STRUCTURAL ANALYSIS FOR STRENGTH AND FATIGUE LIFE OF HALF COUPLING WELDMENT FOR LARGE COOLING WATER PIPES

Transcription

1 STRUCTURAL ANALYSIS FOR STRENGTH AND FATIGUE LIFE OF HALF COUPLING WELDMENT FOR LARGE COOLING WATER PIPES KUNAL S BHATT, SARBJEET SINGH SANDHU, TARUN KUMAR SHARMA ITER-India, Institute for Plasma Research, Ahmedabad, India kunal.bhatt@iter-india.org SHRISHAIL B PADASALAGI, ADITYA PRAKASH, MAHESH JADHAV ITER-India, Institute for Plasma Research, Ahmedabad, India Abstract ITER cooling waters system consists of large piping network to remove the heat load of about 1100 MWatt through various branched connections. Many of the branches are connected to main pipes by half coupling full penetration weld joints. There is requirement is to have full penetration for all the joints, however quality classification (QC-2), recommends only 10% testing of the total weldments. In view of this it is expected that there can be some joints with little or no penetration. The above requirement demands for the structural strength and fatigue life assessment to ascertain that components is not failing even if there is no weld penetration. The design by analysis approach is considered for structural and fatigue life assessment, for maximum expected loads combination case. The weld joint is structurally qualified using ASME code. Fatigue life of weld joint is calculated using both ASME Section VIII Div.2 and RCC-MR RR The maximum stress and fatigue life observed for full penetration case is 92 MPa and cycles as per ASME and cycles as per RCC-MR. Whereas, in no penetration case, the stress is 188 MPa and fatigue life is cycles as per ASME and 1500 cycles as per RCC-MR. It is concluded in the paper that weld joint is safe for both the case in most severe load case combination. 1. INTRODUCTION One of the objectives of successful operation of ITER is to remove the excess heat that the ITER Tokamak and plant auxiliary systems will produce. All of this heat needs to be dissipated to the environment. The ITER Cooling Water System (CWS) is designed to carry out this objective and the same is divided in to the sub-systems such as Tokamak Cooling Water System (TCWS), Component Cooling Water System (CCWS), Chilled Water System (CHWS) and an open-loop Heat Rejection System (HRS) [2]. These cooling water systems have large piping networks to accomplish the objective of removing the heat. The large piping network consists of many branches connected to the main pipe by half coupling full penetration weld joints as shown in figure OBJECTIVE FIG. 1. Branch pipe connected to main pipe by half coupling weld joint For quality control requirement, the half coupling weld joints are qualified as per Quality classification (QC-2) procedures. QC-2 recommends inspection only 10% of total weld joints. As remaining 90% of the weld joints do not undergo inspection, there are the possibilities that some of the weld joints may have partial or no weld penetration, which may adversely affect the strength and life of welded joint. Therefore these weld joints need to 1

2 FIP/P3-32 be assessed for structural strength and fatigue life in order to ensure safe working under operating load conditions. The objective of current paper is to assess strength and fatigue life of weld joint in case of no penetration. 3. WORKFLOW Two cases were considered for the assessment. The first is No penetration joint, in which it is assumed that weld material has not at all penetrated the joint. This case is preferred over partial penetration due to its conservative nature. Second case is that of Full Penetration, in which it is assumed that weld material has fully penetrated in joint, as per the requirement. This case is considered for comparison purpose. Finite element analysis has been performed for both full weld penetration and no weld penetration cases to assess and compare structural strength and fatigue life [1]. Detailed fatigue assessments are carried out for both the joints using ASME [4] and RCC-MR [5] codes. Comparison is discussed at the end. 4. FINITE ELEMENT ANALYSIS Main Run pipe and branch pipe along with half coupling is considered for the assessment. Two cases of Penetration are separately modeled using ANSYS Design Modeler Software [7] Modeling of full and no penetration joint FIG. 2. Sectional view of half coupling weld joint A three dimensional Model of both Full penetration and No penetration is immaculately prepared as per original drawing, with particular attention towards modeling of weld region. For Full penetration weld joint, Welded section is modelled in Single beveled groove, with complete weld penetration up to the inner periphery of half coupling beyond root region of weld. FIG. 3. Mesh details of full penetration weld joint

