IR 05000282/2011010: Difference between revisions

From kanterella
Jump to navigation Jump to search
(Created page by program invented by StriderTol)
(Created page by program invented by StriderTol)
Line 1: Line 1:
{{Adams
{{Adams
| number = ML111610249
| number = ML11195A336
| issue date = 06/09/2011
| issue date = 07/13/2011
| title = IR 05000282-11-010 & 05000306-11-010; on 05/13/21011 - 05/20/2011; Prairie Island Nuclear Generating Plant, Units 1 and 2, Other Activities
| title = Prairie Island. Units 1 & 2, Supplemental Information Regarding Inspection Report 05000282/2011010; 05000306/2011010 (EA-11-110)
| author name =  
| author name = Schimmel M A
| author affiliation = NRC/RGN-III/DRP
| author affiliation = Northern States Power Co, Xcel Energy
| addressee name = Schimmel M A
| addressee name =  
| addressee affiliation = Northern States Power Co
| addressee affiliation = NRC/Document Control Desk, NRC/NRR
| docket = 05000282, 05000306, 07200010
| docket = 05000282, 05000306
| license number = DPR-042, DPR-060, SNM-2506
| license number = DPR-042, DPR-060
| contact person =  
| contact person =  
| case reference number = EA-11-110
| case reference number = EA-11-110, ir-11-010, L-PI-11-071
| document report number = IR-11-010
| document type = Letter, Licensee Response to Notice of Violation
| document type = Inspection Report, Letter
| page count = 43
| page count = 20
}}
}}


Line 19: Line 18:


=Text=
=Text=
{{#Wiki_filter:UNITED STATES NUCLEAR REGULATORY COMMISSION REGION III 2443 WARRENVILLE ROAD, SUITE 210 LISLE, IL 60532
{{#Wiki_filter:XcelEnergy" JUL 1 3 2011 L-PI-1 1-071 U S Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555-0001 Prairie Island Nuclear Generating Plant Units 1 and 2 Dockets 50-282 and 50-306 Renewed License Nos. DPR-42 and DPR-60 Supplemental Information Reqarding NRC Inspection Report 05000282/2011010:
-4352 June 9, 2011 E A-11-110 Mr. Mark Site Vice President Prairie Island Nuclear Generating Plant Northern States Power Company, Minnesota 1717 Wakonade Drive East Welch, MN 55089
05000306/2011010 (EA-11-110)
By letter dated June 9, 2011 (Agencywide Documents Access and Management System (ADAMS) Accession Number ML111610249), the NRC transmitted Inspection Report 05000282/2011010; 05000306/2011010 for the Prairie Island Nuclear Generating Plant (PINGP). In the inspection report, the NRC identified one finding and apparent violation related to battery charger performance with a preliminary significance of White for Unit 1 and a preliminary significance of Green for Unit 2, and provided the opportunity to request a Regulatory Conferenc By letter dated June 14, 2011 (ADAMS Accession Number ML111662049), Northern States Power Company, a Minnesota corporation, doing business as Xcel Energy (hereafter "NSPM") requested a Regulatory Conference which will be held on July 28, 2011.This letter transmits documents which support the July 28, 2011 Regulatory Conferenc Enclosure 1 provides document V.SPA. 11.016, "Simulator Runs for Battery Charger Significance Determination Process".


SUBJECT: PRAIRIE ISLAND NUCLEAR GENERATING PLANT, UNITS 1 AND 2
Enclosure 2 provides a document entitled, "Battery Depletion Calculation Load Profile".
, NRC INSPECTION REPOR T 05000 282/2011010; 050003 06/2 011010 PRELIMINARY WHITE FINDING


==Dear Mr. Schimmel:==
Enclosure 3 provides document V.SPA. 11.006, "Evaluation of Battery Depletion Differences".
On May 20, 2011, the U.S. Nuclear Regulatory Commission (NRC) completed a n inspection at your Prairie Island Nuclear Generating Plant, Units 1 and 2. The enclosed report documents the results of this inspection, which were discussed on May 20, 2011, with you and other members of your staff.


The inspection examined activities conducted under your license as they relate to safety and compliance with the Commission's rules and regulations and with the conditions of your license. The inspectors reviewed selected procedures and records, observed activities, and interviewed personnel.
If there are any questions or if additional information is needed, please contact Mr. Dale Vincent, P.E., at 651-388-1121.


This report documents a finding for Unit 1 that has preliminarily been determined to be White or a finding with low-to-moderate increased safety significance. In addition, this same finding was preliminarily determined to b e Green , a finding of very low safety significance
* 1717 Wakonade Drive East * Welch, Minnesota 55089-9642 Telephone:
, for Unit 2.
651.388.1121 Document Control Desk Page 2 Summary of Commitments This letter contains no new commitments and no revisions to existing commitments.


As documented in Section 4OA5 of this report both trains of safety-related battery chargers were not capable of performing their safety function from initial installation in 1994 due to being susceptible to failure during certain design basis events.
Mark A. Schimmel Site Vice President, Prairie Island Nuclear Generating Plant Northern States Power Company -Minnesota Enclosures (3)cc: Administrator, Region III, USNRC Project Manager, PINGP, USNRC Resident Inspector, PINGP, USNRC ENCLOSURE1 V.SPA.11.016 SIMULATOR RUNS FOR BATTERY CHARGER SIGNIFICANCE DETERMINATION PROCESS 15 pages follow C Xcel Energy Prairie Island Nuclear Generating Plant Probabilistic Risk Assessment V.SPA.1 1.016 Simulator Runs for Battery Charger Significance Determination Process March 2011 Document Review Form Document or Analysis Title: V.SPA.11.016-
Simulator Runs for Battery Charger SDP Revision of Document or 0 Analysis: Date: 3/11/2011 Preparer:
,,,t- Date: 3/11/2011 Mark A. Brossart ,1 Reviewer: Jayne E. Ritter Date: 3/11/2011 PRA Acceptance Date: 6/07/2011 Ashley Peterman V.SPA.11.016 Simulator Runs for Battery Charger Significance Determination Process Table of Contents Page Number T able of C ontents .............................................................................
1.0 P u rp o se ....................................................................................
2.0 Simulator Development Process ........................................................
3.0 S im ulator R uns ............................................................................
3.1 Sim ulator "Initial R un" R esults .....................................................
3.2 Simulator "Follow-up Run" Results ..............................................
4 .0 C o nclu sio n s ................................................................................
5 .0 R e fere nce s ..................................................................................
1 2 2 3 3.5 7 8


This finding was assessed based on the best available information, including influential assumptions, using the applicable Significance Determination Process (SDP).
===Attachments:===
Attachment 1 -Simulator Prior Knowledge Statement
..................................
(2 pages)Attachment 2 -Simulator Scenario Security Agreement
...............................
(2 pages)File Prepared By: Mark Brossart Date: 3/11/2011 File Reviewed By: Jayne Ritter Date: 3/11/2011 V.SPA.1 1.016 Simulator Runs for Battery Charger Significance Determination Process Page 1 of 8 V.SPA.11.016 Simulator Runs for Battery Charger Significance Determination Process 1.0 Purpose: A simulator scenario was developed to support the arguments used in the Significance Determination Process for the Battery Charger past operabilit The results of the scenario will not be used as part of the PRA success strategy but may be used to demonstrate, in a simulated environment, that there is adequate time with significant margin to perform the recovery actions credited in the PRA success strateg It is intended that site management may use this information in discussing the site response with the Regulator as a demonstration that the PRA recovery strategy assumptions are reasonable.


Upon identification of this issue and after interaction with the NRC, you concluded that a design ated operator position needed to be established to ensure that a specific individual could perform actions to recover the battery charger(s) prior to the safety
The simulator run will determine the time it takes operators to: 1. Diagnose and respond to re-energizing a safeguards 4.16KV bus via the opposite unit bustie breaker during a LOOP event.2. Perform the actions necessary to restart the battery charger following a lock up.Information pertaining to this calculation folder can be found in the following directory:
-related batteries being depleted. Lastly, during your past operability review you concluded that there was reasonable doubt that the battery chargers would have performed their safety function if called upon prior to October 22, 2010
J:\PRA\Calc Folder Writeups\SPA Folders\ V.SPA.11.016
, (the date the designated operator position was established).
-Simulator Runs for Battery Charger SDP 2.0 Simulator Development Process: 1) The simulator scenario is based on the limiting accident sequence developed for the Battery Charger SDP (Reference 5.3). The scenario includes the following conditions:
a. Dual unit LOOP (Initiating Event).b. The 12 Battery Charger experiences a lock up during the load sequencing restoration process.c. The Emergency Diesel generator (EDG) Dl experiences a lockout. As a result of the LOOP and D1 failure, Bus 15 will not be energize This will cause the emergency lighting to transfer from its normal AC source to the 12 Battery (DC source) resulting in a more rapid battery depletion and minimizing the allowed operator response time. (Reference 5.1)d. The inverters will fail to the DC source as a result of a known issue with voltage overshoot associated with the D1ID2 voltage regulator This will result in more rapid battery depletion and minimizing the allowed operator response time.e. The Non-License Operator (NLO) used to restore the battery charger is physically located in a location that was expected for the scenario (Screenhouse for DDCLP monitoring)
but trying to maximum travel time to the battery rooms.2) Sequence of Events: Dual Unit LOOP Both units experience a simultaneous LOOP. D1 is locked out causing Bus 15 to not be immediately available to power safeguards loads. As a result of the 12 Battery Charger's susceptibility to lock up, the DC loads for B Train will be powered from the 12 Battery as 12 Battery Charger locks up during the load sequence restoration process. See PRA Calculation Folder V.SPA. 11.003 for details on the impact on the 12 Battery depletion time. In order to maximize the time for the NLO to respond to the battery room, the NLO was sent to monitor the Diesel Driven Cooling Water Pumps (DDCLPs) in the Plant Screenhouse.


Because of the compensatory actions taken, no current safety concern exists.
V.SPA.1 1.016 Page 2 of 8 Simulator Runs for Battery Charger Significance Determination Process 3) Expected Crew Responses:
a. Enter IE-0 and initiate restoration of Bus 15 at Step 3 per 1C20.5 AOPI (Reference 5.4).b. Perform actions necessary to restore the 12 Battery Charger per the Alarm Response Procedure ARP C47024-1105 (Reference 5.5) and 1C20.9 AOP4 (Reference 5.6).4) Positioning of NLO: The decision was made to pre-stage the NLO initially in the Records Room and then upon the announcement of Reactor Trip dispatch him to monitor the Diesel Driven Cooling Water Pumps.Alternate locations were evaluated but discounted for the following reasons. With a Loss of Offsite Power event the assumption was that the duty crew would dispatch operators to monitor all running diesels. If the NLO was doing Attachment J of IESO. I (Reference 5.7), he would have been in closer proximity to make copies of procedures and go to the battery rooms. The D I or D2 rooms were considered but it was determined it would take less time to get to the battery rooms from the Dl or D2 rooms considering that the operator would need to make a copy of the procedure and would use the water treatment office to do so. Remote locations in the auxiliary building were not considered since the Turbine Building Operators would not have been utilized or sent to the auxiliary building tasks during this scenario.5) Two simulator runs were performed:
The initial run was performed on March 5 th, 2011 and a follow-up run on March 8 h, 2011. The purpose of the initial run was to demonstrate that the assumptions made in the PRA success strategy are reasonable with normal Control Room staffing including a Shift Manager (SM), Shift Supervisor (SS) and Shift Technical Advisor (STA). Also included were two Unit 1 Reactor Operators (RO), 1 Unit 2 RO and a Unit 1 Turbine Building NLO. Following the initial run for this scenario, it was determined that a more limiting scenario would be to assume that Unit 2 was in a sufficiently complex recovery that the extra Unit 2 RO would not be available to support the Unit 1 response.Consequently, the follow-up run was performed with a crew of a Shift Manager, Shift Supervisor, two Unit I ROs and a Unit 1 Turbine Building NLO 6) Simulator Run Data Collection:
Simulator timing data was collected for both simulator runs. There were three individuals collecting timing data (PRA Engineer in booth, Operations SM in booth, Engineering Supervisor in the plant).3.0 Simulator Runs 3.1 Simulator "Initial Run" Results (March 5th% 2011)The initial simulator run demonstrated the following risk-significant operator actions.a. The restoration of Bus 15 from Bus 25 via bustie breaker per 1C20.5 AOP4 (Reference 5.8)occurred at approximately 24 minutes into the event.b. The restoration of the battery charger per 1C20.9 AOP4 (Reference 5.6) occurred at approximately 43 minutes into the event.These actions were demonstrated in an initial simulator run performed on March 5, 2011. The crew consisted of Darrel Lapcinski (SM), Mark Davis (SS), Jim Kapsh (U1 RO), Mark Jenkin (STA), Mike Murphy (U1 Lead RO), Evan Bystrom (UI NLO) and Joe Julian (U2 Lead RO).See Table 1 for a detailed event description and tining for this simulator preliminary run.V.SPA. 11.016 Page 3 of 8 Simulator Runs for Battery Charger Significance Determination Process ThhIA 1-~imsIIAtnr fuRl unit LOOP Fv~nt ~wlth EDfl Dl Lnnk-niit~
~r~An,~arin
~/fl~/~flhl Time from Time Start of Event Event/Action q 8:50:15 0:00:00 Trigger 1 -LOOP event (Dual-unit)
with EDG D1 Lock-out 8:50:15 0:00:00 Enter 1E-0 8:50:45 0:00:30 Unit 1 RX Trip Announced to Plant Immediate Actions per 1 E-0 completed, Bus 15 is de-energized and 8:51:15 0:01:00 was reported to the Unit 1 SS.8:52:10 0:01:55 13 Charging Pump started Unit 1 SS instructed Unit 1 Lead RO to enter 1C20.5 AOP1 per 1 E-0, 8:53:10 0:02:55 Step 3 RNO 8:53:32 0:03:17 Verified that SI is not required Unit 2 Lead RO instructed to perform actions in 1C20.5 AOP1 per 8:54:00 0:03:45 Unit 1 SS/Lead RO Discussion of RCS condition between Unit 1 SS and RO (potential de-pressurization of RCS and LOCA conditions, watch for SI 8:55:00 0:04:45 actuation conditions)
8:55:45 0:05:30 Transition to 1 ES-0.1 8:58:30 0:08:15 Dispatch outplant operator to check EDG D1 and D2, locally Crew Brief, SS restored normal alarm response protoco Then directed the ROs to perform a board walkdown and report any alarms 8:58:30 0:08:15 that are unexpected for the LOOP/reactor trip.8:59:52 0:09:37 Sump C and RHR pit high level alarms were identified as abnormal.RCDT indicated zero. This was addressed as being caused by the 9:00:51 0:10:36 loss of Bus 15.Unit 1 Lead performed board walkdown and reported that Bus 11, 12, 9:01:00 0:10:45 13 and 14 are de-energized Per Step 2.4.2.B of 1C20.5 AOP1, report from outplant operator that the Bus 15 Undervoltage signal was lit and Bus 15 was not locked 9:01:45 0:11:30 out.A discussion about the 12 battery charger being de-energized with 9:02:30 0:12:15 bus 16 available and at power was conducted.


This finding is also an apparent violation of NRC requirements and is being considered for escalated enforcement action in accordance with the NRC Enforcement Policy. The current Enforcement Policy can be found at the NRC's Web site at http://www.nrc.gov/reading
9:03:00 0:12:45 Started 122 CR Chiller and fans Instructed Turbine building Operator to investigate 12 battery Trouble 9:03:50 0:13:35 alarm.SM announced that Alert was declare Escalation criteria would 9:03:55 0:13:40 occur if Bus 16 was lost.9:06:30 0:16:15 SM makes plant announcement that an Alert was declared Dispatch outplant operator to investigate why 12 Battery is Dischargin Outplant operator was located in the 12 DDCLP room 9:08:00 0:17:45 (Screenhouse)
-rm/doc-collections/enforcement
Crew Brief. The crew is working through ES-0.1. The inter-unit 9:10:40 0:20:25 cross-tie to bus 15 is nearly complete.Report back from TB operator on status of 12 Battery -12 Battery is 9:11:30 0:21:15 discharging.
. In accordance with Inspection Manual Chapter (IMC) 0609, we intend to complete our evaluation using the best available information and issue our final determination of safety significance within 90 days of the date of this letter. The SDP encourages an open dialogue between the NRC staff and the licensee; however, the dialogue should not impact the timeliness of the staff's final determination.


Before the NRC makes its enforcement decision, we are providing you an opportunity to either: (1) present to the NRC your perspectives on the facts and assumptions used by the NRC to arrive at the finding and its significance at a Regulatory Conference, or (2) submit your positi on on the finding to the NRC in writing. If you request a Regulatory Conference, it should be held within 30 days of the receipt of this letter and we encourage you to submit supporting documentation at least one week prior to the conference in an effort to make the conference more efficient and effective. If a conference is held, it will be open for public observation. The NRC will also issue a press release to announce the conference. If you decide to submit only a written response, such submittal should be sent to the NRC within 30 days of the receipt of this letter. If you decline to request a Regulatory Conference or to submit a written response, you relinquish your right to appeal the final SDP determination; in that, by not doing either you fail to meet the appeal requirements stated in the Prerequisite and Limitation Sections of Attachment 2 of IMC 0609. Please contact John Giessner at (630) 829
TB Operator proceeded to the Water Treatment office to get a copy of 9:12:00 0:21:45 1 C20.9 AOP4.9:13:25 0:23:10 U2 RO announced that he was ready to cross-tie bus 15.Unit 2 Bus 25 connected to Bus 15. Bus 15 re-energized per 1C20.5 9:14:00 0:23:45 AOP 4.V.SPA.1 1.016 Page 4 of 8 Simulator Runs for Battery Charger Significance Determination Process Table 1-Simulator Dual Unit LOOP Event (with EDG D1 Lock-out)
-9619 in writing within 10 days of the date of this letter to notify the NRC of your intended response.
Scenario -3/05/2011 Time from Time Start of Event Event/Action TB operator reported that he is using 1C20.9 AOP4 and investigating 9:16:00 0:25:45 12 Battery Charger operation.


If we have not heard from you within 10 days, we will continue with our significance determination and enforcement decision. The final resolution of this matter will be conveyed in separate correspondence.
9:16:26 0:26:11 Announced that bus 15 loads can be restored.9:18:00 0:27:45 Control Room Update: Entered 1 C20.9 AOP4 for 12 Battery Charger 9:21:31 0:31:16 Began efforts to restore battery room cooling 9:22:34 0:32:19 Swapped charging pump suction from VCT to RWST Identified that letdown was not in service. Directed U2 RO to place 9:25:04 0:34:49 letdown in service.TB Operator reports back to control room information from Attachment A of 1 C20.9 AOP4. TB Operator recommended completing action in Attachment B of 1 C20.9 AOP4 (re-starting 12 9:27:00 0:36:45 Battery Charger)Reported the pressurizer level is 36% and rising with minimum 9:28:00 0:37:45 charging.Unit 1 SS directed TB operator to re-start 12 Battery Charger per 9:28:00 0:37:45 1C20.9 AOP4, Attachment B.9:30:00 0:39:45 Entered AOP to repower battery room cooling from D3 TB operator open AC input breaker for 12 Battery Charger, waited, 9:32:00 0:41:45 then reclosed breaker.Report from TB operator that 12 Battery Charger has been restored.Output is 250 amps. At Step 8 of 1C20.9 AOP4 and will continue to 9:33:00 0:42:45 monitor.9:33:50 0:43:35 Letdown established Unit 1 Lead RO confirmed with ERCS that 12 Battery Charger has 9:36:00 0:45:45 been restored.9:39:00 0:48:45 End of simulator run.3.2 Simulator "Follow-up Run" Results (March 8 th, 2011)The follow-up simulator run demonstrated the following risk-significant operator actions.a. The restoration of Bus 15 from Bus 25 via bustie breaker per 1C20.9 AOP4 occurred at approximately 17 minutes into the event.b. The restoration of the battery charger per 1C20.9 AOP4 occurred approximately at 51 minutes into the event.These actions were demonstrated in a follow-up simulator run performed on March 8, 2011. The crew consisted of Steve Ingalls (SM), Doug Larimer (SS), Greg Boek (UI RO), Jack Edwards (U I Lead RO)and Rick Lodermeier (Ul NLO).See Table 2 for a detailed event description and timing for this simulator preliminary run.V.SPA.1 1.016 Simulator Runs for Battery Charger Significance Determination Process Page 5 of 8
...Table 2-Simulator Dual Unit LOOP Event (with EDG D1 Lock-out)
Scenario -.3/08/2011 Time from Time Start of Event Event/Action 15:24 Crew has duty 15:35:06 0:00:00 Trigger 1 -LOOP event (Dual-unit)
with EDG D1 Lock-out 15:35:15 0:00:09 Enter 1E-0 15:35:31 0:00:25 Unit 1 RX Trip Plant Announcement Immediate Actions per 1 E-0 completed, Bus 15 is de-energized, SI 15:36:20 0:01:14 not actuated or required.15:36:33 0:01:27 1 E-0, Step 1 Reactor trip verified 15:37:21 0:02:15 1 E-0, Step 2 Turbine trip verified 1E-0, Step 3, Bus 16 is energized, Bus 15 has Undervoltage alarm 15:37:36 0:02:30 and is de-energized Turbine trip verified 15:38:00 0:02:54 12 DC System Trouble Alarm actuated on simulator Unit 1 SS Directs Unit 1 Lead RO to enter 1C20.5 AOP1 (Re-15:38:39 0:03:33 energizing 4.16KV Bus 15) to restore bus 15 15:39:00 0:03:54 Charging Pump in operation 15:39:03 0:03:57 1E-0, Step 4, Check SI actuated -no. Check if SI is required (RNO)15:40:03 0:04:57 DDCLPs running SAT Unit 1 Lead calls turbine building operator to investigate Undervoltage on Bus 15 protective relays per Step 2.4.2.B in 1C20.5 AOP1 (Re-15:40:44 0:05:38 energizing 4.16KV Bus 15)15:40:51 0:05:45 Transition to 1 ES-0.1 15:40:51 0:05:45 SM reports that all safety functions are GREEN 15:41:39 0:06:33 1 ES-0.1, Step 2 entered 15:42:42 0:07:36 1ES-0.1, Step 3 entered 15:45:00 0:09:54 1ES-0.1, Step 4 entered Turbine Building operator calls back control room to inform them that there are degraded and undervoltage relays on Bus 15 Load 15:45:00 0:09:54 Sequencer per 1C20.5 AOP1 Step 2.4.2.B 15:46:00 0:10:54 Alert declared, 577 completed 15:46:27 0:11:21 1 ES-0.1, Step 5 entered Unit 1 RO calls turbine building operator to perform 1 ES-0.1, 15:46:55 0:11:49 Attachment J (Isolate Unit 1 Moisture Separator Reheaters)
15:47:00 0:11:54 SS and Lead RO discuss re-energizing Bus 15 from crosstie 15:47:34 0:12:28 1ES-0.1, Step 6 entered 15:52:00 0:16:54 Cross-tie between Bus 15 and Bus 25 closed. Bus 15 is re-energized Unit 1 RO reports that the Reactor Makeup Flow deviation alarm is in.Discussed with Unit 1 SS and it was decided to open the RWST to 15:54:00 0:18:54 charging pump suction (MV-32060).


