Ermi,
In response to "I'm glad that we agree about the reduction of the radiated field of a linear antenna next to a conducting wall by a reverse current induced in the wall." I'd like to add that I don't think we do. My allowing for the possibility of the wall affecting the radiation is based on the effects of capacitance and perhaps on changes in the antenna radiation resistance and not on reflections or induced currents directly.
Here's why. I don't believe either the image antenna or director/reflector models work well here because both are founded on the assumption of passive structures which are much larger in terms of wavelength than we are assuming. They may guide us in our thinking but we have to be careful in drawing firm conclusions based on them.
For example, suppose a current is induced in the wall and is flowing vertically. Would I not see this current if I stood on the other side of the wall? And, if so, wouldn't it produce a radiated field coming toward me? One interpretation can be that the wall thereby appears transparent to the incident wave and will not stop it. Unfortunately this is only my speculation but it does provide another question which could be considered.
The energy incident to the wall, absent dissipation in the wall, has to go somewhere, either be reflected or transmitted. This is the basis of the image model and the principle upon which directors and reflectors are analyzed, but again, this applies for elements larger than we are discussing.
I am not saying that your conclusions are wrong nor that my speculation is right. I am just allowing that they may be. A good model or math analysis might help us decide.
Neil
It seems that from a practical application perspective the following might be considered:
A compliant elevated dipole has the potential to be more useful than a ground surface mounted compliant base loaded vertical if a large structure is close to the base loaded vertical.
An elevated dipole may be operated without an RF ground Earth connection as one leg of the dipole is connected to RF ground. Lightning protection would still be required.
The leg of the dipole connected to RF out is loaded with adequate inductance for the shorter element, however the additional loss resistance due to the larger loading coil is compensated for by the absence of ground loss.
Larger diameter dipole elements would further reduce loss resistance due to the loading coil by because slightly less inductance is needed for large elements. The chart below indicates what this might mean:
The greatest benefit seems to fall between 1.5 and 3 inches in diameter. Going up to 6 inches diameter only shaves off another 80 uH.
Might be an interesting experiment...
My submission does not address any of the specifics so far discussed. Instead I herein express thoughts aroused by this discussion. The idea of an "image antenna" reminds me of a mirror, which is able to reflect an "image" that is exactly like the source, except that the image is 180-degrees flipped (out of phase) on the horizontal plane. As far as we know one's mirror image does not "cancel" or "reduce" the source, but light reflection is a different form of energy than RF radiation. But look at acoustics, where surfaces indeed produce image effects that do alter the quality of the source in complex and compound ways, varying with the density of the materials and the distances involved. This is everybit like the proposition put forward regarding antenna behavior near certain types of surface.
These are only thoughts and should not be taken for actual science.
Ermi said in the first post in this thread:
"This means that the elevated vertical dipole requires a loading coil with twice the inductance that is needed for a monopole above ground. So, the dipole has more loss resistance due to the loading coil than the monopole above ground; but the dipole also has no ground resistance (since there is no connection to ground). The increased loading coil loss just about balances out the absence of ground loss; and so the elevated dipole works about as well as a monopole above ground without any nearby structures."
The question is how do you connect a vertical dipole that has "no connection to ground"? One way is to mount the transmitter at the center of the vertical dipole and power the transmitter and co-located audio source from internal batteries. In real life you need to power the transmitter and feed the audio source remotely, and you want to provide a heavy wire ground for lightning protection. So, a path to ground is a fundamental requirement.
With a ground path, the current in the grounded side of the dipole is minimal. Most of the current flows in the upper element of the dipole and in the ground path. This is similar to a 1.5 meter antenna elevated at the same distance. But, the loading coil needs more reactance than for a 3 meter vertical antenna at the same height, so coil losses will be higher.
There is a very important reason why a 1/4 wavelength vertical, or in our case, a shortened vertical antenna, is the best performing antenna. Hams have struggled with this for decades when designing antennas for the 160 meter ham band (about 1.8mHz). They haven't come up with a better solution other than the marginally better center-loaded, top-loaded and top-hat designs.
I know everyone would love to see an antenna design that doesn't require a ground, but so far, there isn't one.
Phil B
Gents,
For years, the broadcast industry has used Panel Reflector Antennas for VHF and above. If you place an antenna/radiator just short of a half wave apart you can actually produce gain in the antenna away from the panel toward the radiator. The same antenna can also reduce radiation in a particular direction. What you have here is a very basic directional array. By reducing the effect of the reflecting or absorbing surface you can directional-ize the signal. The previous examples in this post speak more to the absorptive properties of materials rather than reflectivity. Because of the very long wavelength of medium wave signals of Am BC, the close proximity of materials is more absorptive that reflective. Just something to play with in your spare time. I have experimented with directional panel reflectors in ham radio and broadcasting for years. Good luck.
