ZVS arcstarter w/ferrite core / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

ZVS arcstarter w/ferrite core

Move Thread

LAN_403

Billybobjoe

Wed Aug 06 2008, 02:23AM

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

If anyone remembers I started a thread dealing with superimposing a HF high voltage over a high current welding . . . current. See here

I wasn't impressed with the spark gap setup and wanted to go solid state. So I gathered some parts:

I casted 200 small ferrite rods (with polyester resin) into a core about 1 1/4" in diameter and 12' long. Around this is 26 turns of 4 AWG wire and 13 turns of 20 AWG wire inside vinyl aquarium tubing filled with mineral oil (mainly just to pull the wire through but offers some corona resistance).

This "coupling coil" as I'll call it it to be put in series with a welding current and driven with a high voltage high frequency source. So I constructed a ZVS "Mazzilli" flyback style driver for a ferrite transformer from a solid state microwave oven. The output was then meant to be fed into the primary of the "coupling coil". This was the plan, but I haven't managed to get any arcs or sparks out of the secondary (4 AWG wire).

I believe there are two main problems:

1) Not enough voltage from the SS MOT. I haven't found a definitive answer about how to obtain the maximum voltage from a ZVS driver besides raising the input voltage (right now I use 24v from two lead acids in series which gives maybe 1KV straight out of the MOT at a decent hot arc current). IIRC I read somewhere on the net that a larger inductor increases output voltage. You can see the inductor I use now in the pic. It is a ferrite core with ~30 turns on it (yes I know I need some large gauge magnet wire). None of the ferrite toroidal inductance calculators I've used seem to give any useful answer about the inductance of my particular inductor so I can't match it to the ZVS schematic. I also have never seen a definitive answer about what type of core to use. Yellow ATX, ferrite, air? I suppose it depends on the frequency which leads me to my next point.

2) Coupling Obviously coupling needs increased. This involves a few questions. I understand this can be accomplished by increasing the frequency. How exactly is this done with a ZVS driver (I would assume capacitance changes)? Also, does anyone think I'd get any appreciable change in coupling if I were to remove the secondary and half of the primary, then rewind the secondary (all of this on the coupling coil) so it is half the length of the core, but in two layers. Then wind the primary on the other half of the core. As of now the primary (white wire) is almost 1/2" from the core because of the thick 4 AWG wire underneath it.

Notes:

â€¢Tesladownunder, I noticed on your site you ran a SS MOT with a Mazzilli driver - what size capacitor were you using? I've tried between .68 to 4 uf with only slightly different results.

â€¢I'm looking for at least 2500 VAC on the output of the SS MOT and at least 4KV on the secondary of the coupling coil.

â€¢ I know it may seem like I haven't done much experimenting with cap. and inductor values but my resources are limited and before wasting any more money I'd like to get some input on how to maximize the voltage and frequency of the MOT from people who understand this circuit:

Myke

Wed Aug 06 2008, 05:35AM

Registered Member #540 Joined: Mon Feb 19 2007, 07:49PM
Location: MIT
Posts: 969

For increasing the output voltage you would decrease the number of turns and increase the capacitance. Thing is, When decreasing L and increasing C I would assume that the current it would draw would go up. To increase the freq you would decrease the C or the L (note that when I refer to L, I am talking about the primary of the transformer).

...

Wed Aug 06 2008, 08:02AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

I do not see how you plan to couple the output of the SS mot with like 4kv output into a 13 turn coil on a highly gapped ferrite core.... I calculate that you have about 50uH maybe a little more, but that would still give you a reactance (essentially AC resistance) of under 10 ohms at 100KHz. So if you want to put 4kv into that circuit you will need to put out hundreds of amps of current, which that driver cannot handle.

To make this work you will definiteley want to switch over to some kind of pulsed circuit, either pulsing on thye HV side (ie, sparkgap +capacitor) or the low voltage side (IGBT/SCR/etc+capacitor).

Steve Conner

Wed Aug 06 2008, 09:15AM

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

Yeah, I think Terry Fritz's SiSG topology would be a good bet for this, see the forum archives and the Pupman list.

Steve Ward

Wed Aug 06 2008, 07:02PM

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

Yep, you are using the wrong type of drive system for this. You need something that really whacks the spark coil, and the ZVS push-pull driver is more of a constant high power drive, which is NOT desirable. Either go with straight up capacitive discharge into the inductor (like a SGTC, but maybe using silicon to switch it), or you might want to look at the DRSSTC which can also drive the heck out of low impedance circuits, though its considerably more advanced and likely to break.

Really, the spark gap drive you had originally was probably the safest bet, maybe it just needed some extra tweaking.

quicksilver

Thu Aug 07 2008, 01:53PM

Registered Member #1408 Joined: Fri Mar 21 2008, 03:49PM
Location: Oracle, AZ
Posts: 679

The coil in a ssMOT is not the ideal as you now know. What TDU did was begin with the core & bobbin from that & essentially make his own LOPT. I'm fairly sure it wasn't simple. The Mazzilli would be fantastic on a very beefy, simple AC flyback....remember the notation dealing with current? VERY few LOPT can deal with such a thing as a Mazilli driven at full tilt. (I found out fast!).

I'm sure he will reply but I believe that TDU pulled the "Lutz-wire" type secondary apart (on the SS-MOT) and wound his own after a great deal of pondering. If you check his page, he made a secondary from VERY thin wire.
The great thing about the SS MOTs are the parts or the collective power of the inverter in use for something entirely different than drawing arcs. In fact now there are about 50% SS MWO due to the cost savings - but if those parts were to be bought separately, via HV part suppliers, they may cost much more than the copper in a MOT.....but you're still only getting about 2.5Kv out of both.

Billybobjoe

Thu Aug 07 2008, 03:34PM

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

Well you learn something every day. I always thought coupling became easier at high frequencies but not quite the reasoning. So as the frequency increases, so does the reactance of the the primary, so it makes it more practical (ie w/ less turns) to reach the desired impedance? And a pulsed setup allows one to reach high pulse currents which satisfy the demands of a primary with a low inductance?

I'll guess I'll go with a SGTC design. On tis site

"The power supply for the L/C/Gap in the Cyber-TIG stabilizer looked *strongly* like a TV flyback." and ". . . inside the primary was a stick of 4-5 ferrite rods - wrapped in fiberglass tape". So my core is still useable? I'd like to use a small NST or maybe OBIT for the HV transformer and would like as much coupling as possible. What kind of frequencies would be necessary to still use the core?

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