Jump to content


Photo
- - - - -

Yagi Antenna for Wifi + Expected Range? + Things to Know?


  • Please log in to reply
10 replies to this topic

#1 TheFunk

TheFunk

    SUP3R 31337

  • Binrev Financier
  • 186 posts
  • Country:
  • Gender:Male

Posted 04 July 2012 - 09:48 PM

I'd been looking into designs for a Yagi antenna with a 17dbi output, in the hopes that I might increase my Wifi's range a bit. However, I came across a problem; my house is surrounded on all sides by what can only be described as a ton of foresty-goodness (trees and whatnot). Not being one to be told that something is impossible, I had another idea. My idea is to put the antenna on the roof and then point it, above the treeline, towards the only tall thing around, the top of a nearby mountain. I'd then take the same type of antenna and attempt, after a brief hike, to access my home Wifi from atop the mountain. My question is, in open air, how much range can I expect to get out of a yagi of this sort? The peak, according to a few online resources is about 1800 ft above sea level, which for simplicities sake is about how far I would estimate that my house is from the base. In addition, we'll also assume that my house is about at sea level (it's definitely not). So, a little bit of elementary school Pythagorean theorem work, and we find that in order to broadcast to the top of the mountain, I'll need to have a signal that can reach approximately 2500 ft (and some change) from it's source, or nearly half of a mile if you prefer. I've toyed with the idea of an amp of some sort, but I'd like to keep this project safe, and not go shooting microwaves at the top of a mountain like some sort of mad scientist or something.

If anyone has a few suggestions for what I can do, on a budget, to get this kind of range/functionality, I'm open to all ideas. I appreciate you taking the time to read this, and to my fellow U.S. Citizens, Happy Independence Day!

As a side note Firefly has been on the Science channel all day. Heck yes.

EDIT: In case anyone is wondering about intentions, I'd just like to do this for a single day, in order to say that I could. A half of a mile is a long distance though.

Edited by TheFunk, 04 July 2012 - 10:03 PM.


#2 systems_glitch

systems_glitch

    Dangerous free thinker

  • Moderating Team
  • 1,644 posts
  • Gender:Male

Posted 05 July 2012 - 07:26 AM

I did a similar medium-haul wireless link between two houses back home to share a (very expensive at the time) satellite Internet connection. The antennas I used were "cantenna" style waveguides made from 4" steel conduit with an end welded on and a mounting lug on the side. The tubes of the antennas were around 15" long, and we were able to get an open-air link of over one mile between them.

The hardest part was keeping the far end stable in the tests -- one antenna was strapped to a house chimney, the other was on a camera tripod. At that distance, the slightest misalignment would really degrade performance.

#3 TheFunk

TheFunk

    SUP3R 31337

  • Binrev Financier
  • 186 posts
  • Country:
  • Gender:Male

Posted 05 July 2012 - 09:04 PM

I did a similar medium-haul wireless link between two houses back home to share a (very expensive at the time) satellite Internet connection. The antennas I used were "cantenna" style waveguides made from 4" steel conduit with an end welded on and a mounting lug on the side. The tubes of the antennas were around 15" long, and we were able to get an open-air link of over one mile between them.

The hardest part was keeping the far end stable in the tests -- one antenna was strapped to a house chimney, the other was on a camera tripod. At that distance, the slightest misalignment would really degrade performance.


That's really interesting, did you use some sort of amp by the way? The way I was thinking I would do it, would be to run the cable from the homebrew antenna to an Alfa AWUS036H (1000mW) USB adapter, and then from there to my laptop. The house setup would just be from an access point router connected to the antenna on the roof. However if you got that kind of distance out of your cantennas alone, who knows what I could get with a setup like that. Perhaps some testing.... Thanks for the response!

Did you use a satellite dish along with your cantenna? I've heard that works really well.

Edited by TheFunk, 05 July 2012 - 09:13 PM.


#4 systems_glitch

systems_glitch

    Dangerous free thinker

  • Moderating Team
  • 1,644 posts
  • Gender:Male

Posted 06 July 2012 - 06:14 PM

No amps or reflector dishes, just waveguide (non-Yagi) cantennas and 200 mW Senao 2511 PCMCIA cards. Stability of the antennas and no interfering vegetation are key -- 802.11(b, g) are 2.4 GHz, which happens to get absorbed by water. It probably helped a lot that the conduit we used for the waveguide can was cut (as close as possible with a chop saw) to the right length for the chosen diameter, rather than picking from existing can sizes.

If you want to make your own waveguides from metallic tubing, but lack welding skills/equipment, head down to your nearest well-stocked plumbing supply house and see what they carry for DWV (Drain-Waste-Vent) copper. It's expensive, but you can usually get it in big sizes, and end caps are available. DWV copper can be soldered with a blowtorch from the hardware store.

