ML20042B863

Stress Analysis for Demonstration of Operability of Purge & Vent Valves During Dba.

ML20042B863
Person / Time
Site:	Three Mile Island
Issue date:	03/08/1982
From:	Elam B, Jasani K, Rochino A, Shiau A, Wilson R GENERAL PUBLIC UTILITIES CORP.
To:
Shared Package
ML20042B862	List: ML20042B861 ML20042B863
References
RTR-NUREG-0737, RTR-NUREG-737, TASK-2.E.4.2, TASK-TM 283, 283-R02, 283-R2, NUDOCS 8203260220
Download: ML20042B863 (20)

Text

'ATTACID1EUT 1 , u g u ,,:;g ,

283 gh] TDR NO. REVISION NO. _ _,2 TECHNICAL DATA REPORT I 16 CNVNYNO.k[207If_ PAGE OF DEPARTMEf'T/SECTION Eng hem ing & Des @n Three Mile Islanu, Unit 1 RELEASE DATE 9/30/81 REVISION OATE 3/ / 2* _

DOCUMENT TITLE: .

Stress Analysis for Demonstration of Operability of p.i r n o ,nn vnni valvon nnrino nonian nanin Accidonts ORIGINATOR SIGNATURE DATE APPROVAL (S) SIGNATURE DATE i 1 .>.t . . . _ . ?

/). C. &a q/j{l?l N. M. Jasani k)f AI ,&* 9llNIN.

A. C. Shiau A. P. Rochino 7[/6[f/

. lWn 'k/tY/ry A

APP,ROVAL FOkEXTERNAL DISTRIBUTION DATE h hk. k i\

o DISTRIBUTION ABSTRACT:

• R. C. Arnold a. Brief Statement of Problem G. R. Capouanno P. R. Clar< In accordance wita Reference 6.3 it is request-D. K. Croneberger ed that the analysis be performed to verify B. D. Elam that containment purge and vent valves are R. F. Evers capable of performing their intended function K. M. Jacant without damage to eritical valve components R. d. Keaten during the Design Basis Accident - Loss of G. P. Miller Coolant Accident (DBA-LOCA) loads ana that the J. P. Moore, Jr. valves will close when fluid dynamic torques A. P. Rocnino are introduced.

A. C. Shiau D. G. Slear According to Reference 6.6, the valves to be B. G. Wallace analyzed are:

F. Weinzimmer R. F. Wilson 1. Spring Closed valves All-V-1A anu AH-V-lD 7, y {g*fg (Reference 6.7).

h,lA/. M4M[G 2. Motor operated valves AH-V-1B and Ad-V-lC (Reference 6.7).

The stress analysis results for demonstration of operability of purge and vent valves per NRC guidelines dated 9/27/79 (Reference 6.8) are documented in Section 3.0 of this TDR.

b. Summary of Key Results

1. The valve operator.3 are ante to resist tne reaction of LOCA induced fluid dynamic tor-ques-for valve opening angle up to 30 from closed position (see Taole 3 in Appendix).

2. Tne calculated stress levels of the valve components under combined seismic and LOCA conuitions meet tne code allowable stress limits, or tne LOCA allowable limits of 8203260220 820322 90% of tne yield strengtn except tne stress 5 PDR ADOCK 05000289 PDR p Accoco30 7.s i

Page la in the shaft. The calculated shaft stress is 2% over the code allowable stress limit (See Table 5 in Appendix).

Our engineering judgement is that this slight overstress f*. the shaft would not create a failure situation,

c. Conclusion

1. The actuator works in cooperation with the fluid dynamic torque to close the valve. Bearing friction and seat / disc friction are the only significant effects which restrain valve closure except that, for All- V1B/1C, the motor operator speed limits the closure rate.

/ ..

2. The closing ability of the spring closed valves is assured if the valve opening is limited to 30' plus 1.75* tolerance or less, (AH-V-1A and AH-V-1D).

The closing ability of the motor operated valves is assured if the valve opening is limited to 30* plus 3.29' tolerance or less (All-V-1B and All-V-lC),

d. Recommendations Pending any further analysis and NRC acceptance the valves should be limited as follows:

1. 30* plus 1.75* tolerance open or less from the closed position for All-V-1A and All-V-lD.

