ML051220367
| ML051220367 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse  |
| Issue date: | 04/28/2005 |
| From: | Bezilla M FirstEnergy Nuclear Operating Co |
| To: | Document Control Desk, Office of Nuclear Reactor Regulation |
| References | |
| 3121, LAR 03-0017, TAC MC3058 | |
| Download: ML051220367 (13) | |
Text
FENOC FirstEnergy Nuclear Operating Company 5501 North State Route 2 Oak Harbor, Ohio 43449 Mark B. Bezilla Vice President - Nuclear 419-321-7676 Fax: 419-321-7582 April 28, 2005 Docket Number 50-346 License Number NPF-3 Serial Number 3121 United States Nuclear Regulatory Commission Document Control Desk Washington, D. C. 20555-0001
Subject:
Supplemental Information Regarding License Amendment Application to Change Design Requirements for Emergency Diesel Generators (License Amendment Request No. 03-0017, TAC No. MC3058)
Ladies and Gentlemen:
By letter dated May 3, 2004 (Serial Number 3006), the FirstEnergy Nuclear Operating Company (FENOC) submitted a license amendment application for the Davis-Besse Nuclear Power Station (DBNPS). The proposed amendment would change the facility as described in the DBNPS Updated Safety Analysis Report (USAR) to modify the design requirements for the emergency diesel generators (EDGs). Specifically, the proposed amendment would allow a departure from the regulatory position of Safety Guide 9, "Selection of Diesel Generator Set Capacity for Standby Power Supplies," for the frequency and voltage transient during the EDG automatic loading sequence. By facsimile on October 19, 2004, the NRC requested additional information regarding the proposed amendment. The FENOC responses to this request are provided in.
Should you have any questions or require additional information, please contact Mr. Henry L.
Hegrat, Supervisor - Licensing, at (330) 315-6944.
Docket Number 50-346 License Number NPF-3 Serial Number 3121 Page 2 The statements contained in this submittal, including its associated attachments, are true and correct to the best of my knowledge and belief. I declare under penalty of perjury that the foregoing is true and correct.
Executed on:
XA__h_
By::
MarB. Bezilla, Vice Prey nt - Nuclear MAR Attachments cc:
J. L. Caldwvell, Regional Administrator, NRC Region III N. Dragani, Executive Director, Ohio Emergency Management Agency, State of Ohio (NRC Liaison)
J. B. Hopkins, DB-1 NRC/NRR Senior Project Manager C. S. Thomas, DB-I NRC Senior Resident Inspector Utility Radiological Safety Board
Docket Number 50-346 License Number NPF-3 Serial Number 3121 Page I FENOC Response to NRC Request for Additional Information Amendment to Change UFSAR to Modify Design Requirements for EDGs Related to EDG Voltage and Frequency While Loading Davis-Besse Nuclear Power Station la.
How long have you experienced this loading sequence transient problem?
Response
The condition was identified during the review of data from a test performed on February 2, 2003, during the thirteenth refueling outage (13RFO). It is not known how long the voltage and frequency responses of the Emergency Diesel Generators (EDGs) have not met their design requirements during design basis transient loading conditions. Since there have been no major design changes to the EDGs or to the major loads, it is believed that the loading sequence transient problem has existed since initial plant startup.
lb.
When was the problem discovered?
Response
The lack of analysis of the EDG voltage and frequency response under transient loading conditions was identified as part of a Latent Issues Review of the EDGs performed during 13RFO. The EDG Latent Issues Review was completed in December 2002. The loading sequencing transient problem was confirmed during surveillance testing of EDG-1 on February 2, 2003.
lc.
How was the problem discovered?
Response
The 2002 EDG Latent Issue Review identified deficiencies with a 1992 design calculation that documented the EDG loading. The purpose of the calculation was to demonstrate that EDG-1 and EDG-2 would be able to pick up the loads listed in the EDG Loading Tables. The 2002 EDG Latent Issue Review determined that the methodology used in the calculation was inadequate. In addition, it was determined that the calculation did not: (a) evaluate the EDG voltage and frequency response that would occur during the automatic load sequencing associated with a design basis Loss of Offsite Power (LOOP) coincident with a Loss of Coolant Accident (LOCA) and (b) verify that the response meets the design criteria specified in the Updated Safety Analysis Report (USAR). Furthermore, it was determined that the impact of the voltage and frequency response on the emergency bus loads also was not evaluated. This condition was documented in the Davis-Besse Corrective Action Program in the form of a Condition Report.
