H-Bridge Circuit Opinions

[ssylee]

I have two different H-bridge circuits that I would like to show you, attached in the message. More specifically, what are the major differences between H-bridges with intermediate BJT drivers vs. those without? Thanks in advance.

 

[dougy83]

The left circuit has an IC to provide the level translation from the inputs (I'm assuming they're TTL level) to the mosfet gate level (6-12VDC). It has 2 inputs and can switch the load in 2 directions and off.

The right circuit uses BJTs to provide the level translation. I notice that the wiring of the mosfets is incorrect and they will short out the supply. You'll also notice that there's only a single input, which means (if the mosfets were wired correctly) that the load can never be switched off (which is why they have capacitively-coupled the load).

 

[phoenox]

It looks to me like the right circuit will turn Q1 and Q4 on, or Q2 and Q3 on at any given time. This will not create a short. There are no flyback diodes on this h-bridge. That will allow voltage spikes from the load to damage the mosfets.

The top side diodes in the left circuit are backwards. This will create a short across the h-bridge as soon as it is turned on.

 

[MikeMl]

I have two different H-bridge circuits that I would like to show you, attached in the message. More specifically, what are the major differences between H-bridges with intermediate BJT drivers vs. those without? Thanks in advance.

They are the SAME in that NEITHER will WORK!!!

 

[ssylee]

They are the SAME in that NEITHER will WORK!!!

I tried them both in real life before and both worked.

 

[MikeMl]

I tried them both in real life before and both worked.

Your definition of "working" must be different than mine

How do you explain that in the Left circuit, Q1, Q2, D1, and D2 are wired the wrong way around. The diodes are always forward biased, as are the Drain to Source diodes in the PFets?

In the right circuit, the sources of the P-Fets Q1, Q2 are operated at -6V, which means the drain to source diodes of the NFets, Q3, Q4 are always forward biased (ie they are connected backwards)?

 

[ssylee]

Thanks for the suggestions, dougy83 and phoenox. I have also used this circuit before, where the inputs are from the L293 driver, with one input being the inverted signal of another input from the output of driver. It would be great to get comments about it. Thanks.

 

[MikeMl]

The 10K resistors used to discharge the gate capacitance of all 4 FETs creates a slow turn-off which causes the Fets to dissipate a lot of power as they turn off. The goal when designing gate drivers for H-bridges is to turn-on/turn-off the Fets as rapidly as possible. This requires a gate driver which can both charge and discharge the Fet gate capacitance rapidly.

At least the polarity of the Fets in this circuit (compared to the other two you posted) is correct...