Don't skimp on your pigtails and jumpers, and minimize the number of connectors between the card and antenna. Each connector is rated with a certain loss, usually in tenths of a dB -- the really cheap ones can be 1 dB or more loss /per connector/! Absolutely no connector converters in the line...buy RF jumpers with the right terminations on the end. For example, if you need to go from RP-SMA to N, don't use a RP-SMA -> RP-SMA jumper with a RP-SMA -> N converter, just order a RP-SMA -> N jumper. Usually you can't avoid going from the card to some intermediate termination; for example, u.fl to RP-SMA bulkhead. The intermediate termination makes waterproofing the RF source easier, too.

#5 TheFunk

TheFunk

    SUP3R 31337

  • Binrev Financier
  • 186 posts
  • Country:
  • Gender:Male

Posted 09 July 2012 - 10:37 AM

Sorry to be a bother, as you've already been more than helpful, but I've been doing my homework so that when my project finally comes to fruition I'll have the best end result I can manage.

On your advice, I'll be avoiding adapters as much as possible. My network card has an RP-SMA connection, so I'm in luck there as well. The only thing I'm stumped on is connecting the driven element (a folded dipole made from 14 AWG copper) to a cable or connector. My instant thought is to solder the copper to an N female chassis mount connector (something like this, with the connection looking like this), then connect that to a short n-male to rp-sma pigtail (such as this). A lot of the designs I see for yagi-udas however, just strip a piece of rp-sma coaxial cable and make a connection similar to this (except with the shielding running to the other piece of copper). I guess my question is, are both of those connections possible/functional ways to minimize signal loss?

Edited by TheFunk, 09 July 2012 - 10:39 AM.


#6 systems_glitch

systems_glitch

    Dangerous free thinker

  • Moderating Team
  • 1,644 posts
  • Gender:Male

Posted 09 July 2012 - 10:46 AM

I guess my question is, are both of those connections possible/functional ways to minimize signal loss?

Yes, both of those termination methods are completely valid. I generally prefer to use N connectors on my antennas. That way, they can be connected to future projects with minimal effort. RP-SMA is pretty universal as an antenna connection though, so soldering a RP-SMA pigtail right to the antenna would not be too restrictive.

I'd avoid those Rosewill jumpers if you can. I pick up most of my RF stuff from the MIT Flea, during the summer months anyway. Otherwise, I usually buy from Netgate. Here's their page of RP-SMA -> * jumpers:

http://store.netgate...RP-SMA-C95.aspx

#7 TheFunk

TheFunk

    SUP3R 31337

  • Binrev Financier
  • 186 posts
  • Country:
  • Gender:Male

Posted 16 July 2012 - 08:24 PM

Reporting in:

My equipment arrived on Friday night and I finished the antenna Saturday morning. I tested her out yesterday and she works perfectly! Pictures and/or a video of the results to follow later this week. In the end I used an n connector to connect the pigtail to the dipole, which I found to be both extremely simple, and highly effective. Thank you for all your advice systems_glitch, and to anyone wishing to attempt this project in the future, I'd be happy to help however I can. Now that I have a general idea of how this process works, I think I'll be making a few more antennae for various frequencies. :)

Edited by TheFunk, 16 July 2012 - 08:26 PM.


#8 systems_glitch

systems_glitch

    Dangerous free thinker

  • Moderating Team
  • 1,644 posts
  • Gender:Male

Posted 16 July 2012 - 10:12 PM

Excellent! Glad to hear it's coming together for you! Do post pictures, and if you've got any hard data on signal strength, troughput et c. related to distance, that would be good as well.

#9 TheFunk

TheFunk

    SUP3R 31337

  • Binrev Financier
  • 186 posts
  • Country:
  • Gender:Male

Posted 21 July 2012 - 05:09 PM

I'm running a few more comprehensive tests now, but here's what I did yesterday. The connection seems pretty solid from just about anywhere around my house, I'll check the throughput later. The video is pretty bad, but hey, it gets the job done.

Oh yeah, and your avatar picture. I've been trying to figure out how you did it. My guess is alcohol vapor in a bottle plus a match? It's pretty cool looking.



Edited by TheFunk, 21 July 2012 - 05:12 PM.


#10 systems_glitch

systems_glitch

    Dangerous free thinker

  • Moderating Team
  • 1,644 posts
  • Gender:Male

Posted 21 July 2012 - 08:40 PM

Oh yeah, and your avatar picture. I've been trying to figure out how you did it. My guess is alcohol vapor in a bottle plus a match? It's pretty cool looking.


Yes, we have a winner! Used to fire them out of our dorm window in college, we recorded a lot of videos and slowed them down to see how different nozzles affected the burn. This one just came out really cool with the neon blue.

#11 PurpleJesus

PurpleJesus

    Dangerous free thinker

  • Members
  • 1,578 posts
  • Gender:Male
  • Location:800

Posted 01 August 2012 - 11:04 PM

Just a thought. Instead of going to the local plumbing shop, How about a local HVAC shop? 3" flue pipe for a gas water heater would be close and alot cheaper then copper.




BinRev is hosted by the great people at Lunarpages!