Cantenna proj. question

I've seen several "cantenna" projects on the 'net. They vary from 
using a large coffee can, a cookie can, to one for orange juice. One 
thing I've noticed over all projects is that a can with a flat bottom 
(not one with those strengthening ripples in it) is preferred.

I know that the wavelength specs (1/4?) determine the length of the 
"probe" (wire or tubing soldered to inside of the connector) and the 
distance that is positioned from the rear of the can. It is not clear 
to me, however, how/if the size of the can effects performance.

When choosing can size, is bigger better? One design uses a 
heater/air conditioner duct "funnel" attached to the can to increase 
gain. Would starting with a larger diameter can and eliminating the 
funnel be a better design?

Of course, maximum gain, given the limitations of the project (no 
high-tech evaluation test equipment), is the goal. I want to know how 
can size effects gain.

Another question is about the coax. I know that in order to put this 
antenna on the roof, I'll need 20 feet, or so, of coax. "Pigtail" 
(small diameter stuff) isn't sufficient for this task, I know, so the 
beefy RG-8-type stuff ( *not* RG-8; I just don't remember the 2-plus 
GHz spec, right now...) is required.

Am I dreaming? Is it possible to capture a signal and run it 20 feet 
and still have sufficient signal to make it worth my while?

And a basic question about his whole process: Is there a better way 
to capture a remote WI-Fi signal than to use an external antenna and 
run it down a coax and plug it into my wireless card in my PowerBook? 
Another option would be a bridge (which would convert it to wired 
medium), yes?

Looking for feedback, constructive criticism, and alternative ideas.

Thanks,
_______________________________________________
BAWUG's general wireless chat mailing list
[unsubscribe] http://lists.bawug.org/mailman/listinfo/wireless