H bridge cross conduction (TIG inverter)

Or relevance to sci.engr.joining.welding is a question of how much current "leak" would not trigger turn off of the high frequency arc starter.

======================================================================

So, now I have a working H bridge for my TIG inverter project. For those who missed my previous posts, I have a nice DC TIG welder that I am trying to convert into advanced squarewave AC/DC welder.

This bridge cross conducts, that is, there are brief moments when all of its IGBTs conduct. You can think of it as a short condition lasting for a very short moment.

For a welder with a big ass inductor, that is considered generally a good thing, as it reduces spikes at turnoff (since there is never a true turnoff from the standpoint of the inductor).

The amount of cross conduction depends on resistance of the resistor that is placed on the circuit that discharges the gate (the off resistor).

I am trying to determine just how much cross conduction do I need, which would be adjustd by adjusting the off resistor.

I gathered some data.

With almost no off resistance (off resistor shorted with a 0.5 ohm alligator lead):

20V OCV -> 0.035A cross conduction 30V OCV -> 0.047A cross conduction 38V OCV -> 0.050A cross conduction

With 100 Ohm off resistance:

20V OCV -> 0.1 A cross conduction

That means that with 100 ohm resistance, in a TIG welder with 85V OCV, the amount of cross conductance could be approximately 0.42 amp.

That's really not a lot for a 200A welder, HOWEVER, I am concerned about the HF arc starting.

My approximate calculations suggest that this short moment, for 100 ohm turn off resistors (see below), the cross conduction interval amounts to approximately 10 microseconds for every switch. I may be way off base here.

I also hope that with more realistic frequencies, such as 200 Hz, the cross conduction would decrease as the number of shorts per seconds decrease. At 200 Hz, for example, I can hope for only 0.08A cross conduction. At 500 Hz, my highest setting, I would expect about 0.21 A cross conduction.

I have a on demand HF arc starting, where HF is turned off when the main welding arc starts, and turned back on when the welding arc extinguishes.

My c ``HIGH FREQUENCY SWITCH -- in the AUTO position, the high frequency starter circuit is energiuzed when there is open-circuit voltage at the output terminals. Once an arc is established, the high frequency circuit is automatically deenergized. (the TIG/STICK switch must be in the TIG position).''

That suggests to me that paltry 0.42A are not going to substantially lower the voltage from OCV to a much lower voltage and turn off HF. Am I wrong here?

i