In every reference book I've seen, an electric dipole antenna has basically a donut pattern, and as things are added to increase the gain (directors, reflectors etc) the pattern changes. I've always understood this as saying that the only way to get more gain out of an electric dipole is to change the antenna pattern, reducing the beamwidth in one or the other or both axis.

Correspondingly, it seems like this is saying that increasing the gain may keep the volume enclosed by the antenna pattern (in 3 dimensions) constant, and the gain modifiers change the shape of that space.

Obviously I'm speaking ONLY of the gain of the antenna itself, no amplifiers etc, and that the initial dipole is already optimal, not built out of wet string or other lossy material. :)

Am I understanding this correctly? Is this more of a physics question than an EE question?

  • 1
    \$\begingroup\$ This is tantamount to asking if the law of conservation of energy can be broken. "higher gain" antennas don't increase the available power, they just steal it from other directions. \$\endgroup\$ – user_1818839 Dec 7 '20 at 16:40
  • \$\begingroup\$ Possibly so. I'd just never seen it called out explicitly in the ARRL antenna books, or anywhere else. \$\endgroup\$ – user103218 Dec 7 '20 at 16:41
  • \$\begingroup\$ Ok, what if the initial dipole were significantly lossy. Obviously the pattern would be the same, just shorter radius, and the directors/reflectors etc would have the same effect as they would on the efficient dipole. Correct? \$\endgroup\$ – user103218 Dec 7 '20 at 16:44
  • \$\begingroup\$ That's explicitly a different question. You can go from lossy to less lossy, but the original question excluded that. \$\endgroup\$ – user_1818839 Dec 7 '20 at 16:47
  • \$\begingroup\$ Understood, just exploring the space. \$\endgroup\$ – user103218 Dec 7 '20 at 16:54

It depends what you mean by "gain" of an antenna.

Some people use this term as a synonym for the directivity of the antenna. You can't change the directivity of any antenna (dipole or any other kind) without changing its radiation pattern.

That's because the directivity is a description of its radiation pattern, that tells you how much power it sends in its "brightest" direction compared to the total power it emits. Improving the directivity gain of an antenna means changing the radiation pattern to emit more in the desired direction and less in other directions.

Other people use the term gain for a combination of the directivity and the efficiency of the antenna. You can improve the efficiency of the antenna by making out of higher-conductivity materials or larger-diameter wires, not surrounding it with lossy dielectrics, etc. But your stipulation that the antenna is "not built out of ... lossy material" seems to indicate you already considered this and can't improve in this area.

My understanding is that this definition of antenna gain does not include effects due to impedance matching between the feedline and antenna (source), so improving your matching network is not a way to improve the antenna gain.

Summary: You can not improve the gain of the antenna without either changing its radiation pattern or improving its construction to make it less lossy.

  • \$\begingroup\$ Ok, thanks, I just had never seen this called out explicitly. \$\endgroup\$ – user103218 Dec 7 '20 at 16:56

It's pretty clear that for a transmitting antenna, you only have so much power, and so the more you send in one direction, the less you send in another.

But what about a receiving antenna. Shouldn't there be a way to make it "bigger" and so intercept more energy? Intuitively there should be, but the engineering always says "no". And it's interesting to take a moment to understand why.

Research quickly lead to https://ham.stackexchange.com/questions/1836/how-to-increase-antenna-aperture which is an interesting read.

My explanation of the issue would be as follows:

The concept at play is the antenna's aperture, or basically the effective perpendicular cross section of signal in intercepts. Can't we make that bigger?

Well, no.

We can't really make the dipole bigger without changing its resonant frequency, and if operate it in an overtone mode (not uncommon, eg, using a 7 MHz antenna as 21 MHz) it's already going to have different behavior.

But what if we get a big field and build a bunch of dipoles and add their energy together? We should get the sum of the energy all receive, right?

Well, sorta, but there's a problem: if we add signals from distinct points in space where they have different path lengths to the transmitter and thus different phases, in some combinations they will add constructively, and in others they will interfere destructively. So to make that work, we'd have to adjust the phase delay of the feed from each antenna so that they add constructively. Which means we've just re-invented the phased array antenna. And since a given set of delay adjustments only work for a signal coming from one direction, phased array antennas are directional... it's pretty much why they get built.

TL;DR no, you can only do this by making the antenna system directional. You then have the choice of mechanical rotation or electronic rotation.

And anyway, in weak signal work, directionality is considered a good thing, because (provided you can readily aim it) it lets you get more of the signal you want, and less of the extraneous energy you don't.

Let's also not forget that directionality is a question in three dimensions. A typical antenna for ground-wave modes intended to be geographically omnidirectional would still typically want to be spatially directional in the sense that it is optimized horizontally towards the horizon, rather than up into the sky. Unless spacecraft communication or skywave propagation is the goal, "up" is more a direction of waste or one from which undesired noise arrives - even aircraft are less "up" than "over" by the time range becomes a challenge.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.