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.

  • Resolved

INA225-Q1: Sensing 0.085mA and Detecting no load(0mA)

Guru 20610 points

Replies: 11

Views: 2771

Part Number: INA225-Q1

Hi,

My customer requires follows;

・Sensing 0.085mA~8mA

・Requiring detection of no load(0mA)

・ADC is 10bit, VDD=3.5V.

I think that INA225-Q1 can achieve above requirement if Max. error[0.085mA] - Min. error[0mA]  is larger than accuracy of ADC(3.42mV) as attached file.

Current Sense Amp.xlsx

Is my understanding correct?

Please let me know if you have any question.

Best Regards,

Kuramochi

  • Dear Kuramochi,

    I am looking at this with my team right now and will get back to you with a recommendation. Thank you for all of your information.

    What is the common-mode voltage (Vcm) of your system? Basically since this is a high-side application, what is the voltage at the IN+ pin? This will affect your Ib and could make your system more accurate if it is lower than 12V since the 85uA Max Ib spec was determined when the Vcm = 12V.

    Also, does this device need to be Automotive Qualified? We do have parts with lower Ib such as: INA138/168, INA216, INA270, INA199-Q1, INA210, but not all of these will work with detecting your differential voltage and Vcm requirements. If you would like to read up on sensing low differential voltages (<20mV) you can visit the following post on our FAQ page.

    Best,

    Peter Iliya

    Current Sense Applications

  • In reply to Peter Iliya:

    Kuramochi,

    The INA225-Q1 will not work for your scenario of measuring 0A, 85uA, and 8mA simply because the input bias current can be max +85uA. This means that at worst, the INA225 will siphon or sink 85uA of current into the IN+ pin. If you ILOAD is 85uA, then the current passing through your shunt resistor effectively becomes zero.

    Additionally, once your system opens and the ILoad = 0A, then your Vout will not be 0.338V, but actually 0.305V. I calculated this with the following equation:

    Vout = [ Ios*Rsh*1.01 + Vos_max + dVos/dT*100degC  ] * gain*gain error  + Vref

    where Ios = difference of Ib+ and Ib-, gain = 25V/V, gain error = 0.0015, Vref = 0.3V, Rsh = 15Ohm, Vos_max = 150uV, dVos/dT = 0.5uV/degC. I also took into account the 1% tolerance of the shunt resistor.

    If you would like to read about the theory behind this, other current shunt monitors with lower Ib, and other alternative low-current solutions you can go here:

    Peter Iliya

    Current Sense Applications

  • In reply to Peter Iliya:

    Peter-san,

    Thank you for your help.

    >The INA225-Q1 will not work for your scenario of measuring 0A, 85uA, and 8A simply because the input bias current can be max +85uA.
    >This means that at worst, the INA225 will siphon or sink 85uA of current into the IN+ pin.
    >If you ILOAD is 85uA, then the current passing through your shunt resistor effectively becomes zero.

    I think that load current 85uA can be sensed if bias current typical value is supposed when designing schematic.
    What do you think?

    >Additionally, once your system opens and the ILoad = 0A, then your Vout will not be 0.338V, but actually 0.305V.
    > I calculated this with the following equation:
    >Vout = [ Ios*Rsh*1.01 + Vos_max + dVos/dT*100degC ] * gain*gain error + Vref

    Should you consider not only Ios but also Ibias?

    Best Regards,
    Kuramochi
  • In reply to Kuramochi Tadahiko:

    Kuramochi-san,

    When designing these systems you want to make them as robust as possible. So even though they typical values in datasheet may work, it is best to design according to the maximum and minimum values as you can be most confident (99.7%, 6-sigma) that the INA225-Q1 will not have Ib outside of the maximum and minimum specs.

    For the Vout equation when ILOAD = 0A, I am considering Ibias because Ios = Ibias+ - Ibias-. Instead of writing Ios, I could have written (-Ibias+ - Ibias-), but this will end up being the difference of the two currents anyway,

    Peter Iliya
    Current Sense Applications
  • In reply to Peter Iliya:

    Peter-san,

    Thank you for your advice.

    >When designing these systems you want to make them as robust as possible. So even though they typical values in datasheet may work, it is best to design >according to the maximum and minimum values as you can be most confident (99.7%, 6-sigma) that the INA225-Q1 will not have Ib outside of the maximum and >minimum specs.

    I understood.

    >For the Vout equation when ILOAD = 0A, I am considering Ibias because Ios = Ibias+ - Ibias-.

    >Instead of writing Ios, I could have written (-Ibias+ - Ibias-), but this will end up being the difference of the two currents anyway,

    Sorry, I can't understand what you say.

    I'm understanding flow of current as below, is this misunderstanding?

    (And I'm understanding that we can neglect Ioffset because it is much smaller than Ibias.)

    Best Regards,

    Kuramochi

  • In reply to Kuramochi Tadahiko:

    Kuramochi-san,

    I think it is best to generally understand current flow according to this picture:

    • There are actually two input bias currents: Ibias+ (Ib+) and Ibias- (Ib-).
    • These input bias currents are modeled as positive current sources into the IN+ and IN- inputs of the device. This is how input bias current is modeled not only current shunt monitors, but also op-amps, instrumentation amplifiers, transimpedance amplifiers and such. Thus, if Ib positive then it is flowing into the amplifier input and if it is negative it will be flowing out.
    • Ios = input offset current = Ib+ - Ib-. It is simply the difference of the two input bias currents, thus it is an indicator of how well matched the internal resistors inside the part are.

    Keep in mind that if the the Vbus opens up and the source current goes to 0A, you will have the following situation:

    Vout = [ Ib+*Rsh*1.01 + Vos_max + dVos/dT*100degC ] * gain*gain error + Vref = 0.338A and this is greater than the 0.305V you may see when ILOAD = 85uA.

    Peter Iliya

    Current Sense Applications

  • In reply to Peter Iliya:

    Peter-san,

    Thank you for your explanation!

    I understood.

    I suppose to lost not supply but load on this application as follows.

    So I calculate shunt current as not Ioffset but Ibias-(≒Ibias).

    Best Regards,

    Kuramochi

  • In reply to Kuramochi Tadahiko:

    Kuramochi-san,

    You are correct. If the Load is removed, then the current flowing across Rshunt will be Ib-. Although you still will have the problem of not being able to determine the difference between 85uA and 0A at least not with a robust solution. You will need to choose a part with lower max Ib.

    Peter Iliya

  • In reply to Peter Iliya:

    Kuramochi-san,

    Have you solved your issue? The INA226-Q1 (10uA Ib typical) and INA220-Q1 (20uA Ib typical) would be your best options to satisfy your original requirements.

    You can additionally find low current measurement (10uA - 100mA) designs with non-current shunt monitors here:
    http://www.ti.com/tool/TIPD104
    http://www.ti.com/tool/TIPD156
    http://www.ti.com/tool/TIPD129

    Best,

    Peter Iliya

    Current Sense Applications

  • In reply to Peter Iliya:

    Hi Peter-san,

    I'm sorry for late response.

    Yes, my customer decided to evaluate INA225-Q1.

    Though IN226-Q1 and INA220-Q1 has low bias current surely, they don't have guaranteed value(max/min).
    So I think customer can not choose them.

    Best Regards,
    Kuramochi

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.