• State
  • Replies 3 replies
  • Subscribers 34 subscribers
  • Views 53 views
  • Users 0 members are here
Support feedback
Options
Options
Related

DP83869EVM: No SGMII coming from DP83869 chip

Marvn Arcangel
Genius 13690 points
Part Number: DP83869EVM
Other Parts Discussed in Thread: DP83869,

Hi Experts,

Hello, we are trying to connect to the DP83869EVM using ethernet and send SGMII out of the 4 SMA. We can see ethernet traffic going into the DP83869 chip using an oscilloscope from the ethernet cable, but it doesn’t come out of the chip and therefore no signal to the SMA. We have jumpers on OPMODE 1 and 2, none on OPMODE 0 nor MIRROR_EN. We populated C12, C14, C15, and C17 and removed R31, R38, R45, and R47. We have the SW modes normal operation (0000). What possible issues could there be causing the chip not to work?

Regards,

Marvin

  • 0
    TI__Mastermind 31490 points

    Hi Marvin,

    I believe this question is related to another query I've been working on. Here was my latest response with replies for context:

    The jumper settings you describe should be correct. Essentially, it would be the default jumper settings, except with Opmode 1 and 2 jumpers populated

    One thing I am curious on: Is the signal just a flat line at 0V when probing near C12 and C14, or is there a signal?

    • I was just seeing a flat line around 0V near those caps.

    Another thing: If there is no connection on the RJ-45 I suspect you may see SGMII auto negotiation, but no real data. Is that ok for this test? The PHY typically takes data from the MDI/copper connection and transmits it on the SGMII lines.

    • The main concern was making sure that all the jumper connections are correct. I think we could verify the SGMII signal when connected to the system in a few days

    Let me see if I can test this on my EVM in our lab. I believe you would see more than a flat 0V on the outputs, but I'll check to be sure.

    Best,

    Shane

  • 0 in reply to Shane Hauser
    Prodigy 10 points

    Thank you for creating the post, this is our issue. We have tested the board in our system and we could not establish a connection. 

    Yes when we test on the chip side of C12 and C14 we see no signal whatsoever with differential probes. On the lines from the PC going into the DP83869 we can measure what is assumed auto negotiation(Figure 2-11 in snla443a). If you could test those output lines from the DP83869 in your lab to confirm what we should see that would be very helpful.

     In addition to the jumpers, we have also wrote to registers, could you verify these are correct?

    Wrote 0x000E to register 0x0018 

    Wrote 0046 to register 01DF to change to SGMII-copper mode according to PDF snla318

    Wrote 0000 to register 0009 to change to 100Mbps

    Thanks a lot,

    Garett

  • 0 in reply to garett carlson
    TI__Mastermind 31490 points

    Hi Garett,

    I reworked my EVM for SGMII (populate C12/14/17/15, depop fiber resistors) and used the default jumper settings with OPMODE 2 and 1 jumpers populated. In this configuration I can see a signal on the SGMII outputs near the SMA:

    This is with nothing plugged into the RJ-45 side, so no data is transferring through the PHY. Even after writing the same registers you describe I still see the same signal coming from the SGMII outputs.

    Can you confirm our jumper settings are the same? If so, how are you probing the signal? I'm using a 10Mohm high-impedance single ended probe.

    Best,

    Shane