Text

Flowserve RAL-7483 Rev 0 Design Modification Report Assy 93-14347, 10/26/2004
	ML050140247
Person / Time
Site:	Hope Creek
Issue date:	10/26/2004
From:	; Flowserve Corp
To:	; Office of Nuclear Reactor Regulation
References
	TAC MC5111 RAL-7483, Rev 0
	Download: ML050140247 (59)
	v • d • e

FLOWSERVE

, e Anchor/Darling Valves SWA7P Valves Edward Valves Vahtek Control Products Worcester Valves Flow Control Division RAL-7483 Rev 0 Design Modification Report PSEG Nuclear LLC Hope Creek Generating Station Customer P.O. No: 4500257459 Size 20 Class 900 Carbon Steel Gate Valve S.O. 31042-02 Assembly Drawing 93-14347 Prepared by:

Date:

Reviewed by:

Date:

Approved by:

Date:

"I, the undersigned, a Registered Professional Engineer experienced in the design of valves, verify that to the best of my knowledge, information, and belief, this design report complies with the requirements of the ASME Code for Nuclear Power Plant ComponentsSection III, 1974 Edition with Summer 1975 Addenda and the conditions of the design specifications. Pursuant to paragraph NCA-3350 of said Code, this certification is solely for the purpose of complying with paragraph NB-3560 of said Code and is not lo be construed as involving, modifying, or changing contractual relationships or legal liabilities."

Certified By:

Date:

Certified Dy:

Date:

REPORT NO.:

RAL-7483 FLOAISERVE REV. 0 DATE: 10/2612004 PAGE i Flow (:ontrnl Division Rale:igh. XVC Design Source:

Design Inputs:

Customer P.O.: 4500257459 Design Specification: BFN-50-C-7105 Rev. 7 ASME Section III, 1974 Edition, Summer 1975 Addenda.

Literature Searches:

Not Applicable Assumptions:

None Computer Calculations/Programs:

Valve Data:

ANSYS Workbench 8.1 Size 20 Class 900 Carbon Steel Gate Valve Sales Order 31042-02 Assembly Drawing 93-14347 Rev. D

REPORT NO.:

RAL-7483 FLOV SE W E REV. 0 DATE: 1012612004 "I

tPAGE ii Flow Control Division Rale~igh. NVC RECORD OF REVISIONS Rev.

Pages Description 0, 10-25-04 All Initial Issue

- REPORT NO.:

RAL-74B3 FLOVWSERVE REV. 0 DATE: 1012612004 PAGE 3 Flow Control Division Raleigh. XC 1 DISCUSSION This report is prepared to provide analysis results that support a proposed modification (see ) to the subject valve. The goal of the modification is to reduce the effect that vibration due to pump induced chugging has on the valve's bevel gear actuator cover assembly. Based on historical damage, the original configuration is subject to fatigue damage and fretting contact due to the vibration. It is assumed that, because the valve's natural frequency is in a range that matches chugging frequencies, significant accelerations can result.

This analysis may be considered a supplement to the existing Class I design and seismic analysis prepared by Anamet Laboratories Inc., Lab No. 79.016. It is concluded here that the valve can withstand the specified chugging loading in addition to the seismic loading addressed in the Anamet report.

The analysis was performed using the finite element based design simulation package called Workbench by ANSYS, Inc. along with the solid model design package Pro/E. The procedure involved creating the solid model of the valve extended structure in Pro/E, porting the model into Workbench, setting up analysis scenarios, solving the model, and preparing a report. contains the report that was generated from Workbench that documents the analysis performed. The report contains six scenarios: the first three address the proposed modification, while the last three address the original state. Each set of three scenarios address (1> natural frequencies of the valve extended structure, (2) response of the extended structure due to a broad spectrum acceleration input at the base of the extended structure, and (3) stress analysis due to conservative accelerations based on broad spectrum response at the top of the extended structure.

In addition to the workbench analysis, the capscrews that mount the actuator cover are investigated in section 2 using classical methods using input from the Workbench results. Only the original configuration is analyzed. The modified configuration is not addressed because the added struts add significant support due to their attachment to bevel gear housing, which greatly relieves the load on the capscrews.

