• State Resolved
  • Locked Locked
  • Replies 15 replies
  • Answers 1 answer
  • Subscribers 103 subscribers
  • Views 425 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.

Original question:

AM3352: ADC detection ability

AM3354: Inquiry about ADC (Linux SDK)

Takayuki Miyazaki
Genius 10120 points
Part Number: AM3354

Hi Experts,

Our customer developed their system with AM3354BZCZA100 for a long time ago. Customer is using internal ADC, they are checking the voltage with built-in ADC. Recently, customer noticed that,  ADC’s value shows “0” Occasionally even though the voltage is output normally when checking voltage repeatedly. Customer is using TI Processor SDK 1.0.0.3 (very old version). Customer mentioned, this issue occurs with STEPDELAY (default value 0x098) in linux-driver, and it seems this issue was resolved with 0xFFF. Customer would like to know the reason why this issue was resolved with STEPDELAY (large value configuration). Therefore, customer is asking following inquiries.

  1. Customer is storing ADC data into FIFO (inside the CPU). When customer set large value to STEPDELAYxx(Open Delay bits), does it extend the effective period of the data written from the ADC to the FIFO?
  2. Or does it extend the validity period of the data that the user reads from the FIFO?
  3. According to TRM, customer is considering It seems to extend the validity period of the data written from the ADC to the FIFO. Is that correct?

Customer would like to know the reason why this issue was resolved with STEPDELAY (large value), as well. Sorry, I just translated customer’s inquiries.

I’ll really appreciate if you will share expert’s advice/comments on their inquiries.

Best regards,

Miyazaki

  • 0
    TI__Guru* 80625 points

    Hello Miyazaki-san,

    First off, please note that SDK 1.0.0.3 is too old for us to answer questions about the software on the forums. If I recall correctly, there were some changes to the Linux ADC driver between then and now, but I cannot remember if those changes were just for the ADC touchscreen driver, or also for "regular" ADC driver. It is possible that you may find some improvements in the ADC driver of more recent Linux SDKs.

    I am sending your thread to a hardware engineer to discuss the details of what different register settings are doing to the ADC hardware.

    Regards,

    Nick

  • 0 in reply to Nick Saulnier
    TI__Guru*** 141509 points

    Hello Miyazaki-san,

    Our customer developed their system with AM3354BZCZA100 for a long time ago. Customer is using internal ADC, they are checking the voltage with built-in ADC. Recently, customer noticed that,  ADC’s value shows “0” Occasionally even though the voltage is output normally when checking voltage repeatedly.

    I am trying to understand your inputs and have the below questions:

    Did you mean customer did not see this issue for long time and seeing this now. Is the issue observed on specific board.

    Can you please confirm if customer has taken care of the ADC related errata recommendations on the hardware?

    Regards,

    Sreenivasa

  • 0 in reply to Kallikuppa Sreenivasa
    TI__Guru*** 141509 points

    Hello Miyazaki-san,

    Please note the updated query regarding the ADC interface circuit:

    Can you help confirm if there were any changes made around the ADC interface circuit that could change the output impedance of the source, that is connected to the AM335x ADC inputs.

    Regards,

    Sreenivasa

  • 0 in reply to Kallikuppa Sreenivasa
    TI__Guru*** 141509 points

    Hello Miyazaki-san,

    Please note the updated query regarding the ADC interface circuit:

    Can you help confirm if there were any changes made around the ADC interface circuit that could change the output impedance of the source, that is connected to the AM335x ADC inputs.

    Regards,

    Sreenivasa

  • 0 in reply to Kallikuppa Sreenivasa
    TI__Guru*** 141509 points

    Hello Miyazaki-san,

    Please note the updated query regarding the ADC interface circuit:

    Can you help confirm if there were any changes made around the ADC interface circuit that could change the output impedance of the source, that is connected to the AM335x ADC inputs.

    Regards,

    Sreenivasa

  • 0 in reply to Kallikuppa Sreenivasa
    TI__Guru*** 141509 points

    Hello Miyazaki-san,

    Please note the updated query regarding the ADC interface circuit:

    Can you help confirm if there were any changes made around the ADC interface circuit that could change the output impedance of the source, that is connected to the AM335x ADC inputs.

    Regards,

    Sreenivasa

  • 0 in reply to Kallikuppa Sreenivasa
    TI__Guru*** 141509 points

    Hello Miyazaki-san,

    Please note the updated query regarding the ADC interface circuit:

    Can you help confirm if there were any changes made around the ADC interface circuit that could change the output impedance of the source, that is connected to the AM335x ADC inputs.

    Regards,

    Sreenivasa

  • 0 in reply to Kallikuppa Sreenivasa
    TI__Genius 10120 points

    Hi Sreenivasa,

    Thank you for your advice.

    I'm asking your questions to our customer. Once I'll receive their comments, I'll share them with you,

    Best regards,

    Miyazaki

  • 0 in reply to Takayuki Miyazaki
    TI__Guru*** 141509 points

    Hello Miyazaki-san,

    Thank you for the note.

    Regards,

    Sreenivasa

  • 0 in reply to Kallikuppa Sreenivasa
    TI__Genius 10120 points

    Hello Sreenivasa,

    I’m sorry for my slow response. Since I received customer’s feedback, I’d like to share them with you.

    Customer mentioned, there were Not any changes/modification around ADC interface circuit and it seems they have taken care of  the ADC related errata. Customer observed this behavior suddenly and they mentioned they did nit see this issue for a long time. Therefore, I asked customer what’s failure rate. This failure rate is, one or two times per 100 tests with the same board. However, customer mentioned they sometime were not able to observe this issue even if they have done 300 tests..

    Customer mentioned, this issue occurs with STEPDELAY (default value 0x098) in linux-driver, and it seems this issue was resolved with 0xFFF (large value). Therefore, if possible, could you clarify STEPDELAY register bit function, please?

    I appreciate your help.

     Best regards,

    Miyazaki

  • +1 in reply to Takayuki Miyazaki
    TI__Guru*** 141509 points

    Hello Miyazaki

    Thank you for the inputs.

    Please refer below i received from the experts

    The Open Delay value should not have any influence on the measurement.  It simply inserts delay before the step begins to execute.  However, the Sample Delay determines how long the sample capacitor is charged before a conversion begins.  If the source impedance is too high it may not  be able to charge the sampling capacitor enough before the conversion begins. 

    Regards,

    Sreenivasa

  • 0 in reply to Kallikuppa Sreenivasa
    TI__Genius 10120 points

    Hi Sreenivasa,

    Thank you for your clarification.

    I'll share your comments with customer, and I'd like to wait their feedback, for a while.

    Best regards,

    Miyazaki

  • 0 in reply to Takayuki Miyazaki
    TI__Guru*** 141509 points

    Hello Miyazaki

    Thank you.

    Can you also please check if they see the issue with any specific channel?

    Regards,

    Sreenivasa

  • 0 in reply to Kallikuppa Sreenivasa
    TI__Genius 10120 points

    Hi Sreenivasa,

    Since I did not receive any further question about STEPDELAY and their feedback, I'd like to close this ticket at thi time.

    Thank you for your help.

    Best regards,

    Miyazaki

  • 0 in reply to Takayuki Miyazaki
    TI__Guru*** 141509 points

    Hello Miyazaki

    Thank you for the note and appreciated.

    Regards,

    Sreenivasa