What should an open source plasma cnc controller have ? At what price?

I think they will be ok. we modelled the thermal heat rise. So will see how we go.

rodw wrote: Tommy has done one with 2 parallel ports and a THCAD.

It can be done with one, but needs shared limits/home switches.
1 probe
1 encoder/THCAD
2 for Y axis
1 for X and Z
so the arcok has to be derived form THCAD
Not ideal, but it sure works properly.

anyone has experience with arduino hal integration?
Nick on cnc-club.ru has some work on this.
www.cnc-club.ru/forum/viewtopic.php?f=15...8&view=unread#unread

emergent.unpythonic.net/01198594294

It gives
6 analog inputs
6 PWM outputs
6 digital io

Not real time, so no reason to bother with.

tommylight wrote: Not real time, so no reason to bother with.

I just want to ask.
[strike]For arc-voltage sensing will it not be fast enough?[/strike] After some study, USB is a big no-no.


Can we use modbus for more io?

update:
can we use this?
USB is the problem or serial communication?

I'm curious what you have in mind for this as a project. Are you thinking of just spec'ing some known to work components with some working config files? Or full on control panel box, optimal layout of components, assembly instructions, etc.?

AgentWD40 wrote: I'm curious what you have in mind for this as a project. Are you thinking of just spec'ing some known to work components with some working config files? Or full on control panel box, optimal layout of components, assembly instructions, etc.?

Both.
I am pretty sure some users more versed in mechanical work would gladly pay a bit more to have a fully assembled box than risk releasing the magic smoke.
But i am sure i do not want to end up like some other vendors that are mentioned often here for putting 500$ of stuff in a box and asking 3000$ for it with no support and with no bloody shielding and proper grounding.
I would rather spend two days assembling and two more days testing on a real machine then ship knowing that it is fully functional and ready to use. I prefer the "set it and forget it" way of doing things.
As a general idea, a single box with electronics wired and tested would be the easiest and cheapest way as shipping anything from here cost an arm and a leg.
I would like to to the latter but that would require a team and would raise the price sky high.

tommylight wrote: Both.
I am pretty sure some users more versed in mechanical work would gladly pay a bit more to have a fully assembled box than risk releasing the magic smoke.
But i am sure i do not want to end up like some other vendors that are mentioned often here for putting 500$ of stuff in a box and asking 3000$ for it with no support and with no bloody shielding and proper grounding.
I would rather spend two days assembling and two more days testing on a real machine then ship knowing that it is fully functional and ready to use. I prefer the "set it and forget it" way of doing things.
As a general idea, a single box with electronics wired and tested would be the easiest and cheapest way as shipping anything from here cost an arm and a leg.
I would like to to the latter but that would require a team and would raise the price sky high.

Thats great.

I want you to endup with parport and Thcad.

Can you suggest known good pci Parport?

pinder wrote:

tommylight wrote: Not real time, so no reason to bother with.

I just want to ask.
For arc-voltage sensing will it not be fast enough?

Can we use modbus for more io?

Not really, USB has very bad latency and very often does nothing for long periods of time, having another device on a USB HUB will mess up even more that latency as every USB HUB has two shared ports.
And yes ModBus can be used for more IO but again not real time.
A much better way would be to use ESP32 as IO and use multiplex/demux to a parallel port, that would still remain in the real time domain.

No eBay APP ID and/or Cert ID defined in Kunena configurationI know that I have purchased 5 of these and all work fine.


But probably no good for you