• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 33 subscribers
  • Views 403 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.

DP83869HM: MDIO gives wrong data on query from DP83869HM

Souvik Sarkar
Prodigy 100 points
Part Number: DP83869HM

Hello All,

I am currently using 4 numbers of DP83869HM ICs in my design. 2 of which are having VDDIO as 3.3V and 2 of them are having VDDIO of 1.8V. These are connected as shown in the figure below

Now the ones connected to 3.3V are working fine. There are no data inconsistencies on them.

But the 1.8V ones are giving unstable data. While probing the "mii info" command from uboot of NXP processor, the vendor ID keeps on changing on every probe as shown below.

PHYs 0x0A and 0x0E are 3.3V PHYs and PHYs 0x0B and 0x0F are 1.8V PHYs. You can see the inconsistency in OUI and other details.

I have also attached the PDF with the schematic of the 1.8V PHYs. Please help with this.

Schematic of 1.8V DP83869 PHYs.pdf

Thanks and Regards,

Souvik

  • 0
    TI__Mastermind 39825 points

    Hello,

    This is a very unique use case for the PHYs. As MDIO is bi-directional and MDC is an input to the PHY, I would be concerned with the setup and hold time, as well as reflections that would be coming out. I would advise probing the SMI lines on 1.8V VDDIO and compare to 3.3V VDDIO to understand where the source of the behavior is coming from.

    Sincerely,

    Gerome

  • 0 in reply to Gerome Cacho
    Prodigy 100 points

    Hello Gerome,

    Thank you for your insight. As can be seen from the below figure, the main concern might have been the rising waveform of the MDC signal in-spite of the pull-ups.

    Observing the above waveform, we increased the pull-up from 2.21K to 1K on MDC. Thereby,  one of the 1.8V PHYs was being detected correctly. We could not increase the pull-up more because of current limits in both the PHY and the processor. 

    Once, we changed the MDC/MDIO configuration at the processor from open-drain to push-pull configuration, both the PHYs are now being detected correctly. There are no more issues with data inconsistency.