QtPlasmaC release.
15 Mar 2021 09:00 - 15 Mar 2021 22:10 #202319
by phillc54
QtPlasmaC release. was created by phillc54
QtPlasmaC is now fully functional and is able to be used on plasma tables. It has all the functionality of PlasmaC plus some additional features. A couple of bugs with eoffsets which only show up due to the way the plasmac component uses eoffsets were also kindly fixed by Dewey (dgarrett) in 2.9 branch during the testing phases of QtPlasmaC.
There will be no more new development work done on any of the PlasmaC configs, all new work is limited to the QtPlasmaC GUI and config plus any supporting files.
The QtPlasmaC config is designed to work on LinuxCNC master branch and will work on Python2 or Python3 builds.
Documentation is available from linuxcnc.org/docs/devel/html/plasma/qtplasmac.html
There are helper scripts provided to migrate from an existing PlasmaC configuration which leaves the existing configuration untouched. This gives the user the ability to revert to the original PlasmaC config should any anomalies occur due to the migration. Details of these scripts are in the documentation.
Many thanks to Greg (snowgoer540) for his tireless testing, suggestions and encouragement, as well as all his work on the documentation.
Also a big thank you to Chris (cmorley) for his wonderful QtVCP infrastructure that made this possible and also for the changes he made to fit in with the odd way we do things.
There will be no more new development work done on any of the PlasmaC configs, all new work is limited to the QtPlasmaC GUI and config plus any supporting files.
The QtPlasmaC config is designed to work on LinuxCNC master branch and will work on Python2 or Python3 builds.
Documentation is available from linuxcnc.org/docs/devel/html/plasma/qtplasmac.html
There are helper scripts provided to migrate from an existing PlasmaC configuration which leaves the existing configuration untouched. This gives the user the ability to revert to the original PlasmaC config should any anomalies occur due to the migration. Details of these scripts are in the documentation.
Many thanks to Greg (snowgoer540) for his tireless testing, suggestions and encouragement, as well as all his work on the documentation.
Also a big thank you to Chris (cmorley) for his wonderful QtVCP infrastructure that made this possible and also for the changes he made to fit in with the odd way we do things.
Last edit: 15 Mar 2021 22:10 by phillc54. Reason: add a missed thank you
The following user(s) said Thank You: tommylight, Clive S, EW_CNC, robertspark, Franklin, CNCFred, Mud, snowgoer540, web123456, paulsao
Please Log in or Create an account to join the conversation.
- snowgoer540
- Offline
- Moderator
Less
More
- Posts: 2383
- Thank you received: 776
15 Mar 2021 10:08 #202323
by snowgoer540
Replied by snowgoer540 on topic QtPlasmaC release.
Many thanks are owed to you as well, Phill. It's hard to believe this has been in development since the end of September. There are countless hours of coding here, and probably even more wading through my borderline OCD bug reports. Thank you for letting me be part of the process; I've learned a ton.
I am really proud of how it's turned out. I look forward to see where the future takes the project.
I am really proud of how it's turned out. I look forward to see where the future takes the project.
Please Log in or Create an account to join the conversation.
15 Mar 2021 13:06 #202346
by rodw
Replied by rodw on topic QtPlasmaC release.
I think this is really a milestone. Now we have a good purpose built GUI for plasma and a full featured controller suite.
I think we are still only scratching the surface with what linuxcnc can do with plasma. I'd like to think future development brought some high def features to entry level plasma machines. This will require some changes to the core motion control code but I have a working example. I just have no time.
Given the ability to control current in real time (which probably can't be done in with a Hypertherm), we should be able change alter the current to keep voltage constant when negotiating a corner. With the power of external offsets, we could also offset the torch path in real time (no gcode) so the wider kerf at slower velocity could be compensated by moving the torch further away from the programmed path.
I might add that Grotius has experimented with direct current control of Thermal Dynamics/ESAB machines using a digital potentiometer. Also throw in new member Beefy's work measuring the plasma current, we have the beginnings of PID based current control...
So if anybody knows a bit about coding and is looking for a challenge, I have the ideas! Just let me know!
I think we are still only scratching the surface with what linuxcnc can do with plasma. I'd like to think future development brought some high def features to entry level plasma machines. This will require some changes to the core motion control code but I have a working example. I just have no time.
