Path: utzoo!utgpu!water!watmath!watcgl!awpaeth
From: awpaeth@watcgl.waterloo.edu (Alan Wm Paeth)
Newsgroups: sci.electronics
Subject: Re: phone dialing (touch tone synthesis)
Message-ID: <4937@watcgl.waterloo.edu>
Date: 20 Jun 88 16:06:30 GMT
References: <361@tiger.oxy.edu> <3310002@wdl1.UUCP>
Reply-To: awpaeth@watcgl.waterloo.edu (Alan Wm Paeth)
Organization: U. of Waterloo, Ontario
Lines: 26

> ... Original touch tone phones had discrete transistors
>and LC tone generator circuits switched by a matrix of contacts
>corresponding directly to the rows and columns...

Original touch tone phones had *ONE* transistor -- there is nothing
prohibiting an oscillator's generating two simultaneous frequencies, and
Bell's introduction of solid state electronics into the consumer phone
followed the KISS "keep it simple, stupid" philosophy to its extreme.
Phone system design in North America has always followed this design
principle -- clever circuits and architecture minimize the the "per circuit"
complexity of each line, thereby allowing a system of around 10^8 telephones to
be maintained by a huge decentralized pool of minimally trained maintainers
("the phone men").

As a second example, consider that a consumer phone line (one twisted pair)
allows simultaneous (full duplex) conversation, yet each speaker hears his
own voice in the headset at a much lower volume than the remote sender, all
this done without the benefit of digital, let alone solid state components.
Moreover, each phone contains no power supply -- the two leads to the exchange
supply power, bidirectional data and also allow for remote ringing, yet each
set contains only a handful of 1950's parts [but no vacuum tubes :-)].
I don't think most folks could draw the circuit off the tops of their heads.

     /Alan Paeth
     Computer Graphics Laboratory
     University of Waterloo