[TCML] psstc troubles-h-bridge

[jason hill <effuzionz@xxxxxxxxx>]

this post goes along w the thread "need help psstc." had noise in gate drivers, filtered that, now having trouble with our h-bridge which is explained quickly below. im sure i missed a few key points in analyzing all aspects of the problem. stress and fatigue may be a factor. 

fyi- this is being used for audio modulation. circuit employs a tl494 which is getting a clean signal now and the gate drivers are tc4422's. r-c low pass filters at gate driver outputs, initially set to about 80 ohms w a .0033 cap to ground. output signal at r-c node (butterworth passive low pass filter). signal out of gate drivers is clean, signal at gates of FET's after GDT's is just as clean. filters need adjustment when analyzing signals at gates which drives resistance down in order to get correct signals. using potentiometers to set resistance. 

alright! so we have done a fair amount of filtering and have gotten precisely the correct wave-forms through our GDT's. we have a clean square wave w distinct on and off states due to some passive r-c low pass filtering that we have set up using very low resistance and .0033F cap at the outputs of the gate drivers. phasing is correct on the GDT's and the waveform at the mosfets gates (g-d) are 20v p-p which should clearly be plenty to turn the mosfets on and off cleanly through our h-bridge. 
now we have set things up to check the output of the h-bridge under no load w 2 probes set up in differential mode and grounds connected to eachother. we place our hv dc power supply on the correct rails and begin to power these rails in 5v increments. as we start to ever so slightly bring this voltage up we get this clouded signal that is definitely not a square wave or a resemblance of any other kind of waveform. we have our circuit set up for push-pull mode which once fully energized should give us a square wave that should have a rather large voltage swing. this clearly is not the case. we are resuming testing on the mosfets to see if they are infact still good and tediously checking all connections. 
has anyone else had this problem? when we increase the voltage on the rails the g-d signal gets stretched out and pulled in all directions...
fix one problem and move to the next! cant help but laugh at this point but any advise would be greatly appreciated.

just a quick question....with our mosefets in the h-bridge circuit, meaning we can not easily remove them, should the same rules apply when testing them in the circuit or should we take them out in order to properly test them? im sure i will find this answer as i do some more research and my head begins to clear....working out these problems thus far has pushed my capacity for the night. 
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla