Fast, Cheap, and Easy Wifi Cantenna

This instructable is on how to make a wifi wave guide antenna for FREEEEE! (Assuming you already have a few things.) 

Step 1: Materials

Can (I just used a #10 can)
Knife
USB WIFI dongle
USB extension cord (my dongle came with one)
That's it.

Comments

[Chad Baxter]

Don't touch the can...six volts of electricity flows thru that USB dongle. It would be best if you insulate it

[talhaijaz]

wow! it is good i like it

[ederfel]

This article has some very useful information about can dimensions and other aspects which affect the gain:

http://www.h-online.com/features/Tin-can-aerial-materials-747480.html

In particular, it advises smooth sided tins, and using a wire to avoid excessive intrusions into the tinspace.

[XboxModz]

Yes. SMOOTH is best. My latest Cantenna (I have experimented with a lot of different can sizes/types) is the best one that I have built to date. It has smooth sides (no ripples). It was an X-mas cookie container that measures 3" Dia x 10" Long. PERFECT for a Cantenna.

I didn't use any dongle tho. I built it the traditional way using the "N" female chassis mount connector and copper wire probe.

[knownassociate]

In reading the comments I saw several questions regarding the exact placement of the usb stick. This is a link that will calculate that for you according to the measurements of your can

http://www.turnpoint.net/wireless/cantennahowto.html

That site refers to a cantenna made to connect to a wifi card installed into a pc, but as it worked well, the two must not be too dissimilar.

[Douglas 99]

Found that this #10 USB Cantenna works great, found that it work even better if you add another #10 can to it, just cut out the bottom and duct tape it on, or hot glue gun works good also. Even better directional signal for me! Make sure you slowly rotate around the neighborhood, you will be amazed what this pulls in. I pull in 3 signals without the Cantenna, I can pull in about 30 when I put the same dongle into the cantenna, I put it about an inch (2.5cm) off the bottom of the can just like in the above picture, But my USB case is NOT removed. Also the higher you can mount it the more or better signal pulls in., mine works better on the second floor pointed out the window. Roof mount even better, make your mast so you can rotate the cantenna. also plastic lid on the open end helps to keep it weather proof. anyone find any other improvements let me know, I love this Cantenna, it is simple and easy. [e-mail]

[lnascimento2]

Worked fine! i've upload the image of the result of my wifi cantenna.

[omgmissles1012]

i see that your can is cut open, so i assume if i do that itll work like this, |_ -->, so the line going up represents the tin can that is left there, and the side that is open is were the wifi is directed?

[pgwbali]

I live in Bali and use a wireless modem (uses a SIM) would this instructable increase the signal strength?

[pcooper2]

Yes, in a way. It doesn't increase transmitter power or receiver sensitivity, but it concentrates the available power into a narrow, directional beam that can reach farther than the omnidirectional antennas used in most Wi-Fi gear.

[Greasetattoo]

Would mounting to the top of my house make a big difference?
Higher up, better??

As, there is a school about block away from my house.
My laptop the way it sits, does not pick it up!

[sidorthatotherguy]

You might be able to pick it up, I pick up a couple from around 200 yds away, but as for on your roof, usb cables aren't meant to go more than like 6 feet I think.

[gburnaska]

USB's max is around 18 feet... thats why the longest USB cable is only 16 feet

[Sounds4cc]

I have successfully coupled 2 - 16 footers and a 6 footer together from my laptop to an electronic flow monitor. worked fine.

[gburnaska]

no doubt you can go longer.... just not if you want speed.... try 35 feet to a HDD and transfer a 2 gig file.... if you want any chance in keeping the 480 speeds you should use a repeater

[pcooper2]

Not likely to work. USB works by the host (computer) polling devices on the bus. Responses from devices need to arrive back at the host within a certain length of time. At about 1 nanosecond per 9 inches of cable, exceeding the maximum cable length allowed by the USB specification even by a couple of feet is likely to cause communication over USB to fail.

[StuNutt]

The "spec" for USB2 is 6 METRES! I'm currently using a wifi connection to my hotel with a 6m cable to put my USB adapter in a better position to receive the distant hotel AP (other end of the corridor)

Might give this "cantenna" a try too though.

[psykopenguin]

For USB 2.0 or earlier, the maximum length of a standard cable is 5 metres (16.4041995 ft). For 3.0 there's no exact number but we figure 3 metres (9.84251969 ft).

[sdobbie]

You can get a USB cable much longer than that if you use active repeater extender cables. I got a 5 meter on off ebay which works with everything except external hard drives.

