Plasma Torch Height controller (not so simple voltage divider)

[ArcSpark]

Hi all,
I'm new here so first I will introduce myself:
I am 29 and live in the Netherlands and work as a technician E&I
Hobbies include welding, CNC and "basic" electronics

I just finished the build of my practice no-budjet CNC engraver.
http://www.youtube.com/watch?feature=player_profilepage&v=piWEYOGqN44

Now I know I can, I want to make a CNC plasma cutting table.

The mechanical and electrical part is no problem, so I will save that for later.

The problem lies in the torch height measurement. The torch of the plasma cutter must stay at the same height during cutting, even when the plate bends because of the heat.

The trick now is to create a control system that allows the torch to follow the "curved" plate. There are special kits for sale that can do this but they are very expensive. so there is no fun in doing that.
These kits work with an arc voltage measurement, the arc voltage is a measure of the distance between the torch and the plate. If you know the distance the control system is just a small trick.

My own idea is to use a voltage divider for scaling of the open arc voltage to a 0-5v voltage.
This I can read as an analog signal into an Arduino and then use 2 outputs (up, down) to adjust the height.
The problem lies in filtering the signal and protecting my Arduino to the high peaks.
A plasma torch puts out anything from 0 to 200/300 Vdc (negative BTW) and it's very dirty. A voltage divider gets this down to the required 0 to 5 for the arduino, but I need to make sure no spikes go above this.

My knowledge of electronics is limited, two years ago I have done a basic course electronics because at work it sometimes comes in handy.

Maybe some of you already have had this problem before and could give me some helpful tips. Thanks for your help!

Greetings ArcSpark

 

[duke37]

I have not used a plasma torch or an arduino so am not qualified to give a definitive answer.
One problem will be to get the system stable, having a low pass filter to get rid of the noise could give sufficient phase shift for oscillation. There are books which deal with this rather complicated subject.
If you sense the arc voltage with a high value resistor, you can then place a zener from the output to ground to limit the voltage spikes. Then feed the current into an inverting amplifier with a feed back capacitor to give a low pass filter. You then have a smoothed signal of the right polarity. Instead of a computer, you could use a pair of comparators to generate the up/down signals, using the computer would enable you to use proportional control.

 

[Laplace]

Indeed, there are many books written on modern control systems theory; however, they do require that you write the differential equations which govern the motion of your physical system prior to designing the feedback electronics for controlling that system. Otherwise you can start with a first-order feedback control circuit with experimentation to vary the amplitude and frequency response until achieving something close to acceptable. It may be close enough. The same considerations apply whether the control circuit is analog or digital but a microcontroller makes a non-linear response easier to achieve.

 

[Laplace]

I would use something like the attached circuit to present the -300 Volt signal to the Arduino ADC input. Note that there will be a small positive offset that you can subtract out following the conversion. The RC filtering is based on a guess about the characteristics of the plasma signal.

 

[jackorocko]

http://www.centricut.com/New_Lessons/lessons_10.html

From what I can tell from the above article, the voltage increases when the torch-to-work distance increases and the voltage decreases when the torch-to-work distance decreases. The arc voltage is the same as the output voltage from the cathode to the anode. The desired voltage output can be figured from known data about the type of material and the thickness of the material you're cutting.

I would really be interested in seeing what you come up with. I have a vested interest in this project. What is cooler then plasma cutting in 3d space??? Really, I can't think of anything

 

[jackorocko]

http://www.plasma777.com/driver/agelkom.htm

Here is another THC, this one looks to use an H-bridge to control the motor. They don't specifically say how they know the distance or set it from what I can see. This H-brige could easily be made to run from the output of a op-amp in differential mode.

 

[Consultant]

Wouldn't this use a circuit like ones developed to automatically adjust the distance between the carbon electrodes in an arc-lamp?

Looking for something like that then, I found this:

http://www.lafavre.us/brush/lamparc.htm

Could that be of some help, or the books mentioned?

 

[ArcSpark]

Thanks for brainstorming so far, i'm in the middle of the mechanical build now.
when i'm ready with welding i'll get back to the electronics.

Thanks so far and i'll let you know how it go's

 

[daniel1987]

Thanks for brainstorming so far, i'm in the middle of the mechanical build now.
when i'm ready with welding i'll get back to the electronics.

Thanks so far and i'll let you know how it go's

How is this going, I'd be interested to know?

 

[ArcSpark]

How is this going, I'd be interested to know?
http://www.kerfdevelopments.com/plasmacutting.html

I'm still working with the metal