From: Mike Poulton
Subject: Re: Problems with MOSFET drivers
Date: 15 Nov 2002 16:14:37 GMT
Organization: MTP Technologies
On 15 Nov 2002, Winfield Hill said:
> Mike wrote...
>> Here's the problem: the drivers explode! Loudly and reliably.
> There are so many things that make me squirm about your goal and
> your circuit, if I understand them correctly, that I don't know
> where to start. OK, try using proper levels and signals at each
> point in your circuit. E.g., for FET drivers this means applying
> a proper logic signal, which you can get from a comparator.
> But before trying any of your intended self-oscillating setup, I'd
> source the FET drivers from an external bench oscillator, using a
> very low-duty-cycle pulse burst, to allow debugging the H-bridge
> and its many parasitics.
I've done that, and it works fine. I can drive the thing with a function
generator and get a pretty decent output when it's at the resonant
frequency. I have not tried applying more than 24V to the H-bridge yet,
but at 24V and 2A it lights fluorescent tubes with no wires from a couple
feet away. I could just apply full power and leave it at that, but the
need for constant tuning is a major disadvantage. Besides, I need my
function generator for other things. Regardless, the power switching is
taken care of -- it's the drivers that seem to be the problem.
The spec sheet for these fet drivers can be found here:
They are rated for input voltage down to -5V and up to Vdd. I can't
imagine that I have exceeded those limits here. Whats the advantage of
adding a comparator here? The fet drivers already have a wide input
voltage range and the exact voltage at which the state changes is not
critical. Adding more propagation delay is probably not advisable since I
already have 60ns (plus 75ns rise/fall) from the fet drivers, so I wanted
to keep the number of signal-handling parts to a minimum.
Not only do I speak for my company, I AM my company!
Live free or die!