Topband: Generation of mixing products (was 1.810 BCB interference)

[Tom Rauch]

> This continued for several overs, and included letters to 
> station managers, with suggestions as to how this spur 
> could possibly be generated. It eventually died on the 
> vine. The "ham" engineer really seemed to be in a fog on 
> the whole topic.

While bad connections are often thought of as causes in 
transmitters, that generally only applies to bad connections 
near or in the receiving end.

Near or in the transmitting antenna any bad connection is 
almost always quickly "healed" by arcing and becomes at best 
a very intermittent problem.

The common cause of this mixing is sufficient RF making it 
back down the feedlines into a transmitter. Good modulation 
linearity or an efficient PA stage demands having a very 
sharp switching waveform at radio frequencies...and this 
means the transmitter is a very efficient mixer for anything 
injected into the transmitter's output device. The 
conversion or mixing loss can be just several dB!

When two stations are very close together or using the same 
antenna any amount of RF backfeeding the PA stage of either 
transmitter will be converted to new frequencies. The 
transmitter's main frequency (and its harmonics) becomes a 
"local oscillator".

Take the 1550 and 1680 case. We have every sum and 
difference of the "local oscillator" and its harmonics, so 
we have a new signal possible every 130kHz up and down from 
the transmitters. Since this spur is 1810 the cause is 
1680*2-1550 = 1810.  This means the mixing causing the 160 
problem is almost certainly in the 1680 transmitter where a 
tiny bit of 1550 is modulating the 1680 PA stage. This is 
because the problem is the *second harmonic of 1680* minus 
the fundamental of 1550.

The first thing they should check is the notch in the 
multicoupler that removes 1550 from the 1680 transmitter. 
This would be the parallel tuned circuit that is in series 
with a series tuned circuit going to the 1680 input. If they 
sent a copy of the multicoupler schematic I could tell them 
how to adjust it. If the notch isn't deep enough then the 
best thing to add would be a 1810 trap on the 1680 
transmitter or an additional stage of 1550 notch on the 1680 
transmitter.

 It isn't simple to notch at the antenna but it could be 
done. It is much easier to notch at the offending 
transmitter.

73 Tom