Given the ability to control current in real time (which probably can't be done in with a Hypertherm), we should be able change alter the current to keep voltage constant when negotiating a corner. With the power of external offsets, we could also offset the torch path in real time (no gcode) so the wider kerf at slower velocity could be compensated by moving the torch further away from the programmed path.
I might add that Grotius has experimented with direct current control of Thermal Dynamics/ESAB machines using a digital potentiometer. Also throw in new member Beefy's work measuring the plasma current, we have the beginnings of PID based current control...
So if anybody knows a bit about coding and is looking for a challenge, I have the ideas! Just let me know!
The following user(s) said Thank You: tommylight
Please Log in or Create an account to join the conversation.
- tommylight
- Away
- Moderator
Less
More
- Posts: 18826
- Thank you received: 6324
15 Mar 2021 15:04 #202355
by tommylight
Replied by tommylight on topic QtPlasmaC release.
I have a stinking feeling the current will need to be increased when cutting a corner, just ask the mill guys.
Please Log in or Create an account to join the conversation.
15 Mar 2021 19:54 #202398
by rodw
I think of it differently. A plasma cutter is a constant current device and ohms law applies.
If travel speed and torch height provide the resistance (which is constant at a constant height and velocity), there is an equilibrium between resistance, voltage and current, the plasma generates the correct power to cut the material and everybody is happy. Increase the torch height, you increase the resistance so voltage goes up to achieve constant current.
If the torch slows down on an arc and we lock the height, we know voltage goes up, so resistance is increasing. I think if cutting an arc shape, we could lock the height and vary current to achieve constant voltage. This should result in the same power being generated so the kerf width should remain unchanged.
The other thing is we know that not everybody's acceleration is identical so there is no one size fits all when it comes to the point at which cutting velocity slows. Its the actual accelleration that determines the arc radius limit (the centripetal limit) at which velocity must be slowed. But given cut velocity x & y axis acceleration, this limit can be calculated in advance. We could then improve Plasmac's hole cutting algorithm.
If your head does not hurt when reading this, then reread it. Its hard to understand becasue we are talking about the collision of the physics of motion with ohms law...
Replied by rodw on topic QtPlasmaC release.
I have a stinking feeling the current will need to be increased when cutting a corner, just ask the mill guys.
I think of it differently. A plasma cutter is a constant current device and ohms law applies.
If travel speed and torch height provide the resistance (which is constant at a constant height and velocity), there is an equilibrium between resistance, voltage and current, the plasma generates the correct power to cut the material and everybody is happy. Increase the torch height, you increase the resistance so voltage goes up to achieve constant current.
If the torch slows down on an arc and we lock the height, we know voltage goes up, so resistance is increasing. I think if cutting an arc shape, we could lock the height and vary current to achieve constant voltage. This should result in the same power being generated so the kerf width should remain unchanged.
The other thing is we know that not everybody's acceleration is identical so there is no one size fits all when it comes to the point at which cutting velocity slows. Its the actual accelleration that determines the arc radius limit (the centripetal limit) at which velocity must be slowed. But given cut velocity x & y axis acceleration, this limit can be calculated in advance. We could then improve Plasmac's hole cutting algorithm.
If your head does not hurt when reading this, then reread it. Its hard to understand becasue we are talking about the collision of the physics of motion with ohms law...
Please Log in or Create an account to join the conversation.
- robertspark
- Offline
- Platinum Member
Less
More
- Posts: 915
- Thank you received: 216
15 Mar 2021 20:09 #202399
by robertspark
Replied by robertspark on topic QtPlasmaC release.
the voltage goes up when the torch slows because the kerf gets wider (more material has been cut / blown away than the optimum kerf width).
lowering the amperage allows you to lower the feedrate OR, to put it another way>>>> lowering the feedrate allows you to lower the amperage..... The cut shape (bevel) does change and optimum will still be the optimum feedrate.
For full control, you may need to control the amperage and the air pressure too..... that can be done, via the same way that hypertherm do it with an electronic pressure regulator, such as one of these (via 4-20mA controls signal)
uk.rs-online.com/web/p/pneumatic-regulat...ALw_wcB&gclsrc=aw.ds
lowering the amperage allows you to lower the feedrate OR, to put it another way>>>> lowering the feedrate allows you to lower the amperage..... The cut shape (bevel) does change and optimum will still be the optimum feedrate.
