INA3221 compression with low currents

Hello Giulio,

Thanks for considering to use Texas Instruments.  When you say shunt unbalance, are you referring to the input bias currents?  According to figure 12 of the datasheet, that will matter more for higher common mode voltages, especially if you have large tolerance filter resistors between the shunt and input pins.

As for the low output swing. This is not an analog output device, but rather an I2C output, so you will not have a low output swing limitation (or at least you will not have a limitation relating to shunt, but rather your communication lines).  However there is an input swing limitation.  Or a minimum current level or sense voltage that can be resolved.  The dominant factors for this minimum resolvable current are the shunt LSB size (40uV) and the input referred offset votlage.  It turns out in both of your cases that the offset voltage (≤80uV) actually may be higher than the LSB step size (40uV typ).  Below are calculations I used to determine the min current requirement imposed by either requirement. For offset I assume you can accept 10% error, which would be found at the lowest measurement and would only get better with increasing current. Offset Error (%/100) = Vos/Vsense_min=Vos/(Rshunt x Imin).

Table 1: Imin for 10% error from offset, assuming worst offset

Table 2: Imin according to LSB, assuming no sources of error present

Hello Giulio,

I spoke with one of our designers and he confirmed this a different architecture than the INA301 and he said that as it leverages an ADC for the measurement you will not have the same swing to ground limitation you would have with our typical amplifier current shunt monitors.  He also confirmed that Vos will be the limiting factor for the minimum current level that can be measured.  As I stated above Vos is influenced by CMRR, so you will need to include that in the calculation.