• State Resolved
  • Locked Locked
  • Replies 12 replies
  • Answers 2 answers
  • Subscribers 33 subscribers
  • Views 536 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.

DP83TD510E-EVM: Register values are reset after device restart - bootstrapping?

Peter Csato
Prodigy 50 points
Part Number: DP83TD510E-EVM
Other Parts Discussed in Thread: USB-2-MDIO, DP83TD510E

Dear TI E2E Team,

We are currently using two DP83TD510E-EVM boards for single-pair Ethernet evaluation.

In order to make it work we needed to modify the clock signal of TD83TD510E to 25MHz.

This was done using USB-2-MDIO and modifying register's value at address 0x0017 from value 40A1 to the new 4021 value.

Unfortunately however this value is reset back to 40A1 after a device restart (e.g., power supply restart).

If I understand it well the DP83TD510E is programmed through the MSP430. So if we would like to have this register set upon bootup we need to somehow write a bootup code to set the registers of TD83TD510E. I've already heard of bootstrapping but I got a bit lost.

I would appreciate if you could help me with startup configuration of TD83TD510E.

Thanks,

 Peter

  • 0
    TI__Mastermind 40285 points

    Hello,

    The new user's guide should be able to remedy this query. Please take a look.

    Sincerely,

    Gerome

  • 0 in reply to Gerome Cacho
    Prodigy 50 points

    Hi Gerome,

    Looking at the changelog of the new user's guide I did not find any part of the document that explains bootstrapping. There are some new chapters explaining resistor population but as we are currently exploring the possibilities and the final setup (clock signals etc.) are not decided yet, we would focus more on software and would not really solder resistors on a daily basis.

    Maybe I missed something in the user's manual but is there any other solution besides resistor population? Please help me change the register value at boot time.

    I would also appreciate for some detailed explanation of chapter 3.7. At least for me it is not clear what happens if I change a resistor connected to e.g. RX_DV (pin 26).

    Appreciate your help, kind regards,

     Peter

  • 0 in reply to Peter Csato
    TI__Mastermind 40285 points

    Hello,

    Please remove the LED_0 jumper. That should put DP83TD510E in RMII master mode which will give the 50MHz signal to the RMII slave device, DP83822.

    Sincerely,

    Gerome

  • 0 in reply to Gerome Cacho
    Prodigy 50 points

    Hi Gerome,

    LED_0 jumper was removed, unfortunately nothing changed.

    There are two 25MHz crystals on the PCB (Y4 and Y5), both of them are 25MHz.

    We would really prefer a software solution, is it possible?

    Thank you,

     Peter

  • 0 in reply to Peter Csato
    TI__Mastermind 40285 points

    Hi Peter,

    I am confused by your last post. You had stated you would prefer a software solution, but in your initial thread mentioned the 4021/40A1 fix. Isn't this the software solution you would be looking for?

    If you are not looking for that fix, then hardware strapping would be the only other method of getting this to work. DP83822 would need to be a RMII master/slave, while DP83TD510E has to be the corresponding mode. This can be accomplished by analyzing the bootstraps on both devices and configuring the EVM accordingly. As such, please refer to SNLU271B for directions on configuration.

    Sincerely,

    Gerome

  • 0 in reply to Gerome Cacho
    Prodigy 50 points

    Hi Gerome,

    I still prefer a software solution. The main goal is the 40A1 --> 4021 change on startup.

    If this is not possible, I still need help with the hardware strapping. LED_0 removal did not work.

    Please explain in details either a software solution or a hardware solution. Software solution is preferred is possible!

    Thank you,

     Peter

  • 0 in reply to Peter Csato
    TI__Mastermind 40285 points

    Hi Peter,

    Could you please implement steps in section 2.3 as well as ensure that other steps in SNLU271B are followed?

    Sincerely,

    Gerome

  • 0 in reply to Gerome Cacho
    Prodigy 50 points

    Hi Gerome,

    I will re-implement the steps described in 2.3 and will get back to you with the results at the end of the week or next week.

    Thank you,

     Peter

  • 0 in reply to Peter Csato
    TI__Mastermind 40285 points

    Hi Peter,

    Understood. Will look forward to your response.

    Sincerely,

    Gerome

  • 0 in reply to Gerome Cacho
    Prodigy 50 points

    Dear Gerome,

    Thank you for your patience. I followed chapter 2.3 in the new User's Manual and it works now, thank you!

    Could you please tell me what is the reason behind removing R199 and R200? They are in connection with pin 28 (LED_2/TX_ER) of DP83TD510E, so by removing these resistors pin28 will not be connected to voltage (1/2)*VDDIO_A.

    Also what is the purpose of removing LED_0 jumper, J20?

    I have not modified DP83822 so register 0000 value is currently 3100 instead of 0100. Could you please also explain the difference between 3100 and 0100?

    Thank you,

     Peter

  • +1 in reply to Peter Csato
    TI__Mastermind 40285 points

    Hi Peter,

    The resistor removals were due to redundancy between jumper strapping and resistor strapping. You can notice that some nodes are repeated and are used in either a resistor strapping network, or a LED jumper strapping network. Desoldering these resistors removes the redundancy and puts the strapping at the expected value (instead of the redundant resistor creating an equivalent "smaller" value, which would throw off the strapping due to different-than-expected voltage).

    For DP83822, this difference is due to PHY potentially connecting at 100mbps via auto-negotiation. Instead, we disable auto-negotiation and force 10mbps to be the speed.

    Sincerely,

    Gerome

  • 0 in reply to Gerome Cacho
    Prodigy 50 points

    Hi Gerome,

    Thank you very much for your explanation. Now everything is clear, and the dev. boards are up and running fine.

    The case can be closed, thank you again!

    Kind regards,

     Peter