High Voltage "Arc Starter" / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

High Voltage "Arc Starter"

1 2 next

Move Thread

LAN_403

Billybobjoe

Mon Jan 07 2008, 09:26PM

Registered Member #396 Joined: Wed Apr 19 2006, 12:55AM
Location: Pittsburgh, PA
Posts: 176

I've recently discovered an interesting concept used to initiate high current welding arcs using high frequency/voltage superimposed on them.. I've found several websites dealing with this-

- check out the "more pictures" link too.

- Pupman forum discussion

- another schematic

The concept is simple enough but there are some things I'm not clear on.

Lets look at this schematic from the first link:

I suppose C1 is for blocking EMI and the potentiometer for adjusting frequency, otherwise the left part of the circuit is very SGTC-like.

I understand the resistor and capacitor array on the right is to pass the high frequency HV, but for DC operation only, why are both a resistor(s) and a capacitor necessary?

Also, the label "FROM POWER SOURCE" on the schematic seems to indicate that both a full wave rectifier and large electrolytic capacitors for ripple filtering would not be affected by the many thousands of khz pulsing throughout the system. IS it safe to assume that the 10 uf cap. and resistors will keep that from happening?

And would the 10 uf cap. need to be able to stand high frequency even though it is practically "invisible" to a high frequency current?

Thanks guys - it just seems like this is overcomplicated in places and not complete in others.

Steve Ward

Tue Jan 08 2008, 12:18AM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

I suppose C1 is for blocking EMI and the potentiometer for adjusting frequency, otherwise the left part of the circuit is very SGTC-like.

I dont understand why its a pot (instead of normal resistor), but it would mainly serve to dampen the system (slightly lowering the frequency, but not by much) perhaps to protect the capacitor in the event of the system firing but the output electrodes are not able to strike an arc (in which case the system has pretty high Q).

I understand the resistor and capacitor array on the right is to pass the high frequency HV, but for DC operation only, why are both a resistor(s) and a capacitor necessary?

Well, i would have not used the 10 ohm resistor at all, and instead just a small 1uF or so capacitor across the supply rails. The 10 ohms might be to reduce the current of really high frequencies, but i cant see its use, and would have just shorted it out myself.

IS it safe to assume that the 10 uf cap. and resistors will keep that from happening?

Yes, the RC basically looks like 10 ohms to high frequencies (id have made it look like even less impedance) and this will effectively short out the high frequencies across those points because the current is fairly low.

And would the 10 uf cap. need to be able to stand high frequency even though it is practically "invisible" to a high frequency current?

It shouldnt be much current from the RF generator, but it should probably be able to handle several amps peak current. I would use a polypropylene type cap to be really safe.

Thanks guys - it just seems like this is overcomplicated in places and not complete in others.

I agree. I would have done things slightly different, but i already talked about that.

Billybobjoe

Wed Jan 09 2008, 08:24PM

Registered Member #396 Joined: Wed Apr 19 2006, 12:55AM
Location: Pittsburgh, PA
Posts: 176

Thanks! A few more questions though - You recommend a polypropylene cap to pass the HF, but what about the several thousand uf of electrolytics for the ripple filtering? I'd like to make this as stable as possible, and I don't think there are polypro. caps that large.

Now that I think about it the 10uf cap in the schematic isn't necessary at all if one was going to have filtering caps before it - right?

And you would recommend keeping that resistor on the spark gap side?

Ooh, and here's and interesting one: On the Pupman forums they talked about why multiple gaps are used in series but I didn't see a conclusive argument - any ideas? Quenching? Or doesn't that apply here?

Steve Ward

Wed Jan 09 2008, 09:46PM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

Thanks! A few more questions though - You recommend a polypropylene cap to pass the HF, but what about the several thousand uf of electrolytics for the ripple filtering? I'd like to make this as stable as possible, and I don't think there are polypro. caps that large.

Calculate the impedance of 10uF at a few hundred khz (or whatever the RF is). Its very small, so the law of diminishing returns begins to take over quickly! Also, a PP cap will have lower ESR and ESL (equivalent series resistance/inductance) making its true impedance close to that of just a capacitor. Lytics would probably be OK here, but a HF rated cap is better, even if its value is magnitudes lower.

Now that I think about it the 10uf cap in the schematic isn't necessary at all if one was going to have filtering caps before it - right?

Im not sure how welders are physically assembled, but i would worry about the lead inductance going back to the power supply. If you add several feet of wire, its significant at RF. The small 10uF cap can be mounted right near the ignition point, keeping the loop for the RF to pass through very small. Any extra inductance in the RF path is just wasting voltage, whether its significant in practice is something i cant really answer.

And you would recommend keeping that resistor on the spark gap side?

Depends on how robust your components are. The resistor might just save them from some extra abuse. I dont have enough direct experience to say whether or not the resistor is absolutely needed, but id probably have about 1 ohm there to be sure at first.

