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.

DS15MB200: DS15MB200 being used as a SGMII multiplexor with two Intel I210-IS

Part Number: DS15MB200

Hi

We are currently developing a board that uses 2 x I210-IS devices that are configured in SGMII mode and these are then connected to the DS15MB200 so that we can change the destination PHY for each device.

We have found that after a few days of successful tests sending packets, the SGMII RX interface (MUX to I210) seems to break permanently while the SGMII TX (I210 to MUX) seems to continue to work. 

This suggests that the TI MUX is maybe driving the SGMII RX incorrectly.

1. Is there a reference schematic of the DS15MB200 being used as an SGMII MUX?

2. Is there a reference design that uses the DS15MB200 and the Intel I210-IS?

3. Can you suggest why the MUX is permanently breaking the SGMII interface? We have AC coupled the TX and RX lines.

4. We think that this may be related to the drive strength or VOD (Differential Output Voltage). Can you suggest a way to lower this so that the SGMII interface does not break?

Thank you

  • Hi Luke,

    This is one of the older parts and unfortunately there is not a reference design.

    1). When you say "break permanently". Does it mean when it breaks you have to power down and up to get it going again? Or you have to replace the SGMII input.

    2). Does this occur on every part or some parts? Assuming you run your test the same length of time.

    3). Since you are AC coupled, when this occurs - permanent damage - is the output of the DS15MB200 that is connected to SGMII interface active?

    Regards,, Nasser

  • Hi Nasser

    Thank you for getting back to me.

    1) The SGMII input on the Intel device is damaged and we have to replace the chip. The TI mux remains functional. We have run the test a few times and had to replace the Intel chip each time.

    2) This occurs on every part. We suspect the Intel device input is very sensitive.

    3) Yes, the MUX output is active. We transmit terabytes of data over days and at some point the Intel chip becomes permanently damaged. 

    Regards 

    Luke

  • Hi Luke,

    Thanks for explanation. It is more clear to me now. 

    As you said Intel part seems to be sensitive. Suggestion is to turn turn off pre-emphasis on TX side. Also, i don't know if you are pulsing output enable pins. Maybe you want to have this tied high permanently. I am trying to reduce any current build up on the receiver side. Hope these help.

    Regards,, Nasser