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.

RGMII trace length

Hi,

I saw in many documents the trace length of RGMII as 6 inches. For a PCB trace to act as a lumped element the length is based on rise time. If we take clock period , one cycle = 1/125 MHz = 0.008 micro seconds = 8 nano seconds.

If we take rise time as 10% it will be around 0.8 nano seconds and hence a transmission length of 0.8 inches approximately is possible without transmission line effect.

But I couldnt find out how the maximum trace length is estimated at 6 inches.Please help me in this regard.

Regards,

Bhavya

  • Hi Bhavya,

    Rise and Fall time is actually clearly specified in the RGMII specification. 750ps is the MAX rise/fall time for a RGMII transmitter.
    Also, the RGMII spec has a fairly low far-end receiver capacitance so any heavy input buffer will also limit your max distance.
    This is why 6 inches is a good rule of thumb. Of course you are welcome to experiment and try and exceed this number.

    Kind regards,
    Ross
  • Hi Ross,

    Is this 6 inch of maximum length with or without termination? How did you reach at precisely "6 inches"?

    Say for example , with/without termination how did you reach at the maximum possible length.

    Regards,

    Bhavya

  • Hi Bhavya,

    No termination though... these are CMOS drivers and they should not have high load termination.
    You should do a straight connection without far-end termination. It is recommended that you use 50 ohm routing.

    Again, there are too many factors to say what the max length is because it is material dependent, layer dependent and load capacitance dependent.

    6 inches is the standard consensus. If you want to go further, you are more than welcome but we will not give any guarantee.

    Kind regards,
    Ross