Custom Mazzilli's "ZVS" driver with IRF540 / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Custom Mazzilli's "ZVS" driver with IRF540

1 2 next

Move Thread

LAN_403

climatex

Tue Oct 20 2009, 10:47AM

Registered Member #2012 Joined: Sat Mar 07 2009, 10:22AM
Location: Slovakia
Posts: 45

Hi,
I've decided to try the ZVS driver with IRF540's (100V, ~0.055R) and FR154 diodes (1.5A, 150-250ns, 400V) because the IRFP250's are pretty hard to obtain here. The input voltage should be 30V Â± 5V and any spikes should the FETs turn into heat by their zener diode (35*4=140V, somewhere above limit). The flyback I'll be using is a 18kV AC w/o rectifier. I also arranged Mazilli's scheme by myself;

.

What do you think? Will it work fine?

Regards
inflater

Myke

Tue Oct 20 2009, 03:37PM

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

I think that the MOSFETs will break down because if I remember correctly, the peak voltage across the FETs is around 2pi*Vin. There are some pretty cheap IRFP250s on ebay from the Thai shop. They worked well for my ZVS flyback driver.

EDIT: oop. If that's the equation then they should be fine then.

Dr. Dark Current

Tue Oct 20 2009, 03:49PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Myke wrote ...

I think that the MOSFETs will break down because if I remember correctly, the peak voltage across the FETs is around 2pi*Vin.

They say its pi*Vin (but I've come to the 2*sqrt(2) figure instead), so with 30V supply they slould be OK.

If the peaks are higher than the FET's breakdown, they will get hot but probably wont fail. However if the supply voltage is greater than half the FET's breakdown, they will fail for sure

MRacerxdl

Sat Oct 24 2009, 01:34AM

Registered Member #989 Joined: Sat Sept 08 2007, 02:15AM
Location: SÃ£o Paulo, Brazil
Posts: 476

Yep, its ok, I did it once with IRF540 and a 24V Supply, and as I remeber I used 1n4007 (normal retifier diodes)

Look:

Its not a good design and I made it fast, so its crappy thing. But it works =)

Good look to your work =)

climatex

Sat Oct 24 2009, 02:24PM

Registered Member #2012 Joined: Sat Mar 07 2009, 10:22AM
Location: Slovakia
Posts: 45

Well, I'm waiting for the 24V transformer to come by mail, so I tried to power it up with a PC PSU (at +12V it's rated 15A max), directly + and - to the filter capacitor. The result? Bridge rectifier starts to get VERY hot along with the first MOSFET (the second is barely warm). Also, if I use a thin wire for +12V from the PSU, it starts to glow red and its insulation starts to smoke after ~2 seconds, so I stopped it. No oscillations, the HV transformer just makes an audible click when the power is applied (and shut down), like if DC was applied to it. I've tried to insert a small neon bulb to the output, it blinks when the power is applied or cut off.

I don't know what's wrong with it. If we take in order the fact that I could make some errors in the printed circuit board scheme, what could be wrong with it?
I believe the voltage's too low, I'll try the transformer when it arrives to me. But I think that won't fix the problem, as it should work with 12V. I can upload some photos of the PCB either today's evening or tomorrow, but it's a bit meesy...

Thanks, your responses or hints are highly appreciated...

climatex

Sat Oct 24 2009, 05:03PM

Registered Member #2012 Joined: Sat Mar 07 2009, 10:22AM
Location: Slovakia
Posts: 45

I found the error along with dr. Kilovolt's help. I had D-S reversed on the second MOSFET, and also in my scheme

On 12V the arc starts at 2-3mm and can be drawn to max. 2.5cm.. We'll see how the IRF540's will work with the 24V transformer (slightly more than 30V on the output). The 540's have somewhat room temperature when it's running, and the heatsink is at room temperature, too. Yay.

climatex

Mon Oct 26 2009, 06:56PM

Registered Member #2012 Joined: Sat Mar 07 2009, 10:22AM
Location: Slovakia
Posts: 45

Here's the first version without any special tuning whatsoever: 330n cap MKT, IRF540's, input voltage 30-35V. The arc ignites at 0,5cm and can be stretched to max. 4,5cm.. Current draw is 1.9A standby, 4-8A when arcing, 10-11A peak at the max. arc length (4,5cm only).

It's pretty weak, I think I should give it min. 550n resonant capacity.. I don't know if that would help, but I'll try it.

Comments, hints welcome.
Inflater

climatex

Tue Oct 27 2009, 06:27PM

Registered Member #2012 Joined: Sat Mar 07 2009, 10:22AM
Location: Slovakia
Posts: 45

And here's the 550n cap, along with the more stable inductor.

The arc is more yellow and brighter than before, but I can't draw it too far, max. 4cm, I dunno why this happens; maybe 1000M filter cap is not enough; I've also tried a DST - the arc starts at @1cm (0,5cm more than the ac flyback).
Anyways tomorrow I'll try 8 windings (4+4) on the primary, it should increase the output voltage. (I hope).

Anyways thanks for keeping this discussion active. I mean, the last four posts are made by me.. gee. How interesting topic it is. </sarcasm>

Woods

Mon Nov 09 2009, 03:02PM

Registered Member #2459 Joined: Thu Nov 05 2009, 11:17AM
Location:
Posts: 2

Hi all, I'm also attempting to construct a Mazzilli ZVS driver but i'm having some trouble. I have used the components recommended on the schematic, the only one im not sure about is the Zener diode from the gates to the -ve input rail.

Is this a Zener diode or a schottky rectifier diode? At the moment i'm using a 1n5349B 12V zener diode (and the circuit doesnt work...)

Also this is my first post, nice website and forum!

Dr. Dark Current

Mon Nov 09 2009, 03:52PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Yes it is a zener diode. make sure you didnt put them in the other way round, the anode must be connected to gnd, cathode to gate.
Also make sure you have the fast feedback diodes "crossed" e.g. anode to gate 2 cathode to drain 1 and vice versa.

1 2 next

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