Capturing a very limited frequency range is easy, but trying to concoct the ultimate wide band radio aerial is perhaps like the search for the Holy Grail. There are many approaches. Which one you take will depend on your knowledge, skills, facilities, resources, proposed usage and personal preference.
Modern HF receivers are sufficiently sensitive to pick up signals from the other side of the world on the proverbial piece of wet string. To my mind, sensitivity is no longer an issue the way it was two or three decades ago.
On the other hand, the radio spectrum is much more crowded than it was 2 or 3 decades ago. We are nowadays positively bathed in an impervious smog of electromagnetic energy. Our radio environment is polluted by a plethora of television, local radio, private mobile radio and - perhaps above all - by the ubiquitous mobile telephone heads, which now litter almost every high-rise roof top and grow like trees along motorways and railway lines. The traditional sources of interference like the car ignition, the vacuum cleaner and the washing machine are of course still with us. And what about this thing right here in front of me. It generates a constant concoction of square waves and pulse trains ranging from the 75 Hz time base of its monitor to the 2 GHz of its CPU's internal bus clock. A personal computer is not exactly Mr Clean when it comes to radio noise.
All these things contribute to the interference and noise that can potentially drown every distant signal I try to snatch from the mêlée. To my mind, therefore, it is selectivity that is now the issue.
Modern receivers are very selective. My receiver allows me to choose between a 6 kHz and a 2·5 kHz signal passband. It uses mechanical filters in the IF (intermediate frequency) section. These provide very precise and well shaped signal passbands. It is also a double conversion superhet. This pretty well guarantees that I will not get any unwanted images of strong signals from other parts of the radio spectrum.
However, when I look at the RF (radio frequency) section of my receiver - the part where the raw signal actually enters the receiver - I see only very crude filtering. It comprises only two switched filters for the whole of the LF/MF/HF range. One allows through all the RF energy coming in from the aerial that is between 100 kHz and 1·7 MHz. The other allows through all the RF energy coming in from the aerial between 1·7 and 30 MHz. That's an awful lot of unwanted energy hitting the receiver's first active device - namely its combined RF amplifier and mixer. Quite different from the old days when good HF receivers had one or two fully tuned RF amplifier stages - but these are not practical for the automatic search and scan functions demanded by today's short wave listeners.
I realise a modern active device like a MOSFET (metal oxide semiconductor field-effect transistor) can tolerate a wide range of input signal strength. Nevertheless, I am sure that presenting it with as strong and as clean a signal as possible will never do any harm and can only help when it comes to prizing a weak signal from the cacophoney of the HF radio spectrum. And this is the aerial's job.
An aerial can provide two kinds of selectivity - frequency selectivity (favouring signals within a desired frequency range) and directional selectivity (favouring signals coming from the a desired direction). These two kinds of selectivity reinforce each other. Their effects add. Consequently, the best kind of aerial is one that provides both kinds of selectivity.
The text book aerial is the electric dipole. It is a length of wire laid in line with the electic vector of the desired signal. This is perpendicular (at right angles) to the direction from where the signal is coming. The wire is cut to a length that will make it electrically resonant at the frequency of the desired signal.
The field strength of the electric vector of the signal wave varies in a cycle. The wire feels only a weak electric field that gradually builds up in strength until it reaches a maximum as the wave passes by. [The strength of an electric field is measured in volts per metre.] The electric field then starts to diminish (fade) back down to zero. It then starts to grow again in the opposite direction (the opposite polarity) up to a maximum again and then starts to fade again all the way to zero. Then the cycle repeats. It repeats at the frequency of the signal.
The presence of the signal's electric field along the direction of the wire induces a potential difference between one end of the wire and the other. If the signal's electric field were 30 micro volts per metre, and the wire were 20 metres long, then there would be a voltage of 600 micro volts across the whole length of the wire. This causes a current to flow along the wire, just as if you connected a very weak battery across it.
The current comprises all the free electrons in the metal of which the wire is made being propelled to one end of the wire by the force the signal's electric field exerts upon them. The electric field rises to a peak, falls back to zero, then rises and falls again in the opposite direction - all in a repeated cycle. This causes the metal's free electrons to swill back and forth from one end of the wire to the other.
One way of visualising this is to imagine the wire as a motorway with big car parks at each end. One end is suburbia. The other end is the commercial centre of a city. People commute along the motorway each day.
The problem with the dipole, of any given length, is that it is only resonant over a very small range of frequencies. I therefore think that, in view of the vast frequency range of the HF bands, the magnetic loop or coil aerial is a much better option for the short-wave listener. I describe this kind of aerial in the following essay.