Digital simulations of a stationary and a linear weld
- PDF / 2,516,283 Bytes
- 10 Pages / 612 x 792 pts (letter) Page_size
- 84 Downloads / 206 Views
I. INTRODUCTION
ACCORDING to David and Debroy,[1] “losses of life and property damage due to catastrophic failure of structures are often traced to defective welds.” One major source of weld defects is hot cracking. Factors affecting the hot-cracking susceptibility of an alloy fall into two categories; metallurgical and mechanical. The metallurgical factors are controlled by the composition and solidification morphology of the weldment, whereas the mechanical factors are controlled by thermal stresses and strains. The latter occurs in the material as it goes through its intense thermal cycle, which causes the weld to solidify rapidly. Since these cracks are detrimental to the quality of the weldment and, ultimately, the workpiece, it is highly desirable to be able to develop a proper welding procedure so that hot cracking can be avoided. In order to achieve this goal, it is first necessary to understand the physical processes that occur during welding. The nature of arc welding does not allow direct observation during the welding process, and physical observation of the weld is limited to solidified welds. Thus, accurate computational simulations are needed to provide a better understanding of the transient phenomena that are present during the welding process. In earlier computational models, a number of restrictive assumptions were used. For example, a prescribed weld-pool profile, an undeformable weld-pool surface, a stationary heat source, and a two-dimensional (2-D) simplification of a three-dimensional (3-D) problem were commonly assumed. Extensive reviews of the earlier studies were provided by Sozou and Pickering,[2] Oreper et al.,[3] Thompson and Szekely,[4] and Tsotridis et al.[5] Therefore, only the recent works directly relating to the present study are reviewed here. In 1988 Zacharia and co-workers[6,7] developed a 3-D transient model to simulate the flow and the thermal condition in the weld pool during the moving gas-tungsten-arc (GTA) as well as the gas-metal-arc (GMA) welding process. The model incorporated a deformable surface and allowed for mass addition and surface evaporation. It was found that the Marangoni forces were dominant and that the surface deformation can retard the Marangoni effects. They also
D.K. AIDUN and G. AHMADI, Professors, are with the Department of Mechanical and Aeronautical Engineering, Center for Advanced Materials Processing, Clarkson University, Potsdam, NY 13699-5727. J.J. DOMEY, Project Manager, is with Corning Inc. Manuscript submitted October 16, 2000. METALLURGICAL AND MATERIALS TRANSACTIONS B
showed that the simulated surface deformation (surface rippling and the weld “crown”) is in agreement with observation. Therefore, the model developed by Zacharia, et al. removed many of the earlier major limitations and provides a realistic computational model for calculating weld-pool characteristics. Tsao and Wu[8] presented a transient model that simulates both GMA and GTA welding processes. They were able to account for the additional thermal energy associated with a
Data Loading...