The broad spectrum acceleration input (0.3 g's) is based on accelerometer measurements.

Measurements were taken at low frequency and high frequency pump speeds, and the high frequency speed resulted in the higher acceleration. The 0.3 g's input envelopes the square root of two times RMS measured values, which all parties involved have agreed is adequate input for evaluating fatigue damage. In addition, results are such that it may be concluded that response to peak input is not a critical design concern without performing additional analyses.

Due to the valve's extended structure height, the input acceleration is greatly amplified. The analysis uses the response at the top of the structure as input for the static stress analysis. This is conservative as the amplification decreases rapidly from structure top to bottom. For instance, results show that the amplification at the base of the stem protector is about half that at the top of the stem protector. Thus, results are extremely conservative for portions lower on the structure such as the yoke legs a-n'd ldyd;even so, stress is shown to be lovw in these structures.

Section 1.1 provides an overview of results. It is concluded that the proposed modification does not significantly increase the topworks natural frequency: the natural frequency of the original configuration meets the specified criteria. In addition, deflections are reduced slightly, but it may not be significant enough to reduce the possibility of fretting contact. However, it is demonstrated that the modified assembly improves the ability of the assembly to withstand the measured high cycle loading.

The given stress limits are based on allowed fatigue stress at one million cycles, which is considered infinite fatigue life, from the ASME Boiler and pressure vessel code,Section III appendicies, Appendix

9. Note that these fatigue curves are conservative by the greater of a factor of two on stress or a factor of 10 on number of cycles (RE Section III Code Appendix III).

0 REPORT NO.:

RAL-7483 FLOWSERVE REV. 0 DATE: 1012612004 PAGE 4 Flow Cnimtm lDivision Raleigh. NC 1.1

SUMMARY

Description Value Limit Criteria (original/modified)

Natural Frequency [Hz]

First: 94. / 106.

> 90 Hz Second: 97.6 /115.

Peak Acceleration Response Gy:.91 /0.65 N/A

[g's]

G,,

17.1 / 15.5 Gm: 12.2 /10.35 Stress Range in Topworks (Conservative),

6.0 /4.0 12.5 other than bolting [ksil (re fig. 1-9.2)

Capscrew Stress Range [ksi]

14.98 / N/A 20.

(re fig. 1-9.1)

Deflection at Top [in]

0.54 /0.36 N/A Other

Conclusions:

Because stress in the added struts is low at the locations of required welding, it is not required to provide weld metal equal in cross-section to the area of the strut. Conservatively, since the maximum topworks stress is one third the limit, it is permissible to provide weld area equal to only one third the strut area. However, it is important to place the welds at the corners of the struts for maximum strength.

Because the handwheel is loosely connected to the impactor by a retaining ring, it is highly recommended to remove the handwheel so as to remove potential fretting contact.

1V

4 REPORT NO.:

RAL-7483 FLOWSERVE REV. 0 DATE: 1012612004 NoPAGE 5

Iglow Lontrol Division RAeigh. VC 2 ACTUATOR COVER MOUNTING BOLTING ANALYSIS The basic approach is to determine the axial force due to bending and thrust for one bolt located at the maximum distance from the pivot point. This assures that the maximum force is considered.

Reaction forces are determined using data from the solid model and from the analysis input.

Assumptions:

1.

The stiffness of the flange is much greater than the stiffness of the bolts. Hence, Hooke's law and bolt spacing geometry are used to determine bolt forces.

2.

The bending pivot point is located on the bolt circle.

Maximum Reaction Force Due to Moment (Al) for One Bolt:

1.

Sum of Moments about pivot = 0 yields:

K M

=

g2F;Xj, K=n/2 -1I.....................................2.1 ToLa1 oF NB BoiLs 7

rA/

Bolts X1 k

Top View Showing Typical 601 PivoL 6

X/,-Po i nL Ff t

Xi 2F1k 2F1 2F1

R 1 t t

Thr-usL IT I d VIe Someng A I

1 Side View Showing Assumed Bolt Deflections and Reaclion Forces Bolt Pattern

.1.

