Path: utzoo!telecom-request
Date: Thu,  6 Jun 91 18:37:03 CDT
From: Al L Varney <varney@ihlpf.att.com>
Newsgroups: comp.dcom.telecom
Subject: Re: Pet Peeve About Newer Modems (was Telephone Keypads)
Message-ID: <telecom11.432.5@eecs.nwu.edu>
Organization: AT&T Network Systems
Sender: Telecom@eecs.nwu.edu
Approved: Telecom@eecs.nwu.edu
X-Submissions-To: telecom@eecs.nwu.edu
X-Administrivia-To: telecom-request@eecs.nwu.edu
X-Telecom-Digest: Volume 11, Issue 432, Message 5 of 7
Lines: 20

In article <telecom11.426.1@eecs.nwu.edu> jmason2@utcs.utoronto.ca
(Jamie Mason) writes:

>	Anyone know *why* 35ms (or 36ms) seems to be a universal limit
> on modern exchanges?  Is this designed in?

Yup, designed in.  In order to meet the goal of less than one
occurrence of false digit simulation in 1500 calls from room noise,
etc., the digit tones have to be present for a period of time with no
other significant frequencies, except dial tone must be ignored.  Thus
the standard for requiring 40 msec. as a lower bound, for both the
digits and the inter-digital interval.  Switches should reject digits
shorter than 23 msec., and since the loop can distort signals somewhat
(Right, Toby?), the transmitter should use about 50 msec.  Switches
use 50 msec. when using DTMF to PBXes.


Al Varney, AT&T Network Systems, Lisle, IL  varney@ihlpf.att.com