It always confused me why 100 watts out of your transmitter could radiate 200 watts coming from a high-gain antenna.

Why is the term gain used instead of directionality?

  • $\begingroup$ Did you mean directivity? $\endgroup$ – Phil Frost - W8II Mar 24 '14 at 12:06
  • $\begingroup$ 200W EIRP from a 100W transmitter isn't "high gain". That's just 3dB of gain -- a dipole in free space has about that much gain over an isotropic radiator, and a dipole one half-wavelength above Earth has about 8dBi of gain. $\endgroup$ – Phil Frost - W8II Mar 28 '14 at 13:21

The directional antenna doesn't actually radiate 200 Watts, it just focuses the actual mere 100 Watts so that along some optimal directional axis, a receiver would see same power as if an imaginary 200 watt transmitter using an imaginary isotropic antenna were instead the source.

Which makes it easier to categorize the radiated power from that "high gain" antenna in that optimal direction using just one number, instead of 2 or more.

An additional 100 Watts isn't created, more like taken away from other less optimal directions in the antenna's radiation pattern.


I'm going to assume you mean directivity, not directionality. Then:

Why is the term gain used instead of [directivity]?

These are two related but different things.

Gain measures the ratio of input electrical power to output electromagnetic radiation (or the reverse, for receiving). It is a spherical function, that is: an antenna has gain defined for any direction, azimuth and elevation. When we discussing just "gain" as a single number and not a function, it's usually understood that we mean maximum gain: the gain of the antenna in the direction of peak sensitivity.

From the law of conservation of energy, a passive antenna (one without any amplifier or other source of energy) can't have an average gain greater than 1. The only way it can radiate more energy in one direction is to radiate less energy in another direction. The total power output remains the same.

Directivity is a measure of the extent to which an antenna does this. If the gains are normalized such that the peak gain is 1, then directivity is 1 divided by the average gain over all directions. An isotropic antenna, one that radiates equally in all directions1, has a directivity of 1.

It's possible to have a high directivity, low gain antenna. Simply build a very lossy, very directional antenna. This antenna might have a peak gain of -10dBi. The average gain might be -60dBi. That is, it receives and transmits much better in one direction, and thus has a high directivity, but even in that direction it's very poor. In practice we might actually call this device a dummy load or a heater.

To build a high gain, low directivity antenna you must have an amplifier. An amplifier with a low directivity antenna could be considered a high gain, low directivity system.

Assuming that we can point our antennas appropriately, then peak gain is a useful number for the calculation of a link budget. We might then care about directivity if we consider all energy received from unintended directions as "noise": then maximizing directivity will minimize noise. Conveniently, if we can make a passive antenna more directional without also making it less efficient, then we maximize both peak gain and directivity.

1: A coherent isotropic radiator does not exist. Even a dipole radiates more in some directions than others.

  • $\begingroup$ "An amplifier with a low directivity antenna could be considered a high gain, low directivity system." This sentence made me think of the final RF power amplifier stage of any transmitter. Without any amplification, you'd be stuck with, at most, the sum of the power output of the oscillator stage and the modulation input. Not a whole lot of power there, though I suppose it could be of interest as an experiment for QRPP operators... $\endgroup$ – a CVn Mar 24 '14 at 15:09

I'll leave the more technical explanations up to someone else, but here's hopefully a good-enough explanation:

My assumption is that you know, but just in case, let's start with what "gain" is. Gain in the example you've presented represents a focusing of power (e.g. 3dBi, or it doubled). There are lots of analogies that I could paint, but let's go with an explosion, since it will try to radiate its power in all directions similar to a basic "perfect" antenna. If you put a cannon ball next to a certain amount of gun powder, both suspended in air, and ignite the powder, it will push the cannon ball some. Put them both on the ground, and the same amount of gunpowder will push the cannon ball even more because the ground is directing the force to go only out and up instead of in all directions. Put them both in a cannon, and the ball will go significantly farther than either because all of the force is being allowed to go in only one direction. Remember, you have the same amount of gun powder in each case, but the effective force on the ball is different because of how the explosion was focused.

Gain, then, is a relative description of how much radiation is being directed or focused, and therefore how much more effective power it has. To get the relative value (dBi), we usually compare antennas to a perfectly omnidirectional one (similar to the gun powder suspended in air). So an antenna with 3dBi will have double the power in the most powerful direction.

Sometimes, for some people, knowing how much it's focused is enough by itself, sometimes the direction the power is focused in is implied by the antenna type, but sometimes it could be purposefully not given to make the antenna sound better than it is--or to make it sound better to a larger group of people.

So, knowing both the gain and direction is probably the most useful when choosing the right antenna for your needs, but knowing just one or the other is only giving part of the story.

  • 3
    $\begingroup$ "3dBi" doesn't mean "doubled". That would be just "3dB". 3dBi is an absolute unit of gain and it means 3dB of gain over a theoretical isotropic radiator. Likewise, all units of dBi aren't relative. We don't usually compare these to an "omnidirectional" antenna (the word you should use is isotropic, which is different), rather the i in dBi explicitly says an isotropic radiator is the reference. $\endgroup$ – Phil Frost - W8II Mar 24 '14 at 12:03
  • $\begingroup$ Another common reference is dBd, where the d shows the reference of comparison to be the peak, free-space gain of a center-fed, 1/2-wave dipole. That dipole has a peak, free-space gain of 2.15dBi. $\endgroup$ – Richard Fry Feb 27 at 13:13

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.