from the onset of weld segment 4, whereas the region close to weld segment 3 undergoes cooling, as illustrated in Figure 13(a). This compensates for the subsequent distortion caused by the weld segment 4, involving a transversal weld across the header. Considering the torsional distortion, it seems important to allow for a large (negative) angle at the time the welding of the last segment starts. This is the situation depicted in Figure 13(b). The cooling after welding of this segment will induce considerable thermal contractions at the regions at the right part of this figure, corresponding to the computed development between 80 and 100 seconds of the torsional distortion angle seen in Figure 12. The results from cases E and F illustrate clearly that the welding sequence may have a pronounced influence on the distortions. Use of welding sequence 2 (case E) resulted in a significantly reduced tilt angle compared to case A. The bending of the header induced by the first weld segment is seen to be "absorbed by" the gap between the parts. However, the torsional angle is predicted to be very large in case E, and therefore an improved welding sequence was looked for by performing a series of simulations. After a few trials, the welding sequence 3 applied in case F was found to be optimal with respect to both kinds of angular distortions. The splitting of the first weld into two segments (compared to sequence 2) was essential in order to obtain a uniform gap distance prior to local fusion along these weld segments. This, in turn, prevented torsional distortion. Figure 13: (a) Fringe plot of heating/cooling rate after 51 seconds. (b) Temperatures after 69 seconds (right) from case D. Distortions are magnified by a factor of 10. In general, it is possible to compensate for the resulting weld distortions by mounting the parts in the fixture in an offset position, unless the welded parts form a closed loop. The exact magnitude of the offset positions can be determined from practical welding experiments, or preferably, by weld simulations provided that the latter are sufficient accurate. Canceling the distortions, by applying weak clamping or by applying an optimal welding sequence, can be an attractive alternative. Generally, it is considered as advantageous to apply a welding sequence, which implies that the distortions at the interface (including generation or closure of gaps) can compensate for the welding induced deformations of each of the welded parts. SUMMARY In this article a process model (WELDSIM) for welding of age-hardening aluminum alloys has been outlined. The model consists of a thermal, a microstructure and a mechanical sub-model implemented in a FEM framework. Examples have been shown on verification of the simulation model against distortion measurements on real welding experiments. Very good agreement was obtained between computer simulations and welding experiments. Results from a modeling case study on welding of a windshield frame have been presented. A given welding sequence resulted in considerable angular distortions when rigid clamping was assumed. However, by selecting an alternative welding sequence and keep the rigid clamping, the computed distortions were almost eliminated. Alternatively, the distortions were also eliminated, by applying the original welding sequence combined with a more compliant clamping. This case study illustrates how a simulation model like WELDSIM can be a valuable tool in the pre-production phase in fabrication of welded automotive parts in aluminum. ACKNOWLEDGEMENTS The authors wish to acknowledge the financial support from Hydro Automotive Structures and The Research Council of Norway. Moreover, Mr. Peter B. Jacobsen and Mr Flemming Damborg at Hydro Automotive Structures Tønder, Denmark is acknowledged for their support regarding the automotive case study presented in the paper. 1. REFERENCES Myhr, O.R.; Grong, Ø.; Klokkehaug, S.; Fjær, H.G.; Kluken, A.O. 1993. A Process Model for Welding of Al-Mg-Si Extrusions Part I: Precipitate Stability, Science and Technology of Welding and Joining, Vol 2, (6), 245-253. 2. Myhr, O.R.; Klokkehaug, S.; Fjær, H.G.; Grong, Ø.; Kluken, A.O. 1998. Modelling of Microstructure Evolution, Residual Stresses and Distortions in 6082-T6 Aluminum Weldments, Welding Journal, Vol 77, (6), 286-292. 3. Myhr, O.R.; Klokkehaug, S.; Fjær, H.G.; Grong, Ø.; Kluken, A.O. 1999. Modelling of microstructure evolution and residual stresses in processing and welding of 6082 and 7108 aluminium alloys, Proceedings of the 5th International Conference on Trends in Welding Research, ASM International, 233–238. 