Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!usc!wuarchive!mailrus!accuvax.nwu.edu!nucsrl!telecom-request
From: jst@cca.ucsf.edu (Joe Stong)
Newsgroups: comp.dcom.telecom
Subject: LADS Circuits
Message-ID: <2310@accuvax.nwu.edu>
Date: 19 Dec 89 09:10:00 GMT
Sender: news@accuvax.nwu.edu
Reply-To: jst@cca.ucsf.edu (Joe Stong)
Organization: Computer Center, UCSF
Lines: 45
Approved: Telecom@eecs.nwu.edu
X-Submissions-To: telecom@eecs.nwu.edu
X-Administrivia-To: telecom-request@eecs.nwu.edu
X-Telecom-Digest: Volume 9, Issue 584, message 10 of 10


I would like to hear of people's experiences with LADS (solid copper 4
wire or 2 wire circuits only for sites local to the same central
office).  In particular, I'm curious about what you can get through
them in the way of frequency response.

My understanding at the moment is that T1 is delivered on a regular 4
wire connection to one's building.  Are there equalization networks
along the "subscriber loop" of a T1 that make it "better" than a LADS
circuit?

Here's the scenario: I'm wondering if I/others can connect a LADS
circuit to an ARCnet board on a PC, running the Phil Karn (KA9Q)
software, with the ARCnet-TCP/IP driver in place (don't worry, we're a
non-commercial outfit) to get a TCP/IP connection between home and
work.  The 1 megabaud (nominal?) rate from the ARCnet board should be
lower rate than a T1 at 1.544 megabaud; will it make it in terms of
frequency response?  If this setup would work, it would let everybody
use relatively CHEAP interfaces and software to make a FAST
connection.

An old story: Back when 9600 baud modems were horribly expensive, we
arranged a remote terminal on a LADS circuit for a customer in Arcata,
CA.  The "couplers" (not modems) that got purchased seemed to be
transformer coupled, and mostly seemed to convert to/from some RZ
coding scheme at a fixed baud rate (9600) onto the LADS line.  The
whole setup was horribly sensitive to electrical noise (changing the
interior wiring that PacBell had done from silly "quad wire" (NOT
twisted pair) to twisted pairs helped some, but it was still subject
to garbage occasionally.

Would current-loop couplers have been better for this 6 mile circuit?
(Traditional sillyness, the offices were only 2 blocks apart, but the
phone company would only make the connection into the office on the
other side of town, and back out again) I notice that some other
high-speed devices use RS422 (like microwave transmitters), but I've
had better luck with current loop lines than differential or
single-ended lines.

At least in San Francisco, LADS circuits are extremely cheap, namely
$200 initial installation and $18 a month, when I checked a couple of
years ago.


       Joe Stong       jst@cca.ucsf.edu