[RFI] High Noise Levels on 80m
w8ji at contesting.com
Tue Dec 19 22:21:13 EST 2006
> In a 500Hz CW bandwidth, I typically have about S0 or S1
> noise on all
> bands except 80m. Sometimes noise jumps up to as high as
> S6 on other
> bands, but most of the time it is pretty low. 80m, on the
> other hand,
> is S7 to S8 24/7/365.
Several factors enter the equation on 80 (and especially
160). Ground wave goes much further, antennas are closer to
nearfield noise sources in terms of wavelength or antenna
dimensions, and noise is conducted along wiring much
By the way ground wave can only be vertically polarized on
low frequencies. This is because the earth shorts any
horizontal electric field, and when the time-varying
electric field is taken to zero so is the magnetic field.
Thus the only thing that propagates along the earth any
distance is vertically polarized. Keep this in mind.
> Question 1 - I know that 80m has more noise than 40m, but
> should it
> really be 6 or 7 S-Units higher than 40m? That seems like
> a lot.
Well, the antenna even if at the same height is twice as
close to other things in terms of wavelength. Worse yet, you
have the ends near ground.
Now all dipoles respond to vertically tilted waves to a very
noticable extent, and that response is maximum off the areas
of the end. It only goes perfectly horizontal broadside to
the antenna. The Inverted Vee dipole, while perfectly
horizontal in polarization broadside to the ideal antenna
like any dipole, has increasing vertical response as we move
around towards the ends. The Inverted Vee dipole actually
has enhanced vertical propagation off the ends.
So you have three things against the antenna being quiet:
1.) The frequency is lower so the antenna is electrically
closer to any noise source
2.) The ends are dropped very low, so the antenna has
enhanced vertical response in directions near the end
3.) The antenna itself is probably closer to noise sources
Now there is one more issue. Many people do not use good
baluns on dipoles and so the feedline and everything
connected to the feedline is part of the actual radiating
system. This means the radio chassis, the power line
connected to the radio chassis, and everything generating
noise on that power line has a direct wire connection to the
antenna. This can be a major noise contributor in today's RF
> Question 2 - Normally, the noise level in a 2.4KHz SSB
> bandwidth is
> considerably higher than in a 500Hz or 300Hz CW bandwidth.
> I can see
> the S-Meter drop several S Units when I switch between
> wide and narrow
> bandwidths. I do NOT see this same drop on 80m. There is
> no drop
> when I switch bandwidths or if there IS a drop then it is
> less than 1
> S-Unit. This doesn't make sense. Can anyone explain to
> me why this
> would happen?
If the noise is broad spectrum random non-coherent noise the
detected noise power is directly tied to filter bandwidth.
Half the bandwidth is half the noise power. Until you get
down to the BW of the signal S/N will keep getting better.
If you are not seeing that relationship and the receiver
bandwidth is really narrowing, it could be your noise has a
phase relationship that varies with frequency. Noise a few
kHz away might not add, it might even subtract. It could be
your 80M noise has a very high peak to average power ratio
and when you narrow the filters the noise bangs the filters
or some other circuit so hard it distorts or rings, keeping
the average level up. DSP radios with all the selectivity
following ten stages and a none-too-costly A/D converter
trying to sort out all those rapidly changing voltage levels
without overflowing like a toilet in Tijuana can be (and
often are) a problem. You might try using as much
attenuation as you can in case very high sharp peaks on the
noise (compared to averaged power of the noise) are
overloading or overdriving something in the receiver.
More information about the RFI