ML20081E030
| ML20081E030 | |
| Person / Time | |
|---|---|
| Site: | Vermont Yankee File:NorthStar Vermont Yankee icon.png |
| Issue date: | 03/15/1995 |
| From: | Pelletier J VERMONT YANKEE NUCLEAR POWER CORP. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| BVY-95-33, NUDOCS 9503210117 | |
| Download: ML20081E030 (5) | |
Text
-
[j ;* ;,
-*f;
~ HVERMONT YANKEE -
LNOCL,EAit POWER : CORPORATION J
k Ferry Road, Brattleboro. VT 0530N002 y
[
ENGINEERING OFFICE ?
660 MAIN STREET BOLTON, MA 01740 I
(508) 779-6711-March 15,1995 -
BVY 95-33 U.S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, D.C. 20555
References:
See Attachment A
Subject:
Vermont Yankee Core Spray injection Valve Pressure Locking Evaluation in response to your request, Vermont Yankee is providing the results of our evaluation of pressure locking of the Core Spray injection valves. Our evaluation concludes there is reasonable assurance -
that the Core Spray injection valves will perform their required safety function, even if the valves are affected by pressure locking.
Pressure locking reviews were performed by Vermont Yankee in 1993 and 1994 consistent with the methodologies provided in References (c), (d) and (e). The reviews identified the currently closed injection valves, V14-11A and V14128, as being susceptible to pressure locking following a rapid -
depressurization during a large break LOCA. This is because high pressure water can become trapped in the bonnet resulting in forces which oppose movement of the disc from the seat and may prevent the valve from opening. Valve V1411A is considered to be more susceptible to pressure locking.
because high pressure water is known to be against the valve due to leakage past check valve V14-13A.
Vermont Yankee fully Intends to resolve the pressure locking concerns during the upcoming Refueling Outage by modifying the normally closed valves to preclude occurrence of this event. The Refueling Outage is scheduled to commence on March 17, 1995. To determine operability in the Interim, Vermont Yankee has performed a calculation which shows that sufficient thrust is available to open the -
valves, even if they are affected by pressure locking in the bonnet. --This calculation was performed using available industry equations and Vermont Yankee actualln-situ dynamic and static test data. The calculation and the basis for the inputs is further described in Attachment B to this letter.
Vermont Yankee has also reviewed our Individual Plant Examination (IPE) to assess the risk significance of the failure of the Core Spray injection valves. In this assessment, we conservatively assumed that the Core Spray system would fall for all accident initiators analyzed in the IPE. These events include large, medium, and small break LOCAs and transients with and without a scram. No other IPE assumptions were changed. The results showed that the calculated Core Damage Frequency (CDF) increased from the IPE baseline value of 4.3E-06 per year to 1.6E-05 per year. While no regulatory acceptance criteria has been established for CDF results, Vermont Yankee believes that the plant could be operated safely under these circumstances for the following reasons:
I 9503210117 950315 DR ADOCK 05000271 ll p
PDR 1
1
\\
~
"s VERMONT YANKEE NUCLEAR POWER CORPORATION 3
U.S. Nuclear Regulatory Commission March 15,1995 Page 2.
1.
The USNRC's draft, proposed safety goal for CDF is 1.0E 04 per year. Even with the Core -
Spray system assumed to always fall, our calculated CDF is approximately 6 times lower than the safety goal.
2.
Industry-wide basellne CDF's range from approximately 4E-06 per year to 8E-05 per year, with a mean of approximately 2.4E-05 per year. Even with the Core Spray system assumed to always fall, our calculated CDF is lower than the industry mean.
3.
A factor of ten increase in CDF is proposed by the USNRC Regulatory Review Group, " Risk Technology Application, Volume 4," as a threshold for judging the importance of a system in
" maintaining safety at the current estimated level". Even with the Core Spray system assumed to always f all, our calculated factor of four increase in CDF is less than the threshold increase.
