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.

ADS131A04: ADS131A04 dynamic range question

Part Number: ADS131A04

Hi Team,

 

Here is South China FAE Sunny, we saw that ADS131A04 can support 0.1% accuracy at 1:10000 dynamic range, that means the dynamic range of ADS131A04 should be better than 140dB(0.1% need 60dB at lease, while 1:10000 is 80dB). But from datasheet we know the best dynamic range of ADS131A04 is 119.49dB(1kHz sampling rate). So do you have some comments on why we say ADS131A04 can support 0.1% accuracy at 1:10000 dynamic range?

Thanks very much.

 

Regards, Sunny

  • Hello Sunny,

    The dynamic range specification is a ratio that uses the full scale range and noise of the device and is not a function of accuracy as shown in section 8.1 of the datasheet. As you have stated, this range can vary greatly based off the configuration of the device with a maximum range of 124.55dB as shown on Table 3. The issue with the calculation of 140dB, however, is the conversion of 0.1% to 60dB of dynamic range. The "0.1%" is a measure of the error of the device when the dynamic range is at 80dB, this feature is not meant for calculation of the max dynamic range of the device. If you would like greater clarification, I have linked an excellent blog series below that focuses on the differences of ADC accuracy vs. resolution.

    e2e.ti.com/.../adc-accuracy-part-1-is-accuracy-different-from-resolution

    e2e.ti.com/.../adc-accuracy-part-2-total-unadjusted-error-explained

    Best,
    Hasan Babiker
  • Hi Hasan,

    Thanks for your reply, yes I know that resolution is different from accuracy, but my point is if we need to measure signal at 0.1%, we need get an dynamic range at least 60dB, because only get 60dB dynamic range, we can see 0.1% change of signal. I think this is premise of accuracy, I know we still need good offset/INL/gain error and etc.
    So it even make me confused, if we need get 0.1% as 80dB whole rage, let make 1V full scale as example, that means we need to get 0.1% accuracy at 100uV (-80dB to 1V)input, so first we need to get 0.1uV change firstly, that means noise should below 0.1uV, even got this resolution can`t guarantee the accuracy, but I think we need to get that resolution firstly, so that why I say it need 140dB dynamic range.
    So Do you have some more comment on my understanding?
    Thanks again.

    Regards, Sunny
  • Sunny,



    I think that Hasan has explained it correctly, but let me explain with the same example you use.

    If you have a full scale of ±1V (assuming you have a 1V reference), then the error that you may see is 1mV because the gain error is 0.1%. With a dynamic range, you're measuring the noise versus the full scale. This dynamic range is 10000:1 and this is the noise compared to the full scale. For that dynamic range, it comes from the equation 2 in the datasheet (without the 20logx). That would be 71uV with a gain of 1.

    In the end, the device could perform better than that. Depending on the operation, the dynamic range is better for most modes of operation. In the end, "-Better Than 0.1% at 10,000:1 Dynamic Range" just means that the noise error will be smaller than the gain error.



    Joseph Wu
  • Hi Joseph,

    thanks for your comments, but I still has some concern, i will double check with customer, and give you feedback later.

    Thanks.

    Regards, Sunny