• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 34 subscribers
  • Views 75 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.

DP83867CS: DP83867CS SGMII TX tuning

Seppo Pyorret
Prodigy 20 points

Part Number: DP83867CS
Other Parts Discussed in Thread: 66AK2L06, , DP83869

System information:

  • SOC: 66AK2L06
  • PHY:  DP83867CS
  • Diff pair impedance 100 ohms in PCPB
  • 100nF series capacitors close to SGMII RX
  • 100nF series capacitors close to PHY RX

 

Now our eye diagrams looks as follows, measured at SOC RX and immediately after the 100nF series caps and with full communication load:

 

We have observed that for close to 4 day continuous operation, about 50% of the boards have no rx align code errors. The other half we get one rx align error code per each hour of operation. Nothing deadly serious but we would like to tune the PHY SGMII TX to SOC to make this connection more robust.

 

Now we would like to tune the DP83867CS SGMII TX to SOC but according to datasheet there are no tuning values for slew rate, de-emphasis etc. Are there any hidden registers in PHY that would allow tuning?

  • 0
    TI__Mastermind 37295 points

    Hi Seppo,

    When you see the RX align error, do you notice any value in register 0x0015 of the PHY? This would tell us if the RX error is coming upstream from the PHY (cable/link partner) or whether it is coming from the SGMII.

    Register 0x0015 is an error counter on the RX coming from the link partner. I would also be curious to know if swapping the link partner or the cable has any effect on the align error you're seeing.

    Best,

    Shane

  • 0 in reply to Shane Hauser
    Prodigy 20 points

    Hi Shane,

    Sorry for my late reply. Reg 0x0015 shows no errors. The issue is completely on SOC RX side. We would like to boost the PHY TX to SOC but it appears that PHY has no parameters to increase the swing or any other parameter.

  • 0 in reply to Seppo Pyorret
    TI__Mastermind 37295 points

    Hi Seppo,

    Thanks for confirming. If we're looking at ways to increase the SGMII swing, consider trying register D6:

    This register is present on a similar PHY (DP83869). While this is not described in the DP83867 datasheet, both of these PHYs share enough similarities in the register map that this register may be present on DP83867 as well. Can you see if modifying this register has any impact on the SGMII swing?

    Best,

    Shane