HPCS) and the B diesel-driven fire pump) has increased recently. However, the unavailability for these systems remains within the error band of the unavailability distribution. Therefore, there are no equipment reliability issues that would adversely impact the SAMA analysis.

2. Plant data issues were identified during the expert panel reviews, but were resolved in the version of the model used for the SAMA analysis. Thus, there are no equipment reliability issues that would adversely impact the SAMA analysis.

The response to RAI 1.c is clarified as follows (strikethrough indicates deleted text, underline indicates added text.)

The maintenance rule system health reports indicate no unresolved equipment reliability issues that would adversely impact the SAMA analysis. Also, no unresolved plant data issues were identified during the expert panel reviews of the model updates or during the expert panel review of the Level 2 cutsets were resolved in the model used for the SAMA analysis.

Therefore, there is reasonable assuranGe that there are no equipment reliability degradation issues since the August 2006 data update that would adversely impact the SAMA analysis.

2. RA11.e The initial response to RAI 1.e does not provide adequate information to explain the approximate 40% difference between the core damage frequency (CDF) from the Level 1 and Level 2 quantifications, specifically the requested, "Describe specific contributions to the approximate 40% difference in CDF, such as some of the non-minimal cutsets or other reasons." Relative to the three reasons given for the differences in results, elaborate on the to GNRO-2012/00117 Page 2 of 12 following:

a. While there are uncertainties in the minimal cut set upper bound technique for cutset quantification, particularly when it involves terms that are close to 1, to the best of our knowledge, this always results in an overestimate of the true result and is most significant for large early release frequency (LERF) or other Level 2 calculations that usually have a large number of events involving values close to 1. Therefore, the minimal cut set upper bound technique for cutset quantification would not appear to be a contributor to the Level 2 CDF result being less than the Level 1 CDF. Please explain.

b. While it is expected that the Level 1 sequence by sequence calculation would result in non-minimal cutsets, the third paragraph of the RAI response in correcting the footnote to Table E.1-7 states, "The total CDF from the level 1 model presented in Table E.1-7 is slightly higher than the single top solution in which non-minimal cutsets are subsumed." This indicates that the elimination of non-minimal cutsets is not a major factor in the Level 2 result being lower. Could quantification of the One-TOP Level 1 CDF model or alternatively combining all the Level 1 sequence cutsets and minimizing and then quantifying possibly justify what is termed "slightly" in the above quotation?

c. The response states "The level 2 LOSP recoveries in the cutsets are different than the level 1 recoveries, which is lowering the percent contribution of an SBO in the level 2 model." It is not clear what is meant by this. There should be a consistent evaluation of the recovery of the loss of offsite power (LOSP) in the two models. If it is meant that there is less credit for LOSP recovery in the Level 2 model because core damage has occurred, this should not impact the CDF. Please explain.

The difference in results from the two models leads to conflicting and inconsistent information in the SAMA submittal. For example, Table E.1-1 says that the CDF contribution from LOSP is 14% of the CDF but the contribution from station blackout (SBO) is 36%. Except for the relatively small contribution to LOSP from a consequential LOSP, the SBO contribution is a subset of LOSP contribution. Further, the Case 1 result, which is based on the same Level 2 model as the LOSP contribution discussed previously, indicates that SBO contributes 13.6% of the CDF. Again, the SBO contribution should be something less than the LOSP contribution.

Provide further support for the Level 2 model giving a valid result for the total CDF and explain differences from the Level 1 results using specific examples of these reasons including requantification using different techniques or assumptions. Provide assurance that the Level 2 model result is not missing important sequences and/or cutsets.