3 For No Weld Penetration, fillet weld is only modelled in single beveled groove section, and the weld does not penetrate beyond root region. FIG. 4. Mesh details of no penetration weld joint 4.2. Meshing Complete model with Main Run Pipe, Branch Pipe, Half coupling and Weld joint are meshed for both the cases using ANSYS Finite Element Package, v 17.0 [7]. Complete model is meshed with hexahedral SOLID 186 elements. Mesh Quality control is strictly enforced by keeping a check on mesh parameters such as element skewness ratio, aspect ratio, jacobian ratio etc. For weld region, a very detailed fine mesh is created, by ensuring a minimum of six layers of elements across the thickness of weld region. To avoid contact elements formulations, conformal mesh is generated between weld joint, half coupling and pipes. For the main pipe smooth mesh transition is obtained from fine mesh near weld region to coarse mesh at regions far away from weld in order to limit the model size. Mesh sensitivity check is performed to arrive at optimized mesh model. Mesh sensitivity is a convergence of results for specific mesh density value. TABLE 1. MESH PARAMETERS Parameters Full Penetration No Penetration No. of Elements Element type Hexahedral Hexahedral Skewness ratio FIG. 5. Meshed model 3

4 FIP/P Material properties The material considered is ASTM A 53 Gr. B, Type S [6]. The properties considered are listed in the table below. TABLE 2. MECHANICAL PROPERTIES Material property value Density rho (kg/m3) 7883 Modulus of Elasticity E (GPa) 203 Poisson s Ratio mu Allowable stress Sm (MPa) Loads TABLE 3. LOAD COMBINATIONS Load combination/single loads X(mm) Y(mm) Z(mm) Rx(degrees) Ry(degrees) Rz(degrees) P1+T1+Dw+D 1.4e-2 3.5e e-3 1e-4 9.3e-2 Expansion 1.4e-2 2.4e e-3 1e e-2 Where, P1 - Design pressure, T1 - Operating temperature, Dw - Dead weight, D - Nozzle displacement For conservative approach, only most severe load case combination are considered [3]. Loads are applied in terms of displacements in all six (translation & rotational) degrees of freedom. Loads are applied in terms of displacements at the node at half coupling joint location. The displacements are obtained from CAESAR tool for above mentioned load cases. Main pipe is considered fixed at both the ends. 5. RESULTS TABLE 4. FULL WELD PENETRATION Load case Von Mises Bending Membrane Peak General membrane Fatigue Life (ASME) Fatigue Life (RCC-MR) P1+T1+Dw+D Root e5 2e5 Toe Expansion Root e5 >10e6 Toe TABLE 5. NO WELD PENETRATION Load case Location Location Von Mises Bending Membrane Peak General membrane Fatigue Life (ASME) Fatigue Life (RCC-MR) P1+T1+Dw+D Root e Toe Expansion Root e Toe CONCLUSIONS a) For both the load cases, von mises stresses in root region of weld joint increased twice fold approximately, in case of No weld penetration as expected. This is due to stress concentration at weld irregularity.

5 Significant increase in membrane and peak stresses is also observed. Minimum factor of Safety for Full weld penetration joint is 1.74, whereas it is reduced to 1.54 in case of No weld penetration joint. b) For fatigue life, significant reduction in fatigue life is observed in both the cases for No weld penetration joint. Maximum of 58 % reduction in fatigue life is observed as per ASME. Whereas, as per RCC-MR, 99 % reduction in fatigue life of weld joint is observed. This can be attributed to high peak stresses and their treatment in ASME and RCC-MR codes. c) ASME is a non-nuclear code and calculates fatigue life based on stress life approach. It uses bending and membrane stress components, which shows significant variations in present two load cases, therefore there is ~30% to 58% decrease in fatigue life; for the case of no weldment penetration, the minimum life of cycles are obtained for P1+T1+Dw+D at toe region. Fatigue cycle at root are quite higher compared to toe. d) RCC MR covers Fatigue assessment using strain life approach, which is based on total stress intensity. Also strength reduction factors were provided for full penetration and no penetration welds, which plays a significant role in life assessment. Minimum life prediction as per RCC MR is 1000 cycles for Expansion case in Root region, which in full penetration case have life greater than 1e9 cycles. This drastic reduction in fatigue life cycle in RCC-MR can be attributed to fact that it is a nuclear code with stringent design validation requirements. e) The piping section, being a non-nuclear component, is designed as per ASME, therefore fatigue cycle evaluation as per ASME is valid for same. Fatigue Life as per RCC-MR is calculated only for comparison purposes. f) Further, if only stress are considered there is no issue in the weld strength for any of the provided load combination for the no weldment penetration case under consideration. REFERENCES [1] P. Dong, J. K. Hong, The Master S-N Curve Approach To Fatigue Of Piping And Vessel Welds, Welding in the World January 2004, SpringerLink, Volume 48, Issue 1, pp [2] System Design Description Document (DDD) Heat Rejection System, DDD-PBS 26.HR. [3] Load Specification for Cooling Water System, ITER_3YGYH7_v5_2. [4] ASME Boiler and Pressure vessel code Sec VIII Div 2. [5] RCC-MR RR [6] ASME Code for Process Piping B [7] Ansys Theory Reference. 5