2. 30* plus 3.29* tolerance open or less from closed position for AH-V-1B and AH-V-1C.

t a

c.,---

TDR NO'

~

EsOSTE 283 TITLE III-l - Stress Analysis for Demonstration of Operability PAGE OF of Purge and Vent Valves During Design Basis Accidents y REV

SUMMARY

OF CHANGE APPROVAL DATE 1 Page la - Valve angle opening tolerances added. WL {/dity4cg Is i t'l ti pn%a 1 Page 3 - Wording changes in Paragraph D. )//C (S/ E,_., ,,f. jf, 1 Page 6 - Conservative assumptions added. M 'd Q10pOlt twen.n it j iy iI 1 Page 7 - Valve angle opening tolerances added d'l0 lil hl e l d.1)p\h $3!!i re 1 Page 8 - Valve angle opening tolerances addedb ]/d d.\\\Raaty 14 Ulbl chtiff.

i l

l l

l l

l A000 0017 9 80

E N 283 TITLE TMI-1 Stress Analysis for Demonstration of Operability PAGE OF i

of Purge and Vent Valves during Design Basis Accidents 11 REV

SUMMARY

OF CHANGE APPROVAL D' ATE 2 Page la - Wording changes in paragraphs "c" and O [ [b "d" per plant engineering request, k ,

y t

l l

l l

e A000 0017 9 80

Three Mile Island Unit 1 Stress Analysis for Demonstration of Operability of Purge and Vent Valves During Design Basis Accidents Table of Contents Title Page Abstract

a. Brief Statement of Problem 1

b. Summary of Key Results 1

c. Conclusion la

d. Recommendations la Section 1.0 Purpose and Summary 2 2.0 Methods 4 3.0 Results 6 4.0 Conclusion 7 5.0 Recommendations 8 6.0 References 9 7.0 Appendix 10 i

-iii-

1.0 PURPOSE AND

SUMMARY

The purpose of this TDR is to document tne results of the analysis for the containment purge and vent valves regarding structural adequacy to withstand the fluid dynamic torques which would occur during the faulted condition of a loss of coolant accident (LOCA) within the containment vessel and the design basis seismic loads.

According to Reference 6.6, the valves to be analyzed are the air operated valves AH-V-1A and AH-V-lD (Reference 6.7), and the motor operated valves AH-V-1B and AH-V-lC (Reference 6.7). These valves are Pratt 48 inch butter-fly valves, model RlA. Stress analysis is performed to show the structural adequacy for a valve opening of 30 or less from the closed position.

In summary, the NRC guidelines for demonstration of operacility of purge and vent valves dated 9/27/79 (Reference 6.8) has been incorporated in this evaluation.

A. Considerations Per NRC guidelines (Reference 6.8) the following considerations have been addressed.

1. Valve closure time during a LOCA will be less than or equal to the no flow time demonstrated during shop tests, since the fluid dynamic effects tend to close the valve.

2. To qualify valves for an opening of 30* or less from the closed position, the maximum differen-tial pressure across the valve per Reference 6.6 are used in the analysis.

3. Worst case is determined as a single valve clo-sure with containment pressure on one side of the valve and atmospheric pressure further downstream.

4. Containment back pressure will have no effect on closing the valve.

5. The subject valves do not use accumulators.

C. There are no torque limiting devices for tne air operated valves All-V-1A and All-V-lD. The se t-tings of the torque limiting devices for the l

electr ic motor operated valves All-V-1B and All-V-lc are compatible witn the torques required to operate the valve during the design basis conditions.

i

7. The effect of upstream piping is ignored as a conservative approach.

8. The valve disc and shaft orientation does not affect torque calculations.

B. Stress Analysis Stress analysis of the valve components under com-bined seismic and IDCA conditions is performed using the design rules for Class 1 valves as detailed in Paragraph NB-3540 of Section III of the ASFE Boiler and Pressure Vessel Code (Reference 6.1, hereafter referred to as the Code). The calculated stress levels are conpared to code allowables, if possible, or the IDCA allowables of 907 of the yield strength of the material used.

C. Operator Evaluation In evaluating the structural integrity of the valve operators, the calculated torque during IDCA is conpared with the maxinun torque rating of the operator per manufacturer's data.

D. Sealing Integrity Decontamination chemicals have very little effect on EPT and stainless steel seats. Molded EPT seats are generically known to have a maxinun cirmulative radiationresistanceof1xg0 rads at a maxinun incidence temperature of 350 F.

