I joined the Precision Machining Research Consortium (PMRC) while completing my undergraduate curriculum. During the following three years I not only learned the operation and programming of multi-axis CNC equipment but also gained experience in a wide range of advanced topics including digital manufacturing and equipment monitoring.
One of my main tasks was the validation of experimental gcode generated from a new voxel-based computer-aided manufacturing software (CAM) called SculptPrint. Our team generated toolpaths and exported gcode from this software which was uploaded to a 4.5 axis millturn. Aluminum and Delrin parts were then machined to evaluate runtime and qualitatively judge the efficiency of machine tool kinematics dictated by the gcode. The main focus in these evaluations was tool accessibility and 5-axis orientations. The hybrid octree structure that SculptPrint uses to represent part geometries yields a “What-You-See-Is-What-You-Get” (WYSIWYG) effect when programming. The discrete nature by which volume is depicted faithfully recreates toolmarks and gouges within the software before any machining has even taken place. Of greater concern was the inverse-kinematics required to send the tool along certain trajectories where it does not impact the part or chuck.