Since the NRC has not made a final determination in this matter, no Notice of Violation is being issued for this inspection finding at this time. Please be advised that the number and characterization of the apparent violation described in the enclosed inspection report may change as a result of further NRC review. In accordance with 10 CFR 2.390 of the NRC's "Rules of Practice," a copy of this letter, its enclosure, and your response (if any) will be available electronically for public inspection in the NRC Public Document Room or from the Publicly Available Records System (PARS) component of NRC's document system (ADAMS). ADAMS is accessible from the NRC Website at http://www.nrc.gov/reading
15:55:00 0:19:54 MV-32060 is opened.Unit 1 Lead RO is in ARP 47024-1105 for 12 DC System Trouble 15:55:00 0:19:54 Alarm 15:55:55 0:20:49 1ES-0.1, Step 7 entered 15:56:00 0:20:54 Normal alarm response protocol re-established 15:56:34 0:21:28 1ES-0.1, Step 8 entered V.SPA.1 1.016 Simulator Runs for Battery Charger Significance Determination Process Page 6 of 8 Table 2-Simulator'
-rm/adams.html (the Public Electronic Reading Room).
Dual Unit;LOOP Event (with EDG D1 L"ck-0ut)
Scenario 13/08/2011 Time from,-Time Start of Event Event/Action Unit 1 Lead RO Recognized that 12 Battery is dischargin Battery is supplying load. This was discussed with Unit 1 SS and Lead RO was told to instruct outplant operator to look at 12 battery 15:58:45 0:23:39 charger.Unit 1 Lead RO call outplant operator to investigate 12 battery 15:59:21 0:24:15 charger 15:59:27 0:24:21 1 ES-0.1, Step 8.b -RO instructed to place letdown into service Outplant Operator called from 12 Battery Room -Information transmitted to Lead RO was: 0 amps DC output (charger), 117 V DC, 16:06:09 0:31:03 Low Voltage Lit, and AC Light ON.12 Battery Charger information from ouptlant operator was discussed 16:07:21 0:32:15 with Unit 1 SS 16:08:33 0:33:27 Entered 1 C20.9 AOP4 (Failure of 12 Battery Charger)1C20.9 AOP4, Attachment A, dispatched to Turbine Building 16:10:00 0:34:54 Operator 16:12:27 0:37:21 1ES-0.1, Step 9 entered 16:13:57 0:38:51 1ES-0.1, Step 10 entered 16:14:55 0:39:49 1ES-0.1, Step 11 entered 1 ES-0.1, Step 11.a.RNO.b, RO calls outplant operator to open VC-1 -16:17:21 0:42:15 1 16:19:49 0:44:43 1ES-0.1, Step 11.C entered Unit 1 Lead RO instructed outplant operator to restart 12 Battery 16:22:00 0:46:54 Charger per 1C20.9, AOP4, Attachment B.Per 1 ES-0.1, Step 11 .C.RNO.C, Unit 1 SS discussed that battery 16:23:50 0:48:44 room cooling must be placed in service per C20.16 AOP1.16:24:00 0:48:54 Discussed using TSC support staff for re-start Battery Room Cooling 16:26:00 0:50:54 Restarted 12 Battery Charger 16:28:00 0:52:54 Used ERCS to confirm 12 Batter Charger was restarted.


Sincerely,/RA/
16:30:00 0:54:54 End of simulator run.4.0 Conclusions The purpose of these simulator runs was to demonstrate that the actions assumed in the PRA success strategy were reasonable given the time availabl The limiting depletion time for the 12 Battery for this event is just over three hours assuming Bus 15 remains de-energized (PRA Calculation V.SPA. 11.003).The depletion is much slower once Bus 15 is restored and the emergency lights are no longer a load on the 12 Battery (PRA Calculation V.SPA.1 1.009). The simulator runs demonstrated that Bus 15 can be re-energized via the bustie breaker in less than 30 minutes during a LOOP event and the battery charger can be restored in less than an hour. Once the charger is restored the event is over. The assumptions in the PRA success strategy can be concluded to be reasonable.
Steven West, Director Division of Reactor Projects Docket Nos.
: 50-282; 50-306; 72-010 License Nos.: DPR-42; DPR-60; SNM-2506


===Enclosure:===
V.SPA.1 1.016 Simulator Runs for Battery Charger Significance Determination Process Page 7 of 8 5.0 References 5.1 PRA Evaluation V.SPA. 11.003, Prairie Island Batter, Depletion Study PRA LOOP with Emergency Lighting and ISI Steady State Test Loads, Revision 0.5.2 PRA Evaluation V.SPA. 11.009, Prairie Island Batter, Depletion Study PRA LOOP with ISI Steady State Test Loads, Revision 0.5.3 PRA Evaluation V.SPA. 11.012, Batteny Charger SDP: Accident Sequence Analysis, Revision 0.5.4 1C20.5 AOP1, Re-energizing 4.16kV Bus 15, Revision 12.5.5 1C20.9 AOP4, Failure of 12 Battery Charger, Revision 10.5.7 1ES-0.1, Reactor Trip Recover,, Revision 25.5.8 1C20.5 AOP4, Reenergizing 4.16kV Bus 15 Via Bustie Breakers, Revision 3W.V.SPA.1 1.016 Page 8 of 8 Simulator Runs for Battery Charger Significance Determination Process AfKc6hment Page 1 ofA,-XceIEnergy" IPrior Knowledge Statement SIMULATOR SCENARIO:
Inspection Report 05000282
Scenario performed in support of the Prairie Island Battery Charger SDP conducted on March 8h, 2011 and commehcing at approximately 1500 I confirm that I had no prior knowledge of the details used in the above simulator scenari Further, I confirm that I had no knowledge of the details of the simulator run performed on March 5t, 2011. I understand that had I had prior knowledge of the scenario it would impact the legitimacy and use of the information gathered in support of the Battery Charger SDP.PRINTED NAME JOB TITLE / RESPONSIBILITY SIGNATURE DATE NOTE 2.5.6.7.8.9.10.NOTES:
/2 011010; 050003 06/2 011010
fa Pageof 1 SXcelEnergy-
Prior Knowledge Statement SIMULATOR SCENARIO:
Scenario performed in support of the Prairie Island Battery Charger SDP conducted on March 8th, 2011 and commencinq at approximately 1500 I confirm that I had no prior knowledge of the details used in the above simulator scenari Further, I confirm that I had no knowledge of the details of the simulator run performed on March 5t, 2011; I understand that had I had prior knowledge of the scenario it would impactthe legitimacy and use of the information gathered in support of the Battery Charger SDP.PRINTED NAME JOB TITLE / RESPONSIBILITY
_VNATURE DATE NOTE 2.3.4.5.6.7.8.9.9i0- T_______1___
10.NOT NOTES:.j AwIa.cXv~wo 2Z Page 1 of/A -I XceIEnergy SIMULATOR SCENARIO SECURITY AGREEMENT SIMULATOR SCENARIO:
Scenario performed in support of the Prairie Island Battery Charger SDP conducted on March 8e, 2011 and commencing at approximately 1500 During the post-critique I agreed that I would NOT discuss any aspects associated with the contents of this simulator run with any other operator until informed that it was acceptable to do so. I understand that violation of the conditions of this agreement may impact the legitimacy and use of the information gathered should subsequent runs of this scenario become necessary.


===w/Attachment:===
PRINTED NAME JOB TITLE / RESPONSIBILITY SIGNATURE DATE NOTE 1. .j _ .T-f)3. 1 t e 1kOA 3/5.6.7.8.9.10._NOTES:--
Supplemental Information cc w/encl:
Page;kofT 4, SXeelEnergyl SIMULATOR SCENARIO SECURITY AGREEMENT SIMULATOR SCENARIO:
Distribution via ListServ
Scenario performed in support of the Prairie Island Battery Charger SDP conducted on March 8t". 2011 and commencing at approximately 1500 During the post-critique I agreed that I would NOT discuss any aspects associated with the contents of this simulator run with any other operator until informed that it was acceptable to do so. I understandthat violation of the conditions of this agreement may impact the legitimacy and use of the information gathered should subsequent runs of this scenario become necessary.


Enclosure U.S. NUCLEAR REGULATORY COMMISSION REGION III Docket Nos:
PRINTED NAME JOB TITLE / RESPONSIBILITY SIGNATURE DATE NOTE 2.3.4.5.6.7.8.9.10. _NOTES: U ENCLOSURE2 BATTERY DEPLETION CALCULATION LOAD PROFILE 3 pages follow Batterv DeDletion Calculation Load Profile Purpose: The purpose of this document is to summarize the basis for the load profile used to determine the battery depletion time at the Prairie Island Nuclear Generating Plant (PINGP). Battery depletion time was needed to support the risk assessment of the battery charger lock up condition that could occur for the old battery chargers during loss of offsite power and safety injection accident scenario For the purposes of the risk assessment evaluations, battery depletion is defined as the time for the batteries to reach minimum design voltage when supplying the required loads.Discussion:
50-282; 50-306; 72-010 License Nos:
Although battery depletion is defined as battery voltage reaching minimum design voltage when the batteries are supplying required loads, the risk assessment depletion calculations were developed as best-estimate evaluations so that the actual plant and operator responses to the analyzed events could be realistically evaluate As a result, use of design-basis inputs for DC system loading would not be appropriat Rather, a load profile based on actual, expected system response was develope The load profile used in the risk assessment depletion calculations consists of three parts: a first minute transient portion, a steady state portion, and a final minute transient portion. Development of each of the load profiles is discussed below.A one hour design-basis battery loading profile is developed in the design-basis battery sizing calculations to meet the battery sizing requirements in the Updated Safety Analysis Report (USAR). The purpose of the load profile developed to meet the USAR requirements is to determine a worst case and conservative load profile, thereby resulting in a conservative battery size and greater battery capabilit The load profile developed for the depletion calculations was modeled after the design-basis load profile with some adjustments to reflect expected system response.The design-basis load profile developed for the USAR battery sizing requirements contains a first minute transient portion, a steady state portion, and a final minute transient portion. However, the design-basis load profile and calculation contain conservative assumptions to ensure that the calculation and required battery size has adequate margin to the minimum size required by the USAR.In the design-basis load profile, the first minute transient portion of the load profile contains various changes in the DC load due to multiple breaker operations and device actuation The first minute profile modeled in the risk assessment battery depletion calculations is the same as the first minute profile used in the design-basis calculations with a slight adjustment downward for the inverter loads. The adjusted inverter loading was determined from reviewing historical operating data taken during regular inverter surveillance Use of the adjusted design-basis load profile continues to result in a conservative first minute profile in the depletion calculation The loading includes many smaller loads that are assumed energized in the calculation which would actually de-energize on an event. These are further described below.The steady state portion of the battery depletion load profile was modified from the design-basis load profile. As stated above, the design-basis load profile contains conservative assumptions to provide additional margin to actual system respons One of the major assumptions which results in an overly conservative steady state load in the design-basis load profile is that various loads such as lights, relays, solenoid valves, etc. are modeled as energized during the entire scenari While this assumption is appropriate for use in the design-basis evaluations that are to determine conservative battery sizes, this assumption is overly conservative for an evaluation of actual expected system respons As a result of these conservatisms, use of the design-basis steady state profile would result in much higher loading than what would be expected in a real event. This is not consistent with a best estimate evaluation.
DPR-42; DPR-60; SNM-2506 Report No s: 05000282/20 11010; 050 003 06/20 11010 Licensee: Northern States Power Company, Minnesota Facility: Prairie Island Nuclear Generating Plant, Units 1 and 2 Location: Welch, MN Dates: May 13 through 20, 2011 Inspectors:
K. Stoedter, Senior Resident Inspector P. Zurawski, Resident Inspector C. Brown, Reactor Engineer L. Kozak, Senior Reactor Analyst Observer: S. Lynch, Nuclear Safety Professional Development Program Participant Approved by:
John B. Giessner, Chief Branch 4 Division of Reactor Projects Enclosure


=SUMMARY OF FINDINGS=
A review of the major contributors of the steady state load estimated that the actual load would be reduced by approximately 50% or more from the design-basis steady state load values when actual equipment operation is considere This review identified various DC system components that are modeled as a constant energized load which would actually de-energize on an event or have only momentary actuation For example, solenoid valves which de-energize to their fail safe position are modeled as being energized for the entire scenari It was also identified that dual indicating lights (on/off, open/close, etc.) were both included as energized for the entire scenario when only one light would be energized at a time. Reactor protection relays were also modeled as energized for the entire scenario when the relays would de-energize on a reactor trip.The Integrated Safety Injection (SI) Test simulates an SI with a Loss of Offsite Power (LOOP).During this test, the various devices which are required to respond to an event are actuated.Correspondingly, the DC load profile during the Integrated SI Test is representative of the load profile that would be expected during an actual event. The loading that occurs during the Integrated SI test was compared to the actual load profile that occurred during a past inadvertent SI event. The two profiles were approximately equal, showing a difference of less than 2 amps.The steady state DC load value observed during the Integrated SI Test also includes additional loads, such as inverters, that are present on Unit 1 due to operating conditions that are not present in the current design-basis calculation Therefore, use of the worst case best estimate steady-state load profile from either the Integrated SI Test or the inadvertent SI event provides a slightly conservative but realistic representation of the expected system response to a LOOP or SI event.This steady state load value was then adjusted upward for use in the risk assessment depletion calculation This upward load adjustment was to account for swing loads that could be present in scenarios evaluated in the risk assessment but would not be seen in the Integrated SI Test.The swing loads accounted for an additional load of approximately 4 -6 amps depending on the battery system being evaluate An upward adjustment of 7.5% of this adjusted steady state load was also added to provide additional conservatis This percentage increase resulted in an additional load of approximately 9 amps applied to 11 and 12 batteries, approximately 3 amps applied to 21 battery, and approximately 2 amps applied to 22 battery.This worst case scenario was a failure of Train A AC which is the normal power source for the emergency lights. To account for this scenario, an additional adjustment to the steady state load described above was added to the 12 and 22 batteries to account for the emergency lights. This resulted in an increase of 110 amps for the 12 battery and approximately 66 amps for the 22 battery at the evaluated voltage These adjustments result in a final steady state load profile that is slightly higher than would be expected but that can be used for the best-estimate battery depletion calculations.
IR 05000 282/2011010; 050003 06/2011010; 05/13/21011


- 05/20/2011; Prairie Island Nuclear Generating Plan t, Units 1 and 2; Other Activities.
The final minute of the load profile in both the design-basis load profile and the battery depletion load profile is a continuation of the steady state load described above with one additio In the last minute of the profile, it was assumed that operation of one 4160 VAC circuit breaker would be require The breaker operation is a momentary load with a relatively high load demand.Application of this load in the last minute of the profile is conservative as it causes a relatively significant drop in battery voltage due to the high current and length of time the battery has been deplete Since operation in the last minute is the most conservative time after the first minute due to the length of time the battery has already been depleted, the modeled momentary load is a bounding case for other miscellaneous momentary load operations which may occur. The assumption of a breaker operation results in a conservative last minute profil Summary: The load profile developed for the battery depletion calculations is a realistic representation of the actual expected system response during a LOOP or SI event. Development of the load profile for the risk assessment was based on the design-basis evaluation summarized in the USAR with modifications to remove overly conservative assumptions that are not appropriate for a best-estimate evaluatio Removal of the conservatisms combined information from the design-basis calculations as well as test data to obtain an accurate and conservative representation of the DC system load during an event. This resulted in battery depletion times which are representative of actual battery capability.


This report covers the review of a potential common cause failure of the safety-related battery chargers. The inspectors identified one apparent violation (AV) with a preliminary significance of White for Unit 1 and a preliminary significance of Green for Unit 2. The significance of most findings is indicated by their color (Green, White, Yellow, Red) using Inspection Manual Chapter (IMC) 0609, "Significance Determination Process" (SDP). Findings for which the SDP does not apply may be Green or be assigned a severity level after NRC management review.
References 1. PRA Evaluation V.SPA.1 1.006, Rev. 0, "Evaluation of Battery Depletion Differences" 2. Calculation 91-02-11, Rev. 2, "PI Battery 11 Calculation" 3. Calculation 91-02-12, Rev. 4, "PI Battery 12 Calculation" 4. Calculation 91-02-21, Rev. 2, "PI 21 Battery Calculation" 5. Calculation 91-02-22, Rev. 2, "PI Battery 22 Calculation" 6. PRA Evaluation V.SPA.1 1.003, Rev. 0, "Prairie Island Battery Depletion Study PRA LOOP with Emergency Lighting and ISI Steady State Test Loads" 7. Prairie Island Updated Safety Analysis Report, Section 8, Rev. 31, "Plant Electrical Systems" 8. Corrective Action Program Action Request #01270104, "Non conservative assumption in Unit 1 Battery Calcs" ENCLOSURE 3 V.SPA.11.006 EVALUATION OF BATTERY DEPLETION DIFFERENCES 20 pages follow Xcel Energy Prairie Island Nuclear Generating Plant Probabilistic Risk Assessment V.SPA.1 1.006 Evaluation of Battery Depletion Differences March 2011 Document Review Form Document or Analysis Title: V.SPA.1 1.006: Evaluation of Battery Depletion Differences Revision of Document or 0 Analysis: Date: Preparer:
Robert Flaaen .i Date: ,/0E.Zo ij Reviewer Greg Kvamme Date: 3 ' /ibof* " '
,//,,/PRA Acceptance Review: Jayne E. Ritter Prairie Island Nuclear Generating Plant (PINGP)Evaluation of Battery Depletion Differences Table of Contents 1.0 Purpose ...................................................................................................
3 2.0 Evaluation of DC System Loading ..........................................................
3 2.1 Comparison of Calculated Load to Observed Load ....................................
3 2 .1.1 11 D C S yste m ......................................................................................................
..3 2 .1.2 12 D C S ystem ......................................................................................................
..4 2.1.3 2 1 D C S ystem ......................................................................................................
..6 2 .1.4 22 D C S ystem ......................................................................................................
..7 2.2 DC Circuit Comparisons
.............................................................................
9 2 .2 .1 11 D C S ystem ......................................................................................................
..9 2.2.2 12 D C S ystem ...................................................................................................
..10 2.2.3 2 1 D C S ystem ...................................................................................................
..10 2.2.4 22 D C S ystem ....................................................................................................
..11 3.0 Evaluation of Loading Used in Battery Depletion Calculations
.............
13 3.1 Station Blackout (SBO) Depletion Calculation
............................................
13 3.2 Safety Injection on Offsite Power (SI) Depletion Calculation
......................
14 3.3 Loss of Offsite Power (LOOP) Depletion Calculation
................................
15 3.3.1 LOOP with Actual Inverter Loading ....................................................................
15 3.3.2 LOOP with Actual Inverter Loading and Failure of Opposite Train AC ..............
16 4.0 Other Influences on Battery Depletion Timing ......................................
18 4.1 Minimum Allowable Battery Terminal Voltage ......................................
18 5.0 Conclusion
.............................................
18 6.0 References
..........................................................................................
19 1.0 Purpose The purpose of this evaluation is to document and provide justification for the differences in battery depletion times calculated for Probabilistic Risk Assessment (PRA) purposes for a Station Blackout (SBO) scenario, Loss of Offsite Power (LOOP) scenario, and Safety Injection on Offsite Power (SI) scenari This evaluation will review the battery loading that was assumed in the calculations and describe the differences.


