IGBT driver board

[ChiefsBuggy]

Paul,

I ran across this little ditty... very interesting... I think. [Note to self, get a life]

Check out page 60 of http://www.sun-fly.com.cn/cn/UploadF...2153019195.pdf

Originally Posted by Hewlett Package: Designer Guide

"This circuit includes a bootstrap circuit for providing output power to the HCPL-316J gate drive optocouplers, thereby Figure 4. Bootstrap Circuit for Power Control System eliminating the need for isolated power supplies or dc-to-dc converters. It can be modified to suit other HP gate drive
optocouplers and current/voltage sensing isolation amplifiers."

I think the gimmick is the lower drive stage has to be initially enabled to bootstrap the upper stage. I'm still analyzing it on paper, but I like what they have so far. I sent a message to AVAGOTECH asking for something like this. Wouldn't you know, right after I hit send, Google re-found this.

[dcb]

You might want a transformer with a tap to simplify creating a -5v source, i.e. 1:1 + 1:0.3 if you want +15 and -5v outputs to hold the gate high and low with prejudice.

[ChiefsBuggy]

Originally Posted by dcb

...to hold the gate high and low with prejudice.

would that make it a bigot-igbt? Ok that was a stretch, but its making me laugh.

[MPaulHolmes]

Originally Posted by ChiefsBuggy

would that make it a bigot-igbt? Ok that was a stretch, but its making me laugh.

You are hilarious. hahaha.

Hey! I've got it all almost completely breadboarded (the isolated DC-DC). But now we are going for a picnic, since it's 71 degrees today!!! Tomorrow it will rain again. Adam had done some tests on the MIC4451 driving some igbts, 0v to 15v, and they had pretty long turn on/turn off times. So, the driver stage might need some beefcakeness. There are readily available IGBT drivers that are $10-$15 from Powerex, so that wouldn't be the end of the world I guess.

I read somewhere that the bootstrap approach only works for the lower power levels, but I have no idea why, because I have no idea what the bootstrap approach is. haha. I need to read your link!!!

I would have gotten more done last night, but I was playing "Angry Birds". Holy cow that's a fun game! These pigs keep stealing the birds eggs, so you have to launch the birds at the pigs and destroy them.

[jackbauer]

Guys whats the story with the desat detection on the vla igbt driver? I'm a bit concerned about having the collector (M-) connected to the board as its at traction voltage levels. As far as i know desat is for short circuit protection. Is that really a concern in an ev?

edit: regards the driver chip. I have used a UCC27....someting or other to drive 2 of the powerex igbts with no probs so far. +15v dc dc on the board.

[MPaulHolmes]

It seems to me that you could do a near equivalent of the desat detection in software. If the overcurrent was tripped, mark the time. If current doesn't get under control in a certain amount of time, open the main contactor, because a tripped overcurrent circuit without reducing current would imply a shorted igbt I suppose. It would be slower than the desat detection I think, but that may not matter.

[ChiefsBuggy]

I went through the math in order to drive a big honking device like a pm600cla060... What I came up with was --- at 15KHz PWM, and 23V of Vge swing (15v to -8v) and 750 nF of Cies, and a 5 ohm Rg (and in the background, 5000 nC of gate charge -- good guess for high pack voltage)... the IGBT driver needs to supply 5A peak current (instantaneous current at 30Khz) and 75ma average current.

What this doesn't take into account is the inductance of the wiring. This is very topologically different depending on the installation. If we had a PCB that mounted directly to the IGBT's, we could control it.

Also, I haven't calculated parallel IGBT's or MOSFETS... but it could be done.

Why did I do this? I'm trying to come up with a open-source DC-DC supply for beefy IGBT's. I am playing off of Paul's and Flan's [sic] domino about the transform based DC-DC supply.

[sawickm]

I ran across this simple solution for driving IGBT modules - http://www.mornsunamerica.com/pdf/dualigbt.pdf

-Mark

[ChiefsBuggy]

Paul, how did the oscillator driving the transformer do towards a.dc-dc converter?

[MPaulHolmes]

Good, but I forgot to drive both directions, so I was only getting half of the voltage on the output side. So I ordered new parts such as a dual driver and also an attiny for driving the signal with 50% duty. I'll test that this week. It should do a couple hundred milliamps without any trouble.