Proma THC 150 setup issues

Been a lurker for a few years since I started with LinuxCNC on a scratch built router. Most of the time the stepconfig was enough but with the plasma cutter on my latest build I've dipped into a bit more of the manual config stuff. I'm struggling to get the config working for the proma thc thats posted on proma's site as well as discussed here by tommylight.

I believe I have all the pins hooked up correctly and I have motion jogging fine in joint mode but when I switch to world mode I end up getting joint errors when jogging as well as the maximum speeds for certain axis being exceeded. I copied over most of the axis information from my original stepconfig that is working perfectly but I assume i'm missing something or have incorrect values in the INI for the axis configuration. I've attached the modified INI.

The machine being setup, scratch built in the garage;

Gantries with dual motors benefit from using the master branch which has the joint/axis branch merged into it. I know a lot of gobble goop but in any case that branch handles gantries correctly.

Very nice machine and I like the mobility which is a must in my shop for many things.

JT

Nice setup.
There is a post in that thread that mentions editing a line in stepper.hal
in line:
#setp stepgen.2.maxvel [AXIS_2]MAX_VELOCITY
You need to remove # , save and start Lcnc.

tommylight wrote: Nice setup.
There is a post in that thread that mentions editing a line in stepper.hal
in line:
#setp stepgen.2.maxvel [AXIS_2]MAX_VELOCITY
You need to remove # , save and start Lcnc.

Thanks for the good words guys. Tommy, I've went back and removed that #. Jogging X or Y works perfectly in joint or world mode, Z only jogs in joint. With that max velocity being uncommented, I get an immediate joint 2 following error as well as one I've never seen before which is " need to be enabled in teleop mode for teleop move". Only happens with the Z axis.

The only things i can point a finger at is the acceleration of the Z axis might be to much, so lover it to 10 and stepgen acceleration to 12.5.
I did the calculations and all looks fine, but since Z axis is controlled by thc.hal that might be the reason.
Also the velocity as far as math is concerned is all good, but again controlled by thc so might not be able to follow that quickly.
As experience taught me, i never push Z axis above 900mm/minute (15mm/sec or 0.6inches/sec), that has proven the best value for thc with proma as it is most stable during cutting (no bunny hopping), and is enough to follow over 5cm of difference at a distance of 30cm during cutting at 3meters/minute.
If you do not mind, set it as mentioned above and give it a try.
As for teleop error, i never got that so, if you still get that after changing settings, give us a shout.

HommemadeNJ (NJ is for New Jersey, right?), i have to go get some zzzzz, i am up since 8AM and it is 3:35 AM now.
Later,
Tom

Yes sir Northern / Central NJ in the states

I ran one last test and just packed it in for the night myself. I tried your suggestions to slow down the Z and still getting the same joint following error for the Z axis but I noticed something interesting.

Lets say I set the parameters so the Z axis will run at 50in/min. If I then switch to world mode and keep the slider at the bottom of the screen at 50in/min or under I don't see the joint error. However if I raise it to say 100in/min and try to move the Z, the Z appears that its only moving 50in/min but the velocity on the screen will show as if its trying to run at 100in/min and it looks like it is throwing the joint error due to the mismatch in speed.

So the Z velocity limit that we un-commented is acting properly but maybe somewhere else in the program it still thinks that the velocity from that slider is the actual speed and the f-error value falls way outside of my min/max causing the joint error.

I guess I can increase the f-error values of the Z to massive numbers as a test. Will fiddle with it in the morning.

Just checked your ini again, last night i missed that you changed the min_ferror for Z axis. You can not change that, it has to waaaay higher ( i have it set at 500). Z axis is controlled by thc.hal so it will not work with your setting.

B.T.W, i have a brother living in NJ.

tommylight wrote: Just checked your ini again, last night i missed that you changed the min_ferror for Z axis. You can not change that, it has to waaaay higher ( i have it set at 500). Z axis is controlled by thc.hal so it will not work with your setting.

B.T.W, i have a brother living in NJ.

Ah yes that did the trick, probably should have known not to touch that from the comment directly above it in the .ini file. I was a bit confused though by the behavior of the Z axis when pressed in world mode but that's why you mentioned in the other thread only give the jog button short pulses. I think I am all set now, still waiting on some water table additive but next step is to test with the THC, I've had it running before without issue with the simple touch and go sub. I'll be sure to post up a video of everything running once I get it sorted. Thank you for all the help.

Where abouts in NJ is your brother?

When you do get to testing it, follow this short procedure so you do not have to hunt around for problems that might arise:
---on the Proma THC
--set the hysteresis at 2,
--set the voltage difference at 2 (or 1 i can not remember as all 4 of mine are gone to various machines)
--set the voltage to 140
--put it in test mode ( press both buttons till you see TST) and check if the UP/DOWN, ARC OK signals are in order and corresponding to up/down arc ok from the THC
Now on LinuxCNC check the "switch travel" and adjust it accordingly, also the pierce height (GAP)set to 3mm
Press F9, it should move the torch down till it hits the material and continue moving down till it triggers the floating switch, move up to pierce height and fire the torch. If all is well, it should pierce and start moving Z axis down slowly. Stop with F9 or ESC.
Now set you jog speed to something corresponding to your test metal thickness ( for 3mm--1/8inch set it to 1500mm/m or 60IPM and 40 to 50 Amps) .
Again press F9 and when you get the ARC OK press the arrow key to move, watch if the torch is going up or down and the aprox distance from the metal. If all is OK, you should have the torch moving to high so just set the THC to a lover voltage and test again.
After 2 or 3 tries you should get nice cuts so now you can move on to testing actual cutting and finding a nice cutting height and feed speed.
Only after all this you can test and set the THC following speed, so set the metal plate at an angle aprox 3 to 4cm high on one side and flat to the table on the other. Do a test cut climbing the plate and dropping down. Set the "step size" lower or higher so you get it to follow that slope at a distance of at least 30cm (12inch) smoothly without to much of up/down torch movement.
That is it, now all you need is to set the Amps, feed and rarely voltage on the THC depending on the thickness of the material.
It will take a bit of testing to get nice clean cuts without to much dross.

He is in Clinton.