Wireless not connecting...

[Gene Heskett]

On Friday 31 May 2019 08:24:05 am Mike Marchywka wrote:

> On Fri, May 31, 2019 at 01:33:57PM +0200, Liam Proven wrote:
> > On Fri, 31 May 2019 at 13:25, Oliver Grawert <ogra at ubuntu.com> wrote:
> > > distros should really consider simply shipping paperclips
> > > alongside these isos images ... since it seems to make all these
> > > firmware packages and the related driver mess obsolete ;)
> > >
> > :-D
> >
> > I found the tip online. I was amazed it worked and worked so well.
> >
> > https://discussions.apple.com/thread/4037028?answerId=18677374022#18
> >677374022
> >
> > Sorry, my mistake. Not a paper clip -- the wire was too thick. A
> > twisty-tie, with the end of the plastic stripped off.
>
> On your way to making a conformal dielectric antenna :) Hi tech and
> not know it... The reason for pointing this out though is not just to
> use jargon but curious if there is any practical application here. RF
> fields can be sensitive to some things and if the front end is
> instrumented properly maybe there are sensor applications here- put a
> coil on it and make a metal detector :)
>
Antennas can be spooky things at wifi frequencies.  To be correct and 
stable, the circuit between the amplifier and the antenna should be 
designed to improve the match between the antennas actual measured 
characteristics in terms of both its resistance and its reactance. It 
would be extremely rare to do a smith chart that shows both 50 ohms, and 
a purely resistive reactance, eg whether it looked capacitive, or 
inductive at the test frequency. 

I might add that the test gear to make those measurements runs into the 
30 to 50k$ range and that translates to virtually no maker having it, 
and likely not knowing how to run it if they did.

Its generally considered that an  infinitely thin wire exactly 1/4 
wavelength long is around 36 ohms. Fatter, or the wrong length can make 
it very reactive, and diddle the ohmage from just 2 or 3 ohms to as high 
as almost 1000 ohms  Its also modified by conductive stuff within 10 
wavelengths, which for wifi frequencies is several feet and of course 
its thicker than our imaginary thin wire. 

So the circuit on the back side of that jack on the router isn't always a 
good match, and this is true regardless of the direction, which here 
means transmit and receive. A poorly designed circuit can cost you 30 
db.  Its rarely that bad, but 10 db is quite common. Throw in that 
generally there is no ground plane. Such a wifi antenna needs around 4 
sq ft of ground plane to be able to describe it as good. Usually the 
best you are going to get is the size of the ground plane of the pcb in 
that router.

You can mount it high on a wall, with the stick(s) on top pointing up, 
and thats going to be about the best you can do as you will have made it 
into a sort of half wave driven near the center of that half wave.  And 
the directionality is concentrating the best signal area into a donut 
laying horizontally around it.

Best you're going to get given the limits imposed by the packaging.

[...]

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>