Since adding a reflector and a director horizontally to a dipole (as in a three element Yagi) causes the beam to reflect horizontally, what should happen if reflectors above and below the dipole were added?

A dipole less than a half a wavelength above the ground radiates up. But, if one were to add reflectors above and below the dipole (or even a standard three element Yagi) would not that cause the beam to be horizontally directed?

I am imagining this for an HF antenna, but I presume the theory would apply to any type of Yagi or dipole antenna.

  • $\begingroup$ I think what you're describing is a corner reflector antenna. You don't say what band you had it mind, I'd guess HF. The most common corner reflectors are UHF TV antennas, but the same principle applies to HF. $\endgroup$
    – Duston
    Oct 2, 2018 at 13:26
  • $\begingroup$ By "adding reflectors", do you mean adding another wire element, as the "reflector" in a Yagi? Or do you mean adding a conductive plane as a reflector? $\endgroup$ Oct 2, 2018 at 13:46
  • $\begingroup$ Yes, Dustin, that is what I had in mind. Thank you for your response. Why aren't these used to improve HF antenna directionality and to reduce cloud heating? $\endgroup$
    – MarqTwine
    Oct 2, 2018 at 13:47
  • $\begingroup$ Phil, I had imagined another element above and below and below the active element in a Yagi or a dipole, but using a conductive plane would probably work better. $\endgroup$
    – MarqTwine
    Oct 2, 2018 at 13:51
  • $\begingroup$ I think they're not used on HF for several reasons, the biggest of which (pun intended) is size. Also, I don't think the improvement in gain is worth the effort, and as others have mentioned, you may actually want a high radiation angle for certain paths. $\endgroup$
    – Duston
    Oct 3, 2018 at 16:25

2 Answers 2


In a Yagi, the beam direction is the result of constructive interference between all three elements. Typically the elements are spaced about a 1/4 wavelength. The reflector is ahead about 90 degrees (a 1/4 oscillation) in phase, and the director behind about 90 degrees. The phase of each element is a consequence of its spacing and length.

When viewed from the front of the beam where gain is greatest, the reflector is a 1/4 wavelength farther away, and the director a 1/4 wavelength closer. This counteracts the phase shift just described, and the result is the wavefronts from each element are arriving in phase from this perspective. The constructive interference is what gives the antenna high gain in this direction.

Behind the beam, this is reversed: the relative distances of the elements adds another 90 degree phase shift on top of the 90 degree shift they already had, meaning each individual wavefront arrives 180 degrees out of phase. These opposite waves cancel, meaning the antenna has very low gain (ideally zero gain) in this direction.

enter image description here
By Chetvorno [CC0], from Wikimedia Commons

The reason for the radiation pattern of a dipole over ground is similar: it's the consequence of the wavefronts from two radiating elements arriving in phase or out of phase. Only in this case the 2nd element is the image antenna that exists as a consequence of the ground. As a simplification, whenever an antenna exists on one side of a conductive plane (like the ground), we can imagine that plane as a mirror, and pretend there's an identical antenna on the other side of the mirror. The only difference is the phase of this image antenna is opposite.

So consider a dipole a quarter wavelength above ground. This makes the image antenna a half wavelength away from the dipole, and 180 degrees out of phase.

At low elevation angles, the distance to the real dipole or the image antenna is equal, so there's no phase change. So the wavefronts arrive 180 degrees out of phase, and cancel. So no radiation here.

Directly overhead, the image antenna is a 1/2 wavelength away, adding 180 degrees of phase to it. Now the wavefronts from the real dipole and the image antenna arrive in phase, so there is high gain in this direction.

Raise the dipole to a half wavelength, and now the relative phase and distances of the image antenna result in a null directly overhead, and a maximum at 30 degrees above the horizon.

Now to your question: could adding more elements to a dipole that is less than a quarter wavelength high improve gain at low elevation angles?

Well, depending on just where they were added, and their relative phase to the main element, sure. However, since you're already starting from a situation where some of the elements aren't in the right phase, you'll never get better than a compromise solution. And at least some of these elements will need to be at least a half-wavelength high. If the support structure for holding an element that high exists, why not just put a simple dipole up there and skip the reflector business?

Furthermore, proximity to the ground has another detrimental effect: restive losses. The ground isn't a perfect conductor, so whenever there's current in the ground, transmitter power is being lost. This is why we put down radials for example. Getting the antenna higher reduces the current density in the ground, and thus the loss. To have some kind of antenna close to the ground which is still efficient will require improving the ground conductivity somehow.


Following the pattern of a traditional parasitic reflector element as used in a Yagi-Uda array, there is more benefit to adding elements in the horizontal plane of the driven element than there is in adding them in the vertical plane. This is even more true if, as you suggest, one contemplates adding two parasitic elements.

Your observation that, "A dipole less than a half a wavelength above the ground radiates up," is key. Were one to add two "reflectors" at a typical spacing of ~0.1-wavelength above and below the driven element, all of the elements would still be less than a half-wavelength above ground, so the opportunity to lower the radiation angle is minimal.

  • 1
    $\begingroup$ I mostly agree, Brian. Orient the dipole and reflector (and any director) all in a horizontal plane for DX. However, increasing the gain pointing straight up can be very useful for local (NVIS, within a few hundred miles) QSOs. I know a ham who placed a reflector under a 75m dipole during Field Day and immediately started breaking pileups. I was there, and can testify that it worked because I was operating another station on the same band. :-) $\endgroup$ Oct 2, 2018 at 23:31

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .