40n60's / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

40n60's

1 2 next

Move Thread

LAN_403

Desmogod

Fri Mar 03 2006, 12:30AM

Registered Member #139 Joined: Sat Feb 11 2006, 11:01AM
Location: Perth, Western Australia
Posts: 358

Having had my FET's stolen, I'm now going to use my 40n60 IGBT's
What is the best way to get the 20+V gate drive needed from UCC's?
1.5:1 GDT?

HV Enthusiast

Fri Mar 03 2006, 01:24AM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

Use the gate drivers in push-pull configuration (provides +15/-15 to gate transformer) and a 1:2:2 gate transformer design. This is the typical configuration used in most DRSSTCs.

Desmogod

Fri Mar 03 2006, 01:35AM

Registered Member #139 Joined: Sat Feb 11 2006, 11:01AM
Location: Perth, Western Australia
Posts: 358

so one inverting, one non inverting, and a 1:2:2?
Lovely.
Is it ok to wind, say, 10 turns of single wire, then 20 turns of bifilar(for a half bridge)?

HV Enthusiast

Fri Mar 03 2006, 02:18AM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

Desmo wrote ...

so one inverting, one non inverting, and a 1:2:2?
Lovely.
Is it ok to wind, say, 10 turns of single wire, then 20 turns of bifilar(for a half bridge)?

The best arrangement should either be:

1. Trifillar winding (3 turns intertwined)

2. Layering with the primary winding the middle layer

However, layering does increase leakage inductance. Best bet is trifillar wound. Start with 3 equal strands of wire. Wind 20 turns, then you can cut the primary wire in half paralleling it. This is the easiest way to do it.

Dr. SSTC

Fri Jul 04 2008, 04:17AM

Registered Member #1407 Joined: Fri Mar 21 2008, 07:09AM
Location:
Posts: 222

how did you get your fets stolen

Steve Conner

Fri Jul 04 2008, 09:59AM

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

My DRSSTC worked fine with a 1:1:1 GDT, I never found the need to go to 1:2:2.

Steve Ward

Fri Jul 04 2008, 06:42PM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

My DRSSTC worked fine with a 1:1:1 GDT, I never found the need to go to 1:2:2.

Yeah, but you baby those things and probably dont even exceed their peak current rating.

The issue here is desaturation of the Vce junction from excessive currents. The datasheet provides Vce vs Vge vs Ic graphs, extrapolate from them to figure out what gate drive voltage you need. I dont think the 40N60s would like 600 to 1000A with only 15Vge, and so you must overdrive the gate to keep them from desaturating.

GeordieBoy

Fri Jul 04 2008, 09:43PM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

> Yeah, but you baby those things and probably dont even exceed their peak current rating.

That may be true, but Steve's coil produces some very impressive sparks for it's size. I guess that keeping within device maximum ratings is just something that is ingrained into us engineer types. Out of interest Steve C, how many sets of those IGBTs have you been been through so far...? ;)

Marko

Fri Jul 04 2008, 09:57PM

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

Out of interest Steve C, how many sets of those IGBTs have you been been through so far...? ;)

If I can speak for his name - I'm pretty sure he never replaced his IGBT's, and actually I think he mentioned it somewhere here not-so-long ago.

Staying within ratings is surely a good way to keep your IGBT's - but if I wanted to acheive spark lengths like Steve W. does on his DRSSTC1 on same coil size, it would be rather troubling to find the IGBT's that I wouldn't overrate.

Those would most likely need to be bricks, but their problem would be that they would be rather slow for typical frequencies around 100kHz. So I don't know how well would they perform.

Steve Conner once said that small and medium DRSSTC's are pushing what any IGBT can do - which is probably right..

Marko

Steve Conner

Sat Jul 05 2008, 12:36PM

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

I've never blown one. I'm still running the four that I started with.

Steve Ward is right, I effectively "babied" them by putting them into a coil about half the size that most people would build with a fullbridge of 40N60s. That way, I could generate sparks that looked impressively big compared to the coil (my record is 4.4x the secondary coil length) without even exceeding the manufacturer's 400A peak rating. I just set the limiter to 400A and forgot about it.

Like Richie mentions, I just don't see the point in overrating. When you do it, you're basically saying that you understand IGBT reliability and failure mechanisms better than the guys who make them, which of course you don't.

Yes, I did say that small and medium coils are right on the edge of what any IGBT can do. If you make the device bigger so that conduction losses don't kill it, then the increased switching losses kill it instead. If it weren't for Fairchild's latest generation of fast IGBTs, we couldn't have made tabletop sized coils at all.

With big coils the frequency is lower and the switching losses less of a problem.

1 2 next

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