Aerials and their Q

Most aerials used in amateur radio are resonant; either self resonant, where the antenna is a multiple of half wavelengths long, or brought to resonance by one or more components such as in an Aerial Tuning Unit. This is because resonant aerials are very efficient for their size. This in turn is because each “input” of current injected into the aerial flows to and fro for approximately ‘Q’ cycles until all its energy is radiated. This may be envisioned as the electric charge bouncing from end to end of the aerial, (even if one end of the aerial is the image in the ground). Also, the voltage, (at the ends of the aerial), builds up to Q times the input voltage at the centre of the current maximum, which is often the feed point. In contrast, a non resonant aerial, such as a “Beverage”, or a “Rhombic”, (used at HF), or a horn, (used at microwave frequencies), usually has to be several wavelengths in size to function efficiently. However, these non resonant types can usually cover a broader bandwidth than resonant types, often extending to many octaves.

Recently, a short article was written in the Newsletter about the meaning of “Q factor”, and this was applied mainly to tuned circuits. It was shown that the higher the Q, the lower was the loss and the narrower the bandwidth of a tuned resonant circuit. Can similar reasoning be applied to aerials? Resonant Aerials have a “preferred operating frequency” around their resonant frequency and this defines their “preferred operating bandwidth”, so do they have a Q?, and should we endeavour to make it as high as possible? The answer is complicated.

Firstly, only single resonant circuits can be said to have a Q factor. Two or more coupled resonant circuits have a more complicated response to an input frequency than the usual “bell shaped plot” of volts or current against frequency. As explained in a previous article, coupled circuits may have a relatively flat response over a restricted frequency band, or even a double or multi-humped response. In the context of aerials, reflectors and/or directors count as coupled tuned circuits.

Secondly, there are far more factors to take into account when describing the performance of aerials, than with simple resonant circuits, and therefore their behaviour with changing frequency is more complicated. Aerials are primarily designed as “transducers” between the voltage and current in their feeder wires, and the electro-magnetic wave which they produce in space. Factors such as impedance match, gain, “directivity”, and reducing radiation in unwanted directions have to be considered.

A simple aerial can have a unique Q factor. By “simple” in this context is meant having only one radiating element, such as a dipole or a quarter wavelength monopole over a ground plane. Once another resonant element, such as a reflector or director is introduced, the whole concept of its resonant or operating frequency and its bandwidth are amenable to different interpretations. For example, an aerial may be required to have low near sidelobes, (for direction finding for example), or a minimum or maximum beamwidth, or best possible front to back ratio, or maximum gain irrespective of all the other desirable features over its operating bandwidth. Not only may optimising one feature degrade some or all of the others, but each desirable characteristic may be at its best at a different operating frequency! Aerials such as Yagis and cubical quads can have complicated responses within their operating bandwidths to say nothing of multi-banders.

Let us concentrate on simple aerials containing only one resonant element, such as dipoles, monopoles with ground planes, Hertzian doublets and Marconi aerials together with their tuning systems where required. These definitely have a Q factor which can be defined. The first three of these may be described as “all metal aerials” as, in an ideal situation, the electric field flows from one metal part to another metal part with no resistive or loss inducing component in its path. The Marconi aerial is an exception to this because, although it is basically a monopole or inverted ‘L’ over a ground plane, the ground plane is usually natural earth, not the idealised earth system which could be provided, for example, by a whip aerial over an all metal car roof at VHF, where there is very little wasteful resistive loss.

In general, aerial systems which use the natural ground or earth as part of their resonant system incur significant wasteful loss, even if several deep ground spikes are used. If circumstances permit, it is always better to use highly conductive, (i.e. copper), radial wires over the ground adjacent to the aerial wire.

Resonant aerials have two main loss mechanisms; wasteful resistive loss and radiation resistance. The wasteful resistive loss of the ground connection, (where this is used), can be several Ohms to several 10s of Ohms, (where only one or a small number of earth spikes are used). The loss in an Aerial Tuning Unit, (where this is used), is also wasteful and can amount to several dBs, particularly when it is required to match a high impedance aerial to 50 Ohms. Loss in the actual aerial wire is usually only an Ohm or two, unless excessively thin wire is used. Such losses serve no useful purpose. However, radiation resistance is entirely another matter. This is the “damping”, of the resonant circuit, or extraction of energy from the aerial system caused by turning the voltages and current in the aerial wire into radiating electro-magnetic waves, and is what is required. So for maximum efficiency, the radiation resistance should be large compared to the wasteful Ohmic losses.

In an all metal system such as a dipole well away from ground, virtually all the loss is due to radiation resistance. Can anything be done to lower the Q and thereby increase the bandwidth without introducing dissipative loss? Fortunately there is, and it can be achieved by increasing the diameter of the conductors. However, this has to be done in a rather exaggerated manner to obtain a modest reduction in Q and an increase in bandwidth.

In a normal aerial, whether wire at HF, or metal rods or tubes at VHF, the Q due to radiation damping is about 10, and the usable bandwidth about 10%. By making the diameter of the conductors about one quarter of the aerial length, the Q can be reduced to about 5. Aerials of this type, where each arm of the dipole is constructed of several wires connected in parallel are known as “cage aerials”. They can sometimes be seen as vertical dipoles on the control tower of civil and RAF airfields, and are designed to cover the whole of the VHF airband from 118 to 136MHz, a proportional bandwidth of 14.2%. Although not very applicable to amateur radio, the use of two or three spaced wires may be a useful improvement on the wider bands such as 160, 80 and 10 metres. Why “fat” aerials should have a broader bandwidth is not easy to envisage, but an eminent aerial expert once described the fat aerial as “having a greater grip on the aether”!

Of course, aerials tuned against natural ground can have low Qs and thus a broader bandwidth because of dissipative loss due to the resistance of the ground, but this is not an efficient way of achieving wide bandwidth. Unless the broad bandwidth is required simultaneously, i.e. at opposite ends of a band, it is perhaps better to use an ATU to tune the aerial to each required frequency.
John, G0NVZ