Another quick update to show some progress made this evening. I’ve plugged my inverse kinematics code into a quick and dirty motor controlling serial comms link to the Arduino and can now move the head of the delta robot to arbitrary XYZ coordinates. Here’s a video showing the plotting of slightly misshapen square (need some micro-switches to finished the calibration routine)
I should be able to get some more speed out of it by improving the Arduino comms link (it’s sending individual steps to the motors).
I’ll get around to doing a more descriptive post once I’ve got G-Code interpretation working (all the python code is done, just have to plug the bits together).
Here’s a quick video showing the first movement of the robot:
The video is only short because one of the grips for the toothed belts worked loose and it ended up crashing, fortunately there was no damage.
The plan is to use the delta robot as a 3D printer (like makerbot,reprap, ect). To keep the inertia of the head low I’ve decided on using a Bowden extruder. The Bowden extruder works on almost the same principle as a bikes brake cable, the plastic filament to be melted is driven through a tube to the hot end where it is melted. This reduces the weight of the printing head as it moves the filament driving stepper motor away from the heated head.
The following shows a design I’ve been working on for the cold-end, which drives the plastic filament down the Bowden tube to the hot-end. The design is based around a Nema17 stepper motor, it also uses 626 bearings (because I have a bunch spare), a handful of 6mm bolts, and two springs (for tensioning purposes).
Not shown in the above image are the springs used to tension the idler bearing carriage against the driven spindle.
The following shows a cut-away view exposing the driven spindle and idler bearing.