• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 98 subscribers
  • Views 218 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.

[FAQ] AM6442: Rise/Fall time requirement for RGMII(0.75ns)

Shota Mago
Expert 8110 points
Part Number: AM6442
Other Parts Discussed in Thread: DP83867E, DP83869HM, TMDS64EVM

Hi team,

I'm asking same question on interface forum. But I need help from Sitara team as well because it seems related to drive strength for RGMII of AM6442.

My customer will use AM6442 in conjunction with DP83867E and DP83869HM. But only AM6442's RGMII output side can't achieve 0.75ns rise/fall time which is required by RGMII specification.

These PHYs are used at EVM of AM6442, but my customer has been struggling with achieving this slew rate requirement.

Could you please provide any advice to achieve 0.75ns slew rate(e.g. adjusting output impedance/drive strength from register, changing I/O voltage etc.)?

Best regards,

Shota Mago

  • 0
    TI__Guru**** 189029 points

    Hello Shota Mago

    Thank you for the query.

    Could you please help confirm the IO supply customer is using.

    Regards,

    Sreenivasa

  • 0 in reply to Kallikuppa Sreenivasa
    TI__Expert 8110 points

    Hi Sreenivasa,

    1.8V is used for IO supply.

    Best regards,

    Shota Mago

  • 0 in reply to Shota Mago
    TI__Guru**** 189029 points

    Hello Shota Mago

    Thank you. Please refer inputs from the expert.

    RGMII is source synchronous, where we are using IOs of the same design with similar rise/fall characteristics and we define PCB trace length matching such that the signals should arrive at the attached device with enough margin to always meet the setup/hold requirements defined for an attached PHY. 

     There is a trade-off on signal rise/fall times.  The signal are likely to cause radiated emission problems if the rise/fall is too fast.  So we have tried to find a happy median, where the outputs drive the signal fast enough to produce acceptable timing margin without being too fast to cause excess radiated emissions.

    Regards,

    Sreenivasa

  • 0 in reply to Kallikuppa Sreenivasa
    TI__Expert 8110 points

    Hi Sreenivasa,

    Kallikuppa Sreenivasa said:
    There is a trade-off on signal rise/fall times.  The signal are likely to cause radiated emission problems if the rise/fall is too fast.  So we have tried to find a happy median, where the outputs drive the signal fast enough to produce acceptable timing margin without being too fast to cause excess radiated emissions.

    Does it mean TMDS64EVM doesn't meet 0.75ns rise/fall time? In my understanding, the0.75ns(max) rise/fall time is specified by RGMII standard.

    Is it okay that rise/fall time doesn't achieve 0.75ns rise/fall time as long as if there is enough margin for setup/hold requirement?

    If so, could you please provide any evidence which states 0.75ns requirement can be compromized?

    Best regards,

    Shota Mago

  • +1 in reply to Shota Mago
    TI__Guru**** 189029 points

    Hello Shota Mago

    Thank you.

    Shota Mago said:

    Does it mean TMDS64EVM doesn't meet 0.75ns rise/fall time? In my understanding, the0.75ns(max) rise/fall time is specified by RGMII standard.

    Is it okay that rise/fall time doesn't achieve 0.75ns rise/fall time as long as if there is enough margin for setup/hold requirement?

    This is correct.

    Shota Mago said:
    If so, could you please provide any evidence which states 0.75ns requirement can be compromized?

    As i said this is a trade-off and we have the SK and many customer implementations that work.

    Regards,

    Sreenivasa