QtPlasmaC and pncconf with "fake" ohmic probe and arc ok

More
11 Feb 2024 22:49 - 11 Feb 2024 22:51 #293111 by shasse
I'm working to create a fiber laser control system with LinuxCNC, and have gotten to the point where I'm ready to try using QtPlasmaC as a fiber laser control system. I've got the base machine working, including a capacitive sensor that I've described in another thread. I'm using a Mesa 7i92 card and pncconf to configure the base machine. It seems like the default method of choosing ohmic probe and arc OK inputs is by picking I/O pins off the Mesa card. However, in a fiber laser config I think I can just always consider arc OK to be true, and I can simulate an ohmic probe in software using the comp hal component like so in custom.hal:
loardrt comp count=1
addf comp.0 servo-thread
net fake-ohmic-probe-in hm2_7i92.0.encoder.00.velocity-rpm comp.0.in0
setp comp.0.in1 930000.0

and I can do a similar fake "arc ok" by configuring a not component.

However, it seems like I need to pick a ohmic probe pin in the Mesa card configuration in order to enable the ohmic probe capability. Looking through the config files there are some new ones that I am not familiar with for QtPlasmaC. My question is:

How can I enable ohmic probing and use a software-based ohmic probe when using pncconf to configure my machine? I am using LinuxCNC 2.9.2.

Thanks!

Scott
Last edit: 11 Feb 2024 22:51 by shasse. Reason: fix misspell

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

More
11 Feb 2024 22:53 #293112 by shasse
Probably worth adding that the "fake" ohmic probe works because when the laser head touches the workpiece, the capacitive sensor values change so much that the condition can easily be detected. So the capacitive sensor really does act like an analog ohmic sensor in a way.

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

More
11 Feb 2024 23:30 #293114 by tommylight
Is this the ohmic probe that outputs frequency depending on head/lens distance?
Why not use that as probe and THC so it keeps a constant distance from material?
Or is this not required?

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

More
11 Feb 2024 23:48 #293116 by shasse
It's a capacitive sensor that outputs frequency depending on the head/lens distance, and I'm definitely planning on using that for THC. I think I should also be able to use it for ohmic probing though, I just don't understand all of the moving parts to do that with a QtPlasmaC pncconf-generated configuration.

I'm happy to build a script to replace values or whatever I need to do, but there are new config files with QtPlasmaC so I for instance I don't know how to enable ohmic probing without specifying a Mesa pin as the ohmic probe input, which I don't need (or want) since I can do ohmic probing "in software" using a comp component and the capacitive sensor input.

Thanks!

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

More
12 Feb 2024 00:16 #293118 by tommylight
You will need the ohmic pin in hal however you look at it, you just need to omit the Mesa input and add whatever pin comes out of the comp to the existing pins/parameters.

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

More
12 Feb 2024 00:51 #293120 by phillc54
Usually the signal plasmac:ohmic-probe-out nets db_ohmic.out to plasmac.ohmic-probe and qtplasmac.ohmic_led_in.

You could possibly unlink db_ohmic.out and then net the output of your comp to this signal. This way you would totally bypass the debounce and control ohmic directly.
 
The following user(s) said Thank You: tommylight

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

More
12 Feb 2024 04:32 #293124 by shasse
So is the simplest way to configure an actual Mesa input pin to be an ohmic input and then either fix the generated config via something like sed or unlink the 7i92 input pin and net in the comp output I want to actually control the ohmic input? The debounce might still be useful since the encoder input I'm using to make a synthetic "ohmic" probe is from a physical process that could have some bounce. When I grepped through the config it seemed like there were moving parts to enable ohmic probing in general in other config files that this approach would make sure are set up properly. If this is the case its not the cleanest approach, but for now I have plenty of I/O on the 7i92 to burn.

and then for an "arc OK" that is always ok do something similar?

Thanks!

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

More
12 Feb 2024 04:54 #293126 by phillc54
If you want to use the debounce then you could net the output of your comp to the signal plasmac:ohmic-probe which should be created regardless of whether or not you designate a Mesa pin.
Ohmic probe is odd in that it can use either an external comp or the ohmic part of the plasmac comp, so It can look a tad convoluted.

For Arc-OK you would connect your comp output to the signal plasmac:arc-ok-in in lieu of a Mesa pin.
The following user(s) said Thank You: shasse

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

More
12 Feb 2024 05:26 #293130 by shasse
Thanks, I'll try this this week and report back.

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

More
17 Feb 2024 01:11 #293563 by shasse
Just repprting back that plasmac:arc-ok-in and plasmac:ohmic-probe worked great to resolve this issue. Thanks!
The following user(s) said Thank You: phillc54, tommylight

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

Moderators: snowgoer540
Time to create page: 0.154 seconds
Powered by Kunena Forum