Another plasma component...
25 Mar 2019 00:55 - 25 Mar 2019 00:56 #129545
by phillc54
Replied by phillc54 on topic Another plasma component...
John,
Did you do linuxcnc.org/docs/devel/html/code/buildi...g-Build-Dependencies
If you did then I would have thought that all dependencies would have been loaded.
Cheers, Phill.
Did you do linuxcnc.org/docs/devel/html/code/buildi...g-Build-Dependencies
If you did then I would have thought that all dependencies would have been loaded.
Cheers, Phill.
Last edit: 25 Mar 2019 00:56 by phillc54.
Please Log in or Create an account to join the conversation.
25 Mar 2019 08:38 #129558
by rodw
From a system perspective, this is pretty easy. Just provide one more pin that is ORed with the float switch and use the output for probing.
The other thing I had trouble with was the probe becoming enabled while cutting. So becasue I used a gcode touch off routine, I set a digital output from the gcode before the probing sequence and reset it at the end of the routine. I then ANDed that with the probe input so it would only trigger when we were in the probing sequence. I Was thinking I could then sense when the probe/ohmic sensing was triggered while cutting (digital output not asserted) and wondered if I made an immediate offset that was equivalent to the cut height. I've had the gear to add Ohmic sensing here for a month and have not had time to modify the machine to implement it. (I need to fit a bigger drag chain as part of this).
Dunno if my ramblings help...
Replied by rodw on topic Another plasma component...
P.S.
Yes, I would be interested in being able to provide ohmic probing as well.
From a system perspective, this is pretty easy. Just provide one more pin that is ORed with the float switch and use the output for probing.
The other thing I had trouble with was the probe becoming enabled while cutting. So becasue I used a gcode touch off routine, I set a digital output from the gcode before the probing sequence and reset it at the end of the routine. I then ANDed that with the probe input so it would only trigger when we were in the probing sequence. I Was thinking I could then sense when the probe/ohmic sensing was triggered while cutting (digital output not asserted) and wondered if I made an immediate offset that was equivalent to the cut height. I've had the gear to add Ohmic sensing here for a month and have not had time to modify the machine to implement it. (I need to fit a bigger drag chain as part of this).
Dunno if my ramblings help...
The following user(s) said Thank You: phillc54
Please Log in or Create an account to join the conversation.
- islander261
- Offline
- Platinum Member
Less
More
- Posts: 757
- Thank you received: 216
25 Mar 2019 15:13 #129584
by islander261
Replied by islander261 on topic Another plasma component...
Rod
I OR2 the ohmic probe signal and the float switch signal and feed the output to the probe input on my current system. Out of an abundance of caution my ohmic sensing circuit is only connected to the work and the torch shield through a relay (well actually 2 because I only have SPST relays on my chinacom relay board) when actually doing the probing. I do have a GUI button so I can do a test connection for trouble shooting a shorted torch tip (happens more often than you think with the Finecut consumables I use). I did this even though the opto isolator I use is good to 2Kv or so isolation. I know this means that I can't use it for plate contact sensing when I am actually cutting but I don't need that because of my torch break away sensing switch (sensitive enough that the torch mount usually doesn't come all the way off the Z). Your A120 when used with the automation torch doesn't need this because the built in firmware will throw an error and shut off the torch if this happens, on US machines the 85psi led will light indicating this error. I believe we discussed my ohmic sensing setup in some detail in the recent past.
Phill
I didn't notice any problems during my installation. I was actually quite surprised by the lack of them. Thanks to Rod I didn't even have to look into the problem, straight to Synaptic to find the the required library wasn't installed and install it.
Today's adventure is find out why I have lost my hal widgets out of my Glade setup. I went through a bunch screwing around to get them installed a few months ago and now they are gone.
John
I OR2 the ohmic probe signal and the float switch signal and feed the output to the probe input on my current system. Out of an abundance of caution my ohmic sensing circuit is only connected to the work and the torch shield through a relay (well actually 2 because I only have SPST relays on my chinacom relay board) when actually doing the probing. I do have a GUI button so I can do a test connection for trouble shooting a shorted torch tip (happens more often than you think with the Finecut consumables I use). I did this even though the opto isolator I use is good to 2Kv or so isolation. I know this means that I can't use it for plate contact sensing when I am actually cutting but I don't need that because of my torch break away sensing switch (sensitive enough that the torch mount usually doesn't come all the way off the Z). Your A120 when used with the automation torch doesn't need this because the built in firmware will throw an error and shut off the torch if this happens, on US machines the 85psi led will light indicating this error. I believe we discussed my ohmic sensing setup in some detail in the recent past.
