ML17241A337
| ML17241A337 | |
| Person / Time | |
|---|---|
| Site: | Saint Lucie  |
| Issue date: | 05/20/1999 |
| From: | Stall J FLORIDA POWER & LIGHT CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| GL-95-07, GL-95-7, L-99-110, NUDOCS 9905260363 | |
| Download: ML17241A337 (11) | |
Text
CATEGORY 2 REGULAT Y INFORMATION DISTRIBUTIOh SYSTEM (RIDS)
ACCESSION NBR:9905260363 DOC.DATE: 99/05/20 NOTARIZED: NO DOCKET FACIL:50-335 St. Lucie Plant, Unit 1, Florida'Power S Light Co. 05000335 50-389 St. Lucie Plant, Unit 2, Florida Power 8 Light Co. 05000989 AUTH:NAME AUTHOR AFFILIATION STALL,J.A. Florida Power 2 Light Co.
RECIP.N2QTE RECIPIENT AFFILIATION Records Management Branch (Document Control Desk)
SUBJECT:
. Forwards util suppl to GL 95-07 response re pressure-locking C
& thermal binding of safety-related power-operated gate valves, in response to NRC second RAI dtd 990225. A DISTRIBUTION CODE: A056D COPIES RECEIVED:LTR J ENCL SIZE: T TITLE: Generic Ltr 95 Pressure Locking 6 TTiermal Binding of Safety Rela E
NOTES:
RECIPIENT COPIES RECIPIENT COPIES ID CODE/NAME LTTR ENCL ID CODE/NAME LTTR ENCL NRR/DLPM/EATON 1 1 . LPD2-2 PD 1 " 1 GLEAVES,W 1 1 INTERNAL: E CENTER 1 1 NRR/DE/EMEB 1 1 EXTERNAL: NOAC. 1 1 NRC PDR 1 1 NUDOCS ABSTRACT . 1 1 D
0 e
NOTE TO ALL "RIDS" RECIPIENTS:
PLEASE HELP US TO REDUCE WASTE. TO HAVE YOUR NAME OR ORGANIZATION REMOVED FROM'DISTRIBUTION LISTS OR REDUCE THE NUMBER OF COPIES RECEIVED BY YOU OR YOUR ORGANIZATION, CONTACT THE DOCUMENT CONTROL DESK (DCD) ON EXTENSION 415-2083 TOTAL NUMBER OF COPIES REQUIRED: LTTR 8 ENCL 8
Florida Power & Light Company, 6351 S. Ocean Drive, Jensen Beach, FL 34957 May 20, 1999 L-99-110 10 CFR 50.4 10 CFR 50.54 (f)
U. S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555 RE: St. Lucie Units 1 and 2 Docket Nos. 50-335 and 50-389 Second Request for Additional Information Response Generic Letter 95-07 The Florida Power & Light Company (FPL) supplement to our Generic Letter (GL) 95-07, Pressure-Locking and Thermal Binding of Safety-Related Power-Operated Gate Valves, response for St. Lucie Units 1 and 2 is attached. The attached provides a response to your second request for additional information (RAI) dated February 25, 1999. The FPL response was requested within 60 days of the March 4, 1999, receipt date. On May 4, 1999, the NRC back-up Project Manager, Kahtan Jabbour, concurred with an FPL schedule change for the submittal to May 20, 1999.
The NRC issued GL 95-07 on August.17, 1995. The FPL initial responses were submitted by FPL letters L-95-282 and L-96-31 dated October 11, 1995, and February 13, 1996, respectively.
On June 26, 1996, NRC issued its first RAI and FPL responded to that request by FPL letter L-96-191 dated July 31, 1996.
This letter contains FPL commitments in the response to NRC question 2. FPL plans to modify three of the St. Lucie Unit 1 shutdown cooling (SDC) valves (V-3481, V-3651, and V-3652) during the Fall 1999 St. Lucie Unit 1 refueling outage (SL1-16) to increase design margin and to eliminate the potential for pressure-locking concerns. Please contact us if there are any questions about this submittal.
Very truly yours, J. A. Stall Vice President St. Lucie Plant JAS/GRM Attachment cc: Regional Administrator, Region II, USNRC Senior Resident Inspector, USNRC, St. Lucie Plant 99052b03b3 9'70520 PDR ADGCK 05000335 P PDR an FPL Group company
St. Lucie Units 1 and 2 Docket Nos. 50-335 and 50-389 L-99-110 Attachment Page 1 NRC Question 1:
The Florida Power & Light Company (FPL) response to the first Nuclear Regulatory Commission (NRC) request for additional information dated July 31, 1996, states that calculations were used to verify that the Unit 2 pressurizer PORV block valves, V-1476 and V-1477, would operate during a pressure-locking event. These calculations were reviewed during NRC inspection 50-335/97-11, 50-389/97-11 and margin between actuator capability and the thrust required to operate during pressure-locking conditions did not meet ComEd Picslock guidelines. Explain why the current calculation is an acceptable long-term corrective action for pressure locking.
FPL Response 1:
FPL performed a calculation on July 23, 1996, using the "PRESLOK" computer model developed by Commonwealth Edison. The calculation addresses depressurization induced pressure-locking of the St. Lucie Unit 2 PORV block valves and compares the predicted unwedging load to the actuator capability at undervoltage conditions. The predicted unwedging load during this scenario is 9,712 lbs.
The actuator capability at the time of the calculation was 11,090 lbs. and provided a margin of 14.2%.
The actuators of these valves were modified during the Fall 1998 Unit 2 refueling outage (SL-2-11) to increase their thrust capability. The modification changed the gear ratio, which resulted in an actuator capability of 18,369 lbs. at undervoltage conditions. With this capability, the margin to unwedge the valves during the postulated pressure-locking conditions is 89%.
The ComEd Pieslock guidelines require an actuator margin of 20% for stiff valve body assemblies (i.e., greater than 600 lb. Class). This margin is intended to address uncertainty in the model, random variations in static unseating load, and measurement uncertainty in determining static unseating load.
The basis for the 20% iequired margin was provided to the NRC in ComEd's May 29, 1998, iesponse to NRC request 3. The St. Lucie Unit 2 PORV block valves are 3-inch, 1705 lb. Class Westinghouse gate valves. Therefore, the valves have a stiffvalve body per the above criteria and the 20% margin guideline is applicable. As such, the St. Lucie Unit 2 PORV block valves exceed the ComEd Pieslock margin guidelines, and therefore, the Pieslock iesults aie applicable for long-term design justification.
NRC Question 2:
It has been shown that the Unit 1 shutdown cooling hot leg suction valves, V-3480, V-3481, V-3651, and V-3652 are susceptible to piessuie-lochng. FPL has no plans to modify these valves to eliminate pressure-locking, but plans to use an analytical method to iesolve prcssure-locking concerns. Describe the calculation that was used to determine bonnet pressure and the calculation used to determine the thrust icquired to operate during pressure-locking conditions (include margin between actuator capability and pressure-locking thrust).
FPL Response 2:
Following the issuance of Information Notice 96-48 Supplement 1, on July 24, 1998, FPL re-evaluated all the motor-operated valves (MOV) in the NRC Generic Letter (GL) 89-10 program. The guidance provided by the Information Notice was incorporated into the St. Lucie GL 89-10 engineering design
St. Lucie Units 1 and 2 Docket Nos. 50-335 and 50-389 L-99-110 Attachment Page 2 documents. As a iesult of this work, a number of valves were identified as requiring modifications to increase the design margin. Gree Unit 1 shutdown cooling valves (SDC), V-3481, V-3651, and V-3652 were among those identified and are scheduled to be modified to eliminate the potential for pressure-locking during the up-coming Unit 1 refueling outage (SL1-16). The elimination of the potential pressure-locking condition for these valves will significantly increase the available design margin.
The methodology originally used to determine the thrust required to operate the SDC valves during potential pressure-locking conditions includes the margin between the actuator capability and the pressure-locking thrust. The Unit 1 SDC valves, V-3480, V-3481, V-3651, and V-3652 are 10-inch 1500 lb. Class flexible wedge gate valves manufactured by VELAN. These valves were originally evaluated for pressure-locking utiTizing the VELAN pressure-locking method. The total thrust iequuement to open the valve under pressure-locking conditions was calculated by adding the forces necessary to unseat the valve to the force due to a pressure-locking condition. The unseating force was calculated using the methodology provided by EPRI and the maximum allowable closing thrust.
