I've been playing with ground mounted antenna models just to see what I can find out. This is something I couldn't do with the nec2 engine as even with the best real ground, you could only get so close before it became inaccurate. As I now have a license for the nec5 engine, I don't have that limitation.
All of these models are 40 meter ground mounted antenna models with various radial amounts and length. The vertical element is aluminum, and the radials are copper wire, and this is over moderate ground. I'm not going to post the outputs you typically see with models, they literally are all the same shape, just some have a different amounts of gain. Because of this, I will simply post the data in charts.
Also, the vertical element and ground element lengths aren't being adjusted for tuning at all. This is part of the testing as I want to see the changing in said numbers.
One thing I noticed with my first model, the ground mounted radials have little to no current flowing on them. This is because they are touching the earth. Typically, the fewer currents on an element, the less of an effect it has on the overall antenna, although in the past I have noticed some exceptions to that, for example what I called the "blue wire" on the Astroplane antenna, so that definitely isn't universal.
Some hypothesis that I initially had, some of which I posted previously on this and other forums, and the last two are claims I've hears several other people on this forum and other places as well tell others, so I figured, why not put them to the test.
1) I have stated in the past that radials that are on the ground don't work like elevated radials, and instead have the effect of increasing the conductivity of the earth while acting like part of said earth.
2) As such they are not length dependent as the signal has a medium (namely the earth) beyond said radials.
3) As we add radials, I expect efficiency to go up, X to go up (but only a little) and R to drop.
4) I have heard that as you continue to add radials, once you get past 32 radials you get diminishing returns, so I will test this with up to 64 radials. This will be tested.
5) I have also heard that for a given amount of wire, more shorter radials is better than fewer longer radials. This will also be tested.
So to start with 1/4 wavelength radial models, doubling the radials every model starting at 2 radials, ending at 64 radials.
Now for my subjective view on what I am seeing. If anyone has any thoughts, or thinks I am missing something, feel free to bring it up. (Especially if anyone has any idea what is going on with X).
Statements 2 and 5 are not testable with this data.
For statement 1 above, that seems to be right on the mark. Efficiency and gain goes up, and R goes down. This is what is expected.
Looking at the gain data, gain goes up as we add radials, but not at a steady rate. Peak difference as we add radials is going from 8 to 16 radials with a 0.7 dB difference. Then, after that we have a 0.6 dB, and finally a 0.5 dB difference in gain. While we see there is more gain to be had by continuing to increase the radial count, the rate begins to drop off. I would expect this drop off to continue when jumping to 128 radials.
When it comes to efficiency, we see something similar, we continue to see an increase in efficiency, but after 32 radials, the rate at which efficiency continues to go up starts to drop off. A lot of people are going to expect these numbers to be higher, but in my modeling experience they are right were they should be. In modeling and in my experience, the earth itself is the biggest enemy to a radio signal there is. This is where the phrase "height is might" comes from.
Impedance... R followed my prediction, as we continue to add radials the lower R becomes. That being said, we are very close to what many people will say is the limit for R over an ground plane (about 36 ohms), although I would suggest that 36 ohms isn't the limit many people think it is. That, however, would be a topic worth of its own discussion thread.
And finally, looking at X... What the hell? I predicted that X would continue to drop as the additional wires would increase the conductivity of the earth, which would effectively add capacitance to the system, which has the effect of lowering X. There is a reason why I think this way, a specific study on a different type of antenna design. That starts to happen... Then it does the exact opposite... I'm not sure what to make of this, either the added radials have little to no effect on capacitance, or something I'm not seeing is having more of an effect than said additional radials.
I am going to call this part 1. At some point I will add a part 2, and in that section we will look at the question of is more smaller radials better than fewer longer radials? At some point further down the road I would like to do a part 3 where I look at radial layouts that aren't evenly spaced, for example, you have an antenna mounted in the corner of your yard, so you have radials spread over just the part of the antenna that is in your yard.
The DB
All of these models are 40 meter ground mounted antenna models with various radial amounts and length. The vertical element is aluminum, and the radials are copper wire, and this is over moderate ground. I'm not going to post the outputs you typically see with models, they literally are all the same shape, just some have a different amounts of gain. Because of this, I will simply post the data in charts.
