IGBT driver update: I called International Rectifier

I decided to just call IR on the phone. Asked them for a 1,200V gate
driver for large IGBTs, with negative reverse bias. Spoke to a very
knowledgeable and helpful engineer.

He suggested a new chip IR22141SS. This chip has negative reverse
bias, can drive large IGBTs, has 1,200 V isolation, needs only one
supply voltage, etc etc etc. He also suggested to read design tip
DT04-4. I printed the datasheets and the DT, and so far, everything
looks like this chip perfectly suits my needs, except that it is in
SSOP and I have a DIP breadboard.

I think that there are adapters that cam change SSOP24 into DIP.

http://www.accutekmicro.com/product_detail.cfm?Product=DIP Adapters

AK24D300-SSOP.8mm 24 Pin SSOP 0.8mm PITCH to DIP


i

 

And the best part is...

I called their dealer on the phone, and then they called me back. They
were able to order me free samples of these chips (2 chips). The rep,
a super nice woman, said that I should have them on Friday or on
Monday.

i

--

 

Ignoramus1797 wrote...

Out of the frying pan, into the fire. :>)

 

Be careful with them !

When the IGBTs fry - the driver normally goes too .

Graham

 

I will try to be careful indeed. I will make a detailed drawing of the
relevant part (minus signal generator part) and will post it prior to
powering it up.

i

 

Forgot to say. There would not be much to that schematic. The chip has
almost everything, it just needs a diode, bootstrap capacitor and
possibly a few resistors.

i

 

Ignoramus1797 wrote...

If you think a schematic contains all the relevant information,
you will find you are sadly mistaken. (Inductance is one key.)

 

You are right. I will try to actually measure inductance of my
welder. I think that I can do it with the stuff that I have (decade
resistor, signal generator and voltmeter).

i

 

I suspect that Win was trying to alert you to layout issues wrt the gate driver
actually. Read IR's app notes thoroughly !

Graham

 

You are right, DT04-4 discusses layont of the circuit.

i

 

Speaking of IGBTs, I'm thinking about something like this:
http://www.irf.com/product-info/datasheets/data/50mt060wha.pdf
for my induction heater. (Problem being IR doesn't show any
distributors...) If not, then I'm thinking a quad of TO-247
devices of similar total spec.

I was also thinking of a cooked-up circuit something like
http://webpages.charter.net/dawill/Images/Gate Drive.gif
(yeah, so it has a lot of stages... I don't see another easy solution
to get it saturating by the floating rails. Oh and I'll do a DC-DC
converter for the floating supplies) for gate drive, and
http://webpages.charter.net/dawill/Images/Desat Detector.gif
for desat detection.

Any comments?

Tim

 

["Followup-To:" header set to sci.electronics.design.]
On 11 Oct 2005 14:15:58 -0700,

*chuckle*

robert

 

There's a weldor about 5 or 6 feet from my office, welding 1/2" or
so mild steel. I don't know what amps setting he's using, but
every time he strikes an arc, my monitor distorts from the magnetic
field. I don't know how much inductance that translates to, but
it's almost scary, thinking about how to deal with that kind of
current/volts/transients with semiconductors. And it's not
power line droop - the image actually twists! =:-O

Good Luck!
Rich

 

I have a small desk fan sitting on my desk, 1.5 ft away from my
screen. It uses 0.55 amperes of 110V. I need it on cold days because I
am sitting right under an air conditioning opening blowing hot air.

Whenever I turn that fan on, the image on my screen begins to shake.

The reason for relaying this is that I am not sure just how many
conclusions we can make from seeing image distortion.

That said, I want to thank you and another poster for making me aware
of high inductance of the welder and I will try to perhaps delay turn
off enough to see some cross conduction.

i

 

I had a funny experience in the early '60's while working on
alternator regulator designs.

The control chip sat on the ground tab of the package leadframe.
Field current, ~5A PWM'd, flowed under the chip. At every switch of
state all hell broke loose. Turned out that the resulting magnetic
field moved the BJT base current sideways killing the beta.

Redesigned the leadframe to have the control chip on its own
no-current-under stub cured the problem.

...Jim Thompson

 

I read in sci.electronics.design that Ignoramus24489

Depends how much it shakes, but think around 3 A/m or 3.6 uT.

 

yep. Yuck!

always ensure base currents are limited. lest stuff fail during wierdo
start-up, shut-down or interference conditions. in general, such linear
circuits have piss-poor performance. actual thresholds vary widely with
time, temperature and sock colour (or so it seems....). Use a real
reference voltage, and a proper comparator with suitable hysteresis.


AIUI you drive the IGBTs with a fixed duty cycle (almost 50%) variable
frequency square wave. In which case, a transformer + gate resistor
would do just fine. Or, if you want, use the transformer and a couple of
diodes + caps to generate the +/- supply rails, *and* drive a
complementary emitter-follower (thru some Rb) which drives the gate.

likewise the desat detector is equally yuck. use an LM339, and just
measure Vc. there are 2 methods: a resistive divider, or an HV diode and
a pullup to (say) +15V. Either way, note that Vce can be very, very
negative - even though the IGBT has a reverse diode, there is an
inductor between it and where you measure Vc.


Cheers
Terry

 

If I were to develop a unit with those chips, I would get an entire
*TUBE*. And I would probably use them, too. Until your gatedrive works
properly, it doesnt work properly. And gatedrives that dont work
properly cause explosions. Even when you are experienced and careful.

But there is a *LOT* to be learned from such an exercise.

Cheers
Terry

 

solid groundplane underneath the circuit. And keep IGBT gatedrive
connection inductance low - 50nH is really too much, 10nH is better.

if you use a DIP proto-board, it will go *BANG* the first time you try
to use it on the welder.

Cheers
Terry

 

sneaky. bet that wasnt easy to figure out.

Cheers
Terry