Docket Number 50-346 License Number NPF-3 Serial Number 3121 Attachment I Page 2 In order to address the above issues, it was decided to develop an analytical model of the EDGs and connected loads and then calculate the response. To develop the analytical model, more detailed test data on the EDGs than usually collected was required. This test data was obtained during a subsequent Safety Features Actuation System (SFAS) surveillance test conducted in February 2003. Evaluation of this data indicated that the terminal voltage and frequency of EDG-1 dropped below the design criteria specified in the USAR during automatic load sequencing.
This condition also was documented in the Davis-Besse Corrective Action Program in the form of a Condition Report.
- 2.
Provide a diagram that shows the EDG, the loads and the essential substation transformer.
Response
Attached with this response are copies of USAR Figures 8.3-1 and 8.3-2 (see attachment 2). These drawings show the Emergency Diesel Generators, the 4160 Volt loads, the 4160 - 480 Volt essential substation transformers, and the 480 Volt essential MCCs.
3a.
Provide the rating and the associated loading of the essential substation transformer.
Response
The self-cooled 4160 - 480 Volt essential transformers are rated at 1000 kVA at 55 'C rise. The maximum calculated loading on a 480 Volt essential switchgear bus after automatic loading sequencing is approximately 0.74 MW (and 0.52 MVAR), which occurs during the last load step (step 5) associated with a LOOP coincident with a LOCA.
3b.
Provide the magnitude and the duration of inrush current for the essential substation transformer.
Response
The magnitude and duration of inrush current for the essential substation transformer has not been calculated. As discussed in the FENOC license amendment application, the existing USAR Section 8.3.1.1.4.1 states that "there may be a voltage dip below 75 percent of normal lasting for a few cycles due to the essential unit substation transformer excitation inrush." When a transformer is first energized, the magnitude of the inrush current is a function of the supply voltage at the instant the transformer is energized, the residual flux in the transformer core, and the impedance of the supply circuit. The magnitude of the inrush transient can be as much as 8 to 10 times the transformer's full load current. The inrush current transient decays very rapidly (within a few AC cycles); therefore it is not considered in the transient analysis. However, it is
Docket Number 50-346 License Number NPF-3 Serial Number 3121 Attachment I Page 3 addressed in the setpoint calculation for the protective relaying on the buses and EDG output breakers. For the essential substation transformer protective relaying calculations, the transformer inrush current is conservatively assumed to be 10 times full load current for 6 cycles (0.1 seconds).
- 4.
It is stated in Section 4.3 of the submittal that the EDGs are not affected by the momentary voltage and frequency dips below USAR criteria that occur during load sequencing and that the Safety Features Actuation System (SFAS) testing has demonstrated that the engine, governor, and generator and excitation system are capable of operating reliably. However, during the SFAS testing safety-related pumps are not operated at full flow and certain safety-related loads cannot actually be loaded without undue hardship or potential for undesired operation. With regard to this, the staff questions whether the voltage and frequency values may dip below those experienced during the SFAS testing (59% and 95%) under actual accident conditions. Therefore, confirm that the calculated voltage and frequency values envelope accident loading conditions (assuming all pumps are operating at full flow).
Response
It is expected that under actual accident conditions the EDG voltage and frequency may dip below the values experienced during the periodic surveillance test. This is expected because the worst case loading associated with a LOOP coincident with a LOCA will be greater than the load available during the surveillance test. The calculated voltage and frequency values envelope accident loading conditions assuming all pumps are operating at maximum expected flow.
As discussed in the response to Question 1c above, an analytical model of the EDGs was developed to allow evaluation of the EDGs transient response under worst case design conditions. FENOC's conclusion that the EDGs will adequately perform their safety function is based on this analytical model and not solely on the actual test results.
- 5.
It is stated in Section 3.0 of your submittal that in addition to this proposed amendment, corrective actions to improve the EDG voltage and frequency response are being considered. These include (1) upgrading the existing static exciter and voltage regulator, (2) adding a frequency interlock as a closure permissive of the EDG output breaker or incorporating a time delay to the existing voltage interlock, (3) moving loads from SFAS load step I to other load steps, (4) upgrading the existing governor control system, and (5) delaying briefly the start of the service water pump or component cooling water pump following a LOOP only. Please provide your schedule as to when your assessment of these options will be completed and implemented in order to improve EDG voltage and frequency response.
Docket Number 50-346 License Number NPF-3 Serial Number 3121 Attachment I Page 4
Response
An independent contractor has prepared a report which includes an examination of the EDG voltage and frequency response issue. This report includes recommendations to improve the voltage and frequency response of the EDGs during a LOOP coincident with a LOCA and a LOOP-only event. Assessment of the recommendations from this report by FENOC will be completed by the end of April 2005. It is FENOC's intent to make improvements in EDG voltage and frequency response during the next refueling outage (spring 2006). Some recommendations may be implemented in subsequent outages based on design and procurement lead times.
- 6.
With regard to moving loads from load step 1 to other load steps, please provide a list of the loads that would be moved to other load steps.
Response
As noted in the response to Question 5, options to improve the EDG voltage and frequency transient are currently under evaluation. Therefore, a list of loads that would be moved is not provided.
- 7.
It is stated in Section 4.4.2 that the historical reliability of the component cooling water pump motors was examined to determine if the reduced voltage and frequency during surveillance testing had resulted in premature aging of the motors and that this review identified no premature aging of the equipment. Were any other safety pump motors included in this study and if so, which ones?
Response
No pump motors other than the component cooling water pump motors were included in the review of historical reliability data. The review of historical reliability focused on the component cooling water pump motor because it is the only safety-related 4160 Volt motor that starts during the first load step.
Accordingly, it is required to start at the lowest voltage.
- 8.
Additionally, discuss the types of reviews that were performed on the component cooling water pump motors to determine that there were no premature aging effects due to reduced voltage and frequency as a result of surveillance testing.
Response
The review of historical reliability of the component cooling water pump motors was limited to a review of the maintenance history of the motors. The component cooling water pump motors are original plant equipment. The in-service performance of these motors has been very good. There have not been any indications of electrical failure or degradation. Within the last year, two of the three component cooling water pump motors have been sent off site for a shop overhaul. There were no indications of electrical degradation identified.
Docket Number 50-346 License Number NPF-3 Serial Number 3121 Attachment I Page 5
- 9.
It is stated in Section 4.5.1.3, page 16, 2d paragraph, that for a worst case voltage dip at the 4160 V bus of 2500, the voltage at the motor terminals will be less than 70% of rated motor starting. For conservatism, the motor was assumed to be completely stalled for 1.5 seconds. The EDG transient loading calculation shows the bus voltage will rise to greater than 4160 V within this time and remain above 4160 V for the duration of Step 1. Please clarify the following:
9a.
How many pump motors are assumed to be stalled in this first step?
Response
The transient analyses performed to determine the voltage and frequency response determined that none of the energized pump motors are expected to stall during the first load step. The first load step occurs with closure of the Emergency Diesel Generator output circuit breaker. At that time, the motor loads will draw locked rotor current. However, the transient loading analysis indicates that the motors will accelerate and the voltage and frequency will recover. While initial acceleration may be slower than with full voltage, none of the motors are predicted to stall.
The discussion in Section 4.5.1.3 of the FENOC license amendment application regarding "stalled motors" pertains to the evaluation of the impact of the voltage and frequency dips on the protective relays. For these evaluations, it was assumed, conservatively, that each pump motor connected during the first load step would stall for the duration of the dip. This was done because the motor load currents are higher under stalled conditions and, therefore, have a higher potential for inadvertently tripping the protective relays. As discussed in the FENOC license amendment application, it was concluded that the relays will not inadvertently operate due to the voltage and frequency response of the Emergency Diesel Generators during automatic load sequencing.
Docket Number 50-346 License Number NPF-3 Serial Number 3121 Page 6 9b.
If all motors are assumed to be stalled simultaneously for 1.5 seconds and drawing inrush current then how would the safety bus voltage and frequency recover within the same time period?
