ML20206J127

Pressure Locking Thrust Evaluation Methodology for Flexible Wedge Gate Valves

ML20206J127
Person / Time
Site:	Grand Gulf
Issue date:	04/30/1998
From:	Danni Smith ENTERGY OPERATIONS, INC.
To:
Shared Package
ML20137N361	List: ML20206J094 ML20206J120 ML20206J127
References
ME-98-002-00, ME-98-2, NUDOCS 9905120121
Download: ML20206J127 (16)

Text

• '

CNRO-99/00006 Attrchmtnt 4 ENGINEERING REPORT ME-98-002-00 PRESSURE LOCKING THRUST EVALUATION METHODOLOGY FOR FLEXIBLE WEDGE GATE VALVES l

t l

t l

l

)

1 l

~

9905120121 990504 PDR ADOCK 05000416 P PDR _

J

Engineering Report No.: ME-98-002-00 '

Page 1 of 15 ENTERGY OPERATIONS Engineering Report Ftr "Entergy's Pressure Locking Thrust Evaluation Methodology for Flexible Wedge Gate Valves" (Application Guide)

APPLICABLE SITES ANO Unit 1: l X l GGNS: l X l W 3: l X l ANO Unit 2: l X l RBS: l X l ECH: l l Safety Related: x Yes No Prepared by: DangE. Smith M

Date: 11/4/98 1 p 'nsibfe Engineer [/

Reviewed by: =

Date: //[2c[fg

/

@eviewer Reviewed by: M. $64 M Date: #[2</[P Supervifor/ Reviewer Approved by: Date: #/!30/P7

[fesponsible CDE Manager Ifor multiple site reports only)

Engineering Report No.: ME-98-002-00 Page 2 of 15 Page No.

1.0 IN D USTR Y B A C KG RO UND ...................... ........................................................... 3

2. .....................................................................................................................

3. .............................................................................................................

4. 0 G UI D E L I N E S . ..... ... ........ ..... .. . .. .. ... .. .. ... ... ...... . .. ....... ... .... . ........ . ....... .. . ...................... .. 5 5.0 M E TI I O D O L O G Y .. .. . . .... . .. . ... . . ..... .. .... . .. .. . .. ... . .... .. ..... .. .... .. . ........... ... .. .... ........ .......... . 7 .

6.0CIIARTS....................................................................................................................13

l 9

Engine ring Report No.: ME-98-002-00 Page 3 of 15 l l

1.0 Industry Background i 1

l INPO SOER 84-7 reported a number of events involving flexible wedge gate valves that failed to open. These failures have prevented safety-related systems from functioning when called upon. Binding of the valve disc in the closed position due to high pressure water trapped in the bonnet cavity (i.e., pressure ,

locking) represents a potential common mode failure mechanism for these valves. 1 In GL 89-10, the NRC licensees were asked to provide additional assurance of the capability of safety-related MOVs and certain other MOVs in safety-related systems to perform their safety-related functions. In GL 89-10 licensees were asked to review MOV design basis, verify MOV switch settings, test MOVs under i design-basis conditions, improve evaluations of MOV failures and trend MOV l problems. Supplement 6 to GL 89-10 described an acceptable approach to {

address pressure locking of motor-operated gate valves.

l GL 95-07 required licensees take actions as necessary to ensure that safety-related I power operated gate valves susceptible to pressure locking are capable of performing their required safety functions. I Significance:

Binding of flexible wedge gate valves in the closed position due to thermal binding or bonnet pressurization is a significant safety concern for the following reasons: (1) flexible wedge gate valves are used in a variety of applications in nuclear power plant safety systems; (2) the valves may be required to open during or immediately following postulated design basis events; (3) the events that most i severely challenge plant safety (i.e., the design basis accidents) usually involve the most rapid system cool down and depressuri::ation rates and possibly the j largest pressure differentials in and across these valves. Valve operators were not originally sized to open the valves with high pressure fluid trapped in the valve bonnet or when excessive binding forces are applied to the disc. Pressure locking is a phenomenon that can cause the unseating thrust for a gate valve to increase j significantly above levels expected for design basis conditions. This can result in the valve actuator having insufficient capability to open the valve. In addition, this can result in valve damage in cases where the actuator capability exceeds the a valve structural limits. l

Engineering Report No.: ME-98-002-00 l

Page 4 of 15 2.0 Purpose 2.1 Provide guidelines for using Entergy's Pressure Locking Thrust Evaluation Methodology to evaluate MOV actuator minimum required open thrust during pressure locking situations. The Entergy Method provides a conservative estimate of the minimum required opening thrust -

for a flexible wedge gate valve under pressure locking conditions. When the appropriate inputs are specified in accordance with this application guide, Entergy valve test data indicates that the predicted thmst will exceed the measured thrust.