Response to clarification RAI 2 related to RAI 1.e:

At issue is the difference between the level 1 CDF of approximately 2.92E-06/rx-yr and the level 2 CDF of approximately 2.05E-06/rx-yr used as the reference value in the SAMA analysis. The total CDF from the level 2 model utilized for the SAMA analysis is 2.05E-06/rx-yr when each of the 13 release categories (HIE, H/I, H/L, M/E, Mil, MIL, UE, UI, UL, LUE, LUI, LUL, and OK) is solved. However, further investigation into this issue revealed that the OK gate in the level 2 model was not fully developed as a means to quantify the intact sequences. The steps taken to assure the technical adequacy of the level 2 model, (including the self assessment, technical acceptance review and expert panel cutset review), which were applied to the other release

Attach ment 1 to GNRO-2012/00117 Page 3 of 12 categories, were not applied to the OK gate in the fault tree. The OK gate was only used as a place-holder for the residual core damage sequences that succes sfully avoid a release larger than the design basis leakage and was not intended for quanti fication. Rather, the intact sequences must be quantified by subtracting the sum of the other 12 release categories from the total level 1 CDF. However, these facts were not clearly docum ented in the level 2 analysis, nor were they identified during the technical transfer meeting.

Thus, the SAMA analysis incorrectly used the 2.05E-06/rx-yr reference value obtained via quantification of all 13 release categories in the level 2 model. A condition report has been initiate d within the corrective action process to docum ent this error and ensure the OK gate in the level 2 model is not quantified in future applications of the level 2 model.

To correct the analysis, the SAMAs are being re-analyzed using the level 1 CDF as the reference value. Also, the impact of each SAMA on the level 1 CDF is being used to quantify the impact on the intact release category. This reanalysis could not be completed in the required response time of 45 days. Thus, the revised results will be provided in a follow-up letter by Novem ber 17, 2012. When the results of the reanalysis are provided, updated versions of the ER tables and previous RAJ responses that are impacted by the reanalysis will also be provided.

3. RA11. f The discussion of the disposition of Observation 85 should be clarified by providing specific information on the event tree sequences, the meaning of "provid e insight into the Level 2 PRA core-damage binning process," and how these sequences are handled in this process.

The conclusion of each observation discussed is that the dispos ition remains applicable to the PRA used for the SAMA analysis. The assessment of timing of various events factors into many of the discussions (for example, Observations 87, 89, and 97).

Clarify if these assessments were made for the extended power uprate operation. If not, review the assessments and address the power increase effect on the conclusions reached.

Response to clarification RAJ 3 related to RAI 1.f:

Observation 85 and its disposition, as supplied in the RAI 1.f respon se, are repeated here for convenience.

Observation 85 (Element AS-g)

"There is a slight problem in the chronology of the event tree headin gs associated with containment heat remov al or venting and low pressure injection.

The general transient tree shows the sequence for RCIC operating to ask RCIC, RHR, Venting, and then HPCS and Low pressure injection. The result is that sequences are assign ed to vented containment even if low pressure systems fail. This would certain ly not be the case when EOPs are followed. For TQUV sequences, core damage would occur in the sequence at the time of depressurization on HCTL, i.e., - 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />.

Venting would not genera lly occur until 22.5 psig in containment, i.e., about 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.

Therefore, there are "vented" core damage end states in the PRA that are improp erly labeled.

Examples include: T-16, T-20, and T-21."

to GNRO-2012/00117 Page 4 of 12 Observation 85 Disposition "The transient event tree shows several non-minimal sequences with respect to core damage that are not quantified as follows: T-16 is subsumed by T-13, T-24 is subsumed by T-20, and T- 25 is subsumed by T-21. The end states for T-16, T-24, and T-25 are correct (vented containment) and provide insight into the Level 2 PRA core-damage binning process. Therefore, no changes are needed.

Also, the HCTL limit would not occur at 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />, but would occur at a later time (approximately 7 to 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />) with the current EPs. Based on MAAP evaluations, 22.5 psig in the containment would not occur until approximately 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />. However, the operators would likely initiate venting before that time based on whether they believe that containment pressure can be maintained below 22.4 psig. "

Observation 85 was made during the peer review, which reviewed an earlier revision of the general transient event tree than the one used in the current model. The disposition applies to the earlier version of the event tree and was attempting to explain the timing associated with the event tree. However, it appears that both observation 85 and the disposition have misstated some of the sequence numbers. The observation states that T-16, T-20 and T-21 were "vented" sequences and the disposition states that T-16, T-24 and T-25 were "vented" sequences. However, as can be seen below, T-13, T-20 and T-21 were "vented" sequences and T-16, T-24 and T-25 were sequences in which venting failed.

