Last night I was reminded of a rather odd source of RFI. I keep a
older Radio Shack Pro34 in the car that is
locked to 121.5 [Civil Aircraft Crash Beacon frequency] I keep other
useful low and high VHF and UHF
frequencies programed as well. I needed to make some additions and was
reminded that this scanner
leaks the 10.7MHz IF signal like a sieve. That, and it's 200CH
limitation, are the reason why it resides in
the car. I suspect that many hand held scanners, or maybe any plastic
cased scanner or FM BCB radio
may also leak 10.7MHz. If I keep it more the about 3' from my R2000's
there is no problem. Just something
to keep in mind.


Let's say that you find you need a MW rejection filter, so you bought
a nice Kiwa low pass or high pass
filter and life is good. Then you connect your headphones, an external
speaker, connect to the Record/
Line out and find that %@^! MW station 3 miles away now has images
across the dial. Very few receivers
have any RFI filtering on audio inputs. And some receivers have none
on the power input. Our DX398's
suffer from MW overload spurs, in spite of Kiwa filters in the RF
chain, whenever external power is connected.
No big deal. I added RFI filtering in a small Altoids tin with about
3" of cable to the power jack. Similar
treatment for the audio lines solved those issues as well.


I don't know how many modern receivers have non coaxial aux antenna
inputs, but the R2000 has a set of
connectors for low Z (50 ohms) and high Z (500 ohms) that are
unshielded. On several occasions I have
experienced unexpected noise ingress via these connectors. There are
two options for the R2000, either
fabricate a shield, the RF inputs have adjacent ground connections,
and it was very simple to mount.
The other option is to open the R2000 up and bypass the aux RF inputs.
For years I ran with the 2nd mod,
but after I needed the high-Z input for some tests, I built a shield.


Another sneak path can be the AC mains input. I haven't investigated
this one with any complex tests,
but I did connect an AOR 7030+ and a known compliant Class B FCC
switch mode power supply to
the test side of a LISN. The LISN isolates the device under test with
inductors from the AC mains.
I terminated the antenna input on the AOR with a 50 Ohm 0.01% BNC
terminator. When I connected
the Class B switch mode the AOR received broad band noise. By adding a
inline RFI filter to either
the SM supply or the AOR the noise went away.

While I operate my R2000's from DC, I tried the same experiment and it
too allowed RF noise to ingress to
the receive chain across the internal power transformer. My wife, who
solders much better then me, added
CorCom RFI filters to each of our R2000s. (Don;t ask how many we own).
We had to use a CorCom because
my favorite two stage RFI filter is simply too large to fit in the
available space.


A note on telephone installations. The cord that connects the
telephone to the wall or the hand set to the
telephone base is called "silk" cable by telco workers. And it is a
very, very, as in extremely!! POOR!!!!
choice for any application where more it is longer then about 10'. The
4 conductors are side by side and
form a very good antenna and they are very effective as noise
radiators! In the typical installation the
yellow and black are either left floating or grounding. Neither option
is optimal.

In addition, most telephone installations use four conductor, 2
twisted pair, to connect the in house
phones to the NID. [NID Network Interface Device AKA Demarcation Box,
the box where responsibility
of the telephone company ends and the home owner begins.] Unless you
have two lines, I have found
a single unshielded twisted pair to be much quieter then the two
unshielded pair. And shielded cable
should only be tried as a last resort.

Modern telephone installations appear to be using Cat5 or Cat6 cable
to connect the NID to the home.
Unless you need 4 twisted pairs, I suggest this cable be avoided. Even
when the unused conductors
are grounded, significant noise can be radiated from Ethernet cable.


Coming up next, Noise reduction techniques.

Terry