Stacked Collinear Design Issues A 1090 MHz

Hi,

I just got interested in ADS-B and wanted to build my own antenna since it would have cost me 10 EUR instead of buying one for 150. Instead it has cost me a lot of agony and time, since I believe I'm becoming a radio ham as a result...

At first I found some very interesting antenna plans. In particular the very simple (to construct) stacked collinear ones. But as I got more into trying to understand that antenna, I got myself into more trouble than anyone could expect, for such a project. Apparently the theory and documentation behind these antennas are really not very solid or at least not very well documented. In the last two weeks I have Googled and carefully read about these things, and I have come to the conclusion that either (a) the people building these things don't know what they're doing, or (b) I have not understood the first thing about antennas, even after carefully gone through dozens of plans, books (ARRL Handbook/Antenna) and started modeling using 2NEC4 etc.

I posted in another forum, but after more than one week, not a single response. I think my questions are too technical for that place. So I am hoping to find someone more RF oriented here, who may be able to help.

Here are a few of the plans I have been looking into.

http://www.diff.net/media/2011_09_24..._4919.jpg.html
http://www.airlomba.net/index.php?pa...rojpg=coli1090
http://www.multimode.fr/articles/rea...aire-1090-mhz/

They are all essentially the same, apart variations in the measurements.
They consists of the following "elements":

Connector => L/2 + S + 3L/4 + S + <3L/4

where:
L = Lambda (full wavelength)
S = 1-2 turn coil as a "1/4 wave phase stub".

For reference, the "Zero order" (ignoring losses and velocity factors) theoretical design has:
One of the particular models (above), that have shown "success" have the following measurements:
Now as you can see these are very different, as expected. But what was not expected is, that after studying Antenna design, these measurements simply doesn't "measure up" or make any sense (to me). First of all the velocity factors used to get those measurements are different for the different pieces, second they must have used some dreaded kind of "Cutting Formula" for the "S" coils length, but even more surprising, the total length is just ~L/4 and not ~L/2 as one would expect for a "1/4 wave phase stub". But when you realize that it does match if you use 3L/4 elements, it may seem ok... But is it?

My 1st question is:
Why do they use 3L/4 (3/4-wavelength) elements for the two top parts and not 1/2 wave elements in phase?

The reason I'm asking, is because I read somewhere (related to a J-pole) that the 1/4 wavelength "leftovers" on the radiator element (in our case the 3/4) would cancel part of the wave in the far field, resulting in worse gain near the horizon, than otherwise. At least that's how I understood it from reading HERE and HERE.

My 2nd question has to do with the 2 phase stub loops (S).
In some designs the 2 loop's handedness are opposing each other. What would be the effect of this, if any?

It would be cool to see some simulation or model of this, but I don't have that kind of SW (e.g. HFSS etc.) to play with, so if there are any specialists here, I would very much like to know more about this. I tried modeling the coils in 2NEC4, but coils are really a PITA when using Geometry Builder, since you cannot specify total length, so gave up after a few attempts. Here is a picture of my attempt in modeling using Zero order measurements, but with approximate coil measurements according to model (1) above, both RH. You can see the terrible fencing and high angle radiation, although quite descent along the horizon which is important for catching airplanes far away...

My 3rd question is:
How can I best understand this antenna and it's properties to make improvements and actually make it useful?


Thanks in advance,
73, E:V:A

I'm trying to edit my post above, since it may look a bit too long, but I don't see any button for that. Is it not possible?

Anyway, the 3 questions are:

1. Why would one choose to use 3/4-wavelength elements for the two top parts and not 1/2 wave elements in phase?
2. How does opposing coil handedness (in the phasing stub) affect the antenna behavior?
3. How can I best understand this antenna and it's properties to make improvements and actually make it useful?

May not give the answer to some of your questions but definitely gives better explanation for collinear antennas:

http://www.para.org.ph/membersarticl...s%20-%2021.pdf

Hi vfone! Thanks for responding. I have read that article, and yes, it's a great introductory text, but like you mentioned, it doesn't answer my questions, even remotely. The closest I have gotten to any sensible answers is the excellent article by (the late) Cebek:
"The Case of the Curly Collinear", [Cebek, 1999]

Given the difficulty of modeling the electro-dynamic interaction between an inductor (coil) and the radiating elements, I still think I have a fairly good idea of what's going on, but I have not seen any mention of the inductor handedness... So at this point I just wanna understand why they stack 3/4 WL elements and not the much more common 1/2 WL or 5/8 WL, ones? Surely this must be known to most radio amateurs!?