Yet we have an analogue of this in real life, at least to some extent. If you take a 1/4 wavelength ground mounted vertical and put it over 120 ground plane radials you get very close to the same feed point impedance that you get as if the earth was a perfect ground, and modeling shows this as well.
The solution above is the basis of how to deal with such a problem, the main part of the solution is more wires, and a grid pattern works very well. However, they are not needed everywhere, you put them where the currents are going to be naturally the strongest, and that is at the feed point of the antenna. When you get further away from the feed point the currents are spread out over a plane, such as the metal roof of a vehicle, you can use fewer wires, as with a sheet of metal, their will be far fewer currents in this area.
In the past I have played with various models and irl counterparts, in various situations, at least as far as I am able to test, and the modeling software has accurately predicted to within a reasonable margin of error all of the parts I could confirm, which includes the feed point impedance. Things like the radiation pattern I could only confirm to the point that the pattern shown was in the general shape that the modeling software shows, but I only had a cheap field strength meter as a measuring instrument (I don't have a full line of lab grade test equipment unfortunately). I have also confirmed that if you use a grid of wires in place of a sheet of metal, assuming the grid is "fine enough", the difference between the grid and the sheet is very small.
I have also attempted to confirm my original Ford Explorer model's radiation pattern with the same method, and it didn't match up to what my field strength meter said it should. The model showed gain in some directions that the actual antenna didn't. The whole reason I put in the effort of making the second model of the same vehicle, which included things like curves in the design and such, is that the original model's pattern didn't match my actual test measurements. If it wasn't for that live testing, the second model would have never existed. As a bonus, that model also presents a much closer feed point impedance to what I actually get on said vehicle, and tuned resonance on the same frequency within 1/8 inch of length of the actual vehicle. Without more accurate equipment I'm not going to say it is perfect (I wouldn't anyway), but it is close enough to get an idea of what is going on.
I guess this ends up working as a response to an irrelevant comment fourstringburn made as well...
I wasn't going to say anything about this initially as it is the comment of someone who doesn't know me nearly as well as he seems to think, and honestly, it wasn't worth a response. It was pointless, and didn't add to the conversation in any way, shape, or form. In my opinion it says more about the person who said it than anything, especially when he was clearly wrong. You don't get to where I am with antennas on theory alone, but you can't get here with experience alone either, it takes both. The truth is I likely have my hands on far more antennas every year than he does. Some of these are repair jobs after a storm, some of these are custom made for where they are going, but most are pre-made antennas that someone bought and needed help putting up. That is what I get for being on an active and growing ham radio club's antennas team.
I would love to be able to afford a full lab of lab grade equipment to experiment with antennas. Hell, I would love to have more accurate equipment than I do just for testing purposes. Unfortunately I am only one person and don't have unlimited resources at my disposal. Instead I do what I can do with what I have, and that will have to do, at least for the time being.
The DB
