Re: SSTC- feedback vs. none

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: "K. C. Herrick" <kchdlh@xxxxxxx>

A follow-on to my earlier posting today:  The delay problem in TX1 is 
due to slow turn-off of the forward-biased transistor base:collector 
junctions.  As TX1's outputs reach their extremes, the respective 270 
uF capacitors become recharged through those junctions (and that's 
near the tops of the waves so the cessation of normal 
emitter-follower action is immaterial since the IGBT gates have 
already been almost completely driven).  At least in the simulation, 
the junctions apparently remain on while the signals are at their 
peak values, and then have to be turned off again before the base 
voltages can again begin to change.

As I mentioned, I have added schottky diodes across those junctions 
in the actual hardware, which will preclude forward biasing of the 
base:collector junctions.  When next I fire it all up again, that 
will be one of the first things I'll check: whether those diodes are 
turning off fast enough to eliminate the delay.

KCH