The NRC's program for overseeing the safe operation of commercial nuclear power reactors is described in NUREG-1649, "Reactor Oversight Process," Revision 4, dated December 2006.
2.0 Evaluation of DC System Loading A battery loading profile of one hour is developed in the design basis battery sizing calculations (Ref. 6.1, 6.2, 6.3, and 6.4) per requirements stated in the Updated Safety Analysis Report (USAR) (Ref 4.10). The design basis load profile is based on assuming worst case loading for loads to develop a conservative load profile. Additionally, other than major loads (breaker coils, spring charging motors, inverters), smaller loads (lights, misc relays, solenoids)
are assumed on for the entire duration of the analysis or alternating lights and similar components (red/green indicating lights, separate open/close solenoid valves on the same component, etc.) are assumed as both simultaneously on for the entire duration of the analysi This results in a conservative design basis load profile as the devices actually only operate for a short period of time and do not all operate simultaneously.


A.
To obtain the load for these miscellaneous panels which feed such devices as solenoid valves, relays, etc., the design basis calculation took all of the devices on a circuit and added up their individual rated loads, whether the individual devices were on or not. All depletion calculations used the design basis calculation and design basis load profile as a starting point and modified the load profile as necessary to obtain battery depletion times suitable for the specific evaluation being performed.


===Cornerstone: Mitigating Systems===
A comparison of the design basis steady state DC load to the steady state load observed during testing follows in Section 2.1. Additionally, some specific example of circuits are included in Section 2.2 by evaluating the design basis load and reviewing anticipated loading on that circuit based on actual operation of devices.2.1 Comparison of Calculated Load to Observed Load Because of the conservative load profile used in the calculation, the DC load used in the calculation is much greater than that experienced while testing simulated LOOP/SI conditions during Integrated SI testing and also experience during an inadvertent SI event on Unit 2. During these events, the first 30-40 seconds contains a number of transient periods due to breaker closures, valve and relay actuations, etc. However, after 30 seconds, the DC load settles to a steady state value. The steady state value shown in the design basis calculations is during the 1 minute to 59 minute time period.When comparing this to steady state DC load values experienced during simulated or actual event conditions, the calculated value is much higher than what is actually experience As the steady state load is the major influence on the depletion times, the use of a conservative steady state load values on the batteries can result in a conservative depletion time and would not be representative of actual system load or response.2.1.1 11 DC System The design basis calculated load of the 11 DC system at steady state conditions is approximately 123.5 amps (Ref. 6.1). This value does not include inverter loads based on assumptions within the calculatio The design basis inverter loading on 11 battery is approximately 129.5 amps (Ref. 6.1). From the last performance of the Integrated SI test shown below in Figure 1, it can be seen that Page 3 of 19 the steady state load of the 11 DC system experience during the test is approximately 115 amps.Unit I Train A 1R26 Data (10/23/2009)
300 150 Time (Seconds)-Battery Float Current -11 Charger Current -DC System Demand (Charger-Float)
Figure 1 This steady state value from the Integrated SI test is assumed to contain the inverter loads as the inverter loads were not removed from the DC system during the event. This is due to a trip of the inverter AC input breaker on the large voltage overshoot of the EDGs. The basis for this assumption is the difference between the pre-event DC load and the steady state load. This difference (approximately 83 amps) closely resembles the actual inverter load. It would be expected that the steady state load would be approximately equal to or only slightly greater than the pre-event DC load due to the lack of the actuation of loads and generally the same number of operating (running)
loads (indicating lights, continuously energized lights, etc.). This is confirmed by reviewing Integrated SI test data and actual SI data from Unit 2 in which the inverter loads do not remain on the DC system. The comparison of the Unit 2 data showed that the steady state load was only a few amps higher than the pre-event load for both an SI/LOOP and SI loading condition (Ref. Sections 2.1.3 and 2.1.4).The tripping of the inverter AC input breaker on a high voltage has been identified to occur on Unit 1 due to the Unit 1 EDG voltage response during transient loading (Ref. 6.6). This issue has been input into the corrective action program and is being evaluated for additional impact (Ref. 6.11).The profile above in Figure 1 does not include transferable DC panels 17 and 19.These panels are normally fed from the 11 DC system (Ref. 6.12) however, are transferred to their alternate source (21 DC system) during a unit shutdown (Ref.6.13) are then returned to their normal source after a unit startup (Ref. 6.14). The design basis load for these panels on the batteries is approximately 12.4 amps.2.1.2 12 DC System The calculated load of the 12 DC system at steady state conditions is approximately 201.5 amps (Ref. 6.2). This value does not include inverter loads, Page 4 of 19 but does include emergency lightin The design basis inverter loading on 12 battery is approximately 134.8 amps (Ref. 6.2). The design basis emergency lighting load on 12 battery is approximately 110 amps (Ref. 6.2). From the last performance of the Integrated SI test shown below in Figure 2, it can be seen that the steady state load of the 12 DC system is approximately 115 amps. Note that since the 12 battery charger was not included in the test, it is not shown in this Figure.12 DC System Data from 1R26 Test (10/23/2009)
2C o II II I I I I I l 1 l l I IT , I Eb -- 1 1 -I 11 11 11 11I WI iI I + [II III II II M1 II II I II JM I I I I II isIIIiI i I I Ji tN I Seconds-DC Demand Figure 2 This steady state value from the Integrated SI test does not include the emergency lighting, however, it is assumed that the steady state value contains the inverter loads as the inverter loads were not removed from the DC system during the event due to a trip of the inverter AC input breaker on the large voltage overshoot of the EDGs. This is due to the difference between the pre-event DC load and the steady state load. This difference (approximately 92 amps) closely resembles the actual inverter load. It would be expected that the steady state load would be approximately equal to or only slightly greater than the pre-event DC load due to the lack of the actuation of loads and generally the same number of operating (running)
loads (indicating lights, continuously energized lights, etc.).This is confirmed by reviewing Integrated SI test data and actual SI data from Unit 2 in which the inverter loads do not remain on the DC system. The comparison of the Unit 2 data showed that the steady state load was only a few amps higher than the pre-event load for both an SI/LOOP and SI loading condition (Ref.Sections 2.1.3 and 2.1.4).The tripping of the inverter AC input breaker on a high voltage has been identified to occur on Unit 1 due to the Unit 1 EDG voltage response during transient loading (Ref. 6.6). This issue has been input into the corrective action program and is being evaluated for additional impact (Ref. 6.11).The profile above in Figure 2 does not include transferable DC panels 14 and 18.These panels are normally fed from the 12 DC system (Ref. 6.12) however, are Page 5 of 19 transferred to their alternate source (22 DC system) during a unit shutdown (Ref.6.13) are then returned to their normal source after a unit startup (Ref. 6.14). The design basis load for these panels on the batteries is approximately 8 amps.2.1.3 21 DC System The calculated load of the 21 DC system at steady state conditions is approximately 113.1 amps (Ref. 6.3). This value does not include inverter loads.The design basis inverter loading on 21 battery is approximately 94.6 amps (Ref.6.3). From the last performance of the Integrated SI test shown below in Figure 3, it can be seen that the steady state load of the 21 DC system is approximately 36 amps.Unit 2 Train A 2R26 Data (5/17/2010)
200 180 160 140 120 100 80 60 40 E i i i i i i i i i i i i i i i ! i i i i 4 " i i i i i i i i i i I i i ! i I I I I I I I I I I I I I II I I ¶ II I I I I I I I I I I I I CL 11 Ill! I 111111 N 11 W Ill 11111 1MIM 11111111111
............-..... .. ........ ..... ...0-20-40-60-80-100-120 fl1i iiii i i F.11 Sli iiiiii i~ Ti17lYIT F 1 II~.I Seconds i ?U M ý3 R 1-21 Battery Float Amps -21 Battery Charger Amps -21 DC System Demand I Figure 3 Additionally, Unit 2 experienced an inadvertent SI event in 2007. Data taken from that event is shown below in Figure 4. From this data it can be seen that the steady state load of the 21 DC system is approximately 37 amps.Page 6 of 19 4/5/2007 U2 SI Event -Train A DC Loading (1st Minute)Load Step 1 2 3 4 5 6 7 110.0000 90.0000 80.0000 70.0000 60.0000 50,0000 400000 tE 300000 20.000-200000 V I-300000-400000-50.0000 Tlme 1- 21 Battery Float -21 Battery Charger Figure 4 2.1.4 22 DC System The calculated load of the 22 DC system at steady state conditions is approximately 155.6 amps (Ref. 6.4). This value does not include inverter loads, but does include emergency lightin The design basis inverter loading on 22 battery is approximately 103.5 amps (Ref. 6.4). The design basis emergency lighting load on 22 battery is approximately 65 amps (Ref. 6.4). From the last performance of the Integrated SI test shown below in Figure 5, it can be seen that the steady state load of the 22 DC system is approximately 21 amps.Unit 2 Train B 2R26 Data (5/17/2010)
250 200 II I 150 E o .I II-o 1 Seconds 1- 22 Battery Roat Amps -22 Battery Charger Amps -22 DC System Demand Figure 5 Page 7 of 19 Additionally, Unit 2 experienced an inadvertent SI event in 2007. Data taken from that event is shown below in Figure 6. From this data it can be seen that the steady state load of the 22 DC system is approximately 19.5 amps.4/5/2007 U2 SI Event -Train B DC Loading (lt Min)Load Step 1 2 3 4 5 6 7 I 90 80 70 60 50 40 30 20 1 10 L VA o -" ^ -,.,s. V-10-20-30-40-50 .... ....Time 1-22 BteyFlot -22 WeryCha~rer]
Figure 6 Page 8 of 19 2.2 DC Circuit Comparisons This section will look at some of the DC system circuits and compare the design basis loading used in the design basis calculation with estimated loading based on actual device operation to give some example of the conservatisms that are present in the design basis load profile of the batterie The design basis calculations obtain the DC loading information from the Master DC Load List (Ref. 6.15). However, as stated above under Section 2.0, the calculations summed up the rated load of each device on the circuit within miscellaneous load panels to obtain the panel load and did not consider device operatio This resulted in a conservative load value.Reference 6.15 also contained information which identified the expected operation of specific devices on a circuit. This was done by reviewing equipment operation, drawings, control circuits, etc., to try and determine when and if a device would operate during an event. Using this information, a load value which more accurately resembles actual expected load can be estimate This load value is still slightly conservative due to the use of worst case loading of individual devices and conservative device load assumptions where specific manufacturer data could not be located. The following sections will review the design basis load for various panels on the system based on Attachment J of References 6.1, 6.2, 6.3, and 6.4 and compare that to estimated load values based on device operation from Reference 6.15.2.2.1 11 DC System The major contributors to the design basis steady state load of the 11 DC system as shown in Reference 6.1 are panels 13, 15, 17, 19, 151,152, 153, and 191.The design basis load of these panels at nominal DC voltage (125VDC) is tabulated in Attachment J of Reference 6.1. The expected steady state load for a HSD condition, such as during a LOOP, can be estimated using the Reference 6.15 information of whether the device would operate during the event and actually be a load on the DC system. The different load values are compared below in Table 1.DC Panel Design Basis Load Estimated Expected Load % Reduction From Attachment J of From Reference 4.15 Reference 4.1 (Amps) (Amps)15 17.64 7.11 60%17 4.84 3.26 33%19 7.22 4.07 44%151 23.54 14.04 40%152 9.44 6.88 27%153 5.67 1.24 78%191 6.77 2.15 68%Total 75.12 38.75 48%Table 1: Comparison of 11 DC System Loading Panel 13 is not included in the above table since it is not detailed in Reference 6.15. Assuming a 48% load reduction for panel 13 to be consistent with the total load reduction seen by the above panels, panel 13 load can also be considered as shown below in Table 2.Page 9 of 19 DC Panel Design Basis Load Estimated Expected Load % Reduction From Attachment J of From Reference 4.15 Reference 4.1 (Amps) (Amps)13 34.56 16.73* 52%15 17.64 7.11 60%17 4.84 3.26 33%19 7.22 4.07 44%151 23.54 14.04 40%152 9.44 6.88 27%153 5.67 1.24 78%191 6.77 2.15 68%Total 109.68 55.48 49%*Assumed Load Reduction of 48%Table 2: Comparison of 11 DC System Loading with Panel 13 This shows that the design basis steady state load for the 11 DC system is very conservative to expected and observed loading and is not reflective of actual system performance or loading.2.2.2 12 DC System The major contributors to the design basis steady state load of the 12 DC system as shown in Reference 6.2 are panels 14, 16, 161,162, 161,181. The design basis load of these panels at nominal DC voltage (1 25VDC) is tabulated in Attachment J of Reference 6.2. The expected steady state load for a HSD condition, such as during a LOOP, can be estimated using the Reference 6.15 information of whether the device would operate during the event and actually be a load on the DC system. The different load values are compared in Table 3.DC Panel Design Basis Load Estimated Expected Load % Reduction From Attachment J of From Reference 4.15 Reference 4.2 (Amps) (Amps)14 6.74 2.95 56%16 41.59 8.54 79%18 1.25 1.15 8%161 21.35 3.28 85%162 6.72 2.67 60%163 3.41 0.42 88%Total 81.06 19.01 77%Table 3: Comparison of 12 DC System Loading This shows that the design basis steady state load for the 12 DC system is very conservative to expected and observed loading and is not reflective of actual system performance or loading.2.2.3 21 DC System The major contributors to the design basis steady state load of the 21 DC system as shown in Reference 6.3 are panels 17, 19, 23, 25, 251, 252, and 253. The design basis load of these panels at nominal DC voltage (125VDC) is tabulated in Attachment J of Reference 6.3. The expected steady state load for a HSD condition, such as during a LOOP, can be estimated using the Reference 6.15 information of whether the device would operate during the event and actually be a load on the DC system. The different load values are compared in Table 4.Page 10 of 19 DC Panel Design Basis Load Estimated Expected Load % Reduction From Attachment J of From Reference 4.15 Reference 4.3 (Amps) (Amps)17 4.84 3.26 33%19 5.13 4.07 21%25 18.23 3.59 80%251 11.85 5.2 56%252 8.63 5.31 38%253 6.33 1.55 76%Total 55.01 22.98 58%Table 4: Comparison of 21 DC System Loading Panel 23 is not included in the above table since it is not detailed in Reference 6.15. Assuming a 50% load reduction for panel 23 to be consistent with the total load reduction seen by the above panels, panel 23 load can also be considered as shown below in Table 5.DC Panel Design Basis Load Estimated Expected Load % Reduction From Attachment J of From Reference 4.15 Reference 4.3 (Amps) (Amps)17 4.84 3.26 33%19 5.13 4.07 21%23 33.1 16.55* 50%25 14.16 3.59 75%251 11.85 5.2 56%252 8.63 5.31 38%253 6.33 1.55 76%Total 84.04 39.53 53%*Assumed Load Reduction of 50%Table 5: Comparison of 21 DC System Loading with Panel 23 This shows that the design basis steady state load for the 21 DC system is very conservative to expected and observed loading and is not reflective of actual system performance or loading.2.2.4 22 DC System The major contributors to the design basis steady state load of the 22 DC system as shown in Reference 6.4 are panels 14, 18, 26, 261, 262, and 263. The design basis load of these panels at nominal DC voltage (1 25VDC) is tabulated in Attachment J of Reference 6.4. The expected steady state load for a HSD condition, such as during a LOOP, can be estimated using the Reference 6.15 information of whether the device would operate during the event and actually be a load on the DC system. The different load values are compared in Table 6.DC Panel Design Basis Load Estimated Expected Load % Reduction From Attachment J of From Reference 4.15 Reference 4.2 (Amps) (Amps)14 6.82 2.95 57%18 1.25 1.15 8%26 30.39 5.2 83%261 11.14 2.81 75%262 5.52 0.21 96%263 3.54 0.42 88%Total 58.66 12.74 78%Table 6: Comparison of 22 DC System Loading Page 11 of 19 This shows that the design basis steady state load for the 22 DC system is very conservative to expected and observed loading and is not reflective of actual system performance or loading.Page 12 of 19 3.0 Evaluation of Loading Used in Battery Depletion Calculations 3.1 Station Blackout (SBO) Depletion Calculation On the occurrence of an SBO, the batteries supply the emergency lights (where applicable)
and inverters due to a loss of AC power. The SBO depletion calculation used the design basis load profile and modified the inverter (INV) loading to more realistically resemble actual data by removing the conservatism in the inverter load.The inverter load values that were used were based on actual load readings taken during surveillance procedures of the DC system. The approximate inverter load values used in the SBO calculation, the difference from the design basis inverter load, as well as the resultant steady state load in the SBO calculation and is shown below in Table 1. The SBO steady state load includes values with and without the emergency lights (E.L.) where applicable as it was evaluated that during an SBO, the emergency lights are manually shed from the DC system 40 minutes into the event per emergency procedures. (Ref. 6.5)DC System Approximate Inverter Difference In Approximate Steady Load in SBO Inverter Load State Load in SBO Depletion Calculation From Design Depletion Calculation (Ref. 6.5) Basis Calculation (Ref. 6.5)11 DC Inverters 97.7 amps 31.8 amps lower 219 amps 12 DC Inverters 107.2 amps 27.6 amps lower 305 amps (w/ E.L.)1 1 197 amps (w/o E.L.)21 DC Inverters 89.1 amps 5.5 amps lower 201 amps 22 DC Inverters 85.36 amps 18.1 amps lower 239 amps (w/ E.L.)174 amps (w/o E.L.)Table 1 This steady state load that was used in the SBO depletion calculations still contains the conservatisms of various loads being energized during the entire event. When comparing the steady state load used in the SBO depletion calculation to the DC load profile during the Integrated SI Test in Figures 1-4 under Section 2.1, it can be seen that the steady state load used in the SBO depletion calculation is above the steady state load seen under actual conditions and in some cases is above the highest load experience during the entire Integrated SI Test when the Emergency Diesel Generator (EDG) is starting up and there is no AC power.DC Maximum DC System DC System Steady Approximate Steady System Load During Sequence State Load From State Load in SBO From Section 2.1 Figures Section 2.1 Figures Depletion Calculation (ISI Test Data) (includes Inverter and E.L (if applicable))(Ref. 6.5)11 133 amps (2) 115 amps 219 amps 12 235 amps (13 115 amps 305 amps (w/ E.L.)197 amps (w/o E.L.)21 113 amps 37 amps 201 amps 22 145 amps V) 21 amps 239 amps (w/ E.L.)174 amps (w/o E.L.)Table 2 (1) The load value listed here also includes the Inverters and (2) The load value listed includes the Inverters.


N RC-Identified and Self-Revealed Findings  Preliminary WhiteThis finding was determined to be more than minor because it was associated with the design control and equipment performance attributes of the Mitigating Systems Cornerstone. In addition, this performance deficiency impacted th e cornerstone objective of ensuring the availability, reliability, and capability of systems that respond to initiating events to prevent undesirable consequences.
Emergency Lighting.Page 13 of 19 This shows that the steady state load values used in the SBO depletion calculation were still very conservative even with the adjustment of the inverter loads. However, the depletion times reached in the calculation were adequate for the evaluation that was occurring at the time and therefore, no additional conservatism needed to be removed.The SBO depletion calculation (Ref 6.5) provided the following results: Battery Depletion Time Battery 11 165 minutes (2 hours 45 minutes)Battery 12 217 minutes (3 hours 37 minutes)Battery 21 184 minutes (3 hours 04 minutes)Battery 22 239 minutes (3 hours 59 minutes)3.2 Safety Injection on Offsite Power (SI) Depletion Calculation This event occurs on offsite power and therefore, there is no loss of AC power to the emergency lights or the inverters and they remain powered from their respective AC sources. This calculation used the design basis load profile, modified it to represent SI sequencing without a loss of offsite power, and removed the inverters and the emergency lighting (as applicable)
due to no loss of AC power. Table 3 shows the steady state load values used in the depletion calculation and compares them to steady state load values seen from Figures 1 -4 under Section 2.1.DC System Approximate Steady State Approximate Steady Load in SI Depletion States Load From Calculation (Ref. 6.7) Section 2.1 Figures 11 124 amps 115 amps 12 93 amps 115 amps 1'21 114 amps 37 amps 22 91 amps 21 amps Table 3 (1) The load value listed includes the Inverters.


The inspectors performed a Phase 1 SDP evaluation and determined that a Phase 2 evaluation was required because this finding represented an actual loss of safety function of a single train of equipment for greater than the TS allowed outage time. The inspectors performed a Phase 2 evaluation using the pre-solved SDP worksheets for Prairie Island and determined that this finding screened as Red.
As shown above in the table, the steady state load for 12 DC system from Section 2.1.2 is higher than that used in the SI only depletion calculatio The load profile under Section 2.1.2 is a SI/LOOP conditio As discussed in Sections 2.1.1 and 2.1.2, the load profiles for 11 and 12 DC system contained the inverter loads during the entire sequence, as well as during the steady state portion of the profile, due to the inverters'
AC input breaker tripping on an EDG voltage overshoot causing them to be a continuous load on the DC system. The event analyzed under the SI only depletion calculation is when offsite power is availabl The voltage overshoots experienced during Unit 1 EDG sequence loading do not occur when sequencing on offsite power. Therefore, the inverters would not be a steady state load on the Unit 1 DC system for this event. If the inverter loads were removed from the profiles under Sections 2.1.1 and 2.1.2, the steady state load would be slightly higher than the pre-event load of approximately 30 amps for 11 DC system and 20 amps for 12 DC system shown in Figures 1 and 2.Page 14 of 19 The steady state load that was used in the SI only depletion calculations still contains the conservatisms of various loads being energized during the entire event. When comparing the steady state load used in the depletion calculation to the DC load profile during the Integrated SI Test in Figures 1-4 under Section 2.1, it can be seen that the steady state load used in the calculation is very conservativ However, the depletion times reached in the calculation were adequate for the evaluation that was occurring at the time and therefore, no additional conservatism needed to be removed.The SI only depletion calculation (Ref 6.7) provided the following results: Battery Depletion Time To Design Basis Voltage Battery 11 460 minutes (7 hours 40 minutes)Battery 12 710 minutes (11 hours 50 minutes)Battery 21 515 minutes (8 hours 35 minutes)Battery 22 645 minutes (10 hours 45 minutes)3.3 Loss of Offsite Power (LOOP) Depletion Calculation This event was further evaluated into two separate scenarios:
(1) LOOP with Actual Inverter Loading (Ref. 6.8) and (2) LOOP with Actual Inverter Loading and Failure of Opposite Train AC (Ref. 6.9).The initial case (1) evaluated the actual response where only the Unit 1 inverters become a load on the DC system due to tripping of their AC input breaker on a Unit 1 EDG overshoo Unit 2 inverters were assumed to remain on AC power since it is not expected that the Unit 2 inverters'
AC input breaker will trip due to a lack of significant overshoots during EDG sequencing (Ref. 6.6). This case also modified the steady state load to match what is seen during the Integrated SI Test or what was seen during the inadvertent SI event. This removes the conservative assumption in which various loads are continuously energized and more closely resembles actual system loading and response.The second case (2) used the load profile developed under case 1 in which the inverters were a load on the Unit 1 EDGs and the steady state load matched the actual test or event data. Additionally, this case assumed a failure of the opposite train's AC power source. The emergency lighting is fed from Train A AC power and Train B DC power. By assuming a failure of the Train A AC power, the emergency lights become a load on the Train B DC system during the entire event. This is considered the limiting case for DC loading and battery depletion during a LOOP event.3.3.1 LOOP with Actual Inverter Loading This case used the same load profile developed for the SBO calculations except that the emergency lights were removed from the load profile after the EDGs provided AC power to the bus, the Unit 2 inverters were removed from the load profile after the EDGs provided AC power to the bus (the Unit 1 inverters remained on the DC system), and the steady state load was adjusted to approximately match actual test or event data. The missing panels described under Section 2.1 for Unit 1 DC systems were conservatively added to the steady state load values observed from the tests for both units, plus some additional margin.Page 15 of 19 Table 4 shows the steady state load values used in the depletion calculation and compares them to steady state load values seen from Figures 1 -4 under Section 2.1.DC System Approximate Steady Approximate Steady State Load in Case 1 States Load From LOOP Depletion Section 2.1 Figures Calculation (Ref. 6.8)11 130 amps 115 amps 12 128 amps 115 amps 21 50 amps 37 amps 22 30 amps 21 amps Table 4 This steady state load that was used in the Case 1 LOOP depletion calculations still contains some margin to actual values but is much more reflective of actual plant conditions and a more realistic load value.The Case 1 LOOP (with actual inverter loading and Steady State Load) depletion calculation (Ref 6.8) provided the following results: Battery Depletion Time To Design Basis Voltage Battery 11 Greater than 6.5 hours Battery 12 Greater than 7.5 hours Battery 21 Greater than 22 hours Battery 22 Greater than 24 hours 3.3.2 LOOP with Actual Inverter Loading and Failure of Opposite Train AC This case used the same load profile for the Case 1 LOOP depletion calculation discussed in Section 3.3.1, however, due to the assumed failure of the opposite train AC power source, the emergency lights (EL) were included as a load on 12 and 22 batteries during the entire time period. Table 5 shows the steady state load values used in the depletion calculation.


A Phase 3 SDP evaluation was required to assess reasonable credit for recovery by operators. The results of the Phase 3 SDP evaluation showed that this finding was preliminarily determined to be White for Unit 1
DC System Approximate Steady State Load in Case 2 LOOP Depletion Calculation (Ref. 6.9)11 130 amps z 12 238 amps (1)21 50 amps (2)22 96 amps "I Table 5 (1) The load value listed here also includes the Inverters and Emergency Lighting.(2) The load value listed includes the Inverters.
, and Green for Unit 2. No cross-cutting aspect was assigned to this finding because licensee decisions made in regards to evaluating the performance of the battery chargers were made man y years ago and therefore, not reflective of current plant performance. (Section 4OA5.1)
. An apparent violation of Technical Specification (TS) 3.8.4 was identified by the inspectors due to the licensee's failure to maintain the train A and train B direct current electrical power subsystems operable while operating the reactor in Modes 1 through 4. Specifically, the licensee installed safety-related battery chargers which were susceptible to failure during certain design basis events. This issue was entered into the licensee's corrective action program (CAP) as CAP 1250561. Upon identifying this issue, the licensee performed an operability evaluation and determined that the battery chargers remained operable because procedures were in place to recover the battery chargers if a failure occurred. After further interaction with the NRC, the licensee concluded that a design ated operator position needed to be established to ensure that a specific individual would perform the battery charger recovery actions prior to the safety-related batteries being depleted.


Long term corrective actions included replacing all four battery chargers.
(3) The load value listed includes Emergency Lighting This steady state load that was used in the Case 2 LOOP depletion calculations still contains some margin to actual values but is much more reflective of actual plant conditions and a more realistic load value when the emergency lights would be powered from the DC system due to a failure of the opposite train AC.Page 16 of 19 The Case 2 LOOP (with actual inverter loading and Failure of Opposite Train AC Power) depletion calculation (Ref 6.9) provided the following results: Battery Depletion Time To Design Basis Voltage Battery 11 Greater than 6.5 hours Battery 12 Greater than 3 hours Battery 21 Greater than 22 hours Battery 22 Greater than 9 hours Page 17 of 19 4.0 Other Influences on Battery Depletion Timing 4.1 Minimum Allowable Battery Terminal Voltage The minimum allowable terminal voltage is specific to each battery and is determined by the battery sizing calculations (Ref. 6.1 to 6.4). The depletion time for the battery to reach minimum terminal voltage is largely influenced by the steady state load on the battery as described above, however, the minimum allowable terminal voltage also has an affect on the battery depletion time. The minimum allowable terminal voltage values for each battery is listed in table below.Battery Minimum Allowable Terminal Voltage Battery 11 110.34 Vdc (Reference 6.1 )Battery 12 109.55 Vdc (Reference 6.2)Battery 21 110.61 Vdc (Reference 6.3)Battery 22 110.08 Vdc (Reference 6.4)Given similar loading on each battery, the depletion time will be shorter for a battery with a higher minimum allowable terminal voltage limit. For example, in Section 3.3.1 (LOOP with Actual Inverter Loading), the 11 battery and 12 battery have very similar steady state loads (130 amps and 128 amps, respectively).


B. No violations of significance were identified.
However, the depletion times for 11 battery and 12 battery are different (6.5 hours and 7.5 hours, respectively).


===
The 12 battery depletion time is 1 hour longer due to the lower minimum allowable terminal voltage.5.0 Conclusion The evaluation performed above in Section 2.0 showed that the design basis loading profile of the batteries is very conservative when compared to actual data. When performing battery depletion studies to obtain a best estimate depletion time, the use of test or actual event data represents a more accurate reflection of system performance and loading.As the PRA evaluations progressed, significant initiating events were identified which required re-evaluation of the battery depletion calculations in order to identify conservatism As part of that process, conservatisms were removed from the design basis load profile to more accurately match real data.Page 18 of 19 6.0 References 6.1 Calculation 91-02-11, Rev. 2, "PI Battery 11 Calculation" 6.2 Calculation 91-02-12, Rev. 4, "PI Battery 12 Calculation" 6.3 Calculation 91-02-21, Rev. 2, "PI 21 Battery Calculation" 6.4 Calculation 91-02-22, Rev. 2, "PI Battery 22 Calculation" 6.5 PRA Evaluation V.SPA.1 0.013, Rev. 0, "Battery Depletion Calculation" 6.6 PRA Evaluation V.SPA.1 1.005, Rev. 0, "Evaluation of Inverter Source Selection During Emergency Diesel Generator Loading" 6.7 PRA Evaluation V.SPA.1 1.004, Rev. 0, "Prairie Island PRA SI Only Battery Depletion Study" 6.8 PRA Evaluation V.SPA.1 1.009, Rev. 0, "Prairie Island Battery Depletion Study PRA LOOP with ISI Steady State Test Loads" 6.9 PRA Evaluation V.SPA.1 1.003, Rev. 0, "Prairie Island Battery Depletion Study PRA LOOP with Emergency Lighting and ISI Steady State Test Loads" 6.10 Prairie Island Updated Safety Analysis Report, Section 8, Rev. 32P, "Plant Electrical Systems" 6.11 Corrective Action Program Action Request #01270104, "Non conservative assumption in Unit 1 Battery Calcs" 6.12 Operating Procedure C20.9, Rev. 29, "Station Battery & DC Distribution System" 6.13 Operating Procedure 1C1.3, Rev. 69, "Unit 1 Shutdown" 6.14 Operating Procedure 1C1.4, Rev. 51, "Unit 1 Power Operation" 6.15 Master DC Load List (File MasterDCLddlst_05-31-01 .xls Dated 05/31-01 at 1:08pm)Page 19 of 19
Licensee-Identified Violations===
 
=REPORT DETAILS=
 
==OTHER ACTIVITIES==
{{a|4OA3}}
==4OA3 Follow-U p of Events and Notices of Enforcement Discretion==
 
===.1 ===
{{IP sample|IP=IP 71153}}
a. (Closed) Licensee Event Report 05000282/2010
-004:  Battery Charger Inoperability due to Potential Undervoltage Conditions The inspectors reviewed the licensee's response to discovering that safety
-related battery chargers installed in 1994 were susceptible to failure during certain design basis accidents. Specifically, these battery chargers had the potential to stop providing an output, or "lock
-up," if their alternating current input voltage dropped below the nameplate minimum voltage of 90 percent at the battery charger motor control center (MCC). This item was documented as an unresolved item in NRC Inspection Report 05000282/2010005; 05000306/201005.
 
Documents reviewed during this inspection are listed in the Attachment to this report.
 
Inspection Scope This event follow
-up review constituted one sample as defined in Inspection Procedure 71153-05. b. See Section 4OA5.1 below for a discussion of this issue.
 
Findings
{{a|4OA5}}
==4OA5 ==
===.1 Other Activities===
 
a. (Closed) Unresolved Item 05000282/2010005
-05; 05000306/2010005
-05:  Potential for Common Mode Failure of Safety
-Related Battery Chargers The inspectors reviewed the circumstances surrounding the licensee's failure to maintain the both direct current (DC) electrical power subsystems operable during reactor operation in Modes 1 through 4.
 
Inspection Scope
 
====b. Findings====
IntroductionDescription
:  In NRC Inspection Report 05000282/2010005; 05000306/2010005, the NRC documented several issues regarding the safety
-related battery chargers, specifically with the 12 battery charger "locking up" during a simulated loss of offsite power (LOOP) event concurrent with a simulated loss of coolant accident (LOCA).
 
In the same inspection report, the NRC opened an unresolved item to address the potential for a common mode failure of all of the safety
-related battery chargers.
 
:  An apparent violation of Technical Specification (TS) 3.8.4, "DC Sources - Operating ," was identified by the inspectors due to the licensee's failure to maintain both DC electrical power subsystems operable during reactor operation in Modes 1 through 4.
 
4 Enclosure The inspectors reviewed the licensee's evaluation of the potential for a common mode failure. The evaluation contained information that the safety
-related battery chargers had the potential to stop providing an output, or "lock
-up", if the input alternating current (AC) voltage dropped below the nameplate minimum voltage of 90 percent of 480 Volts at the battery charger motor control center (MCC).
 
Specifically, the NRC learned that the lock-up of the battery charger was related to the operation of the silicon controlled rectifiers (SCRs) on the battery charger control circuitry.
 
As a low voltage condition occurred in response to the simulated LOOP/LOCA, the firing angle of the SCRs advanced to maintain output voltage.
 
If the firing angle advanced too far on a low voltage condition, the control circuitry became reverse biased and unable produce any output. The licensee was unable to determine the exact voltage, the duration of voltage dip, and the battery charger loading conditions which caused the lock
-up to occur.
 
In reviewing plant data from the periodic LOOP/LOCA tests, the inspectors determined that at certain points in the loading sequence the input voltage to the battery chargers decreased to less than 90 percent, the design minimum, for all four chargers (two on Unit 1 and two on Unit 2).
 
In addition, the licensee further determined that the LOOP/LOCA tests did not include all possible loads, including the 121 motor
-driven cooling water pump, and other loads such as an instrument air compressor. These loads could further decrease 480V bus voltage and contribute to the battery charger locking up.
 
On October 22, 2010, the licensee completed an operability evaluation and concluded that the chargers could be considered operable but non
-conforming if compensatory measures were put in place.
 
These compensatory measures included revising abnormal and emergency operating procedures, placing copies of needed abnormal operating procedures and tools needed within the battery charger rooms, and establishing a specific designated operator (with no other duties) to perform the manual actions needed to recover the battery chargers if needed.
 
At the end of 2010
, the licensee completed a past operability review of the safety-related battery chargers and concluded that there was reasonable doubt that the chargers would have performed their safety function if called upon during specific design basis accidents.
 
This was documented in LER 2010 00 submitted at the end of January 2011. As a result, the inspectors concluded that the DC electrical power subsystems (specifically the safety
-related battery chargers) had been inoperable since their initial installation in December 1994.
 
In May 2011, the licensee replaced and tested both Unit 1 battery chargers. The licensee planned to replace the Unit 2 battery chargers during the next refueling outage. The licensee's compensatory actions remain in place for Unit 2.
 
=====Analysis:=====
The inspectors determined that the licensee's failure to ensure that the DC electrical power subsystems remained operable during reactor operation in Modes 1 through 4 was a performance deficiency that required an evaluation using th e Significance Determination Process (SDP) described in NRC Inspection Manual Chapter (IMC) 0609. The inspectors also determined that this finding should be assigned to the Mitigating Systems Cornerstone because it impacted systems used in short term and long term heat removal.
 
The inspectors performed a Phase 1 SDP analysis and concluded that the finding represented the actual loss of safety function of 5 Enclosure a single train for greater than its TS allowed outage time which required a Phase 2 SDP evaluation
.
 
The Phase 2 SDP result was potentially greater than very low safety significance. The exposure time was assumed to be 1 year since the battery chargers have been susceptible to failure since they were installed. For the SDP, the initiating events that could result in one or more battery charger failures were determined to be those events where input AC voltage at the battery charger MCC would be less than 90 percent and charger output demand would be high. These initiating events were any non
-station blackout (SBO) LOOP event and any event that resulted in a safety injection (SI) signal. These events included small LOCAs, medium LOCAs, large LOCAs, stuck
-open power operated relief valves (PORVs), a steam generator tube rupture (SGTR), and a main steam line break.
 
The pre-solved SDP worksheets modeled the "Train A" or 11 battery charger. Consistent with the SDP usage rules defined in IMC 0609A, "Determining the Significance of Reactor Inspection Findings
- At Power ," the pre-solved worksheet assumed that a finding involving a battery charger would increase the Loss of DC initiating event frequency. However, this finding only involved failure of the battery charger in response to specific initiating events and would not increase the Loss of DC initiating event frequency. Therefore, the worksheets were individually solved assuming that one train of mitigating equipment would be failed as a result of a battery charger failure. A recovery credit of "1" was applied in all sequences because recovery of the battery charger was possible.
 
The dominant sequence was a LOOP, a failure of all auxiliary feedwater, and the failure of feed and bleed.
 
A Region III Senior Reactor Analyst (SRA) conducted a Phase 3 SDP evaluation to provide a more realistic estimate of the change in core damage frequency (CDF) for the finding. Similar to the Phase 2 SDP evaluation, the exposure time was 1 year and the initiating events that could result in one or more battery charger failures was assumed to be either a non
-SBO LOOP event or an event that resulted in an SI signal. Loss of coolant accidents and SGTR were considered to be the events that would result in an SI signal. The Standardized Plant Analysis Risk (SPAR) model for Prairie Island, Revision 8.15 , was used in the evaluation. The SPAR model represents Prairie Island Unit 1 only, so the results of the Unit 1 evaluation were generally assumed to be applicable to Unit 2. The model was modified to (1) require battery charger operation for DC system success in non
-SBO events because the safety-related batteries cannot function for the entire 24 hour mission time and (2) to account for the potential for common cause failure (CCF) of the battery chargers. The model was solved assuming one charger would fail in response to the applicable initiating events and the opposite train charger had the potential to fail. The dominant cut
-sets were reviewed and the potential for recovery of the failed battery charger was evaluated and applied at the sequence level.
 
The inspectors performed a review of the licensee's abnormal operating procedures and determined that a locked
-up charger could be recovered by locally turning the charger off and then turning it back on. However, the operators would be required to diagnose that the charger ha d locked-up and to restart the charger prior to the safety
-related battery depleting. The DC battery depletion times for each of the batteries were variable due to loading differences. In addition, the depletion times were uncertain due to assumptions regarding operation of equipment in response to initiating events.
 
6 Enclosure In general, the battery life estimates for the Unit 1 batteries were shorter than Unit 2 batteries in non
-SBO LOOP events, due to differences in battery loading. Also, for both units, the battery life for non
-LOOP events was generally longer because AC power was available to carry emergency lighting loads.
 
NUREG/CR-6883, "The SPAR
-H Human Reliability Analysis Method," was used to estimate the human error probability (HEP) for the failure to recover a battery charger prior to safety
-related battery depletion. For this HEP, the SRA considered both diagnosis and action failures but determined that diagnosis was the dominant failure mode. In response to LOOP events, the only performance shaping factor (PSF) that was considered a "performance driver" was stress. During this event operators would be receiving many alarms in the control room and would have a high workload in responding to the event.
 
The initial alarm indicating battery charger failure received in the control room would be "DC System Trouble."  The alarm response procedure instruct ed operators to check the "DC Panel Undervoltage" alarm and if it was also lit to proceed to Abnormal Operating Procedure 1C20.9 AOP3 or 1C20.9 AOP4, "Failure of 11(12) Battery Charger
."  The inspectors and the SRA determined that the "DC System Trouble" alarm would be expected to come in during LOOP events even if the battery charger functioned properly. After some period of time, with all equipment functioning normally, the alarm would clear. If the battery charger failed to function, the alarm would remain lit. The SRA considered that during an event, operators would likely prioritize other alarms associated with the initiating event and other potential complications before attending to the "DC System Trouble" alarm.
 
Once operators entered the Abnormal Operating Procedure (AOP), an operator in the plant would record data associated with the DC system in the battery charger room in order to determine that one or more battery chargers had locked up. Once diagnosis ha d occurred, the operator would reset the charger by opening the input breaker, waiting 10 seconds, and then closing the input breaker. Stress was considered to be "high" given that the dominant sequences involve d a complicated LOOP. All other PSFs were considered nominal. The SRA also concluded that procedures exist ed for resetting the battery charger(s)and that adequate time exist ed for diagnosis and action.
 
Based upon this information, the SRA estimated an HEP for failure to recover the battery charger as 2.2E-2 for most LOOP sequences on Unit 1.
 
For all other events, (LOCAs and SGTR s, Unit 2 LOOP events, and Unit 1 LOOP events that do not involve failures other than battery charger failures) the SRA also considered "time available" to be a performance driver. For these events, the licensee's battery depletion calculations showed that much more time was available for operator to respond to a locked up battery charger(s)prior to safety-related battery depletion.
 
Therefore, "extra time" was considered for diagnosis of the condition. The HEP for failure to recover a battery charger in these scenarios was estimated to be 2.2E
-3. After reviewing the cut
-sets, the SRA determined that for certain LOOP events, recovery of offsite power within the battery depletion time would also mitigate the event by allowing power to be restored to the train that was unaffected by a failed battery charger. To account for this, the SRA applied a factor to the applicable Unit 1 LOOP sequences that represent ed the probability that the LOOP event exceeds 2.5 hours. This value was obtained from NUREG/CR
-6890, "Reevaluation of Station Blackout Risk at Nuclear Power Plants
," Table 3-3. The value is the composite probability of exceedance for all categories of LOOP events. For Unit 2 LOOP sequences, a factor that represented the probability that the LOOP event exceed ed 8 hours was used.
 
7 Enclosure The SRA determined that the risk contribution from seismic events and internal flooding events for this finding was negligible. For internal fire events, the SRA used the licensee's Individual Plant Examination of External Events (IPEEE) to consider fire-induced LOOP scenarios and concluded that there was a small contribution to the risk of the finding from Unit 2 fire scenarios.
 
The potential risk contribution to large early release frequency (LERF) was evaluated using IMC 0609 Appendix H, "Containment Integrity Significance Determination Process." Prairie Island is a 2
-loop Westinghouse pressurized water reactor with a large dry containment. Sequences important to LERF include d SGTR events and inter-system LOCA events. These were not the dominant core damage sequences for this finding and thus the risk significance due to LERF was evaluated to be of very low safety significance.
 
The total delta CDF for Unit 1 was estimated as 1.9E
-6/yr and for Unit 2, 5.2E
-7/yr. This represent ed a preliminary White finding for Unit 1 and a preliminary Green finding for Unit 2. The dominant core damage sequence cut
-set for Unit 1 was a LOOP event followed by common cause failure of the battery chargers and the failure to recover a battery charger. Other important cut
-sets involved a LOOP event, failure of a single battery charger, and failure of the opposite train emergency AC power sources. The Unit 2 results were dominated by the fire contribution which was not fully developed since the initial estimates for total delta CDF, including the fire contribution, were less than 1.0E-6/yr. The SRA reviewed a risk evaluation performed by the licensee for this finding. The licensee concluded that the delta CDF for both units was less than 1.0E
-6/yr (Green). The major differences between the NRC's SDP evaluation and the licensee's risk assessment were HEPs estimated for failing to recover a battery charger and the assessment of the potential for common cause failure of both chargers on a single unit. Both evaluations considered the potential for recovering a charger; however, the licensee's HEP estimate was more optimistic than the NRC's. The NRC believes the potential for operators to miss or misinterpret the alarms during diagnosis of the failed charger was higher than estimated by the licensee. With regard to the common cause failure potential, the licensee assumed that both battery chargers on a single unit could not lock
-up in response to the same initiating event. The NRC concluded that since all the battery chargers were of the same design and would be modeled as part of the same common cause component group in a PRA model, that it was appropriate to treat the potential for common cause failure of both battery charger trains probabilistically, consistent with the "failure memory approach" used in NRC risk assessments. In the risk assessment of inspection findings
, the "failure memory approach" models observed successful components as having a probability of failure rather than concluding that the component would always be successful.
 
The results of the SDP evaluation were sensitive to both assumptions on recovery and common cause failure and therefore, the NRC performed sensitivity evaluations to vary the "best
-estimate" assumptions.
 
In particular, the NRC considered higher probabilities of common cause failure due to concerns that the actual potential for common cause failure was under-represented by SPAR model.
 
The sensitivity evaluations varied the battery charger common cause failure probability and the human error probabilities used the analysis. The results of the sensitivity evaluations were generally higher than the "best
-estimate" SDP evaluation but overall supported the preliminary conclusion of a White finding for Unit 1 and a Green finding for Unit 2.
 
8 Enclosure Inspection Manual Chapter 0305, "Operating Reactor Assessment Program," Section 12.01 , states that the NRC may refrain from considering safety significant inspection findings in the assessment program for a design
-related finding in the engineering calculations or analysis, associated operating procedure, or installation of plant equipment if the following statements were true:
Old Design Issue Review  The issue was licensee
-identified as a result of a voluntary initiative such as a design basis reconstitution; The performance issue was or will be corrected within a reasonable period of time following identification; The issue was not likely to have been previously identified by routine efforts such as normal surveillance or quality assurance activities; and The issue does not reflect a current performance deficiency associated with existing licensee programs, policy or procedures.
 
Based upon the information provided above, the inspectors have determined that this finding did not meet the criteria to be considered an old design issue for the following reasons:  The finding was not licensee
-identified as a result of a voluntary initiative. Although the licensee initiated a CAP document in late September 2010 regarding the possibility of charger lock up during grid voltage fluctuations, NRC prompting was needed and specifically requested during the October 2010 exigent TS change discussions to ensure that the licensee addressed the susceptibility of all chargers to a lock
-up condition during other design basis accidents.
 
The failure of the battery chargers to operate as expected following a design basis event was first discovered in 1996 during the performance of testing which simulated a LOOP/LOCA event. However, the licensee failed to recognize the significance of this issue and dispositioned the item as "use as is." As a result, the issue was not corrected within a reasonable period of time.
 
The finding was likely to be identified by past activities such as surveillance testing. Specifically, the licensee was unable to successfully perform the simulated LOOP/LOCA test following the 1994 battery charger installation. After performing at least two additional LOOP/LOCA tests which resulted in the lock-up of the 12 battery charger, the licensee ultimately changed the LOOP/LOCA test procedure to ensure that the 12 battery charger was turned off prior to performing the surveillance test.
 
No cross-cutting aspect was assigned to this finding
, because licensee decisions made in regards to evaluating the performance of the battery chargers were made many years ago and therefore, not reflective of current plant performance.
 
=====Enforcement:=====
Technical Specification 3.8.4, "DC Sources
- Operating," requires that the train A and train B DC electrical power subsystems be operable in Modes 1 through 4.
 
9 Enclosure With one battery charger inoperable, TS 3.8.4, Condition A, requires that the battery charger be restored to an operable status in 8 hours or that actions be taken to shut the plant down within the following 42 hours.
 
With both battery chargers inoperable, Limiting Condition for Operation (LCO) 3.0.3 requires that when an LCO is not met and the associated actions are not met, an associated action is not provided, or if directed by the associated actions, the unit shall be placed in a mode or other specified condition in which the LCO is not applicable. Action shall be initiated within 1 hour to place the unit, as applicable, in:
Mode 3 within 7 hours; Mode 4 within 13 hours; and Mode 5 within 37 hours.
 
Contrary to the above, from December 21, 1994, to approximately October 22, 2010, the safety-related battery chargers on both Unit 1 and 2 failed to maintain the DC electrical power subsystems operable in Modes 1 through 4. Specifically, under design basis accident conditions, all battery chargers were susceptible to a common cause failure under design basis accident conditions whereby the battery chargers would stop providing an output, or "lock
-up", when their AC input voltage dropped below their nameplate minimum voltage at the battery charger MCC. This is an apparent violation of TS 3.8.4 pending the completion of the final significance determination (AV 05000282/2011010
-01; 05000306/2011010
-01, Failure to Ensure that the Train A and Train B DC Electrical Power Subsystems Remained Operable in Modes 1 through 4)
.
{{a|4OA6}}
==4OA6 ==
===.1 Management Meetings===
 
O n May 20, 2011, the inspector s presented the inspection results to Mr. M. Schimmel, and other members of the licensee staff. The licensee acknowledged the issues presented. The inspectors confirmed that none of the potential report input discussed was considered proprietary.
 
===Exit Meeting Summary===
 
ATTACHMENT: 
 
=SUPPLEMENTAL INFORMATION=
 
Attachment
SUPPLEMENTAL INFORMATION KEY POINTS OF CONTAC
T
: [[contact::M. Schimmel]], Site Vice President
Licensee
: [[contact::K. Davison]], Plant Manager
: [[contact::T. Allen]], Site Engineering Director
- Acting
: [[contact::J. Anderson]], Regulatory Affairs Manager
: [[contact::C. Bough]], Chemistry and Environmental Manager
: [[contact::B. Boyer]], Radiation Protection Manager
: [[contact::K. DeFusco]], Emergency Preparedness Manager
: [[contact::D. Goble]], Safety and Human Performance Manager
: [[contact::J. Hamilton]], Security Manager
: [[contact::J. Lash]], Nuclear Oversight Manager
: [[contact::M. Milly]], Maintenance Manager
: [[contact::J. Muth]], Operations Manager
: [[contact::S. Northard]], Recovery Manager
: [[contact::A. Notbohm]], Performance Assessment Supervisor
: [[contact::K. Peterson]], Business Support Manager
: [[contact::A. Pullam]], Training Manager
: [[contact::R. Womack]], Outage Manager
: [[contact::J. Ritter]], Risk Analyst
: [[contact::J. Giessner]], Chief, Reactor Projects Branch 4
Nuclear Regulatory Commission
: [[contact::T. Wengert]], Project Manager, NRR
LIST OF ITEMS OPENED, CLOSED AND DISCUSSED
05000282/201
010-01; Opened 05000306/2011
010-01 AV Failure to Ensure that the Train A and Train B DC Electrical Power Subsystems Remained Operable in Modes 1 through
(Section 4OA5.1)
05000282/2010
-004 Closed LER Battery Charger Inoperability due to Potential Undervoltage Conditions
05000282/20
0005-05; 05000306/20
0005-05 URI Potential for Common Mode Failure of Safety
-Related Battery Chargers
Discuss ed  None.
Attachment
LIST OF DOCUMENTS REVIEWED The following is a partial list of documents reviewed during the inspection. Inclusion on this list does not imply that the NRC inspector reviewed the documents in their entirety, but rather that selected sections or portions of the documents were evaluated as part of the overall inspection effort. Inclusion of a document on this list does not imply NRC acceptance of the document or any part of it, unless this is stated in the body of the inspection report.
  - Risk Assessment of Operational Events RASP
Handbook; Volume 1 (Internal Events) and Volume 2 (External Events).
Section s 4OA3 and 4OA5 - The Prairie Island Standardized Plant Analysis Risk Model
- NUREG/CR-6890; Reevaluation of Station Blackout Risk at Nuclear Power Plants
- NUREG/CR-6883; The SPAR
-H Human Reliability Analysis Method - INL-EXT-10-18533; SPAR
-H Step-by-Step Guidance; Revision 1
- V.SPA.10.013; Battery Depletion Calculation; November 4, 2010
- V.SPA.11.001; Evaluation of Battery Charger Operation for a Loss of Offsite Power (LOOP) Event; Revision 0; January 17, 2011
- V.SPA.11.002; Evaluation of Battery Charger Operation for a Safety Injection Event While on
Offsite Power; February 25, 2011
- V.SPA.11.003; Prairie Island Battery Depletion Study PRA LOOP with Emergency Lighting and ISI Steady State Test Loads; Revision 0; February 16, 2011
- V.SPA.11.004; Prairie Island PRA SI Only Battery Depletion Study; Revision 0;
February 3, 2011 - V.SPA.11.008; Evaluation of Battery Charger Operation During Bus Crosstie Operation; Revision 0; March 7, 2011
- V.SPA.11.012; Battery Charger Significance Determination Process Fault Tree Analysis";
Revision 0; March 23, 2011
- V.SPA.11.013; Battery Charger Significance Determination Process Accident Sequence
Analysis; Revision 0; March 22, 2011
- V.SPA.11.014; Battery Charger Significance Determination Process Human Reliability Analysis; Revision 0; March 22, 2011
- V.SPA.11.015; Battery Charger Significance Determination Process Quantification Analysis;
Revision 0; March 24, 2011
- V.SPA.11.018; Battery Charger Significance Determination Process Accident Sequence Analysis (121 Cooling Water Pump Sensitivity); Revision 0; March 29, 2011
- V.SPA.11.019; Battery Charger Significance Determination Process Human Reliability
Analysis (121 Cooling Water Pump Sensitivity); Revision 0; March 31, 2011
- V.SPA.11.020;
Battery Charger Significance Determination Process Quantification Analysis (121 Cooling Water Pump Sensitivity); Revision 0; March 31, 2011
- Work Order 9712763; "12 Battery Charger Test during SP 1083"
- CAP 19971622; Intermittent Operation during SP 1083
; December 5, 1997
- CAP 19960452; 12 Battery Charger Intermittent Operation During SP 1083; February 22, 1996
- CAP 1250561; Battery
Chargers may stop
Operating if Undervoltage
Setpoint is Reached; September 21, 2010 - CAP 1252265; Questions Related to Operability
Review and Reportability for CAP 1238842; September 30, 2010 - CAP 1253478; Concerns with the Operability Review from CAP 1238842 on 12 Battery Charger; October 9, 2010
- CAP 1254359; Compensatory Measures not Evaluated Properly; October 16, 2010
 
Attachment
- CAP 1238842; CDBI 2010 Prep SP 1083
Revised without
Proper 50.59
Screening; June 24, 2010 - CAP 1270104; Non
-conservative Assumption in Unit 1 Battery Calculations; February 9, 2011
- Operability Review 1238842
-01; Continued Operability of D2 Emergency Diesel Generator due to Testing Question; October 22, 2010
- Operability Review 1250561
-02; Continued Operability of Safety
-Related Battery Chargers; October 22, 2010 - Alarm Response Procedure C47024; 12 DC System Trouble; Revision 35
- 1C20.9 AOP4; Failure of 12 Battery Charger; Revision 010
-A - 1C20.9 AOP3; Failure of 11 Battery Charger; Revision 9
- 1C20.5 AOP 1; Re
-energizing 4.16 KV Bus 15; Revision 12
- 1C20.5 AOP2; Re
-energizing 4.16 KV Bus 16; Revision 14
- 1C20.5 AOP4; Reenergizing 4.16 KV Bus 15 Via Bus
-Tie Breakers;
Revision 3W
- 1C20.5 AOP5; Reenergizing 4.16 KV Bus 16 Via Bus
-Tie Breakers; Revision 3W
 
Attachment
LIST OF ACRONYMS USE
D AC Alternating Current
ADAMS Agencywide Document Access Management System
AOP Abnormal Operating Procedure
AV Apparent Violation
CAP Corrective Action Program
CCF Common Cause Failure
CDF Core Damage Frequency
CFR Code of Federal Regulations
DC Direct Current
DRP Division of Reactor Projects
HEP Human Error Probability
IMC Inspection Manual Chapter
IPEEE Individual Plant Examination of External Events LCO Limiting Condition for Operation
LER Licensee Event Report
LERF Large Early Release Frequency
LOCA Loss of Coolant Accident
LOOP Loss of Off
-Site Power MCC Motor Control Center
NRC U.S. Nuclear Regulatory Commission
NRR Office of Nuclear Reactor Regulation PARS Publically Available Records System
PORV Power Operated Relief Valve
PSF Performance Shaping Factor
SBO Station Blackout
SC R Silicon Controlled Rectifier
SDP Significance Determination Process
SGTR Steam Generator Tube Rupture
SI Safety Injection SPAR Standardized Plant Analysis Risk
SRA Senior Reactor Analyst
TS Technical Specification
URI Unresolved Item
 
M. Schimmel
    -3-  In accordance with
CFR 2.390 of the NRC's "Rules of Practice," a copy of this letter, its enclosure, and your response (if any) will be available electronically for public inspection in the NRC Public Document Room or from the Publicly Available Records System (PARS) component of NRC's document system (ADAM S). ADAMS is accessible from the NRC Website at http://www.nrc.gov/reading
-rm/adams.html
(the Public Electronic Reading Room).
 
Sincerely,  /RA/
Steven West, Director
Division of Reactor Project
s  Docket Nos.:
50-282; 50-306; 72-010 License Nos.:  DPR-42; DPR-60; SNM-2506  Enclosure:
Inspection Report 05000282
/2 011010; 050003 06/2 011010  w/Attachment:  Supplemental Information
cc w/encl:
Distribution via ListServ
DOCUMENT NAME:
G:\DRPIII\PRAI\Prairie Island 2011 010 Greater than Green Rpt.docx
Publicly Available
Non-Publicly Available
Sensitive  Non-Sensitive To receive a copy of this document, indicate in the concurrence box "C" = Copy without attach/encl "E" = Copy with attach/encl "N" = No copy
OFFICE  RIII    RIII    RIII    RIII    NAME  JGiessner:dtp
PLougheeed for SOrth  LKozak  SWest  DATE  06/06/11  06/06/11  06/06/11  06/09/11 OFFICIAL RECORD COPY
 
Letter to M. Schimmel from S. West dated
June 9, 2011
SUBJECT: PRAIRIE ISLAND NUCLEAR GENERATING PLANT, UNITS 1 AND 2, NRC INSPECTION REPORT
05000 282/2011010; 050003 06/2 011010 PRELIMINARY WHITE FINDING DISTRIBUTION
: Daniel Merzke
RidsNrrPMPrairieIsland Resource
RidsNrrDorlLpl3
-1 Resource
RidsNrrDirsIrib Resource
Cynthia Pederson
Steven Orth
Jared Heck
Allan Barker
Carole Ariano Linda Linn
DRPIII DRSIII Patricia Buckley
Tammy Tomczak
ROPreports Resource
}}
}}

Revision as of 14:57, 18 September 2018

Prairie Island. Units 1 & 2, Supplemental Information Regarding Inspection Report 05000282/2011010; 05000306/2011010 (EA-11-110)
ML11195A336
Person / Time
Site: Prairie Island  Xcel Energy icon.png
Issue date: 07/13/2011
From: Schimmel M A
Northern States Power Co, Xcel Energy
To:
Document Control Desk, Office of Nuclear Reactor Regulation
References
EA-11-110, ir-11-010, L-PI-11-071
Download: ML11195A336 (43)
v • d • e


Text

XcelEnergy" JUL 1 3 2011 L-PI-1 1-071 U S Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555-0001 Prairie Island Nuclear Generating Plant Units 1 and 2 Dockets 50-282 and 50-306 Renewed License Nos. DPR-42 and DPR-60 Supplemental Information Reqarding NRC Inspection Report 05000282/2011010:

05000306/2011010 (EA-11-110)

By letter dated June 9, 2011 (Agencywide Documents Access and Management System (ADAMS) Accession Number ML111610249), the NRC transmitted Inspection Report 05000282/2011010; 05000306/2011010 for the Prairie Island Nuclear Generating Plant (PINGP). In the inspection report, the NRC identified one finding and apparent violation related to battery charger performance with a preliminary significance of White for Unit 1 and a preliminary significance of Green for Unit 2, and provided the opportunity to request a Regulatory Conferenc By letter dated June 14, 2011 (ADAMS Accession Number ML111662049), Northern States Power Company, a Minnesota corporation, doing business as Xcel Energy (hereafter "NSPM") requested a Regulatory Conference which will be held on July 28, 2011.This letter transmits documents which support the July 28, 2011 Regulatory Conferenc Enclosure 1 provides document V.SPA. 11.016, "Simulator Runs for Battery Charger Significance Determination Process".

Enclosure 2 provides a document entitled, "Battery Depletion Calculation Load Profile".

Enclosure 3 provides document V.SPA. 11.006, "Evaluation of Battery Depletion Differences".

If there are any questions or if additional information is needed, please contact Mr. Dale Vincent, P.E., at 651-388-1121.

  • 1717 Wakonade Drive East * Welch, Minnesota 55089-9642 Telephone:

651.388.1121 Document Control Desk Page 2 Summary of Commitments This letter contains no new commitments and no revisions to existing commitments.

