3 ms·

Is there a slicer that can create g code to print in this angular fashion? (So diy'ers could make one)? I wonder what could be used for the belt? This struct

Is there a slicer that can create g code to print in this angular fashion? (So diy'ers could make one)? I wonder what could be used for the belt? This structure actually looks easier to build and seems more sturdy.

IgorPartola 6y ago
It’s not a slicer problem but a firmware one. The slicer tells the printer something like “G1 X30 Y10 E5” and it’s up to the printer firmware to figure out how to move the X axis 30mm, the Y axis 10mm while extruding 5mm of filament. Given the fact that the X kinematic here are identical to a bed slinger, just with no end stops, Y and Z use kinematic similar to a CoreXY setup, this shouldn’t take much at all. I guess this would be CoreYZ? Also, this is Creality which uses Marlin for their firmware. Chances are we will see this in Marlin in no time. As far as ease of build, I suspect that’s not quite the case. The main thing I would worry about is the belt flexing under the print head. If you think about it, the nozzle only moves across the bed in a straight line, left to right. To print an object with a Y dimension you need to move the belt forward and back and depending on how you set up the line on which things are printed if there is flex there you won’t get bed adhesion. You could use the treadmill belt material coated with something that can be printed on. Or whatever belt sander belts are made of before the abrasive is glued on.
- cellular 6y ago
  But the belt only moves forward, never backwards, in the video. So I thought the slicer would need to change. A one-way belt also wouldn't have as much problems with adhesion, or backlash that you mentioned with a two-way belt. Also, the nozzle moves in the Z direction (not just x). Actually the Z direction is really a little z and a little y.
- xsmasher 6y ago
  That doesn't sound right; the slicer outputs paths, and the firmware isn't going translate flat paths into diagonal paths. The head can not travel in a plane parallel to the base plate, so it can't use paths calculated for a regular flat primer.

IgorPartola 6y ago

It’s not a slicer problem but a firmware one. The slicer tells the printer something like “G1 X30 Y10 E5” and it’s up to the printer firmware to figure out how to move the X axis 30mm, the Y axis 10mm while extruding 5mm of filament. Given the fact that the X kinematic here are identical to a bed slinger, just with no end stops, Y and Z use kinematic similar to a CoreXY setup, this shouldn’t take much at all. I guess this would be CoreYZ? Also, this is Creality which uses Marlin for their firmware. Chances are we will see this in Marlin in no time. As far as ease of build, I suspect that’s not quite the case. The main thing I would worry about is the belt flexing under the print head. If you think about it, the nozzle only moves across the bed in a straight line, left to right. To print an object with a Y dimension you need to move the belt forward and back and depending on how you set up the line on which things are printed if there is flex there you won’t get bed adhesion. You could use the treadmill belt material coated with something that can be printed on. Or whatever belt sander belts are made of before the abrasive is glued on.

cellular 6y ago
But the belt only moves forward, never backwards, in the video. So I thought the slicer would need to change. A one-way belt also wouldn't have as much problems with adhesion, or backlash that you mentioned with a two-way belt. Also, the nozzle moves in the Z direction (not just x). Actually the Z direction is really a little z and a little y.
xsmasher 6y ago
That doesn't sound right; the slicer outputs paths, and the firmware isn't going translate flat paths into diagonal paths. The head can not travel in a plane parallel to the base plate, so it can't use paths calculated for a regular flat primer.