Hacker News new | past | comments | ask | show | jobs | submit login
Turbo-boost your WiFi signal - in 15 minutes (eyesandfeet.com)
106 points by luxative on June 10, 2010 | hide | past | favorite | 38 comments



If you want to spend a little bit more time for a better result (while learning a bit) you can homebrew a more solid solution.

Directionals are pretty easy and fun to make out of everyday items, which do the trick quite well. Here's some examples (http://www.turnpoint.net/wireless/has.html) and here's one you can do in maybe 30 minutes (http://www.turnpoint.net/wireless/cantennahowto.html)

Those antennas are good for long-haul point-to-point links, but can also do a pretty good job in point-to-multi, if it's close by. I gave a neighbor internet access for a few months with one of these pointed at his house, without any special gear on his end.

Omni-directionals are a bit more advanced, but there are some good designs out there. I built one similar to this once (http://wireless.gumph.org/articles/homemadeomni.html). Because these are a lot more sensitive to flaws than directionals, it's important to test.

Sector antennas are a good compromise between omni and directional, as it gives you a nice wide area of coverage (http://www.grynx.com/projects/wide-angle-sector-antenna-14-d...)

Keep in mind that just installing DDWRT and cranking up the juice doesn't always result in a better wifi experience (though sometimes it can give access to that one dark spot), and is probably screwing up everyone else's experience. Hooking an ipod up to a bullhorn gives you nice range, but there's a lot less fidelity.


I spent an entire afternoon trying to do this about 2 years ago from some stupid paper cut-out design and tin foil. I never felt so punk'd in my entire life.


I had the same experience, didn't help much at all


I can imagine - must be disappointing when you take the effort and nothing happens. Personally, I had a fantastic improvement in signal strength. In fact, strangely enough, the signal also improved in the room right below the one where the directional+wifi was.


In Buenos Aires, old apartments are made of thick, stone walls that really hinder signals. In a long apartment you have to repeat.

I bought 3 WRT-54GS refurbs from newegg for $29 each, installed DD-WRT, set 2 of the three as Bridged Repeaters, put them all on the same SSID (protip: put the word Mega in front of the old SSID), spread them across my friend's house, and he now has great signal everywhere.


$29 is a steal. One thing to consider in Bridged Repeater mode, from: http://www.dd-wrt.com/wiki/index.php/Repeater_Bridge:

"Also take note of the fact that all repeaters, including this Repeater Bridge mode, will sacrifice half of the bandwidth available from the primary router for clients wirelessly connected to the repeater. This is a result of the repeater taking turns talking to not just one partner, but to two, and having to relay the traffic between them. As long as your internet bandwidth requirements are within this halved bandwidth amount there will be little or no reduction in "speed".

But realistically speaking for general wireless, its just fine.


Why the word 'Mega'.


shameless self promotion. so that every time my buddy has great signal he remembers it's because of his tech-friend's dd-wrt hacking prowess.


Or, buy a Linksys WRT54 router, load Tomato (http://www.polarcloud.com/tomato) and jack up the signal level. Be careful not to exceed the legal limit (or not).


Boosting the Tx power of the router only solves half of the problem. Wifi connections are symmetric; the clients need to transmit back to the router, too. A bigger antenna (even home-brewed like this) boosts Tx and Rx capability, so it's a better solution.


DD-WRT lets you do the same. The power on my Netgear defaults to 71 out of 251 (mW) :)

I'm seriously loving DD-WRT, far far far more capable. And IPtables!


does it really help to boost the signal? i tried before and didn't seem to get much of a difference.


It depends on the quality of the amplifier in your radio. Like a cheap stereo, if you turn it up to max volume, you can hear it louder, but the signal gets distorted.


Haven't been able to test it, just looked for a comparable setting. I'd bet it depends on the router, but I should be able to actually test tomorrow. Want me to?


sure - i think one of the above answered it for me that there is no equivalent boost on the client end which i think is why it hasn't helped me much.


I tried this technique on my WRT54 (running Tomato jacked way up already) and it really did seem to help. But ultimately the stupid cats kept knocking the antennas off so I just set up another router at the other end of the house.