4. Myhr, O.R.; Fjær, H.G.; Klokkehaug, S.; Holm, E.J.; Grong, Ø.; Kluken, A.O. 1999. WELDSIM - An advanced simulation model for aluminium welding. The Ninth International Conference on Computer Technology in Welding, Detroit 1999, 52-63. 5. Myhr, O.R.; Grong, Ø.; Andersen, S.J.; 2001. Modelling of the Age Hardening Behaviour of Al-Mg-Si Alloys, Acta Materialia, Vol 49, 65-75. A COMPUTATIONAL MODEL FOR HEAT AFFECTED ZONE OF REACTOR PRESSURE VESSEL STEEL J.S. Kim*, S.G. Lee, J.S. Park and T.E. Jin ASTRACT A computational model for the heat affected zone(HAZ) of reactor pressure vessel(RPV) steel is developed to predict microstructures and material properties. The microstructures of HAZ are predicted by a combination of the temperature analysis considering multi-pass welding and PWHT, and the thermodynamics-kinetics models for prior-austenite grain growth, austenite decomposition and carbide coarsening. Also, the hardness, yielding strength and tensile strength are estimated by the prediction results for microstructures and the empirical relations. Finally, these prediction results are compared with experimental ones and show the reasonable agreement. KEYWORDS Reactor Pressure Vessel Steel, Circumferential Narrow Gap Weld, Heat Affected Zone, Temperature Analysis, Microstructure, Mechanical Strength INTRODUCTION The metallurgical microstructures of HAZ are changed due to repeated thermal cycle during multi-pass welding and carbides coarsening during PWHT. Especially, in case of SA508 steel, which has been widely used for pressure vessel because of good mechanical properties, the mechanical properties of HAZ may be degraded due to the microstructure change. Therefore, the some studies(Ref. 1~4) has been performed to more detailedly estimate the microstructures and material properties for HAZ of SA508 steel. However, these studies used the experimental approaches, so it has the economic problems to apply the methodologies used in these studies to various welding processes and geometric shapes. Accordingly, the purpose of this study is to develop a computational model predicting the microstructures and material properties for the HAZ of SA508 Gr.3 Cl.1, which has been often used for RPV. The microstructures of HAZ are predicted by a combination of the temperature analysis considering multi-pass welding and PWHT, and the thermodynamics-kinetics models for prior-austenite grain growth, austenite decomposition and carbide coarsening. Also, the aswelded hardness, yielding strength and tensile strength are estimated by the prediction results for microstructures and the empirical relations. Welding Integrity in Nuclear Structures Laboratory (www.wins.re.kr), Korea Power Engineering Company COMPUTATIONAL MODEL Fig. 1 shows a computational model predicting the microstructures and mechanical properties for the HAZ of SA508 Gr.3 Cl.1. As shown in Fig. 1, firstly, thermal analysis is performed to determine the peak temperature and cooling rate in HAZ considering real welding processes such as multi-pass welding and PWHT. Secondly, the microstructures of HAZ are estimated considering various metallurgical factors such as prior-austenite grain growth and austenite decomposition during welding. In additions, the carbide coarsening during PWHT is also estimated. Finally, the changes of mechanical properties due to microstructure changes are predicted. Fig. 1. Computational model to predict microstructures and material properties. Analysis Model TEMPERATURE ANALSYS Fig. 2 shows the analysis model of RPV circumferential narrow gap weld. Base material is SA 508 Gr.3 C1.1 and filler material is L-TEC 44. Submerged arc welding(SAW) method is used in the narrow gap weld. Welding is performed with 81 passes and 39 passes for inner and outer part respectively. Table 1 summarizes the chemical compositions of SA 508 Gr.3 C1.1 for RPV. And Table 2 summarizes the welding parameters of this model. Finite Element Model The finite element for an analysis model is shown in Fig. 3. The numbers of elements and nodes are 830 and 940 respectively. The element property is a 4-node axis-symmetric element (Ref. 5). By using the lumped model(Ref. 6), the finite element model is simplified as the one with 8 and 5 weld layers for inner and outer weld part respectively. Model change technique(Ref. 5) is adopted to simulate multi-pass welding process. |