Please tell me the influence of IB (Input base current) on the current measurement of uA order.
Q1: The effect of Ib on "Current Measurement-1", and the error that occurs.
Q2: At "Current Measurement-2", if the measured current is Ib(typ=10uA) or less, can the device measure correctly?
If “ Ib(+)=Ib(-)= typ 10uA” , Ivss < (Ib(+)+Ib(-) is inconsistent.
Can this measurement circuit measure properly?
Does Ib (input biase current) change depending on the input terminal voltage?
Hello Toshimichi,
Welcome to the forum and thank you for posting this question.
The diagrams you have drawn are basically correct (I will note that there is a 13.46kOhm differential input resistance). Also, the IB will absolutely change with input common-mode voltage (see Figure 12 of datasheet to determine common-mode IB).
The INA226 can still report measurements when measuring uA of current; however, there will be significant offset error. There will be additional offset error due to the tolerance mismatch of the 10-Ohm resistors as well.
Overall while there is significant error, you can calibrate out the offset error with one measurement since the measurements will still be linear.
Here is an analysis using simulation. I would recommend removing 49.9-Ohm protection resistor. I don't a benefit for this resistor so I ignored it for the analysis. As you can see with the three curves, the measurements you would get from option 1 or 2 are offset from the ideal shunt voltage. All three curves have the same slope (or gain), thus if you can correct the measurements with an offset in the firmware, then you can get accurate measurements. Note that when VCM = 0V, the IB drops to somewhere between 0uA and -5uA. I chose -3uA. However, the actual value does not matter so much because the error from it will be calibrated out.
Sincerely,
Peter
