Steve McConner wrote ...

Mary: The TIG arc starter is basically a Tesla coil. And you can use the ignition coil/lamp dimmer thing as the power supply for a Tesla coil, you just substitute it for the neon sign transformer in the regular TC/arcstarter circuit. I assume this is what the original poster plans to do.

Steve McConner wrote ...

If you put ferrite in the HF transformer, it should be a bundle of rods, or something with an airgap. If you used a toroid or an ungapped E core set, the welding current would probably saturate it hard, and it would be no better than an air core.

But then you can argue that if welding current is flowing, the arcstarter has already done its job and doesn't need to work any more.
*SNIP*
You'll probably be able to adjust the HF intensity by twiddling the dimmer, but it will also change the point in the cycle where the arcstarter fires. If your welder's output voltage is passing through zero at that point, results could be dismal.

wrote ...
The answer to "how many turns" - as many as you can tolerate the IR drop of in your welding circuit. The more turns, the better the HF will get coupled.

wrote ...
The capacitor will probably need to be smaller, I doubt an ignition coil can charge as much capacitance as a NST can.

wrote ...

I've AC welded hundreds of stick welding electrodes and never had an arc extinguish for any reason but me being too gung-ho and sticking the electrode too fast. I have yet to see an explanation as to why this is different between SMAW and TIG.

wrote ...

Ok thanks for that answer, next question am I better off wrapping primary next to secondary or intertwining primary and secondary? Or is that too reliant on the core material and design?

wrote ...

The capacitor value in question on the OP is the one to charge the primary side of the ignition coil not the one on the secondary side of the ignition coil (and primary side of the welder coil)

Steve McConner wrote ...

[Automotive spark plugs tend to have resistors inside that make them completely useless as Tesla coil spark gaps.

lightlinked wrote ...

how is the output current varied with the foot pedal? is there some kind of add on module you can get for a stick welder?

wrote ...

That's pretty cool. it looks like a giant dimmer for a welder. although wont that be more of a constant voltage instead of a constant current? i imagine this would be a good addon to one of those crappy two setting flux core wire feed welders

wrote ...
i wonder how the non inverter tig welders like your linde do it? im curious to what arcane magic they got in those giant boxes. i imagine the pedal sets a target current and it has a current transformer watch the output for a feedback loop to a phase angle controller.

wrote ...
Another update, no pics today.

The tig pedal is working good, but I'm going to change the schematic. Using another pot I found that the range was much more linear with the pedal if there was a 15k resistor in parallel with the potentiometer on the pedal. So I'm going to try adding a couple resistors in and see what happens.

I finally got all the parts together for the Hi-frequency arc stabilizer I'll need to avoid scratch-starting and contaminating the aluminum I intend to weld.

I'm basing most of it off of this http://www.casano.com/projects/hfstart/index.html

I want anyone who is going to build something similar to the one linked above to recognize that his schematic is WRONG and INCOMPLETE!

In order for the above circuit to work the tungsten spark gap can't be in parallel with the primary of the welding transformer, it has to be connected in between the welding transformer and the two resistors. I'm thinking that this will also save the ignition coil from the backlash of saturation in the transformer core when you get a high-amperage arc going. Someone on hackaday commented that they even eliminated the 2000 ohm 10W resistor to ground but since this isn't in operation yet I'm hesitant to change things around too much before I start welding with it. If you look at his pics his completed project is done correctly if you need to see how it lays out visually. I sent him an email about it but didn't get a response yet.

His also uses a switch in line with the light dimmer to start the arc and doesn't mention but should REALLY use some sort of capacitor to prevent excess corona from trashing the innards of the welder its attached to.

The casano site used a torroidal (donut shaped) round ferrite core, for mine I'm using a double-E core from a BIG 460v sodium light transformer that blew up. I used my portaband to cut half the transformer windings off of one side and used a punch to drive all the copper out...I probably got 4-5 lbs of copper off of it. I don't know enough about flux to know specifics as to how this will effect the circuit or arc, just that winding it with some 10 ga primary and junk 1/0 welding cable I was able to get arc to jump about 1/32" between the output cables. No visible corona leakage and a pretty short arc jump so I'm going to rewind it.

I know he says that he used 20kv test probe wire for his but I have a much cheaper solution to that problem you can even find locally: ignition coil wire. If the parts store asks if you want resistor wire the answer is no. I got some "racing" wire in bulk for about 1.25/ft from the local napa. Plus racing wire uses a solid, solderable core whereas other wires might use graphite impregnated fiber strands or god knows what and solidly integrating that into your circuit probably won't be as easy. I picked up 25' of racing wire from napa and am waiting for the local Orschelins (kinda like TSC) to get more welding cable in so I can wind the secondary with #4 double insulated welding cable. Hopefully with this arrangement I'll get quite a few more turns on the transformer core and better "saturation" thus more HV energy in the welding line and the ability to jump a bigger gap if I want to.

Because I couldn't get smaller quantities of the "hard to get" items I still have 4 extra 2000 ohm 10 watt ceramic resistors, an extra 100 ohm 10 watt and an extra 500 pf 30 kv doorknob cap if anybody wants to build their own PM me. I'd let them all go for $20 shipped. The resistors aren't necessarily that hard to get but values that large radioshack won't have and digi-key will put the hurt on you for price in small quantities. I also have 2 more blown up sodium bulb transformer cores I'd let go for $20/ea + shipping once I strip the dead copper off them. And 2 more SCR modules I'd let go for $20/ea shipped.

I know welder construction isn't a huge topic here but I had another DIY idea for once the high frequency part is nailed down: I would really like to have AC balance control for welding aluminum. Last night while daydreaming I had this though: if I had a pulse generator running to one driver and also to a NOT gate (inverter) before it hits the other driver (in an inverse parallel gate arrangement of 4 drivers) I could easily take rectified DC from the welder and chop it back into AC before it hits the ground clamp and tig torch. I'm curious how a transformer machine accomplishes AC balance control most of the newer machines that have AC balance control are inverter-based which is far beyond the scope of this project.