• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 48 subscribers
  • Views 602 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.

INA219: Variation is large for current more than 300mA

Satheesh Velu
Prodigy 20 points
Part Number: INA219

Hi TI Support Team,

We are using INA219 in our design.

  • Python Modules to talk to INA219 via FTDI chip
  • Max Current that can flow through : 3A
  • Shunt Resistor : 10mOhm

Calibration Procedure:

  • Using Electronic Load, applied the load at the INA219 and calibrated.
  • INA219 measured valued is cross checked with the Current Probe

The deviation is large. over 200mA when the load changes in real time. Kindly please advise how this can solved.

Thanks,

Satheesh

  • 0
    TI__Mastermind 26335 points
    Hello Satheesh,

    Thanks for considering to use Texas Instruments. When you say: "The deviation is large. over 200mA when the load changes in real time." Do you mean that if the load is fixed, then the INA219 measurement is close to what is expected? Or is it always off by 200mA? If it is only innacurate for a varying load, how fast is your load changing? If it is always inaccurate could you provide some details on the calibration value used, as well as the gain setting and ADC setting you used?
  • 0 in reply to Patrick Simmons
    TI__Mastermind 26335 points
    Hello Satheesh,

    Can you help clarify your issue by answering my questions above or provide me with an update of your debug status?
  • 0 in reply to Patrick Simmons
    TI__Mastermind 27715 points
    Hello Satheesh,

    It's been some time since your last post, so hopefully your issue is resolved. I will close the thread for now. If you need more assistance, please reply or start a new thread to get help.

    Best regards,

    Ian Williams
    Applications Manager
    Current & Magnetic Sensing
  • 0 in reply to Patrick Simmons
    Prodigy 20 points

    Hi Patrick,

    Apologies for my delayed reply. We use Electronic load in our calibration.

    • INA219 is given E-load from 50 to 1.5A in 50mA steps
    • The resulting value is Plotted [Actual Load vs INA219 Measured]
    • The Plot is linear. An equation is used in curve fit to get accurate results

    In the above calibration say I'm applying the E-Load at Point A and do the calibration. In real time Load is at a different location of the circuit point B. The resulting value has variation.

    Do you mean that if the load is fixed, then the INA219 measurement is close to what is expected? Or is it always off by 200mA?

    The load variation is not always off by 200mA. I will send you the variation chart.

    If it is only innacurate for a varying load, how fast is your load changing?

    My board has 4 INA219s. 3 out of 4 are giving near perfect result with the above calibration method I mentioned. So I believe the speed of load change may not have an impact. In fact the load changes rapidly only during the calibration using E-Load. In real time the load does not change much.

    If it is always inaccurate could you provide some details on the calibration value used, as well as the gain setting and ADC setting you used?

    I will send you more of the details in a while.

    Thanks,

    Satheesh

  • 0 in reply to Satheesh Velu
    TI__Mastermind 26335 points
    Hello Satheesh,

    Are you performing high-side measurements with the 1 INA219 that has a large measurement error? If so, I have a possible hypothesis: there is a parasitic path from the bus siphoning off current to something further down stream. I would suspect this in the form of solder bridging from the bus to something else. Is this something you think you might be able to check for? I only suspect this as you said you have 3 other INA219s working fine.

    Aside from the above test, were you able to put together that variation chart?