simple hbridge issue

AlbertHall

Joined Jun 4, 2014
12,345
MOSFETs, like everything else, don't turn on or off instantly so in this circuit their conduction times will overlap giving a shoot through current.
 

WBahn

Joined Mar 31, 2012
29,976
Your simulation results show the current in R1, but the schematic (the one posted with the sim results) doesn't have an R1 (at least not identified as such). Be sure that you schematics have all the needed annotations to communicate your results clearly.

This isn't a good sim setup to measure something like shoot through. You appear to be using ideal voltage sources. Do you have them defined with step changes? That will cause non-realistic results since you will be driving a capacitive load with an ideal step change, which will result in an unrealistic amount of charge injection (which may or may not skew your results too much).

Also note that your gate voltages are NOT the gate-to-source voltages for the upper transistors, so if you have a more realistic rise and fall times for the gate drive signals, the level shift will result in there being finite overlap of the transition times.

A good way to get reasonable drive signals is to use the backend of whatever circuit will be driving the actual H-bridge, or something close. Just making a simple CMOS buffer is far better than using an ideal source. You can then usually drive THAT stage with an ideal signal and rely on it to naturally produce something realistic at its output.
 

crutschow

Joined Mar 14, 2008
34,281
how can this simple hbridge have shootru when its gate voltages are precisely out of phase with no overlap?
There won't be much shoot-through if the gate drive signals have zero output impedance and instant rise and fall time.
But that's not achievable with a real circuit.
Finite drive impedances and finite rise and fall times, means there will be overlap where both the top and bottom MOSFETs on the same side, generating a significant shoot-through current.
That's why typical H-bridge drivers are designed with a dead-time between signals.
 
Top