[comp.dcom.telecom] Metallic circuits

dyer@spdcc.com (Steve Dyer) (11/10/88)

In article <telecom-v08i0166m05@vector.UUCP> lars%acc.arpa@bu-cs.BU.EDU writes:
>The bandwidth of a metallic circuit is probably more like one megabit;
>this is why we are moving towards IDSN: 2 twisted pairs is 1544 kbps =
>24 voice channels.

OK.  Here's my situation.  I've got a metallic (LADS) circuit between
my house and my Internet site.  I am presently running at 19.2kb with a
pair of Gandalf LDS309a limited distance modems.  What are my chances
of ekeing any additional bandwidth out of this line, and what
technology would be necessary to take advantage of it?  I have seen ads
for 56kb DSUs which are also claimed to work on twisted pair, but they
seem to refer to in-house private twisted pair and not metallic lines
from the TelCo.  Just what is the theoretical and practical maximum
bandwidth of a LADS circuit?

--
Steve Dyer
dyer@harvard.harvard.edu
dyer@spdcc.COM aka {harvard,husc6,linus,ima,bbn,m2c,mipseast}!spdcc!dyer

roy@phri (Roy Smith) (11/16/88)

> I've got a metallic (LADS) circuit between my house and my Internet site.
> I am presently running at 19.2kb with a pair of Gandalf LDS309a limited
> distance modems.  What are my chances of ekeing any additional bandwidth
> out of this line

	We have a LADC (is that the same as LADS?) circuit.  About a mile
total loop length of 26 gauge wire (2 pairs).  Using either Black Box
V.35-LDM's or Rad ASM-20's we manage to run 56 kbps with no problems (well,
no problems after many fights with NY Tel to get the bugs out of the line,
which included 2 bad house pairs and a pair with polarity not observed).
Actually, I think the LDM's are running at 64 kbps.  Rad supposedly has a
flavor of the ASM-20 which will run at 128 kbps, as does Black Box.  VIR
has a modem whch will supposedly do 230 kbps over that line, but we havn't
gotten it to work at anything faster than 19.2 kbps.

	Most of the units I mentioned are V.35 modems, but that's a trivial
problem.  I'm sure they can be bought with RS-232 interfaces, and if not,
RS-232 to V.35 converters are trivial.  All the above-mentioned modems are
about $600-700 each.
--
Roy Smith, System Administrator
Public Health Research Institute
{allegra,philabs,cmcl2,rutgers}!phri!roy -or- phri!roy@uunet.uu.net
"The connector is the network"

roy@phri (Roy Smith) (11/20/88)

To update what I said earlier:
> VIR has a modem whch will supposedly do 230 kbps over that line, but we
> havn't gotten it to work at anything faster than 19.2 kbps.

	We tried the VIRs on another loop which is much shorter (we figure
about 2-3000 feet) and it seems to work fine at 144 kbps, but we still
can't get it to work at 230 kbps over any appreciable amount of wire.  VIR
admits that there is a problem with the modem (they have already modified
it a couple of times, although I have no idea what they did to it).  In
fact, on of the engineers at VIR admitted that RAD makes the best modems!
VIR says they will have a redesign of theirs out in a few months which
really should work at 230 kbps over 1-mile loops.

	To be fair, VIR has been very good to us.  They lent us 2 BERTs for
a few weeks so we could test the lines and help us convince NY Tel that
there really was something wrong.
--
Roy Smith, System Administrator
Public Health Research Institute
{allegra,philabs,cmcl2,rutgers}!phri!roy -or- phri!roy@uunet.uu.net
"The connector is the network"