Help with cleaning up gate drive to IGBT buck converter / General Science and Electronics / Forums

Forums

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

« Previous topic | Next topic »

Help with cleaning up gate drive to IGBT buck converter

first previous 3 4 5 6 next

Move Thread

LAN_403

Patrick

Wed Dec 01 2010, 06:48AM

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

wow a schottky diode, hmmm thats interesting.... i was quite aways away from suggesting different diodes, well show me a full scematic and where this diode is and ill advise you.

all i can say is find a diode that works. but that doesnt help much.

Steve Conner

Wed Dec 01 2010, 10:52AM

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

Iamsmooth, that's interesting because it suggests the oscillations are caused by the diode's reverse recovery spike. Schottkies have no recovery spikes, but they don't come in high voltage ratings. That RHRG... diode is pretty decent as these things go. So you just have to live with the oscillations. The circuit will probably work OK.

You will never see perfectly clean waveforms in any SMPS circuit, there are always little glitches and oscillations like this. The trick is telling the difference between the harmless ones that just make you fail your EMC test, and the ones that mean it's going to explode as soon as you crank the variac further. I won't go into this, as it's hard to explain without practical experience, not to mention too much like my day job.

Sometimes a ferrite bead on one of the diode legs helps. Slowing the turn-on of the IGBT as much as possible helps too. And, using a diode no bigger than necessary, or a more modern one (look for devices designed for boost PFC, this is a highly competitive market)

Or maybe you can get hold of a 600 or 1200V silicon carbide Schottky diode from Cree or whoever. Those are really nice

IamSmooth

Wed Dec 01 2010, 01:07PM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

Steve McConner wrote ...

one (look for devices designed for boost PFC, this is a highly competitive market)

Or maybe you can get hold of a 600 or 1200V silicon carbide Schottky diode from Cree or whoever. Those are really nice

I was actually reading about the Cree diodes last night before I went to bed. No one was open at 1am. I was going to do what you alread suggested now that I seem to have localized the problem. I was also going to try ferrite beads on one of the legs as you had suggested. I might try it on the line from the emitter to the return of the fod3180 and see if this helps.

Daedronus

Wed Dec 01 2010, 02:07PM

Registered Member #2329 Joined: Tue Sept 01 2009, 08:25AM
Location:
Posts: 370

Some diodes are designated as having soft recovery, even if they are not Schotty type, also, sometimes RC snubbers are used across diodes.

You can try a RC snubber across the fast non Schotty diode and see if it helps.

IamSmooth

Thu Dec 02 2010, 02:03AM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

So I wanted to see what happens with various maneuvers.

The ferrite bead did not help on the hyperfast diode lead.
The 30v schottky worked great, but I can only go to 30v. After 30v I can see the oscillations start to manifest. I spoke with CREE and I should be able to get 3 of the 20A/1200v diodes that I can run in parallel.

I tried a 0.01uf capacitor across the diode leads and this is what I got:

The pictures in order are gate signal, gate signal with 100v on CE, gate signal with 100v on CE with 0.01uf cap

Patrick

Thu Dec 02 2010, 05:14AM

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

i dont see why your having so much trouble, i really think its your board layout, you have a green power trace traveling across your gate/pin trace, i think your inducing the oscillations. thats why you see the oscillations at 100v and power, but when you idle back they leave.

Steve Conner

Thu Dec 02 2010, 09:45AM

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

IamSmooth wrote ...

I tried a 0.01uf capacitor across the diode leads and this is what I got:

Nooooo, don't do that, it'll only make the problem worse.

IamSmooth

Thu Dec 02 2010, 01:33PM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

Steve McConner wrote ...

IamSmooth wrote ...

I tried a 0.01uf capacitor across the diode leads and this is what I got:

Nooooo, don't do that, it'll only make the problem worse.

How is that? It seems to make it better on the scope tracing.

Patrick wrote ...

i really think its your board layout, you have a green power trace traveling across your gate/pin trace, i think your inducing the oscillations. thats why you see the oscillations at 100v and power, but when you idle back they leave.

The green trace is going across the Common-emitter, not the gate. The problem existed in my first board version without this trace for the snubber surface mount chips. The problem goes away completely when I change the hyperfast diode on the buck board with the schottky. The buck board is separate from the driver board.

Patrick

Thu Dec 02 2010, 02:43PM

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

what about the earth ground trace, between your flood fill and the narrow trace that winds up around past the FOD and back to the snubber? that is a loop waiting to pick up stray fields.... im hard pressed to thinkof any other reason why this circuit doesnt work....

IamSmooth

Thu Dec 02 2010, 04:23PM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

same problem with and without that plane. connected to ground or not connected makes no difference.

Unless someone has another idea, I'm pretty sure it is the diode on the buck. I'm getting a 65ns diode to replace the 85ns and see if that makes a difference.

Again, the circuit does work at the 120v input levels. However, I need to go higher and these oscillations seem like a problem waiting to happen; the oscillations get worse with higher voltages, and I have to get up to 240v. I could try replacing the diode and see if it is just not good.

first previous 3 4 5 6 next

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