Pressure-locking forces were calculated based on the methodology provided by VER¹ A bonnet pressure of 2235 psig (normal operating reactor coolant system (RCS) pressure) was utilized for calculating the pressure-locking forces. Credit was not taken for bonnet pressure decay during the six to ten hours the RCS pressure is reduced to SDC conditions. Ge design actuator capability of 35,425 lbs. was greater than the force required to overcome pressure-locking (32,580 lbs.) by 9%. This design margin also included conservative undervoltage factors, a COF = 0.2, and pullout efficiency.
For additional confidence, the Unit 1 SDC valves were also evaluated for operability using a pressure-locking methodology developed by MPR Associates, Inc. The results of the evaluation showed that there is sufficient actuator capability to overcome the iequired opening thrust for the potential pressure-locking condition. The MPR evaluation results also showed that V-3480 is insensitive to prcssure-locking. This is attributed to the fact that V-3480 was ieplaced in 1994 with a VIZ.ANflex wedge gate valve that has a disc which is significantly stiffer than the discs of the other three SDC valves.
Because of the relatively high stiffness of the valve disc, contact between the disc and body seats tends to be relieved by elevated bonnet pressure because the valve body expands more than the disc. Based on the evaluation results, V-3480 was evaluated using the MPR Associates, Inc. methodology for purposes of long-term design justification. MPR Associates, Inc. calculation number 002156044816 documents the methodology used to predict the opening thrust for flexible wedge gate valves. The model is based on the basic geometry of a flexible valve disc (i.e. two discs connected by a solid hub). To calculate the required stem thrust to open a valve, the model first calculates the amount of wedging interference that occurs when the valve is wedged shut. Using the calculated wedging interference, the model applies force and deflection relations to determine stem thrust required to unwedge the disc (i.e., pull the disc free of the interference). The model accounts for the differences in pressure and temperature conditions between when the valve is closed and when it is to be opened. The model was validated against applicable test data from the EPRI MOV program. The validation effort was documented in MPR Associates, Inc. calculation number 108-085-CSA-07.
St. Lucie Units 1 and 2 Docket Nos. 50-335 and 50-389 L-99-110 Attachment Page 3 For long-term design justification of V-3480, the worst case pressure-locking scenario was used as input to the MPR methodology. The bonnet pressure was assumed to be 2500 psia with the upstream and downstream pressures set at 14.7 psia. The valve internal dimensions were obtained directly from VELAN, the valve vendor, as part of the HPRI MOV Performance Prediction Program effort. The evaluation inputs were provided to MPR Associates, Inc. in the form of a verified FPL engineering evaluation. The results of the pressure-locldng evaluation showed that V-3480 was self-opening during a worst case pressure-locking event (i.e. 6 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> post-LOCA). Because of the relatively high stiffness of the valve disc, contact between the disc and body seats tends to be relieved by elevated bonnet pressure because the valve body expands more than the disc. For V-3480, the required thrust to open under pressure-locking conditions is bounded by the required unwedging thrust under conditions without pressure-locking. The unwedging thrust was previously evaluated to be 17,005 lbs.
The worst case actuator capability for V-3480, including loses for undervoltage and stem degradation, is 32,898 lbs. The actuator capability of 32,898 lbs. is significantly greater than the 17,005 lbs. required to open the valve. Therefoic, there is 93% margin available for the actuator to open V-3480 for the worst case design basis conditions. Based on the results of this evaluation, V-3480 does not require modification.
To summarize, FPL currently intends to modify the St. Lucie Unit 1 shutdown cooling valves (V-3481, V-3651, and V-3652) to eliminate the potential for pressure locking during the Unit 1 refueling outage (SL1-16) scheduled for September 1999. In the interim, the VELAN and MPR Associates, Inc.
methodologies show that sufhcient actuator capability is available to overcome any potential pressure-locking condition. Based on the above, V-3480 will not be modified.
