Calculating "velocity factor" (wifi antennas) / Electromagnetic Radiation / Forums

Forums

4hv.org :: Forums :: Electromagnetic Radiation

« Previous topic | Next topic »

Calculating "velocity factor" (wifi antennas)

1 2 next

Move Thread

LAN_403

Fnord

Sun Oct 30 2011, 04:12AM

Registered Member #2004 Joined: Sat Feb 28 2009, 11:43PM
Location:
Posts: 39

(I'm reusing this topic to ask about velocity factors. Look a little farther down for the relavant post)

I recently made a double biquad using the instructions from martybugs;

...And I'm not even getting a blip from it. There are at least 30 AP's nearby that my little 5dB omni antenna can see, so I think I must be missing something obvious in the construction. I'm just not sure what.

I've attached some pictures. The only way I've differed from the construction is my use of a 2mm aluminum plate, friction-fit to the copper tubing. The tolerances on the element dimension are accurate down to about 0.25mm, and the wire itself is 1.5mm diameter. I haven't added wire supports yet, but it doesn't wiggle around any.

I haven't taken it wardriving... right now I'm just shooting it randomly out the window. I know biquads don't have a huge field of view but you'd think I'd get a weak signal occasionally.

The network adapter I'm using is an Alfa awus036h, rated for 1watt output.

Any ideas?

Mattski

Sun Oct 30 2011, 06:11AM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

Hard to tell in the pictures, but the copper wire going in to the coax center pin isn't touching the pipe is it?

It might also be worth soldering the pipe to the aluminum plate to make sure you've got good contact all the way around. Sometimes the natural oxidation of aluminum can cause conductivity issues.

Also try moving the antenna around, this is a directional antenna so it will perform worse than an omnidirectional antenna if it's not pointed properly, also try changing the orientation because it is polarized.

Fnord

Sun Oct 30 2011, 02:17PM

Registered Member #2004 Joined: Sat Feb 28 2009, 11:43PM
Location:
Posts: 39

It's not touching. Also The parts where the wires cross over aren't touching.

I don't have whatever exotic solder alloy is required to bond aluminum, and it's nearly impossible to solder with normal stuff (I've done it before, but it looks like crap). I can tell it's making good contact since I can just poke it with my MM probes, though.

I've tried changing the polarization, too. It doesn't seem to have any effect.

radiotech

Sun Oct 30 2011, 04:02PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

What is the insulating material at the feed point? Are your coax connectors free of shorts?

Mattski

Sun Oct 30 2011, 05:08PM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

Here's a guide to soldering aluminum:

Because of the thermal mass of a copper pipe and aluminum sheet you'll need a lot of heat. A typical soldering iron probably won't cut it, but a soldering gun or a propane torch would do it, maybe a heat gun also. You also need to abrade the surface to remove the oxide. I'd probably apply flux to it after abrading and before soldering as this should help inhibit more oxidation.

If that still doesn't work then there probably are solders specially formulated for aluminum soldering.

Ash Small

Sun Oct 30 2011, 05:45PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Mattski wrote ...

Here's a guide to soldering aluminum:

It looks like the process described here uses the Alumiweld type aluminium brazing rods. I've heard these are difficult to use, due to the oxidation problems, and I don't think they will solder aluminium to copper.

While I don't claim to know anything about wifi antennas, I'd try to use a mechanical connection wherever possible between aluminium and copper.

Fnord

Sun Oct 30 2011, 07:33PM

Registered Member #2004 Joined: Sat Feb 28 2009, 11:43PM
Location:
Posts: 39

Are your coax connectors free of shorts?

BAM! Problem solved! Thanks.
I knew it was something super simple I was missing. Whoever standardized these rp-sma coax connectors needs to be kicked in the face a few times.

Anyway, now that I've proven to myself it wasn't just the laws of physics trying to screw with me, I think I'll try a big AMOS antenna next. Even if it doesn't get much higher gain, it would still look awesome to have an array of them next to your tv antenna/satellite dish :)

Fnord

Tue Nov 01 2011, 01:34AM

Registered Member #2004 Joined: Sat Feb 28 2009, 11:43PM
Location:
Posts: 39

Bumping this up with a question related to amos antennas;

Basically, I'm not sure how to calculate "velocity factors", which is, as I understand it, the speed the electrons travel through conductors that are within strong electromagnetic influence of each other.

The AMOS design seems pretty simple, but there's one thing I don't like about the plans I'm finding out there. The common practice seems to be using a length of coax cable as the "balun" loop. It looks chunky and ugly, is hard to solder precisely, and looks quite prone to corrosion if it is used outdoors.

I'd like to try making a homemade coaxial balun from copper tubing, but I have no idea what the velocity factor would be for this.

In case anyone knows, here are the exact numbers I'm looking at:

Tubing OD: .25"
Tubing ID: .188"
Inner wire diam: .06"
Insulator: Most likely pvc heatshrink tubing.
I could also use teflon tape, which has an unusually high dielectric strength.

Normal velocity factors for coax cables are listed are from 0.66 to 0.83, where "1.0" would equal the speed of light.

...

Tue Nov 01 2011, 02:52AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

The velocity factor is a little more complicated than that, it is the speed at which a wave propagates down the cable, not the speed that the electrons move (which would be the drift velocity, and is quite slow)

Calculating it is tricky, but for a first approximation you can assume that it is a wave traveling in the dialectric you are using, which would make the velocity factor = c/sqrt(relative dialectic constant) For a closer approximation I would suggest looking for a similar cable and noting the values they measured.

Mattski

Tue Nov 01 2011, 07:33AM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

Fnord wrote ...

I'd like to try making a homemade coaxial balun from copper tubing, but I have no idea what the velocity factor would be for this.

In case anyone knows, here are the exact numbers I'm looking at:

Tubing OD: .25"
Tubing ID: .188"
Inner wire diam: .06"
Insulator: Most likely pvc heatshrink tubing.
I could also use teflon tape, which has an unusually high dielectric strength.

Normal velocity factors for coax cables are listed are from 0.66 to 0.83, where "1.0" would equal the speed of light.

Trying to DIY your own coax cable is going to be difficult and expensive. Your first goal is to get a 50 ohm impedance of the cable, otherwise the two halves of your antenna are going to be fed unequal signal powers. You can get this from the following formula for coax based on the ID and OD and the dielectric:

. Now if you have a completely filled space between inner and outer conductors with a known dielectric and a consistent spacing between the conductors then that formula will work, and velocity factor is 1/sqrt(epsilon_r) (eps_r being the relative dielectric constant). But if you're doing a hand-built cable it's going to be fairly difficult to get a high quality cable, the spacing will be off, there will be a mix of insulator and air between the conductors, and all of this means that your homemade cable will be reflecting signals back to the source and have an unpredictable velocity factor.

It will be easier, higher performance, and probably cheaper to just buy a coax cable, which should come with velocity factor (or at least dielectric constant or dielectric type) specified.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob