Recent PlasmaC Update and No Motion

More
25 Aug 2020 05:38 #179317 by PurduePete
I recently updated PlasmaC. Usually dont do many updates so I was a bit behind. Been struggling to get hypersense working properly so one thing a friend and I tried was updating the plasmac and running ohmic3 component. After the update and making sure that configurator.py ran upgrade, I can't get any motion from the machine after pierce. When starting a program, it will move to position. After piercing the material it does not move from the pierce position at all. It will hover for 3-4 seconds with a read out on arc voltage and showing arc ok. But it hovers long enough to blow a hole so that src voltage cant be established to sufficiently meet the arc ok requirements. If the program is restarted, the arc wont even come close to establishing. Just get valid arc lost error. The arc ok light is on and everything appears good, so I'm baffled.

Does anyone have any suggestions on what to look at? I run mode 0 and have not had a problem before.

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

More
25 Aug 2020 05:56 - 25 Aug 2020 05:58 #179318 by Tesremos
LINUXCNC 2.9.0-pre0-2638-gadd5f0b68
Linux Buster

MESA 7i76e + THCAD(volts) + THCAD(Ohmic-Hypersense)
HyperTherm 45XP


is the OS and LCNC version loaded on the machine currently

(im the friend)
Last edit: 25 Aug 2020 05:58 by Tesremos.

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

More
25 Aug 2020 06:08 #179320 by phillc54
If you start LinuxCNC from a terminal do you get any errors?

What value is Pierce Delay?

A screenshot of your run panel and config panel may help.

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

More
25 Aug 2020 10:55 - 25 Aug 2020 10:55 #179341 by snowgoer540

I recently updated PlasmaC. Usually dont do many updates so I was a bit behind. Been struggling to get hypersense working properly so one thing a friend and I tried was updating the plasmac and running ohmic3 component. After the update and making sure that configurator.py ran upgrade, I can't get any motion from the machine after pierce. When starting a program, it will move to position. After piercing the material it does not move from the pierce position at all. It will hover for 3-4 seconds with a read out on arc voltage and showing arc ok. But it hovers long enough to blow a hole so that src voltage cant be established to sufficiently meet the arc ok requirements. If the program is restarted, the arc wont even come close to establishing. Just get valid arc lost error. The arc ok light is on and everything appears good, so I'm baffled.

Does anyone have any suggestions on what to look at? I run mode 0 and have not had a problem before.

I have several questions that would probably help diagnose what's going on:

What version of the ohmic component were you running before this?

Did you remember to update the <machine>_connections.hal to point it to ohmic3?

Did everything run right before updating?

Are you running the 25K ohm 5W resistor?

When you say you run mode 0, do you mean on the ohmic comp?

At any rate, I have seen similar behavior to what you are describing with ohmic2, before Rod updated it.

One recommendation I have is to use Mode 1, that's what I currently run and it works perfect over here on my end.

Here's my ohmic stuffs from the <machine>_connections.hal file (note that the or2 stuff is just to control the AC relay to turn my ohmic power supply on and off at the appropriate times, if you have it hard wired to be on all the time, you can ignore those lines):
# ---OHMIC signals---
#0V = 139.8 KHz
#5v = 953.9 KHz
setp hm2_5i25.0.encoder.01.scale -1
setp hm2_5i25.0.encoder.01.counter-mode 1
setp hm2_5i25.0.encoder.01.filter 1

loadrt  ohmic3                  names=ohmicsense
addf    ohmicsense              servo-thread
loadrt  or2                     count=1
addf    or2.0                   servo-thread
 
setp ohmicsense.thcad-0-volt-freq    139800
setp ohmicsense.thcad-max-volt-freq  953900
setp ohmicsense.thcad-divide         32
setp ohmicsense.thcad-fullscale      5
setp ohmicsense.volt-divider         4.9
setp ohmicsense.ohmic-threshold      23.5
setp ohmicsense.ohmic-low            21.5
setp ohmicsense.mode                 1 
setp ohmicsense.num-readings         10
net ohmic-vel ohmicsense.velocity-in <= hm2_5i25.0.encoder.01.velocity

net ohmic-true ohmicsense.ohmic-on => debounce.0.2.in
net plasmac:ohmic-enable plasmac.ohmic-enable => ohmicsense.is-probing

net plasmac:motion-type => ohmicsense.motion-type-in
net ohmic:power <= ohmicsense.power-on

net plasmac:ohmic-enable => or2.0.in0
net ohmic:power => or2.0.in1
net or-out or2.0.out => hm2_5i25.0.7i76.0.0.output-01
Last edit: 25 Aug 2020 10:55 by snowgoer540.

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

More
25 Aug 2020 12:30 #179368 by PurduePete
I was cutting some 14 ga with fine cuts. Normal pierce delay is .4s. I changed it to .2 and to 0 without any change. I can screen shot it.

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

More
25 Aug 2020 12:38 #179371 by PurduePete
I was running ohmic before switching to 3. So I skipped 2.

Yes, the connections hal was updated to ohmic3. Also tried changing it back to just ohmic without any effect.

Yes, I am using the resistor.

I should have been more specific, sorry about that. Yes I am running mode 0 for ohmic comp.

My connections looks at minimum similar to this. When using the meter my ohmic voltage reading was offset by about 4v to start with so my values are a bit higher.
loadrt ohmic names=ohmicsense
addf ohmicsense servo-thread

# --- 7i76e ENCODER 2 SETUP FOR OHMIC SENSING---

setp hm2_7i76e.0.encoder.01.scale -1
setp hm2_7i76e.0.encoder.01.counter-mode 1

# --- Configure the component ---

setp ohmicsense.thcad-0-volt-freq    140300
setp ohmicsense.thcad-max-volt-freq  956200
setp ohmicsense.thcad-divide         32
setp ohmicsense.thcad-fullscale      5
setp ohmicsense.volt-divider         4.9
setp ohmicsense.ohmic-threshold      29
setp ohmicsense.ohmic-low            24

#****** depending where you have the second thcad connected, 00 is A channel encoder 1,
#****** 01 is B channel encoder 1, 02 is for phils custom firmware because he blew something up.

net ohmic-vel ohmicsense.velocity-in <= hm2_7i76e.0.encoder.01.velocity

# --- Replace Plasmac’s Ohmic sensing signal ---

unlinkp debounce.0.2.in
net ohmic-true ohmicsense.ohmic-on => debounce.0.2.in
net plasmac:ohmic-enable    =>  ohmicsense.is-probing

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

More
25 Aug 2020 13:17 - 25 Aug 2020 13:18 #179380 by snowgoer540

I was running ohmic before switching to 3. So I skipped 2.

Yes, the connections hal was updated to ohmic3. Also tried changing it back to just ohmic without any effect.

Yes, I am using the resistor.

I should have been more specific, sorry about that. Yes I am running mode 0 for ohmic comp.

My connections looks at minimum similar to this. When using the meter my ohmic voltage reading was offset by about 4v to start with so my values are a bit higher.
loadrt ohmic names=ohmicsense
addf ohmicsense servo-thread

# --- 7i76e ENCODER 2 SETUP FOR OHMIC SENSING---

setp hm2_7i76e.0.encoder.01.scale -1
setp hm2_7i76e.0.encoder.01.counter-mode 1

# --- Configure the component ---

setp ohmicsense.thcad-0-volt-freq    140300
setp ohmicsense.thcad-max-volt-freq  956200
setp ohmicsense.thcad-divide         32
setp ohmicsense.thcad-fullscale      5
setp ohmicsense.volt-divider         4.9
setp ohmicsense.ohmic-threshold      29
setp ohmicsense.ohmic-low            24

#****** depending where you have the second thcad connected, 00 is A channel encoder 1,
#****** 01 is B channel encoder 1, 02 is for phils custom firmware because he blew something up.

net ohmic-vel ohmicsense.velocity-in <= hm2_7i76e.0.encoder.01.velocity

# --- Replace Plasmac’s Ohmic sensing signal ---

unlinkp debounce.0.2.in
net ohmic-true ohmicsense.ohmic-on => debounce.0.2.in
net plasmac:ohmic-enable    =>  ohmicsense.is-probing


Before I go further, if you turn ohmic off, and use the float switch, does it work?

I see a few issues there:

you're missing:
setp hm2_7i76e.0.encoder.01.filter 1

29 and 24 seem way too high. I know you said your voltage is offset by 4 volts, but I'd wonder why. Have you tried verifying the voltage readings? Like complete the circuit for the ohmic (not through material, like a jumper) so it provides 24v from the power supply to the THCAD and see what it reads in hal meter, and verify with an actual voltage meter?

I'd also encourage you to go back to ohmic3, but add the following lines to your hal file:
setp ohmicsense.mode                 1 
setp ohmicsense.num-readings         10
Last edit: 25 Aug 2020 13:18 by snowgoer540.

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

More
25 Aug 2020 13:30 #179385 by rodw
What power supply are you using for ohmic? If you use the recommended power supply you need to fine tune it to output 24 volts with a good multimeter.
The following user(s) said Thank You: snowgoer540

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

More
25 Aug 2020 13:33 #179386 by snowgoer540

What power supply are you using for ohmic? If you use the recommended power supply you need to fine tune it to output 24 volts with a good multimeter.

All that said, I'm not sure I understand why it's preventing the torch from firing or movement after the torch does fire.

But he's describing the exact problems I've seen when I had ohmic2 on the machine. I know CBEAR saw the same problems on his table, but he's not able to get ohmic3 to work (it wont ohmic sense at all on his end for some reason, he hasnt been around for me to help him fix it).

@ROD Why would the ohmic component affect that stuffs?

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

More
25 Aug 2020 13:48 #179389 by rodw

@ROD Why would the ohmic component affect that stuffs?


I think if a touch off does not happen, then motion does not follow. ( feature of PLasmac, not LInuxcnc)

The original ohmic.comp just simply replicated a relay so there was a voltage at which it came on and a lower voltage at which it turns off. so there is hysteresis to prevent false triggering.

The OP should use halshow to monitor the voltage in the circuit and observe if the thresholds are reached on the way up and down. if its set at 29 volts, and it never gets there, then the probe will never fire. and motion won't happen

So the newer version maintains the hysteresis but also does some fancy stuff with moving averages to increase sensitivity. I thought it could even trigger a probe event before the voltage fell below the high threshold.

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

Moderators: snowgoer540
Time to create page: 0.091 seconds
Powered by Kunena Forum