NRC Question 3:
The 180-day submittal states that Unit 2 auxiliary feedwater pump steam admission valves, MV-08-12 and MV-08-13, are not susceptible to pressure-locking because the bonnets are always full of steam
. and willnot prcssure lock because steam is a compressible fluid. The NRC considers that these valves are not susceptible to thermally induced pressure-locking but normally are susceptible to hydraulic pressure locking.
Please provide detailed explanation as to why these valves are not susceptible to hydraulic pressure locking.
FPL Response 3:
Valves MV-08-12 and MV-08-13 aic the steam admission valves aligned to the Unit 2 auxiliary feedwater (AFW) pump steam turbine. A review of the original FPL response to GL 95-07 stated that the subject valves are not susceptible to thermally or hydraulically induced pressure-locking due to the valve and piping configuration. Both valves are installed in horizontal piping and oriented with the valve stem above the horizontal. Therefore, there is no concern with thermally induced prcssure-locking (i.e., condensation collecting in the bonnet region and subsequently pressurizing the bonnet due
St. Lucie Units 1 and 2 Docket Nos. 50-335 and 50-389 L-99-110 Attachment Page 4 to a temperature increase). However, the hydraulic pressure-locking concern requires further clarification.
Hydraulic prcssure-locking can occur in a fiex wedge or double disc gate valve that is subject to steam pressure ifa sudden depicssurization of the adjacent piping occurs. Valves MV48-12 and MV-08-13 are Anchor Darling double disc gate valves. Until the temperature of the valve cools sufficiently to reduce the pressure trapped in the bonnet, pressure-locking may occur. The original icsponse stated that valves MV-08-12 and MV-.08-13 could experience rapid deprcssurization upstream of the valve in the event of a main steam line break or a feedwater line break. A depiessurization of the upstream piping could cause the potential for hydraulic pressure-locking of the effected valve due to steam trapped in the bonnet. Since the design function of these valves is to open on an auxiliary feedwater actuation signal (AFAS) to supply steam to the APW pump turbine, the inability to open a faulted steam supply train steam admission valve that does not have steam in the piping (i.e. depressurized) is of no consequence and does not require any further evaluation. This would be considered part of the initiating failure.
The steam admission valve located in the non-faulted (intact) steam supply train is also subjected to a decrease in upstream pressure albeit to a lesser degree due to main steam and feedwater isolation. A review of Unit 2 Updated Final Safety Analysis Report shows that the most significant decrease in steam generator picssure occurs in the hot zero power steam line break with off-site power available.
This case maximizes RCS cooldown, which decreases the non-faulted steam generator pressure to 150 psia. For Anchor Darling double disk gate valves, the effect of pressure-locking can be calculated by the equation for parallel disc gate valves found in NUREGICR-6611, Results of Pressure-Locking and Thermal Binding Tests of Gate Valves.
Using the highest measured valve disc factor of 0.647 identified in the NUREG, a bonnet pressure equal to 1015 psig (main steam safety setpoint) and assuming the intact steam generator lowest pressure, the required opening thrust was determined to be 14,654 lbs. The actuator calculation for these valves shows that the actuator design capability in the opening direction is 16,011 lbs. Since 16,011 lbs. is greater than 14,654 lbs., 9.26% margin is available to overcome the potential pressure-locking forces. There is additional mMgin available within the conservative inputs, assumptions, and methodology.
NRC Question 4:
The 180-day submittal states that quarterly testing demonstrates that the Unit 1 high pressure safety injection pump suction valves Rom the shut down cooling (SDC) heat exchangers, V-3662 and V-3663, are not susceptible to pressure-locking because the valves are cycled after the low pressure safety injection (LPSI) pump is operated. Are the bonnets of V-3662 and V-3663 pressurized by the LPSI pump during this testing'I Explain why these valves aie not susceptible to pressure-locking following operation of the containment spray pumps for containment spray or surveillance and how this testing demonstrates that the valves will operate with reduced voltage to the actuator.
