Manipulating welder output with Arduino

[Ron66]

If you need a dac you could use an ATXMEGA128 with a built in dac. Maximum voltage for operation is 3.6V so I would run it at 3.3V then you could use either a precision 2.7V reference or waste some of the 12 bit resolution and set the reference to 3.3V. This would be ok if you use a 1% accurate regulator. I don't know if you could use that part with Arduino but I don't care much for the extra layers anyways. I use HPINFOTECH Codevision. They have a pretty good free version. I like it because the code wizard helps you set up all of the registers and generates the initialization code for you. There's a pretty active community, avrfreaks, that can offer a lot of help. My biggest concern with your project is how the high frequency start is going to affect your micro. Lots of EMI energy there. Fun project though. Really opens up what you could use for the motion detection. My guitar pedals from Digitech use a magnet and hall sensor to detect pedal angle with a simple calibration routine that I have to run maybe once a year. My Rolland drum kit uses this pretty neat pressure sensitive resistor strip for the high hat. There's also the possibility of using an optical wheel. Lots of people repurpose the scroll wheel of a mouse for rotary encoding. There's also linear optical encoders. Use two Avr's and a wireless link like zigbee and goodbye cord!

[joshuab]

The ATXmega line is not currently supported by the Arduino IDE and libraries, so no dice there yet. Adding a digital pot will be the simplest thing to do if I run into trouble with the PWM, I think. Regarding HF, I know that'll be a challenge, but I also know it's solvable. After all, the welder itself has tons of electronics in it, and the HF start doesn't bother it. Proper grounding may be required, but I'm sure it can be done.

[joshuab]

Today's work:

[sportbike]

When you are ready to start testing with an actual arc, I'd flip it to lift start at first. That way you can at least see if it is functional and deal with the HF as a seperate issue.

[everlastsupport]

So basically, it seems that you need to put your electronics in a metal box and ground it, or HF is going to potentially play havoc.

Like a Faraday cage.

[joshuab]

When you are ready to start testing with an actual arc, I'd flip it to lift start at first. That way you can at least see if it is functional and deal with the HF as a seperate issue.

This is what I did previously when I was testing the Arduino with a device that was triggered by the UV from the arc. The HF definitely messed with the Arduino, but it was plain naked, so no surprise.

Like a Faraday cage.

Yes, exactly. I work with wireless for a living, and although I don' t do any circuit design at all, I have seen the inside of my share of radios. The amplifier circuitry is always covered by a metal shield, to prevent it from interfering with the rest of the radio. Similar idea, although the HF arc, at thousands of volts, is a different story altogether. Nevertheless, the principle should be the same. I did some research into other devices that deal with the HF signal--aftermarket arc timers that use inductive sensors on the welding leads--and the ones that they sold for use with HF required grounding, so I'm guessing that's the answer. If more advanced hardening of the electronics is required, I may be out of luck.

I wonder whether it would work to ground the device's case to the negative terminal of the welder?

[everlastsupport]

A good filter, a choke. Run on its own power supply. I would not use - terminal. Use a common ground from inside. Where are you picking the ground up now. Did not look.

Ok Mark's turn, on the warranty.

[Rambozo]

Ok Mark's turn, on the warranty.

Can you say...

[DaveO]

I'm not generally a t-shirt slogan type of guy, but I'd wear that one.

Following this conversation has been educational, and as usual I'm impressed by the depth and breadth of knowledge on this board.

[joshuab]

A good filter, a choke. Run on its own power supply. I would not use - terminal. Use a common ground from inside. Where are you picking the ground up now.

The ground right now is coming from the power supply.

[everlastsupport]

The ground right now is coming from the power supply.

That's where it needs to be/stay.

[joshuab]

That's where it needs to be/stay.

Yeah. Maybe an easy workaround would be to use a 120v plug with only the ground wire connected. That way, you could plug into any 120v outlet to ground the box. Heck, at that point, may as well stick in a transformer and just pull from mains power.

[Rambozo]

Yeah. Maybe an easy workaround would be to use a 120v plug with only the ground wire connected. That way, you could plug into any 120v outlet to ground the box. Heck, at that point, may as well stick in a transformer and just pull from mains power.

