• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 2 answers
  • Subscribers 47 subscribers
  • Views 335 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

INA296B-Q1: Output Set Using Resistor Divider

Akihiro Hirata
Expert 5770 points
Part Number: INA296B-Q1

Hello team,

In the D/S of INA296B-Q1, there are below statement highlighted in yellow.
Could you explain more clearly why TI doesn't recommend a single-ended measurement with this configuration?

1. 

Regards,

Hirata

  • +1
    TI__Mastermind 29630 points

    Hello Hirata,

    The reason is due to the reference voltage moving due to impedance and load changes on the reference pin.  Without a buffer the reference voltage will change.  If you would like to measure it single ended please use a buffer like that described in the figure below. 

  • 0 in reply to Javier Contreras4
    TI__Expert 5770 points

    Hello Javier-san,

    Thanks for your asnwer.

    With this resistor divider configuration without OPAMP, Does't this configuration affect gain error, CMRR etc...?

    Regards,

    Hirata

  • 0 in reply to Akihiro Hirata
    TI__Mastermind 29630 points

    Hello Akihiro,

    Yes the configuration without OPAMP does affect gain error as the load on the Ref pin changes with both the Vsense and VCM at the input but the VCM change has minimal impact.  This will impact gain error more.

    Regards,

    Javier

  • 0 in reply to Javier Contreras4
    TI__Expert 5770 points

    Hello Javier-san,

    1. In case of without OPAMP, Do you have any recommendation for the resistance range of the resistor divider? With that recommended case, how much is the spec/accuracy impacted?

    2. Can I think adding OPAMP with resistor divider is much simpler and we recommend the solution with OPAMP?

    Regards,

    Hirata

  • 0 in reply to Akihiro Hirata
    TI__Mastermind 29630 points

    Hello Akihiro-san,

    1.  We can recommend a resistor divider but I would still recommend measuring differentially if it is done.  The Ref pin pulls around 20µA and there is a delta of about +/- 2µA varying with Vsense.  This then becomes a power tradeoff on the R-divider and changes with current.  Also the 20µA will vary from part to part.

    2.  With an opamp driving the input pin will give the the flexibility of knowing the Voltage and being able to measure single ended with not the large error introduced by the input bias current of the Vref pin.  I would recommend this solution.

    Regards,

    Javier 

  • 0 in reply to Javier Contreras4
    TI__Expert 5770 points

    Hello Javier-san,

    1-1. I think you are talking about the reference point (= output voltage at Vsense=0V), but I was asking the gain error especially. Reference point can be calibrated/compensated by the differential measurement, but Gain error cannot be calibrated/compensated by the differential measurement, is this correct understanding?

    1-2. If so, how much the gain error is introduced when the resistance of the resistor divider is what?

    Regards,

    Hirata

  • +1 in reply to Akihiro Hirata
    TI__Mastermind 29630 points

    Hello Hirata-san,

    1)The gain error will be compensated for with the differential measurement.  Making a differential measurement with a resistor divider at the Vref pin will compensate for the Vref chaining due to the loading on the Resistor divider.

    2)Basically there is an input current on the Vref pin ~20µA  that will load your divider and this current changes with input voltage(Vsense)  This load change causes the gain error to change if making a single ended measurement as the Vref voltage will change.  If a differential measurement is not being done I would recommend a buffer.

    Regards,

    Javier

  • 0 in reply to Javier Contreras4
    TI__Expert 5770 points

    Hello Javier-san,

    Oh, I didn't know that gain error can be compensated by differential measurement!

    Thanks for your answer.

    Regards,
    Hirata