0 r
~ ~
St. Lucie Units 1 and 2 Docket Nos. 50-335 and 50-389 L-99-110 Attachment Page 5 FPL Response 4:
Unit 1 valves V-3662 and V-3663 are 4-inch fiex wedge gate valves. A review of plant drawings shows that V-3662 and V-3663 are separated Aom the LPSI pump discharge by normally locked closed valves, V-3452 and V-3453. Valves V-3452 and V-3453 are 10-inch gate motor operated valves that are maintained in acconhnce with the GL 89-10 Program. These valves are not opened during LPSI pump testing. Therefore, the bonnets of valves V-3662 and V-3663 are not exposed to LPSI pump discharge pressure (186 psig) during pump testing.
A review of plant drawings shows that valves V-3662 and V-3663 are exposed to containment spray (CS) pump pressure during CS pump testing. CS pump shutoff pressure is 222.9 psig. Following CS pump operation, the bonnet regions of valves V-3662 and V-3663 could remain pressurized following pump operation. However, a review of the emergency operating procedures (EOP) shows that a CS pump is started prior to opening either valve V-3662 or V-3663. Therefore, a differential pressure will exist across the valve prior to opening regardless ifthe bonnet remained pressurized following previous CS pump operation. The required thrust to open these valves at the maximum differential pressure conditions was previously evaluated at worst case undervoltage conditions as part of the GL 89-10 Program. Based on the procedural instruction provided in the EOPs that require the CS pumps to be started prior to opening these MOVs, no pressure-locking concern exists.
NRC Question 5:
The 180-day submittal states that Unit 1 shutdown cooling heat exchanger isolation valves V-3452, V-3453, V-3456, and V-3457, and Unit 2 shutdown cooling heat exchanger isolation valves, V-3456, V-3457, V-3517 and V-3658, are not susceptible to pressure-locking because for normal SDC system warm-up, the valves are opened with the LPSI pump running. Are these valves required to open for modes of operation other than SDC?
FPL Response 5:
Unit 1 A review of the design basis axtuirements for Unit 1 valves V-3452, V-3453, V-3456, and V-3457 was performed. The review shows that these valves are required to open for SDC warm-up, subsequently closed following SDC system warm-up, and then reopened with the upstream and downstream piping aligned to the RCS for SDC. The engineering documentation review also shows that the appropriate LPSI pump is started before the valves are re-opened, Therefor, a differential pressure will exist across the valve prior to opening regardless ifthe valve bonnet(s) remained pressurized following the previous LPSI pump operation. The required thrust to open these valves at the LPSI pump maximum differential pressure conditions was previously evaluated at worst case undervoltage conditions as part of the GL 89-10 Program.
A review of the EOPs and off-normal operating procedures (ONOP) was performed to verify that the appropriate LPSI pump is started prior to opening these valves. 'Ihe review showed that all the safety significant EOPs and ONOPs start+1 the appropriate LPSI pump prior to opening these valves with the
St. Lucie Units 1 and 2 Docket Nos. 50-335 and 50-389 L-99-110 Attachment Page 6 exception of thee piocedures. Two ONOPs were identified that showed that valves V-3456 or V-3457 may be opened without a pump opemting. The third procedure identified an EOP that started a CS pump prior to opening valves V-3456 or V-3457. A detailed review of the system conditions and configurations for these three exceptions concluded that the valves were capable of performing their design basis opening function. Based on pioceduml instructions provided in the EOPs and ONOPs that require the LPSI or CS pump to be started prior to opening these MOVs, including the exceptions identified, no pressure-locking concern exists.
Unit 2 A review of the design basis nxluiiements for Unit 2 valves V-3456, V-3457, V-3517, and V-3658 was performed. 'Ihe review shows that these valves are required to open for SDC warm-up and operation.
The engineering documentation review also shows that the iespective LPSI pump is in operation at the time the valves are opened. A review of the EOPs and ONOPs confirms that the appropriate LPSI pump is started prior to opening the valves. Therefore, a differential pressure will exist across the valve prior to opening regardless ifthe valve bonnet(s) remained pressurized following the previous LPSI pump opemtion. 'Ihe required thrust to open these valves at these maximum differential pressure conditions was previously evaluated at worst case undervoltage conditions as part of the GL 89-10 Program. Based on procedural instructions provided in the EOPs and ONOPs that require the LPSI pumps'to be started prior to opening these MOVs, no pressure-locking concern exists.
~ ~
P