Hi Chris and David
I tend to agree with Andy that practice doesn’t necessarily follow theory, especially when it comes to antenna systems because of the many variables that can affect the accuracy of the model. For example, because ground is an integral part of the antenna system it must be included in the model and does affect performance.
If the apex angle is too acute, relative to the vertical (which would be the case if the ends of the 20m dipoles legs are close to the ground relative to its feed point), the radiation pattern would still be similar to the 40m pattern in my article. On the surface all well and good, but the antenna would not appear to work very well; the mode of propagation needs to be taken into account.
On the 3 lower bands (80, 40, and especially 60 metres) we are exploiting near vertical incident skywave (NVIS) propagation where our signals are reflected by layers in the atmosphere almost directly above the antennas location, hence my antennas near vertical pattern (no side lobs). 20 metres is not propagated in this way and the usual practice is to try to get signals as far as possible; with an antenna close to the ground this can be a problem. Horizontal dipoles only a few metres above ground, as is the case with most portable installations, will have little or no radiation towards the horizon.
For 20 metres (and above) the antennas pattern needs to have components that send signals out at lower angles relative to the horizon. The minimum height above ground which begins to provide these angles is approximately 5 metres at the feed and 4 metres at the ends of the dipoles for a 20m dipole (modelled with 4nec2); in effect a shallow apex angle. In practice that means the either the ends of the dipole have to be supported in some way (not many trees on most of the highest summits) or the guy lines have to be very long or (and this is why linked dipoles work) for the 20m dipole to be a “link” in a 80m dipole.
So what I meant when I said other higher bands could be added I should’ve correlated that to say “in the 80m legs not as additional or separate dipoles“. The interactions mentioned are more to do with harmonics and the very complex relationships between the dipoles.
Basically what I’m saying is that if you have 40m as one dipole and 20m as the other the 20 metre dipole won’t be efficient as a radiator unless its feed and ends are high above the ground (difficult to do on most summits). If my design is changed too radically it would be better to model any new arrangement before lengths of wire are wasted and any fruitless trips are undertaken. If you want bands above 40 metres and are looking for DX contacts then I would recommend either the normal linked dipole arraignment or a vertical like the Super antennas MP1. My antenna was designed to specifically take advantage of a relatively short range propagation mode.
Happy to answer any questions here or by e-mail
BTW the vast majority of my portable operating is @ 5W