[57779] in North American Network Operators' Group
Re: DS3 Coax..
daemon@ATHENA.MIT.EDU (Jim Segrave)
Tue Apr 22 14:02:58 2003
Date: Tue, 22 Apr 2003 20:02:26 +0200
From: Jim Segrave <jes@nl.demon.net>
To: Josh Richards <jrichard@cubicle.net>
Cc: nanog@merit.edu
Reply-To: jes@nl.demon.net
Mail-Followup-To: Jim Segrave <jes@nl.demon.net>,
Josh Richards <jrichard@cubicle.net>, nanog@merit.edu
In-Reply-To: <20030422163211.GC13679@cubicle.net>
Errors-To: owner-nanog-outgoing@merit.edu
On Tue 22 Apr 2003 (09:32 -0700), Josh Richards wrote:
>
> Makes sense but...
>
> Before fiber and ADMs were so common, in the days of buried coax for
> DS-3s (between COs and for the occassional large customer), coax was
> obviously being used for inter-building connectivity.
>
> How exactly did the topology differ? Besides fiber, which eliminates
> the issue of ground loops and such altogether, how did the RBOCs actually deal
> with this problem when they didn't have fiber in the ground?
>
> I presume there's an element/architecture that can be used safely. Mike alluded to
> a 'GID' per Sprint. Anyone have a reference/source for this device?
One easy way would be to use baluns and go from 120 ohm balanced on
RJ45s in the buildings to 75 ohm coax between buildings. Or you can
probably find a supplier of 75 ohm 1::1 transformers suitable for data
traffic. Both should provide isolation and allow a floating ground.
--
Jim Segrave jes@nl.demon.net
