• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 33 subscribers
  • Views 80 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.

DP83TC814S-Q1: No waveform output after setting test mode

Ivy Jin
Expert 4565 points
Part Number: DP83TC814S-Q1

Tool/software:

Hi team,

My customer is testing the physical layer of Ethernet and after they set up Test mode, they find no waveform output, may I ask the potential reason for this issue? What is the way to confirm that Test mode was set successfully or failed?  Thanks.

Regards,

Ivy

  • 0
    TI__Mastermind 18930 points

    Hi Ivy,

    This device has 4 different test modes as described in the Open Alliance Testing document (https://www.ti.com/lit/an/snla389e/snla389e.pdf):

    I have some questions:

    • What PMA test is the customer trying to run, what test mode are they trying to enable, and what is the register procedure they have followed?
    • Are the other parts of the PHY's behavior normal? For example, is it verified to be powering on, can link up and send packets, and has register access?

    Best,

    Evan Su

  • 0 in reply to Evan Su
    TI__Expert 4565 points

    Hi Evan,

    Customer needs to test Test mode1, 2, 4, 5. Communication is normal, and the registers are accessible.

    1. After setting the appropriate mode, what is the way to confirm that the PHY is actually working in the mode that is being set?

    2. What is the potential reason why there is no waveform output if it is confirmed that the PHY is operating in the set mode?

    Thank you!

    Regards,

    Ivy

  • 0 in reply to Ivy Jin
    TI__Mastermind 18930 points

    Hi Ivy,

    Ivy Jin said:
    Communication is normal, and the registers are accessible.

    OK, this is good to hear.

    Ivy Jin said:
    1. After setting the appropriate mode, what is the way to confirm that the PHY is actually working in the mode that is being set?

    The only method I am aware of is to re-read the test mode register(s) to ensure that the write was successfully applied. Some customers in the past have written to the wrong registers when they were trying to enable the test mode. I have not heard of cases where the register was set but the PHY did not enter the appropriate test mode.

    Ivy Jin said:
    2. What is the potential reason why there is no waveform output if it is confirmed that the PHY is operating in the set mode?

    It depends on the test mode and the conditions. Please see if the customer can tell us more details about their software and hardware procedures for each test modes they have problem with. For example, I recently helped another customer who said they were not able to get an output waveform on a certain test mode, and it turned out that they did not have a link partner and were looking at the wrong pin.

    Best,

    Evan Su

  • 0 in reply to Evan Su
    TI__Expert 4565 points

    Hi Evan,

    "It turned out that they did not have a link partner" means to connect the host and slave and then measure the waveform on the bus? 

    As shown in the figure, the customer is currently testing the Port directly after the Test mode is configured, and is this OK? 

    Regards,

    Ivy

  • +1 in reply to Ivy Jin
    TI__Mastermind 18930 points

    Hi Ivy,

    Ivy Jin said:
    "It turned out that they did not have a link partner" means to connect the host and slave and then measure the waveform on the bus? 

    Not quite. For the specific test mode the other customer was asking for help with, it was necessary to connect the host and slave, then measure the output on a separate PHY pin that was not the Ethernet bus, but this is not necessarily the same for the other test modes.

    Ivy Jin said:
    As shown in the figure, the customer is currently testing the Port directly after the Test mode is configured, and is this OK? 

    Thank you for the picture, but each test mode is different so we will still need more detailed information on the procedures. The picture setup would be able to check some test modes, but not all.

    For every test mode that the customer has found a problem with, please ask them to:

    • Clarify what exact register writes they are performing to switch the PHY into the specific test mode,
      • Verify with readback that the register writes were successful
    • Clarify what they have performed on hardware (where is the output being measured, was a link partner used, etc.)

    Best,

    Evan Su