2.4GHz antenna design

[NiHaoMike]

I'm designing a high gain (~18dBi or so) 2.4GHz antenna for use with 802.11 hardware. I decided on a Yagi design as I wanted something that takes up little space in a bag. Instead of wire elements extending out of a stick like a TV antenna, I'm thinking of taking a wooden (or plastic?) board, cutting grooves in it, and pressing copper wire into the grooves. Are the calculations different if I were to embed the elements in wood? Dry wood is an insulator but its dielectric constant is much different from that of air.

As of current, my 25 element design is going to be on a board that is 3' long and 3" wide. I'm thinking of cutting it in half and adding a hinge so it folds to a more reasonable 1.5' length for carrying. Am I correct that I'll need a hinge or other mechanism made of a nonconductive material to avoid affecting antenna performance?

I'm going to use homemade Litz wire to increase the efficiency, so they'll be even more fragile than solid wire.

[kaptain_zero]

Argh... My antenna theory is rusty at best, but do you really need 18db gain? The reason I ask is that a collapsible corner reflector antenna would meet the compact, folding and easy to make feature, but only about 12db gain to the front and 20db rejection to the rear.

The most efficient design when you get up in the 2.4ghz and up range is a small parabolic dish..... Easy to get lots of gain with not much material and fairly compact.

Your wooden/plastic idea could be made to work, but at those frequencies... relative humidity would become an issue and might create tuning headaches.

I'd let my fingers do the walking in google, there's plenty of Ham Radio references on antenna design in the 2.4ghz range and up.... Phase 3D (a large elliptical orbit satellite built by hams) had a 2.4ghz down link and decent gain was important to hear that satellite so there was lots of home brew antennas cobbled together for that. I dropped out of the amateur satellite arena about 10 years ago.... so I'm not up on the latest. Amsat.org might have some useful information as well.

Regards

Christian

[Psi]

Most people tend to avoid RF design at frequencies that high, you can do the math and build the antenna/circuit but without very expensive test equipment it's hard to tune and fix issues. It requires a ridiculous amount of trial and error. You might be better off finding a forum that deals entirely in RF design to get a good answer to that question.


There's probably only a small handful of people on here with excellent RF knowledge at frequencies that high.

[Zero999]

The easiest way to design a 2.4GHz antenna is to etch it on a PCB.

[kaptain_zero]

Did some slight poking around on the net, found this site...... http://www.wlan.org.uk/diy-ant.html

Plenty of ideas and information on this topic....

Regards

Christian

[sonicj]

i might suggest checking out the Moxon.... high gain, easy to construct, 50 ohms, tiny!

[NiHaoMike]

i might suggest checking out the Moxon.... high gain, easy to construct, 50 ohms, tiny!

That's not enough gain for my current application, but I'll keep it in mind for some other projects.

[sonicj]

like psi mentioned above, 2.4ghz is very difficult to get right without proper tuning gear! a 18db yagi at 2.4 w/o tuning gear... well... i personally wouldn't try it and I have 2.4 measurement equipment. 

i like hero's suggestion of printing & etching the antenna! i know the moxon doesn't give you the best numbers on paper, but it's design allows for much tighter tolerances for diy, no balun required, sma/n/whatever to be soldered directly to the board.

heres a 2.4 yagi for $12usd if you have enough time for shipping from china. link tough to build one cheaper than that!

patch antennas are portable and pretty easy to fabricate.

16db printable patch antenna link

the "cantenna" is another easy build with potential for high gain.

my 2¢ fwiw,
-sj

[Mechatrommer]

circuit antenna...

from...
http://www.lelong.com.my/broadband-external-antenna-cable-connector-76380369-2011-03-Sale-P.htm
my reference for yagi...
http://www.skyscan.ca/Antennas.htm
GeoffS from https://www.eevblog.com/forum/index.php?topic=2703.msg36602#msg36602 suggesting for cabling...
http://www.tigereye.net.au/vk4ion/nextg_yagi.html

...i just thought i remember this thread sorry for ugly post.

[Kiriakos-GR]

All those replicated ideas from the internet does not help any one.
Its an true waste of time.

Nothing compares next to an true 3Com  Wifi bridge , with connector for external antenna , plus an high gain external antenna.

In a cost of 250$ .

Now there is some restrictions , about the total gain, but if you live in the countryside , you can bend them with out damaging any one.
 
If some one is still in denial to get the proper equipment, he will never see an 24/24 reliable connection .

[Mechatrommer]

1) All those replicated ideas from the internet does not help any one. Its an true waste of time.
2) in a cost of 250$ .
3) If some one is still in denial to get the proper equipment, he will never see an 24/24 reliable connection .

1) its science Kiri!
2) $250? wow, any schematic?
3) i dont need 24/24, just around 5/7 max will be ok.
peace

[Kiriakos-GR]

Open one underground tunnel , and drop in it an UTP cable ..    

This will save you from all the troubles and the dog.  

[scrat]

Search for "can antenna" on google. An oil can or even a Pringles tube and the coax cable can be used for 2.4GHz, even if square section antennas usually work better. This kind of antennas have high directivity, someone used them for "stealing" the wi-fi connection.
The simplest implementation is at 75Ohm, which could be an issue.

I could provide you the math from some notes of the antennas course I attended at university.

[Mechatrommer]

I could provide you the math from some notes of the antennas course I attended at university.

i'll be glad if i have that! pm or here. thanx.

[scrat]

Here it is the simple calculation.

The problem is that of the propagation into a waveguide, so wavelength is not the same as for a wave traveling in the open-space, and depends on the mode involved. Each mode (field conformation) has its wavelength (and propagation speed, of course). Then the ideal situation is when there is only one mode propagating, so above a certain minimum frequency and below a certain other, which are a function of the diameter of the cavity.

Since the back of the can (closed side) imposes a zero to parallel electric field, positioning the pin at a distance equal to one quarter of the length lambda_g gives the best result (emitter antenna is into a point of maximum field strength).

Imposing the minimum and maximum frequencies to stay in between the mode and second mode cut-off:
a = d/2 > 1.83 / (2*pi*f_min*sqrt(mu0*epsilon0))            above the first mode cutoff
a = d/2 < 2.405 / (2*pi*f_max*sqrt(mu0*epsilon0))          below the second mode cutoff

Once the diameter is chosen (for wi-fi, 2.4 to 2.5 GHz, it is easy to find one satisfying the above conditions)
fTE_11 = 1.83 / (2*pi*a*sqrt(mu0*epsilon0))                  first mode cutoff frequency

lambda_g = 1 / ( f*sqrt(mu0*epsilon0)*sqrt(1-(fTE_11/f)^2) )

mu0 and epsilon0 are the magnetic and electrical permeability constants of the air.

For a narrow bandwidth signal, f can be the central value, for wi-fi I'd choose the geometric average f = f_avg = sqrt(f_min*f_max).
In the example on my textbook it isn't mentioned, but IMO there is also an optimal can length, which should be
lambda_g/4 + N*lambda_g/2, with N at least 2 or 3. In that case, one could try to estimate the antenna gain considering it as a circular aperture antenna. I can't give the right formula at the moment, just can say that radiation goes as a Bessel function of the angle.

The maximum in the radiation diagram is in front of the opened side of the can, of course.
Since the reciprocity theorem (I think it is called so), the considerations made for the emission of radiation are true for receiving, too.

Hope it's worth. If the calculation seems complex, it's just me unable to explain, so just ask.