I'm not saying that there's not a case for suppressing common mode as you
and others suggest.
It all comes down to impedance ratios and signal levels. Any choke or filter
is like a pi network or attenuator pad. It is ***always*** more than the
series element alone.
It the line has a very low common mode shunting path impedance, a very small
series impedance adds a very large amount of additional isolation. This is
the way most lines in houses and Ham shacks are, because they connect to
things have low shunt impedances.
Cables that are buried, or very long, establish fairly low impedances.
Ground rods might help, or might not.
Just think of the shield as an antenna you are feeding. Imagine an aerial
span over a road or driveway. If you had a little 20 foot long half square
feed through a buried single wire feeder, that "antenna" would not be a good
antenna at all. It could not greatly affect the rest of the system, because
its small contribution to the shield would be lost in attenuation of the
lossy single wire feeder near earth.
Also, if you have a GOOD system that meets codes, you have every single
shield grounded at the house entrance point. That alone establishes a low
impedance for things conducted in or out on the line shield. It takes a very
minimal series shield impedance to make an effective barrier at that point,
if anything at all. (My system require nothing at all.)
I don't use ground rods on shields, and generally don't use beads, unless
the connector is problematic. I have Beads slipped over cables (just ONE
pass) to eliminate problems from my phono connectors that swap RX antenna to
the contest barn. If a connector comes up to a fraction of an ohm, even a
30-40 ohm bead isolates the line. If it is high resistance or open, nothing
would fix it. So in this case a just a single bead is more than enough. The
"system" determines what is required, not some arbitrary number.
I do have a problem here lately with RFI from an electronic fluorescent
ballast on 160 (10 feet from where I operate the radio), and I'm not sure
how it's getting into my NE Beverage. Big #43 beads on the ballast leads
didn't fix it. But that's another story.
There are two forms of noise egress, common mode and differential mode. As a
matter of fact if you get into the system deep enough, there has to be a
differential mode somewhere at the source or load or the common mode is
meaningless.
Everyone ignores that and the shunting impedances, and that is what gets
them in trouble and require the abnormal impedances some systems might
require.
We can string beads over the wires for a hundred feet and not change a thing
if the excitation is differential mode.
For common mode, we add beads with hardly any change for common mode if the
shunt impedances are high. At the same time, one single 20-100 ohm impedance
bead can clean a system right up with 50 dB attenuation or more, if the
shunt impedances are low.
It seems very few people look at the system, which isn't that hard to do.
73 Tom
_________________
Topband Reflector Archives - http://www.contesting.com/_topband
|