This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

INA3221: Using to sense/monitor negative supply

Part Number: INA3221
Other Parts Discussed in Thread: ISO1540, TIDA-00313, INA270, INA2332


I am using the INA3221 in my circuit to sense and monitor a few supplies. One of those is a negative supply. I understand using the device with positive supplies, but was curious if this device could be used with a negative supply, and how that might be accomplished? Thank you everyone.

  • Hi Michael,

    Its more of a challenge, but it can certainly be done. I just want to clarify whenever you will be measuring a mix of positive and negative voltages or only negative.

    Assuming they are all negative, the device can be powered directly from one of the negative rails and a ISO1540 digital isolator used to perform level shifting to the MCU, residing on the positive supply side. We have a reference design, TIDA-00313 that uses this approach and might be worth looking into.

    For a mix of positive and negative supplies, I would use an inverting amplifier with a gain of -1 to level shift the required negative voltage. Naturally output voltage will be limited by the output swing of the op-amp and its supply voltage, so a rail to rail variant may be preferred.

    The diagram below shows how to achieve this.

    The gain is given by (R2/R1 * -VIn), therefore 1/1 * -(-VIn) = VIn

  • Carlos,

    Thanks for the reply. The INA3221 provides the ability to monitor not only voltage, but current as well. Both are measured with a current sense resistor between the + and - Vin rails on the device. Don't know all the internal particulars of how it does both, but that is the connection. The device states it can handle supplies from 0 to 26V, but I have a -1.2V supply to monitor both the voltage and the current. An FPGA takes that information and provides a status as a result.

    The only potential way I see of doing this for -1.2V is to use a separate voltage monitor for that function and a separate current monitor that could handle negative voltages. The INA270 appears to be able to do that (monitor current for negative supply). That seems overly complicated, so was hoping there might be some way to use the INA3221 for this purpose.

    One thought I had was reversing the side where the load is for the negative voltage, but unsure if that is doable with this device.

  • Hi Michael,

    As per my original question, can you clarify whenever the INA3221 is measuring a mix of positive and negative voltage voltages or just one type?
    The -1.2V rail can be measured with a shunt breaking the ground for the negative voltage (since that is essentially the positive of the negative voltage source). To remove any uncertainties, I can test this for you and provide the results back, but it will take me a few days to order the parts.
    Since adding a current sense resistor will affect the output voltage regulation, what is the required maximum voltage drop that your application can support?
  • Carlos,
    I am measuring 4 voltages total. +1.2V, +2.5V and +3.3V along with -1.2V. I have the other three voltages on one INA3221, so I would be having the -1.2V as the only supply on a 2nd INA3221.

    If you can do that that would be appreciated. I believe I have some time before needing to finalize my design and right now this is the only question mark about it.

    Thank you,
  • Hi Mike,

    Given your negative rail is too low to power the INA2332 I would go with the op-amp level shifting approach as discussed above.
    Isolating would be simpler, but it would add the expense of an additional rail to power the INA3221 and the ISO1540.

    Allow me a week to order the EVM and perform some measurements and I will get back to you on this issue.
  • Hi Mike,

    As promised, here are my findings:

    The INA3221 has a measuring range of +/-163.5mV (it clips beyond that) and can measure independent negative and positive values on each of its 3 channels. I am satisfied this matches the advice I provided earlier to use a ground measurement on the negative supply.

    For this test I used a differential measurement.

    For the voltage measurement, as discussed,

    • An opamp working with a gain of -1 can be used to convert the negative voltage  to positive, and then fed into the ADC. The device wont be able to report the bus voltage since it its internal ADC is connected to VIN-, which on a low side application is zero, and hence why it is marked as a high side monitor.
    • Given your application only uses one channel for the second INA3221, it is possible to take advantage of the negative reading capabilities to sample both current and voltage on the negative rail, using a suitable voltage divider to bring the 1.2V rail to the readable range of the part (160mV) on any of the two unused channels. When doing this size the value of the resistors to minimise the error introduced by the 10uA bias current on each input. I recommend the total resistance of the voltage divider not exceeding 470R.

    I hope you find this information useful. Please advise should you require any further help.