• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 282 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.

AM263P4: Can the ADC_CAL pins be used as 2 extra analog input readings

Theo Kachelski
Prodigy 95 points
Part Number: AM263P4

Tool/software:

Hi,

I understand that the ADC_CAL pins (0 and 1) are connected as channel 6 and channel 7 for every internal ADC Mux.  My assumption is that, in theory, you should connect 2 known constant voltages to these pins to then be able to calibrate out differences between the internal ADC muxes. This is evident by the table included below taken from the technical reference.  

My questions are:

Can they be used as regular single-ended analog inputs? 

Can I connect a variable signal to the ADC_CAL0 pin and just read it on every adc mux?

Is the raw ADC count measurement for the ADC_CAL pins available to be fetched from the ADC peripheral? 

Is there any auto-calibration that occurs on the ADC front end to try to automatically calibrate out the difference in the ADC_CAL measurement between cores that could be affected by a changing signal on the ADC_CAL pin?

  • 0
    TI__Guru*** 135355 points

    Hello Theo,

    Theo Kachelski said:
    Can they be used as regular single-ended analog inputs? 

    No they cannot.

    Theo Kachelski said:
    Can I connect a variable signal to the ADC_CAL0 pin and just read it on every adc mux?

    No, the internal routing of the ADC_CAL signals doesn't enable this, they are designed to be used as part of the ADC conversion process to guarantee accuracy. 

    Theo Kachelski said:
    Is the raw ADC count measurement for the ADC_CAL pins available to be fetched from the ADC peripheral? 

    Not to my knowledge, I am not aware of this being stored in a register that can be read - but I will look to verify.

    Theo Kachelski said:
    Is there any auto-calibration that occurs on the ADC front end to try to automatically calibrate out the difference in the ADC_CAL measurement between cores that could be affected by a changing signal on the ADC_CAL pin?

    I'll have to ask about this with a design expert, let me get back to you later this week.

    Best Regards,

    Ralph Jacobi

  • 0 in reply to Ralph Jacobi
    Prodigy 95 points

    Ok Interesting... just to make sure you are referring to the ADC_CAL pins not the REFHI/REFLO pins.  This below screenshot seems to suggest that the ADCCAL pins are part of the same mux used to sample the analog inputs as opposed to the REFHI/REFLO pins that are circled in yellow. 

    I do not understand how, without any autocalibration in the HW itself, these ADC_CAL pins can be used if the raw ADC value is not stored in an accessable register.  Can you refer to me something to explain how these would work if the value is not accessible.

  • 0 in reply to Ralph Jacobi
    TI__Guru*** 135355 points

    Hello Theo,

    I'm waiting to get more details from our design experts so I can present a comprehensive summary. I expect to have that feedback by end of the day tomorrow.

    Best Regards,

    Ralph Jacobi

  • 0 in reply to Ralph Jacobi
    TI__Guru*** 135355 points

    Hi Theo,

    I didn't get the feedback I was expecting today, I'm pushing to get them tomorrow.

    Best Regards,

    Ralph Jacobi

  • 0 in reply to Theo Kachelski
    TI__Guru*** 135355 points

    Hello Theo,

    Okay so I got better educated on the exact structure of this ADC and one area that tripped me up is that what we call an ADC_CAL pin on this device isn't named properly at all. It's not for calibration in the sense you and I would think of it. It's not connected to any calibration circuitry but is instead used for an unrelated purposes that are TI internal details.

    Now then as for how the internal connections are actually routed, there is a little good and a lot of bad from your perspective. On one hand, our designer does believe that they could be selected through the mux. However, unlike the actual ADC channels we are advertising, the ADC_CAL pins are connected across many ADCs which means a lot of degradation in performance. There is also absolutely zero testing done on such a configuration and no data at all about what functionality remains if used in that manner.

    Therefore it is not a use case that is considered valid for this device and any usage of the ADC_CAL pins in such a manner is strongly not recommended. If you wish to do so at your own risk, there won't be any support from TI in terms of configuration, debug, or performance analysis.

    If you have any further questions or clarifications, I can try and answer based on the discussion I had and what I learned through it.

    Best Regards,

    Ralph Jacobi