Help can't figure out why the mosfet driver refuses to work

[30penny]

I originally built this circuit with a timing signal into a continuous toggle jk flip flop then into mosfet driver and mosfets and the circuit worked as expected no issues. But when I remove the timing pulse and jk flip flop and replace with lm311 and voltage translator ltp the fet driver stops working. I can put a scope on the driver inputs coming from the translator and it is getting pulses referenced to com but it's as if the fet driver doesn't see them or is at a different level. I'd start pulling my hair if I had some, any help much appreciated. 

[bikeron]

There is no ground for the driver, pin 4.. it is tied to "com" which is -50V...

[30penny]

My understanding of it is that it’s a floating well and it’s supposed to be connected to the negative rail. That’s what the translator transistor is for. It’s works when I build it. The get driver needs to be at the same ground potential of the lower mosfet. 

[bikeron]

I believe that the driver model expects that the ground of the IC is the ground of the rest of the circuit. Its a model issue not a simulator issue.

[30penny]

That's the answer I was expecting and hoped I didn't hear. Oh well. Tt's a fairly standard configuration for split supply halfbridge circuits, it's odd that this is the way they chose to model it. Do you know of any more accurate models?