I am trying to drive two motors in parallel with an H-bridge circuit (using MOSFETs). At this point, I am not sure whether I will be switching the H-bridge at high frequency (20kHz which is out of hearing range) or not, but for now I would like to design my circuit with the capability of high frequencies. I would like to use a driving chip by the name of TC4469. It is a “Logic-Input CMOS Quad Driver”. I looked at the datasheet and found a circuit diagram that I don’t completely understand. I have attached a picture of the circuit that can be found at the bottom of this post (I hope). Alternatively, I added a link to the datasheet, where the circuit can be found on page 11. Here is the link to the datasheet: http://ww1.microchip.com/downloads/en/DeviceDoc/21425C.pdf
The diodes in this circuit are the part that’s confusing me. I know that the gate of a MOSFET has capacitance and needs time to charge and discharge (more so at high frequencies). To my knowledge, these diodes are currently bypassing the gate resistor and allowing the gate of the MOSFET to charge without any current limiting. When discharging, the current has to go out through the resistor. I am confused by this because I thought the diodes should be placed in the other direction, so there would be current limited on the way into the gate, and then the diode would allow the current to quickly discharge.
Could someone tell me why these diodes are placed in this direction? Also, shouldn’t there be resistors and diodes on the bottom two MOSFETs as well?
Looking at the datasheet, I'm guessing what you're going to be using are two drivers? One driving each motor, and then cycling it on/off with PWM to control rotational rate. This seems to be how the datasheet wants you to do it (odd).
While it's a different part, you may wish to look at the datasheet for the L293 H-bridge, it has a better example of how you want to configure the diodes: http://www.ti.com/lit/ds/symlink/l293.pdf
What I can't tell from the 4469 DS is whether there is built in diode protection of the circuit. In many cases I've had a lot of problems as a result of back EMF frying mosfet and BJT components, it's likely the diodes and resistors are providing this in the example circuit.
The diodes/resistors ensure that a minimum voltage will reach the inverted side of three of those logic gates (pins 4, 6, 9) instead of everything being sunk into high-capacitive gates in the H-bridge mosfets. Or in other words it's to avoid shoot-through. It has nothing to do with back-EMF protection; in figure 5-1 it's pretty clear that none is being used there.
That circuit only works from a stable 12V supply.
Why not use all n-channel FETs and high-low drivers, then the MOSFET supply can be any voltage
you want, only the drivers need a stable 12V. Typical motor supplies are not stable and not
spike-free (the 18V zener they show in the circuit is there to catch spikes and is requred to
stop the motor frying the driver chip!)
Look at the datasheet for the HIP4081A to see how this sort of thing is normally done - with
proper MOSFET drivers you get protection functionality like shoot-through prevention and
under-voltage protection, which you need in the real world.
This chip is a line-driver, not a MOSFET H-bridge driver.
Thanks for all of the information. I think I am going to try control my motors with the HIP4081A as MarkT suggested. It sounds like a superior method.