Mark A. Schimmel Site Vice President, Prairie Island Nuclear Generating Plant Northern States Power Company -Minnesota Enclosures (3)cc: Administrator, Region III, USNRC Project Manager, PINGP, USNRC Resident Inspector, PINGP, USNRC ENCLOSURE1 V.SPA.11.016 SIMULATOR RUNS FOR BATTERY CHARGER SIGNIFICANCE DETERMINATION PROCESS 15 pages follow C Xcel Energy Prairie Island Nuclear Generating Plant Probabilistic Risk Assessment V.SPA.1 1.016 Simulator Runs for Battery Charger Significance Determination Process March 2011 Document Review Form Document or Analysis Title: V.SPA.11.016-

Simulator Runs for Battery Charger SDP Revision of Document or 0 Analysis: Date: 3/11/2011 Preparer:

,,,t- Date: 3/11/2011 Mark A. Brossart ,1 Reviewer: Jayne E. Ritter Date: 3/11/2011 PRA Acceptance Date: 6/07/2011 Ashley Peterman V.SPA.11.016 Simulator Runs for Battery Charger Significance Determination Process Table of Contents Page Number T able of C ontents .............................................................................

1.0 P u rp o se ....................................................................................

2.0 Simulator Development Process ........................................................

3.0 S im ulator R uns ............................................................................

3.1 Sim ulator "Initial R un" R esults .....................................................

3.2 Simulator "Follow-up Run" Results ..............................................

4 .0 C o nclu sio n s ................................................................................

5 .0 R e fere nce s ..................................................................................

1 2 2 3 3.5 7 8

Attachments:

Attachment 1 -Simulator Prior Knowledge Statement

..................................

(2 pages)Attachment 2 -Simulator Scenario Security Agreement

...............................

(2 pages)File Prepared By: Mark Brossart Date: 3/11/2011 File Reviewed By: Jayne Ritter Date: 3/11/2011 V.SPA.1 1.016 Simulator Runs for Battery Charger Significance Determination Process Page 1 of 8 V.SPA.11.016 Simulator Runs for Battery Charger Significance Determination Process 1.0 Purpose: A simulator scenario was developed to support the arguments used in the Significance Determination Process for the Battery Charger past operabilit The results of the scenario will not be used as part of the PRA success strategy but may be used to demonstrate, in a simulated environment, that there is adequate time with significant margin to perform the recovery actions credited in the PRA success strateg It is intended that site management may use this information in discussing the site response with the Regulator as a demonstration that the PRA recovery strategy assumptions are reasonable.

The simulator run will determine the time it takes operators to: 1. Diagnose and respond to re-energizing a safeguards 4.16KV bus via the opposite unit bustie breaker during a LOOP event.2. Perform the actions necessary to restart the battery charger following a lock up.Information pertaining to this calculation folder can be found in the following directory:

J:\PRA\Calc Folder Writeups\SPA Folders\ V.SPA.11.016

-Simulator Runs for Battery Charger SDP 2.0 Simulator Development Process: 1) The simulator scenario is based on the limiting accident sequence developed for the Battery Charger SDP (Reference 5.3). The scenario includes the following conditions:

a. Dual unit LOOP (Initiating Event).b. The 12 Battery Charger experiences a lock up during the load sequencing restoration process.c. The Emergency Diesel generator (EDG) Dl experiences a lockout. As a result of the LOOP and D1 failure, Bus 15 will not be energize This will cause the emergency lighting to transfer from its normal AC source to the 12 Battery (DC source) resulting in a more rapid battery depletion and minimizing the allowed operator response time. (Reference 5.1)d. The inverters will fail to the DC source as a result of a known issue with voltage overshoot associated with the D1ID2 voltage regulator This will result in more rapid battery depletion and minimizing the allowed operator response time.e. The Non-License Operator (NLO) used to restore the battery charger is physically located in a location that was expected for the scenario (Screenhouse for DDCLP monitoring)

but trying to maximum travel time to the battery rooms.2) Sequence of Events: Dual Unit LOOP Both units experience a simultaneous LOOP. D1 is locked out causing Bus 15 to not be immediately available to power safeguards loads. As a result of the 12 Battery Charger's susceptibility to lock up, the DC loads for B Train will be powered from the 12 Battery as 12 Battery Charger locks up during the load sequence restoration process. See PRA Calculation Folder V.SPA. 11.003 for details on the impact on the 12 Battery depletion time. In order to maximize the time for the NLO to respond to the battery room, the NLO was sent to monitor the Diesel Driven Cooling Water Pumps (DDCLPs) in the Plant Screenhouse.

V.SPA.1 1.016 Page 2 of 8 Simulator Runs for Battery Charger Significance Determination Process 3) Expected Crew Responses:

a. Enter IE-0 and initiate restoration of Bus 15 at Step 3 per 1C20.5 AOPI (Reference 5.4).b. Perform actions necessary to restore the 12 Battery Charger per the Alarm Response Procedure ARP C47024-1105 (Reference 5.5) and 1C20.9 AOP4 (Reference 5.6).4) Positioning of NLO: The decision was made to pre-stage the NLO initially in the Records Room and then upon the announcement of Reactor Trip dispatch him to monitor the Diesel Driven Cooling Water Pumps.Alternate locations were evaluated but discounted for the following reasons. With a Loss of Offsite Power event the assumption was that the duty crew would dispatch operators to monitor all running diesels. If the NLO was doing Attachment J of IESO. I (Reference 5.7), he would have been in closer proximity to make copies of procedures and go to the battery rooms. The D I or D2 rooms were considered but it was determined it would take less time to get to the battery rooms from the Dl or D2 rooms considering that the operator would need to make a copy of the procedure and would use the water treatment office to do so. Remote locations in the auxiliary building were not considered since the Turbine Building Operators would not have been utilized or sent to the auxiliary building tasks during this scenario.5) Two simulator runs were performed:

The initial run was performed on March 5 th, 2011 and a follow-up run on March 8 h, 2011. The purpose of the initial run was to demonstrate that the assumptions made in the PRA success strategy are reasonable with normal Control Room staffing including a Shift Manager (SM), Shift Supervisor (SS) and Shift Technical Advisor (STA). Also included were two Unit 1 Reactor Operators (RO), 1 Unit 2 RO and a Unit 1 Turbine Building NLO. Following the initial run for this scenario, it was determined that a more limiting scenario would be to assume that Unit 2 was in a sufficiently complex recovery that the extra Unit 2 RO would not be available to support the Unit 1 response.Consequently, the follow-up run was performed with a crew of a Shift Manager, Shift Supervisor, two Unit I ROs and a Unit 1 Turbine Building NLO 6) Simulator Run Data Collection:

Simulator timing data was collected for both simulator runs. There were three individuals collecting timing data (PRA Engineer in booth, Operations SM in booth, Engineering Supervisor in the plant).3.0 Simulator Runs 3.1 Simulator "Initial Run" Results (March 5th% 2011)The initial simulator run demonstrated the following risk-significant operator actions.a. The restoration of Bus 15 from Bus 25 via bustie breaker per 1C20.5 AOP4 (Reference 5.8)occurred at approximately 24 minutes into the event.b. The restoration of the battery charger per 1C20.9 AOP4 (Reference 5.6) occurred at approximately 43 minutes into the event.These actions were demonstrated in an initial simulator run performed on March 5, 2011. The crew consisted of Darrel Lapcinski (SM), Mark Davis (SS), Jim Kapsh (U1 RO), Mark Jenkin (STA), Mike Murphy (U1 Lead RO), Evan Bystrom (UI NLO) and Joe Julian (U2 Lead RO).See Table 1 for a detailed event description and tining for this simulator preliminary run.V.SPA. 11.016 Page 3 of 8 Simulator Runs for Battery Charger Significance Determination Process ThhIA 1-~imsIIAtnr fuRl unit LOOP Fv~nt ~wlth EDfl Dl Lnnk-niit~

~r~An,~arin

~/fl~/~flhl Time from Time Start of Event Event/Action q 8:50:15 0:00:00 Trigger 1 -LOOP event (Dual-unit)

with EDG D1 Lock-out 8:50:15 0:00:00 Enter 1E-0 8:50:45 0:00:30 Unit 1 RX Trip Announced to Plant Immediate Actions per 1 E-0 completed, Bus 15 is de-energized and 8:51:15 0:01:00 was reported to the Unit 1 SS.8:52:10 0:01:55 13 Charging Pump started Unit 1 SS instructed Unit 1 Lead RO to enter 1C20.5 AOP1 per 1 E-0, 8:53:10 0:02:55 Step 3 RNO 8:53:32 0:03:17 Verified that SI is not required Unit 2 Lead RO instructed to perform actions in 1C20.5 AOP1 per 8:54:00 0:03:45 Unit 1 SS/Lead RO Discussion of RCS condition between Unit 1 SS and RO (potential de-pressurization of RCS and LOCA conditions, watch for SI 8:55:00 0:04:45 actuation conditions)

8:55:45 0:05:30 Transition to 1 ES-0.1 8:58:30 0:08:15 Dispatch outplant operator to check EDG D1 and D2, locally Crew Brief, SS restored normal alarm response protoco Then directed the ROs to perform a board walkdown and report any alarms 8:58:30 0:08:15 that are unexpected for the LOOP/reactor trip.8:59:52 0:09:37 Sump C and RHR pit high level alarms were identified as abnormal.RCDT indicated zero. This was addressed as being caused by the 9:00:51 0:10:36 loss of Bus 15.Unit 1 Lead performed board walkdown and reported that Bus 11, 12, 9:01:00 0:10:45 13 and 14 are de-energized Per Step 2.4.2.B of 1C20.5 AOP1, report from outplant operator that the Bus 15 Undervoltage signal was lit and Bus 15 was not locked 9:01:45 0:11:30 out.A discussion about the 12 battery charger being de-energized with 9:02:30 0:12:15 bus 16 available and at power was conducted.

9:03:00 0:12:45 Started 122 CR Chiller and fans Instructed Turbine building Operator to investigate 12 battery Trouble 9:03:50 0:13:35 alarm.SM announced that Alert was declare Escalation criteria would 9:03:55 0:13:40 occur if Bus 16 was lost.9:06:30 0:16:15 SM makes plant announcement that an Alert was declared Dispatch outplant operator to investigate why 12 Battery is Dischargin Outplant operator was located in the 12 DDCLP room 9:08:00 0:17:45 (Screenhouse)

Crew Brief. The crew is working through ES-0.1. The inter-unit 9:10:40 0:20:25 cross-tie to bus 15 is nearly complete.Report back from TB operator on status of 12 Battery -12 Battery is 9:11:30 0:21:15 discharging.

TB Operator proceeded to the Water Treatment office to get a copy of 9:12:00 0:21:45 1 C20.9 AOP4.9:13:25 0:23:10 U2 RO announced that he was ready to cross-tie bus 15.Unit 2 Bus 25 connected to Bus 15. Bus 15 re-energized per 1C20.5 9:14:00 0:23:45 AOP 4.V.SPA.1 1.016 Page 4 of 8 Simulator Runs for Battery Charger Significance Determination Process Table 1-Simulator Dual Unit LOOP Event (with EDG D1 Lock-out)

Scenario -3/05/2011 Time from Time Start of Event Event/Action TB operator reported that he is using 1C20.9 AOP4 and investigating 9:16:00 0:25:45 12 Battery Charger operation.

9:16:26 0:26:11 Announced that bus 15 loads can be restored.9:18:00 0:27:45 Control Room Update: Entered 1 C20.9 AOP4 for 12 Battery Charger 9:21:31 0:31:16 Began efforts to restore battery room cooling 9:22:34 0:32:19 Swapped charging pump suction from VCT to RWST Identified that letdown was not in service. Directed U2 RO to place 9:25:04 0:34:49 letdown in service.TB Operator reports back to control room information from Attachment A of 1 C20.9 AOP4. TB Operator recommended completing action in Attachment B of 1 C20.9 AOP4 (re-starting 12 9:27:00 0:36:45 Battery Charger)Reported the pressurizer level is 36% and rising with minimum 9:28:00 0:37:45 charging.Unit 1 SS directed TB operator to re-start 12 Battery Charger per 9:28:00 0:37:45 1C20.9 AOP4, Attachment B.9:30:00 0:39:45 Entered AOP to repower battery room cooling from D3 TB operator open AC input breaker for 12 Battery Charger, waited, 9:32:00 0:41:45 then reclosed breaker.Report from TB operator that 12 Battery Charger has been restored.Output is 250 amps. At Step 8 of 1C20.9 AOP4 and will continue to 9:33:00 0:42:45 monitor.9:33:50 0:43:35 Letdown established Unit 1 Lead RO confirmed with ERCS that 12 Battery Charger has 9:36:00 0:45:45 been restored.9:39:00 0:48:45 End of simulator run.3.2 Simulator "Follow-up Run" Results (March 8 th, 2011)The follow-up simulator run demonstrated the following risk-significant operator actions.a. The restoration of Bus 15 from Bus 25 via bustie breaker per 1C20.9 AOP4 occurred at approximately 17 minutes into the event.b. The restoration of the battery charger per 1C20.9 AOP4 occurred approximately at 51 minutes into the event.These actions were demonstrated in a follow-up simulator run performed on March 8, 2011. The crew consisted of Steve Ingalls (SM), Doug Larimer (SS), Greg Boek (UI RO), Jack Edwards (U I Lead RO)and Rick Lodermeier (Ul NLO).See Table 2 for a detailed event description and timing for this simulator preliminary run.V.SPA.1 1.016 Simulator Runs for Battery Charger Significance Determination Process Page 5 of 8

...Table 2-Simulator Dual Unit LOOP Event (with EDG D1 Lock-out)

Scenario -.3/08/2011 Time from Time Start of Event Event/Action 15:24 Crew has duty 15:35:06 0:00:00 Trigger 1 -LOOP event (Dual-unit)

with EDG D1 Lock-out 15:35:15 0:00:09 Enter 1E-0 15:35:31 0:00:25 Unit 1 RX Trip Plant Announcement Immediate Actions per 1 E-0 completed, Bus 15 is de-energized, SI 15:36:20 0:01:14 not actuated or required.15:36:33 0:01:27 1 E-0, Step 1 Reactor trip verified 15:37:21 0:02:15 1 E-0, Step 2 Turbine trip verified 1E-0, Step 3, Bus 16 is energized, Bus 15 has Undervoltage alarm 15:37:36 0:02:30 and is de-energized Turbine trip verified 15:38:00 0:02:54 12 DC System Trouble Alarm actuated on simulator Unit 1 SS Directs Unit 1 Lead RO to enter 1C20.5 AOP1 (Re-15:38:39 0:03:33 energizing 4.16KV Bus 15) to restore bus 15 15:39:00 0:03:54 Charging Pump in operation 15:39:03 0:03:57 1E-0, Step 4, Check SI actuated -no. Check if SI is required (RNO)15:40:03 0:04:57 DDCLPs running SAT Unit 1 Lead calls turbine building operator to investigate Undervoltage on Bus 15 protective relays per Step 2.4.2.B in 1C20.5 AOP1 (Re-15:40:44 0:05:38 energizing 4.16KV Bus 15)15:40:51 0:05:45 Transition to 1 ES-0.1 15:40:51 0:05:45 SM reports that all safety functions are GREEN 15:41:39 0:06:33 1 ES-0.1, Step 2 entered 15:42:42 0:07:36 1ES-0.1, Step 3 entered 15:45:00 0:09:54 1ES-0.1, Step 4 entered Turbine Building operator calls back control room to inform them that there are degraded and undervoltage relays on Bus 15 Load 15:45:00 0:09:54 Sequencer per 1C20.5 AOP1 Step 2.4.2.B 15:46:00 0:10:54 Alert declared, 577 completed 15:46:27 0:11:21 1 ES-0.1, Step 5 entered Unit 1 RO calls turbine building operator to perform 1 ES-0.1, 15:46:55 0:11:49 Attachment J (Isolate Unit 1 Moisture Separator Reheaters)

15:47:00 0:11:54 SS and Lead RO discuss re-energizing Bus 15 from crosstie 15:47:34 0:12:28 1ES-0.1, Step 6 entered 15:52:00 0:16:54 Cross-tie between Bus 15 and Bus 25 closed. Bus 15 is re-energized Unit 1 RO reports that the Reactor Makeup Flow deviation alarm is in.Discussed with Unit 1 SS and it was decided to open the RWST to 15:54:00 0:18:54 charging pump suction (MV-32060).

15:55:00 0:19:54 MV-32060 is opened.Unit 1 Lead RO is in ARP 47024-1105 for 12 DC System Trouble 15:55:00 0:19:54 Alarm 15:55:55 0:20:49 1ES-0.1, Step 7 entered 15:56:00 0:20:54 Normal alarm response protocol re-established 15:56:34 0:21:28 1ES-0.1, Step 8 entered V.SPA.1 1.016 Simulator Runs for Battery Charger Significance Determination Process Page 6 of 8 Table 2-Simulator'

Dual Unit;LOOP Event (with EDG D1 L"ck-0ut)

Scenario 13/08/2011 Time from,-Time Start of Event Event/Action Unit 1 Lead RO Recognized that 12 Battery is dischargin Battery is supplying load. This was discussed with Unit 1 SS and Lead RO was told to instruct outplant operator to look at 12 battery 15:58:45 0:23:39 charger.Unit 1 Lead RO call outplant operator to investigate 12 battery 15:59:21 0:24:15 charger 15:59:27 0:24:21 1 ES-0.1, Step 8.b -RO instructed to place letdown into service Outplant Operator called from 12 Battery Room -Information transmitted to Lead RO was: 0 amps DC output (charger), 117 V DC, 16:06:09 0:31:03 Low Voltage Lit, and AC Light ON.12 Battery Charger information from ouptlant operator was discussed 16:07:21 0:32:15 with Unit 1 SS 16:08:33 0:33:27 Entered 1 C20.9 AOP4 (Failure of 12 Battery Charger)1C20.9 AOP4, Attachment A, dispatched to Turbine Building 16:10:00 0:34:54 Operator 16:12:27 0:37:21 1ES-0.1, Step 9 entered 16:13:57 0:38:51 1ES-0.1, Step 10 entered 16:14:55 0:39:49 1ES-0.1, Step 11 entered 1 ES-0.1, Step 11.a.RNO.b, RO calls outplant operator to open VC-1 -16:17:21 0:42:15 1 16:19:49 0:44:43 1ES-0.1, Step 11.C entered Unit 1 Lead RO instructed outplant operator to restart 12 Battery 16:22:00 0:46:54 Charger per 1C20.9, AOP4, Attachment B.Per 1 ES-0.1, Step 11 .C.RNO.C, Unit 1 SS discussed that battery 16:23:50 0:48:44 room cooling must be placed in service per C20.16 AOP1.16:24:00 0:48:54 Discussed using TSC support staff for re-start Battery Room Cooling 16:26:00 0:50:54 Restarted 12 Battery Charger 16:28:00 0:52:54 Used ERCS to confirm 12 Batter Charger was restarted.

16:30:00 0:54:54 End of simulator run.4.0 Conclusions The purpose of these simulator runs was to demonstrate that the actions assumed in the PRA success strategy were reasonable given the time availabl The limiting depletion time for the 12 Battery for this event is just over three hours assuming Bus 15 remains de-energized (PRA Calculation V.SPA. 11.003).The depletion is much slower once Bus 15 is restored and the emergency lights are no longer a load on the 12 Battery (PRA Calculation V.SPA.1 1.009). The simulator runs demonstrated that Bus 15 can be re-energized via the bustie breaker in less than 30 minutes during a LOOP event and the battery charger can be restored in less than an hour. Once the charger is restored the event is over. The assumptions in the PRA success strategy can be concluded to be reasonable.

V.SPA.1 1.016 Simulator Runs for Battery Charger Significance Determination Process Page 7 of 8 5.0 References 5.1 PRA Evaluation V.SPA. 11.003, Prairie Island Batter, Depletion Study PRA LOOP with Emergency Lighting and ISI Steady State Test Loads, Revision 0.5.2 PRA Evaluation V.SPA. 11.009, Prairie Island Batter, Depletion Study PRA LOOP with ISI Steady State Test Loads, Revision 0.5.3 PRA Evaluation V.SPA. 11.012, Batteny Charger SDP: Accident Sequence Analysis, Revision 0.5.4 1C20.5 AOP1, Re-energizing 4.16kV Bus 15, Revision 12.5.5 1C20.9 AOP4, Failure of 12 Battery Charger, Revision 10.5.7 1ES-0.1, Reactor Trip Recover,, Revision 25.5.8 1C20.5 AOP4, Reenergizing 4.16kV Bus 15 Via Bustie Breakers, Revision 3W.V.SPA.1 1.016 Page 8 of 8 Simulator Runs for Battery Charger Significance Determination Process AfKc6hment Page 1 ofA,-XceIEnergy" IPrior Knowledge Statement SIMULATOR SCENARIO:

Scenario performed in support of the Prairie Island Battery Charger SDP conducted on March 8h, 2011 and commehcing at approximately 1500 I confirm that I had no prior knowledge of the details used in the above simulator scenari Further, I confirm that I had no knowledge of the details of the simulator run performed on March 5t, 2011. I understand that had I had prior knowledge of the scenario it would impact the legitimacy and use of the information gathered in support of the Battery Charger SDP.PRINTED NAME JOB TITLE / RESPONSIBILITY SIGNATURE DATE NOTE 2.5.6.7.8.9.10.NOTES:

fa Pageof 1 SXcelEnergy-

Prior Knowledge Statement SIMULATOR SCENARIO:

Scenario performed in support of the Prairie Island Battery Charger SDP conducted on March 8th, 2011 and commencinq at approximately 1500 I confirm that I had no prior knowledge of the details used in the above simulator scenari Further, I confirm that I had no knowledge of the details of the simulator run performed on March 5t, 2011; I understand that had I had prior knowledge of the scenario it would impactthe legitimacy and use of the information gathered in support of the Battery Charger SDP.PRINTED NAME JOB TITLE / RESPONSIBILITY

_VNATURE DATE NOTE 2.3.4.5.6.7.8.9.9i0- T_______1___

10.NOT NOTES:.j AwIa.cXv~wo 2Z Page 1 of/A -I XceIEnergy SIMULATOR SCENARIO SECURITY AGREEMENT SIMULATOR SCENARIO:

Scenario performed in support of the Prairie Island Battery Charger SDP conducted on March 8e, 2011 and commencing at approximately 1500 During the post-critique I agreed that I would NOT discuss any aspects associated with the contents of this simulator run with any other operator until informed that it was acceptable to do so. I understand that violation of the conditions of this agreement may impact the legitimacy and use of the information gathered should subsequent runs of this scenario become necessary.

PRINTED NAME JOB TITLE / RESPONSIBILITY SIGNATURE DATE NOTE 1. .j _ .T-f)3. 1 t e 1kOA 3/5.6.7.8.9.10._NOTES:--

Page;kofT 4, SXeelEnergyl SIMULATOR SCENARIO SECURITY AGREEMENT SIMULATOR SCENARIO:

Scenario performed in support of the Prairie Island Battery Charger SDP conducted on March 8t". 2011 and commencing at approximately 1500 During the post-critique I agreed that I would NOT discuss any aspects associated with the contents of this simulator run with any other operator until informed that it was acceptable to do so. I understandthat violation of the conditions of this agreement may impact the legitimacy and use of the information gathered should subsequent runs of this scenario become necessary.

PRINTED NAME JOB TITLE / RESPONSIBILITY SIGNATURE DATE NOTE 2.3.4.5.6.7.8.9.10. _NOTES: U ENCLOSURE2 BATTERY DEPLETION CALCULATION LOAD PROFILE 3 pages follow Batterv DeDletion Calculation Load Profile Purpose: The purpose of this document is to summarize the basis for the load profile used to determine the battery depletion time at the Prairie Island Nuclear Generating Plant (PINGP). Battery depletion time was needed to support the risk assessment of the battery charger lock up condition that could occur for the old battery chargers during loss of offsite power and safety injection accident scenario For the purposes of the risk assessment evaluations, battery depletion is defined as the time for the batteries to reach minimum design voltage when supplying the required loads.Discussion:

Although battery depletion is defined as battery voltage reaching minimum design voltage when the batteries are supplying required loads, the risk assessment depletion calculations were developed as best-estimate evaluations so that the actual plant and operator responses to the analyzed events could be realistically evaluate As a result, use of design-basis inputs for DC system loading would not be appropriat Rather, a load profile based on actual, expected system response was develope The load profile used in the risk assessment depletion calculations consists of three parts: a first minute transient portion, a steady state portion, and a final minute transient portion. Development of each of the load profiles is discussed below.A one hour design-basis battery loading profile is developed in the design-basis battery sizing calculations to meet the battery sizing requirements in the Updated Safety Analysis Report (USAR). The purpose of the load profile developed to meet the USAR requirements is to determine a worst case and conservative load profile, thereby resulting in a conservative battery size and greater battery capabilit The load profile developed for the depletion calculations was modeled after the design-basis load profile with some adjustments to reflect expected system response.The design-basis load profile developed for the USAR battery sizing requirements contains a first minute transient portion, a steady state portion, and a final minute transient portion. However, the design-basis load profile and calculation contain conservative assumptions to ensure that the calculation and required battery size has adequate margin to the minimum size required by the USAR.In the design-basis load profile, the first minute transient portion of the load profile contains various changes in the DC load due to multiple breaker operations and device actuation The first minute profile modeled in the risk assessment battery depletion calculations is the same as the first minute profile used in the design-basis calculations with a slight adjustment downward for the inverter loads. The adjusted inverter loading was determined from reviewing historical operating data taken during regular inverter surveillance Use of the adjusted design-basis load profile continues to result in a conservative first minute profile in the depletion calculation The loading includes many smaller loads that are assumed energized in the calculation which would actually de-energize on an event. These are further described below.The steady state portion of the battery depletion load profile was modified from the design-basis load profile. As stated above, the design-basis load profile contains conservative assumptions to provide additional margin to actual system respons One of the major assumptions which results in an overly conservative steady state load in the design-basis load profile is that various loads such as lights, relays, solenoid valves, etc. are modeled as energized during the entire scenari While this assumption is appropriate for use in the design-basis evaluations that are to determine conservative battery sizes, this assumption is overly conservative for an evaluation of actual expected system respons As a result of these conservatisms, use of the design-basis steady state profile would result in much higher loading than what would be expected in a real event. This is not consistent with a best estimate evaluation.

A review of the major contributors of the steady state load estimated that the actual load would be reduced by approximately 50% or more from the design-basis steady state load values when actual equipment operation is considere This review identified various DC system components that are modeled as a constant energized load which would actually de-energize on an event or have only momentary actuation For example, solenoid valves which de-energize to their fail safe position are modeled as being energized for the entire scenari It was also identified that dual indicating lights (on/off, open/close, etc.) were both included as energized for the entire scenario when only one light would be energized at a time. Reactor protection relays were also modeled as energized for the entire scenario when the relays would de-energize on a reactor trip.The Integrated Safety Injection (SI) Test simulates an SI with a Loss of Offsite Power (LOOP).During this test, the various devices which are required to respond to an event are actuated.Correspondingly, the DC load profile during the Integrated SI Test is representative of the load profile that would be expected during an actual event. The loading that occurs during the Integrated SI test was compared to the actual load profile that occurred during a past inadvertent SI event. The two profiles were approximately equal, showing a difference of less than 2 amps.The steady state DC load value observed during the Integrated SI Test also includes additional loads, such as inverters, that are present on Unit 1 due to operating conditions that are not present in the current design-basis calculation Therefore, use of the worst case best estimate steady-state load profile from either the Integrated SI Test or the inadvertent SI event provides a slightly conservative but realistic representation of the expected system response to a LOOP or SI event.This steady state load value was then adjusted upward for use in the risk assessment depletion calculation This upward load adjustment was to account for swing loads that could be present in scenarios evaluated in the risk assessment but would not be seen in the Integrated SI Test.The swing loads accounted for an additional load of approximately 4 -6 amps depending on the battery system being evaluate An upward adjustment of 7.5% of this adjusted steady state load was also added to provide additional conservatis This percentage increase resulted in an additional load of approximately 9 amps applied to 11 and 12 batteries, approximately 3 amps applied to 21 battery, and approximately 2 amps applied to 22 battery.This worst case scenario was a failure of Train A AC which is the normal power source for the emergency lights. To account for this scenario, an additional adjustment to the steady state load described above was added to the 12 and 22 batteries to account for the emergency lights. This resulted in an increase of 110 amps for the 12 battery and approximately 66 amps for the 22 battery at the evaluated voltage These adjustments result in a final steady state load profile that is slightly higher than would be expected but that can be used for the best-estimate battery depletion calculations.

The final minute of the load profile in both the design-basis load profile and the battery depletion load profile is a continuation of the steady state load described above with one additio In the last minute of the profile, it was assumed that operation of one 4160 VAC circuit breaker would be require The breaker operation is a momentary load with a relatively high load demand.Application of this load in the last minute of the profile is conservative as it causes a relatively significant drop in battery voltage due to the high current and length of time the battery has been deplete Since operation in the last minute is the most conservative time after the first minute due to the length of time the battery has already been depleted, the modeled momentary load is a bounding case for other miscellaneous momentary load operations which may occur. The assumption of a breaker operation results in a conservative last minute profil Summary: The load profile developed for the battery depletion calculations is a realistic representation of the actual expected system response during a LOOP or SI event. Development of the load profile for the risk assessment was based on the design-basis evaluation summarized in the USAR with modifications to remove overly conservative assumptions that are not appropriate for a best-estimate evaluatio Removal of the conservatisms combined information from the design-basis calculations as well as test data to obtain an accurate and conservative representation of the DC system load during an event. This resulted in battery depletion times which are representative of actual battery capability.

References 1. PRA Evaluation V.SPA.1 1.006, Rev. 0, "Evaluation of Battery Depletion Differences" 2. Calculation 91-02-11, Rev. 2, "PI Battery 11 Calculation" 3. Calculation 91-02-12, Rev. 4, "PI Battery 12 Calculation" 4. Calculation 91-02-21, Rev. 2, "PI 21 Battery Calculation" 5. Calculation 91-02-22, Rev. 2, "PI Battery 22 Calculation" 6. PRA Evaluation V.SPA.1 1.003, Rev. 0, "Prairie Island Battery Depletion Study PRA LOOP with Emergency Lighting and ISI Steady State Test Loads" 7. Prairie Island Updated Safety Analysis Report, Section 8, Rev. 31, "Plant Electrical Systems" 8. Corrective Action Program Action Request #01270104, "Non conservative assumption in Unit 1 Battery Calcs" ENCLOSURE 3 V.SPA.11.006 EVALUATION OF BATTERY DEPLETION DIFFERENCES 20 pages follow Xcel Energy Prairie Island Nuclear Generating Plant Probabilistic Risk Assessment V.SPA.1 1.006 Evaluation of Battery Depletion Differences March 2011 Document Review Form Document or Analysis Title: V.SPA.1 1.006: Evaluation of Battery Depletion Differences Revision of Document or 0 Analysis: Date: Preparer:

Robert Flaaen .i Date: ,/0E.Zo ij Reviewer Greg Kvamme Date: 3 ' /ibof* " '

,//,,/PRA Acceptance Review: Jayne E. Ritter Prairie Island Nuclear Generating Plant (PINGP)Evaluation of Battery Depletion Differences Table of Contents 1.0 Purpose ...................................................................................................

3 2.0 Evaluation of DC System Loading ..........................................................

3 2.1 Comparison of Calculated Load to Observed Load ....................................

3 2 .1.1 11 D C S yste m ......................................................................................................

..3 2 .1.2 12 D C S ystem ......................................................................................................

..4 2.1.3 2 1 D C S ystem ......................................................................................................

..6 2 .1.4 22 D C S ystem ......................................................................................................

..7 2.2 DC Circuit Comparisons

.............................................................................

9 2 .2 .1 11 D C S ystem ......................................................................................................

..9 2.2.2 12 D C S ystem ...................................................................................................

..10 2.2.3 2 1 D C S ystem ...................................................................................................

..10 2.2.4 22 D C S ystem ....................................................................................................

..11 3.0 Evaluation of Loading Used in Battery Depletion Calculations

.............

13 3.1 Station Blackout (SBO) Depletion Calculation

............................................

13 3.2 Safety Injection on Offsite Power (SI) Depletion Calculation

......................

14 3.3 Loss of Offsite Power (LOOP) Depletion Calculation

................................

15 3.3.1 LOOP with Actual Inverter Loading ....................................................................

15 3.3.2 LOOP with Actual Inverter Loading and Failure of Opposite Train AC ..............

16 4.0 Other Influences on Battery Depletion Timing ......................................

18 4.1 Minimum Allowable Battery Terminal Voltage ......................................

18 5.0 Conclusion

.............................................

18 6.0 References

..........................................................................................

19 1.0 Purpose The purpose of this evaluation is to document and provide justification for the differences in battery depletion times calculated for Probabilistic Risk Assessment (PRA) purposes for a Station Blackout (SBO) scenario, Loss of Offsite Power (LOOP) scenario, and Safety Injection on Offsite Power (SI) scenari This evaluation will review the battery loading that was assumed in the calculations and describe the differences.

2.0 Evaluation of DC System Loading A battery loading profile of one hour is developed in the design basis battery sizing calculations (Ref. 6.1, 6.2, 6.3, and 6.4) per requirements stated in the Updated Safety Analysis Report (USAR) (Ref 4.10). The design basis load profile is based on assuming worst case loading for loads to develop a conservative load profile. Additionally, other than major loads (breaker coils, spring charging motors, inverters), smaller loads (lights, misc relays, solenoids)

are assumed on for the entire duration of the analysis or alternating lights and similar components (red/green indicating lights, separate open/close solenoid valves on the same component, etc.) are assumed as both simultaneously on for the entire duration of the analysi This results in a conservative design basis load profile as the devices actually only operate for a short period of time and do not all operate simultaneously.

To obtain the load for these miscellaneous panels which feed such devices as solenoid valves, relays, etc., the design basis calculation took all of the devices on a circuit and added up their individual rated loads, whether the individual devices were on or not. All depletion calculations used the design basis calculation and design basis load profile as a starting point and modified the load profile as necessary to obtain battery depletion times suitable for the specific evaluation being performed.

A comparison of the design basis steady state DC load to the steady state load observed during testing follows in Section 2.1. Additionally, some specific example of circuits are included in Section 2.2 by evaluating the design basis load and reviewing anticipated loading on that circuit based on actual operation of devices.2.1 Comparison of Calculated Load to Observed Load Because of the conservative load profile used in the calculation, the DC load used in the calculation is much greater than that experienced while testing simulated LOOP/SI conditions during Integrated SI testing and also experience during an inadvertent SI event on Unit 2. During these events, the first 30-40 seconds contains a number of transient periods due to breaker closures, valve and relay actuations, etc. However, after 30 seconds, the DC load settles to a steady state value. The steady state value shown in the design basis calculations is during the 1 minute to 59 minute time period.When comparing this to steady state DC load values experienced during simulated or actual event conditions, the calculated value is much higher than what is actually experience As the steady state load is the major influence on the depletion times, the use of a conservative steady state load values on the batteries can result in a conservative depletion time and would not be representative of actual system load or response.2.1.1 11 DC System The design basis calculated load of the 11 DC system at steady state conditions is approximately 123.5 amps (Ref. 6.1). This value does not include inverter loads based on assumptions within the calculatio The design basis inverter loading on 11 battery is approximately 129.5 amps (Ref. 6.1). From the last performance of the Integrated SI test shown below in Figure 1, it can be seen that Page 3 of 19 the steady state load of the 11 DC system experience during the test is approximately 115 amps.Unit I Train A 1R26 Data (10/23/2009)

300 150 Time (Seconds)-Battery Float Current -11 Charger Current -DC System Demand (Charger-Float)

Figure 1 This steady state value from the Integrated SI test is assumed to contain the inverter loads as the inverter loads were not removed from the DC system during the event. This is due to a trip of the inverter AC input breaker on the large voltage overshoot of the EDGs. The basis for this assumption is the difference between the pre-event DC load and the steady state load. This difference (approximately 83 amps) closely resembles the actual inverter load. It would be expected that the steady state load would be approximately equal to or only slightly greater than the pre-event DC load due to the lack of the actuation of loads and generally the same number of operating (running)

loads (indicating lights, continuously energized lights, etc.). This is confirmed by reviewing Integrated SI test data and actual SI data from Unit 2 in which the inverter loads do not remain on the DC system. The comparison of the Unit 2 data showed that the steady state load was only a few amps higher than the pre-event load for both an SI/LOOP and SI loading condition (Ref. Sections 2.1.3 and 2.1.4).The tripping of the inverter AC input breaker on a high voltage has been identified to occur on Unit 1 due to the Unit 1 EDG voltage response during transient loading (Ref. 6.6). This issue has been input into the corrective action program and is being evaluated for additional impact (Ref. 6.11).The profile above in Figure 1 does not include transferable DC panels 17 and 19.These panels are normally fed from the 11 DC system (Ref. 6.12) however, are transferred to their alternate source (21 DC system) during a unit shutdown (Ref.6.13) are then returned to their normal source after a unit startup (Ref. 6.14). The design basis load for these panels on the batteries is approximately 12.4 amps.2.1.2 12 DC System The calculated load of the 12 DC system at steady state conditions is approximately 201.5 amps (Ref. 6.2). This value does not include inverter loads, Page 4 of 19 but does include emergency lightin The design basis inverter loading on 12 battery is approximately 134.8 amps (Ref. 6.2). The design basis emergency lighting load on 12 battery is approximately 110 amps (Ref. 6.2). From the last performance of the Integrated SI test shown below in Figure 2, it can be seen that the steady state load of the 12 DC system is approximately 115 amps. Note that since the 12 battery charger was not included in the test, it is not shown in this Figure.12 DC System Data from 1R26 Test (10/23/2009)

2C o II II I I I I I l 1 l l I IT , I Eb -- 1 1 -I 11 11 11 11I WI iI I + [II III II II M1 II II I II JM I I I I II isIIIiI i I I Ji tN I Seconds-DC Demand Figure 2 This steady state value from the Integrated SI test does not include the emergency lighting, however, it is assumed that the steady state value contains the inverter loads as the inverter loads were not removed from the DC system during the event due to a trip of the inverter AC input breaker on the large voltage overshoot of the EDGs. This is due to the difference between the pre-event DC load and the steady state load. This difference (approximately 92 amps) closely resembles the actual inverter load. It would be expected that the steady state load would be approximately equal to or only slightly greater than the pre-event DC load due to the lack of the actuation of loads and generally the same number of operating (running)

loads (indicating lights, continuously energized lights, etc.).This is confirmed by reviewing Integrated SI test data and actual SI data from Unit 2 in which the inverter loads do not remain on the DC system. The comparison of the Unit 2 data showed that the steady state load was only a few amps higher than the pre-event load for both an SI/LOOP and SI loading condition (Ref.Sections 2.1.3 and 2.1.4).The tripping of the inverter AC input breaker on a high voltage has been identified to occur on Unit 1 due to the Unit 1 EDG voltage response during transient loading (Ref. 6.6). This issue has been input into the corrective action program and is being evaluated for additional impact (Ref. 6.11).The profile above in Figure 2 does not include transferable DC panels 14 and 18.These panels are normally fed from the 12 DC system (Ref. 6.12) however, are Page 5 of 19 transferred to their alternate source (22 DC system) during a unit shutdown (Ref.6.13) are then returned to their normal source after a unit startup (Ref. 6.14). The design basis load for these panels on the batteries is approximately 8 amps.2.1.3 21 DC System The calculated load of the 21 DC system at steady state conditions is approximately 113.1 amps (Ref. 6.3). This value does not include inverter loads.The design basis inverter loading on 21 battery is approximately 94.6 amps (Ref.6.3). From the last performance of the Integrated SI test shown below in Figure 3, it can be seen that the steady state load of the 21 DC system is approximately 36 amps.Unit 2 Train A 2R26 Data (5/17/2010)

200 180 160 140 120 100 80 60 40 E i i i i i i i i i i i i i i i ! i i i i 4 " i i i i i i i i i i I i i ! i I I I I I I I I I I I I I II I I ¶ II I I I I I I I I I I I I CL 11 Ill! I 111111 N 11 W Ill 11111 1MIM 11111111111

............-..... .. ........ ..... ...0-20-40-60-80-100-120 fl1i iiii i i F.11 Sli iiiiii i~ Ti17lYIT F 1 II~.I Seconds i ?U M ý3 R 1-21 Battery Float Amps -21 Battery Charger Amps -21 DC System Demand I Figure 3 Additionally, Unit 2 experienced an inadvertent SI event in 2007. Data taken from that event is shown below in Figure 4. From this data it can be seen that the steady state load of the 21 DC system is approximately 37 amps.Page 6 of 19 4/5/2007 U2 SI Event -Train A DC Loading (1st Minute)Load Step 1 2 3 4 5 6 7 110.0000 90.0000 80.0000 70.0000 60.0000 50,0000 400000 tE 300000 20.000-200000 V I-300000-400000-50.0000 Tlme 1- 21 Battery Float -21 Battery Charger Figure 4 2.1.4 22 DC System The calculated load of the 22 DC system at steady state conditions is approximately 155.6 amps (Ref. 6.4). This value does not include inverter loads, but does include emergency lightin The design basis inverter loading on 22 battery is approximately 103.5 amps (Ref. 6.4). The design basis emergency lighting load on 22 battery is approximately 65 amps (Ref. 6.4). From the last performance of the Integrated SI test shown below in Figure 5, it can be seen that the steady state load of the 22 DC system is approximately 21 amps.Unit 2 Train B 2R26 Data (5/17/2010)

