Author Topic: Physical buttons for plasma  (Read 82987 times)

0 Members and 1 Guest are viewing this topic.

Offline Davek0974

  • Active Member
  • Posts: 2,408
    • View Profile
    • DD Metal Products
Re: Physical buttons for plasma
« Reply #440 on: November 26, 2016, 08:19:21 AM »
Hiya

absolutely must have a dead-band in there, mine is +/- 1v - does not seem much but should equate to about 0.5-1mm height.

Don't forget that when re-using code from one plat thickness to another, the speed will vary but the voltage will too.

THC is a compromise it seems, you can have one that will cut corrugated steel sheet but not work too well on fine detail or like mine it will work ok on most jobs as long as the plate does not warp too much - if cutting near the edge the waste can buckle upwards and the torch will follow the waste :) I also have a tip-saver in circuit that detects sharp rises in voltage and locks the down motion out as it thinks the tip has dived into a gap or kerf, then there is Mach Anti-Dive as well which is velocity based i recall.
Mach3 V062, CSMIO-IP/A controller, AC Servo Drives

Offline Hood

  • Mach4 Alpha
  • Posts: 25,775
  • Carnoustie, Scotland
    • View Profile
Re: Physical buttons for plasma
« Reply #441 on: November 26, 2016, 02:05:51 PM »
Thanks, I would have thought deadband was needed as well but miniTHC doesn't have it :(

Yes understood regards thickness feeds and volts, same with materials.

The miniTHC seemed very quick at following once I got it half working with altering the accel and Velocity but as said it trips the servo drive after a bit, having a deadband would I am sure cure this so hopefully Denis will be able to do something there, we will see.

There is an AntiDive setting in the miniTHC which I presume is similar to your Tip-saver function, it seems to work well even though I have just set mid range and not messed with it.

Offline Davek0974

  • Active Member
  • Posts: 2,408
    • View Profile
    • DD Metal Products
Re: Physical buttons for plasma
« Reply #442 on: November 26, 2016, 02:32:22 PM »
Seems odd they left out a dead-band, hopefully they can add one.

Maybe steppers are more tolerant of non-stop rapid moves than servos?
Mach3 V062, CSMIO-IP/A controller, AC Servo Drives

Offline Hood

  • Mach4 Alpha
  • Posts: 25,775
  • Carnoustie, Scotland
    • View Profile
Re: Physical buttons for plasma
« Reply #443 on: November 26, 2016, 02:56:02 PM »
Yes, said it was in their previous THC but they removed it from this one and no problems until now. Maybe most are using steppers and maybe servos are more prone to problems, don't know. Having said that servos  are used on machines that do 3d stuff so for things like photo engraving they do thousands of sharp up/down movements  and no problems.
Maybe I do just need to increase the servo size or maybe even reducing the gearing or having a larger pitch ballscrew would help.
Just don't really want to have to spend more money and time converting if something as simple as deadband would solve the issue.
I am thinking of testing out the internal THC of the CSMIO, only problem is the voltage divider is 50:1 and 20:1 would likely work better for the CSMIO.

I ran a test today cutting some alu, same code as before, 200mm long and 20mm wide rectangular path but this time I angled it at 45 degrees, the THC took it in its stride and that was at 4m/min, so it does handle things well.

Offline mc

  • Active Member
  • Posts: 381
    • View Profile
Re: Physical buttons for plasma
« Reply #444 on: November 26, 2016, 06:59:41 PM »
I would expect a servo to handle those kinds of moves, if tuned well. Could it be a case of the short/sharp moves, combined with a bit instability in the tuning causing the problem?

Could you try reducing the max current/acceleration?
You obviously don't want to reduce it too much that it's not fast enough to respond, but dropping it slightly might be enough to take the edge of the moves which are causing the heating.


Offline Hood

  • Mach4 Alpha
  • Posts: 25,775
  • Carnoustie, Scotland
    • View Profile
Re: Physical buttons for plasma
« Reply #445 on: November 26, 2016, 08:08:19 PM »
It could well be the tuning, don't even recall when I did it, could have been before everything was fitted fully.

The servo is small so it maybe does need an upgrade but it did seem to be handling things nicely. There were some on eBay a while back and I was going to buy a couple as they were cheapish £40 or so but decided I wouldn't bother ::)

Offline Hood

  • Mach4 Alpha
  • Posts: 25,775
  • Carnoustie, Scotland
    • View Profile
Re: Physical buttons for plasma
« Reply #446 on: November 27, 2016, 03:38:12 PM »
Had a look at the tuning today and it was pretty good but I saw the Low Pass Filter was set very low, not sure why. Anyway I upped it to the default value and things were better but it was still tripping out after a while of running a test programme I made. Loosened the tuning a wee bit and I was able to run the test programme for over an hour so I reckon it should be fine for the moment.
 The motor does get hot but not enough to fault the drive it seems.

I found a motor on eBay that is one size up from the one I have, not too a bad price considering it also has a 5:1 Alpha gearbox on it so I decided to buy it.
Not sure how I will fit it yet, it is a direct replacement for the one I have if I decide to keep the belts or I may alter things and do away with the belts and use the gearbox instead.

Anyway here is it cutting 10mm steel, same code that it was tripping out the last day after about 2/3 of the way round.
https://www.youtube.com/watch?v=xgbLEXSUuSk
Also a couple of pics of the cut.

Offline Davek0974

  • Active Member
  • Posts: 2,408
    • View Profile
    • DD Metal Products
Re: Physical buttons for plasma
« Reply #447 on: November 27, 2016, 04:24:25 PM »
Some nice looking cuts there, it really begs for a dead-band setting though ;)

Mach3 V062, CSMIO-IP/A controller, AC Servo Drives

Offline mc

  • Active Member
  • Posts: 381
    • View Profile
Re: Physical buttons for plasma
« Reply #448 on: November 27, 2016, 06:00:17 PM »
The problem is, if it's the continual movement that was overheating the motor, then the same will likely happen with a larger motor.

It'll be a case of striking the balance between moving too much that the motor overheats, and moving too slowly the torch height doesn't respond quick enough.

Offline Hood

  • Mach4 Alpha
  • Posts: 25,775
  • Carnoustie, Scotland
    • View Profile
Re: Physical buttons for plasma
« Reply #449 on: November 27, 2016, 06:20:13 PM »
Some nice looking cuts there, it really begs for a dead-band setting though ;)


I am thinking that if the CSMIO didn't used the accel from motor tuning then  the THC wouldn't be moving so much but sadly it does so I will just have to see what Denis or CS-Lab can come up with.

The problem is, if it's the continual movement that was overheating the motor, then the same will likely happen with a larger motor.

It'll be a case of striking the balance between moving too much that the motor overheats, and moving too slowly the torch height doesn't respond quick enough.

A bigger motor shouldn't have to use  as much of its available current to do the same job as the smaller motor so my thinking is that should reduce the heat.