2.2 Provide a common screening process for determining when a valve may be susceptible to pressure locking. This logic is presented in Charts 1 and 2.

3.0 References 3.1 Formulas for Stress and Strain, Roark and Young,6th Edition.

3.2 NUREG/CR-5807, Improvements in Motor Operated Gate Valve Design and Prediction Models for Nuclear Power Plant Systems, Wang and Kalsi, May 1992.

3.3 NUREG/CR-0146, Workshop on Gate Valve Pressure Locking and Thermal Binding, Calculation to Predict the Required Thrust to Open a Flexible Wedge Gate Valve Subjected to Pressure Locking, Dana Smith, February 4,1994 3.4 Wyle Laboratories Test Report 45161-0 dated Dec. 15,1997, " Pressure Locking and Thermal Binding Test Program on Two Gate Valves with Limitorque Actuators for Entergy Operations", Vendor Doc # M-J5.08-Ql-45161-0-8.0 1-0.

3.5 Engineering Report ME-98-001-00, " Pressure Locking and Thermal Binding Test Program on Two Gate Valves with Limitorque Actuators."

3.6 NUREG-1275, Volume 9, " Pressure Locking and Thermal Binding of Gate Valves".

3.7 Engineering Report GGNS92-035," Evaluation of Safety Related Gate  !

Valves for Susceptibility to Thermal Binding and Bonnet Pressurization".

Engine: ring Report No.: ME-98 002-00 Page 5 of 15 3.8 Wyle Laboratories Test Report 43008-01 dated Feb. 18,1993," Flow Loop Differential Pressure and Pressure Lock Tests on a 14-inch William Powell Gate Valve for Entergy Operations, Inc., Grand Gulf Nuclear Station", Vendor Doc # M-J5.08-Ql-43008-01-8.0-1-0.

4.0 Guidelines 4.1 The total required thrust for opening any flexible wedge gate valve under pressure locking conditions is dependent on the final wedging force from the previous closing cycle. For a given torque switch setting, the wedging-force from valve closure can vary because the inertial overshoot is affected by the magnitude of the differential pressure across the disc. Typically, the highest wedging force would be introduced when the valve is closed without differential pressure. The static unseating force used in the Entergy Method represents the open packing load and pullout force due to wedging of the valve disc as a result of the previous closure. These loads are superimposed on the loads due to the pressure forces on the disc that occur during pressure locking. The Open Unseating Thrust (Trace Point 09) from the last MOV Static Test Analysis shall be used for the static unseating thrust. Appropriate instrument uncertainty shall be conservatively applied to this value (i.e., take thrust to highest positive value).

4.2 The Entergy Method is used to obtain a bounding minimum required open thrust under pressure locking conditions. Design Basis Pressure Conditions in the valve at the time the valve is required to open shall be obtained. This includes the upstream (P,), downstream (P,,,,), and bonnet pressure (Pm ,). The Entergy method is directionally specific. The highest pressure side shall be designated as the upstream side in the calculation.

4.3 Valve Disc Geometry, including the hub radius (b), mean seat radius (a) and seat angle (0) shall be obtained. When the hub cross-section is not circular (e.g. many Westinghouse gate valve designs), then an effective hub radius which corresponds to a circle of equal area to the hub cross-sectional area should be used.

4.4 Coefficient of Friction should be selected based on Figure 1. If the conditions are determined to be within the predictive range, then a friction factor of 0.4 should be used. If outside the predictive range, actual friction factors may be used, if available. See section 5.9 for the basis for this recommended practice. Figure 1 is based on an evaluation of the test results as presented in Reference 3.5. The friction factor can be

. Engineering Report No.: ME-98-002-00 Page 6 of 15 determined from static test analysis data in accordance with the method presented in Reference 3.5. If static test analysis data determines the actual coefficient to be greater than 0.4, then additional testing or evaluation of valve behavior should be considered since experience indicates that friction factors do not exceed 0.4 on the open stroke for a valve that is functioning properly.

4.5 Poisson's ratio (v) for the disc base material shall be obtained.

Predictive vs Conservative Range 7

-100 h a \

- 80

~ 60 (

a 2

- 40 g e

Conservative [

Range - 20 E Predictive a Range 5

O 0 20 40 60 80 100

{

g Upstream Pressure (as a % of Bonnet Pressure)

Figure I i

I

J Engineering Report Ns.: ME-98-002-00 i Page 7 of 15 .I i

i 1

5.0 Methodology 5.1 The flexible discs act like two uniform thick, flat circular plates with a fixed hub connection at the center.

l 5.2 The analysis will be broken down into four load cases and then numerically added together.