REPORT NO.:

RAL-7483 FLOW'SERVE REV. 0 DATE: 10/26/2004 PAGE 6 FlowA Contriol Division Raleigh. NC Maximum Reaction Force Due to M: (Cont'd)

2.

Applying Hooke's Law:

F

=

EA,(611L)

Fj1F,

=

(A,6/(Aj6.5)

(AXd(Aj)........................

2.2 where A]J

= Boltarea = 2AB forall ij E

=

Young's Modulus L

=

Bolt Length (All Assumed equal)

3.

Maximum Moment Induced Bolt Force:

FK

=

M(AKXV1/(T,2X/Ad........................

2.3 For a four-bolt flange with all bolt areas equal, this reduces to F, = Al/f(X).

Stress in Bolt:

a

=

FKIAB + Thrus/l[nAa]........................

2.5 T

=

Shear/[nAB] + 2*Torquel[ndAa].......................

2.6

REPORT NO.:

RAL-7483 FLOVJS ERVE REV. 0 DATE: 1012612004

>.PAGE 7

IFlow lontrol Division Raleigh. NC 2.1 COVER CAPSCREWS ANALYSIS RESULTS (ORIGINAL)

Top Works Data Parameter Description W

Weight XV Vertical C.G.

XH Horizontal C.G.

GRII Acceleration (resultant of horizontal values)

GRV Acceleration (resultant of horizontal values) 68.

12.9 2.6 21.0 lb in In g's Rld Solid Model Solid Model Solid Model Analysis Analysis

.91 g's External Forces Vi, Sh TN TI MN M

QN Tc Bolting Properties n

Ni D

BC A8 St d

Be X2 B(

X3 B(

rameter Description hear Force (WGRH) trust Force (WGRv) oment (WGRHXv) orque (WGRHXH) rameter Description Lumber of Bolts olt Size ress Area of One Bolt 1t Circle Diameter olt Group I Pivot Alt Group 2 Pivot-Alt Group 3 Pivot Value 1428.

61.88 18,421.

3713.

Value 6

.25 3.182E-2 12.75 7.509 18.13 25.64 bnits lb lb in-lb in-lb Units in in2 in in in in Bolting Statics Analysis Results Parmeter escrintion r, K Max. Bolt Force From Moment Bolting Stress Analysis Results CT Bolt Axial ri

B6lt Slhear Ss<,AX Stress Intensity SI, Stress Range (SUAjx2)'

Value Knin 667.3 lb RQE 2.3 B&L 2.5 2.6 Value 21.29 10.53 29.95 14.98 Units ksi ks'i -

ksi ksi

' Since the capscrews experience tension only, it is appropriate to divide the resulting stress by two to yield the appropriate alternating stress range for entering the design fatigue curve.

REPORT NO.:

RAL-7483 FLOWSERVE REV. 0 DATE: 10/26/2004 PAGE 8 Flow Control Di~vision l.eg.N ATTACHMENT 1 SKETCH OF PROPOSED MODIFICATION

1-~REPORT NO.:

RAL-7483 FLOWSE RVE REV. 0 DATE: 1012612004 PAGE 9 Ilox. (Cont)rol Divisioti Rea ik h. iVC FindJ Descriplion McIeriol Oly A

Plale (3 x.375)

C.S.

4 B

Box Iron (2x2x.25)

C.S.

A 8.5 NOTE:

MODIFICA7ION INCLUDES REMOVAL Or THE HANDWHEEL (26 LOS).

NET ADDED WEIGHT EQUALS 30 LBS.

01 TOPWORKS MODIFICATION SIZE 20 CLASS 900 GATE VALVE

REPORT NO.:

RAL-7483 F LOaiW SE WvE REV. 0 DATE: 1012612004 PAGE 10 Flown Contrul Division Raleaigh. NC ATTACHMENT 2 ANSYS SIMULATION REPORT

I l., Summarv This report documents design and analysis information created and maintained using the ANSYSO engineering software program. Each scenario listed below represents one complete engineering simulation.

Scenario 1 Based on the Pro/ENGINEER' assembly 'zvhomelrfarreIlA31042 (20-900-FW)AFTXED.ASM.2".

Considered the effect of body-to-body contact, acceleration and structural supportS.

Calculated harmonic results.