Phill
I didn't notice any problems during my installation. I was actually quite surprised by the lack of them. Thanks to Rod I didn't even have to look into the problem, straight to Synaptic to find the the required library wasn't installed and install it.
Today's adventure is find out why I have lost my hal widgets out of my Glade setup. I went through a bunch screwing around to get them installed a few months ago and now they are gone.
John
Please Log in or Create an account to join the conversation.
- islander261
- Offline
- Platinum Member
Less
More
- Posts: 757
- Thank you received: 216
25 Mar 2019 16:46 #129589
by islander261
Replied by islander261 on topic Another plasma component...
All
I don't know why this happens. When I try to open any of the plasmac glade files by right clicking on it in the file browser and selecting Glade to open it I don't get all the hal widgets in the left pane and I get error messages where they should be on the preview screen. Now if I start Glade from start menu or from the desktop short cut it starts with all the widgets loaded and if I open a plasmac Glade file from the file menu it opens correctly. Oh well I don't have time to fight this.
A bigger problem is that some of the files won't open in Glade... I opened some of those files with a text editor and found some names to contain special characters. So for instance the file plasmac_buttons.glade works when you run linuxcnc and select a configuration that uses it but if you try and open it with Glade you get nothing, not even a warning. I think this may have something to do with the problem:All of the files that won't open in Glade have a similar problem with special characters in the file names.
John
I don't know why this happens. When I try to open any of the plasmac glade files by right clicking on it in the file browser and selecting Glade to open it I don't get all the hal widgets in the left pane and I get error messages where they should be on the preview screen. Now if I start Glade from start menu or from the desktop short cut it starts with all the widgets loaded and if I open a plasmac Glade file from the file menu it opens correctly. Oh well I don't have time to fight this.
A bigger problem is that some of the files won't open in Glade... I opened some of those files with a text editor and found some names to contain special characters. So for instance the file plasmac_buttons.glade works when you run linuxcnc and select a configuration that uses it but if you try and open it with Glade you get nothing, not even a warning. I think this may have something to do with the problem:
<interface>
<!-- interface-requires gladevcp 0.0 -->
<!-- interface-requires Pàì5þU 0.0 -->
<!-- interface-requires gtk+ 2.12 -->
<!-- interface-naming-policy project-wide -->
John
Please Log in or Create an account to join the conversation.
25 Mar 2019 17:11 #129591
by rodw
Replied by rodw on topic Another plasma component...
John , glad you are up and running. Yes, my Ohmic sensing is somewhat based on your ideas shared earlier except I'm using a 24 volt isolated slimline DIN rail power supply instead of your 12 volt wall wart. The biggest problem is as a carry over from the machine upgrade is there is not enough room in my drag chain any more so I have to upgrade the tray it sits on to upgrade it. I've been able to fold up a new one but its not installed yet.
Please Log in or Create an account to join the conversation.
25 Mar 2019 20:59 #129612
by phillc54
That is a problem that I have run into with Glade quite often.
Just delete that entire line with a text editor.
Cheers, Phill.
Replied by phillc54 on topic Another plasma component...
All of the files that won't open in Glade have a similar problem with special characters in the file names.
That is a problem that I have run into with Glade quite often.
Just delete that entire line with a text editor.
Cheers, Phill.
The following user(s) said Thank You: islander261
Please Log in or Create an account to join the conversation.
- islander261
- Offline
- Platinum Member
Less
More
- Posts: 757
- Thank you received: 216
25 Mar 2019 21:21 #129615
by islander261
Replied by islander261 on topic Another plasma component...
Phill
Thanks, I am quite accomplished at causing Glade to crash but this is new one on me. That is why my first move is to always open a Glade file in a text editor and check for funny stuff.
John
Thanks, I am quite accomplished at causing Glade to crash but this is new one on me. That is why my first move is to always open a Glade file in a text editor and check for funny stuff.
John
Please Log in or Create an account to join the conversation.
25 Mar 2019 21:43 - 25 Mar 2019 23:02 #129619
by phillc54
Replied by phillc54 on topic Another plasma component...