5.3 The high side pressure loads on the hub area will be transferred directly to the low side disc.

5.4 Boimet Pressure Forces The internal pressure will be analyzed using Ref. 3.1 Table 24 Case 2d.

This will provide the force, Qa, in units oflbs/in exerted on the seat ring by the disc because of the internal pressure. This load case is shown in Figure 2.

O at

$ qt qt Q

at j

N psig psig 7

%N ^ Si r g3g3g a J

%+ '

\ 91

+%

psig g s- s s s Figure 2 Calculate the force on the perimeter of the disc hub, (Ob,) as; using (Case 2d):

C2L 3 - Cslu Qs,.q,a C2Co - C1Cs'

Engineering Report Ns.: ME.98 002-00 Page 8 of 15 Where:

r  %

. C2 = p2r1 + 21n ,3~

3 1

4

<aj < b> _

~

C3 = - 4 b 'b +1 In a +

'b-

-1 -

4a

<a, .b _ga,

~

/ b'

C=I -

2 1 + v + (1 - v)k a) .

'1 - b

b l+v 1-v C9 = - < In a + -

a 2 b 4 s,a r r ,.,3 4 r ,.,3 2 r ,,,3 2 '

Ln = ; < l + 4 ,., s 2 - 5 -4 2+ In , -

64 ga, ga, <a, _

ga, _ ro

' ~

r ,.,3 4' r,,3 2 Ln = ; < l ;_p 1 1 + (! + v)ln , -

4 4 <a, s a,_

r...

Then the force on the perimeter of the disc at the seat ring, (Qa,) is; using (Case 2d):

Q,i = G6, 9' 2 a 2a(a -r.')

5.5 Differential Pressure Forces The external pressures, on the high pressure side and on the low pressure side, will be analyzed using Ref. 3.1 Table 24 Case 2d and Case 1b in combination.

5.5.1 Reduction in Bonnet Pressure Forces First the reduction in seat forces imposed by bonnet pressure are calculated using Case 2d This load case is shown in Figure 3.

Engineering b ' t No.: ME-98-002-00 Page 9 of 15 v _

i 75 y 92 / 93

% y psig Q a psig 2 Y )

High Pressure Low Pressure

/ I \

q2 f 2 3 x & 92 psig /

) psig hK L f Q.,

Figure 3 Forces are transferred as a perimeter line load to the disc hub, because of the pressure acting against the disc on the high pressure side. Calculate the force on the perimeter of the disc hub, (Qb,) as; using (Case 2d):

C2Ln - CsLu Qs, = qsa C2C9 - C3Cs l

Where the equations for the C and L variables are given above. )

The hub perimeter line load, (Qb2), will be used to determine Ga2, which reduces the bonnet pressure forces at the perimeter of the disc. Calculate the force on the perimeter of the disc at the seat ring,(Ga2) as; using (Case 2d):

Q,, = G3, 9' 2 a 2a (a -r.2)

Substitute the low pressure, q,, to obtain Gb, and pa, for the low pressure side disc. l l

l l

1

. Enginasting Report No.: ME-98-002-00 Page 10 of 15 5.5.2 DP Forces on Low Pressure Side Disc From Hub Area -

. This load case is shown in Figure 4 and accounts for the differential pressure on the hub area that was left out of the Case -

2d equations. This results in a net increase in the force to the low

. pressure side disc.

75- J a

92 N

' o b*

-)-: r,s j x q3 g

psig h i

( psig bd g

/

liigh Pressure Low Pressure h

Figure 4 For this case, convert the differential area load to a perimeter line load at the hub,(Qb) as:

'1 Qu . (y,-y,)nb' l

2nb l

d

)

4

1 l

Engineering Report No.: ME-98-002-00 Page iI of 15 1 5.5.3 Total Force on Low Pressure Side Disc Seat I

The Case Ib analysis is used to analyze the transfer of hub perimeter forces calculated in 5.5.1 and in 5.5.2 to the seat ring on the low pressure side disc. Case 1b is considered to be appropriate since the disc hub will provide a rigid " guided" type of j configuration in the center of the disc. The load case is shown in i Figure 5. l l

I

$ 0., \

D' O.

d \ \ l

/ N /N / g, r g3g3g a

) ( Y%3 b2

~ s.

