My weekend plans: Machining. @nvk @mattodell @k3tan @stephanlivera
@nvk @mattodell @k3tan @stephanlivera aaaand that's Saturday... CAM + Machining the front half. Done.
@esixtyone @nvk @mattodell @k3tan @stephanlivera !!
What CNC mill is that?
@pete @nvk @mattodell @k3tan @stephanlivera that would be the one I built from scratch. My "fuck this 2020 mainstream media + Covid crap" project. Amazing what you can do if you shut the noise off. It's 95% complete. Still a little final finishing to do, as with all good projects.
@esixtyone @nvk @mattodell @k3tan @stephanlivera Nice! What approach did you use? Like, CNC-only ballscrews or acme threads?
How accurate is it?
@esixtyone Damn, that looks really nice. And with proper cable trays and coolant!!
How did you mount those linear rails exactly? And what are they mounted on? You can get anywhere form decent to excellent accuracy depending on how you mount linear rails.
2.2kw? I guess that's not being run from a wimpy North American 15A plug... :)
@pete lol. We have real electricity in Australia. The frame is constructed from 6mm thick rectangular steel section. The gantry is 150x75mm channel, which was precision ground on the front face - my only outsourced job on this build. Its heavy stuff.
@esixtyone Nice! That's definitely an easy way to do it. :) One other alternative is to mount the linear rails on a thin layer of epoxy (there's special high-dimensional-stability grades specifically for this purpose). The epoxy then takes up any imperfections in the base, allowing the linear rails to sit true and flat. But if you can get outsourced grinding, that's a much easier option.
@pete the biggest issue is not having a flatness reference. I've seem videos showing the rails are often a little off or bent. I was going to make a grinding jig.... But no reference.
@esixtyone @pete Interesting choice to have the spindle move in two (or three?) dimensions instead of moving the table. From what I hear that might not be optimal for rigidity but probably good enough for home use cases? I've mostly seen such designs in CNC routers so far that don't experience high cutting forces. Did you ever test rigidity with larger end mills or deeper cuts?
@sebx2a the ali extrusion on z has a small flex. The rest is steel. And your looking at 25mm rails and 20mm screws. Its more space efficient and easier for flood cooling. The datrons are gantry. I haven't pushed my luck. That spindle is not very capable... And it's on loan from good friend.
@sebx2a oh but... 10mm end mill in ali at typical (lighter) hsm s&f's is workable.
@sebx2a @esixtyone IIUC the big problems with a moving gantry design is: 1) The lack of rigidity stacks up. So movement in the outermost axis allows the next axis to move, and so on. 2) If the gantry is moving you have to trade off rigidity/weight and speed/power requirements.
@pete @sebx2a the stackup between tool and work is the same. The compromise is in how heavy the spindle + associated reduction / drive can be. And machine footprint; Gantry is very space efficient. Enclosing a more traditional design for flood cooling is a pain too. This one has been designed from the start to be wet. Every design has it's +/-.
@esixtyone @sebx2a I was thinking of angular rigidity actually, which is worse in a typical moving gantry. But thinking about it more, it's quite hard to do an apples-to-apples comparison, because there are so many different possible design goals... so I'm not going to claim I'm right. :P
@pete made with basic tools, without a precision flat surface reference.... And I'm actually pretty amazed. I don't have a specific number but it is more precise than I need. Accuracy is as good as I can measure. Approach... Gantry. Steel. Designed for flood cooling, and rigid enough for working steel.
Currently using a little 2.2kw spindle. Upgrade incoming.