I am running a 1000mm x-carve with an added rotary axis.
You can see videos of it running last week at https://www.facebook.com/a.malcolm.stanley/videos/10161249801350464/ and https://www.facebook.com/a.malcolm.stanley/videos/10161249792485464/
Basically to get going I took a stock X-carve, put it on a blue kreg table with a split top so I could mount the rotary below it, and clamped it in place. It has worked fine for the past year while I worked out the rotary axis part of the equation, which turned out to be operationally a little more complex than I had imagined.
I now think I understand the rotary workflow,and have it to the point where the x-carve is now starting to be visible as a weak link. To correct this, I am thinking of taking the next step and specifying a custom integration specifically for rotary axis implementation.
To keep things simple®, I am thinking of an integration/ combination of the X-carve X-axis and gantry, your screw drive implementation instead of belts, or Phil’s 9mm belt kit if price is an issue, and custom end plates that bolt to the pre-drilled holes on the top deck of the Kreg Table cross bars. That would give me a rigid X, and then I could specify a plate setup for the lower deck of the table cross bars to bolt the rotary to. Z would be a CNC4Newbies Z.
Since I have a working Rotary CNC I think I can CAD this up with real world measurements verified to work, so good starting point. I wondered if you had a CAD file for your screw drive that could be integrated in? Also, since you are making custom end plates already, I wondered if you would be interested in being source for the end plates we would need for this? I sort of see a multi-supplier BOM approach where you source
- the table from I guess Amazon
- the x-carve X as a parts BOM from Inventables,
- the rotary from e-bay,
- the z from CNC4Newbies
- the screw drive and maybe the end plates from you,
- optionally / alternatively 9mm belts from Phil,
- controller could be user choice, inventables controller does not support rotary but any real controller should do…
- sender would be user choice but I would reco Mach3 since I know it works,
- G-code from Deskproto
Eventually the whole thing could doubtless be kitted but not sure that would make sense logistically.
Anyway, this is a long email. Wonder what you think? let me know. /malcolm