On Wed, 15 Jul 1998, Max Eskin wrote:
> Hey! A great way to bring this thread back on topic! How do those
> electronic switches work, and what is the approximate network topology
> of the system? Is there a reference guide to it somewhere? I'd love to
> take a look.
Well, it wouldn't be on topic. Telephony is a whole 'nother subject. Try
calling your local telco and say you're a student interested in learning
more about the phone network. If they ask if you're a phone phreak, say
you've never even heard of that term.
Or, a good book to start on would be _Understanding Telephone
Electronics_. Its a Sam's book but you can also find a Radio Shack
branded version at Radio Shack (they all carry it AFAIK). I recommend it.
Just a quickie, the phone network design has many parallels to the
Internet, or a basic datacom network topology. You've got your end
offices (sort of like level 3 routers), tandem switches (sort of like
level 1 baclkbone routers). Ask me more questions in e-mail.
> As far as I know, a telephone rings because when it's on hook, any
> current on the phone line is connected to the ringer. When someone is
> trying to connect, the company sends a periodic AC signal to the phone.
When a phone is on-hook, the ringer circuit is connected in parallel with
the line. The tip and ring (telephone-speak for positive wire and
negative wire, or commonly color-coded as green and red respectively,
except when dealing with a multiple pair cable) each have a capacitor in
series with the ringer. Like so:
+ (tip) - (ring)
| |
|--||--[BELL]--||--|
| |
| \ |
|______o \o______|
Hookswitch
The capacitors allow current to flow during the AC ring cycle (about 90 to
120v) which allows the bell to ring. Once the phone goes "off-hook"
(someone picks up the hookswitch and closes the loop) the telco senses
loop current and ceases the ring current, and puts 12v DC at ~20mA on the
line. When the phone is on-hook, there is a 48v potential on the line.
> If the caller ID data signal was sent first, it would cause jittering in
> mechanical ringers, and a maddening noise in electronic ones. When the
Nope, the caller ID data is a 1200bps data burst. Not enough current to
activate the ringer.
> ring is detected, the caller ID can 'pick up' the phone line briefly to
> check the data signal, which should be extremely brief. The station
> knows to keep ringing the phone if the user has caller ID. This is my
> theory.
Bzzt. But thanks for playing :) The caller ID box camps on the line with
a high impedance monitor circuit (basically has capacitors in parallel
with the line like the ringer circuit, with over-voltage protection to
keep the nasty ring cycle away from it. This allows the AC signal to pass
but prevents a DC loop which would look like an off-hook to the phone
company). If you ever see a telephone technician, ask him to demonstrate
his/her lineman set for you. It allows you to camp on a line and listen
to the conversation without actually taking the line off-hook.
Sorry for the off-topic. Please send all further queries to e-mail.
Sam Alternate e-mail: dastar_at_siconic.com
-------------------------------------------------------------------------------
Ever onward.
September 26 & 27...Vintage Computer Festival 2
See
http://www.siconic.com/vcf for details!
[Last web page update: 07/05/98]
Received on Wed Jul 15 1998 - 22:45:12 BST