Prediction of weld pool and reinforcement dimensions of GMA welds using a finite-element model

  • PDF / 1,629,941 Bytes
  • 11 Pages / 576 x 792 pts Page_size
  • 4 Downloads / 208 Views

DOWNLOAD

REPORT


I.

INTRODUCTION

T H E gas metal arc (GMA) welding process is rapidly becoming popular for semiautomatic and automatic welding applications where commercially important metals, such as carbon steels, stainless steels, aluminum, and copper alloys must be joined. By selecting the correct electrode type and size, shielding gas, and weld operating parameters, high quality welds can be made in all positions with this process. However, experimental determination of the correct welding procedure for each new application can be very time consuming and costly. In automated GMA welding applications, empirical relations that describe the interaction of process variables and their influence on average weld dimensions, such as weld pool width and depth, and reinforcement bead height and width, are required for the development of process control algorithms. 11,21Experimental studies to determine such relationships require substantial effort, and the results are usually limited to the range of parameters studied. I3,4,51 Changes of material, plate thickness, or other parameters would require repetition of the experiments to derive new equations. Clearly, a flexible mathematical model of the process, which is capable of accurate prediction of the mean fused metal dimensions in GMA welding, would be valuable for the rapid development of welding procedures and empirical equations for control algorithms in automated welding applications. In GMA welding, where metal is deposited into the E. PARDO, formerly Postdoctoral Fellow, Depamnent of Mechanical Engineering, University of Waterloo, Waterloo, ON, Canada N2L 3G 1, is now Adjoint Professor of Mechanics with INTEMA, Facultad de Ingenieria, Universidad de Mar del Plata, Mar del Plata, 7600 Argentina. D.C. WECKMAN, Assistant Professor, is with the Department of Mechanical Engineering, University of Waterloo, Waterloo, ON, Canada N2L 3G1. Manuscript submitted November 10, 1988. METALLURGICAL TRANSACTIONS B

weld pool from the electrode, the type of metal transfer mechanism plays an important role in determining the final weld profile. However, the metal far from the arc heat source can be regarded as a solid with constant properties, and analytical solutions of the linear heat transport equation can be constructed which accurately describe temperatures in the plate. Rosenthal [6] first published some closed form solutions for the temperatures within welded solids subjected to point and line heat sources, and Rykalin et al.lTl have devoted two books to this subject. Eagar and Tsai ]8j have extended these solutions to include Gaussian-distributed heat sources. From Rosenthal's equations, Adams r9J and others t1~ have derived simple approximated expressions for peak temperatures, size of the heat-affected zone (HAZ), cooling rates, and other important welding considerations. These analytical approaches have not been sufficiently accurate for the prediction of weld pool geometry in GMA welding and cannot be used to predict weld reinforcement shape. With numerical techniques, many of the si