Thread: dialup issues
View Single Post
  #10   Report Post  
Posted to sci.electronics.repair
Dave Platt Dave Platt is offline
external usenet poster
 
Posts: 379
Default dialup issues

In article , mike wrote:

My question is if the problem is in the premises wiring what could be
the most likely cause of something like this? As I previously
mentioned I can't "hear" any problem but of course that doesn't mean
that one doesn't exist. Before I start dismantling my inside wiring it
would be nice if I had a clue as to what I was looking for. The phone
company by the way has been no help with this.


I've had premises-wiring problems from all sorts of causes:

- Humidity and moisture condensation in the wall outlet box
(affecting a wall box located near our washing machine). Surface
contamination (dirt and mold buildup) plus moisture equals leakage,
which was apparently enough to cause problems.

- Corrosion of the connecting screws, wire-end ring or fork
terminals, or the wire ends crimped to the terminals... again,
probably due to moisture and contamination over the years.

- Loose screw-and-terminal connections in the connection boxes. In
many cases the screws are coarse-thread self-tapping varieties,
screwed into relatively soft plastic; over the years they can work
loose or (if loosened and tightened repeatedly) tear or wear out
the plastic into which they are fastened.

- Corroded, dirty, or bent metal "fingers" in the RJ-11/12 jacks.

- Too many branch circuits within the house... e.g. it's wired for "a
phone in every room". The usused jacks act as unterminated stubs,
and cause signal reflections (near-end echo) which can interfere
with the incoming signal. My guess is that near-end echo which is
inaudible or unobjectionable to the human ear (too low in
amplitude, too near in time to the original transmission) can still
degrade signal quality in a full-duplex modem transmission.

- Phones on other branch circuits on the line placing their ringer
impedance across the line - again a form of near-end echo but with
slightly different characteristics.

- Bad connections (e.g. sets of wires merely twisted together, perhaps
with wire nuts, rather than being either soldered or cross-connected
with a good "punch-down" system of some sort.

- Use of old-style non-twisted-pair "station cable" or (worse yet)
flat "satin" cable for the premises wiring. This leads to poor
rejection of RFI and hum. Running phone wire right by power wiring
can induce hum, and this makes life more difficult for the modem.

I can't give any objective estimate as to the amount of trouble any
one of these can cause, but I believe that all of them can have some
effect.

When looking at and diagnosing your existing premises wiring, I'd
start by looking at all of the connectors - they're probably the
weakest link. Check the wire itself (wherever it's accessible) to
look for dubious connections, breaks (rats and mice sometimes chew up
the cable), or unneeded branch connections which could be eliminated.

Pulling new wire is probably a measure of last resort but might be
necessary. The closer you can get to an "Ethernet-class" phone wiring
from your modem's connection point back to the MPOE, the better. A
single run of twisted-pair cable (CAT-3 or better), punched down onto
good-quality insulation-displacement-type RJ-11 terminations, with no
branch circuits and thus nothing else sharing the line, is probably
the best approach I can recommend.

--
Dave Platt AE6EO
Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!