ZCS resonant driver idea for (HV) ferrite power transformers / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

ZCS resonant driver idea for (HV) ferrite power transformers

1 2 3 4 next

Move Thread

LAN_403

Dr. Dark Current

Mon Oct 29 2007, 09:25PM

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

Hi, I had this idea of a resonant driver for driving high power ferrite transformers, with the main objective of decreasing wasted power in the switching transistors for the same power throughput.

I thought of putting a capacitor in series with the primary winding to form a resonant LC circuit, which would be excited at resonance so the transistors would switch at zero current, minimizing switching losses. This would be done by taking a feedback from the LC tank circuit by using a current transformer.

Here is a block diagram of my idea:

Firstly, will this even work?

And secondly, will there be any advantages of using such approach as opposed to normal driver without the resonant cap?

-----
On somewhat related topic: Lets say you have 4 Mosfets and you can connect them either in full bridge or half bridge (with 2 parallel fets per 'leg') configuration. If you do the math you'll see that in both case the total dissipation is the same. So I wonder why do people use full-brigdes with FETs if it's easier to make a half-bridge of them?

Marko

Mon Oct 29 2007, 09:30PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Firstly, will this even work?

No, because without primary inductance bridge will be operating into a nearly dead short and blow up.

Just think of what DRSSTC's do.

Dr. Dark Current

Mon Oct 29 2007, 09:40PM

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

Marko wrote ...

Firstly, will this even work?

No, because without primary inductance bridge will be operating into a nearly dead short and blow up.

Just think of what DRSSTC's do.

OK.

For some reason I thought it will somehow self-current-limit itself, but probably I was wrong.

So is there any other way to do what I want (transistors switching at zero current driving a transformer) other than the "mazilli" circuit (for which, at mains voltage, you need very high voltage transistors) ?

Electroholic

Mon Oct 29 2007, 11:06PM

Registered Member #191 Joined: Fri Feb 17 2006, 02:01AM
Location: Esbjerg Denmark
Posts: 720

Why not look at SLR?
check Finn's CCPS thread

Steve Conner

Tue Oct 30 2007, 09:42AM

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

Yes, this is very closely related to the SLR converter. The SLR is just your idea, but operating at half the resonant frequency to keep the resonant rise under control.

You may also be interested in current-fed inverters, which can drive a short-circuit all day. They use a DC bus choke instead of a DC bus capacitor.

Dr. Dark Current

Tue Oct 30 2007, 02:39PM

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

Steve Conner wrote ...

Yes, this is very closely related to the SLR converter. The SLR is just your idea, but operating at half the resonant frequency to keep the resonant rise under control.

You may also be interested in current-fed inverters, which can drive a short-circuit all day. They use a DC bus choke instead of a DC bus capacitor.

As to the SLR converter, can it be realised with current transformer feedback, or do I have to run it fixed-frequency? (I also need it to be able to run into open load, on which I read that SLR converters don't like too much right?)

Marko

Tue Oct 30 2007, 06:15PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

I thought about jan's second question... really I don't see either any special benefit of a fullbridge apart from little less transformer copper loss.

So, is it in most cases just more practical to use a halfbridge of bigger mosfets than a fullbridge?

SLR is good for cap charging, where it's always loaded with low impedance, but I don't thing it's good to load it with unstable load like an arc. You might have seen arcs from big CCPS's, but it's usually just a fraction of their output power and still appears like relatively low impedance to the supply.
High resistive might lead the supply into hard switching which would be especially bad with high peak currents involved.

You may also be interested in current-fed inverters, which can drive a short-circuit all day. They use a DC bus choke instead of a DC bus capacitor.

Steve, is it possible to build a current-fed ZVS bridge inverter like the circuit used in CFL's?

Would then switches stand only pi/2 DC bus voltage as I thought?

Dr. Dark Current

Tue Oct 30 2007, 07:24PM

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

Well what I really want to do is to get the biggest and most scary arcs from four or even more flybacks (I had 10 identical flyback secondaries now I still have 6 left, just find the cores).

After my IGBT flyback driver exploded at fraction of the power it should have handled, I realised there could be something going on in the driver which may not be as simple as I thought, so instead the idea was to drive the bridge "more properly" with more predictable results.

Steve Conner

Tue Oct 30 2007, 07:57PM

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

Yes, it is possible to build a current-fed ZVS fullbridge. I don't know the equation for the switch off-state voltage, but pi/2 seems about right

Marko

Thu Nov 01 2007, 04:50PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Hi...

It's clear to me as day that guys like Steve don't have time for my noob ramblings - but please, don't get pissed on me if I write some kind of nonsense and point it out so it isn't spread further - OK?

After my IGBT flyback driver exploded at fraction of the power it should have handled, I realized there could be something going on in the driver which may not be as simple as I thought, so instead the idea was to drive the bridge "more properly" with more predictable results.

Jan, have you tried measuring your primary current?

If you did, you may have been surprised. You may have noticed that if you keep the output of your transformer shorted, it doesn't generate much heat, and your DC power consumption is also very low at that point, comparable to open transformer. And you probably also noticed that your power consumption is biggest of all when arc is most stretched out just before it extinguishes.

But what do you think is happening to your primary circuit?

Your transformer can be looked at like an ideal transformer with series inductance, which is your leakage or ballast inductance and it is essential for operation of the circuit. (Otherwise shorted output would quickly blow it up).

When your transformer is shorted it effectively disappears leaving the inverter driving solely the leakage inductance. That is a hazardous condition because due to this low inductance you have massive reactive current circulating through your primary circuit, dissipating lots of power in your devices and wiring while most of energy is actually returned to DC bus cap with each cycle.

It's nothing surprising there is you find the current in your switches to be tens of times larger than input DC current. Combined with hard switching this is very bad situation.

Arc isn't really a good load either - it's impedance is still quite low compared to output impedance of the inverter and it still creates very poor power factor.
It would be most economic (from switch current rating view) if you matched Var of your primary ballast inductance to your expected power output, keeping cos fi about 50 but that wouldn't really work well with crazy load like arcs.

I think that was the same reason why Steve W.'s hard switched HV inverter (before SLR CCPS) failed despite being quite robust design.
And I also believe this is one of reasons why he flames people when they compute power of their inverters by multiplying current with voltage rating of IGBT's!

SLR topology is interesting because it eliminates turn-off loss completely - IGBT's still conduct high peak currents, but that isn't really that bad anymore at all!
IGBT's also have entire half cycle to turn off so even slower devices can work quite efficiently.

Problem is that SLR doesn't seem to like large arc loadings too much either - it's mainly for stable low impedance loads (charging capacitors).

Current-fed ZVS topologies from the other side have completely different approach - what they actually do is putting a ''power factor correction'' cap in between the inverter and transformer - so all reactive current is constrained in this LC circuit and inverter, if kept at resonance, always feels it as a resistive load.

We must also note that parallel LC represents low impedance to anything far from it's f0 so it requires supply from a current source, which is imitated by a DC link choke.

For bridges, the choke is actually split onto source of each mosfet and wound in same direction so fluxes *don't* cancel out during shoot through.

I really don't know anything more about this but it would be interesting thing to try out.

Current-fed circuits are, as I think, best of all for driving widely variable loads like arcs.

Marko

1 2 3 4 next

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