Is there anyone interested in building machine controllers? I am designing a new one for the taig and other small machines.
I can share my experience with setting up mach3 and general knowledge of stepper motors, controllers, relays and the parallel interface.
I want to explore EMC2 (Linux CNC) with the Axis GUI, alternatives to parallel breakout boards and closed loop stepper motors.
To be frank my knowledge of electrics is pretty weak, I need some input on how to layout my controller box so it is neat, tidy and reliable. Also it would be really great if someone would be willing to check my designs and component selection for obvious issues.
Have been working on my Taig CNC conversion based around a standalone FPGA based controller from China (which seems to work with gcode postprocessed for mach 3). It’s similar to this one but also has inputs for a MPG handwheel which seems to work great. Really wanted to get away from needing a separate PC and the connectivity/pulse frequency issues.
Would be great to get involved to see what I can contribute and learn (and hopefully also motivate myself to finish my own conversion…) Have some spare cabling, connectors and 48v power supply if needed.
I opted for DSP stepper drivers, but wish I’d known more about the closed loop steppers earlier as they seem interesting to work with.
I’m willing to help out if/when I can! Not a whole lot of time, but I do enjoy machine controllers.
The Sherline CNC uses LinuxCNC - feel free to play with it an experience the pain first hand. Perhaps this project can be a good use for the Sherline, and enough reason to keep it around a while longer?
Ok. I am probably not ready for building another Taig even though they seem to be very neat. But to build a deeper understanding around the function of 3d printers absolutely!
My interest is less about G-Code running motors in a typical cartesian space, and more about just running arbitrary motors in the process automation machines I’m building. However there’s so much overlap with CNCs and 3d printers that I’ve been researching those as well. My approach has been to use ubiquitous NEMA17 and NEMA23 steppers, but I’m open to using regular DC motors along with closed loop control as well.
I’ve built a few Mechaduinos, studied their schematics, hacked on their firmware, and am now pretty familiar with what they can do and what their limitations are. The summary of my experience is that they are really freaking cool as a learning toy, and still somewhat disappointing for most real applications (although this could change as the firmware improves).
I’m at the space pretty frequently (and am a keyholder) so would gladly meet with everyone in this thread. It sounds like we are all coming at this from different angles, but that our goals overlap.
So @lukecyca corrected me on this. I thought you wanted to design a
controller board from scratch, which is a very difficult undertaking.
Sounds like you want to do good box design with existing controller boards
and relays and such. I’m totally willing to help out with this! I’m not
physically in the space much lately, but feel free to post component
selections and ask questions here and I can weigh in.
Yes! Just the box,components, layout, interfaces, and how to hook it all up to something to drive it.
@lukecyca The controller software will be just one section, and you will be doing something different, no big deal. I am waiting on some closed loop Nema 34s for testing, the manufacturer promised the moon, it would be good to see how they compare to the mechaduino.
@handy 3d printers run the same kind of hardware usually just in a little different configuration and usually with the computer build in. I really want to try out the beagle bone with the machine kit.
@rsim I will see if i can get it installed… It should make that machine more manageable.
No experience in this area, but wondering if you’ve tried the ClearPath servo motors. Mechaduino, but more polished and professional (and more expensive ;)). Ben Krasnow has featured one on a couple of his Applied Science videos for motion control.
I’ve been watching those videos and paid special attention to the servo that he’s using to move his highspeed camera. It looks awesome. I want one! I’ve researched them but haven’t ordered one.
As you know, he also used a Mechaduino to pull focus on the camera while it moved.
So @lukecyca [2] corrected me on this. I thought you wanted to design a
controller board from scratch, which is a very difficult undertaking.
Sounds like you want to do good box design with existing controller boards
and relays and such. I’m totally willing to help out with this! I’m not
physically in the space much lately, but feel free to post component
selections and ask questions here and I can weigh in.
Yes basically a walk through of existing common designs that work satisfactory and maybe a few that doesn’t work so well, and why.
Thanks for replying.
BTW There seems to be a problem with this my email address and Discourse. I tried changing it in my settings and it seemed to work, but then when I recovered my Google address I saw that Discourse was having problems accepting the new email adress In use for VHS at Nili.ca.
I do not wish to use GmaIl any more, so : Are you able to correct this issue? You once changed me to gmail on talk when another email provider locked me out…
Kind regards Handy.
Visit Topic [3] or reply to this email to respond.
IN REPLY TO
handy [4]
January 18
Ok. I am probably not ready for building another Taig even though they seem to be very neat. But to build a deeper understanding around the function of 3d printers absolutely!
Visit Topic [3] or reply to this email to respond.
I have all the components, but an old PC case isn’t working out as well as I had hoped as a controller enclosure, so I’ve been putting off the final assembly. Figuring out where to install limit switches seems somewhat non-trivial also, but that’s further down the priority list.
Would there be any benefit, or problem with running each of the drivers with a separate power supply? I have found individual 48v 8amp power supplies cheaper than a single 48v 24amp.
That should be no problem. You will want to tie their negative terminals together so that are all referenced to each other.
With some PSUs you could tie their +48v terminals together as well, running them in parallel to achieve the sum of their current capacity on one single big rail and drive one larger load. However many regular PSUs are not designed for this, and their regulation circuitry will interfere with each other. If you have three separate motor controllers, you can easily just drive them independently as you suggest.
I think we have some fairly nice 48v power supplies at the space in the bins below the electronics test gear. Possibly left over from @jon’s pinball machine.
Perhaps this project can be a good use for the Sherline, and enough reason to keep it around a while longer?
