which combination is better drv8313 + ina240 or drv8313 + acs712 for current sense torque control?
What MCU will you be using? ACS712 needs 5V, so if your ADC can’t read 5V range directly then the other one would probably be better.
As I learned yesterday, hall sensors are less accurate than shunts. Assuming you’re using the 5 amp ACS712, its datasheet says 21mV noise and 185mV/A sensitivity, which works out to 113mA noise (interestingly the higher current versions have lower noise in mV, so they also work out to ~110mA). I don’t have enough experience to say how much that matters, but as with most things, I assume the faster you want to spin, the more it matters.
Hall sensors do have the advantage of not needing the ADC synchronized to the PWM, if your MCU is not on the list of ones where low-side sensing is already implemented (third table, half way down the page).
I’m pretty sure the INA240 would be way more accurate if you’re able to synchronize the ADC to the PWM cycles. If not it could be quite the mess (or at least it was on my INA181, which is a bit shittier than the INA240). So I agree with dekutree64 that would be the criterion I’d base the choice on.
Here’s unsynchronized current samples on the INA181:
This is (similar to what I got) synchronized:
