When we take a new Wi-Fi router from its box, the stock antenna is a short plastic stub with a reverse SMA plug on one end. More recent and more fancy routers have more than one such antenna for clever tricks to extend their range or bandwidth, but even if the manufacturer has encased it in mean-looking plastic the antenna inside is the same. It’s a sleeve dipole, think of it as a vertical dipole antenna in which the lower radiator is hollow, and through which the feeder is routed.
These antennas do a reasonable job of covering a typical home, because a vertical sleeve dipole is omnidirectional. It radiates in all horizontal directions, or if you are a pessimist you might say it radiates equally badly in all horizontal directions. [Brian Beezley, K6STI] has an interesting modification which changes that, he’s made a simple Yagi beam antenna from copper wire and part of a plastic yoghurt container, and slotted it over the sleeve dipole to make it directional and improve its gain and throughput in that direction.
Though its construction may look rough and ready it has been carefully simulated, so it’s as good a design as it can be in the circumstances. The simulation predicts 8.6 dB of gain, though as any radio amateur will tell you, always take antenna gain figures with a pinch of salt. It does however provide a significant improvement in range, which for the investment put in you certainly can’t complain at. Give it a try, and bring connectivity back to far-flung corners of your home!
We’ve covered quite a few WiFi Yagis here over the years, such as this rather extreme wardriving tool. But few have been this cheap.
Thanks to London Hackspace Radio Club for the tip.
57 thoughts on “A Simple Yagi Antenna For Your Wi-Fi Router”
I did something similar once, when our office internet was out and I knew someone in the building across the street. I boosted a wireless router in bridge mode by placing it in the bottom of a mesh steel wastebasket propped up on a chair and pointing out the window.
…Ok, what I did was absolutely nothing like the article. Good to see someone actually simulate and design a directional antenna modification rather than balancing pieces of tin foil and drawing circles around the router with salt.
wgr614 was a great router. sad the ddwrt software stoeed going for it years ago :(
ddwrt is still available and so is openwrt.
DD-WRT was awesome in 2008, but a lot has changed in hardware / software / security.
Check out ASUS Merlin for 2017.
I am hugely skeptical of 8.6dB of gain. Normal 9dBi yagi antennas have a lot more elements to reach that number – the driven element, the reflector, and at least three directors. This one only has one director.
Maybe some empirical testing would be worth mentioning.
you read the article right?
“The signal to my wireless adapter card dropped 6 dB when I removed the parasitics from the router dipole.”
I think that is his point. The article (did YOU read it?) says that the simulation predicts 8.6 dB of gain. That statement is misleading at best, as an ideal Yagi antenna yields 7.5 dB gain (see http://www.radio-electronics.com/info/antennas/yagi/yagi-antenna-gain.php and other references) over a dipole, IN ONE DIRECTION. But if you are planning to keep the same coverage in other directions when using this mod, you will find out where that extra signal came from. It is a “zero-sum game” in this case. You will get more signal in one direction and lose exactly the same amount from areas where you used to have a good signal.
I quote my words above: “though as any radio amateur will tell you, always take antenna gain figures with a pinch of salt” :)
parabolic wifi antenna made from cardboard and alfoil, with downloadable pdf! http://www.instructables.com/id/DIY-WIFI-Antenna-Reception-Booster/
I don’t know much about this but doesn’t the wifi standard have legally constrained rf emission power limits? And wouldn’t directing the emissions with an antenna like this breach the allowable power? Perhaps no-one cares but if you were close to your neighbours wouldn’t you make their signal to noise ratio worse? Do we need to be careful not to be antisocial with this kind of modification?
The transmit power is unchanged, only the EIRP increases (in the main direction). But best to use it so that it directs the power onto your own premises – You do normally not want to supply your neighbor with your wifi connection :-)
There is an FCC limit on EIRP as well as output power, but consumer access points are well below it, so a bit of a boost should be ok.
I looked it up: 30dbm power into up to 6db of antenna gain. With more antenna gain, the power needs to be reduced. Since this antenna probably doesn’t have more than 6db gain, and the router is likely below 30dbm, it’s ok.
What does it do to the 5GHz emissions of a dual-frequency router?
Nothing good :)
Be aware that, to remain legal, if you use a 10dB antenna then you must reduce the transmit power by 10dB. Hence the received power will be unchanged.
OTOH, you can use a high gain antenna on the receiver without restriction; this reduces the interference you suffer due to other transmitters.
The key point is that the transmit power is defined as EIRP (effective isotropic radiated power), i.e. the equivalent power that would be radiated by an isotropic antenna. The reason is that the power limits are defined to contain the chance of interfering with other users.
There is no transmitter and receiver in wifi. All cards are both transmitter and receiver, so you could not only use a receiver antenna.
I think you missed the point that Wireless LAN is not unidirectional. While it is a half-duplex technology, there are no receivers and transmitters. The devices on each end are transceivers, which means they receive _and_ transmit. If you use a high gain antenna on either end, it will not only boost the received signal, but the transmitted one aswell.
I entirely understand that, having been involved in (and using) WLANs since before 802.11 existed!
Your point does not affect nor negate my point: you cannot *legally* increase the power transmitted in one direction using a directional antenna.
You can have separate antennas for tx and rx, but in most situations that won’t be advantageous.
I’ve been able to configure antennae routing before where 1x interface is RX and the secondary is TX….or was my radio lying to me?
WiFi emissions rules vary by country. The US rules are based on EIRP (as referenced in an earlier post) with a maximum of a specific signal level. I’ve been through them pretty thoroughly a few times and I don’t recall ever seeing a non-measured universal regulation such as “you can’t increase the power transmitted in one direction using a directional antenna”. Are you referencing the rules of some other country? If you think that applies in the US, please cite chapter & verse. A number of very much type-approved directional antennas frequently used in commercial WiFi in the USA disagree with your statement as it is written…
Unless your transmit power I already below the FCC limit.
Even if it’s it, it’s a pretty small concern, realistically speaking. Unless you’re in a sensitive RF area no one is likely to come looking.
No. The rule is your EIRP cannot go over a certain amount. To assume that you cannot to any degree focus the energy in one direction more than the stock antenna already did is to assume that the EIRP of the router with the stock antenna was already sitting right at maximum.
This is unlikely to be true.
So your argument “you cannot *legally* increase the power transmitted in one direction using a directional antenna.” is wrong.
Now.. if you want to use some specific numbers including the actual power output of his router and the actual db gain of his new antenna you may or may not end up with a point.
It’s dBi. Please understand dB by itself means bugger all. It needs to be compared to something so we get an absolute value. Some examples of correct use:
0dBi + 8dB = 8dBi
0dBm + 0dBm = 3dBm
0dBm + 3dB = 3dBm
It’s my pet hate when people use dB for absolute values. It’s meaningless by itself.
Very true :)
0dBm + 0dBm != 3dBm
Oh yes it does.
1mW = 0dBm
0dBm + 0dBm = 2mW = 3dBm
And this is why dB is an insane system of units. You get equations such as 0+0=3. Who ever thought that was a good idea?
In EU the limits of use for 2.4ghz band is specified in e.irp, or contains the antenna assembly so you may violate the law if you change the antenna to something with too high gain. See http://www.etsi.org/deliver/etsi_en/300300_300399/300328/02.01.00_30/en_300328v020100v.pdf for details
Find the wlan adaptor with the highest receive sensitivity -92dBm -96dBm. Get a high gain antenna like a 15 element yagi. Turn down your transmit power until eirp is 100mW (when in EU) and you can sustain a link to a router 1.5 mi away.
Hmm. if you pry open the original antenna, you will find a coax cable of which the top part is stripped to only the core wire. the stripped length is the antenna. giving the rest of the cable a yagi director/reflector has no use.
The article claims that this one is a sleeve dipole, which is what you describe plus a 1/4 wave cylinder extending downward. If that’s the case, it needs the full length of the elements. If it’s really a 1/4 wave above ground, each element would need a ground plane if they were half the length.
Stupid question from a non-radio guy: How do you use this with a two-antenna (presumably dipole) router?
It’s conceivable that a two-antenna router is already doing some phased-array magyyks, in which case you probably just don’t.
If, on the other hand, it’s using one for receive and one for transmit, point them both in the same direction. Same if it’s a dual-band router with one antenna for 2.4 GHz and one for 5 GHz, but you’ll need a different set of parasitics for the 5 GHz antenna.
In most cases, the two antennas are for diversity TX/RX. In other words, the receiver uses whichever antenna receives the strongest signal. This is done because there are lots of dead spots around your home/office at those high frequencies. As you move your phone or laptop, etc. you will sometimes have an object blocking the signal, but with two antennas, it very unlikely the path to both antennas will be blocked at the same time. If you want to modify the antenna by adding elements to obtain some gain in one direction, you can add them to either antenna, as long as your WiFi router is configured for diversity TX/RX.
Yoghurt instead of yogurt? Interesting misspelling at least.
I won’t be doing this because my AP reaches outside my entire house and the antennas are built in.
It’s not a misspelling at all. It’s how it is spelled in the UK, where the author is from, much the same as color/colour, plow/plough, etc.
Yogurt is less painful.
It’s the stuff we eat in the UK. Much tastier than yogurt, which is clearly inferior.
Very clever hack. For those concerned about EIRP, know that the guy who did this is an amateur radio operator and is therefor legally able to run higher power levels (effective or actual) than unlicensed individuals as long as the transmissions meet all applicable laws.
I expect the relevant laws allow higher power transmissions in ham radio bands, not in ISM bands.
There is overlap for 802.11b channel 1. It’s in the ham allocation.
That would also require following all the other ham radio rules.
In the US that would mean:
– no codes or cyphers meant to obscure the meaning of the communication
– nothing of pecuniary interest to the user or the user’s employer
– id every 10 minutes
That last one is pretty easy, just set the ssid to your ham callsign.
Those first two though.. kind of goes against how most of us use WiFi doesn’t it?
Of course, I’m just talking about US laws. I don’t actually know what the laws are in the UK but I bet there are pretty similar!
Why is no one 3D printing these already?! I WANT :D (I don’t actually have a AP like this anymore, but if I did, I would…)
Because it would take many times longer to print than to measure out and cut, never mind the design time. Not every problem needs to be ‘solved’ by a 3D printer.
then we’ll CNC them! or… laser cut!!
If I wanted to find Wifi transmitters on a certain frequency, what kind of antenna would I use? In my opinion I’d need an antenna which is as directional as possible. I don’t know very much about antenna design and RF in general, and so far I haven’t found satisfactory solutions on the internet.
My basic idea would be to use my Ubertooth One to monitor the signal strength of a particular frequency, and then correlate that with certain events/frames.
The frequencys are so close, arn’t they? Would cost too much for almost no benefit. I could be very wrong though.
The wavelength is small enough where a collinear(composed of tandem “sleeved dipoles”) wouldn’t be too long and would provide the gain to the entire 360 degree horizontal plane instead of a 30-90 degree beam.
You could put that router at the focal point of an old satellite dish…Forgetting the legalities of this stuff for a moment, have done plenty of tinkering in my day… if you wanna get serious with extreme range, try an Ubiquiti Bullet M with Air max +28dbm screwed into a 24dbi dish, you can configure as an access point, station or repeater. This stuff isn’t that expense, and once you’ve worked with this stuff you’ll never go back… I have 3 of them on my house here in Mexico and they work for miles.
This is what I need. How ever, I have a cable set-up (Xfinity) supplied by the company; and the router does not have an antenna that I can see. So, what can I do to increase my range?
Ideally: buy a cable modem and a router with an actual external antenna. Return the modem router combo abomination you have been renting and save a few dollars each month on your bill too.
Next best option: return the abomination and get them to rent you a plain modem. Buy the router yourself as seen above.
Not so ideal: Go into the setup of the Comcast router and turn the wireless part off. Buy the router as above. Plug it’s input into the output of the Comcast router. Configure the Comcast router to have the ip address of your router as a DMZ.
Least ideal: Just like above but the Comcast router also is active on WiFi. That might be necessary if you are part of their ‘on the go’ program. At least try to put them on separate channels or something.
I’d like to do this for a dual band router. Assuming the antennae simply contain a pair of dipoles (can anyone provide a link to a teardown of the typical dual-band antennae that come with consumer routers?), is this just a matter of going through the same design calculations for the 5 GHz band (if anyone has done these already, a link would be much appreciated) and adding the appropriately measured and spaced pieces of copper wire? Or will the 2.4 GHz and 5 GHz elements interfere with each other such that this won’t work?
Please be kind and respectful to help make the comments section excellent. (Comment Policy)