antenna-theory.com

[wolberine]

Hello,

I'd like to ask about antennas near metal surfaces. If I have a device in free space that contains a pcb monopole, I can measure its gain in an anechoic chamber to be highest when the DUT and receive antenna have the same polarization (let's say the maximum gain is -5dBd). However, if the monpole is placed on a metal surface (parallel to the metal surface, not perpendicular), the gain seriously degrades (-20dBd).

I was wondering why this might happen.

I can see two potential reasons:

1) Image theory: The monpole parallel to the metal surface will produce an image with opposite currents. The field produced by the image currents will cancel the field produced by the actual currents on the antenna and reduce the amplitude in the far-field. As the antenna moves away from the metal surface, this effect becomes negligible.

2) Antenna detuning: The antenna will actively couple to the metal surface, producing a change in the input impedance. Maximum power is no longer delivered to the load and the far-field received signal strength is reduced.

Do you think these theories make sense? Does one theory dominate the situation? Are there other/different explanations?

Thanks!

[bigSteve]

Both of your theories are correct.

Image theory absolutely applies in this case, the image "dipole antenna" will have the opposite direction, and will cancel in the far-field because they are close together (close being defined as less than 0.2 wavelengths, roughly). Hence, you can't take a nice antenna, and place it on a metal surface and expect it to radiate.

The second theory is actually a consequence of the first. It is not that the antenna is detuning (which implies a shift to a different resonant frequency). Rather, the metal near the antenna will introduce a significant capacitance (reactance component) to the antenna impedance, which will cause the energy to be reflected at the antenna terminals because it won't be matched to a real impedance (50 Ohms).

Other explanations....
Similar to the image theory, you can assume the dipole radiates a wave away from the metal surface, and another wave toward the metal surface. The second wave is reflected (sign change) and since the phase between the two waves are similar (because the surface is close to the antenna), the fields will almost cancel.

[wolberine]

Are there good design practices for small form factor antennas near metal surfaces?

For designs which require being near metal objects and surfaces I can think of a few design options:

1) Don't horizontally polarize an antenna near a metal surface

2) Use a different type of antenna (are there types that are robust near metal surfaces or is this more dependent on frequency/electrical distance?)

3) Increase the "distance" from a metal surface through physically moving the object or through placing a high dielectric or magnetic material between the DUT and the metal surface

[bigSteve]

To your ideas:

1) Use a vertically polarized antenna, relative to the metal - yes this will significantly help

2) Using a different type of antenna: All antennas have different radiation patterns, so regardless of polarization, you could use one with a higher directivity, such that the peak gain is away from the surface (and the minimum radiation in the direction of the metal). An example of a small form factor antenna of this type is a patch antenna, with the ground plane nearest the metal. The metal shouldn't negatively affect the patch antenna.

3) Increase the distance is good - at about a quarter wavelength you will get some reflected energy reinforcement (i.e. won't cancel). If you can't increase the distance, you could try using a higher frequency (which will be electrically farther even for the same physical separation). High dielectrics will also work, but may detune your antenna and will have some loss associated with them.