Here's what I am working on, and someone please tell me if I'm doing this right or wrong, but be kind and gentle.

Center frequency 146.520MHz

Driven element is 1/8 wave 10.25" long set at 6.125" from the front. Reflector element is 5/8 wave at 15.3" and 15.3" from driven element to the back. Director element is 1/2 wave at 12.25" and set at the front.

Also, trying to figure out how to connect my SO239.

Trying to make it as small as possible for easy bugout if it comes to it, but also out of strong and durable materials.

Am I doing this all wrong?

Still trying to figure this out...


After finding more information, I have found new knowledge...

This project ends up being in a Quad beam format with a reflector, driven and director in a 1/8WL configuration.

Here's the info:

Element Lengths: Driven = 10-1/16” Reflector = 10-9/16” (5% larger) Director = 9-9/16” (5% smaller)

Element separation: Reflector to Driven = 2.5” Driven to Director = 1.5”

It's a 5" design.

I also realize that tuning this will not be easy.

Hopefully, this helps?

EDIT: Typo correction SO239.

EDIT: I have added a photo of the 3 elements - Like I said, it is very small.

IF it works, then it does... If not, then I have gained a good amount of knowledge from the experiment. Isn't that what HAM is all about?

enter image description here

  • 1
    $\begingroup$ welcome to the site! Sounds like you are trying to make a yagi, but not quite sure. Would be great if you could edit your post to include a little more info on what you're looking for the end result to be. $\endgroup$
    – webmarc
    Mar 5, 2021 at 19:07
  • 1
    $\begingroup$ This sounds like something new, so it's difficult to envision exactly what you describe. A picture would be worth a large number of words. $\endgroup$
    – Brian K1LI
    Mar 6, 2021 at 14:18
  • $\begingroup$ Thank you, Brian... It's a 3 element mini-quad 1/8WL (1/32WL per side) 2m beam antenna that will fit into the space of 5" for a bugout kit with easy breakdown. So, if I'm out backpacking, or whatever, it will easily fit into a backpack. Most of the projects I have seen are comparably larger and heavier. Smallest I have seen is ~39" Yagi Design. $\endgroup$
    – user18968
    Mar 6, 2021 at 16:16
  • $\begingroup$ Is your purpose to get better gain than rubber-ducky gain in some direction(s?) or to get a good forward/backward quotient? (The second could be useful if you want a small antenna that can help you locate a transmitter.) Or do you have some other design goal? I think we need to know this to answer your question. $\endgroup$
    – EdvinW
    Mar 6, 2021 at 22:34
  • $\begingroup$ The whole idea, Edvin, is to make a compact 2m 1/8 wave (1/32 WL per side) miniature quad beam that goes against the norms and something so small that it will fit into a backpack without any problems. Check back in a few, as I have just taken a picture of the project and will be editing my stack. It's an experiment $\endgroup$
    – user18968
    Mar 6, 2021 at 23:24

2 Answers 2


Yagi antennas require an exact set of lengths and spacings to work properly.

The lengths are all near a half wave but there is some subtle tuning to get the currents to the appropriate phase.

Spacing can be between a quarter and a half wave but different spacings affect the optimum lengths.

So your example of a 5/8 wave reflector won't work as a reflector, and 12 inch spacings sound a bit too small. The driven element needs to be about 1/2 wave long as well, slightly shorter if it's a folded dipole.

I recommend looking at some standard yagi designs from page 67 onwards in this book - Antennas in practice for some ideas. I've built a few of these and they've worked well.

For example: The smallest one has these parameters:

  • Reflector of 0.482 wavelengths (39.25"), 0.2 wavelength spacing (16.55")
  • Driven element is a folded dipole 0.389 wavelengths long, about 34 Ohms
  • One director, 0.442 wavelengths (36.58"), 0.2 wavelength spacing

This has a gain of 9.1 dBi. You can see the subtle change in length between reflector and director. At this frequency these probably need to be correct to about 1/4" at the most. The document has more detail about correcting for different element diameters and if they're connected to a metallic boom.