In addition, no surveillance or other work activities are planned prior to the Refueling Outage which would affect the operability of either the Core Spray or Low Pressure Coolant injection systems or their support equipment. The LPCI Injection valves are not considered susceptible to pressure locking because the globe valves are normally closed and the gate valves are normally open at Vermont
~
Yankee. Given that we are planning to shutdown in the very near future and the relatively low risk significance if Core Spray were Inoperable, we conclude that the plant can operate safely until the planned refueling outage.
We trust that the enclosed information is satisfactory; however, should you have any questions or desire any additional information on this issue, please oo not hesitate to contact us.
Sincerely, VERMONT YANKEE NUCLEAR POWER CORP.
A%~
ames P. Pelletier Vice President - Engineering Attachments cc: (with Attachments)
USNRC Regional Administrator, Region I l
USNRC Resident Inspector, VYNPS USNRC Project Manager, VYNPS l
. ~
4-ATTACHMENT A REFERENCES a)
License No. DPR 28 (Docket No. 50-271) b)
Letter, USNRC to [All Licensees), " Safety-Related Motor Operated Valve Testing and Surveillance (Generic Letter No. 89-10) - 10 CFR 50.54(f)," NW 89-144, dated June 28,1989, c)
Letter, USNRC to [All Licensees], " Generic Letter 89-10, Supplement 6, Information on Schedule and Grouping, and Staff Responses to Additional Public Questions," NVY 94-51, dated March 8,1994.
d)
NRC Information Notice 92-26 " Pressure Locking of Motor-Operated Flexible Wedge Gate Valves," NVY 92-072, dated April 2,1992.
e)
AEOD Special Study S92-07, " Pressure Locking and Thermal Binding of Gate Valves," dated December 8,1992.
f)
" Calculation to Predict the Required Thrust to Open a Flexible Wedge Gate Valve Subjected to Pressure Locking," Entergy Operations, Inc., Presented at the NRC Public Meeting, New Orleans, LA, February 4,1994.
g)
EPRI Report TR 103229-V2, "EPRI MOV Performance Prediction Program, Gate Valve Model Report," Revision 2 September 1994.
h)
NUREG/CR 5807, " improvements in Motor-Operated Gate Valve Design and Prediction Models for Nuclear Power Plant Systems," May 1992.
Page A-1 l
g ATTACHMENT B VERMONT YANKEE CORE SPRAY INJECTION VALVE PRESSURE LOCKING EVALUATION Valves V14-11 A and V14-128 are the currently closed injection valves in the "A" and "B" Low Pressure Core Spray Subsystems, respectively. These valves are both 8 inch, 900 lb class, carbon steel, Walworth flexible wedge gate valves with Limitorque SB-2 motor actuators. The disc and valve seating surfaces are faced with Stellite 6.
Vermont Yankee has performed a calculation which shows that sufficient thrust is available to open the valves, even if they are affected by pressure locking in the bonnet. This calculation was performed using available industry equations and actual Vermont Yankee In-situ dynamic and static test data. The results of the calculation are as tollows:
For V14-11 A: Required Thrust
= 29670 lbf Available Thrust
= 29921 lbf M!nimum Margin 251 lbf
=
For 014128: Required Thrust
= 26464 lbf Available Thrust
= 29529 lbf Minimum Margin 3065 lbf
=
As discussed below, the actual margins are greater due to the conservative assumptions utilized in our calculation:
1.
The calculation was performed using the methodology developed by Entergy Operations, Inc. [ Reference f]. Pressure locking testing performed by Entergy indicates that this methodology provides the best available method for prediction of the forces required to open the valves under pressure locking conditions. This is the only method available with any test data to support it.
2.
