Date   : Fri, 13 Jun 1986 03:58:13 GMT
From   : Larry Lippman <larry%kitty.UUCP@seismo.css.gov>
Subject: Re: phone line surges and spikes

In article <3607@reed.UUCP>, kamath@reed.UUCP (Sean Kamath) writes:
> 
> Well, there has been a lot of talk lately about line voltages, and I
> don't know if this has come up.  I know it is possible to buy surge
> suppressors for AC & phone lines.  I know there are such thing as phone
> line spikes (I hear them at 1 am on my ultra-cheap phone from time-life
> books, or whatever.), so the question is, is it as easy as installing a
> MOV in the line?  I'm sort of worried what might be getting rammed into
> my internal modem.  Anyone got any ideas?

Most telephone line transients result from lightning strikes causing 
current to be dissipated through the sheath of the telephone cable to 
ground. This INCLUDES cable which is buried underground. A voltage is 
induced in the individual telephone cable conductors due to the HUGE 
currents and magnetic fields which accompany a lightning strike (to 
give you an idea how large the currents are, lightning strike currents 
are measure in KILOamperes, with a typical lightning strike being 100 
kA).

Except in extreme cases, the lightning voltage itself therefore never 
hits the individual conductors (which would cause destruction of 
dielectric), but its propagated through capacitance and inductance 
between the cable sheath and its pairs. The resulting transients are 
"longitudinal" in nature, being from each wire of a pair to ground; 
these transients are NOT generally from conductor-to-conductor.

What does the above mean in practical terms? For one, placing a MOV 
(Metal Oxide Varistor) just across tip and ring will do almost zip for 
providing transient protection. You need TWO MOV's, one from tip to 
ground, and one from ring to ground. And I mean GROUND, like a #12 AWG 
wire to a water pipe - NOT the third wire of an AC power outlet. Because 
the MOV is not going to be very effective unless a low impedance path 
to earth ground exists.

So, it is possible to build your own telephone line surge protector 
using two MOV's as indicated above. However, it is not quite that 
simple since the MOV's have to be carefully selected. First, the MOV 
should exhibit a leakage current flow at 50 volts DC of LESS THAN 500 
microamperes, which is equivalent to a resistance of at LEAST 100,000 
ohms. At 130 volts DC, the leakage current should still be less than 
1,000 microamperes. Exceeding these values can CAUSE TROUBLE with your 
telephone line, such as: (1) false ring tripping; (2) hum and noise due 
to longitudinal unbalance; and (3) trouble indications at the telephone 
company central office due to automatic line insulation testing 
equipment thinking your line is in trouble. The effective breakdown 
potential of the MOV should be around 180 volts DC to preclude any 
false trouble indications.

In my opinion, any product which claims to be a telephone line surge 
protector and consists of a single MOV across tip and ring is ALMOST 
WORTHLESS. You need to go between the conductors and ground, for the 
reasons mentioned above.

I don't want to get off on a tangent here, but there is something which 
is better than a a MOV: a gaseous discharge tube, especially what is 
called a "three element" gaseous discharge tube. These devices exhibit 
an infinite impedance in the "off" state, and conduct much faster than 
a MOV. In telephone parlence, these are referred to as "rare gas 
protectors". However, the cost of such a component is several times the 
cost of a MOV.

==>  Larry Lippman @ Recognition Research Corp., Clarence, New York
==>  UUCP  {bbncca|decvax|rocksanne|rocksvax|watmath}!sunybcs!kitty!larry
==>  VOICE   716/688-1231                {rice|shell}!baylor!/
==>  FAX     716/741-9635 {G1, G2, G3 modes}        seismo!/
==>  "Have you hugged your cat today?"             ihnp4!/