RS485 For Hypertherms

This be my last OT comment on this thread, I think we have kinda destroyed it...

rodw wrote: Yes as you know I have one in progress

What would be nice while you are playing around with state tags are realtime equivalents of halui.program.is-idle, halui.program.is-paused and halui.program.is-running. This would enable the removal of a few kludges in the plasmac component related to the timing of these userspace pins.

phillc54 wrote: This be my last OT comment on this thread, I think we have kinda destroyed it...

rodw wrote: Yes as you know I have one in progress

What would be nice while you are playing around with state tags are realtime equivalents of halui.program.is-idle, halui.program.is-paused and halui.program.is-running. This would enable the removal of a few kludges in the plasmac component related to the timing of these userspace pins.

I think that would be a separate project as I'm dealing with the interpreter here. I think halui is managed from task via messages. I have not been playing with that stuff yet.

thefabricator03 wrote: I tried using a USB converter but could not get it to work. Probably because my wiring was wrong. I might try out the USB to RS485 converter again on the weekend if I get time.

I am using a PCI card but it is suppose to only be a RS232 card but it is somehow working with my Hypertherm 125 and PlasmaC.

If you do end up using the reference document I made, let me know if you find it useful and if anything needs to be changed.

Stefan,

Did you ever find time to mess around with any of this? I am 98% complete with updating the user guide, and I wanted to touch on the RS485 diagram/topics a bit.

Did you ever figure out how you use a RS232 card without a RS232-RS485 converter? Or do you think that card has the ability to be set to RS485 natively.

At any rate, I think we determined that your setup had something funky going on with the wiring, I just cant recall what it was. Any info you can shed back on this would be great!

Thanks,
Greg

For the record, here's where I ended up with the edits:

linuxcnc.org/docs/devel/html/plasma/plas...ml#rs485_connections

I added reference to the internal wire color, so that users could track that aspect of it down if they had issues wiring it as suggested. That solves the ambiguity that seems to exist between the way mine was wired and the way yours was wired. I did consult several videos and pictures of the serial kit, and my hypertherm wire order (internal to the machine) matches everything I saw.

At any rate, since the wire colors are standard, I figured it was the easiest way to move forward.

Hello,

In the past was a discussion about real time communication with Hypertherm 45xp for amperage control. At the corner, when the speed decrease to decrease the amperage for a good quality. Do you think that is possible.

Alexandru.

Here's Hypertherm's official stance on the matter:



Which I take to read as "it's possible, but we wont support it, try at your own risk, YMMV". Or maybe "We dont recommend this because it makes nice cuts that way and we cant sell you the good stuff then."

Either way, I do know one of the CNC controller brands that uses linuxcnc has this feature. So, it must be possible/they think it works well enough, but I 've not experienced it.

I do admit, I do find it intriguing and I wouldn't mind experimenting with it to see if it's worthwhile, worst case would be it doesn't work.

Me too! I am on the list for testing if somebody make the communication. In the last discussion the problem was that the delay is big on rs485 com, but if the delay is in milliseconds, i don't think is so important for cutting

alexandrub wrote: Me too! I am on the list for testing if somebody make the communication. In the last discussion the problem was that the delay is big on rs485 com, but if the delay is in milliseconds, i don't think is so important for cutting

I think it will be important. My table gets to 10.24 m/min in 30 ms. at a more normal 2-3 m/min its only slowing for 10 ms

The Telwin machines provide 0-10 v control of current
The Thermal dynamics have a potentiometer that controls current. It could be used for such control if replaced by a digital pot.
Grotius has done this.
The big plasma cutters provide control by the ethernet bus hence ethercat prevails.
I am sure Hypertherm have restricted the baud rate to preclude this being done on low end machines.

In a perfect world you would swap from maintaining constant voltage by altering height to maintaining constant voltage while varying current while decelerating and accelerating.

snowgoer540 wrote: I do know one of the CNC controller brands that uses linuxcnc has this feature. So, it must be possible/they think it works well enough

I would love to see their specs on this as to whether they have guaranteed response times or is it just typical sales speak which is normally worth less than what you paid for it.