1
$\begingroup$

I have seen sketches of 3-element 2m yagi antennas in which the driven dipole elements are staggered i.e., one element is positioned slightly in front of the other along the length of the (plastic) horizontal beam. these elements also "overlap" slightly, as they protrude through the beam pipe for an inch or so and the coax is connected directly to the elements at the ends of these protrusions.

I've been told that the staggered elements raise the feedpoint impedance to furnish a correct impedance match to RG-58/U coax without requiring a balun transformer. Is this a true claim? And do the elements need to be driven at their overlapping ends? Or does this overlap matter at all? In short...

Where can I find detailed design rules for staggered dipoles like this? I would like to try this design in a 5-element Yagi.

In the image linked below, you can see the "stagger" (one element is in front of the other) and the "overlap" (the elements cross over the horizontal beam midpoint).

enter image description here

$\endgroup$
  • $\begingroup$ How staggered - how far apart are the driven elements? $\endgroup$ – tomnexus Jul 4 at 18:13
  • $\begingroup$ Can you, please, refer us to a sketch of this antenna or add a sketch to your question? Terms like "staggered" and "overlap" can be used by different people to describe different things. $\endgroup$ – Brian K1LI Jul 4 at 18:20
  • $\begingroup$ will try to get a picture. PICTURE ADDED! oboy. $\endgroup$ – niels nielsen Jul 4 at 18:52
  • $\begingroup$ You refer to driven "elements" in your question. I wonder if you are describing a log periodic cell with reflector and director? These are often fed directly without a balun which would fit in with your description. Driven elements in this case, are often mounted as you describe on a single insulated boom or, with alternate dipole halves on parallel metal booms. Have a look at ."Images for log periodic antenna design" on google $\endgroup$ – R Johnson Jul 5 at 8:44
0
$\begingroup$

Thanks for bringing this technique to our attention. I applied the concept to a simple half-wave dipole to gauge the impact.

Spacing the halves of a 1-inch diameter, 40m dipole $.001\lambda$ apart (about 3.3-inches), overlapping them by 36 inches and feeding them with a segment that joined the staggered dipole halves at the midpoint of the antenna, the feedpoint impedance increased from $72\Omega$ to $92\Omega$ with no change in the resonant frequency. Note that each wire comprised a large number of identical-length segments to avoid possible computation errors. Reducing the element diameters to 1/2-inch reduced the impedance to $82\Omega$, again with no change in the resonant frequency. This could be used to raise the feedpoint impedance of the driven element of a parasitic array, which is ordinarily lowered by inducing currents in the parasitic elements.

I don't know of any closed-form expression to calculate this behavior in pursuit of a working design. Since the resonant frequency wasn't affected by the staggered feed, it doesn't appear that the principle of a gamma match is applicable. It might be possible to build a traditional yagi design with the staggered driven element and simply trim the overlap until you reach your impedance target.

A balun will still be required to prevent unbalance currents from traveling down the outside of the coax shield.

$\endgroup$
  • $\begingroup$ @brianK1L1 in NEC watch out for long and short segments meeting near the feedpoint - I think 5:1 is ok. Perhaps this is what you meant with "many segments". Also the right angle at the feedpoint is a little scary. Check the total integrated power in the RP? Or keep the overlapped model but gradually cut the overlap back until you have a zigzag dipole. I'm not sure a comparison to a straight dipole is quite fair. $\endgroup$ – tomnexus Jul 4 at 23:36
  • $\begingroup$ thanks for the simulations. My intent is to perform experiments for various values of stagger and overlap in a yagi with 5 elements. I have a friend here who has an antenna analyzer which I intend to use. output variables will be feedpoint impedance, resonance, and bandwidth. comparison standard will be the same antenna with an ordinary dipole without stagger or overlap. I think an interesting paper could be got out of this, especially if accompanied by simulations of the stagger designs on 2 meters to get directionality data. -Niels $\endgroup$ – niels nielsen Jul 5 at 5:42
  • $\begingroup$ @nielsnielsen That sounds like a great idea for a paper, Niels. When you write your paper, please add a comment to this question on SE so we can follow your work. $\endgroup$ – Brian K1LI Jul 5 at 10:55
1
$\begingroup$

The staggering in the drawing is small, 0.7", and the overlap is not specified. This makes me think it's done for mechanical reasons.

The feedpoint of a dipole or yagi is difficult to support mechanically - both halves need to be insulated but we'll supported. One neat solution to this is to stagger them somehow, so they can both pass through the boom and be supported there.

I've built broomstick-and-wire-and-glue yagis and staggered the driven elements by 10 mm or so, just to provide mechanical strength. Both halves of the driven element can go through the boom. Then I can solder the coax on one side and its strong and neat.

Yagis can be designed for various feedpoint impedances. I can't simulate your one now, but it might well be designed for 50 Ohms. If you're planning to feed your yagi directly with coax, this is what you'd do. 15 to 25 Ohms is more common, and might give slightly better performance, but a yagi designed for 50 will work fine and is nice and simple to feed. Impedance is mostly affected by reflector spacing and length.

There will be no significant impedance transformation from the feed staggering and overlap in your drawing.

With a direct coax connection you still need a balun. This could be a small coil of coax, or a 1/4 wave sleeve on the coax.

$\endgroup$
0
$\begingroup$

It is a design for a 3el yagi. the feed overlaps because the "wires" are metal rods thru the beam which allows for adjustments. there is a 4el design out there using this method

$\endgroup$

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.