Looks like a Barbie (tm) TV aerial, but it seems to work.
It is based on calculations from the Java applet on http://fermi.la.asu.edu/w9cf/yagipub/index.html . I started from one of the examples (70 cm, 12 elem) and scaled it to 2400 MHz as explained. I used 1.5 mm² domestic wire. I drilled small holes in the wood to put the wires in and I glued them in. One can see pieces of matches glued to fix the 'N' connector. I just used 8 elements for a first test because my wooden stick wasn't any longer...
I am not sure whether it is a good idea to use wood, wires, folded dipole, etc... (the antenna should be adapted to the coax ?300 Ohms - 50 Ohms, ?balun) But the idea is worth a try. I just tested it in my apartment and it seems to be as good as a tin can or a BiQuad. ( http://www.saunalahti.fi/~elepal/antenna2.html, http://trevormarshall.com/biquad.htm, http://www.saunalahti.fi/elepal/antenna4.html, http://martybugs.net/wireless/biquad/).
I made two longer ones (66.5 cm, 20 elements). One with domestic electrical tubing (gray) that does not work and another one with a wooden stick (7x7 mm²) that gives me three dB's more (I expected five). Maybe, it is because the wood is thicker? I don't know yet what is important and what is not. What would make a good boom? Is it reproductible?... At first sight, it is easy to build, cheap and gives many dB's.
Branislav M. asked me how I did wire the dipole on the 'N-Connector'. Well, it is probably not the way to do... (I should use a Ballun, do it more carefuly,...). But it is the way I did it. And it seems to work (at least with a 2.5 meters pigtail). I have no idea how better it would be if it was 'well' done.
The 20 elements has been tested on 9.5 km (with a CardboardHorn on the other side) (Liege20020615) :
Here is the comparison with other antennae :
Beware of polarization!. It matches a parallel Marconi (Lambda/4 piece of copper)(all antenna wires in the same plane). Horizontal ones would go well with vertical Slotted Wave Guides (http://trevormarshall.com/waveguides.htm)
The original page is on http://ReseauCitoyen.be/?UdaYagi (sorry, it is in french, but there are other DIY antennas on http://ReseauCitoyen.be/?HomeMade and some experiments on http://ReseauCitoyen.be/?LongShots ; note that we use hierarchical lines to ease the navigation within our wiki (something like >Hardware>Antennas>Yagi) ).
Elem length distance (mm) 1 61 0 (reflector) 2 60 19 <- driven elem, it is a 'folded dipole'... 3 56 26 4 55 40 5 53 60 6 53 84 7 52 112 8 52 144 (9 51 178) (my stick is shorter, (10 51 215) for a first test, it is good enough) (11 51 255) (12 50 296)
NB: I don't know where to link it on this wiki. If somebody can help. Thanks! (photos are still in Belgium; a local copy would be welcome)
- -- xof, Liege, Belgium
See also :
- http://reseaucitoyen.be/?UdaYagi2 (in French with NEC file)
- http://www.xaviervl.com/Antenne/Frisko/index.html (in french) : 9 dBi with 4 paper clips and an ice cream stick :-). Any idea to make something simpler?
This one is based on the Frisko design :
Elem length distance (mm) 1 58 0 (reflector) 2 53.5 20 <- driven elem, it is a 'folded dipole'... 3 52 35 4 51.5 50
The wire is 0.8 mm brass (model-making)
During the Belgian Microwave Roundtable 2002, I got the opportunity to 'measure' it (with a Network Analyser, thanks ON4AOD). The S11/SWR plot is on http://users.skynet.be/chricat/antennes/UdaYagi-swr-300h.png. (But don't ask me what it exactly means for the moment. It seems it is not so bad somewhere in the ISM band...;-)
Another crazy idea is to use Tetra Brik (Milk or Juice carton) as a wave guide antenna. It contains enough aluminium to reflect waves. See http://ReseauCitoyen.be/?BoiteDeLait . It is also easy to add a horn and get more dB's.
Lg = 1 / SQRT(1/(L0*L0) - 1/(Lc*Lc)) and Lc = 2 * width
width is here 9.5 cm -> Lg is ~16.6 cm at L0=12.5 cm (2400 MHz); The rod is thus at ~4 cm from the bottom of the brick. The rod is Lambda/4, about 3 cm. It is 'fragile', but so easy to build... (and you can already go far away, probably more than 2 km with two of them).
For further adventures with horns, see CardboardHorn
07sep2002, another Tetra Brik horn was used with a SlottedWaveguide (this one) on a ~15 kilometers link in Belgium (Louvain-La-Neuve -- Waterloo). The bigger CardboardHorn (see the picture) allowed 300 KB/sec.
For the skeptics, I just ran some tests. I compared my antennae about 30 meters away from a (vertical) Lambda/4 inside my apartement (it goes through a window and there are a few tree leaves in the path, so don't extrapolate the distance).
Ups and downs show the polarization effect (vertical/horizontal). This (short) Uda-Yagi is the best :-). Of course, a longer one will still be better.
Measurement are done with the iwspy of the wireless tools on Linux (http://www.hpl.hp.com/personal/Jean_Tourrilhes/Linux/Tools.html and plotted with http://www.gnuplot.info/). I also use a front-end (http://ReseauCitoyen.be/scripts/dynspy-0.07.txt (a perl script, rename -> .pl) found on http://ReseauCitoyen.be/?TestTools to log the measurements and see the dynamics.