About a month ago, I received a series of e-mails from a guest on this website which deals with the issue of attenuation of the signal from a Part 15 AM antenna by a vertical conductive plate that is small compared to wavelength. I had discussed that issue extensively in this thread with radio8z. Said guest also communicated with radio8z and at least one other regular user of this website about that subject.
The guest said that his NEC simulation showed that a grounded conductive plate 40 feet long, 10 feet high, and 1.5 feet above ground (to simulate the siding on a mobile home), has very little effect on the radiation pattern and gain of a 3 m monopole above ground if the monopole is just a few feet from said plate. At 1650 kHz, the attenuation was calculated to be about 8 dB at a distance of one foot, 2.6 dB at 6 feet, and there is hardly any attenuation at all at 20 feet. This is a very remarkable result, suggesting that a subwavelength obstruction can be very close to a Part 15 AM antenna (in the near field), and not cause significant interference.
I met with Prof. Valentin Trainotti and discussed this NEC simulation with him. He considered the results to be surprising, and he said that NEC is not a good enough program to be relied upon to give the correct results. NEC can only analyze wire antennas. The 40 X 10 foot plate was simulated by a wire grid with 12 sections in the horizontal direction and 3 sectioms in the vertical direction. Trainotti said that the results have to be verified by a program, such as wipl-d, that can analyze conductive plates, and not just wires.
Later, Trainotti analyzed the problem with wipl-d, but he thought that he would get better results using another electromagnetic analysis program that can analyze plates: FEKO. Here are Prof. Trainotti's FEKO results at 1 MHz for a 3 m monopole over ground separated from a vertical 12 x 3 m grounded plate by a distance d:
If d = 0.3 m, attenuation due to the presence of the plate is 9.4 dB.
If d = 1 m, attenuation is 4.98 dB.
If d = 3 m, attenuation is 1.65 dB.
If d= 9 m, attenuation is .21 dB.
In the FEKO model of the system, no loss resistance exists. Attenuation is entirely due to the reduction in radiation resistance as the monopole gets closer to the plate. At 1650 kHz, which was used for the NEC simulation, the attenuation is somewhat greater than at 1 MHz, which was used for the FEKO simulation.
We see that the preliminary results of the NEC simulation approximately correctly conclude that keeping a Part 15 AM antenna several feet from a conductive surface much smaller than a wavelength will not result in much interference from said conductive surface. The departure of the NEC simulation from the correct results is that NEC substantially, but not overly greatly, underestimated the amount of attenuation caused by the plate.
If I mount my base coil loaded antenna on my boat, will it get reflective interference from the aluminum mast of a sailboat on the other side of the dock finger? The antenna mast is about 6" dia. ... actually, it's not round, but something between a rectangle and an ellipse cross-section), and about 15 feet away from where the antenna mast would likely be.
There is a grounding question for a marine mounting system too, which I asked on another thread about ground stuff, but I never got an answer ... I'll try again ... if still no answer, maybe I'll try a new topic.
TIA ...
In the early 1970s, the FCC no longer permitted small boats to use the marine band just above 2 MHz, and required them to use the present VHF marine band. Before the changeover, my boat had a transceiver with the calling and distress frequency 2182 kHz. My antenna was a 12 foot continuously-loaded vertical fiberglass whip, which I was able to place in the horizontal position when the boat was at the dock. The ground connection was to a small porous aluminum block at the bottom of the hull. The theory of the porous block was that it greatly increased the surface area of the ground block to the water. The block worked, but I doubt the theory behind it. The pores probably got clogged pretty quickly. A brass sheet about a foot squared should work just fine, I'm talking about salt water. You'd practically have to have a metal hull in a lake or river.
From the calculations inspired by the initial efforts of the guest, and ultimately done by Prof. Trainotti, the Part 15 AM antenna will not be much affected by nearby conductive objects in the near field, provided that the distance to the objects is not extremely close. The mast 15 feet away should not cause a problem.
I have had poor results using the cold water inlet to my house as a ground, and runninga 3 m antenna just a couple of feet from the wall, I don't have aluminum siding, but my concrete block walls are conductive enough to cause problems. I would expect that 15 feet should be enough of a separation to give good results.