Re: BEWARE of dead gate drive IC that still works

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Justin Hays by way of Terry Fritz <twftesla-at-qwest-dot-net>" <pyrotrons2000-at-yahoo-dot-com>

Hi All.

Sorry I'm a little late on the follow-up.

Yes, I'm pretty sure that due to its leakage inductance, the gate
isolation transformer is kicking back into the gate drive IC and
causing failure. 

Malcolm, you mention you read that these IC's are made specifically
for drive into capacitive (MOSFET gate) loads. This is completely
reasonable (it is a gate drive IC!) but I know they can reliably
drive isolation transformers also. I have seen Maxim and
International Rectifier application notes where their chip's drive
them directly. The app was usually half-bridge based SMPS's where the
the upper MOSFET's gate drive could not be ground referenced (for
obvious reasons). Which is cool, they only used one transformer as
the lower FET doesn't require isolation ; )) This is unique, but I
wonder what effect this has on the ouput waveform, given that there
is now asymmetry introduced into the switching (MOSFET gate)
waveforms.

Anyhow, I will say that the essence of electronics is to find a WAY ;
) This could be a simple as using resistor to temporarily save the IC
(as Richie Burnett points out to me), OR, as Dave Sharpe's excellent
suggestion, by using a couple of shottky diodes. I like the 1.5 ohms
output current of the MAX4420 and 4429 chips in a single SOIC (and
DIP8) package....I'd much rather use them than discrete
components......although, we will see what happens!

<<<<<<<<<<
(snip from Dave Sharpe)
On the ouput of the driver chip to the gate drive transformer (GDT), 
install two high speed schottky diodes in a doubler configuration. 
The anode of the lower diode goes to the ground and cathode to output
of the driver. The second diodes anode is wired to output of driver
and cathode is to +Vcc operating chip / GDT.  Must be a high speed
schottky like 1n5819-1n5822 family.  The transformers leakage
inductance is kicking back into chip at power up or down and driving
output above (or below) supply rails.
>>>>>>>>>>

I'll be _____ if this isn't exactly what is happening. I have seen
this diode arrangement before in SMPS and motor control designs that
I reverse engineered, and I'm thinking so that's what they're for!

Thanks to everyone for your input.

A great list Terry*.

Justin Hays
KC5PNP
Email: justin-at-hvguy-dot-com
Website: www.hvguy-dot-com

*It's Chip's list ;-)