• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 623 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.

Need Information -> INA226

Edw
Prodigy 40 points
Other Parts Discussed in Thread: INA226, INA226EVM

Hi,

I would like to ask regarding INA226 (http://www.ti.com/product/ina226&lpos=Middle_Container&lid=Alternative_Devices). I refer to specification - SBOS547 JUNE 2011 page 21 and 22. If there is a math overflow in Current Register (Address 0x04). What will be the value? Will this register always be having 32767 (0x7FFF) if it is above 32767? Will it always be having -32768 (0x8000) if it is below -32768?

I want to understand whether the value will be cap and fixed value will always be represented when there is math overflow.

So far, I have tested with TI INA226 that I have and the Current Register will always give me -32768 if it is below -32768. I would like to confirm the behaviour.

Thanks.

  • 0
    TI__Mastermind 25825 points

    Hello Edward,

    Sorry for the delay.

    The best way to detect for overflow is to monitor the OVF bit.  When there is math overflow, the OVF bit will go high.  Therefore the contents of the current and power registers may be invalid.  Can you please explain why you're interested in the contents of a register that is overflowed?  Details concerning your application and what this behavior represents in your system would be helpful so we can further our understanding of the problem you're trying to solve.

  • 0 in reply to Pete%20Semig
    Prodigy 40 points

    Hi Pete Semig,

    Thanks for the reply.

    When OVF bit is 1, how can I know whether this is an overflow in Current or Power and whether it hits the maximum or minimum?

    For current register, I have checked and found that Current Register will always give me -32768 if it overflows below -32768.

    To me, this is still logical because it is capped at the minimum. It is ok as long as during overflow (<-32768), Current Register does not give me value like 0, 1000, 32767 or any value that is far away from -32768.

    Thanks.

  • 0 in reply to Edw
    TI__Mastermind 25825 points

    Hello Edward,

    The OVF bit is set upon overflow in the current register.  Since the power register uses the current register for its calculation, the value in the power register will be inaccurate even though it may look reasonable by not hitting a max/min value. 

    I'm still not clear on why you're interested in the current and power register contents after an overflow has occurred.  The contents will not be reliable.  The solution is to properly determine the calibration register contents.  Details of how to set the calibration register can be found in the datasheet.  Additionally, you may download the EVM software and use it to quickly determine the calibration register contents based on your system.

     

     

  • 0 in reply to Pete%20Semig
    Prodigy 40 points

    Hello Pete Semig,

    Thanks for the reply.

    I have already adjusted the calibration register for normal application. This application continuously uses Current for some computation. However, under certain condition, there is a very short period of high inrush current that exceeds the range of current adjusted at INA226. I do not want to adjust the calibration to have a bigger range to cover all sort of possible short period high inrush as this affects the resolution and accuracy. So, for this kind of situation, where overflow occurs, I would like to continue to use the value of Current Register for computation to reduce inaccuracy in the computation. It is OK as long as it presents a logically capped value. I cannot change computation solely based on overflow bit because it does not tell whether it is above maximum or below minimum or at which register. Hope this explains clearer.

  • 0 in reply to Edw
    TI__Mastermind 25825 points

    Hello Edward,

    Thanks for the additional information.  Using an INA226EVM I observed that the shunt voltage register stops incrementing at 0x7FFF and decrementing at 0x8000.  Since the current register contains the product of the shunt voltage register and the calibration register, I would expect the current register to behave in a similar manner.  This appears to be consistent with your experiments.

    Please let me know if you have any further questions.

  • 0 in reply to Pete%20Semig
    Prodigy 40 points

    Hi Pete Semig,

    Thank you.