• State
  • Locked Locked
  • Replies 12 replies
  • Subscribers 81 subscribers
  • Views 476 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.

TMS320F2800137: By default, these pins function as analog pins and the GPIOs are in a high-impedance state. So AIO default as analog pins, but actually we must configure the register GPHAMSEL.

Shaoxing Ke
Expert 3055 points
Part Number: TMS320F2800137
Other Parts Discussed in Thread: C2000WARE

Hi Teams,

My customer is working on a schematics based on F280013x and want to check if there is anything changed compared to F28002x. 

We found that like AIO232 must be configured as analog function, otherwise the analog sampling port will be 0.3V higher than the actual value. In addition, floating the 14 Pin can measure 3.3V high level.

(The other AIO pins also have the same high-level 3.3V phenomenon. If the code is not configured, the sampling will be 0.3v higher)

We can understand that AGPIO needs to be configured through the register GPHAMSEL, but we don't understand why AIO also needs to be configured? The default AIO in F280025 is an analog function, has the AIO port changed in F280013x? The customer needs details, otherwise the chip cannot be mass-produced F2800137. Thanks.

Like AIO232 must be configured as analog function, otherwise the analog sampling port will be 0.3V higher than the actual value. In addition, floating the 14 Pin can measure 3.3V high level. So doubt any change here？

 

From the picture below, By default, these pins function as analog pins and the GPIOs are in a high-impedance state. So AIO default as analog pins, but actually we must configure the register GPHAMSEL.

Could you please explain this difference in description TRM? Thanks.

  • 0
    TI__Mastermind 19795 points

    Hi Shaoxing,

    Thanks for your question for this. I am working with the analog experts to ask about this. Can you let me know the following information:

    1. Have any other AIOs have been observed like this from the customer? Or is the issue only with AIO232?
    2. Is this for the 64PM package, or 64VPM package? Or another package?

    Regards,

    Vince

  • 0 in reply to Vince
    TI__Expert 3055 points

    Hello Vince,

    F2800137 64PM package.

    We test the marked yellow AIO pin (10, 11,12,13,14) has this phenomenon(the sampling will be 0.3v higher), believe that all AIOs to be the same.

    Also suggest customer both configure the all AIO by using the register GPHAMSEL. Thanks.

    Is the customer urgently needing to explain the difference between the two chip AIOs? (F280025PM Vs F2800137PM)

  • 0 in reply to Shaoxing Ke
    TI__Mastermind 19795 points

    Hi Shaoxing,

      

    The VPM package is only different in that it has a VREGENZ for external supply usage. Otherwise identical to PM package.

    I am still following up with experts to figure out if there are any other items that need to be confirmed on this. We will update accordingly.

      

    Regards,

    Vince

  • 0 in reply to Vince
    TI__Mastermind 19795 points

    Hi Shaoxing,

    Team is looking into this in detail, please expect a response by Wednesday next week.

    Regards,

    Vince

  • 0 in reply to Vince
    TI__Expert 5000 points

    Shaoxing,

    The GPHAMSEL is set to '1' by default on these AIO pins.  Can you have the customer confirm one more time they are still seeing GPHAMSEL=1 in CCS even initially when the 0.3v rise is seen.  There could be some unexpected software in their system which has changed the default from 1 to 0; and this could explain this behaviour.

    Thanks,
    Jason

  • 0 in reply to Jason Whiles
    TI__Expert 3055 points

    Hello Jason,

    We confirm that GPHAMSEL is set to '1' by default, S/W software not configured in customer's code. But actually 0.3v rise on AIOs, if we disconnect the AIO pin (10, 11,12,13,14) to left pin floating, just test the AIO pin can see 3V3 always.

  • 0 in reply to Shaoxing Ke
    TI__Mastermind 19795 points

    Hi Shaoxing,

    What Jason was recommending is to please take a screenshot of the GPHAMSEL register values when the 0.3V rise is seen on the AIOs. He is saying this because another function may have accidentally set the AIOs to GPHAMSEL=0.

    Can you provide this screenshot?

      

    Please note, we are still looking into this from the design side in the meantime, but would like to have the customer provide this in parallel to save debug time.

    Regards,

    Vince

  • 0 in reply to Vince
    TI__Expert 5000 points

    Shaoxing,

    You can also have the customer check GPHPUD values.  The rise of 0.3v sounds like a weak pull up.

    Best regards,

    Jason

  • 0 in reply to Jason Whiles
    TI__Expert 3055 points

    Hi Jason,

    We check the Pin 14 as A1, as below:

  • 0 in reply to Shaoxing Ke
    TI__Mastermind 19795 points

    Hi Shaoxing,

    We have tested our devices on several platforms and cannot replicate this issue. Can you provide customer schematic of the pins? This seems like an external hardware problem. We have verified within design as well that there are no differences between F28002x and F280015x/13x devices with regards to AIO pins standard connections.

    Regards,

    Vince

  • 0 in reply to Vince
    TI__Expert 3055 points

    Hi Vince,

    I can test the EVM can see the 0.2/0.3 V. And customer disconnect the pin to their input circuit, just test the input circuit for the right value.

  • 0 in reply to Shaoxing Ke
    TI__Mastermind 19795 points

    Hi Shaoxing,

      

    Thanks for your follow-up. Can you do the following with the customer, and the project they are using?

    1. Please use one of the ADC examples in C2000Ware unmodified on their board. See if the same 0.2/0.3V issue exists (it should not, even without example code, the default device has been tested to not have this issue).

    2. Please send over the customer code.

    As a note: the issue described is not possible from the design side either, unless the pull-ups were accidentally enabled in the customer code. We have had design verify that boot-up of AIO is same as F28002x.

    Regards,

    Vince