This study's objective is to develop a theoretical model and efficient control algorithm for microprocessor control of the arc length in automated gas metal arc welding systems. The system performance is to experimentally evaluated. Control of the arc length is important since it can greatly affect the quality of the well produced. The Department of Commerce indicates that half of the gross national product is associated in one way or another with welding and joining processes. Gas metal arc welding is among the fastest growing arc welding processes. Development of this methodology proposed should have a substantial impact on the advancement of automatice gas metal arc welding.