Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!ub!kitty!larry
From: larry@kitty.UUCP (Larry Lippman)
Newsgroups: sci.electronics
Subject: Re: Telephone Lines
Summary: Characteristic impedance of telephone lines
Message-ID: <4651@kitty.UUCP>
Date: 29 Jan 91 05:09:31 GMT
References: <4574@mindlink.UUCP> <1991Jan26.163014.1986@athena.cs.uga.edu>
Organization: Recognition Research Corp., Clarence, NY
Lines: 51

In article <1991Jan26.163014.1986@athena.cs.uga.edu>, mcovingt@athena.cs.uga.edu (Michael A. Covington) writes:
> >While in a meeting, one technical "guru" declared that my statement was wrong
> >when I said that a telephone was a form of transmission line.  The "guru"
> >declared that, not only was I totally wrong, but that he and a senior
> >engineer in the company had "proven" that a telephone line was not a
> >transmission line nor could it be modelled as such.  So, is a telephone line
> > a form of transmission line or am I totally wrong as this "guru" declared?

	With respect to the original author's article, this alleged "guru"
doesn't seem to know much about the definition of a "transmission line"!

> It certainly is a transmission line; in fact, long phone lines have
> 88-mH inductors inserted into them periodically to keep the transmission
> line impedance closer to 600 ohms (which is the source impedance of a
> telephone set).

	A non-loaded transmission line of sufficient length does a nice job
of establishing a characteristic impedance (Zo) at a given frequency *without*
the use of series inductors (loading coils).  The purpose of loading coils
is to "counteract" the effect of distributed capacitance and reduce the
insertion loss at frequencies below 3,000 Hz.  A loaded transmission line
has a comparatively flat curve of insertion loss -vs- frequency up to a
cutoff frequency determined by the loading design and cable characteristcs.
Such a cutoff frequency is usually between 3,000 and 3,500 Hz.

	It is important to understand that Zo of a telephone transmission
varies with the frequency.  The impedance specification commonly used in
communication apparatus generally refers to a frequency of 1 kHz.

	It is also important to realize that Zo from an equipment design
standpoint is merely a *compromise*, represents an engineering guideline,
and is not some magic absolute figure.

	Non-loaded telephone transmission lines are generally considered
to have a Zo of 600 ohms @ 1 kHz.  More specifcally, 22 AWG cable has a
Zo of around 560 ohms, and 24 AWG cable has a Zo of around 760 ohms.  The
finer the gauge, the higher the Zo (which should suprise no one).

	Loaded telephone transmission lines for voice frequency service
are generally considered to have a Zo of 900 ohms @ 1 kHz with H-88 loading
which uses 88 mH loading coils space every 6,000 feet.  More specifcally,
22 AWG cable with H-88 loading has a Zo of around 1,050 ohms, and 24 AWG
cable with H-88 loading has a Zo of around 1,080 ohms.  Note the small
difference in Zo between the same wire gauges loaded versus the larger
difference unloaded.  26 AWG cable with H-88 loading has a Zo of around
1,170 ohms, which is why some subscriber loop apparatus, such as 2W/4W
terminating sets, is also available in a 1,200 ohm impedance.

Larry Lippman @ Recognition Research Corp.  "Have you hugged your cat today?"
VOICE: 716/688-1231       {boulder, rutgers, watmath}!ub!kitty!larry
FAX:   716/741-9635   [note: ub=acsu.buffalo.edu] uunet!/      \aerion!larry