Whilst designing and producing tooling for large scale composite parts, I came to appreciate the cost, drawbacks and inefficiencies of the existing methods available. A lot of the tooling produced was for prototype parts and the cost of using tooling board could not always be justified, so 25 or 28mm UDF was often used. As I had access to a spare 6-axis robot arm, I started developing a pellet extruder to mount on it. The robot only had basic digital i/o capability, so to interface it with a computer, I designed and built a modular serial to parallel interface, connected via CAN. It eventually ended up with 96 input and 24 output bits, allowing 4-axis positional control of the robot (XYZ plus head rotation, with feedrate control). I later also reverse engineered the Fanuc IO protocol to remove the need for all of the IO modules, needing only the RS-485 duplex interface of the Fanuc controller.
I pivoted the design philosophy away from the pellet extruder, to using UV curing resin as the medium. At the time, a couple of companies were also developing pellet extruder systems, but none were developing UV based systems. The UV resin was better suited to prepreg tooling, having much better thermal properties, cost, and machinability. It also had the benefit of being able to be foamed, allowing for variable density tooling – further reducing cost. At the time UV resin was not widely used in 3D printing ( consumer LCD printers didn’t exist, only SLA and DLP ), so work was also done to develop suitable resin formulations.
A small FDM printer was also modified for use in the lab for testing of resin formulations. The resin viscosity is managed during printing with a heated hose for delivery to the print head, and peltier based cooling at the head to reduce slumping whilst curing.
