NOISE ADVANTAGE OF LONG WIRE ANTENNA FAR END TERMINATION

[glovisol]

BEVERAGE ANTENNA TERMINATION - 2

- THE FAR END TERMINATING RESISTOR

Test results compared at higher frequencies: Real part.

[glovisol]

BEVERAGE ANTENNA TERMINATION - 2

- THE FAR END TERMINATING RESISTOR

Test results compared at higher frequencies: VSWR.

[glovisol]

BEVERAGE ANTENNA TERMINATION - 3

- THE FAR END TERMINATING RESISTOR - MEASUREMENT DATA SUMMARY


Below is the data summary across 0 - 29 MHz in 1 MHz steps for |Z| and VSWR with a 390 Ohm load resistor, 3.6 m deep ground stake at the far end and 1.8 m depth ground stake at receiver's end. Straight dotted lines show parameter trends. As already mentioned, the graphs show that the ideal far end termination is greater than 390 Ohm and is in the order of 450 Ohm for more balanced VSWR across the HF frequency range. Figure 1 shows large |Z| variations and also large equivalent parallel capacitance variations at frequencies below 4 MHz: in fact the antenna is one wavelength long at 3.8 MHz and is too short at lower frequencies. Figure 2 shows that VSWR increases with frequency, but is always below 3.

Figure 3 shows the original antenna layout before modification. The near end (receiver) termination will be discussed next.

[glovisol]

NOISE PERFORMANCE COMPARISONS - 1

Another example of the noise advantage of modified terminations, on 14 MHz this time, near midday today, with highest man made noise.

[glovisol]

NOISE PERFORMANCE COMPARISONS - 2

More detailed comparison @ 7 MHz, after sunset, 19:10 GMT. The screens (taken seconds from one another) demonstrate that, with the modified termination setup, received signal level increases while noise level goes down by a very significant amount. This condition of course happens only if man made noise is present. If there is no man made noise around, then there is no noise differential between the two setups, while the increase in received signal is due to the better efficiency of the floating counterweight, as compared to the efficiency of the ground stake as antenna return at the receiver's end.

In other words it is not that the antenna "eats up the noise": it simply does not pick up the man made noise, if present. Therefore the noise differential between the two setups is not constant, but increases or decreases according to the man made noise present at tha instant.

In the next posts the antenna modifications will be described and discussed.

[glovisol]

NOISE PERFORMANCE COMPARISONS - 3

Shown below is noise & signal level differentials @ 3.8 MHz in early morning with little noise.

[glovisol]

NOISE PERFORMANCE COMPARISONS - 4

Shown below is noise & signal level differentials @ 21 MHz with very weak propagation at midday.

In compare 1 we have -110.9 dBm of....noise at 21.300 MHz, with compare 2 we have -105 dBm with solid copy at the same frequency. This really makes the difference.

[glovisol]

LOOKING AT THE ORIGINAL ANTENNA PLANT SCHEMATIC

The antenna work described in this thread was caused by the desire of improving my antenna system and is based on the following ideas:

1) I could improve the antenna by increasing the length to the limit of space available, e.g. from 135 to 160 m.
2) Reception was at times severely limited by man made noise, therefore I needed ways to limit noise pick-up.
3) My experience, dating back to the fifties, made me very suspicious of antenna returns to earth ground, as far as noise is concerned. The proven fact that I had the best noise performance with low capacitance isolation transformers indicated that my antenna return (e.g. stake ground) was probably giving me a lot of common mode garbage.
4) Concerning ground return efficiency (not considering the noise issue) I was aware that every antenna return has its own efficiency and different grounds (or different antenna returns) deliver different signal levels, depending on how efficient they are.
5) With my Network Analyser I could find the optimum far end termination resistance and obtain precision impedance data for building the isolation trnsformer.
6) Finally the likely existence of noise voltages between the antenna far end termination (point H) and the ground stake at the receiver's (point K) suggested that the way was to look at possible improvements in this area.

[sdrom33]

Glovisol it looks to me perhaps there is something wrong in your results. If you have a more sensitive antenna, it will also pick up more noise as well as the signal, just as Mike2459 shows in the reference in "Receive antenna & impedance "matching". Also I do not understand what you mean by counterweight.

[glovisol]

Hi sdrom33, there is nothing wrong with my measurements: this can be easily understood by considering we are only dealing with the suppression of LOCAL MAN MADE NOISE. Perhaps I did not clearly explain this point. It is entirely true that nothing can be done to minimise Galactic and/or Atmospheric Noise: they come from afar and follow parallel paths to the wanted radio signals, so they cannot be distingushed or separated by the antenna, apart from polarisation or directivity considerations. In this case it is very true that the better the antenna, the more the random noise received and only the signal stronger that this noise blanket will be received. On the other hand, apart from strong/exceptional atmospheric perturbations, this type of noise is a constant and depends on general propagation conditions.

Local noise is a different animal altogether: in general it comes from nearby machinery or devices and it can reach the antenna by way of electric or magnetic fields, it can be conducted by electrical lines and also through ground currents. Therefore, if we eliminate the ground from the receiving equation, we stand good chances of improving our receiving system. It is now more than two weeks that, having optimised the antenna far end termination, I get a consistent minimum noise advantage of 6 dB or better, at any time of day or night, when I replace the antenna ground return at the receiver's from the ground stake to a counterweight 80 m long and this happens at every frequency 1.8 to 29 MHz.

The counterweight is simply another conductor. If we were dealing with a 1/4 wavelength monopole, the counterweight would be our ground plane extenting for a circle of 1/4 wavelength radius at the base of the monopole: no connection to stake ground is necessary for this antenna to work. In the case of the Beverage antenna the ground plane for directivity is already formed by the poor ground surface underneath it, so the return termination of the receiver's input can be a counterweight. We can form a counterweight with a suitable floating conductor positioned in places away from the bottom surface of the long wire. Ideally this conductor should have little capacitance to earth and be large enough to have as good reception efficiency as possible. Now if we are able to make this efficiency higher than the efficiency of the ground stake, then we have lower man made noise because we pick up less of it and a larger received signal because our antenna is now more efficient. Please understand I have invented nothing: I have just removed some deficiencies of my antenna and made it better. This is what I am going to describe next.