I am in the habit of saving every couple of minutes when using Glade as crashing is one thing it does very well...
Cheers, Phill.
EDIT1: Also I should have checked for those errors before pushing to Git, sorry about that I'll get right on it...
EDIT2: Done!
Cheers, Phill.
EDIT1: Also I should have checked for those errors before pushing to Git, sorry about that I'll get right on it...
EDIT2: Done!
Last edit: 25 Mar 2019 23:02 by phillc54.
Please Log in or Create an account to join the conversation.
26 Mar 2019 00:33 - 26 Mar 2019 00:34 #129631
by phillc54
Replied by phillc54 on topic Another plasma component...
John, Rod,
After reading your ohmic setups I am thinking:
At start of probing, ohmic-enable would switch on, the first to sense ohmic-probe / float-switch would set the stock height then ohmic-enable would switch off.
Should I also have user selectable inputs to:
Cheers, Phill.
After reading your ohmic setups I am thinking:
- ohmic-probe (input) - this would be a separate input for the probe.
- ohmic-enable (output) - this would be on only while probing, off at all other times.
At start of probing, ohmic-enable would switch on, the first to sense ohmic-probe / float-switch would set the stock height then ohmic-enable would switch off.
Should I also have user selectable inputs to:
- enable/disable the float-switch as a secondary probe.
- enable/disable ohmic-enable for testing as per Johns setup.
Cheers, Phill.
Last edit: 26 Mar 2019 00:34 by phillc54.
Please Log in or Create an account to join the conversation.
- islander261
- Offline
- Platinum Member
Less
More
- Posts: 757
- Thank you received: 216
26 Mar 2019 02:20 #129635
by islander261
Replied by islander261 on topic Another plasma component...
Phill
The ability for the operator to easily check for a shorted torch is very important, it should be easy like a button and indicator on the GUI. When using ohmic sensing the float switch is always used as a Z minimum limit if it isn't ORed with ohmic signal.
I really tried hard to get just the float switch to work for probing. I tried many combinations of counter balance springs and never could get it right. If I had enough force negation with the springs then the float assembly would oscillate and over shoot when under THC. When I had it stiff enough so there was no oscillation or over shoot then the thin sheet is deflected a random amount. Actually at Z accelerations that are fast enough to cut corrugated sheet the response of the torch and float assembly get to be a problem.
Please find attached the Gcode probing subroutine I currently use. It is not perfect. The biggest problem is if the probing problem isn't fixed with a retry the subroutine exits with an error code when it should just jump back to the top with the program paused and the error message displayed. This is in imperial units because it was never met for an international audience. The distance sanity check works ok for the thin sheet I use, it has never been tested on thick material that will not deform or rock under the pressure from the torch.
The best for someone doing hobby or job shop work will be a way to select whether to use ohmic sensing or the float switch. This could be a check box in a setup pane and/or a gcode command sent by the PP that depends on the selected tool.
Do you know if anyone has used the plasmac component on a live machine?
John
The ability for the operator to easily check for a shorted torch is very important, it should be easy like a button and indicator on the GUI. When using ohmic sensing the float switch is always used as a Z minimum limit if it isn't ORed with ohmic signal.
I really tried hard to get just the float switch to work for probing. I tried many combinations of counter balance springs and never could get it right. If I had enough force negation with the springs then the float assembly would oscillate and over shoot when under THC. When I had it stiff enough so there was no oscillation or over shoot then the thin sheet is deflected a random amount. Actually at Z accelerations that are fast enough to cut corrugated sheet the response of the torch and float assembly get to be a problem.
Please find attached the Gcode probing subroutine I currently use. It is not perfect. The biggest problem is if the probing problem isn't fixed with a retry the subroutine exits with an error code when it should just jump back to the top with the program paused and the error message displayed. This is in imperial units because it was never met for an international audience. The distance sanity check works ok for the thin sheet I use, it has never been tested on thick material that will not deform or rock under the pressure from the torch.
The best for someone doing hobby or job shop work will be a way to select whether to use ohmic sensing or the float switch. This could be a check box in a setup pane and/or a gcode command sent by the PP that depends on the selected tool.
Do you know if anyone has used the plasmac component on a live machine?
John
Please Log in or Create an account to join the conversation.
Moderators: snowgoer540
Time to create page: 0.172 seconds