Half-Bridge Coilgun Exploding Rectifier Problem... / Electromagnetic Projectile Accelerators / Forums

Forums

4hv.org :: Forums :: Electromagnetic Projectile Accelerators

« Previous topic | Next topic »

Half-Bridge Coilgun Exploding Rectifier Problem...

Move Thread

LAN_403

maruiwlt

Thu Dec 13 2012, 11:48PM

Registered Member #859 Joined: Mon Jun 25 2007, 10:23PM
Location:
Posts: 11

Hello all, I'm working on a half-bridge coilgun and have gotten it to fire safely at <200 volts, but at 300volts the rectifiers that should be recovering the back emf are exploding.

They are UF1007 ultra fast rectifier diodes. I'm quite certain that it isn't the recovery speed since it is as fast as the UF4007 which is common in half-bridge coilgun designs.

I've come across a few schematics that put multiple rectifiers in parallel. Is this something I should try?

2Spoons

Fri Dec 14 2012, 02:24AM

Registered Member #2939 Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615

From the datasheet

Non-Repetitive Peak Forward Surge Current :8.3ms Single half sine-wave Superimposed on Rated Load : 30 A

Are you exceeding this?

I would always change to bigger diodes before running multiple diodes in parallel.

maruiwlt

Fri Dec 14 2012, 02:44AM

Registered Member #859 Joined: Mon Jun 25 2007, 10:23PM
Location:
Posts: 11

Ah that's probably it... I have a few 1n5407 diodes that are rated to 200A surge but are not ultra-fast types.

Do you know what the recovery time affects in a circuit like this? I'm assuming the slower switching time will cause the inductor to build up a higher voltage while current isn't flowing. Then the diode or IGBT may explode due do the spike... lose/lose I guess :/

I don't have time to order new diodes so I'll try to parallel a few uf1007's first. Better to have those explode again than a voltage spike destroying my IGBTs.

Thanks!

Yandersen

Fri Dec 14 2012, 03:12AM

Registered Member #6944 Joined: Fri Sept 28 2012, 04:54PM
Location: Canada
Posts: 340

No matter which diode you will use, there will be no spike. Switching time is a time for diode to close when polarity inverses (yes, it is still conductive for some periode of time even being in inverse direction, and the higher the current was the longer the diode will keep it's doors open). This characteristic should be minimum for diodes used in switching-type power converters applications. For half-bridge it doesn't make any difference. Use slow strong ones.

Ash Small

Fri Dec 14 2012, 07:46AM

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

Yandersen wrote ...

No matter which diode you will use, there will be no spike. Switching time is a time for diode to close when polarity inverses (yes, it is still conductive for some periode of time even being in inverse direction, and the higher the current was the longer the diode will keep it's doors open). This characteristic should be minimum for diodes used in switching-type power converters applications. For half-bridge it doesn't make any difference. Use slow strong ones.

You are talking about reverse recovery time here aren't you, Yan? Would schottky diodes be a better alternative here, as they have no reverse recovery time?

EDIT: Having re-read your post, I think you are saying that the switching time isn't what is causing them to explode, it's the current rating that is causing them to explode.

Yandersen

Fri Dec 14 2012, 08:06AM

Registered Member #6944 Joined: Fri Sept 28 2012, 04:54PM
Location: Canada
Posts: 340

Schottky will be good. I doubt it has no recovery time though, but it doesn't matter in this application.

Of course it is overrated current - do you believe that they got scared of HV that much? :))

Ash Small

Fri Dec 14 2012, 09:36AM

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

Yandersen wrote ...

Schottky will be good. I doubt it has no recovery time though, but it doesn't matter in this application.

Of course it is overrated current - do you believe that they got scared of HV that much? :))

Yes, I'd assume it is too much current. I was on my first coffee this morning after a phone call woke me up, so wasn't really with it, although I don't really understand what happens when diodes reverse-conduct while there is a charge carrier depletion region at the junction (all that 'charge carrier and holes' stuff never really made much sense to me. I do struggle with understanding how semi-conductors actually work, but never had a problem understanding valve (vacuum tube) technology, or any other particle physics, for that matter. I'ts just the NPN-PNP stuff that baffles me.)

Yandersen

Fri Dec 14 2012, 07:13PM

Registered Member #6944 Joined: Fri Sept 28 2012, 04:54PM
Location: Canada
Posts: 340

Don't worry, neither I understand it that deep. I just read datasheets to know the characteristics and wikiing what it means. :)

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