On Fri, 21 Oct 2005 19:17:50 -0500,
ibuprofin@painkiller.example.tld
(Moe Trin) wrote:
>In the Usenet newsgroup alt.internet.wireless, in article
><a11hl1p7lt7oum1q15m5am78ikpr1q4ifa@4ax.com>, Jeff Liebermann wrote:
>
>>However, I now have a way to make the Wi-Fi signal go around corners
>>and through floors. Waveguide, as 4" dia aluminum dryer vent hose:
>Cough, cough, cough!!!
>Geez, Jeff - warn people when you're going to pull a stunt like that.
Well, ok.
Warning. The following wonderful and clever idea may prove
to be no better than useless when actually installed and tested.
The author offers no guarantees of functionality or performance
and will not be responsible for any subsequent or consequential
damages to one's finances or sanity.
Happy?
>>The feed at one end is exactly like the traditional 1/4 wave Pringles
>>can antenna feed.
>A TE11 mode is about the best, and yes this does work.
Yep. I've done a bit of WG plumbing with WR-90 at X-band in the
distant past. Actually, I think I still have some of that stuff
buried somewhere. I even have some 2K25 klystrons. Cleaning the
garbage out of the choke joints seemed to fix most of the loss
problems.
>The MAJOR
>problem is that the hose is very easy to damage, and the VSWR goes
>out of sight.
Actually, the stuff I bought is rather solid construction. It's
fairly thick and stiff. Seems quite durable. I figure it can be used
between floors, around corners, in overhead ceilings, and other places
where it's unlikely to trampled or punctured.
>Traditionally, use of circular waveguide is very
>limited for two other reasons. First is the lack of polarization
>stability. You may put a wave in "up and down' motion at one end,
>but it's anyone's guess what's going to come out the other end.
>The second problem is mode skiping, as there is little to prevent
>higher order modes. The main use I've seen for circular waveguide
>is rotary joints - invariably using the TM11 mode (the equivalent
>of a probe in the exact center of the end cap). The transition
>between non-circular waveguide and circular is usually a "hole"
>between the sidewall of the non-circular guide, and the end of
>the circular guide.
Hmmm.... Well, this is a bit different because one end of the WG is
directly connected to a polarization insensitive antenna. I was
thinking of just a big cone forming a horn antenna. It should accept
or belch any polarization.
The feed end is a problem. The 1/4 wave probe will probably screw up
badly as the polarization varies. I don't think it's all that
important because the polarization from the client radios isn't all
that well controlled. They come from all directions and reflections.
If it's a problem, I was thinking of a full wave loop antenna mounted
perpendicular to the base about .75 wavelengths above the base. If
that doesn't play, I have a few ceramic backed patch antennas that
should easily fit and are somewhat circularly polarized. If
polarization sensitivity really becomes an issue, I'll install two
cross polarized 1/4 wave probes and feed them to each of the two
access point diversity antennas.
>>That means that there's very little attenuation in the waveguide.
>Extremely crude guess - should be under 2.5 dB/hundred feet. Copper
>or silver plate would reduce the losses significantly, possibly
>below 1.0 dB/C'.
There was a web site showing a Wi-Fi waveguide using galvanized
downspout. If that worked, this should be MUCH better.
Methinks for this exercise, I'll abandon 40 years of RF experience and
resort to just building it and see what happens. My neighborhood
wireless LAN as a dead spot created by a very large tree near the
cental access point antenna. I was going to add a 2nd access point
but it would be more interesting to see if the flex waveguide can
deliver the signal around the tree.
>>I'll burn some time this weekend modeling this idea and see if I can
>>find a more efficient end feed (probably just a bigger horn).
>What blend of aluminum is this?
I have no idea. Since it's fairly flexible, it's probable close to
dead soft (pure) aluminum. I'll see if I can get some details from
the manufacturers.
>There's a substantial difference in
>the conductivity between 1100 (nominal 60% compared to .999 Ag) and
>6061 (nominal 40% compared to .999 Ag). That conductivity has a
>large effect on attenuation, as suggested above.
I didn't know that. That's also assuming that it's not plated,
anodized, or alodined. I'm not sure that this stuff has on the
surface to keep the high humidity dryer exhaust from etching the
aluminium in powder.
>Carrying this though a little further - 4" galvanized steel dryer vent
>is going to be powdered dog poop compared to solid aluminum of any
>kind. The only advantage is that it would be stronger, and somewhat
>less liable to mechanical damage.
I won't damage it. I promise.
Somewhere in cyberland is a long rant I wrote on why the lousy surface
conductivity of the average coffee can does not necessarily make a
suitable microwave antenna. I also wrote about illuminating the HVAC
ducting with wi-fi in a hospital for improving the coverage. That
worked but proved unpopular with the administration.
However, even if the results for this waveguide are less than optimum,
it does have potential for ducting RF between floors, around corners,
and in awkward locations.
--
Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
831.336.2558 voice
http://www.LearnByDestroying.com AE6KS
http://802.11junk.com Skype: JeffLiebermann
jeffl@comix.santa-cruz.ca.us jeffl@cruzio.com 
Any inexpensive ways to increase range indoors ? 
Linear Mode
