[RFI] Inter-station interference question.

David Robbins k1ttt.dave at gmail.com
Thu Nov 21 06:52:11 EST 2019

Since adding the stubs didn't do anything you should be looking either
before them or after them... 

things before the stubs to check are all the cables and their connectors,
case leakage from swr bridges, leakage on power to filters, band decoders,
band pass filter switching, radio power lines, etc... I found horrible
coupling between two older ICE 419 bandpass filters because there was no
filtering on the band select lines, the rf coupled from one filter directly
into the other via the band decoder.  Another one was daiwa swr meters that
use a floating full wave bridge on the dc input, any rf on the 12v supply
gets rectified and can do nasty things.

After the stubs you have all the normal external culprits that won't be
affected by any filtering, corroded rain gutters, rusty fence wire, etc...
but now there are also wall warts that can pick up rf, rectify it, and also
modulate it with their switching frequency making wide band noise.

David Robbins K1TTT
e-mail: mailto:k1ttt at arrl.net
web: http://wiki.k1ttt.net
AR-Cluster node: telnet://k1ttt.net:7373

-----Original Message-----
From: RFI [mailto:rfi-bounces at contesting.com] On Behalf Of Martin Sole
Sent: Thursday, November 21, 2019 04:22
To: rfi at contesting.com
Subject: [RFI] Inter-station interference question.


This is about a two radio inter-station interference problem. A bit
different perhaps but still RFI of a sort Looking for some input to help me
understand a bit better what is happening and maybe correct some errors in
my thinking.

I will describe the setup:

Two radios are FT2000, two amplifiers are OM2500, output filters are 4O3A
high power band-pass filters followed by dual, suitably spaced, stubs.
Antenna is an Optibeam, 2 element short 80, 3 element short 40, and 5
element full size 20 interlaced on a common boom. All feed points have large
#31 chokes installed at the feed point. Antenna is at approximately 50m and
feeder length to the station is approximately 49 metres of heliax type half
inch followed by 36m of RG214, 85m in total.

The 403A 80m filter swept with a VNWA3E shows -55dB at the second harmonic
(40m). The dual stubs show approx -55db at 40m, essentially no attenuation
in the band at 80m of course. Total attenuation around 90dB as far as can be
figured at the 7MHz second harmonic

With no filters or stubs fitted and the antennas connected to the radios
directly a 5 watt signal on 80m shows a harmonic measuring about S8 on the
40m radio representing about 116dB being the combination of the harmonic
attenuation in the radio/amplifier and the isolation between the antennas.

Inserting the 4O3A filter in the 80m feed line after the radio shows the
harmonic is reduced from the previous S8 to around S4 (circa 20dB?). 
Addition of the stubs after the filter shows the harmonic to be unchanged
from the S4 with the 4O3A filter alone. Testing in isolation with a Rig
expert AA230 to an SDR running on a battery laptop showed an attenuation of
the second harmonic of approximately 17dB this seems very much in line with
that seen on the radios.

The 2 element Optibeam 80m antenna when swept at the second harmonic of 40m
and at the end of the feed line shows 22-j65. The output of the
OM2500 with the tube replaced with a 2k2 load resistor and tuned to 80m
shows approximately 26+j413. Sim-Smith simulates the stub performance
identically as measured with the VNWA3E. Changing the input and output
impedance to match the values measured as above shows even more attenuation
at the second harmonic (40m).

Irrespective of identifying an ideal location for insertion of the stubs,
fully appreciated from reading the various W2VJN and K9YC articles on stub
location, I would expect to see even in the worst case a marked reduction in
the harmonic signal level over and above that we are experiencing. Having
measured the stubs and shown them to work as predicted and having tried a
couple of locations for the stub a metre or two either way with no
discernible change coupled with the poor apparent filter attenuation
suggests something else is the prime culprit.

Thoughts and suggestions very much appreciated.

Martin, HS0ZED

RFI mailing list
RFI at contesting.com

More information about the RFI mailing list