Might want to use or build a universal supply so that it can run from 120 or 240. That would make things easier with a dual voltage welder.

[Ron66]

I remember when I first got my PP256 hooked up. I had no ground at all. After a couple of minutes of trying to weld my wife comes out and says "I don't know what your doing out here but everything in the house is flashing and beeping!" I drove a 6' ground rod outside of the door and use a 6ga wire and heavy duty alligator clip to ground the system. No problems since.

[sportbike]

I remember when I first got my PP256 hooked up. I had no ground at all. After a couple of minutes of trying to weld my wife comes out and says "I don't know what your doing out here but everything in the house is flashing and beeping!" I drove a 6' ground rod outside of the door and use a 6ga wire and heavy duty alligator clip to ground the system. No problems since.

Wow...you my want to investigate the grounding on your house and your other electrical equipment

[Ron66]

House is very well grounded. In addition to the power company ground I have 4 40ft ground rods that tie into the lightning rod system along with an external surge suppressor that ties into the main breaker panel. There is no earth ground connection in the power cord for the PP256 however which think is the bigger issue.

[everlastsupport]

House is very well grounded. In addition to the power company ground I have 4 40ft ground rods that tie into the lightning rod system along with an external surge suppressor that ties into the main breaker panel. There is no earth ground connection in the power cord for the PP256 however which think is the bigger issue.

If there is not ground post on the back of the machine, one option to try is to ground the outer case and see if that helps. But sounds like it is coming back into the panel.

[performance]

Can you say...

If Joshua can do what he's doing, then I imagine he's pretty sharp on fixing anything related to the operation and I don't think we'd be adverse to sending him the parts to fix it. Something like this, which is carefully documented is easier to see where some modification went off the rails would be more likely to be something we'd feel better about warranty. NOW, if he accidentally jabs a screwdriver between the main internal power components and curls his toes and blows a part,...I am sure he would be honest enough to admit it.

[sportbike]

House is very well grounded. In addition to the power company ground I have 4 40ft ground rods that tie into the lightning rod system along with an external surge suppressor that ties into the main breaker panel. There is no earth ground connection in the power cord for the PP256 however which think is the bigger issue.

Wow...well, maybe you are too grounded :-)

[joshuab]

If Joshua can do what he's doing, then I imagine he's pretty sharp on fixing anything related to the operation and I don't think we'd be adverse to sending him the parts to fix it. Something like this, which is carefully documented is easier to see where some modification went off the rails would be more likely to be something we'd feel better about warranty. NOW, if he accidentally jabs a screwdriver between the main internal power components and curls his toes and blows a part,...I am sure he would be honest enough to admit it.

Thanks for the vote of confidence, Mark. One reason I feel confident about this project is that it doesn't involve opening the welder up at all. I think it's much less likely that I'm going to damage something by mucking around with the pedal input, than if I was mucking around in the internals.

I do notice, however, that depending on how I set the voltage on the pedal control pins, the welder's output display will go outside its normal bounds. The welder is rated for 160 amps; it maxes out at 96 amps on 110v; but when I was still figuring out how to wire things up, I saw the display read as high as 190 amps--on 110v! Obviously, I never tried to strike an arc, nor would I, but I wager that if I did, the welder would attempt to comply, and would probably burn something out. For this reason, I have set the code to never allow the output on the welder's pins to go above about 2.5 volts. It seems that as that voltage on the control pins increases, the welder continues to attempt to increase output accordingly, even if that output is outside of the welder's normal operating range. So anybody who is following in my footsteps should be sure to put in a similar safeguard.

Currently, I'm manipulating the welder's output by putting a PWM voltage directly on the pins, but it would be better if I could use a digital potentiometer instead. That way, there would be no possible way to cause the welder to go outside its operating limits. The problem is, I can't find a digital pot that perfectly fits my needs... but that's a story for another day.

No update videos for now, folks. I spent a good part of one day trying to get a digital pot working, only to find out that it was not going to work at all for my needs. If you like, you can read more about my "search for the perfect digital pot" here. Yesterday, I got some mowing done and did a little soldering on the project. So the code hasn't advanced at all, but I'll certainly update you when I have more!