250 200 II I 150 E o .I II-o 1 Seconds 1- 22 Battery Roat Amps -22 Battery Charger Amps -22 DC System Demand Figure 5 Page 7 of 19 Additionally, Unit 2 experienced an inadvertent SI event in 2007. Data taken from that event is shown below in Figure 6. From this data it can be seen that the steady state load of the 22 DC system is approximately 19.5 amps.4/5/2007 U2 SI Event -Train B DC Loading (lt Min)Load Step 1 2 3 4 5 6 7 I 90 80 70 60 50 40 30 20 1 10 L VA o -" ^ -,.,s. V-10-20-30-40-50 .... ....Time 1-22 BteyFlot -22 WeryCha~rer]

Figure 6 Page 8 of 19 2.2 DC Circuit Comparisons This section will look at some of the DC system circuits and compare the design basis loading used in the design basis calculation with estimated loading based on actual device operation to give some example of the conservatisms that are present in the design basis load profile of the batterie The design basis calculations obtain the DC loading information from the Master DC Load List (Ref. 6.15). However, as stated above under Section 2.0, the calculations summed up the rated load of each device on the circuit within miscellaneous load panels to obtain the panel load and did not consider device operatio This resulted in a conservative load value.Reference 6.15 also contained information which identified the expected operation of specific devices on a circuit. This was done by reviewing equipment operation, drawings, control circuits, etc., to try and determine when and if a device would operate during an event. Using this information, a load value which more accurately resembles actual expected load can be estimate This load value is still slightly conservative due to the use of worst case loading of individual devices and conservative device load assumptions where specific manufacturer data could not be located. The following sections will review the design basis load for various panels on the system based on Attachment J of References 6.1, 6.2, 6.3, and 6.4 and compare that to estimated load values based on device operation from Reference 6.15.2.2.1 11 DC System The major contributors to the design basis steady state load of the 11 DC system as shown in Reference 6.1 are panels 13, 15, 17, 19, 151,152, 153, and 191.The design basis load of these panels at nominal DC voltage (125VDC) is tabulated in Attachment J of Reference 6.1. The expected steady state load for a HSD condition, such as during a LOOP, can be estimated using the Reference 6.15 information of whether the device would operate during the event and actually be a load on the DC system. The different load values are compared below in Table 1.DC Panel Design Basis Load Estimated Expected Load % Reduction From Attachment J of From Reference 4.15 Reference 4.1 (Amps) (Amps)15 17.64 7.11 60%17 4.84 3.26 33%19 7.22 4.07 44%151 23.54 14.04 40%152 9.44 6.88 27%153 5.67 1.24 78%191 6.77 2.15 68%Total 75.12 38.75 48%Table 1: Comparison of 11 DC System Loading Panel 13 is not included in the above table since it is not detailed in Reference 6.15. Assuming a 48% load reduction for panel 13 to be consistent with the total load reduction seen by the above panels, panel 13 load can also be considered as shown below in Table 2.Page 9 of 19 DC Panel Design Basis Load Estimated Expected Load % Reduction From Attachment J of From Reference 4.15 Reference 4.1 (Amps) (Amps)13 34.56 16.73* 52%15 17.64 7.11 60%17 4.84 3.26 33%19 7.22 4.07 44%151 23.54 14.04 40%152 9.44 6.88 27%153 5.67 1.24 78%191 6.77 2.15 68%Total 109.68 55.48 49%*Assumed Load Reduction of 48%Table 2: Comparison of 11 DC System Loading with Panel 13 This shows that the design basis steady state load for the 11 DC system is very conservative to expected and observed loading and is not reflective of actual system performance or loading.2.2.2 12 DC System The major contributors to the design basis steady state load of the 12 DC system as shown in Reference 6.2 are panels 14, 16, 161,162, 161,181. The design basis load of these panels at nominal DC voltage (1 25VDC) is tabulated in Attachment J of Reference 6.2. The expected steady state load for a HSD condition, such as during a LOOP, can be estimated using the Reference 6.15 information of whether the device would operate during the event and actually be a load on the DC system. The different load values are compared in Table 3.DC Panel Design Basis Load Estimated Expected Load % Reduction From Attachment J of From Reference 4.15 Reference 4.2 (Amps) (Amps)14 6.74 2.95 56%16 41.59 8.54 79%18 1.25 1.15 8%161 21.35 3.28 85%162 6.72 2.67 60%163 3.41 0.42 88%Total 81.06 19.01 77%Table 3: Comparison of 12 DC System Loading This shows that the design basis steady state load for the 12 DC system is very conservative to expected and observed loading and is not reflective of actual system performance or loading.2.2.3 21 DC System The major contributors to the design basis steady state load of the 21 DC system as shown in Reference 6.3 are panels 17, 19, 23, 25, 251, 252, and 253. The design basis load of these panels at nominal DC voltage (125VDC) is tabulated in Attachment J of Reference 6.3. The expected steady state load for a HSD condition, such as during a LOOP, can be estimated using the Reference 6.15 information of whether the device would operate during the event and actually be a load on the DC system. The different load values are compared in Table 4.Page 10 of 19 DC Panel Design Basis Load Estimated Expected Load % Reduction From Attachment J of From Reference 4.15 Reference 4.3 (Amps) (Amps)17 4.84 3.26 33%19 5.13 4.07 21%25 18.23 3.59 80%251 11.85 5.2 56%252 8.63 5.31 38%253 6.33 1.55 76%Total 55.01 22.98 58%Table 4: Comparison of 21 DC System Loading Panel 23 is not included in the above table since it is not detailed in Reference 6.15. Assuming a 50% load reduction for panel 23 to be consistent with the total load reduction seen by the above panels, panel 23 load can also be considered as shown below in Table 5.DC Panel Design Basis Load Estimated Expected Load % Reduction From Attachment J of From Reference 4.15 Reference 4.3 (Amps) (Amps)17 4.84 3.26 33%19 5.13 4.07 21%23 33.1 16.55* 50%25 14.16 3.59 75%251 11.85 5.2 56%252 8.63 5.31 38%253 6.33 1.55 76%Total 84.04 39.53 53%*Assumed Load Reduction of 50%Table 5: Comparison of 21 DC System Loading with Panel 23 This shows that the design basis steady state load for the 21 DC system is very conservative to expected and observed loading and is not reflective of actual system performance or loading.2.2.4 22 DC System The major contributors to the design basis steady state load of the 22 DC system as shown in Reference 6.4 are panels 14, 18, 26, 261, 262, and 263. The design basis load of these panels at nominal DC voltage (1 25VDC) is tabulated in Attachment J of Reference 6.4. The expected steady state load for a HSD condition, such as during a LOOP, can be estimated using the Reference 6.15 information of whether the device would operate during the event and actually be a load on the DC system. The different load values are compared in Table 6.DC Panel Design Basis Load Estimated Expected Load % Reduction From Attachment J of From Reference 4.15 Reference 4.2 (Amps) (Amps)14 6.82 2.95 57%18 1.25 1.15 8%26 30.39 5.2 83%261 11.14 2.81 75%262 5.52 0.21 96%263 3.54 0.42 88%Total 58.66 12.74 78%Table 6: Comparison of 22 DC System Loading Page 11 of 19 This shows that the design basis steady state load for the 22 DC system is very conservative to expected and observed loading and is not reflective of actual system performance or loading.Page 12 of 19 3.0 Evaluation of Loading Used in Battery Depletion Calculations 3.1 Station Blackout (SBO) Depletion Calculation On the occurrence of an SBO, the batteries supply the emergency lights (where applicable)

and inverters due to a loss of AC power. The SBO depletion calculation used the design basis load profile and modified the inverter (INV) loading to more realistically resemble actual data by removing the conservatism in the inverter load.The inverter load values that were used were based on actual load readings taken during surveillance procedures of the DC system. The approximate inverter load values used in the SBO calculation, the difference from the design basis inverter load, as well as the resultant steady state load in the SBO calculation and is shown below in Table 1. The SBO steady state load includes values with and without the emergency lights (E.L.) where applicable as it was evaluated that during an SBO, the emergency lights are manually shed from the DC system 40 minutes into the event per emergency procedures. (Ref. 6.5)DC System Approximate Inverter Difference In Approximate Steady Load in SBO Inverter Load State Load in SBO Depletion Calculation From Design Depletion Calculation (Ref. 6.5) Basis Calculation (Ref. 6.5)11 DC Inverters 97.7 amps 31.8 amps lower 219 amps 12 DC Inverters 107.2 amps 27.6 amps lower 305 amps (w/ E.L.)1 1 197 amps (w/o E.L.)21 DC Inverters 89.1 amps 5.5 amps lower 201 amps 22 DC Inverters 85.36 amps 18.1 amps lower 239 amps (w/ E.L.)174 amps (w/o E.L.)Table 1 This steady state load that was used in the SBO depletion calculations still contains the conservatisms of various loads being energized during the entire event. When comparing the steady state load used in the SBO depletion calculation to the DC load profile during the Integrated SI Test in Figures 1-4 under Section 2.1, it can be seen that the steady state load used in the SBO depletion calculation is above the steady state load seen under actual conditions and in some cases is above the highest load experience during the entire Integrated SI Test when the Emergency Diesel Generator (EDG) is starting up and there is no AC power.DC Maximum DC System DC System Steady Approximate Steady System Load During Sequence State Load From State Load in SBO From Section 2.1 Figures Section 2.1 Figures Depletion Calculation (ISI Test Data) (includes Inverter and E.L (if applicable))(Ref. 6.5)11 133 amps (2) 115 amps 219 amps 12 235 amps (13 115 amps 305 amps (w/ E.L.)197 amps (w/o E.L.)21 113 amps 37 amps 201 amps 22 145 amps V) 21 amps 239 amps (w/ E.L.)174 amps (w/o E.L.)Table 2 (1) The load value listed here also includes the Inverters and (2) The load value listed includes the Inverters.

Emergency Lighting.Page 13 of 19 This shows that the steady state load values used in the SBO depletion calculation were still very conservative even with the adjustment of the inverter loads. However, the depletion times reached in the calculation were adequate for the evaluation that was occurring at the time and therefore, no additional conservatism needed to be removed.The SBO depletion calculation (Ref 6.5) provided the following results: Battery Depletion Time Battery 11 165 minutes (2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> 45 minutes)Battery 12 217 minutes (3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> 37 minutes)Battery 21 184 minutes (3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> 04 minutes)Battery 22 239 minutes (3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> 59 minutes)3.2 Safety Injection on Offsite Power (SI) Depletion Calculation This event occurs on offsite power and therefore, there is no loss of AC power to the emergency lights or the inverters and they remain powered from their respective AC sources. This calculation used the design basis load profile, modified it to represent SI sequencing without a loss of offsite power, and removed the inverters and the emergency lighting (as applicable)

due to no loss of AC power. Table 3 shows the steady state load values used in the depletion calculation and compares them to steady state load values seen from Figures 1 -4 under Section 2.1.DC System Approximate Steady State Approximate Steady Load in SI Depletion States Load From Calculation (Ref. 6.7) Section 2.1 Figures 11 124 amps 115 amps 12 93 amps 115 amps 1'21 114 amps 37 amps 22 91 amps 21 amps Table 3 (1) The load value listed includes the Inverters.

As shown above in the table, the steady state load for 12 DC system from Section 2.1.2 is higher than that used in the SI only depletion calculatio The load profile under Section 2.1.2 is a SI/LOOP conditio As discussed in Sections 2.1.1 and 2.1.2, the load profiles for 11 and 12 DC system contained the inverter loads during the entire sequence, as well as during the steady state portion of the profile, due to the inverters'

AC input breaker tripping on an EDG voltage overshoot causing them to be a continuous load on the DC system. The event analyzed under the SI only depletion calculation is when offsite power is availabl The voltage overshoots experienced during Unit 1 EDG sequence loading do not occur when sequencing on offsite power. Therefore, the inverters would not be a steady state load on the Unit 1 DC system for this event. If the inverter loads were removed from the profiles under Sections 2.1.1 and 2.1.2, the steady state load would be slightly higher than the pre-event load of approximately 30 amps for 11 DC system and 20 amps for 12 DC system shown in Figures 1 and 2.Page 14 of 19 The steady state load that was used in the SI only depletion calculations still contains the conservatisms of various loads being energized during the entire event. When comparing the steady state load used in the depletion calculation to the DC load profile during the Integrated SI Test in Figures 1-4 under Section 2.1, it can be seen that the steady state load used in the calculation is very conservativ However, the depletion times reached in the calculation were adequate for the evaluation that was occurring at the time and therefore, no additional conservatism needed to be removed.The SI only depletion calculation (Ref 6.7) provided the following results: Battery Depletion Time To Design Basis Voltage Battery 11 460 minutes (7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> 40 minutes)Battery 12 710 minutes (11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br /> 50 minutes)Battery 21 515 minutes (8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> 35 minutes)Battery 22 645 minutes (10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> 45 minutes)3.3 Loss of Offsite Power (LOOP) Depletion Calculation This event was further evaluated into two separate scenarios:

(1) LOOP with Actual Inverter Loading (Ref. 6.8) and (2) LOOP with Actual Inverter Loading and Failure of Opposite Train AC (Ref. 6.9).The initial case (1) evaluated the actual response where only the Unit 1 inverters become a load on the DC system due to tripping of their AC input breaker on a Unit 1 EDG overshoo Unit 2 inverters were assumed to remain on AC power since it is not expected that the Unit 2 inverters'

AC input breaker will trip due to a lack of significant overshoots during EDG sequencing (Ref. 6.6). This case also modified the steady state load to match what is seen during the Integrated SI Test or what was seen during the inadvertent SI event. This removes the conservative assumption in which various loads are continuously energized and more closely resembles actual system loading and response.The second case (2) used the load profile developed under case 1 in which the inverters were a load on the Unit 1 EDGs and the steady state load matched the actual test or event data. Additionally, this case assumed a failure of the opposite train's AC power source. The emergency lighting is fed from Train A AC power and Train B DC power. By assuming a failure of the Train A AC power, the emergency lights become a load on the Train B DC system during the entire event. This is considered the limiting case for DC loading and battery depletion during a LOOP event.3.3.1 LOOP with Actual Inverter Loading This case used the same load profile developed for the SBO calculations except that the emergency lights were removed from the load profile after the EDGs provided AC power to the bus, the Unit 2 inverters were removed from the load profile after the EDGs provided AC power to the bus (the Unit 1 inverters remained on the DC system), and the steady state load was adjusted to approximately match actual test or event data. The missing panels described under Section 2.1 for Unit 1 DC systems were conservatively added to the steady state load values observed from the tests for both units, plus some additional margin.Page 15 of 19 Table 4 shows the steady state load values used in the depletion calculation and compares them to steady state load values seen from Figures 1 -4 under Section 2.1.DC System Approximate Steady Approximate Steady State Load in Case 1 States Load From LOOP Depletion Section 2.1 Figures Calculation (Ref. 6.8)11 130 amps 115 amps 12 128 amps 115 amps 21 50 amps 37 amps 22 30 amps 21 amps Table 4 This steady state load that was used in the Case 1 LOOP depletion calculations still contains some margin to actual values but is much more reflective of actual plant conditions and a more realistic load value.The Case 1 LOOP (with actual inverter loading and Steady State Load) depletion calculation (Ref 6.8) provided the following results: Battery Depletion Time To Design Basis Voltage Battery 11 Greater than 6.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> Battery 12 Greater than 7.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> Battery 21 Greater than 22 hours2.546296e-4 days <br />0.00611 hours <br />3.637566e-5 weeks <br />8.371e-6 months <br /> Battery 22 Greater than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> 3.3.2 LOOP with Actual Inverter Loading and Failure of Opposite Train AC This case used the same load profile for the Case 1 LOOP depletion calculation discussed in Section 3.3.1, however, due to the assumed failure of the opposite train AC power source, the emergency lights (EL) were included as a load on 12 and 22 batteries during the entire time period. Table 5 shows the steady state load values used in the depletion calculation.

DC System Approximate Steady State Load in Case 2 LOOP Depletion Calculation (Ref. 6.9)11 130 amps z 12 238 amps (1)21 50 amps (2)22 96 amps "I Table 5 (1) The load value listed here also includes the Inverters and Emergency Lighting.(2) The load value listed includes the Inverters.

(3) The load value listed includes Emergency Lighting This steady state load that was used in the Case 2 LOOP depletion calculations still contains some margin to actual values but is much more reflective of actual plant conditions and a more realistic load value when the emergency lights would be powered from the DC system due to a failure of the opposite train AC.Page 16 of 19 The Case 2 LOOP (with actual inverter loading and Failure of Opposite Train AC Power) depletion calculation (Ref 6.9) provided the following results: Battery Depletion Time To Design Basis Voltage Battery 11 Greater than 6.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> Battery 12 Greater than 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> Battery 21 Greater than 22 hours2.546296e-4 days <br />0.00611 hours <br />3.637566e-5 weeks <br />8.371e-6 months <br /> Battery 22 Greater than 9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br /> Page 17 of 19 4.0 Other Influences on Battery Depletion Timing 4.1 Minimum Allowable Battery Terminal Voltage The minimum allowable terminal voltage is specific to each battery and is determined by the battery sizing calculations (Ref. 6.1 to 6.4). The depletion time for the battery to reach minimum terminal voltage is largely influenced by the steady state load on the battery as described above, however, the minimum allowable terminal voltage also has an affect on the battery depletion time. The minimum allowable terminal voltage values for each battery is listed in table below.Battery Minimum Allowable Terminal Voltage Battery 11 110.34 Vdc (Reference 6.1 )Battery 12 109.55 Vdc (Reference 6.2)Battery 21 110.61 Vdc (Reference 6.3)Battery 22 110.08 Vdc (Reference 6.4)Given similar loading on each battery, the depletion time will be shorter for a battery with a higher minimum allowable terminal voltage limit. For example, in Section 3.3.1 (LOOP with Actual Inverter Loading), the 11 battery and 12 battery have very similar steady state loads (130 amps and 128 amps, respectively).

However, the depletion times for 11 battery and 12 battery are different (6.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> and 7.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />, respectively).

The 12 battery depletion time is 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> longer due to the lower minimum allowable terminal voltage.5.0 Conclusion The evaluation performed above in Section 2.0 showed that the design basis loading profile of the batteries is very conservative when compared to actual data. When performing battery depletion studies to obtain a best estimate depletion time, the use of test or actual event data represents a more accurate reflection of system performance and loading.As the PRA evaluations progressed, significant initiating events were identified which required re-evaluation of the battery depletion calculations in order to identify conservatism As part of that process, conservatisms were removed from the design basis load profile to more accurately match real data.Page 18 of 19 6.0 References 6.1 Calculation 91-02-11, Rev. 2, "PI Battery 11 Calculation" 6.2 Calculation 91-02-12, Rev. 4, "PI Battery 12 Calculation" 6.3 Calculation 91-02-21, Rev. 2, "PI 21 Battery Calculation" 6.4 Calculation 91-02-22, Rev. 2, "PI Battery 22 Calculation" 6.5 PRA Evaluation V.SPA.1 0.013, Rev. 0, "Battery Depletion Calculation" 6.6 PRA Evaluation V.SPA.1 1.005, Rev. 0, "Evaluation of Inverter Source Selection During Emergency Diesel Generator Loading" 6.7 PRA Evaluation V.SPA.1 1.004, Rev. 0, "Prairie Island PRA SI Only Battery Depletion Study" 6.8 PRA Evaluation V.SPA.1 1.009, Rev. 0, "Prairie Island Battery Depletion Study PRA LOOP with ISI Steady State Test Loads" 6.9 PRA Evaluation V.SPA.1 1.003, Rev. 0, "Prairie Island Battery Depletion Study PRA LOOP with Emergency Lighting and ISI Steady State Test Loads" 6.10 Prairie Island Updated Safety Analysis Report, Section 8, Rev. 32P, "Plant Electrical Systems" 6.11 Corrective Action Program Action Request #01270104, "Non conservative assumption in Unit 1 Battery Calcs" 6.12 Operating Procedure C20.9, Rev. 29, "Station Battery & DC Distribution System" 6.13 Operating Procedure 1C1.3, Rev. 69, "Unit 1 Shutdown" 6.14 Operating Procedure 1C1.4, Rev. 51, "Unit 1 Power Operation" 6.15 Master DC Load List (File MasterDCLddlst_05-31-01 .xls Dated 05/31-01 at 1:08pm)Page 19 of 19

Retrieved from "https://kanterella.com/w/index.php?title=IR_05000282/2011010&oldid=1051219"