Joints?? Is there a limiting number and configuration

More
06 May 2022 13:20 #242135 by robertspark
This is just am "I was wondering question"?

I was looking at the Voron 2.4, and noticed that it uses 7x stepper motors.

4 for the Z axis,
1 for the X, 1 for the Y and 1 for the extruder.

I was wondering it if was possible to control a Voron with linuxcnc ..... this would require the use of 4 joints for the Z axis.

The benefit of this arrangement is that it is possible to dynamically level the z axis across the bed

I am aware that you can slave axis in linuxcnc, but I thought (or have only ever seen) two joints on 1 axis.

I did wonder about making a Voron slightly wider and running 2x independent extruders on their own X (or Y axis) {known as an IDEX}
pipe dream at the moment.... but I was just wondering:
www.lpfrg.com/guides/benefits-of-independent-dual-extruders/

Please Log in or Create an account to join the conversation.

More
06 May 2022 13:36 #242138 by BigJohnT
You can have as many joints you want per axis up to 9 and best of all my configuration tools will make just such a machine configuration.

JT
The following user(s) said Thank You: robertspark

Please Log in or Create an account to join the conversation.

More
06 May 2022 15:19 #242149 by andypugh
You can have 9 axes that are controlled by G-code and kinematics.

Each Axis can control a number of joints.

You can define "extrajoints" that take part in homing but are otherwise not controlled by G-code.

I don't know if you can control these extrajoints with kinematics.

But, you can definitely control the 7-joint machine described. The bit I am not sure about is whether joints > 8 can be controlled by kinematics.
The following user(s) said Thank You: BigJohnT, robertspark

Please Log in or Create an account to join the conversation.

More
06 May 2022 15:31 #242150 by andypugh

The bit I am not sure about is whether joints > 8 can be controlled by kinematics.

github.com/LinuxCNC/linuxcnc/blob/master.../motion/motion.c#L57
Suggests not.

Without recompiling you can only have 16 joints:
github.com/LinuxCNC/linuxcnc/blob/master...tion/emcmotcfg.h#L25
The following user(s) said Thank You: BigJohnT, robertspark

Please Log in or Create an account to join the conversation.

More
06 May 2022 16:28 #242153 by robertspark
Thanks very much gents,

Wow, I am impressed as that was a very lot larger than I'd thought and shows the flexibility / expandability of the default basic codebase.

Please Log in or Create an account to join the conversation.

  • tommylight
  • tommylight's Avatar
  • Away
  • Moderator
  • Moderator
More
06 May 2022 19:12 #242163 by tommylight
Thinking of it, it is perfect use case for LinuxCNC, 4 separate joints with 4 switches for Z axis, adjust the offsets for each of them, and never ever level the bed after that! :)
And homing takes 3 seconds, not a full minute or more with BLTouch or other sensors that do mesh leveling in several passes.
Nice idea Robert, thank you.

Please Log in or Create an account to join the conversation.

More
07 May 2022 13:59 #242234 by smc.collins
I don't really see the benefit of having 4 Z axis motors and screws. However I do see the benefit of building a very rigid table structure using dual bearing blocks on each corner space below the table on linear rails with a centrally mounted single ball screw for the Z axis, This sound like a nightmare to hold tolerance with. Better to build the required rigidity into the structure than try to compensate for a poor design. Also having a single piece table made from a casting like Iron would really be ideal. Aluminum could work to , or steel, or any sufficiently rigid thermoplastic with a dense material filled into it and CNC machined etc. However, if the goal is to produce superior quality 3d prints, dampening of the machine needs to be taken more seriously, particularly if the XY configuration is attempting to run at high speeds. 

I could sketch up some idea in cad if anyone is interested

Please Log in or Create an account to join the conversation.

More
07 May 2022 18:23 #242244 by robertspark
it's not for poor design... remember the bed is heated and can warp

Please Log in or Create an account to join the conversation.

More
07 May 2022 19:20 #242245 by smc.collins
My experience with printing with abs and similar plastics is that they really need a heated temp controlled enclosure, and if the entire printer is contained in the enclosure and all the mechanicals are steel, you're not going to have the issues you will with a mixed metal system like steel and aluminum

Please Log in or Create an account to join the conversation.

More
09 May 2022 23:35 #242426 by andypugh

Also having a single piece table made from a casting like Iron would really be ideal. Aluminum could work to , or steel, 

Interestingly, Prusa disagree with you :-)

The Prusa XL is designed with a bed made of several smaller beds, with a small expansion gap between them. 

The leadscrew-at-each corner approach is something that I have sen on a very fancy hydraulic press at work. (Google "Laycock 60 ton press") It's a reasonable way to keep things square, though it is kinemeatically redundant, so the screws are almost guaranteed to be trying to twist the bed, even if only slightly. 

I have always fancied making a lathe with two ballscrews to keep the carriage square (and with adjustable squareness). That would probably require some arc-shaped gibs and such, though. 
 

Please Log in or Create an account to join the conversation.

Time to create page: 0.072 seconds
Powered by Kunena Forum