Should I worry about shoothrough? / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Should I worry about shoothrough?

1 2 3 4 next

Move Thread

LAN_403

Ash Small

Sun Apr 24 2011, 12:26PM

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

I'm designing a full bridge HV supply and I'm wondering if I can get away with using one gate drive transformer with four secondaries appropriately wired.

operating frequency will be 25kHz, assuming the GDT produces a square wave output (which it won't, it will take some time for the output to change polarity I assume), all four IGBT's could be 'on' for around 250 nS.

The datasheet says they can withstand 'short circuit' for 10 uS, does this mean I don't need to worry about shoothrough, or should I use PWM to drive two GDT's, each with two secondaries instead?

(EDIT: I plan to use a push-pull transformer as the gate drive, if this makes any difference)

Adam Munich

Sun Apr 24 2011, 06:49PM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

From my experience making half bridges, gate drive ICs seem to make the mosfets run cooler, mainly because they output a much cleaner square wave than a GDT could ever hope to.

I like the FAN7392 myself, it is very easy to use.

Dr. ISOTOP

Sun Apr 24 2011, 06:52PM

Registered Member #2919 Joined: Fri Jun 11 2010, 06:30PM
Location: Cambridge, MA
Posts: 652

I use a single GDT on my full bridge, driven by two gate drivers. I've run it at 1 KW and have yet to see IGBT failures (except from human error, of course).

Ash Small

Sun Apr 24 2011, 07:03PM

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

That's an interesting point, Grenadier. I understand IGBT's run cooler than MOSFET's so GDT's should pose less of a problem?

Also, it's that very point about non-square wave output from a GDT, shoothrough should be less?

This will be my first 'proper' transformer, ie not a flyback, so I want to keep things simple. I also want to avoid IC's for now.

I'm planning to 'scope it, running with limited current/voltage when I test it, so that I can see exactly what is happening, but if people say shoothrough WILL be a problem, I'll design in some adjustable PWM so I can control it (but that means four more transistors with heatsinks, and two GDT's instead of one.

Patrick

Sun Apr 24 2011, 08:17PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

As for shoot through, increase your dead time to max, then decrease till you see the current spike on your scope, theres all kinds of stratagies and tricks to cope with this. SSTCoilers are all over the H-bridge and its capabilities / liabilities. The IC's are nice because they have guarenteed outputs, the GDT's are variable on drive from op-amps, magnetic material, winding, and other factors.

Grenadier wrote ...

From my experience making half bridges, gate drive ICs seem to make the mosfets run cooler, mainly because they output a much cleaner square wave than a GDT could ever hope to.

I like the FAN7392 myself, it is very easy to use.

I support these comments from Grenadier whole heartedly. I use the "FAN" series too, from Fairchild.

Ash Small wrote ...

... I understand IGBT's run cooler than MOSFET's...

This is not as simple as you have stated. It depends, MOSFETs are better at very low voltage and very high current, and faster in the transistions. IGBT's are better at the 400V and up region where the current is less. There is a region where their almost the same, as at approximatley 200-400V. There are some exceptions, and you must calculate your specific needs your purpose, and the IGBTs/MOSFETs you can get and what not.

I think using IC's does keep it simple... but go for your thing then see for yourself.

Ash Small

Sun Apr 24 2011, 09:02PM

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

Thanks for your input Patrick.

I'll bear the FAN series in mind for the future, but initially I'm planning on using two transistors in an astable multivibrator to drive a push-pull centre-tapped GDT. If I need PWM I'm planning to drive a couple of mono-stable multivibrators from the astable, with pots or variable caps, but if shoothrough isn't going to be a problem I won't have to use the extra four transistors, etc.

I'm eventually planning to run it on rectified 240, around 340 V,

I'd like to avoid IC's to start with, especially if I can get the gate driver to work with just two transistors.

I'm just wondering if shoothrough is likely to be a problem?

(The IGBT's I'm planning to use are these:)

Adam Munich

Mon Apr 25 2011, 12:08AM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

To answer your question, yes shoothrough is likey going to be an issue, mainly because I'm not so sure that such a circuit is even capable of making a clean square wave. If you are talking about the 2 transistor astable, that doesn't output anything close to a square at all.

Why all the hate on IC's?

Patrick

Mon Apr 25 2011, 03:38AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Ash Small wrote ...

Thanks for your input Patrick.

I'll bear the FAN series in mind for the future, but initially I'm planning on using two transistors in an astable multivibrator to drive a push-pull centre-tapped GDT. If I need PWM I'm planning to drive a couple of mono-stable multivibrators from the astable, with pots or variable caps, but if shoothrough isn't going to be a problem I won't have to use the extra four transistors, etc.

I think you need to use amps to drive the GDT, waht freq are you using?

if you use the amp ic's to make the GDT have good rise times, then why not skip all that and use a GD IC?

you should look here:

example of the driver you want, notice the mosfet amp, i think this is the best design ive seen. some people use op amps, dont know how good that is.

other schematics, related.

James

Mon Apr 25 2011, 04:22AM

Registered Member #3610 Joined: Thu Jan 13 2011, 03:29AM
Location: Seattle, WA
Posts: 506

I had a lot of trouble with gate drive ICs. They seem to be quite delicate and most of them are expensive. The worst thing about it is if a single mosfet fails, all of them will fail, the IC will blow, and often all the rest of the semiconductors in your circuit will also be destroyed by the cascading failure.

At first I had trouble with GDTs, but having studied this

and found some nice cores I'm getting excellent results. Nice clean square wave, isolation that protects the rest of the circuit in the event of a mosfet failure, and I can use cheap low voltage gate drive ICs to drive the transformer. My SSTC uses a GDT I wound and works great, I've tried it from around 250kHz to 650kHz with no problems.

I can't see myself ever bothering with high side drive ICs again unless I absolutely need a very wide duty cycle range.

Dr. ISOTOP

Mon Apr 25 2011, 05:33AM

Registered Member #2919 Joined: Fri Jun 11 2010, 06:30PM
Location: Cambridge, MA
Posts: 652

It's a sad fact of life that MOSFET drivers hate shorts

Also, if you wind a GDT, be careful with your choice of core; you should scope its output before using it.
The vast majority of the cores found in ATX power supplies do not work.

1 2 3 4 next

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