Valves at outside ambient temperature below 60 F, if not properly adjusted, may have leakage due to ther-mal contraction of the elastomer, however, during IDCA conditions, the valve internal tmperature muld be expected to be higher than ambient which tends to l increase scaling capability after valve closure. The

presence of debris or damage to the seats m uld obviously i

impair sealing. To ensure sealing integrity, the 'DII-1 l preventive maintenance program requires that the valve j seats be cleaned and inspected periodically for damage and deterioration and replace if required. The valves being containment isolation valves are also required to pass the local and Integrated Irak Rate Tests.

l l The seats on three of the four valves were replaced prior to perfonning and passing the Integrated Irak Rate Test on July 5, 1981.

l l

l i

2.0 METHODS This investigation consists of fluid dynamic torque cal-culations, valve stress analysis, and operator evaluation.

2.1 Torque Calculations The torque of butterfly valve at any opening position is the summation of fluid dynamic torque and bearing fric-tion torque at any given disc opening angle.

Bearing friction torque is calculated from the following equation:

Tg = O P(A) (U) (.5d)

Where, AP = Pressure differential, psi.

A = Projected disc area normal to flow, in2, U = Bearing coefficient of friction.

d = Shaft diameter, in.

Fluid dynamic torque is calculated from the following equation:

TD=Ct ' OP Where, D = Valve diameter, inches.

6P = Pressure differential, psi.

Ct = Torque coefficient.

l The detailed torque calculations are documented in Reference 6.9.

2.2 Valve Stress Analysis This analysis used the design rules for class 1 valves described in paragraph NB-3540 of Section III of the ASME Boiler and Pressure Vessel Code (Ref erence 6.1) . The requirements for class 1 valves are much more explicit

! -4 _

._ - _- - - _ ._ - . _ - . . . = _ _ - . _.

• than for either class 2 or 3 design rules. The analysis is conservative .since the design rules for class 2 and 3 valves are exceeded by tne rules for class 1 valves.

l Valve components are analyzed by hand caculations under the assumption tnat the valve is either at maximum fluid i

dynamic torque or seating torque during the LOCA condi-

tions against tne maximum design pressure or the maximum 1 differential pressure across the valve per Reference 6.6.

?

The.SSE seismic accelerations are simultaneously applied

! in each of tnree mutually perpendicular directions.

A natural frequency analysis is performed for valve com-ponents in Reference 6.9. Based on the frequency results, the seismic loads are conservatively taken as 1.5 times of the acceleration levels given in Reference. ,

6.4. The acceleration constants gx, gy and 9z represent accelerations in the x,y and z directions

.t respectively. The coordinate system is defined as the x axis along the pipe axis, the z axis along the shaft

< axis, and the y axis mutually perpendicular to the x and z axes. Valve orientation witn respect to gravity is taken into account by adding an equivalent 19 load to the seismic load in the proper direction. The acceleration constants used are summarized in Table 1 in Appendix.

The detailed stress analysis is given in Reference.6.9.

The calculated stress values are compared to code allow-ables, if possible, or LOCA allowables of 90% of the yield strength of the materials used. Code allowable stress levels are Sm for tensile stresses and 0.6 Sm for shear stresses. Sm is the design stress intensity value as defined in Appendix I, Table I-1.1 of Section III of the Code. The valve component materials are listed in Table 2 in Appendix.

2.3 Ooerator Evaluation The maximum operating torque for valve due to flow under specified LOCA conditions as calculated in Section 2.1 is used to ver ify the structural adequacy of the operator.

i The valve operator structural evaluation is based on a comparison of tne calculated torque against tne operator

. ability to resist the reaction of LOCA induced fluid dynamic torques per manufacturer's data (Reference 6.5).

3.0 RESULTS The results for torque calculations are sumnrized in Table 3 in Appendix. The maxinun torque absorption capability based on manufacturer's advice is also presented in the Table. The evaluation shows that the opgrators are structurally adequate for valve opening angle up to 30 from closed position. Table 4 in Appendix shows the mininun valve body wall thicknesses versus code required mininun thicknesses. All the valves satisfy the mininun wall thickness requirement of the Code.

, The calculated stress levels of the main elements of the valves are listed in Table 5 in Appendix. The results indicate that the valve .

components stresses meet the code allowable stress limits, or the IDCA allowable limits of 90% of the yield strength except the shaft stress.

The shaft stress is 2% over the code allowable stress limit (see Table 5 in Appendix). However, based on our engineering judgnent, the 2%

overstress in the shaft will not create failure situation.

