• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 82 subscribers
  • Views 644 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.

TMS320F28062: ADC Offset Calibration

Antonio Fadhel
Genius 9410 points
Part Number: TMS320F28062


Hi Team,

My customer traced back an issue in their system to the ADC offset calibration of their TMS320F28062:

"According to a few days of investigation this can be tracked down to the offset calibration of the ADC – both from the factory and obtained through re-calibration function AdcOffsetSelfCal().

Specifically, on a unit with poor offset the value stored in ADCOFFTRIM was 17 whereas a unit with good offset the value was 1.

I changed the offset routine to use an ADC pin with a known 3.0V Precision Reference and it seems to greatly reduce the variance in two “identical” units.

Is there a problem calibrating offset using VREFLO and pin B5 as in the examples in the technical manual?"

I would not expect this to be an issue with the methodology (VREFLO to pin B5) and suspect ground noise, but they have not verified this yet. Any ideas on what else could be causing this?:

"I have not checked if our ground is noisy or not but we achieve high offset even when being supplied clean benchtop 3.3V and no inverting switching going on.

We do notice variation in the 3.3V regulator we use to power our micro. We are using the internal VREG of the TI so we choose to feed 3.3V into VDDA, VDD3VFL, VDDIO, and VDD pins.

Additionally, we are using the internal 3.3V reference for the ADC.

If our external 3.3V regulator used to power the chip has variation between boards, will this change the internal 3.3V reference supplied to the ADC? Maybe this is what is causing the difference in offsets?"

Thanks,

Antonio

  • 0
    TI__Guru* 81560 points

    Antonio,

    So long as the external voltage supply is within spec, it shouldn't be impacting the ADC in the manner the customer is observing.  Using the internal connection to VREFLO on ADCINB5 is the preferred method to calibrate out the offset of the device, and is the same manner we would use in the factory environment to populate the default value that gets loaded to ADCOFFTRIM on boot.

    Just to confirm, the value stored in OTP and the value returned when the customer runs the re-cal are both 17?

    I'd like to ask the customer to load the ADCOFFTRIM register with 97 and manually sample the VREFLO connection on ADCINB5 and report back the conversion.(I think this is done in the re-calibration code as well, so if they want to capture the ADCRESULT from that intermediate step that is fine as well).

    Can we also get the relevant ADC settings, i.e. ADC Clock(or just give me the system clock and all the clock DIV values) and ACQPS I can understand if there is anything there as well.

    If possible could customer connect VREFLO externally to a different ADC channel and sample that to see if there is any difference in the value seen when using the internal connection?

    Best,

    Matthew