The pressure in the valve bonnet is assumed to be equal to the normal reactor operating pressure of 1005 psig. The pressure on the reactor side of the valves is assumed to be equal to 300 psig, the low pressure open permissive setpoint for the valves. The pressure on the pump side of the valves is assumed to be equal to 275 psig, the pump minimum flow discharge pressure corrected for elevation. Vermont Yankee considers the use of normal reactor operating pressure to be conservative as it assumes zero leakage from the valve bonnet during the reactor depressurization.
3.
The motor output torque is assumed to have been reduced due to increased temperatures in the area of each valve. The EO LOCA temperatures are assumed. Vermont Yankee considers this to be conservative because the valves are required to open in the initial stages of a LOCA, prior to Reactor Building heatup. Lower area temperatures would result in higher motor output torques.
4.
The AC bus voltages are assumed to be equal to the degraded grid relay setpoint of 420 VAC. This is consistent with Generic Letter 89-10 recommendations. Vermont Yankee considers this to be conservative because higher AC bus voltages are expected during either minimum grid conditions or Emergency Diesel Generator operation. As a bounding case, Vermont Yankee evaluated the plant voltage distribution system using 110 kilovolts on i
Page B-1
p th's 115 kV line. Based on NEPEX studies, this is the lowest voltage that could exist on the 115 kV line before the grid becomes unstable and a loss of normal power occurs (resulting i
in a switch to the emergency diesel generators). Even in this extreme case the 480 VAC bus voltages are greater than 427 volts. Use of 427 VAC bus voltages would increase the available opening thrust by approximately 4000 lbf.
5.
A stem to stem nut coefficient of friction of 0.15 is assumed. Vermont Yankee considers this to be conservative based on actual Vermont Yankee in-situ dynamic and static test data. This data shows the stem to stem nut coefficient of friction at control switch trip to be i
less than 0.12. Vermont Yankee considers data at control switch trip to be applicable to the pressure locking conditions due to the similar load profiles. In both cases, the load at the stem to stem nut interface will be rapidly applied. This compares with a more gradual loading under dynamic conditions which would result in lubricant migration from the stem to stem nut interface and a higher coefficient of friction. Use of a 0.12 stem to stem nut coefficient of friction would increase the available opening thrust by approximately 4000 lbf, In addition to the 4000 lbf from item 4 above.
6.
A coefficient of friction between the valve disc and the seat contact surfaces,, of 0.317 is assumed. This value is obtained from actual Vermont Yankee in-situ static test data for valve V1411A, adjusted for possible equipment inaccuracles, and using the methodology provided in NUREGICR-5807. NUREG/CR 5807, Equations 2.3 and 2.4, are combined to determine the coefficient of friction from the static wedging and unwedging forces.
Vermont Yankee considers static wedging and unwedging forces to most closely represent pressure locking conditions. In both cases, valve internal friction is dominated by the flat-on-flat, valve disc to seat contact friction, which is properly represented by NUREG/CR-5807, Equations 2.3 and 2.4. For the remainder of the opening stroke valve internal frictions are dominated by the valve guide slot to guide contact frictions, which cannot be 4
accurately determined from static test results or analytical methods. The coefficient of friction between the valve disc and the seat contact surfaces, p, of 0.317 is consistent with the results of testing performed under the EPRI MOV Performance Prediction Program for Stellits 6 on Stellite 6, flat on flat sliding friction in water. Also, pressure locking testing performed by Entergy showed the use of the NUREG/CR-5807 equations to be consistent with the actua! test results.
Vermont Yanket recognizes that alternate methods exist to determine the coefficient of friction between the valve disc and the seat contact surfaces, p. As additionalinformation, Vermont Yankee has performed sensitivity studies which show that if the available thrust is increased by utilizing the identified conservatism in items 4 and 5 above, a p in excess of 0.61 may be assumed without impacting the operability of both valves V14-11 A and V14-128. A p of 0.61 is the bounding value for Stellite 6 sliding friction identifled in Table 5.1.1-2 of EPRI Report TR-103229-V2 [ Reference g].
Page B-2
.