DRV8844: current derating for higher voltages

Hi Rick,

Thanks for your quick response.

So just to make sure I'm understanding correctly, the only impact of increasing the supply voltage would be that the extra power dissipated contributes to self-heating of the chip, thereby increasing Rdson and lowering the max allowable current?

If the FETs' limits are imposed mainly by temperature, does that also mean that if I'm only using the chip to control a single BDC Motor (so effectively only using half the package), that the current limit might be pushed a little higher?

thanks,

Ryan

Hi Ryan,

So just to make sure I'm understanding correctly, the only impact of increasing the supply voltage would be that the extra power dissipated contributes to self-heating of the chip, thereby increasing Rdson and lowering the max allowable current?

Yes, that is correct.

If the FETs' limits are imposed mainly by temperature, does that also mean that if I'm only using the chip to control a single BDC Motor (so effectively only using half the package), that the current limit might be pushed a little higher?

Yes, that is true to a point. The limit is the overcurrent trip point of 3A.

If you are only controlling a single BDC motor, you can configure the outputs in parallel as described in section 8.1 of the datasheet. This would reduce the power in the FETs by ~50%.

Oh, I've just seen in section 8.1 of the datasheet that the outputs can be connected in parallel to increase drive current. So in my case, driving a single BDC motor, I'd tie OUT1 - OUT2, and OUT3 - OUT4? Would that then effectively double the current limit?

Oh, looks like we replied at the same time. So in the parallel configuration, the power dissipation is deceased by about 50%, but there's still an overcurrent threshold at 3A? So basically, connecting in parallel with make things run cooler, but still won't be able to drive more than 3A?

Hi Ryan,

Oh, looks like we replied at the same time. So in the parallel configuration, the power dissipation is deceased by about 50%, but there's still an overcurrent threshold at 3A? So basically, connecting in parallel with make things run cooler, but still won't be able to drive more than 3A?

You basically get your choice in parallel mode.

For the same current, the power is decreased to ~50% due to the parallel FETs.Or you can increase the current to approach 5A peak (3.5A RMS).

The overcurrent is measured per FET so as long as the current is relatively balanced per FET the OCP should occur at approximately 6A in parallel mode.

ok, thanks for clearing that up.

I have one more question regarding parallel mode: in section 8.1, I find the use of the word 'paired' kind of confusing. When it says OUT1/OUT2 must be paired, and OUT3/OUT4 must be paired, does that mean OUT1 and OUT2 should be shorted and connected to one side of the motor, and OUT3 and OUT4 are shorted and connected to the other side? Or does it mean that OUT1 and OUT2 should be paired in the sense that they're one driver pair, so OUT1 goes on one side and OUT2 on the other, and then the same thing for OUT3/OUT4?

Hi Ryan,

"paired" means shorted and connected to the load (one side of the motor in your case). This only applies to configuring the outputs as two parallel independent half bridge configurations. Pairing the outputs as described allows proper current measurements using a sense resistor from SRCxx to GND.

In parallel full bridge mode, you can connect any two outputs to one side of the motor and the remaining two outputs to the other side of the motor. Choosing the two connections may help layout. If using a sense resistor, short SRC12 and SRC34 and connect to one side of the sense resistor.

Will I also be tying the corresponding 'INx' and 'ENx' pins together then, to ensure the control is synchronized? For example, assuming I tie OUT1 to OUT2, would I also tie IN1 to IN2, and EN1 to EN2?

Hi Ryan,

Yes, you would.

Good point.