The specific event tree sequences mentioned in the observation and disposition were defined as follows.

T-13 Transient, success of on-site power, failure of power conversion system, failure of high pressure core spray, success of RCIC, failure of the SPC mode of RHR, success of manual depressurization, failure of containment cooling, success of containment venting, failure of high pressure core cooling, and failure of low pressure core cooling T-16 Transient, success of on-site power, failure of power conversion system, failure of high pressure core spray, success of RCIC, failure of the SPC mode of RHR, success of manual depressurization, failure of containment cooling, failure of containment venting, failure of high pressure core cooling. and failure of low pressure core cooling T-20 Transient, success of on-site power. failure of power conversion system. failure of high pressure core spray. success of RCIC. failure of the SPC mode of RHR, failure of manual depressurization. failure of containment cooling, success of containment venting, failure of high pressure core cooling, successful depressurization, and failure of low pressure core cooling T-21 Transient. success of on-site power, failure of power conversion system. failure of high pressure core spray. success of RCIC. failure of the SPC mode of RHR, failure of manual depressurization, failure of containment cooling, success of containment venting. failure of high pressure core cooling, and failure of depressurization to GNRO-2012/00117 Page 5 of 12 T-24 Transient, success of on-site power, failure of power conversion system, failure of high pressure core spray, success of RCIC, failure of the SPC mode of RHR, failure of manual depressurization, failure of containment cooling, failure of containment venting, failure of high pressure core cooling, successful depressurization, and failure of low pressure core cooling T-25 Transient, success of on-site power, failure of power conversion system, failure of high pressure core spray, success of RCIC, failure of the SPC mode of RHR, failure of manual depressurization, failure of containment cooling, failure of containment venting, failure of high pressure core cooling, and failure of depressurization The words, "provide insight into the Level 2 PRA core-damage binning process," in the disposition mean that the containment status (vented in T-13, T-20 and T-21; failed in T-16, T-24 and T-25) is provided in each Level 1 sequence description for use in determining the applicable containment event tree.

Although the disposition of Observation 85 indicated that no changes were needed, further evaluation did result in modification of the general transient event tree to address this observation. The containment venting question was moved after those for high pressure core cooling and low pressure core cooling as recommended in the observation. In addition, questions about low pressure injection via the standby service water crosstie and via firewater were broken out as separate questions in the revised event tree. This revised event tree structure has been retained in the Rev. 3 extended power uprate (EPU) model general transient event tree. Thus, the observation has been resolved in the model used for the SAMA analysis. Since the observation has been resolved and the specific sequences listed above no longer exist in the general transient event tree, these sequences do not have to be handled in the Level 2 core damage binning process.

The dispositions provided in the RAI response for the peer review observations are those for the model that was peer reviewed, not for the Rev. 3 EPU model. However, a statement was provided confirming that these observations do not need to be addressed in the model used for the SAMA analysis. Additional discussion of the power increase effect on the timing-related conclusions is provided below.

Observation 87 (Element AS-18)

"Sequence #37 - The fire protection injection is used for successful RPV makeup following an SSO without HPCS but successful RCIC and SORV. This is judged by the Certification Team to result in loss of injection in approximately 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />. This means the 8 fire protection hoses must be aligned by this time under SSO conditions. No thermal hydraulic basis could be found to assess the HEP allowed time for the fire water cross tie for this sequence. It is judged that this operator action for this sequence has a much higher HEP than 0.013."

Observation 87 Disposition for Rev. 3 EPU Model In the Rev. 3 and Rev. 3 EPU models, this operator action, P64-FO-HE-G (long term),

"Operator Fails to Align Firewater System for Injection", was reassessed to have an HEP of 0.011. The allowable time for this action in the Rev. 3 model (6 hrs) and Rev. 3 EPU model (5.7 hrs) is based on the time to core damage for an SSO with RCIC operation (time to battery depletion or suppression pool heatup).

to GNRO-2012/00117 Page 6 of 12 In the Rev. 3 EPU model, the available action time decreased due to the power increase, but this did not change the HEP value.

Thus, the conclusion that changes are not necessary due to this observation remains applicable to the probabilistic risk assessment (PRA) used for the SAMA analysis.

Observation 89 (Element AS-18)

"Cutset 11 - LOOP initiator with ECCS suction strainer clogging and fail to recover offsite power within 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />. This must assume RCIC operates for 8 - 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> and HPCS cannot be used from the CST. Are both of these assumptions realistic?"

Observation 89 Disposition for Rev. 3 EPU Model The observation was a misinterpretation of a particular cutset from the peer review version of the model and is not applicable to the Rev. 3 EPU model. The disposition stated that the purported assumptions were not made in the peer reviewed model. The Rev. 3 EPU model also does not make the purported assumptions. In the Rev. 3 EPU model, RCIC is assumed to run for less than 6 hrs prior to failure due to battery depletion or suppression pool heatup. Also, the model assumes that HPCS and RCIC are initially lined up to the CST and will switch to the suppression pool after the CST is exhausted. The offsite power recovery terms in the Rev. 3 EPU model have been appropriately reduced to reflect the shorter time periods available due to the increased power level. Thus, the conclusion that changes are not necessary due to this observation remains applicable to the PRA used for the SAMA analysis.

Observation 97 (Element TH-8)

"MAAP indicates that HCTL at 1000 psig (-120°F) is reached in 2 to 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />. 185°F is reached in 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />. Both of these are reached before the 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> assumed battery life is exhausted in the SBO analysis. The requirement to maintain RCIC for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> would appear to be compromised by this assumption. The ability to reach 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> does not appear to be justified."

Observation 97 Disposition for Rev. 3 EPU Model In the Rev. 3 EPU model, RCIC is assumed to run for less than 6 hrs prior to failure due to battery depletion or suppression pool heatup. This is consistent with the MAAP runs performed for the increased power level. Therefore, the conclusion that changes are not necessary due to this observation remains applicable to the PRA used for the SAMA analysis.

4. RAls 2.c and 2.d Although the response to RAI 2.c describes the representative sequence for each release category, it does not address the justification for choosing the sequence with the highest frequency versus the sequence with a higher source term and a lower but still important frequency. Unless the highest source term for important contributors to a release category is used for the base case and SAMA specific analysis, it is possible for the SAMA benefit to be underestimated. This could occur if a particular SAMA primarily affects the frequency of a to GNRO-2012/00117 Page 7 of 12 sequence with a higher source term and lower frequency.

The response to RAI 2.d discusses a sensitivity study for the HighlEarly (HIE) Release Category (RC) using a lower frequency but higher source term alternate.

a. Was the highest source term used in this study for the important contributors to this RC? If not, justify the source term used for the sensitivity study.

b. Provide the results of this sensitivity study to support the statement that there is no change in the cost beneficial status of SAMAs. Include the MACCS2 results or Level 3 information similar to that provided in Table E.1-13 for the new HIE RC, the maximum averted dollar risk results for the revised base case as well as the results of the cost benefit analysis for each SAMA using the revised RC risk results.

c. Identify other RCs where the source term for the representative sequence is less than that for another important sequence and justify that the use of the selected source term does not underestimate SAMA benefits.

d. The third paragraph of the response states, "This increase to a high release was due to the failure of the drywell which had not been preViously accounted for in the nodal analysis." Explain the statement that "drywell failure was not accounted for."

Response to clarification RAI 4 related to RAls 2.c and 2.d:

a) In the sensitivity analysis described in response to RAI 2.d, the highest Csi (Cesium Iodide) source term for the important contributors to the HIE release category was used.

b) The source term from Modular Accident Analysis Program (MAAP) case GG10500, which was used in the sensitivity analysis described in response to RAI 2.d, is being used in the reanalysis of the SAMAs described in response to clarification RAI 2. As indicated in the response to clarification RAI 2, when the results of the reanalysis are provided, updated versions of the Environmental Report (ER) tables and previous RAI responses that are impacted by the reanalysis will also be provided.

c) For all other release categories, the highest Csi source term for the important contributors to the release category was used in the original analysis. These source terms remain unchanged in the reanalysis described in response to clarification RAI 2. Thus, the SAMA benefits are not underestimated.

d) The statement that drywell failure had not been "previously accounted for in the nodal analysis" means that containment event tree node Cl, drywell intact, is successful for the scenario in MAAP case GG10503 because it does not include an early, energetic failure of the drywell. This would generally result in a release of medium magnitude. However, under the conservative binning rules, although the scenario does not include an early, energetic failure of the drywell, it was classified as a "high" release because it does include failure to isolate containment (normally open 10" vent path not isolated) and failure of containment sprays.

to GNRO-2012/00117 Page 8 of 12

5. RAI2.g The RAI response indicates that the MAAP case chosen to represent no containment failure (NCF) is MAAP run GG10502D and is an intact accident scenario with radionuclide releases consistent with design leakage rates. The response to RAI 2.c indicates that this MAAP case was chosen for RC Low-Low/Early (LUE), which is intended to represent containment failure end points. Describe this case and how its use for both release categories would affect the SAMA analysis results.

Response to clarification RAI 5 related to RAI 2.9:

MAAP case GG1 0502D is a loss of makeup at high pressure with containment isolation successful, low pressure injection available, no safety relief valves, no suppression pool cooling, with containment sprays and residual heat removal heat exchangers available, no igniters, and with upper pool dump. Release of drywell gases to the suppression pool serves to scrub the radioactive constituents from the gases resulting in lower fission product release.

The release fraction of Csi from MAAP run GG10502D is 0.001% which is appropriate for the radioactive release severity of both Low-Low/Early (LUE) and NCF release categories. MAAP run GG 10502D has the highest source term that is representative of the LUE release scenarios.

Also, as indicated in the response to RAI 2.g, including the NCF release category using this case has a negligible impact on the SAMA analysis.

The base Level 3 analysis was conducted without the NCF release category based on the assumption that this release category was negligible to the SAMA analysis. The sensitivity analysis described in the response to RAI 2.g was conducted and confirmed this assumption.

Nevertheless, the NCF release category, using the source term from MAAP case GG10502D, as described in response to RAI 2.g, is being used in the reanalysis of the SAMAs described in response to clarification RAI 2. As indicated in the response to clarification RAI 2, when the results of the reanalysis are provided, updated versions of the ER tables and previous RAI responses that are impacted by the reanalysis will also be provided.

6. RA13.d While the response to this RAI indicates that changes to the site since the individual plant examination of external event (IPEEE) impacts the site drainage characteristics and thus the IPEEE recommendations are no longer valid, it is not clear if the specific recommendations are impacted or not. Specifically address the current applicability of each of the five IPEEE recommendations. Note that the last recommendation addresses the adequacy of a flood barrier for the Standby Service Water A equipment hatch.

Response to clarification RAI 6 related to RAJ 3.d:

1. Remove the wooden foot bridge crossing the northwest ditch near its upstream end.

The wooden foot bridge crossing the northwest ditch near its upstream end has been removed.

to GNRO-2012/00117 Page 9 of 12

2. Remove the 15" corrugated metal pipe located in the small auxiliary ditch parallel to the northwest ditch (at the same approximate location as the duct bank crossing the northwest ditch). Regrade the area to provide a gradual transition between the yard upstream, and the auxiliary ditch.

The 15" corrugated metal pipe was not removed. However, the parking lot to the south and west of the auxiliary ditch has been graded and resurfaced. Also, the northwest ditch flow considers the presence of the auxiliary ditch.

3. Re-hang the security fence gates west of the control building to ensure that approximately 5" of gap exist between the gate and the road.

The gap between the gate and road surface is less than 5 inches for one of the gates and approximately 5 inches on the other gate. The gate is not expected to have a significant impact on run-off flow.

