DESIGN OPTIMIZATION OF THE TUBING ASSEMBLY …system under OBE and SSE conditions obtained from the geographical data along horizontal X, Y and vertical Z direction that is 2/3rd of

542

Int. J. Mech. Eng. & Rob. Res. 2014 K Sandhya Rani and V Krishna Reddy, 2014

DESIGN OPTIMIZATION OF THE TUBINGASSEMBLY STRUCTURE FOR SEISMIC

CONDITIONSK Sandhya Rani1* and V Krishna Reddy2

*Corresponding Author: K Sandhya Rani, [email protected]

In this paper, desired data from the modal analysis will be obtained from ANSYS using tubingassembly structure model. The FRS data is obtained from the geographical region where thisstructure will be mounted. Those data will be used for performing response spectrum analysis,which is essential data for designing a structure. Therefore, it was possible to determine whetherthe structure is safe through. In my current project the scope of work includes, Creating 3dmodel of the tubing assembly structure. Modal analysis to study the fundamental frequencies ofthe structure Dynamic analysis: To evaluate seismic response of the system under OperationBasis Earthquake (OBE) and Safe Shutdown Earthquake (SSE). The response spectra usedfor SSE in X, Y, Z directions are given as input. Calculation of stresses in the beam elements asper ASME section, sub-section NF for the above load cases are done, also Verification ofadequacy of welded/nut bolt joints are done. In this project “g” values at five locations wheremaximum accelerations occur are computed. Design optimization of the structure was donebased on the results obtained.

Keywords: Tubing assembly, Seismic conditions, FRS data, ANSYS

INTROUCTIONEven if the ground peak acceleration is a smallfraction of gravity (e.g., 0.2 g) the effects ofearthquakes on tubing assembly aresignificant and have to be carefully taken intoaccount in its design. A seismic event is in fact,in many cases, the most demanding loadingcondition, in particular for the interfacestructures (e.g., supports) which must be sized

ISSN 2278 – 0149 www.ijmerr.comVol. 3, No. 4, October 2014

© 2014 IJMERR. All Rights Reserved

Int. J. Mech. Eng. & Rob. Res. 2014

1 M.Tech Student, Department of Mechanical Engineering, Krishnachaitanya Institute of Technology & Sciences, Markapur, AndhraPradesh, India.

2 Professor, Department of Mechanical Engineering, Krishnachaitanya Institute of Technology & Sciences, Markapur, Andhra Pradesh,India.

for high strength, and often also for highstiffness.

The two main reasons why these events areof so much importance are as follows.

1. Seismic motion has a relatively broadbandspectral content which encompasses thetypical natural frequencies of structures(e.g., 1 to 33 Hz) with therefore spectral

Research Paper

543


amplifications likely to take place. Spectralamplification factors are a function of bothdamping and natural frequency of thestructures and are established byenveloping charts, called design responsespectra. The typical value of accelerationamplification factors (for structures withnatural frequency between 1 and 33 Hz) isabout 3.

2. Structures are usually weaker when loadedby horizontal inertial loads, even when theseare a small fraction of gravity.

The tubing assembly is a mounting structureused to mount 25 numbers of transmitters forcontrol systems in nuclear plants. In this paper,the tubing assembly structure is designed towith stand seismic vibrations.

METHODOLOGYA detailed Finite Element model was built withshell, beam and mass elements to idealizeall the components of the tubing assembly.Modal analysis was carried to capture thenatural frequencies and their massparticipations.

A Response Spectrum Analysis (RSA) wassimulated to evaluate seismic response of thesystem under OBE and SSE conditionsobtained from the geographical data alonghorizontal X, Y and vertical Z direction that is2/3rd of horizontal and the Square Root of theSum of the Squares (SRSS) method was usedto combine the total response in eachdirection. The stresses in the structuralmembers were calculated and comparedagainst the allowable stresses as per ASME

section, sub-section NF for the following loadcases:

• Level A condition: dead weight

• Level B condition: dead weight + OBE

• Level C condition: dead weight + SSE

Finite element analysis was carried in thefollowing steps:

• Static (dead weight) analysis of tubingassembly for the original model.

• Modal analysis for natural frequenciesand mass participation for the originalmodel.

• Response spectrum analysis for originalmodel under given OBE and SSEcondition.

• Static analysis (dead weight) of tubingassembly for modified model.

• Modal analysis for natural frequencies andmass participation for the modifiedmodel.

• Response spectrum analysis for modifiedmodel under OBE and SSE condition.