Also, the vertical element and ground element lengths aren't being adjusted for tuning at all. This is part of the testing as I want to see the changing in said numbers.
One thing I noticed with my first model, the ground mounted radials have little to no current flowing on them. This is because they are touching the earth. Typically, the fewer currents on an element, the less of an effect it has on the overall antenna, although in the past I have noticed some exceptions to that, for example what I called the "blue wire" on the Astroplane antenna, so that definitely isn't universal.
Some hypothesis that I initially had, some of which I posted previously on this and other forums, and the last two are claims I've hears several other people on this forum and other places as well tell others, so I figured, why not put them to the test.
1) I have stated in the past that radials that are on the ground don't work like elevated radials, and instead have the effect of increasing the conductivity of the earth while acting like part of said earth.
2) As such they are not length dependent as the signal has a medium (namely the earth) beyond said radials.
3) As we add radials, I expect efficiency to go up, X to go up (but only a little) and R to drop.
4) I have heard that as you continue to add radials, once you get past 32 radials you get diminishing returns, so I will test this with up to 64 radials. This will be tested.
5) I have also heard that for a given amount of wire, more shorter radials is better than fewer longer radials. This will also be tested.
So to start with 1/4 wavelength radial models, doubling the radials every model starting at 2 radials, ending at 64 radials.
| Radials | Gain | Efficiency | SWR | R | X |
| 2 | -2.5 | 16.59 | 1.08 | 50.7 | 3.66 |
| 4 | -2.3 | 17.9 | 1.08 | 49 | 3.55 |
| 8 | -1.8 | 20.32 | 1.11 | 46.3 | 3.45 |
| 16 | -1.1 | 23.8 | 1.18 | 43.1 | 3.45 |
| 32 | -0.5 | 27.53 | 1.26 | 40.3 | 3.55 |
| 64 | 0 | 30.51 | 1.31 | 38.6 | 3.71 |
Now for my subjective view on what I am seeing. If anyone has any thoughts, or thinks I am missing something, feel free to bring it up. (Especially if anyone has any idea what is going on with X).
Statements 2 and 5 are not testable with this data.
For statement 1 above, that seems to be right on the mark. Efficiency and gain goes up, and R goes down. This is what is expected.
Looking at the gain data, gain goes up as we add radials, but not at a steady rate. Peak difference as we add radials is going from 8 to 16 radials with a 0.7 dB difference. Then, after that we have a 0.6 dB, and finally a 0.5 dB difference in gain. While we see there is more gain to be had by continuing to increase the radial count, the rate begins to drop off. I would expect this drop off to continue when jumping to 128 radials.
When it comes to efficiency, we see something similar, we continue to see an increase in efficiency, but after 32 radials, the rate at which efficiency continues to go up starts to drop off. A lot of people are going to expect these numbers to be higher, but in my modeling experience they are right were they should be. In modeling and in my experience, the earth itself is the biggest enemy to a radio signal there is. This is where the phrase "height is might" comes from.
Impedance... R followed my prediction, as we continue to add radials the lower R becomes. That being said, we are very close to what many people will say is the limit for R over an ground plane (about 36 ohms), although I would suggest that 36 ohms isn't the limit many people think it is. That, however, would be a topic worth of its own discussion thread.
And finally, looking at X... What the hell? I predicted that X would continue to drop as the additional wires would increase the conductivity of the earth, which would effectively add capacitance to the system, which has the effect of lowering X. There is a reason why I think this way, a specific study on a different type of antenna design. That starts to happen... Then it does the exact opposite... I'm not sure what to make of this, either the added radials have little to no effect on capacitance, or something I'm not seeing is having more of an effect than said additional radials.
I am going to call this part 1. At some point I will add a part 2, and in that section we will look at the question of is more smaller radials better than fewer longer radials? At some point further down the road I would like to do a part 3 where I look at radial layouts that aren't evenly spaced, for example, you have an antenna mounted in the corner of your yard, so you have radials spread over just the part of the antenna that is in your yard.
The DB