4. Grade down and remove the access road, the raised berm parallel to the access road, and the curbs adjacent to the access road where they cross Culvert No.1, such that elevations above the culvert do not exceed 132.7 ft above mean sea level (MSL).

The raised berm has been removed. The curb around the guard building is still in place, but the elevation directly above the culvert is approximately 132.1 feet MSL. Drainage from the south access road and the heavy haul road will flow into the 100 year ditch or into the ditch north of the south guard building. The ditch to the north of the guard building has a culvert (corrugated metal pipe) that drains directly into the 100-year ditch.

Culvert No 1 is inspected annually to assure the culvert and ditch are adequate to maintain drainage away from the plant. Also, severe weather procedures direct the performance of inspections in anticipation of heavy precipitation.

5. Replace the C8x11.5 channel forming the flood barrier across the SSW A equipment hatch opening with one having a minimum depth of approximately 13".

The C8 has not been replaced with another member. However, for the design basis event the height of the C8 is adequate. In the event that there is not adequate freeboard, internal barriers inside the pump houses will protect the equipment from any infiltration. Additionally, the floors are sloped toward the doors and there is an extremely low potential that debris will block the drains. Contingency actions are available to place sand bags in front of the doors.

7. RA15.a Review of the Phase I SAMA screening raises the following questions.

a. SAMA 9, reduce DC dependence between high-pressure injection and Automatic Depressurization System (ADS), is said to be addressed by SAMAs 27 and 28. These SAMAs make use of portable generator to supply DC power to buses or panels.

Consider a SAMA that would provide a charging system (without a new generator) and battery that would make the high-pressure core spray (HPCS) independent of the other DC buses.

to GNRO-2012/00117 Page 10 of 12

b. SAMA 36, enhance DC power availability by providing a direct connection from the diesel generator, the security diesel, or another source to the 250 V battery chargers or other required loads, is said to be addressed by SAMA 27. This SAMA makes use of a portable generator. Consider a SAMA that would provide the necessary connections but without the expense of a new portable generator, or explain why this is not feasible.

c. SAMA 42, install key-locked control switches to enable AC bus cross-ties and modify procedures to enhance the reliability of the AC power system, cites SAMA 12 as being similar. SAMA 12 addresses AC bus cross-ties but does not specifically address installing key-locked control switches or enhancing procedures. Consider these improvements to the current GGNS situation, or explain why this is not feasible.

d. SAMA 74, provide capability for alternate injection via the reactor water cleanup (RWCU), is dispositioned as already installed on the basis or procedures that direct use of the RWCU for alternate shutdown cooling. The purpose of this SAMA is improved injection capability not heat removal. Consider the use of the RWCU system for injection, or explain why this is not feasible.

e. SAMA 144, modify containment flooding procedure to restrict flooding to below the top of the active fuel, is dispositioned as already installed based on the Boiling Water Reactor Owners Group (BWROG) guidelines that directs flooding to above the top of the active fuel. Depending on the physical configuration, pressurization of the drywell as a result of flooding may require drywell venting. The stated purpose of this SAMA is to reduce the drywell pressurization and prevent the resulting venting from happening if the coolant level is restricted to below the top of the active fuel (but still adequate to cool the core debris). Evaluate if this is possible for the GGNS arrangement and if so consider such a SAMA.

f. SAMA 160, institute simulator training for severe accidents, is dispositioned as already installed with the statement that the technical support center and control room would be manned in a severe accident evolution to provide additional support by personnel familiar with SAGs. If the GGNS simulator does not include severe accident scenarios, provide a cost benefit analysis for this SAMA.

Response to clarification RAI 7 related to RAI 5.a:

a) The division 3 direct current (DC) power system already contains a charger and battery which provide the HPCS with control and motive power independent of the other DC buses. Thus, the proposed SAMA is already installed.

b) The DC power system provides control and switching power for alternating current (AC)-

powered components and also provides a source of power for emergency components in station-blackout (SBO) situations. In an SBO, all off-site and normal on-site AC power has been lost so the emergency diesel generators are not available. Since GGNS does not have a security diesel generator or another source of power, a portable generator is the only remaining option to enhance DC power availability in SBO situations.