For full control, you may need to control the amperage and the air pressure too..... that can be done, via the same way that hypertherm do it with an electronic pressure regulator, such as one of these (via 4-20mA controls signal)
uk.rs-online.com/web/p/pneumatic-regulat...ALw_wcB&gclsrc=aw.ds
Please Log in or Create an account to join the conversation.
15 Mar 2021 20:27 #202403
by rodw
Exactly! The resistance increases as the arc is stretched looking for material to burn.
You'd have to review your cut charts to see how important pressure was by comparing between amperage values on the same material. Often, the same pressure is used. I also wonder how responsive a pressure valve would be in real time? I think current control would be much more responsive. I did get excited when somebody on the forum here mentioned a plasma brand that had 0-10 volt control of current. I can't remember the brand.
Replied by rodw on topic QtPlasmaC release.
the voltage goes up when the torch slows because the kerf gets wider (more material has been cut / blown away than the optimum kerf width).
Exactly! The resistance increases as the arc is stretched looking for material to burn.
You'd have to review your cut charts to see how important pressure was by comparing between amperage values on the same material. Often, the same pressure is used. I also wonder how responsive a pressure valve would be in real time? I think current control would be much more responsive. I did get excited when somebody on the forum here mentioned a plasma brand that had 0-10 volt control of current. I can't remember the brand.
Please Log in or Create an account to join the conversation.
- robertspark
- Offline
- Platinum Member
Less
More
- Posts: 915
- Thank you received: 216
15 Mar 2021 21:08 #202410
by robertspark
Replied by robertspark on topic QtPlasmaC release.
air pressure and current are static on all air plasma systems...... well static during the cut that is, you get to set them before
you can get the electonic regulators to adjust within 0.1 sec, normally the changes being made are small so the response factor can be lower.
_________________________________________________________
to cut you need power and that is an optimum number (like welding ~40A/mm)
power, not exactly / just current or voltage..... if you lower the cut speed, the voltage rises.... but if you lower the current in sync with lowering the feedrate the voltage "should" stay roughly the same because P=V*I.... negating the requirement for locking the torch height..... given the amperage is the bit you are changing and are holding the voltage static......
thats the theory.....
whether it makes the cut any better, that cutting without it..... or whether it causes increased consumable wear than cutting without it..... those are different factors.....
afterall if there is little to no benefit in the end product.... whats the point of go faster stripes...
you can get the electonic regulators to adjust within 0.1 sec, normally the changes being made are small so the response factor can be lower.
_________________________________________________________
to cut you need power and that is an optimum number (like welding ~40A/mm)
power, not exactly / just current or voltage..... if you lower the cut speed, the voltage rises.... but if you lower the current in sync with lowering the feedrate the voltage "should" stay roughly the same because P=V*I.... negating the requirement for locking the torch height..... given the amperage is the bit you are changing and are holding the voltage static......
thats the theory.....
whether it makes the cut any better, that cutting without it..... or whether it causes increased consumable wear than cutting without it..... those are different factors.....
afterall if there is little to no benefit in the end product.... whats the point of go faster stripes...
Please Log in or Create an account to join the conversation.
- tommylight
- Away
- Moderator
Less
More
- Posts: 18826
- Thank you received: 6324
15 Mar 2021 21:27 - 16 Mar 2021 10:59 #202416
by tommylight
Replied by tommylight on topic QtPlasmaC release.
Deleted as it was off topic.
Sorry Phill.
Sorry Phill.
Last edit: 16 Mar 2021 10:59 by tommylight.
Please Log in or Create an account to join the conversation.
15 Mar 2021 22:09 #202422
by phillc54
Replied by phillc54 on topic QtPlasmaC release.
This topic is about the QtPlasmaC release, can we please keep it to that...
I forgot to mention a big thank you to Chris (cmorley) for his wonderful QtVCP infrastructure that made this possible and also for the changes he made to fit in with the odd way we do things.
I forgot to mention a big thank you to Chris (cmorley) for his wonderful QtVCP infrastructure that made this possible and also for the changes he made to fit in with the odd way we do things.
The following user(s) said Thank You: tommylight, snowgoer540
Please Log in or Create an account to join the conversation.
Moderators: snowgoer540
Time to create page: 0.106 seconds