  • $\begingroup$ I understand what you are getting at, but I'm attempting to keep this as compact as possible to fit into my bugout kit without sacrificing space. Part of the problem is that other people keep pointing me to other's that have already built something using their own math or someone else's. Total length of this project is barely over 15-1/4". $\endgroup$
    – user18968
    Mar 5, 2021 at 0:25
  • 1
    $\begingroup$ You should model your yagi in EZNEC or 4nec2 @PatriotCountry. Then you'll find out exactly how it performs. If you ask nicely (by editing the question, not in a comment please), someone might do it for you. $\endgroup$
    – rclocher3
    Mar 5, 2021 at 2:33
  • 1
    $\begingroup$ @PatriotCountry Compact and A working Yagi might be two different things here. I think that EZNEC comes with a sample 3-element Yagi that you can easily adjust for 146.52 MHz. Download the free demo $\endgroup$ Mar 5, 2021 at 3:32
  • 1
    $\begingroup$ @PatriotCountry a two-element yagi with 15" spacing is possible, if the element lengths are correct. They will be roughly as above, 39" reflector and 32" driven element. The reflector length is not negotiable, if it's wrong then it's not a yagi. $\endgroup$
    – tomnexus
    Mar 5, 2021 at 14:26
  • 1
    $\begingroup$ @PatriotCountry The loops in a quad beam are ~1WL perimeter (1/4 WL on a side). If you make them half the size, then it won't behave anything like a beam and it will have terrible SWR. If you want smaller, design for 70cm or 33cm instead of 2m. $\endgroup$ Mar 6, 2021 at 12:24

I applaud your attempt to satisfy a need. Unfortunately, your design will not achieve your goals.

Here is a model of your design:

enter image description here

The pattern shows none of the front-to-back directivity you want from a "beam" antenna. In fact, the pattern is not affected at all by the addition of the "reflector" or the "director". The pattern has about 1.4-dB gain over an "isotropic" or point-source radiator and the feedpoint resistance is very low, about 0.2-$\Omega$, with a lot of capacitance reactance, as one would expect from an electrically-short, bent, "doublet" radiator. It would be very difficult to match this impedance to a coax feedline.

The parasitic elements don't influence the radiation pattern because no current flows in them. No current flows because, like the driven element, they also possess very high reactance, which impedes the flow of RF current.

If one closes the driven element to form a loop, its pattern changes:

enter image description here

The maximum gain of this antenna is about -2.5-dB; that is, only about half the incident power is radiated in the preferred direction, which is perpendicular to the axis of the "beam" ... and even less in other directions. This is the same pattern as the driven element only because, again, no current flows in the "reflector" or "director." This is the pattern you would expect from a single loop (the driven element) that is, electrically, very small. This fact is reinforced by the very low feedpoint resistance of less than 0.1-$\Omega$ with a very large series reactance.

Closing the "reflector" and "director" loops has very little affect on the current that flows in the elements; it merely changes the sign of the very high series equivalent reactance. The closed loop elements of a quad must have a circumference on the order of a wavelength at the operating frequency to perform the way you intend. Non-closed elements, like those of a Yagi-Uda type array, must have lengths on the order of a half-wavelength. Reducing the size of a quad loop element by a factor of 8 defeats the element behavior on which the array depends, regardless of whether the elements are parasitic or individually driven.

You could rotate the elements and feed them with currents that would produce a directional pattern:

enter image description here

But the maximum gain in the favored direction is still negative and the feedpoint impedance is still very low, making it very difficult to match to a coax feedline. This arrangement is similar to the arrays of small loops used for directional receiving antennas on the lower frequency bands where wavelengths are much longer. Low band receiving antennas profit from their ability to reject noise from non-favored directions, improving the signal-to-noise ratio of the signals from the desired direction. Negative gain in a receiving antenna can be compensated by the addition of preamplifiers of sufficiently low noise figure.

Like many budding hams, I traveled the same intellectual road. Finally, I was forced to realize that you (or, at least, most of us) can't break the laws of physics. The only way I have seen to accomplish your goal is to use an array of fractal antenna elements which, essentially, fold a wavelength of wire into a much smaller area than the classic quad-type loop occupies.

enter image description here

  • 1
    $\begingroup$ Excellent, Brian! +1. This should have received more upvotes. $\endgroup$ Mar 8, 2021 at 16:32
  • 2
    $\begingroup$ QSL, Mike. I hope the OP will use this exercise as a good example of the importance of antenna modelling. $\endgroup$
    – Brian K1LI
    Mar 8, 2021 at 19:49

You must log in to answer this question.