What would I do for a Cisco WRT54G2 V1 with internal antennae? The thing is flat, should I sit the entire unit in a 'bowl' and direct it at my office?


An alternate method I've used is building an antenna on the computer side if you can't build it on the router side. You can get wireless adapters which are in the from of a usb thumb stick. Coupled with an extension cord to connect to a PC, you can put the entire adapter at the focal point of a large dish.


Reflected waves can help or hurt (if echoes interfere with themselves). So even without a intentional reflector like this, be aware that in challenging environments, repositioning your antennas just a little (new angles, a few feet further or closer to a signal-reflective surface) can make a big difference.


I just made a poor quality one of these with plain old paper and aluminium foil. My iwlist scan results:

Frequency:2.437 GHz (Channel 6)

Before: Quality=40/70 Signal level=-70 dBm

After: Quality=47/70 Signal level=-63 dBm

Not very exciting, but not bad. Lots of access points in my block make connectivity a real problem, so I'll take whatever I can get.

My roughly-cut parabolic reflector isn't super-parabolic, make of that what you will: maybe a better shape would be markedly better, maybe it means yours doesn't have to be perfect either shrug

My tips if you want to try yourself: tape the tabs in place but try not to put creases into the foil when you do. If it's not already clear, only put foil on the back reflector (not the curved support)


If your dBm readings are correct, your signal strength increased by 10^((-63-(-70))/10) = 5 , which is not so bad.


I have to wonder if the parabolic shape really matters, or if a large conducting panel of any shape would suffice. I have no idea, really, but would love to hear from anyone who has.


Any reflecting surface will improve it in the reflected direction, the parabolic shape is just to improve the distance further and give better aim.

Just about anything reflecting works, my personal favorite directional antennas use Pringles / coffee cans or a metal colander.


I prefer the salad bowl method:

http://www.youtube.com/watch?v=LY8Wi7XRXCA


exactly what i was thinking, classic


On the same page of the cutout design (http://www.freeantennas.com/projects/template2/index.html) there is what the author of the post call a "scary graph". Looks like a cool Windows based signal monitor thing. Does anyone know it?


It gives you more bars but does it give you more speed?

And if this actually works, why don't commercial products build this right in? Couldn't you reflect more signal with a much smaller reflector if you're up close to the transmitter?


I made a reflector from the same template a few years back, worked great. The reason it works is it improves both transmit and receive. The reason why a regular consumer AP doesn't come with that sort of antenna is it's directional, basically you trade wide coverage for better coverage over a narrower angle and most people have a hard enough time just standing their omnis straight up.

For more commercial work there are plenty of options: http://www.l-com.com/category.aspx?id=2073


However if you place it in one corner of your house and install a reflector with a 90-degree coverage angle you should be able to increase your signal strength up to ~4-fold with a perfect reflector as you're 1/4 your coverage area. Realistically getting a 50% increase in strength is great.

Using materials like aluminium or copper would greatly increase the performance of your reflector. If you by chance had a copper mesh, it would kick the tin foil a mile. I personally have access to sheet aluminium so if I were ever to build one of these I'd definitely be making it from that and not foil. (IIRC reflectivity in metals is improved with thickness, with these tin foil reflectors you're probably still getting a strong enough signal at close range from the side of the reflector you don't want signal.


yeah, a smaller reflector close the transmitter wouldn't do much to help it as a receiver. I'll bet there's an optimal size reflector and distance from the antenna.

and, the same would be true for your laptop too, right? a reflector at both ends.


Well this would only work if your wifi router was on the far side of the house and you want to direct the signal to the rest of the house. If it was in the middle of the house it wouldn't work as well.


If it were in the middle of the house, this issue wouldn't arise in the first place.


Unless you had a REALLY big house :P


Absolutely. My wifi (pic in the article) was in one far corner of the house.



I don't see a turbine anywhere. There's not even any amplification (boost). It's just a directional antenna.


Thanks dude


Good way to cook the children.

Lots of routers make annoying high pitched noises, even without that the flashing lights are a distraction - I wouldn't put one in anybodies bedroom (unless I could turn the lights off and be sure it's silent).




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: