I’ve barely even started printing, but I got an idea in my head looking at the design and thinking about 4 axis milling… The really neat thing about this design (besides the emphasis on inexpensive parts) is the gantry. I don’t know of another CNC design that directly supports the tool head in both X and Y direction. Anyway, my (probably terrible idea) is this:
Use the gantry design of the MPCNC, but instead of having the tool head on the gantry, have the cut table mounted upside down on 2 rails in direction. Use a Z axis on a fixed base underneath the gantry, with 2 Z motors. Raising only lead screw will change the angle of the tool head, allowing for diagonal cuts in rough 45 to 135 degrees (an entirely new Z design would be needed to allow this). Your cut area would only be 1/4 the size of the gantry, but it seems like it could be incredibly rigid and allow for true 4 (or even 5) axis milling. Plus cleanup is a breeze since all the dust just falls down.
I’m sure I’m missing some very good reasons why this wouldn’t work… Thoughts?
Also, 3 lead screws seems like you could move the tool head similar to a delta printer, giving 5 axes and maybe even making this design more stable than just 2 lead screws…
Ah, yeah that makes sense… If I wanted to drive the table on it I really would have to scale up the gantry. Unfortunately keeping the gantry moving the toolhead makes the 4-5 axis goal a lot more difficult (though in a lot of ways still easier than more traditional CNC designs). Maybe an upper and a lower gantry moving independently would provide enough stability to angle the toolhead and have it actually be able to cut something?
Software is definitely a problem… Though it’s also an excuse to start working on a 5 axis CNC “slicer” (is that the correct term when applied to CNC?) since there don’t seem to be much in the way of open source work in that area. I’m thinking rendering the toolpath in a voxel-based system to simulate cutting. The basic math is pretty simple, but I’ll need to do some research on cutting path optimization… Though I could probably have a relatively naive algorithm that would produce good results in sub-optimal time.
this is a slight tangent and a bit of a thread hijack but it’s related to the title!
Something occurred to me when i was putting mine together in its new home in the workshop (i have my kitchen back now, always a landmark step in MPCNC building i think)
can you assemble the machine upside down? like, not the outer frame, but hang the X/Y steppers backwards and upside down so the cross part/central hub hangs underneath. sorry if i’m explaining this badly.
I’m not really an engineer with a capital e or anything, so this is probably a dumb thought and i look forward to hearing why it’s a stupid idea but i feel like this would help with rigidity, having gravity help out, and then you’re a little closer to the work surface too.
feel free to tell me what i’m missing and laugh at me for missing it