Consideration of a SAMA to provide a direct connection from an emergency diesel generator to the chargers or other required loads in a non-SBO situation is impacted by the reanalysis being provided in response to clarification RAI 2 and, therefore, will be to GNRO-2012/00117 Page 11 of 12 discussed in that response.

c) Phase I SAMA 12 (from NEI 05-01) and Phase I SAMA 42 (from Pilgrim) both recommend improving the 4.16 kilovolt (kV) bus cross-tie ability. As noted in the question, Phase I SAMA 12 is more general, stating simply to improve the cross-tie ability, while Phase I SAMA 42 specifically mentions installing key-locked control switches and enhancing procedures.

Both of these Phase I SAMAs have been evaluated in Phase II SAMA 6. Each of the Engineered Safety Features (ESF) buses already has the capability of sharing the same ESF transformer feed. The division 3 DG13 may be cross-tied to either division 1 bus 15M or division 2 bus 16AB during an SBO. These instructions are contained in the Loss of AC Power Off-Normal Event Procedure and include several actions to manipulate breakers and to lift leads to bypass interlocks. Phase II SAMA 6 evaluates improving the ability to accomplish these cross-tie actions.

The implementation cost for Phase II SAMA 6 reflects the physical and procedural changes necessary to facilitate these actions, including installation of key-locked control switches to preclude the steps to lift leads. Thus, Phase II SAMA 6 encompasses the recommendations in Phase I SAMA 12 and Phase I SAMA 42.

d) Consideration of a SAMA to use the RWCU system for alternate injection is impacted by the reanalysis being provided in response to clarification RAI 2 and, therefore, will be discussed in that response.

e) As indicated in the question, the coolant level must be maintained at a level adequate to cool the core debris. Thus, coolant level should not be restricted to below the top of the active fuel unless the operators have indication that the fuel has slumped to a lower level or breached the RPV.

If core debris has breached the RPV, the RPV and Primary Containment Flooding severe accident procedure directs containment flooding to the minimum debris submergence level with containment venting only as necessary to maintain that level. If the core debris has not melted through the RPV, the RPV and Primary Containment Flooding severe accident procedure directs RPV and containment flooding to various levels depending on physical conditions. The levels include top of active fuel, bottom of active fuel, and lower levels defined by the minimum debris retention injection rate.

Containment venting occurs only as necessary to maintain the target water level. Thus, venting is minimized, and Phase I SAMA 144 is already installed.

f) Severe accident scenarios are included in simulator training of operators and emergency response personnel. Thus, Phase I SAMA 160 is already installed.

to GNRO-2012/00117 Page 12 of 12

8. RA15.c Describe whether it is feasible to manually open the HPCS minimum flow line isolation valve (1 E22F012-C) in time to prevent HPCS failure. If the manual actions are determined feasible, consider the cost benefit of such a procedure.

Response to clarification RAI 8 related to RAI 5.c:

The proposed action to manually open the HPCS minimum flow line isolation valve is not feasible because it cannot be completed in time to prevent HPCS pump failure. Therefore, the recommended SAMA is not cost-beneficial.

Attachment 2 to GNRO*2012/00117 List of Regulatory Commitments to GNRO-2012/00117 Page 1 of 1 This table identifies actions discussed in this letter for which Entergy commits to perform. Any other actions discussed in this submittal are described for the NRC's information and are not commitments.

TYPE (Check one) SCHEDULED ONE- COMPLETION CONTINUING COMMITMENT TIME DATE COMPLIANCE ACTION (If Reauired)

From clarification RAI2 related to_RAI 1.e in this letter:

To correct the analysis, the SAMAs are being re-analyzed using the level 1 CDF as the reference value. Also, the impact of each SAMA on the level 1 CDF is being used to quantify the impact on the intact release category. This reanalysis could not November 17, be completed in the required response time -J 2012 of 45 days. Thus, the revised results will be provided in a follow-up letter by November 17, 2012. When the results of the reanalysis are provided, updated versions of the ER tables and previous RAI responses that are impacted by the reanalysis will also be provided.