INA300: About small current sensing

Hello Tsukui,

In the meantime could you provide more schematic and system information. If 40mA is the threshold current, what will be the current typically when not in an over-current operation? Does this have to be a low-side or high-side measuring scheme?

You can also read more about the error associated with low-current measurements here in our article:
e2e.ti.com/.../3593.how-vos-and-ib-affect-small-current-measurements

Sincerely,

Peter Iliya
Current Sense Applications

Tsukui-san,

Overall the INA300 can detect this, but the design should be chosen carefully. The primary specs that should be considered for detecting 40mA are:
- Input bias current (Ib):
- Input voltage offset (Vos)
- The noise adjustment factor (NAF) of internal comparator
- The Absolute Maximum Differential Input Voltage Range (-40V to 40V).

The Ib and Vos meet the requirement depending on shunt resistor chosen.

As long as the shunt resistor chosen creates a differential input no greater than 40V or less than -40V during high current, but also generates a voltage at least greater than or equal to 5mV during the 40mA. I recommend a 5mV shunt or sense voltage because this will be your VLIMIT voltage at the LIMIT pin and this voltage should be reasonably high. Once the shunt voltage trips the 5mV limit (disregarding hysteresis) then the INA300's ALERT pin will go high and signal an overcurrent (40mA) is being sensed.

If the VLIMIT pin is too low around 1mV, then the internal amplifier inside the part will be driving its output stage very close to ground and thus operating very outside of it linear output range. If the 10us delay is chosen in the part, then you will have to account for the NAF, which is 500uV added onto your trip voltage. If your trip voltage was 1mV, then you could be tripping the threshold sporatically. This is why I recommend VLIMIT > 5mV.

Most of the specs in the datasheet were taken with the VLIMIT =10mV so a 5mV limit should be a good compromise. You could go for a higher sense voltage, but you run the risk of using a higher shunt resistor that will burn more power and require higher power dissapation ratings for times when your system has a high current, if that is the case.

Also here is a TI Design using the INA300 to detect 50mA so it should be able to detect 40mA. A lower current was not chosen since the design required measuring up to 20A in tandem with the INA225.

http://www.ti.com/tool/TIDA-00328
 
Please post back to the forum with your other questions.

Sincerely,
Peter Iliya
Current Sense Applications