The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> High-Speed I/O Design >> Baseline wander modelling
https://designers-guide.org/forum/YaBB.pl?num=1280845482

Message started by currant on Aug 3^rd, 2010, 7:24am

Title: Baseline wander modelling
Post by currant on Aug 3^rd, 2010, 7:24am

Hello,
I am trying to simulate baseline wander effect in the ethernet 100 Mbit/s transmit path. I use test pattern from TP-PMD standart doc, scrambler with seed=390, ideal transformer with first winding
inductance 350uH on transmit and load ends and transmition line in the form of 2wire mtline.
When I do AC analysis, I see second order HPF response with f3db≈65 kHz.
But I can't see any big and long base line offset. I see only little (no more than 50 mV) and short (1-3 us) offset.
Did anybody simulate this effect? With what problem could I confront?
Many Thanks.

Title: Re: Baseline wander modelling
Post by love_analog on Aug 7^th, 2010, 1:35pm

you may have tested too short a pattern. Remember the worst case will be a long string of 1s and thats when you see the max effect. Simulate with that input stream and you will surely see the issue

Title: Re: Baseline wander modelling
Post by currant on Feb 9^th, 2011, 6:57am

Hello,
after a long work on another theme, I returned to the 100 Mbit/s Ethernet project.
I corrected binary-to-MLT-3 converter, and after applying BLW worst-case test pattern to the input of 4b/5b converter I saw a 53 bit time zero on output of scrambler. But I see baseline drift only if scrambler out connects directly to bin-to-MLT-3 converter.
In my structure after scrambler there is a NRZ-to-NRZI converter, as in datasheet of NS DP83816 and when I apply output of NRZ-to-NRZI block to input of bin-to-MLT-3 there is no any significant baseline drift.
Does anybody know, Do I need a NRZ-to-NRZI converter between scrambler and bin-to-MLT-3 converter?
Many Thanks.