Response
Section 4.5 of the FENOC license amendment application summarizes the impact of the voltage dips on protective relays. As discussed in the FENOC license amendment application, for the purposes of evaluating the impact on the protective relays, it was conservatively assumed that minimum voltage or maximum current will exist for the duration of the dip. This conservative assumption does not represent the actual condition of the motors, which analysis has shown will accelerate under accident conditions.
The assumption of stalled motor conditions was only used in evaluating the impact on protective relays. In determining the recovery time of safety bus voltage and frequency, no motors were assumed to be stalled for 1.5 seconds.
None of the energized pump motors are expected to stall during the first load step.
The actual transient analysis and surveillance test data show that the loads accelerate and the voltage and frequency recover.
9c.
If all motors are drawing inrush current then how does the voltage at the motor terminals recovers to 100%?
Response
When an induction motor is started, it initially draws inrush (or locked rotor) current. Provided that sufficient voltage is provided to accelerate the motor, the motor current draw will diminish as it accelerates. The reduction in motor current draw coupled with the EDG voltage regulator response enable the voltage to recover. The actual transient analysis and surveillance test data show that the loads accelerate and voltage recovers.
9d.
Was actuation of random and permanently connected loads considered during the first step?
Response
The analysis of the first load step includes the permanently connected loads.
For the purpose of this discussion, "random" is considered to mean loads that are intermittently connected (i.e., loads that are periodically connected to the essential busses and may or may not be connected during EDG loading). The analysis conservatively assumes that "random" loads, such as temperature control ventilation fans and some heat tracing loads, are connected to the Emergency Diesel Generators. Motor operated valves which receive a signal to operate are also part of the loads which are included in the first step.
Docket Number 50-346 License Number NPF-3 Serial Number 3121 Page 7 1O.a. What is effect of the voltage and frequency transients on the response time of the safety-related loads?
Response
Some EDG loads are safety-related, but do not have response times. For those loads with response times, Section 4.7 of the submittal, "Response Times of Essential Loads," addressed the SFAS functions defined in Section 3.3.2.1 and Table 3.3-3, which are the functions that are required for operability. FENOC determined EDG voltage and frequency and SFAS equipment response times using procedure DB-SC-03114, SFAS Integrated Time Response Test. DB-SC-03114 is performed during plant outages. Response time penalties have been added to the observed data, to correct for differences between the test operating conditions and conditions that would be possible during an accident. There is sufficient response time margin to account for the expected design basis LOOP/LOCA voltage and frequency responses.
1Ob. Do the response times of the safety-related pumps still meet the times specified in the accident analysis (assuming that the voltage and frequency transients have no adverse effect on the pump motors)?
Response
Yes, the response times of the safety related components still meet the required SFAS response times specified in the Technical Requirements Manual. These response times bound the response times assumed in the USAR Chapter 15 safety analyses.
Docket Number 50-346 License Number NPF-3 Serial Number 3121 USAR Figures 8.3-1 and 8.3-2 AC Electrical System One-Line Diagram, Sheets 1 and 2 (2 pages follow)
THIS PAGE IS AN OVERSIZED DRAWING OR FIGURE THAT CAN BE VIEWED AT THE RECORD TITLED:
"DAVIS-BESSE NUCLEAR POWER STATION AC ELECTRICAL SYSTEM ONE-LINE DIAGRAM E-, SH.1" FIGURE 8.3-1 REVISION 24, JUNE 2004 WITHIN THIS PACKAGE.....
D-O1
THIS PAGE IS AN OVERSIZED DRAWING OR FIGURE THAT CAN BE VIEWED AT THE RECORD TITLED:
"DAVIS-BESSE NUCLEAR POWER STATION AC ELECTRICAL SYSTEM ONE LINE DIAGRAM E-1 SH. 2" FIGURE 8.3-2.
REVISION 24, JUNE 2004 WITHIN THIS PACKAGE.....
D-O2
Docket Number 50-346 License Number NPF-3 Serial Number 3121 COMMITMENT LIST The following list identifies those actions committed to by the Davis-Besse Nuclear Power Station (DBNPS) in this document. Any other actions discussed in the submittal represent intended or planned actions by the DBNPS. They are described only for information and are not regulatory commitments. Please notify the Supervisor - Licensing (330-315-6944) of any questions regarding this document or any associated regulatory commitments.
COMMITMENTS DUE DATE Assessment of the recommendations from this report April 30,2005 by FENOC will be completed.