[R-390] Re: Line filter issues?

[Bill Smith]

Welcome to the world of low-frequency interference troubleshooting!

You will run across a number of strange stories and experiences, all as grey
as the plains of Kansas in the Wizard of Oz.  They're all caused by the
Wicked Witch of the East and her switching broomstick.

First of all, the noise sources you are hearing are generally produced by
square waves.  Sources are a sparking contact, a light-dimmer turned almost
all the way on (maximum current switching = good transmit power), a
defective part that is arcing, etc.

The signal has many harmonics, and is generally coupled to the power line in
the house.   Of course it travels right outside the house to the power pole
and down the street.  It generally is a very complex wave, thus has many
peaks and nulls.  You'll have quite a time mapping it with a transistor
radio on broadcast frequencies.  Phasing and intensity will give you all
sorts of indications, particularly when filtered through the ferrite-loop
antenna of the radio which is itself very directional.   Try walking down
the street some time with a bad street light.  You'll discover all sorts of
peaks and nulls that seemingly have nothing to do with the offending pole
(assuming you have already spotted the dimming light).

The secret is to use your 2m handheld tuned to the aircraft band.  Or at any
rate, a high-frequency receiver with an AM detector.  The higher harmonics
are attenuated somewhat, and are more easily adsorbed thus don't travel as
far.  Thus you will have a better chance of zeroing in on the real source of
the noise.

 If you can attach a small 3-element beam to the antenna, so much the
better.  This is essentially what is used by the electric company when a
representative comes out with a noise sniffer.  The unit is a regenerative
receiver tunable from 300-350 MHz mounted on the end of a 4-element beam.

There are two ways of eliminating noise.   One is to divert it, adsorb it;
another is to attempt to ignore it.

Diversion, if you will, can be accomplished by good grounding.  Remember,
though, that any wire exhibits inductance with length.  Even a 3-foot ground
wire from a receiver to earth ground may be too long to provide a good
"drain" because of its inherent inductance.  Naturally, a 20-foot ground
line from a second story to ground will not improve things, in fact, will
not be effective except perhaps for frequencies below, say, 500 KHz.  A good
ground is never-the-less the first thing to install.

Another way is to attempt to adsorb the noise before it gets into the
receiver.  Radio Shack interference filters can be very effective.  I don't
have a catalog number handy, but look for the square devices.  They are
actually two "C" shaped sections held together by a snap-together plastic
shroud.  Wire is wrapped through them (as many turns as possible).   Make
sure the ends of two ferrite halves meet together, otherwise  currents
cannot not circulate in the ferrite and the filter won't work as well as it
might.  They can be installed in line cords, speaker wire, control wire,
even coax can be wound through them.   If you install them on the back of a
radio, keep the wiring between the filters and the radio as short as
possible.  The wire between the filter and the radio is unsheltered antenna!

They are also great for suppressing noise from a computer.   Start by
grounding the computer case.  Then,  unplug everything from the computer,
turn it on, turn on the receiver, and plug in computer cables one by one,
noting any increase in noise.   Install filters on offending connections as
close to the computer as practical.  If you can't get them to make a
difference, somehow the noise is traveling through another path and you'll
have to search to find it.

You can attempt to "ignore" noise by use of common-mode approaches.  This
approach makes use of something called a balanced line.  You will notice
that the antenna input connections to the R-390 are balanced, and so are the
audio output connections.  A full explanation of this approach is better
found in text books, but basically the idea is that noise will be induced
equally in two balanced lines.  If the lines are connected properly, the
noise can be nulled out.  Wire telephone lines make good use of this
approach.

Unfortunately, Collins took the balanced line approach very seriously and
established a ground at center of the two balanced antenna inputs.  In fact,
instructions to adjust a capacitor divider that establishes this ground are
provided in the alignment instructions.   The ground point in the receiver
is away from the back of the cabinet, and there are all sorts of sneak paths
that may be particular to an installation which disturb the balance at RF
frequencies.  At any rate, while not perfect, it doesn't hurt to attempt to
feed the receiver with a balun at the balanced antenna terminals.   You will
know if you are effective, if you can short a section of coax that feeds the
antenna line to the balun and the receiver falls absolutely silent on all
frequencies.   The receiver here is quiet, although some strong broadcast
stations can still be heard when the receiver is tuned on frequency.

