• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 920 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.

ADC12DJ3200: challenges with background calibration

Vincenzo Pizzolante
Mastermind 22776 points
Part Number: ADC12DJ3200

hi all,

a customer of mine is using the ADC12DJ3200 and is facing some challenges with the background calibration.

They have used background calibration and it was creating DC steps at the signal (when converters were changing). This degrades the performance. By disabling background calibration they noticed better performance even when operating at our full temperature range (-40degC to 55degC).

From the behavior of the device in tests up to now, background calibration is not a must.

The questions are:

  1. how to decide enabling background calibration? For example if temperature changes by 20degC?
  2. Would you consider adequate enabling background calibration for a few seconds and then disabling it again until the event we decide occurs?

 thanks a lot in advance

KR

Vincenzo

  • 0
    TI__Genius 12780 points
    Hi Vincenzo,

    Your question is being reviewed, and we will be back with you soon.

    Best Regards,

    Dan
  • 0 in reply to dBrock
    TI__Mastermind 43930 points

    Hi Vincenzo

    Your customer is correct. The mechanisms used to implement the background calibration mode do result in small DC shifts when the active cores swap.

    Background calibration mode is recommended for applications requiring continuous up-time, for example communication links. These applications are usually less sensitive to the effects of ADC core swapping.

    Foreground calibration is recommended for applications that do not require continuous up time, and for applications that are sensitive to the  swapping of ADC cores and minor changes in DC gain/offset that can occur with continuous background calibration.

    Figures 10, 11, 12 and 13 of the ADC12DJ3200 datasheet help show the performance differences for Foreground and Background calibration. Figure 10 shows the performance differences between:

    1. Background Calibration
    2. Foreground Calibration with calibration done at each temperature point.

    Figures 11, 12 and 13 show performance with:

    1. Foreground Calibration at each temperature point
    2. Foreground Calibration at 25C ambient only

    These last 3 figures help illustrate the degradation in performance caused by temperature shifts if the device is not re-calibrated.

    I hope this is helpful.

    Best regards,

    Jim B

  • 0 in reply to Jim Brinkhurst1
    TI__Mastermind 22776 points
    hi Jim

    thanks a lot for the reply.
    Customer's application is a point to point communication link. They are impacted by these DC steps,/changes so from what I understood, they should use the temperature as event and/or use some periodical background calibration task. Would this be appropriate?
    or would you recommend not using the Backg. Calib. at all?

    thanks a lot in advance
    KR
    Vincenzo
  • 0 in reply to Vincenzo Pizzolante
    TI__Mastermind 43930 points

    Hi Vincenzo

    The most straightforward method is to use Foreground Calibration and re-initiate a calibration only if the temperature has changed enough to start impacting system performance, but only if there is system downtime that allows the ADC output to be muted (all 0s) momentarily.

    Can you get some information regarding the expected ambient temperature variation of the ADC board, the availability of downtime for re-calibration and the required ADC performance parameters? (sample rate, input frequency range, minimum SNR, SFDR, THD performance needed, etc.)

    Best regards,

    Jim B

  • 0 in reply to Jim Brinkhurst1
    TI__Mastermind 22776 points

    hi Jim,

    thanks a lot.

    Customer's app operates in a range of -40oC to 55oC with no downtime at all.

    So for sure they will stay with a foreground calibration, but several months after they may have a totally different environment.

    Sample Rate = 2800MHz

    Input Frequency Range = 0 - 750MHz

    Min SNR = subject of discussion.

    If we consider a on board loop to be tested in an oven at full temp range, we must be above 43dB at 4 QAM (we are about 45dB+/-1dB) and 41dB at 256QAM (we are 43dB+/-1dB). From figures 10-13 we can have about 48dB with no calibration at all after initial foreground calibration. Thus it may be OK for to leave ADC as initialized.

    Customer already did some test in oven and results have shown that there is no significant change in SNR due to temperature, but this has not been tested with a large number of systems.

    I really want to thank you for your valuable support

    KR

    Vincenzo