[psykopenguin]

Well you could do that but it's not really so much a single cable at that point anymore much like USB over CAT5(e) but you are correct that you could technically go beyond that point. :)

[sdobbie]

So how can cat5 cables be hundreds of meters long but usb cables stop working after a few meters?

[pcooper2]

USB cables work fine over hundreds of meters, but the communications scheme does not. The host polls devices on the bus and according to the standard they are expected to respond to the host within an allotted time. The electrical signal propagates along the cable at something close to the speed of light, about 1 nanosecond per 9 inches of length. Exceeding the maximum allowable cable length even by a couple of feet will delay the communications enough to cause timing errors, i.e., communications failure.

[eleers]

Cat5 cables can only be 100 meters... you start losing signal due to interference and electrical resistance after that. Fiber cables are faster and can be much longer, but they're way more expensive.

[imBobertRobert]

probably, but wifi isnt usually that strong, and since so many schools have the wifi in the center, and have really thick walls, i say probably not.

[skrubol]

You don't mention any measurements in this. Logically you want the antenna part of the dongle to be in the middle of the can, but the distance from the back is important.
IIRC 1/4 wavelength is optimal (or is it 1/2 wavelength?) For 2.4 GHz this is about 3cm (or 6cm.)

[fmeow]

What kind of speed do you get using this method? and why did you remove the plastic cover of the wifi dongle?

[sidorthatotherguy]

I stepped on the dongle and the case cracked so I removed it. The Netgear software says 54 mbps and 4 bars. I can play games without any lag whatsoever.

[fmeow]

great, thx, I had tried to build a cantenna before, but it didn't work, and I had no idea why.

Glad to see it's so simple, will be making one of these shortly... thanks again for posting this.

[нјовичић]

how to do it with USB that have own antenna?
I have this one
http://images.mondoplast.ro/Adaptor-wireless-USB-54Mbps-TP-Link-TL-WN422GC_291_3.jpg

[sidorthatotherguy]

I am not sure, but I think there are other instructables for things like that.

[Doug64]

Fiber cables aren't faster, but actually slower than copper.
The typical nominal velocity of propagation in fiber is .68 times the speed of light in a vacuum, while copper is more like .72. The difference is that you can run a longer distance with fiber, dependent upon what version of copper to fiber convertors are used.

[lourens01]

When I left transmission in 1998 we transmitted 2.5Gbits per second on fibre. This translated to 30720 voice conversations. There is no way that you do this with copper. Copper has an impedance which increase as the frequency increase. In other words; the higher the speed in copper the higher the impedance. This puts a limit on copper in terms of speed and distance. It is therefore wrong to say copper is faster than fibre, because you can transmit at higher speeds using fibre regardless of distance.

[Doug64]

Ahh, there's the difference. You're talking data rate, and I'm talking speed. Two entirely different things.
Data rates on fiber are up to 100Gbits in data centers now.

[Mick Gibson]

With my "n" series card, I and getting over 1/2 miles away. I have over 42 sites, with 30+ sites one channel. I used 3 db 8 antennas.

[atombomb1945]

You might get even more of a focused signal for longer distances if you were to put the dongle in the bottom center of the can rather than off to the side. How far away are you from your WiFi router in your home and what kind of walls do you have between the two?

[ghurse]

Increasing the gain of an antenna in effect "Focuses" the signal.. placing the Dongle/signal feed in the centre will not achieve this..

The "Cantenna" is a Waveguide antenna. The feed point is a specified fraction of a Wave guide-wavelength in front of the reflector (the back of the can). this is calculated from the dia of the Can in relationship to the frequency of operation.. and is fed from a side because the signal is "Launched" into the guide, it isn't radiated like in a Dish or reflective array (Yagi).

You can optimise the length of the tube to get maximum gain.. this is a odd (?) mutiple of g-wavelength/2... in 12GHz system seen people build 'Periscope" style setups.. oh.. And attaching a "Horn", in the form of a cone can improve the capture (apparent gain) of the Cantenna..

[sidorthatotherguy]

I'd say I'm about 30 yds (not too far), and the walls are just the usual insulated outside walls. We also have some kind of fiber cement siding. Hardiplank I think it's called. I will have to try putting the dongle at the bottom some day. Thanks for commenting!

[StuNutt]

Why did you (need to?) remove the plastic casing from your USB adapter?

[sidorthatotherguy]

i didnt. i stepped on it and it broke.

[cournal]

this is fantastic!!
will this work for the itouch 4g aswell? i'm trying to figure out how i can get wifi in my itouch wherever i go. im addicted to the google maps and i wan to know where i am and ofcourse using texting app.