|Thread Links||Date Links|
|Thread Prev||Thread Next||Thread Index||Date Prev||Date Next||Date Index|
Dieter – in the past, link segments that are compatible with (that is met by) ISO/IEC channel specifications have been used, rather than link segments that are identical to ISO/IEC channel specifications.
The specification you propose (𝑅𝑙 (𝑓) = 24 − 5log(f) 10<f<20 ) crosses the 19dB line at 15.849 MHz. So, I think a reasonable solution would be to modify the return loss equation to align with both clauses 184.108.40.206 and 220.127.116.11, unless there is a PHY reason not to.
Return loss >= min ( 18 dB, 24-5log10(f) )
Slide 3 asserts that ‘In the range of 10 MHz to 20 MHz IEEE 802.3 links and ISO cabling standards specify a slope for return loss.’ and proposes changing our adopted baseline ‘To harmonize it is proposed to specify the same slope as all others between 10 and 20 MHz’.
Doing a quick check I don’t see much evidence to support this, specifically:
100BaseT1 has constant R.L. up to 20MHz:
1000BaseT1 has a slope beyond 10MHz:
1000BaseT has constant R.L. up to 20MHz:
So the proposed change seems to be harmonizing specifically with 1000BaseT1, is there some good reason to do this ?