• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 211 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.

DDC316: Signal noise due to Range Error Match between side A and side B of the same input

Bernhard Gaus
Prodigy 50 points
Part Number: DDC316

We did some intensive tests with the DDC316 and noticed a signal noise due to Range Error Match (0.2%FSR typ., 1%FRS max.) between side A integrator and side B integrator of the same input.

This meas with RANGE[1:0]=11, max. input charge 12pC, we have a signal noise of 131LSB@0.2% or 655LSB@1%.

Is there any solution to eliminate this signal noise?

Would it be an option to calculate the averge of to sample (one side A and one side B sample)?

Thanks in advance.

Bernhard

  • 0
    Prodigy 50 points

    Is there any update!

    We did some intensive tests with the DDC316 and noticed a signal noise due to Range Error Match (0.2%FSR typ., 1%FRS max.) between side A integrator and side B integrator of the same input.

    This means with RANGE[1:0]=11, max. input charge 12pC, we have a signal noise of 131LSB@0.2% or 655LSB@1%.

    Is there any solution to eliminate this signal noise?

    Would it be an option to calculate the averge of two sample (one side A and one side B sample)?

    Thanks in advance.

    Bernhard

  • 0 in reply to Bernhard Gaus
    TI__Expert 4630 points

    Hi Bernard,

    You are right, you have to correct side A and B separately and then put them together.

    In more detail for any other reading this post, even and odd samples are taken with two different integrators, each with their offset and gain errors. Because of that, even samples and odd samples don't "line up" beyond a certain accuracy. The normal procedure is to run the device without input signal (left floating), and measure/record the average of even samples and the average of odd samples, separately. Then apply a signal and again do the same (measure the average to extract the gain error). Then apply those correction factors to each of the two streams (even/odd) in the real time data. One can now mix them.

    With that, one can compute the noise of the whole sequence. Otherwise, one is not measuring the noise but the back and forth jumps due to offset/gain. Alternatively, one could measure the noise of the two sides separately. They should match... (if noise is all what one would like to measure).

    Regards,
    Edu