Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

  • State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 182 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.

DP83822H: Forced Link Mode Side Effects

Tom D
Prodigy 21 points
Part Number: DP83822H

Team, could you help with a couple questions of mine? 

We are interested in running the DP83822 PHY in a forced configuration (forced crossover, full duplex, 100Base-TX). While we understand the higher level implications of this, such as the need to properly configure the link partner, we wanted to inquire to see if we need to be considering less obvious effects. Below are a couple questions:

  • Are there any PHY DSP operations effected? The IEEE standard and datasheet don't provide much insight here, but it appears like the PHY is at least performing a couple tasks: Input and Baseline wander compensation, Digital Adaptive Equalization, Clock Recovery.
  • Which PHY DSP functions are performed only once on link up, if any? How does forcing the link impact these functions?
  • Are there any other implications or settings associated with forcing the link that we should be considering?

Regards,

Tom

  • 0
    TI__Mastermind 19621 points

    Hi Tom,

    Using the DP83822 Ethernet PHY in a forced link mode is completely okay. Like you mentioned, just need to make sure the link partner is set at the same speed, MDIX, etc.

    Regards,

    Alvaro

  • 0 in reply to Alvaro (Al-vuh-roe) Reyes
    Prodigy 21 points

    Hey Alvaro, 

    Thank you for the response. That answers part of my question (it is ok to run in full forced mode). However, could you address the questions regarding how the PHY's DSP functions on this chip? Having this additional insight will allow us to more rigorously test our hardware to prove the robustness of the link. Additionally, could you clarify what you mean by Master/Slave? I don't believe that the master/slave concept exists for 100Base-TX, only for other standards like 1000Base-T.

    -Tom

  • 0 in reply to Tom D
    TI__Mastermind 19621 points

    Hi Tom,

    For Robust Link up we recommend auto neg.

    If you want 100 Mbps, we recommend keeping auto-neg on and de-advertising other modes (Write Reg 0x0001 = 6049)

    If using a forced mode and link is good, all DSP functions will be performed as usual. 

    Regards,

    Alvaro