The following conservative assuuptions were made in the analysis:

1. It was assuned to have an instantaneous reactor building pressure of 50.5 psig maxinun. Ilowever, the magnitude of the actual dynamic torque will be substantially less than was used in the analysis because the air-operated valves AH-V1A and AH-V1D will close in less than 2 seconds and the motor-operated valves AH-V1B and NI-VlC will close in less than 5 seconds long before the reactor building pressure of 50.5 psig is attained in approximately 10 seconds after a IDCA event. The pressure buildup in containment after 2 seconds is approxinntely 16 psig and after 5 seconds is approximately 35 psig.

2. Even though it is highly unlikely that both the seismic and IDCA conditions happen sinultaneously, the analysis was based on the worst case of having two abnonnal conditions occurred together.

3. Throttling effects from the inside containment valves RI-V1B and AH-V1C were not used. 7he analysis assumes that the inner valvas fail wide open and that the outer valves AH-V1A and AH-VlD will have to close against the fluid dynamic forces.

., v - . - - - - - - - - - - - - -

4.0 00tCwSI0ti i

Allthevalvesagestructuaflyadequateifthevalveopeningangle is limited to 30 pig 1.75 tolgrance or less for NI-V-1A and NI-V-1D valves and 30 plus 3.29 tolerance or less for NI-V-1B and Mi-V-1C valves from closed postion.. 'Ihis is based on consideration of combined effects of IDCA, pressure load, and DM seismic loads.

Structural adequacy is assured for the operators and the valve cmpenents.

l l

f i

l i

(

I

i. .. __ ___ _ , , - ,_ _ _ _ _ _

5.0 RE00tfEHRTIONS

1. To ensure structural integrity, ghe valve opening nust be limited to 30 plus 1.75 tolerance open or less fmn the closed position for MI-V-1A and NI-V-1D valves.

2. To ensure strgetural. Integrity, the valve opening nust be limited to 30 plus 3.290 tolerance open or less from closed position for NI-V-1B and NI-V-1C valves.

I I

i l

l

_p_

6.0 REFERENCES

6.1 ASME Boiler and Pressure Vessel Code,Section III, 1980 Edition.

6.2 Steel Valves, ANSI B16. 34-1977.

6.3 R. F. Evers' letter to D. K. Croneberger dated October 21, 1980.

6.4 Letter from R. M. Rogers of Gilbert Associates, Inc.

to D. G. Slear dated July 16, 1980. " Seismic Response Curves for the Reactor Building Shell for TMI Unit 1."

6.5 Valve Applicable Data from Manufcturers, A. C. Shiau letter to A. P. Rochino dated 9/16/81.

6.6 J. P. Fritzen's letter to J. R. Holstrom of Henry Pratt Company, dated May 4, 1979.

6.7 Gilber t Associates, Inc. drawing #4192, E-311-861 Rev. 10, and drawing #4192, E-311-833, Rev. 19.

6.8 The NRC Guidelines for Demonstration of Operability of Purge and Vent Valves, dated 9/27/79.

6.9 TMI-l Purge and Vent Valve Analysis Calculation Book, Calculation No. 1302X-322C-A44.

b i

l .

t l

7.0 APPENDIX Tables 1 tnrough 5 are presented in this Appendix.

TABLE 1 SEISMIC LOADS DIIUDCTIGI 10YTir.TMTIOti IIVEIS OF

/CPPTTIMTIGI Sluft Axis is Vertical Shaft Axis is IIorizontal (NI-V-1B and A!!-V-lC) (Ni-V-1A and NI-V-lD)

Values Given Values Used Values Given Values Used in Ibf. 6.4 in the Analysis in Pcf. 6.4 in the Imalysis g

x 0.5g 0.759 0.5g 0.75g (pipe axis) g y 0.5g c: 0.75g (0.25+1) g (0.375+1)g gg (0. 25+1) g (0.375+1)g 0.5g 0.75g (shaft axis) l

. _ .. _.___ __ _=___ __ ..__ . . .

TABIE 2 bMTERIAIS IVR VALVE C0ff0tHNTS l

1

VALVE COMPOtHNIS FETERIAIS FOR ALL VALVES f

i ,

Body A91M A-36 l 4

Disc AS'IM A-36 Shaft AS'IM A-276, Type 316, Condition A Shaft Key AISI C1045, C.D. Stl, i

Disc Pins AS'IM A-276, Type 316 i 30ttorn Cover Plate AS'IM A-36 4

1 Thrust Bearing AS'IM B-164, Condition A

=

Operator Bolts _ SAE Gr. 2 Trunnion Body ASTM A-36 L j Trunnion Bolts SAE Gr. 2

+

)

4 h

4

] . . .

