RE: [802.3ae_Serial] Cable plant values?
Piers - thanks.
1. On the attenuation value, the only concern might be some legacy cable
plants may not comply with 3.5. However, I completely agree there is no
budget issue and the difference in attenuation with such short link
would get lost in measurement accuracy anyway. So, there is no technical
issue, just maybe some less than perfect consistency with the
spreadsheet.
Issue closed.
2. Table 52-24. My mistake was forgetting to include Additional
insertion loss allowed values, which put me in error up to 1 dB. Once I
include them, I am within 0.1 dB roundoff error for my purposes. I am
not concerned about about the hundreths of dB stuff.
Issue closed.
Thanks again, Tom
Doug - should 10GFC use 3.5 dB instead of 3.75 in Table 22? As Piers
points out, it makes no difference in budgeting.
-----Original Message-----
From: piers_dawe@agilent.com [mailto:piers_dawe@agilent.com]
Sent: Friday, May 17, 2002 10:42 AM
To: Doug.Coleman@corning.com; Lindsay, Tom
Cc: owner-stds-802-3-hssg-serialpmd@majordomo.ieee.org;
stds-802-3-hssg-serialpmd@ieee.org
Subject: RE: [802.3ae_Serial] Cable plant values?
Tom,
38.11.1 Table 38 -12 for 62.5 MMF has 3.75 dB/km and says "This value of
attenuation is a relaxation of the standard (IEC 60793-2, type A1b,
category
less than or equal to 3.5 dB/km)."
It looks like the 10GE insertion losses are compatible with 3.75 dB/km
(see
below). As we have only 33m, the difference is 0.25*0.033 = 0.01 dB: in
the
noise. So the 10GE tables are compatible with 3.5 dB/km too.
I think the numbers in tables 52-10 and 52-24 are consistent:
62.5 um MMF 50 um MMF
Table 52-10 160 200 400 500 2000
Additional insertion loss allowed
1.0 0.8 0.5 0.5 0.0 dB
10GEPBud3_1_16a.xls 0.84 0.81 0.63 0.57 0 dB
Updated values for model 0.96 0.83 0.51 0.48 0 dB
Table 52-24
Channel insertion loss 2.6 2.5 2.2 2.3 2.6 dB
10GEPBud3_1_16a.xls 2.44 2.44 2.37 2.37 2.59 dB
Updated values for model 2.56 2.46 2.24 2.28 2.59 dB
The updated values give equal unallocated margin across the fiber types
at
the same spectral characteristics (840 nm).
Piers
> -----Original Message-----
> From: Doug.Coleman@corning.com [mailto:Doug.Coleman@corning.com]
> Sent: 17 May 2002 17:42
> To: Lindsay, Tom
> Cc: owner-stds-802-3-hssg-serialpmd@majordomo.ieee.org; 802.3ae Serial
> reflector (E-mail)
> Subject: Re: [802.3ae_Serial] Cable plant values?
>
> Tom:
>
> Legacy 62.5 um fiber attenuation was specified as 3.75 dB/km
> at 850 nm as
> defined in TIA-568 Rev A. TIA-568 has recently been updated
> (2001) as
> Rev. B that modified the 850 nm attenuation to 3.5 dB/km for
> 62.5 um fiber.
> Also, TIA-568 Rev B now includes 50 um multimode fiber with
> an attenuation
> of 3.5 dB/km at 850 nm.
>
> Hope this helps.
>
> Doug
>
> "Lindsay, Tom" <tlindsay@stratoslightwave.com>
> To: "802.3ae Serial reflector (E-mail)"
> <stds-802-3-hssg-serialpmd@ieee.org>
> Sent by: owner-stds-802-3-hssg-serialpmd@majordomo.ieee.org
> cc: "Doug Coleman (E-mail)" <doug.coleman@corning.com>
>
> Subject: [802.3ae_Serial] Cable plant values?
>
> 05/17/02 12:07 PM
>
> Questions
>
> 1. Table 52-25 quotes 3.5 dB/km max attenuation at 850nm, whereas the
> spreadsheet uses 3.75. 10GFC also quotes 3.75 in its
> documentation. Which
> value is correct? All 3 sources agree on 3.5 dB for 50 micron fiber.
>
> 2. Table 52-24 shows channel insertion losses. Can someone
> explain where
> the values come from for the 850 nm variants? Except for
> 2000 MHz-km, I
> get significantly different values. The values in the table are not
> consistent with the model nor with the budget tables for the PMDs.
>
> Thanks, Tom
> Stratos NW
> (425) 672-8035 x105
>
> (See attached file: Tom Lindsay @ work.vcf)
>
>