this will do the trick... new IGBTs

This forum is for specialized infomation important to the construction and safe operation of the high voltage electrical supplies and related circuitry needed for fusor operation.
Post Reply
scientificfun
Posts: 42
Joined: Wed Dec 12, 2007 7:08 pm
Real name:

this will do the trick... new IGBTs

Post by scientificfun » Wed May 28, 2008 9:16 pm

I blew the igbts in the driver for my hf hv xray tank in my last try- arcover at 30kV...
I hope the multiplier is still intact but I guess so-

this wil not happen again- new IGBTs 1200V 400A full bridge plus over current detection is the solution I guess
So even if I force arcovers at high power (for showing off) the IGBTs should not blow (they really shouldnt- 6 pcs were 150 usd...) even if they are not zero current switching

Will take a while (and lots of questions) to build the driver because it should have current measurement/ over current detection and current feedback for resonace... further more it will have lots of options so I can not only use it to bring power to my tank- also it will act as a nice drsstc.. (at least I hope so...)

greets
Attachments
IMG_0095.JPG
IMG_0095.JPG (53.52 KiB) Viewed 1596 times

bpaddock
Posts: 54
Joined: Tue Oct 10, 2006 11:39 pm
Real name:
Contact:

Re: this will do the trick... new IGBTs

Post by bpaddock » Fri May 30, 2008 1:40 am

> I blew the IGBTs in the driver...new IGBTs 1200V 400A full bridge...

What lets the Magick Smoke out of IGTS, FETS and SCRs in most
cases is turn them on to slowly, causing 'Spot Heating' of the
die.

Think of a FET as hundreds of thousands of very small resistors
all in parallel, where each one can be turned on and off
individually. The 'resistors' closest to the gate turn on first,
and as the gate potential spreads across the die the rest turn
on.


With a slow gate turn on, a few of the small resistors nearest
the gate are trying to carry all of the load, which they can't
do, so they burn up, but the device does not fail quite yet. The
next time the device is turned on, which may be only milliseconds
away, or days away depending on the application, some more of the
resistors further in burn up. When the point is reached that
there is simply not enough of the 'resistors' left to carry the
load is when the Magick Smoke escapes.

This is why the parts generally run "for a while" before failing.
If it fails as soon as you fire it up the first time, you either
had a catastrophic short in the load, or the gate drive really
sucked big time.

There needs to a be a few amps of current pumped in the gate of the larger
parts, for short periods of time, to get the gate potential to spread
across the entire die as fast as possible.

You also want to get the thing turned off as fast as possible.

Digix
Posts: 58
Joined: Wed Oct 10, 2007 12:18 pm
Real name:

Re: this will do the trick... new IGBTs

Post by Digix » Mon Jun 02, 2008 2:23 pm

Controll logic is most important to keep them alive.

if you will try to control them with no deadtime (just feed square wave to transformer)
they probably will blow up no matter what ratings, because of shoot through currents.
I blew up few mosfets in that way before i even connected some load.

scientificfun
Posts: 42
Joined: Wed Dec 12, 2007 7:08 pm
Real name:

Re: this will do the trick... new IGBTs

Post by scientificfun » Mon Jun 02, 2008 4:59 pm

ok- so what to do for deatime? two gdt (one for each half of the bridge) 2x2 driverchips and then switch inverted (with deadtime)? I thought about that also to be able to control power- just make positive AND negative pulses from the bridge short? I.e. on a 10% off 40% on b 10% off 40% and so on for low power?
does anybody have a circuit description for that?
greets and thanks

Digix
Posts: 58
Joined: Wed Oct 10, 2007 12:18 pm
Real name:

Re: this will do the trick... new IGBTs

Post by Digix » Tue Jun 03, 2008 12:22 pm

it would be better if you could post your schematic.
all I can say is, that if you drive GDT with such waveform as here it will be bad.
http://users.tkk.fi/~jwagner/tesla/SSTC ... signal.jpg
correct waveform is this:
http://rayer.ic.cz/teslatr/sstc/photos/briosc06.jpg
it does not matter how do you make it and seems these is no big need to aim for superfast drivers.
almost any standard SMPS IC can do it properly, for example sg3535, (tl494 is not good for mosfets)

AnGuy
Posts: 208
Joined: Tue Jul 20, 2004 11:49 pm
Real name:

Re: this will do the trick... new IGBTs

Post by AnGuy » Tue Sep 09, 2008 12:30 am

"ok- so what to do for deadtime?"

You can use gate drives that have built in deadtimes, use gate drives that have programmable deadtimes, or use a PWM controller that supports a deadtime feature. PIC 18F (and newer) all have PWM controllers with programmable deadtime.

Some gate drivers such as IRS20184 (http://www.irf.com/product-info/datashe ... rs2108.pdf) have programmable deadtimes via a resistor.

Another factor is how large the gate charge is required for the switch (Mosfet or IGBT) is required to react saturation. If you fail to charge the gate fast enough you will create excessive amounts of heat in the switching device. You want to switch the gate on and off fast. However, fast switching can also create large voltage spikes caused by induction leakage. Induction leakage is very high in HV transformers because of the poor coupling between the primary and secondary transformer windings (necessary because of the large amount of insulation required). To drive those large IGBT you're going to need some hefty gate drivers. Othewise they run pretty hot and waste a lot of power too or you blow the gate drivers. I doubt the IRS21084 will be capable of providing enough current. IXYS makes some very high current gate drivers for IGBTs. I would look at what they have too offer. I would also keep a close eye on voltage spikes on your primary side. Make sure you use snubbers to bring down the spikes well below your IGBTs maximium switching voltage.

Best of Luck!

Post Reply