Hmm valid point . I just check the spec of my motors. According to the vendor those motor could reach 1350 RPM with 1/2 step loosing 85% of the tork (from 0.9Nm to 0.13Nm) with 24V supplied. According to this calculation on Cartesian Design I will get around 110mm/s per axis. With Delta with proper arms keeping the angle within the peek velocity multiplier I should be able to get around 200mm/s or more. Could someone with more knowledge with Linear Delta design to confirm.
With parallel connection I could reach around 2700RPM with those motors but they will consume 60W on 24V each for 0.2Nm of tork and 1/2 step. It may work but it will require those Stepper drives for CNC machines that are 50$ each and a lot of manual wiring for Smoothie Board.
Deltas have LOWER top speed limits than Cartesian printers when you consider the entire build area. When the arms are approaching horizontal at the edge of the build plate, the linear carriage must move much faster than the end-effector. Ratios of 2-3x are common at the edge of the build plate. This is why very few people ever build deltas with screws. When they do, it’s either EXTREMELY slow, or uses very long lead screws like 25mm/rev.
Thanks Ryan, so the only way forward is Cartesian design.
With 3 stages, you could build a very nice Cartesian XY bridge gantry (like a Makerbot Replicator 1/2/2x) or a Cartesian “portal” design (like a CNC router). Then do something else for the Z stage. These could be strong enough for milling.
I have portal design CNC and it don’t accept aggressive acceleration. The weight of the portal is really huge plus the height of the portal generate huge momentum. So I am thinking for Gantry like you sad 1/2/2 like Makerbot or Like Ultimaker and if I go for Makerbot Y will need to carry the weight of X axis, I need to check if there will be significant difference between the weight of both designs.
Yeah, it is advantageous to get the Z stage separate from X and Y. Portal is really best with extremely large workpieces, the XY bridge gantry approach is better when the build plate is not too large.
Any way you put that together, those are overkill for a machine that will only be a printer. I say, plan a light milling spindle too.
Depending on your budget, it may be awesome to use some 120VAC servos instead. Much better performance over a stepper.
My wife will kill me if I also bay servos but anyway the biggest problem is that I don’t see firmware distribution that would benefit from close loop systems. Even “pro” printers use stepper motors…
According to the milling, you can’t construct machine that will effectively do both. It will be compromise for printing and compromise for milling. From other hand this 3 pieces are superior for light 5/6 axis mill but the problem there is with CAM software. In the world of CNC above 2.5D almost nothing is free/open source.