DP83869EVM: DP83869EVM: 100Base-Tx amplitude

Hello Viktor,

I was able to discuss with my team on this topic. We believe that the deviation is tied to the board design. This board was made a while back, and may not be the most optimal that it could be. 

I would like to emphasize that these deviations from expected levels are marginal, and that there should not be any functional issues that arise from this. I would encourage to conduct functional tests on this EVM as well as any custom board you may have and if there are any functional issues, we can happily assist further.

Sincerely,
Gerome

Hi,

well I must disagree,becosue differecne in amplitude are in many cases cause by LAN transformer. Yes, each board has different LAN that would explained it. But there aree difference also in one board with crossed / not crossed cable. so everything is identical......I can not Imaginate that it can be something else than PHY it self....the same issue we already have with  DP83...22 PHY....amplitude of PHY was very different piece be piece......I have our first prototype on my dest and I also see this behavior....I measure just two samples where one is ok (deviation in amplitude is about 5mV) but in case of second piece I see 30mV between direct and crossed cable......if you do not believe me I can do measurement on the PHY directly....this will prove my suggestion I believe

Hi Viktor,

Are you seeing functional issues (such as link inability, BER) that are associated with these deviations on the EVMs or on the prototype boards?

Sincerely,
Gerome 

no we do not see any issue wit communication. Only problem we see is that the variation of outtu amplitude do not expect our expectation. there fore we would like to know what is the definition of output amplitude over the sample / voltage / temperature.....I measure at random samples about 90mV difference between channels......if I will take to the cound that ECT has limit in 100mV and there is some precison of measurement probe.....I see that there is issue to full fill the ethernet compliecne test

Hi Viktor,

Please have your FAE reach out to me over email for histogram data. This is only available under NDA only.

Sincerely,

Gerome