Hello Guys,
To know sure how much a vertical antenna with radials would have interaction with a horizontal yagi below (and vica versa) you would have to model each situation.
Since not all can do that, it is fair to say in almost all cases the groundplanes will have (slight) negative infuence.
THE SOLUTION:
Make sure all or at least most elements of the beam are dc grounded to the boom.
If this isnt the case allready mechanical, you can short-circuit the center pieces of the elements towards the boom without getting into trouble.
Some manufacturers have the elements isolated on the boom, often the non-metal"clamps" are better in that way as elements tend to brake near the boom.
And there is a thing called boom-effect.
This is more important on higher frequencies, but what it comes down to is:
As soon as years go past, the electrical bonding between element and boom get less.
Therefor the optimum element lenght changes slightly. (but on 11 meters you would need to be really precise in order to measure this.)..
Now, if the elemenents are already isolated and optimised in that way, nothing will change.
Oke, back to that Vertical with radials ontop of a beam...(were at elements grounded at the boom)
Then loose all the groundradials on the 5/8 wave and place the 5/8 wave just above the yagi.
The entire yagi will work as a "radials" wich is a better system them just 4 radials 3..4 feet long
You migth need to retune the vertical slightly.
Oke, then the imax/a-99 etc. kind of antennas can just be placed above the yagi without effecting the yagi in either way.
There is a slight advantage aswell since those antennas do get better with the additional "yagi-groundplane"- system.
@eddie, oke to be precise and i know you like to be...just some thougths for you:
Have you considered to model the boom as mechanical part of the antenna in this case.
Have you thougth of a split source in order to "feed".
antennas are 3dimensional and have many aspects. We could show a beautifull elevation pattern but the azimuth pattern migth be screwed. So you migth wanna take a look at that aswell.
Last thing here, but the first to take a look at is the SWR of the beam and how that is influenced.
The first indicators there is something "wrong" is the SWR. A good example is the vertical beam attached to a metal mast, that often gives SWR problems..without modeling you know the antenna has some issues.
Kind regards,
Henry 19SD348
All about antennas