Material Properties:All the components of the tubing Assembly aremade using steel. All the components of thetubing Assembly are assigned as per thebelow material properties. The total weight ofthe tubing Assembly for Finite Elementsimulation is 350 Kgs.

Table 1: Material Properties

544


RESULTSFrom the above analysis the deflections andstress results are plotted. Maximum deflectionof 0.05 mm is observed.

Maximum Von Mises stress of 12.97 MPawas observed as shown in the below figure.

Figure 1: 3D Model of Tubing AssemblyUsed for Analysis

Figure 2: Boundary Conditions Applied onthe Tubing Assembly for Static Analysis

STATIC ANALYSIS OFORIGINAL MODEL (DEADWEIGHT)

Figure 3: Total Deflection Plot of theTubing Assembly for Level A Condition

Figure 4: Von Mises Stress Plot for Level ACondition

MODAL ANALYSIS OFTUBING

Table 2: Frequencies and MassParticipation

3D MODELING OFTUBINGASSEMBLY

545


Mode Shapes RESPONSE SPECTRUMANALYSIS OFTUBINGASSEMBLYLevel B: Dead Weight + OBE

Figure 5: Mode1 @ 16 Hz



Figure 8: Total Deflection Plot of theTubing Assembly for Level B Condition

Figure 9: Von Mises Stress Plot of theTubing Assembly for Level B Condition

Level C: Dead Weight + SSE

Maximum deflection of 0.4 mm is observed.


Foundation Bolt Calculation

The reactions at the anchoring locations:

546


The bolt with maximum reaction forces ischecked for strength. The bolts are checkedfor tension and shear as per ASME section III,NF 3324. In the present case M8 bolts areused. The maximum reaction force obtainedfrom the analysis in the tensile direction(vertical) is used to find out tensile stress onthe bolt and the resultant of shear forces inother two directions for shear stresscalculation.

22 /84.16 mmN

mmboltfoundationofAreaNforceTensileFt

Resultant shear force = Frs = Sqrt (Fx2 +Fy2)

2

2 /24.2 mmNmmboltfoundationofArea

forceShearResultantFs

From the above results, it is observed thatthe stresses are within the allowable limits;however it is observed that the bending of thestructure is dominant in the first and thirdvibration modes (Table 1). From the Table1 itis observed that 83% of mass participationexists at the frequency of 16.1 Hz in Y-dir and87% of mass participation exists at thefrequency of 28.6 Hz in X-dir. Design changesare incorporated to shift the dominant bendingmodes (1 and 3) by adding structural memberat the centre.

Results of Modified Tubing AssemblyFrom the above analysis the deflections andstress results are plotted for the modified

Figure 10: Total Deflection Plot of theTubing Assembly for Level C Condition

Figure 11: Von Mises Stress Plot of theTubing Assembly for Level C Condition

Figure 12: Reactions at AnchoringLocations

Table 3: Reaction Forces at AnchoringLocations

547


tubing assembly. Maximum deflection of 0.052mm is observed for level A condition.

all the three directions up to 30 modes arelisted in the Table.

RSA OF MODIFIED MODELLevel B Condition - Dead Weight + OBE

Figure 13: Total Deflection for Level ACondition

Figure 14: Von Mises Stress for Level ACondition


MODAL ANALYSIS OFMODIFIEDTUBINGASSEMBLYModal analysis was carried out to determinethe natural frequencies and mode shapes of astructure in the frequency range of 0-33 Hz.Eigen values and their mass participations in

Table 4: Frequencies and MassParticipation

Figure 15: Total Deflection Plot Level BCondition

Figure 16: Von Mises Stress for Level BCondition

548


The maximum stresses in the ModifiedTubing Assembly for level C are comparedagainst allowable stresses as per ASMEsection, sub-section NF. Maximum VonMisesstress of 6.75 MPa was observed

Level C Condition - Dead Weight + SSE

The maximum stresses in the Modified TubingAssembly for level C are compared againstallowable stresses as per ASME section, sub-section NF. Maximum Von Mises stress of17.32 MPa was observed.

The Reactions at the BoltingLocations

Figure 17: Total Deflection Plot Level CCondition

Figure 18: Von Mises Stress for Level CCondition

The bolt with maximum reaction forces ischecked for strength. The bolts are checkedfor tension and shear as per ASME section,NF 3324. In the present case M8 bolts areused.

The maximum reaction force obtained fromthe analysis in the tensile direction (vertical) isused to find out tensile stress on the bolt andthe resultant of shear forces in other twodirections for shear stress calculation.