The same approach can be used on the power line and the audio lines.  Power
Isolation transformers are used in some commercial broadcast installations
to balance power lines.  There, the center-tapped secondary of the isolation
transformer is grounded.  If you have an isolation transformer without a
center tap, a virtual ground can be established with capacitors (.5 mfd, AC
rated) from each secondary line to ground.  The capacitor value isn't
critical, but make sure the caps are rated for very high peak voltages.
Good to protect them with MOV's too.

Audio lines are probably better protected with Radio Shack filter chokes,
but hum (developed from "ground loops") and some noise can be eliminated by
use of transformer input to external amplifiers.

Hope this helps, or at least provides food for discussion.

73 de Bill, AB6MT
billsmith@ispwest.com





----- Original Message -----
From: bw
To: r-390@mailman.qth.net
Sent: Tuesday, August 05, 2003 5:11 PM
Subject: Re: [R-390] Re: Line filter issues?


I don't have much help for you here, but I had a similar problem a while
back. I have a nice 6' rack that had my R-390A in it near the top. There was
a SP 600 beneath that, and a tube amplifier below the amp. A variac sat on
the base of the rack below all of the gear with a lot of metal between it
and the top of the rack. I had bought the Quantum QX Pro loop which is
incredibly sensitive on MW/LW. Well, the rack made a great reflector in the
shack with the loop located on a lower shelf unit so I put the loop on top
of the rack. I have a swivel chair that sits kind of high up and I could
rotate the loop well enough for the time being. I was getting terrible
buzzing everywhere and the loop was just about unusable in this setup. I
finally moved the loop and my LF receiver to the other end of the house
where it is very quiet there. A few months later I brought the QX Pro back
to do some testing with the R-390A again. The buzzing was gone. The only
thing that I know of changing was the variac location. I had moved it to the
side of the rack, sitting on the floor. Anyway, the buzzing is gone now. My
guess was that the variac was causing the problems but I never checked into
it.


Barry non-Hauser


> Here's more on the issue. A couple of weeks ago, when testing a small
> homebrew amplifier/speaker combination to use with the diode load input, I
> noticed
> some buzzing coming from the speaker. At this point, the hum was not
audible
> from the set's audio on any band, with much of it unnoticeable if there
was
a
> good
> signal from the receiver, or if the speaker was moved away from the set;
it
> would go away completely when unplugging the variac setup I always use to
> power
> up my radio, and was still present when using no variac whatsoever.
However,
> during a recording session of a local BC oldies station, the hum began
coming
> through over the BC station with force. At first, I attributed it to the
fact
> that I had taken off the Utah plate for display purposes, but putting it
back
> on made no difference whatsoever. Earlier today, I decided to see if the
noise
> was still coming through, and it was still present, so I began turning off
> various appliances in the house (none of the other line-powered BC sets
I've
> tried have exhibited this interference), but this had no effect on the
> buzzing. I
> then tried using a small transistor radio tuned to the same station to
trace
> the source of the buzzing. I got clear reception until I got within a few
feet
> of the receiver. I thought that it was perhaps the homebrew variac setup
> which was causing the interference, but trying a commercial variac instead
had
> no
> effect, and unplugging the R-390A from either variac got rid of the
buzzing
> altogether. I can live with the problem for now, since it doesn't appear
to be
> affecting most of the SW bands, but I would like to find a solution which
> wouldn't affect the look of the set too much (i.e. I'd rather not have a
IEC
> connector present in the back of the set, but if there's no way around it,
so
> be it;
> I'll just have to replace the molded plug on the new IEC cord with the
> military-style three-prong plug from the old one!). How hard is it to
access
> the old
> line filter without displacing the rest of the radio? TIA.
> -Adam
> (BTW: I tried sending this twice from AOL Web Mail, as suggested in the
> NoMIME FAQ, and it was rejected as spam by qth.net's filters both times.
> Strange...)
>
> Adam Vaughn

_______________________________________________
R-390 mailing list
R-390@mailman.qth.net
http://mailman.qth.net/mailman/listinfo/r-390