DP83848I: DP83848 fails after repeated plugging and unplugging

Hi Alon,

When testing with the suspected non-faulty part, are there any issues with communication?

Reviewing schematic with team and will get back to you ASAP with feedback.

Thank you,

Evan

Also, does the customer suspect part damage occurred after plugging/unplugging the cable multiple times?

Thank you,

Evan

Hi，

It is suspected that plugging and unplugging is damaging the device, and it is also suspected that there is a problem with the circuit design. Is there a problem with the customer's circuit?

Hi Alon,

I do not see any concerns with the MDI-side connection, but still waiting for more feedback from team members with more PoE expertise.

If possible, please measure the voltage on TD/RD pins of PHY when plugging in cable and communicating. I'd like to confirm that the transformer is isolating the signal and power on the line to avoid PHY damage.

Thank you,

Evan

Hi Evan，

Measure the power-on waveforms of each pin as shown below. Please help with the analysis.Thanks.

Hi Alon,

Are these waveforms with the faulty DP83848, or the one that is working for communication?

The signal levels do not appear to violate max rating of MDI pins, so the cause for PHY damage is not clear.

For the suspected faulty part, please confirm:

• Does probing across RBIAS resistor read ~1V?
• Are PHY registers accessible?

Thank you,

Evan

Hi Evan，

A device that is normally capable of communicating

• Does probing across RBIAS resistor read ~1V? 1.194V
• Are PHY registers accessible? yes

Hi Alon,

If registers are accessible for the PHY on a board that is failing to communicate, then it seems there is another device on board that may be damaged/failing (some passive component on PoE network?)

Thank you,

Evan

Hi Evan,
The experimental failure of the AB-BA crossover follows the damaged DP83848, is the conclusion so far that there is nothing wrong with the schematic and PCB design? Whether we can only suspect that the ESD damage

Hi Alon,

To clarify, there are two criteria I'd like to check to confirm PHY health vs. system health:

1) PHY registers are accessible

2) Cable-side communication to a link partner

Is the damaged DP83848 failing both (1) and (2)?

I do not see any concerns in the schematic, but I am confirming with a colleague with more PoE expertise.

Can you please also share the layout for our review? Altium files are preferred if possible (can email to e-mayhew@ti.com for private share).

Thank you,

Evan

Hi Evan，

The schematic and PCB have been sent to you, but there has been no feedback, please help to give the review results.

Hi Alon,

Please expect feedback today.

Have you been able to confirm notes (1) or (2)?

Thank you,

Evan

Hi Evan，

1) PHY registers can be accessed,
registers can be accessed
2) Communicate with the cable side of the link partner
Cannot communicate

Thanks

Hi Alon,

Thanks for confirming. We have shared feedback over email.

Best Regards,

Evan