"Inazuma" Mk.1 DRSSTC / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

"Inazuma" Mk.1 DRSSTC

previous 1 2 3 4 next

Move Thread

LAN_403

thedatastream

Thu Jul 26 2007, 01:23PM

Registered Member #505 Joined: Sun Nov 19 2006, 06:42PM
Location: Yorkshire!
Posts: 329

I ran the coil up this lunchtime in the workshop, blowing up two full sets of transistors in the process. The first set went with a whimper, but the second set gave a very good account of themselves, generating what looked to be about an 18"+ arcs to air with a 200V, 2A input with a rather splendid noise. It was pretty short lived, maybe 10 seconds, before it blew.

My colleage took a video on his camera, which I will hopefully post in the next couple of days.

I reckon a bit more careful tuning and less turning up the volts would have been a better idea! I'm quite happy with the measured performance so far, it's about what I was expecting. I will repair the coil, tune it a bit more carefully (I might actually measure something this time) and run it up again.

More IGBTs please!
James

Bennem

Thu Jul 26 2007, 04:57PM

Registered Member #154 Joined: Sun Feb 12 2006, 04:28PM
Location: Westmidlands, UK
Posts: 260

Weh Hey!!......good one James!

Did you blow all four igbts?

look forward to seeing that video!

Mel.

Marko

Thu Jul 26 2007, 09:18PM

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

I ran the coil up this lunchtime in the workshop, blowing up two full sets of transistors in the process. The first set went with a whimper, but the second set gave a very good account of themselves, generating what looked to be about an 18"+ arcs to air with a 200V, 2A input with a rather splendid noise. It was pretty short lived, maybe 10 seconds, before it blew.

Hi

With half bridge of 30N60's and incredibly crappy polyester MMC I managed similar arcs, but ran at nearly a microsecond of on time with arcs nearly CW looking. (boom).

I'd say something is wrong in there. Why did you replace broken IGBT's without debugging them first?

Make sure to fix it, or rather give another set to me

if you just intend to blow them.

Tom540

Thu Jul 26 2007, 09:43PM

Banned on 3/17/2009.
Registered Member #487 Joined: Sun Jul 09 2006, 01:22AM
Location:
Posts: 617

Don't forget to make videos each time you fire it up. I've got all my failures on video.
Speaking of blowing all for igbt's. Shouldn't you replace them all anyway? I tried replacing only two n my first coil and then next time id run it the older two would fail. It was never ending until replacing them all at once. What IGBT's were you using in that coil anyhow. I just bought some STGW40NC60WD igbt's they were only 3.18 each! The FGA40N60's werent cutting it after I made some changes.

thedatastream

Fri Jul 27 2007, 07:05AM

Registered Member #505 Joined: Sun Nov 19 2006, 06:42PM
Location: Yorkshire!
Posts: 329

The first set of transistors had maximum voltage and on time accidentally applied to them as I forgot to check my interrupter settings before pressing the fire button (oops).

I suspect that I'm getting either large voltage spikes or non ZCS due to me not tuning the driver properly. I'm erring on the side of the voltage spikes as it didn't look fantastic when I had it on the bench. I am going to look at ways of reducing these spikes including maybe a new bridge layout, more decoupling or maybe even some transient voltage suppressors before I fire it up again in anger.

Yes, all four IGBTs blew in both cases, dead short drain to source but the gates were OK. I was using STGW30NC60WD but when I tried to order some more from Farnell they had sold out

Instead I spoke nicely to Fairchild and I've got some samples of their HGTG40N60B3 being sent.

Danielle

Sat Jul 28 2007, 08:36PM

Registered Member #632 Joined: Mon Apr 09 2007, 01:09AM
Location:
Posts: 267

with those IGBTs you will need diodes with them because thay dont have them built in I would have recommended the ones I use I have got 68" arcs with them. HGTG30N60C3D from fairchild. I think you have a simpler problem possibly with tunning always when first trying it out have a current limiter set really low and run it at low voltages 20% AC in. and slowly ramp up the power after you tunned the coil.

Marko

Sun Jul 29 2007, 02:19PM

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

Yes, all four IGBTs blew in both cases, dead short drain to source but the gates were OK. I was using STGW30NC60WD but when I tried to order some more from Farnell they had sold out Instead I spoke nicely to Fairchild and I've got some samples of their HGTG40N60B3 being sent.

Huh, I wish if I was cool like you to be able to get such insane samples. But you do need to get diodes for those..

Believe me, this is not the first time somebody blames ''voltage spikes'' for his failures. In all cases the cause was actually trivial.
Do some digging, carefully increase ON time and voltage if you need to run it and I'm sure you'l find the problem...

Steve Ward

Sun Jul 29 2007, 04:37PM

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

Believe me, this is not the first time somebody blames ''voltage spikes'' for his failures.

Its always heat (unless you blew up the gate oxide layer). Heat either from avalanching (which is good at causing very fast hot spots as the dissipated power is BIG) or from too much current (not as likely). If there is one thing i learned in class, its that P-N junctions dont fail like a capacitor dielectric... they avalanche (or tunnel, like a zener diode) and get hot and melt. In fact, there is nothing wrong with running a semiconductor in avalanche mode provided it doesnt dissipate too much power doing so.

thedatastream

Mon Jul 30 2007, 04:04PM

Registered Member #505 Joined: Sun Nov 19 2006, 06:42PM
Location: Yorkshire!
Posts: 329

Thanks very much for all of the insights, I've got several things to try now. I'm just waiting on some parts (diodes and some TVS) and then I will be back in the business of tuning and turning things up slowly!

I'm also going to check that my current limit circuit works - just in case...

Any recommendations for interrupter settings during run up? I know that long bursts at very low power show if the coil is tuning OK but as the power increases what PRF and on time to people tune up with?

Thanks again
James

Marko

Mon Jul 30 2007, 04:09PM

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

It would be *really* good idea to install current limiter. Without it, you need to be really careful with your interrupter. Always start with low variac voltages, tune, settle everything and gradually increase if things look fine.

I think it's good to start with interrupter setting of 50us and PRF of 100-150Hz. Depending on few things, mainly your characteristic impedance, you shouldn't need more than 100-200us of ON-time in the end. As I think.

previous 1 2 3 4 next

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