• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 45 subscribers
  • Views 588 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.

Extremely high precision analog isolation

Richard Denny
Prodigy 10 points
Other Parts Discussed in Thread: ISO124

We have an application that requires signal isolation and an output that is within 0.02% of the input throughout the range -10.24V to +10.24V. Our current design, using the ISO124 isolation amplifier, accommodates only a typical 0.05% gain error, for which we compensate using zero and span adjustments. However, the new specification requires the specified accuracy with NO adjustments. We have considered a digital solution using an ADC- optoisolator-DAC combination, but still are not sure that we can achieve the required accuracy. Has anyone had success with an ADC/PWM/DAC solution?

  • 0
    TI__Mastermind 23815 points

    Hello Richard,

    welcome to our e2e forum and thanks a lot for your question.

    To be frank, trying to achieve an accuracy or gain error of 0.02% without any sort of offset and gain calibration for any kind of signal chain will be extremely challenging, not to say unrealistic.
    It already starts with the initial accuracy of the voltage reference. The best voltage references I know of specify an initial error of 0.025% (max) at Ta=25°C. That does not even include the drift over temperature. The gain error of the ADC and DAC will come on top which is usually also on the order of 0.05% to 0.1% at Ta=25°C.

    Even with calibration it will be challenging enough to achieve a 0.02% (max) accuracy over temperature.

    Regards,
    Joachim Wuerker

  • 0 in reply to Joachim Wuerker
    Prodigy 10 points

    Thank you for your reply. That is the same position that we tried to take with the client. We are hitting a wall and will try to convince them that they need to accept a one-time calibration to their specification on a test bench, with no guarantee that it will meet their specification after installation.

  • 0 in reply to Richard Denny
    TI__Mastermind 23815 points

    Hi Richard,

    you are welcome.
    As of today, this is unfortunately the situation we are in. Even achieving 0.1% accuracy without any calibration would be challenging already with the parts available on the market.

    Regards,
    Joachim Wuerker