Gate Driver Problem / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Gate Driver Problem

first previous 6 7 8 9 next last

Move Thread

LAN_403

Graham Armitage

Tue Jan 29 2013, 06:02PM

Registered Member #6038 Joined: Mon Aug 06 2012, 11:31AM
Location: Salado, TX
Posts: 248

Here we go again... So my IGBTs I ordered arrived yesterday and I etched a brand new board to limit parasitic induction. I checked all components were working before soldering them. This finished product looked like this - with IGBT attached and back side without IGBT.

I tested everything without the IGBTs in place and the output the gates looked good, except for some overshoot spikes

When I attach the IGBTs and turn everything on it's fine until I turn on the modulator and the UCC driver popped almost instantly and let the smoke out. I am at a complete loss now. What is it about the IGBT in place that seems to overload the drivers so badly? i though the diodes should prevent the voltage overshoot, could that be a problem? Is it back to the GDT ?

teravolt

Tue Jan 29 2013, 06:11PM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

I don't think that those little drivers are going to be able to handle the capacitance that you are driving. you will need a driver that comes in a TO-220 package

Graham Armitage

Tue Jan 29 2013, 06:16PM

Registered Member #6038 Joined: Mon Aug 06 2012, 11:31AM
Location: Salado, TX
Posts: 248

This is the MiniBrute design by EVR - I believe dozens if not hundreds of these coils have been built. Not sure I want to start second guessing the circuit at this stage. Happy to acknowledge and error on my part though.

If it was a case of not being able to handle the current I would expect the driver to heat up - but it pops (audibly) as soon as the modulator is turned on - like there is a short circuit somewhere. Does not make any sense to me.

teravolt

Tue Jan 29 2013, 06:22PM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

are you using the minibrute driver board

Graham Armitage

Tue Jan 29 2013, 06:25PM

Registered Member #6038 Joined: Mon Aug 06 2012, 11:31AM
Location: Salado, TX
Posts: 248

I am. I etched my own, but it is the same circuit as provided in the design. Completed all the testing including the display board, successfully. Everything works right up until the IGBT is added. So assuming the half bridge is where the problem lies.

teravolt

Tue Jan 29 2013, 08:36PM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

assuming there there is no wiring problems it is a engineering error. what is the zenar value? is it clamping at all? add some resistance in series with driver chip. change your coupling cap to 1uf.

Graham Armitage

Tue Jan 29 2013, 08:54PM

Registered Member #6038 Joined: Mon Aug 06 2012, 11:31AM
Location: Salado, TX
Posts: 248

The Zener's are IN4752 33v. I tested them and they clamp at 33.5v I already have a 1uF bipolar coupling cap. Is it even possible that the problem could exist on the driver board and not on the H-bridge, even though there appears to be a direct correlation?

teravolt

Tue Jan 29 2013, 10:25PM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

I agree, it should actually take some time to heat up and die I would think. Have you checked out under the socket or that part of the board with a fine toothed comb? I'm running out of answers. Maybe there going into thermal runaway and cant handle the load. the power dissipation is only a couple of watts for those devices.

Physics Junkie

Tue Jan 29 2013, 10:34PM

Registered Member #7267 Joined: Tue Oct 16 2012, 12:16AM
Location: Detroit, Michigan
Posts: 407

Bushman wrote ...

The Zener's are IN4752 33v. I tested them and they clamp at 33.5v I already have a 1uF bipolar coupling cap. Is it even possible that the problem could exist on the driver board and not on the H-bridge, even though there appears to be a direct correlation?

I too etched the same H Bridge board for testing. And again, SAME issue we are both having lol... Well now I can safely attach the GDT to both the Driver board and H-Bridge, It's only when I turn on either the modulator OR the signal generator that this problem persists. It just HAS to be an issue somewhere on the driver board.
Bushman, without the GDT's attached, modulator on, and signal generator off, what do you get on the output of the drivers? In your case, since you etched a board design thats known to work, I doubt at all that your problem has to do with the board design/component layout. It's starting to sound like you may need to redo the board, adding a pull down resistor on the modulation signal at the input of the gate enable. Also methinks you might have a diode or capacitor installed backwards somewhere around the gate drive. I Know you've checked your components like a thousand times and I trust your call, I'm just at a loss here too...
My design layout is my own, so my problem most likely is there. However my UCC aren't popping but as you know the regulator is over heating which leads me to believe that if parasitic inductance is not an issue, then I do not have enough energy available to my UCC's and they are giving my poor regulator a hastle by trying to draw the power it needs. I'm wondering if an extra electrolytic cap before the UCC's might help, something like 1000uf or less. I've also redone my board and will happily post the file for anyone to have a look over.

Graham Armitage

Tue Jan 29 2013, 11:40PM

Registered Member #6038 Joined: Mon Aug 06 2012, 11:31AM
Location: Salado, TX
Posts: 248

Yup - I agree, it's sounding more and more like a problem on the driver board. I had to design the layout myself as that does not come with the design instructions on the MiniBrute, so very likely I made a mistake there and it only manifests when there is a load on the driver IC. It's a complicated board so going to take a while to redo it, but running out of other options. I may try Teravolt's suggestion of the resistor on the driver output that may slow the overloading/heating enough for me to try and debug it. I guess this is how we learn

first previous 6 7 8 9 next last

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