_ _ . _ _ _ . . _ _ . _ . ~ __ - - -

1 , ,

i i

TABIE 3 SGTMIU OF 'IOIOUE NIALYSIS FOR ALL VALVES

! Max. Allowable Torque VALVE OPEtIING 'IUI'AL OPEPATING for Operator .(in-lb.)

j NGE ( 0) TOIQUE (in-lb. ) Wu W" (NI-V-1A & -(NI-V-1B &

NI-V-lD) NI-V-lC) 0 (Fully Closed) 64111 70000 153600 5 2667 3 n

10 5901 i

15 9335 20 15033 25 23570 2 30 35069 42500 49000 i

35 56212 i

i 40 79816 l 45 102067 50 117064 i

l 55 138457

! 60 174535 i

! 65 209803  !

I i 70 257926 '

t 75 307347 80 248970 1 I  !

85 155583 Y f 90 (Fully Open) 104 70000 153600 g

i

J TABLE 4 11INIftUf1 BODY WALL THICKNESS VALVE DESIGNATION VALVE SIZE ACTUAL MINIMUIl BODY WALL CODE REQUIRED MIN.

(in.) THICKNESS ( in.) THICKNESS PER ANSI 16.34 (in.)

AH-V-1A, AH-V-1B AH-V-lC, AH-V-lD 48 2.125 0.49 k

s

TABLE 5 Sl29WT OF STRESS NMLYSIS ,-

VALVE STRESS ___ STRESS _IIVEL (osi) ALIDGBLE STRESS COMPOTE 72 IEE A'O SYfDOL AII-V-1A&NI-V-lD NI-V-1B&All-V-lC (psi)

Prirnry I'embrane Pm 990 990 Sm 0.99 12600 , 27000 Prirnry plus secondary Op 2970 2970 an 0.9 G~

stress due to internal 12600 ,

2700[

pressure Pipe Axial Ped 2542 2542 1.5 Sm 0.97 Body reaction 18900 , 27000 stress Bending Peb 9164 9164 1.5 Sm 0.9 f f

~18900 , 27000 Torsion Pet 4773 4773 1.5 Sm 0.9 fy 18900 . 27000 Thermal secondary Qt 1197 1197 Sm 0.9h

, Stress 12600 , 27000 Prirnry plus Sn 13428 13428 3 Sm I 0.9 f secondary stress 37800 > 27000 Disc Cambined Bending S(l) 4643 4643 Sm Stress on disc centerline 12600 Torsional Sheer Stress S(9) 5102 5102 0.6 Sm 12000 Cmbined Shear Stress S(6) 6983 6960 0.6 Sm 12000 Sluft Combined stress S(4) 30748 30734 1.5Sn (shear and bending) 30000 Shaft Key Shear stress on key S(16) 10467 10467 0.9 81000 Disc Pins Shear stress in S(17) 13890 13890 0.9 pins 27000

.~

TABIE 5 (OTE'D)

VALVE STRESS STRCSS_IEVEL_(psi) ALIDG\BIE STIESS -

COMPCG2E IN E AND SYMBOL NI-V-1A&NI-V-lD NI-V-1B&NI-V-lC (psi)

Bearing stress on S(22) 92 169 Sm thrust collar 13600 g Shear stress in S(27) 460 843 0.6 Sm adjusting screw 8160 Cmbined stress in S(28) 3325 6096 Sm retainer bolts 13600 Shear tear-out of S(31) 472 866 0.6 Sm thrust retainer bolts 8160 Shear tear-out of S(33) 1097 1237 0.6 Sm cover bolts thru tapped 7560 holes Cover Shear tear-out of S(34) 818 922 0.6 Sm i Plate cover bolt head thru 7560

[ cover Canbined stress in S(30) 673 942 Sm cover 12600 Shear tear-out of S(42) 962 751 0.6 Sm trunnion bolts in top 7560 trunnion taprxx1 holes Bearing stress of S(43) 3069 2213 Sm Operator trunnion bolt on tapped 12600 lbunting hole in trunnion

'Ibnsion in bolt on top S(47) 5452 4259 Sm trunnion + 12600 S(48)

Shear due to torque S(50) 5364 4252 0.6 Sm on trunnion bolts 7560 ccmbined stress in S(46) 10021 7454 Sm trunnion bolts 12600 Combined stress in S(53) 10198 8037 Sm operator bolts 12600