This is my variation on the Rostock printer. It is pretty much Johann's original design with the following changes:
1) Braid fishing line instead of belts for main axis drives,
2) Threaded drive pulleys to prevent braid tangles,
3) Use of planetary gear box on extruder,
4) Custom Bowden hot-end holder.
The main reason for the braid in place of the drive belts was that I was finding it hard to source GT2 timing belts/pulleys, and braid fishing line was a cheap alternate that a friend suggested I try. I was a bit skeptical that the braid would work, but so far it has proven to be most effective.
The braid line I used was "Spectra Extreme Braid" 0.48mm, with approximately 2.5m needed for each of the 3 axes. I simply tied loops in the ends and used cable ties to tension the lines. To drive the braid I designed and printed a custom drive pulley and a v-groove pulley for the top end idler. 3 of each of these is required. I did not add any support when slicing the pulleys, as the maximum overhang (other than screw holes) is 45 degrees, and the in-fill was 0.3.
For an extruder driver I printed Airtripper's bowden extruder v3 (http://www.thingiverse.com/thing:35404). At the time I was fascinated with gearboxes (with visible gears) so I decided to print Ellindsey's Extruder motor planetary gearbox (http://www.thingiverse.com/thing:39558) to gear the extruder down. Since the gearbox is quite large I ended up mounting it through a cut out in the plywood frame.
For the hot-end end of the bowden unit I designed and printed a custom holder that marries nicely to the rostock platform. It is designed to hold a 1.75mm filament Makergear hot-end and clamps the PTFE tube using a 4mm nut. In I have included support in the design of the top half of the holder, meaning no support is required when slicing the stl.
The hot-end holder was printed in ABS, and seems to standing up to the heat generated for ABS printing. All other plastic parts were printed in PLA, with a layer thickness of 0.25mm.
Things to improve:
- The braid drive pulley was printed with 24 facets per revolution. This needs to be redone with with more facets as there is a noticeable flick on the braid as the pulley rotates.
- Also a smaller diameter might be better as is will improve resolution (more motor steps per mm). Currently my steps per mm is around 35, which is just over 2 real motor steps when the micro-stepping (16) is taken into account. I am not sure how micro-stepping really affects things...
- Add some filament guides.