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.

ADS5401 interleaving correction

Other Parts Discussed in Thread: ADS5407, ADS5409, ADS5401, ADS5404

I'm looking for more information on the ADS5401 interleaving correction.  I've read through the ADS5401 data sheet (SLAS946A - APRIL 2013 - REVISED JANUARY 2014) as well as the data sheets of similar chips (ADS5404, ADS5407, ADS5409), but they don't provide any information on how this correction is performed.  I'm currently evaluating this chip for use in a large-scale, precision experiment, and, therefore, I need to fully understand this auto-correction behavior.  In particular, I'd like to request information on the following two topics:

  1. Is there available documentation on the interleaving correction's circuitry?  It must be some sort of feedback loop, but I'd greatly appreciate any details about how the processing is performed.  The ADS5401 data sheet (page 24) mentions that fine tuning of the algorithm is possible by TI.  Are there available data characterizing the default algorithm, and which settings are possible to be tuned?
  2. Frequency-dependent effects are extremely important for my application.  Is there a known input-signal frequency dependence with regard to this interleaving correction?  The ADS5401 data sheet (page 24) mentions that the correction "yields best performance for input frequencies below 250 MHz" which leads me to believe that frequency-dependent effects do exist.  Specifically, I deal with approximately 10-ns wide pulses which are bandwidth limited at roughly 200 MHz and which I'm digitizing with the ADS5401 at 800 MHz.  Do you expect the interleaving correction to function properly with such small pulses?  And do you think that the correction's behavior will change with a change in the frequency of these short pulses?

I noticed that there's a similar post requesting information on the ADS5409's auto-correction mode from last July, but there hasn't been any response that answers their questions.  Since my evaluation is time-sensitive, I would greatly appreciate any information that one can give off-hand even.  Thank you in advance for any advice on this topic and help to understand this interleaving correction.

  • Hi,

    The details of our correction algorithms are not something that we have been making public.  I have sent this request to the data sheet author and we are looking to see what we can create to provide for this.  At this time I do not personally know the details of the algorithms, but I believe they do employ counters to gather statistics of the samples, so that there is something of a feedback loop and something of a time constant depending on the length of the counter.  I believe we have two such correction algorithms, one to correct for gain mismatch and one to correct for offset mismatch.  There are two register bits in the register space to set or clear to enable the autocorrection because one of those bits clears the counters and the other enables the use of those counters.  I believe just about any such algorithm has the potential to behave badly on some pathological cases, but I do not know if your application is one that would cause problems for our algorithms.  But that is all I know at the moment until we can create a document containing more details. 

    Regards,

    Richard P.

  • Hi Richard,

    Are there are any updates from the data sheet author regarding documentation?  I assure you that my research group and I have no commercial interest in making ADCs; we are evaluating this ADS5401 chip for use in a precision particle physics experiment.  We are also all willing to sign a non-disclosure agreement if needed, and, in that case, perhaps it would be best for us to continue this discussion offline.

    David