How to protect a TIG welder rectifier / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

How to protect a TIG welder rectifier

1 2 3 next

Move Thread

LAN_403

Rick

Sat Nov 10 2012, 07:05PM

Registered Member #4220 Joined: Sun Nov 20 2011, 12:11AM
Location:
Posts: 6

Hello, everyone;

I've learned a lot by lurking here, but now must appeal directly for some advice. I've built a TIG from an AC buzzbox and am having trouble adding a HV start circuit. I don't understand enough about electronics to keep the HV backing up from the coupling coil into the rectifier. In the various designs available on the web they refer to putting a cap across the rectifier output, but I don't understand how a 400v cap can hold up to the 10kv coming out of my OBIT. A diode has been suggested and there are hv diodes out there, but it seems that they'd have to be able to handle the high amp welding current, too. I found this thread in the 4hv archives:

It refers to a choke with contra-wound coils. Would this be essentially a 1:1 xmfr? I could build it out of a MOT.

I'd certainly appreciate any sugestions; I've already fried a couple of rectifiers.

Rick A

GrantX

Sun Nov 11 2012, 06:14AM

Registered Member #4074 Joined: Mon Aug 29 2011, 06:58AM
Location: Australia
Posts: 335

From what I understand, which is admittedly very little, the DIY TIGs I've seen use a high frequency, high voltage supply as an arc starter. With this, they add a low pass filter to the rectifier, which is the 2 coils and caps you mentioned. The caps act to short out the incoming HV, since the capacitance is only seen as a small impedance to the high frequency, conversely the 2 coils have a very low DC resistance, but at high frequency the have a very large impedance, which "blocks" the HV and forces it to short out through that caps, thus saving your rectifier.

This is a pretty basic explanation, and I may be wrong on a few points because I'm not a clever man.

Is your OBIT iron cored / mains frequency, or is it high frequency? (HF ones are quite small and lightweight because of the small ferrite transformer)

I hope thats some help.

Steve Conner

Sun Nov 11 2012, 07:58AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

The fundamental idea is to put a capacitor across the rectifier output. I've seen a few ohms of resistance put in series with the cap to avoid affecting the characteristics of the welding arc.

The capacitor "holds up" to the 10kV just fine. The high voltage is also high frequency, so the capacitor appears as a short circuit to it.

Ash Small

Sun Nov 11 2012, 10:25AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

What I did, and I didn't really know what I was doing at the time is this:

The capacitors are 'hand rolled' polythene and kitchen foil, bound in insulating tape.

The 'choke' is a huge toroidal transformer core, with around half a dozen windings on each leg, wound opposing each other.

the second cap, on the rectifier side of the choke, is probably not required, I just added it as a safety margin.

The HF effectively flows through the capacitor , and is blocked by the 'low pass' choke, thus preventing the HV getting back to the rectifier etc.

Others here can probably advise you regarding the best material for the choke core, and capacitance values. I achieved it by trial and error.

Rick

Sun Nov 11 2012, 07:33PM

Registered Member #4220 Joined: Sun Nov 20 2011, 12:11AM
Location:
Posts: 6

Thank you all for your replies;

I think my OBIT is mains freq, but as I understand it, the spark gaps create the hi freq.

I'm patterning my hv circuit basically after the Miller HF-15-1,

which is simple and only shows the cap across the rectifier outputs, no choke. The enclosed diagram shows a couple of resistors, but it looks to me like they'd only come into play in the AC mode. I can easily make a choke from a MOT core, that's how I made the smoothing choke for the rectifier, but no other designs show it. Wish I understood what I'm doing!

Ash Small

Sun Nov 11 2012, 07:52PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Thanks for posting the Miller schematic Rick, it's very interesting.

From what I understand, the MOT core should do the job, but there are other alternatives. Just make sure when you wind the two windings that you wind them so that the HF 'pulses' cancel each other out.

Rick

Mon Nov 12 2012, 06:02AM

Registered Member #4220 Joined: Sun Nov 20 2011, 12:11AM
Location:
Posts: 6

I wonder if it needs to be as big as a MOT? Smaller would be nice- I've built the rectifier, choke (made from another MOT core) and hv circuit with a pretty big OBIT into the top of a Lincoln tombstone welder and it's getting crowded in there and heavy! I have a 45mf cap across the rectifier; may be enough since I only energize the hv momentarily to strike the arc. Still, anything I can do to protect those 300a diodes is worth doing.

Today I fooled with it some more, assembled everything per the Miller schematic, using a glass cap per Newton Brawn's design, and a length of nichrome wire for a resistor (aviation safety wire is good for making small resistors- one to two ohms per foot). Believe it or not, it worked! Certainly surprised me. The only change I made to what I'd tried before was to remove the ferrite bead core from the coupling xmfr. I had made it that way per Newton Brawn's plans, but I noticed the pix of the Miller unit show an air core.

Ash Small

Mon Nov 12 2012, 09:14AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

45nF or 45mF?

Mine has an air core xformer, I thought they all did!

Mine has ten turns primary and secondary, and is wound around a 'cleaned' effervescent vitamin C tube, which are made from polypropylene, and apparently has no magnetic properties at all.

EDIT: I wouldn't go much smaller than an MOT core for the choke. It needs to be fairly substantial (I think)for the magnetic properties. Also, you need to wind your welding cable (or similar) around it several times, so you need a reasonably sized window. (The choke core is certainly the heaviest, if not the largest part of my setup (The 'hand rolled' caps take up quite a bit of space, too)).

Steve Conner

Mon Nov 12 2012, 10:58AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

I don't think it is worth bothering with a choke. The noise you're trying to choke is differential-mode, so the choke would have to stand the welding current without saturating.

A capacitor and/or TVS across each diode would probably be a better investment.

Ash Small

Mon Nov 12 2012, 11:31AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Steve Conner wrote ...

I don't think it is worth bothering with a choke. The noise you're trying to choke is differential-mode, so the choke would have to stand the welding current without saturating.

A capacitor and/or TVS across each diode would probably be a better investment.

Steve, as I understand it the choke doesn't saturate. The choke is wound so that the fields oppose each other, and cancel out.

I'm not sure how you EE's would describe it

(EDIT: It just needs to be chunky enough to oppose sudden changes in current)

1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob