CNC-USB (PlanetCNC)

[mocha]

I know Hank has the 9 axis mark 2 version of this board from Planet CNC, anyone else using it / tried it / thinking of trying it / got a view on the advantages or otherwise with this type of approach?

[IanS1]

Apologies for reviving an old thread but did you get the 9 axis controller?

Anyone got any experience with this controller and software?

[Kn8]

I have just placed an order for a MKII 9 axis board.
I let you know how I get on when it arrives.

[HankMcSpank]

Well after about 18 months of usage, I still think it's a great bit of software...very clean GUI, nicely thought our features & a dynamic responsive creator at the helm. My only criticism is that the documentation doesn't keep up with the new features, but a quick question on the forum always has a satisfactory outcome.

[IanS1]

I have a interest in this system too, mainly because of the SDK (Software Developers Kit) it's just a pity the SDK hasn't been developed for the MK2 controllers as they could have made a very good base for developing other types of multi axis machines.

[IanS1]

Apologies for reviving this old thread again but I have just jumped in and bought the Planet CNC MK3 9 Axis board. The SDK is now fully supported on the MK3 boards so it'll be great to have a play around with this controller/software. I'll be using it for a small Wire Bending machine project I have planned.

[IanS1]

The controller arrived this afternoon, it's already connected up to 3 AM882 drives and working like a dream. I can see I'm going to have some fun with this.:yahoo:

[Boyan Silyavski]

Please post some video when you can. I have always been thinking about that controller and wondering why is not the most popular considering its price and software price.

[IanS1]

I've only got it set-up on a test bed at the moment so the motors aren't actually driving anything mechanical. So far it seems to be a good controller using CNC-USB software.
.
I've also written a simple Visual Basic program for controlling 3 Axis which works very well too. The SDK isn't very well documented so it's a case of playing and seeing what results I get type of software development!!!, all good fun though and the best way to learn really. Once I've got the VB app a little more polished I will post some details up and hopefully a video.
.
The MK3 can be connected by USB or Ethernet, I've not tried the Ethernet yet but I will do over the coming days.
.

[Boyan Silyavski]

Does that do squaring of slaved axis?

[IanS1]

Unfortunately I have no idea. I will not be using this controller for a router/milling machine.

I'll try to find out for you.

[IanS1]

I can't find any information in the user manual that might suggest this feature is available. I have sent them an email to find out for you.

[Dwayne Dibbley]

Heres my Mk2 running a homing sequence, but I believe that the software doesn't do the zeroing of the slave axis.

quote from planet cnc forums:

Currently this is not planned to be implemented.

Independant homing is dangerous.
If one switch fails during homing then machine might be destroyed. This is not an issue on small low power machines but if you use stronger motors something will break.


https://www.youtube.com/watch?v=2Fp0OKVMT_4

[IanS1]

Here's the reply I got from them. I've have no idea if this solution works well or not, or how difficult it would be to implement.

http://forum.planet-cnc.com/viewtopic.php?f=18&t=1631&p=8094&hilit=auto+squaring#p8094

[Neale]

It's an interesting approach. It is similar to the one suggested in a similar discussion recently about the CSMIO/IP-M, which now has a slave output for dual-motor setups but no squaring capability. In that case, the suggestion was to use something like an Arduino processor alongside the motion controller, driving the stepper drivers by combining the two sets of outputs (from motion controller and Arduino). The Arduino would pick up limit switch signals and be able to execute the homing/squaring sequence, but apart from homing, would do nothing. The "Planet" solution sits in line between motion controller and stepper drivers so in that sense is a bit more intrusive, and because it uses discrete logic gates, circuit complexity limits it to managing step pulses only and the user has to set direction. More margin for error?

Personally, I would go for the Arduino just because it's much easier to set up physically and probably no more expensive (using a pre-programmed bare ATmega chip from an Arduino) and I'm thinking of exploring that when I get my new machine running using the CSMIO/IP-M but there still seems to be a bit of a debate about how much the squaring operation needs to be done, and whether you can't do it manually almost as easily.

[Boyan Silyavski]

Neale,
your solution seems much better. I was a bit confused about that particular solution. Too much electronics and chance to mess.


but there still seems to be a bit of a debate about how much the squaring operation needs to be done, and whether you can't do it manually almost as easily.

I thought so before, but not now. Every reset with steppers results in a positional error. Yeah, for a wood working machine not a big deal, you can adjust from time to time by hand against hard stop. Not so for a more serious machine where more precision is required.

[Neale]

Thanks for the comments. My existing machine is pretty poor anyway, but only has a single X motor so I have no personal experience of the positioning issues and it's interesting to hear what problems other people see. Nearest I get to it is my 3D printer which uses two motors to drive the Z leadscrews; those motors are just wired in parallel from a single driver and within the tolerances of a 3D printer, I have never had any problems with drift of home position between them. However, it's all a bit different with two drivers/two motors, of course, as the drivers themselves can theoretically get out of sync with each other at the microstep level.

Anyway, I quite like the idea of grafting in an Arduino-based solution - I've done a bit of work with them both as the 3D printer controller and programming the microcontroller itself to use on a bare board. I'm sure that that could make quite a neat solution.