Figure 5 l 1

l The difference between Ob2 and Gb, and the hub differential I I

pressure Ob, will be combined and used as Win the Case Ib equation.

i W=(Q62-Qu)+Q64 I

Force on the perimeter of the disc at the seat ring, (Ga,) due to the j differential loading, is; using (Case 1 b): l i

Q,, = -W E a

Where:a = % Valve disc sealing diameter (Mean seat diameter) b = Valve hub radius r=b o

q = pressure over hub area

Engineering Report No.: ME-98-002-00 Page 12 of 15 5.6 Summary of Forces Acting at Each Seat The force on each seat ring due to the four pressure loading conditions will be added together and multiplied by the mean seat ring circumference to deterrnine the total force.

1. Force on the high pressure side seat ring of the valve will be, (Ibs):

Fu = (Gai- G, )(2m)

2. Force on the low pressure side seat ring of the valve will be, (Ibs):

Ft = (Gui- G,i + G,,)(2m) 5.7 Valve Disc Factor The relationship between the Valve Disc Factor and coefficient of friction, p, will be determined using Ref. 3.2 Equation 2.2a.

)

DiscFactor = #

cose +psin 0

)

Where: = coefficient of friction )

0= seat angle, 5.8 Total Pullout Thrust Total required thrust to open the valve will be the addition of the two l forces determined at the seat rings above times the Valve Disc Factor plus the unwedging thrust. Unwedging thrust may be determined by using the last measured static open thrust.

Tww. = (Fu + Ft.)DiscFactor + Tuswstxis a

Engineering Report No.: ME-98-002-00 Page 13 of 15 5.9 Connvative Assumptions The method does not attempt to model the expected reduction in disc seat forces as the valve pressure state transitions from a low-high-low condition to a high-high-low condition. This would require that valve flexibility be accounted for which is beyond the scope of this simplified -

analysis. ' As the transition occurs, the method increases in conservatism since the bonnet pressure force on the seats is assumed to remain constant.'

In reality this component diminishes to zero at the high'-high-low state.

Seat force reduction during the transition is a function of valve / disk structural stiffness.

Test data summarized in Reference 3.5 clearly demonstrates this behavior.

Reasonable prediction occurs for cases where upstream and downstream -

pressures do not exceed 40% of the bonnet pressure. This range is shown -

in Figure 1. For conditions outside this range, the method yields a conservative value for design.

To ensure a conservative estimate in the predictive range,. a 0.4 friction factor should be applied. This provides enough margin to avoid underprediction of required pullout thrust as demonstrated in Reference 3.5. However, outside the predictive range, actual friction factors may be applied, if available, since a large degree of conservatism is introduced by the assumption that bonnet pressure forces on the discs do not diminish ~

with increasing DP.

6.0 Charts 6.1 Chart 1: Hydraulic Locking Review Logic 6.2 Chart 2: Boiler Effect Review Logic m.

Engineering Report No.: ME-98-002-00 Page 14 of 15 Chart 1: HYDRAULICLOCKING REVIEWLOGIC REVIEW SYSTEM NO.

AND THE FLEX WEDGE VALVE DATA P YES ALVE LOCKED IN POSITION YES OR HAVE POWER REMOVED?

NO '

IS VALVE PROVIDED WITH AN- YES OPEN BONNET DRAIN?

NO IS VALVE QUIRED TO OPEN DURIN NO DPCOND YES VALVE IS NOT

- IS PIPING YES $USCEPTIBLE PRESSURE PRIOR TO ,

TO HYDRAULIC OPENING 2 BONNET LOCKING PRESSURE?

NO PRIORITY VAI VE NON-PRIORITY VALVE ES VALVE PERFO YES NO SUSCEPTIBLE TO AN ACTIVE SAFETY SUSCEPTIBLE TO HYDRAULIC LOCKING FUNCTION TO OPEN? HYDRAULIC LOCKING "GL95-07" 4

Engineering Report No.: ME-98-002-00 Page 15 of 15 Chart 2 BOILEREFFECTREVIEWLOGIC REVIEW SYSTEM VALVE RT OF A YES AND GASEOUS VALVE DATA SYSTEM 7

' gg NO NO VALVE INSTALLED BOVE HORIZONTA OSITION' IS YES VALVE LOCKED IN YES POSITION OR HAVE POWER REMOVED 7 NO IS

'ALVE PROVIDED YES WITli A NORMALLY OPEN BONNET DRA NO VALVE IS NOT IS VALVE IN NO SUSCEPTIBLE ICINITY OF A HEAT SOURC TO BOILER WHICH COULD HEATUP EFFECT BONNET FLUID?

YES PRIORITY VALVE NON-PRIORITY VALVE SUSCEPTIBLE TO SUSCEPTIBLE TO ANA ES TY BOILER EFFECT BOILER EFFECT FUNCTION TO OPEN?

"GL95-07"

..