Ooh, and here's and interesting one: On the Pupman forums they talked about why multiple gaps are used in series but I didn't see a conclusive argument - any ideas? Quenching? Or doesn't that apply here?

Typically it helps to quench the spark by spreading the total heat over many smaller areas. A single arc containing all of the energy is much more difficult to extinguish, though it actually has a lower voltage drop... so a properly quenching single gap tends to perform better than many little gaps. It surely does apply here, but how much difference it will make is left as a experiment for the reader

. Just try some different spark gaps and see which one gives the best output. Also try changing that primary resistance (or getting rid of it) and note the effects (should improve output somewhat).

Steve Conner

Thu Jan 10 2008, 10:06AM

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

OK, here are my guesses at the few points that weren't covered:

The large variable resistor is to adjust the strength of the spark, in case it's too strong and starts giving the operator nasty jolts. Also because you want the weakest spark that'll do the job, for the sake of reducing the stress on the other components and making the spark gap electrodes last longer.

The 10 ohm series resistor for AC operation is so that welders with square wave output won't blow the 10uF capacitor out with harmonic currents. (the 10uF cap is indeed to stop RF feeding back into the welder)

Apart from these considerations, the arc starter unit is identical to a Tesla coil. I always wonder when someone will make an OLTC, DRSSTC or SISG arc starter, or even if a commercial welding company has done it already.

This is David Barrett's DIY AC/DC TIG welder

andthis is the schematic of his arc starter unit

Oh, and BTW, welders don't have smoothing capacitors on their outputs! A large capacitor bank won't maintain a stable arc. It would just explode the welding electrode. If they use anything for smoothing, it'll be a choke, to smooth the output current, not the voltage.

Billybobjoe

Fri Jan 11 2008, 02:04AM

Registered Member #396 Joined: Wed Apr 19 2006, 12:55AM
Location: Pittsburgh, PA
Posts: 176

Steve Conner wrote ...

This is David Barrett's DIY AC/DC TIG welder

andthis is the schematic of his arc starter unit

Thanks I don't know how I missed that site - that is impressive!

Steve Conner wrote ...

Oh, and BTW, welders don't have smoothing capacitors on their outputs! A large capacitor bank won't maintain a stable arc. It would just explode the welding electrode. If they use anything for smoothing, it'll be a choke, to smooth the output current, not the voltage.

Wow - that clears some things up - its one of those things that just hits you in the face

This isn't necessarily for welding but the initial arc gap firing would have been frightening with a massive cap. bank wouldn't it?

What size chokes are we talking about here?

Steve Conner

Fri Jan 11 2008, 10:34AM

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

Big f****n chokes! :D

I'm looking into building a TIG welder, and from my research so far, an inverter welder has an output smoothing choke the size of your fist, while an old-fashioned 60Hz unit has an iron-cored lump the size of a paint can that weighs about 40lbs. (A proper old-fashioned welder that is, that needs a crane to move it, not the $200 junk ones.)

Billybobjoe

Fri Jan 11 2008, 08:35PM

Registered Member #396 Joined: Wed Apr 19 2006, 12:55AM
Location: Pittsburgh, PA
Posts: 176

Steve Conner wrote ...

Big f****n chokes! :D

Not what I was hoping - my welding rod style ballast inductor was a PITA so I can't even imagine those chokes. I'm not sure where I'm going with this though so I'm not too worried about regulation.

Steve Conner wrote ...

I'm looking into building a TIG welder, and from my research so far, an inverter welder has an output smoothing choke the size of your fist, while an old-fashioned 60Hz unit has an iron-cored lump the size of a paint can that weighs about 40lbs. (A proper old-fashioned welder that is, that needs a crane to move it, not the $200 junk ones.)

What do you mean by proper old-fashioned? Just a big ass transformer with 100% duty cycle?

Steve Conner

Sat Jan 12 2008, 02:29PM

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

Well, this is what I mean by big old-fashioned:

At the place I used to work, they had a BOC TIG welder the size of a washing machine that ran off 400V two-phase power. It could do AC or DC, TIG or stick welding, up to 300A, and there was a built-in water cooling radiator and pump for the torch.

I saw the schematics and the guts while helping one of the techs fix it one day. Power control was by saturable reactor (another 40lb lump of iron) and the main transformer was the size of a microwave oven. (Not a MOT: a whole oven.) There was a small Tesla coil inside for the HF arc starter. When it was set to DC mode, the smoothing choke was brought into the circuit, making the arc completely silent, even though it only had two-phase input.

The unit had a lifting eye on top for a crane, and needed it.

Wavetuner

Wed May 28 2008, 04:18AM

Registered Member #1500 Joined: Sat May 24 2008, 04:38PM
Location: Ojai, Ca.
Posts: 44

Say, could a circuit like this be used to break down a energy capacitor discharge slightly wide set gap?

1 2 next

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