No can do since the high side mosfets won’t turn on. I had considered soldering a wire to the mosfet output pins and to the motor wire pad so I can run current across the sensor with a battery and resistor, but I don’t know if it would give meaningful information. I would get a current reading from it, but I assume whatever noise was going on before would be superimposed on it.
Reading the datasheet for the 30A sensors used on Valentine’s no-longer-available Qvadrans board, it lists the noise as 20mV, which works out to 25 ADC units. But it looks like it doesn’t use the full output range since the sensitivity is listed as 44mV/A, so 20mV works out to 455mA noise. The sensitivity on my sensors is about the same at 45mV/A, so my previous calculations underestimated the noise in amps. The listed 8mV noise is 178mA, and my GooserCS measurement works out to 394mA.
So apparently hall sensors are not very accurate. That explains why most people go to the trouble of creating those incomprehensible resistor networks, reduced PWM range, and tight ADC timing of low-side sensing.