22 /24.4 mmN

mmboltfoundationofAreaNforceTensileFt

Resultant shear force = Frs = Sqrt (Fx2 +Fy2)

Figure 19: Reactions at AnchoringLocations for the Modified Model

Table 5: Forces at Anchoring Locations

549


2

2 /93.1 mmNmmboltfoundationofArea

forceShearResultantFs

Tensile stress of 4.24 N/mm2 and Shearstress of 1.93 N/mm2 observed in the bolts.

RESULTS AND DISCUSSIONSeismic analysis of the Tubing Assembly forLevel A, level B and level C condition wasperformed. The original model was studied forresponse spectrum for the given X, Y and Zdirections. The stresses and deflections wereobserved to be below the allowable limits.However it was observed that dominantbending modes were existing within the 0-33Hz. A design change was incorporated byadding a structural member to shift the bendingmodes. The maximum stresses in the Originaland Modified Tubing Assembly for Level A,Level B and level C are compared againstallowable stresses as per ASME section, sub-section NF.

For the estimation of the structural integrityof the Tubing assembly, the allowable stressesare applied as follows

Allowable Stress (Ft: tensile, Fv: shear, Fa:compression, Fb: bending)

Ft = 0.6Sy ...(1)

Fv = 0.4Sy ...(2)

Fa = Sy (0.47–kl/r/444), kl/r < 120 ...(3)

Fb, major = Fb, minor = 0.66Sy ...(4)

Combined stress ratio ...(5)

(fa/Fa) + (fbx/Fbx) + (fby/Fby) < 1.0, fa/Fa< 0.15,

Figure 20: Comparison of Level A Stressesfor Original and Modified Model

Figure 21: Comparison of Level B Stressesfor Original and Modified Model

Figure 22: Comparison of Level C Stressesfor Original and Modified Model

CONCLUSIONThe original model was studied for responsespectrum for the given X, Y and Z directions.The stresses and deflections were observed

550


to be below the allowable limits. However itwas observed that dominant bending modeswere existing within the 0-33 Hz. A designchange was incorporated by adding astructural member to shift the bending modes.The maximum stress and translations are wellwithin the acceptable limits of the material. Theanalysis ensures that the Modified tubingassembly and foundation bolts are safe underspecified seismic load prevailing at site.

The Modified tubing assembly is thoroughlychecked under prevailing loads includingseismic acceleration forces and moments onthe flanges etc and is found safe from the pointof structural integrity and operability during aseismic event. The modified Tubing assemblywith all the structural members was found tobe adequately safe for the critical loadingconditions. M8 Bolt sizing are found more thanadequate, safe at most critical loadingconditions and meeting all seismicrequirements. The modified tubing assemblyis SAFE.

REFERENCES1. ASME (2001), ASME Boiler and

Pressure Vessel Code, Section III,Division 1, Appendices, Article N-1000:Dynamic Analysis Methods, AmericanSociety of Mechanical Engineers, NewYork.

2. ASME Section III, Division I, Appendix–N.

3. ASME Section III, Division I, Subsection-NB, NC, ND & NF.

4. Bhardwaj S A (2001), “Broad Steps inEarthquake Resistant Design of aNuclear Power Plant”, NPCIL ReferenceArticle, Vol. 15, Nos. 1-4.

5. Buchholdt H (1997), Structural Dynamicsfor Engineers, Thomas TelfordPublications.

6. Igor J Karassik (2001), Messina, Cooper,Heald, Pump Handbook, 3rd Edition,McGraw-Hill Publications.

7. Richart F E, Woods R D and Hall J R Jr(1970), “Vibration of Soils andFoundations – Prentice – Hall”,International Series in Theoretical andApplied Mechanics, Prentice Hall Inc.,Englewood Cliffs, New Jersey.

8. USNRC (1996), Standard Review Plan forthe Review of Safety Analysis Reports forNuclear Power Plants Report No.NUREG-0800, Section 3.9.3 (Rev.3),ASME Code Class 1, 2, and 3Components, Component Supports, andCore Support Structures, US NuclearRegulatory Commission, Washington DC.

9. Veletsos A S and Verbic B (1973),“Vibration of Viscoelastic Foundations”,Earthquake Engineering and StructuralDynamics, Vol. 2, April, pp. 87-102.

Documents

DESIGN OPTIMIZATION OF THE TUBING ASSEMBLY …system under OBE and SSE conditions obtained from the geographical data along horizontal X, Y and vertical Z direction that is 2/3rd of