At 12:08 PM 28/02/2005 -0600, you wrote:
>My first guess was going to be shoot-through currents, even before I saw the 
>schematic.  It is best to drive the complementary fets with dedicated 
>drivers that are interlocked in some way.  You can get driver chips designed 
>to do this, or you can do it with coordinated sofware pulses, or you can use 
>a simple totem-pole bipolar transistor network to drive them designed so one 
>is always off when the other is on.  As long as they are never on at the 
>same time, it is OK, but if tehy are both on, even for a fraction, things 
>start to get hot.
>..................

>>device and turn on the N-channel device, the P-channel device
>>stays ON for a little while; the resulting shoot-through current
>>is what's causing the P-channel MOSFETs to get hot.
>>
>>Replacing the 4K7 with a 2K2 speeds up the P-channel devices'
>>turn off, reducing the problem somewhat.  To eliminate it
>>altogether I think you need to drive the P-channel devices with
>>something that will turn them off quickly-- maybe a complementary
>>PNP/NPN follower or something like that.

But as Wouter has repeatedtly said, shoot through does not apply because
he's leaving the P channel on, and pulsing the opposite N fet, then doing
the opposite to change direction.


Regards
Roland Jollivet

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist