• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 1499 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Offset current contribution to current shunt monitor error?

Gerrit Barrere
Expert 1245 points
Other Parts Discussed in Thread: INA199

I'm considering using an INA199A1 (gain of 50) as a high-side current monitor.  My shunt resistor voltage drop is only about 6 mV, but accuracy requirements are not high.  CM voltage is 5 V.  

I've looked through the app notes and training videos, which make the contribution of Vos to output error quite clear.  I don't see any treatment of Ios errors though.  The INA199A1 specs a max Vos of 150 uV, which would lead to an output error of 7.5 mV.  Typical is much less.  The typical Ios, however, is .02 uA, which leads to an output error of 20.0 mV through the 1 M feedback resistors in the device.  No maximum is given.

Can this be correct?  It looks like the Ios typical error is much greater than the Vos maximum error.  Please let me know if I'm misinterpreting this somehow.

Thanks,

Gerrit

  • 0
    TI__Expert 4595 points
    Hi Gerrit,
    That is not the correct. IB contribution to the error can be significant when the load current is comparable to the bias current. IOS can contribute to offset if large filter resistors are used in the input. What is the load current range that you are planning to measure?
    Please click on answer if I answered your question.
  • 0 in reply to Rabab Itarsiwala
    Expert 1245 points

    Thanks for the quick response, Rabab. The current being measured is about 200 mA.

    I don't understand why the Ios doesn't affect the output as I mentioned, though. If the input were shorted (zero volts) and the internal opamp had .02 uA offset current (with no other errors), the output would have to go to .02 uA * 1 Mohm = 20 mV to compensate. Please help me understand why that doesn't apply.

    Gerrit

  • 0 in reply to Gerrit Barrere
    TI__Expert 4595 points
    The input offset error should include all internal errors (input referred). For this case if the shunt resistor used was 30m ohm then the IB contribution to the error will be 30m ohm x 28uA = 0.84 uV (input referred).
  • 0 in reply to Gerrit Barrere
    TI__Mastermind 32290 points

    Hi Gerrit,

    The input stage of a current sense amplifier is biased by the inputs.  This input bias currents do not flow through the feedback resistors.   Figure 21 in the INA199 illustrates the bias circuitry and is seen below.  This is the reason that internal the IOS do not contribute a high error unless you add filter resistors as described in the data sheet in the INPUT FILTERING section.

    Regards,

    Javier Contreras

  • 0 in reply to Javier Contreras4
    Expert 1245 points
    Thanks, Javier and Rabab.

    Does the opamp in the figure represent an ideal one, with no error currents? In that case I can see where error currents from the Bias block would pass almost exclusively through Rshunt and have little effect on the output. In other words, the error currents in the data sheet refer to the entire INA199 and not to the internal opamp.
  • 0 in reply to Gerrit Barrere
    TI__Mastermind 32290 points
    Gerrit,

    Yes the IB and IOS in the datasheet do refer to the entire inputs and not to the internal opamp. Please keep in mind that this IB and IOS is at one point with Vsense = 0V with VCM or Vin+ = 12V. Even though IB and IOS change with Vsense and VCM change the biggest errors due from these IB changes is if a input filter resistors are added.

    Regards,
    Javier
  • 0 in reply to Javier Contreras4
    Expert 1245 points
    Thanks to both Javier and Rabab for quickly answering my questions!