[TriLUG] 2.4 Ghz signal distance issue - your thoughts?

Mack.Joseph at epamail.epa.gov Mack.Joseph at epamail.epa.gov
Sun Jun 19 16:06:52 EDT 2005


Joseph Mack PhD, High Performance Computing & Scientific Visualisation
LMIT, Supporting the EPA Research Triangle Park, NC 919-541-0007 Federal
Infrastructure Contact-Ravi Nair 919-541-5467 - nair.ravi at epa.gov,
Federal Visualization  Contact - Joe Retzer, Ph.D. 919-541-4190 -
retzer.joseph at epa.gov

trilug-bounces at trilug.org wrote on 06/12/2005 06:47:45 PM:

> All:
>
> I have a network on the Carolina Outer Banks.  I have one wireless
> link connecting two parts of the network toghether.  The distance
> between the two access points is roughly 300 yards.

this is more than the standard antennas will handle, but
should be well within range with the two yagis being
line-of-sight.

> Both OpenWRT boxes are connected to 14 Dbi Yagi antennas (antenex
> units).  The problem is the signal fades out, and on a regular basis
> which isolates the Cisco 1200 from the main network (and
> thus isolates
> the users who complain about this, loudly.   I have the
> antennas moved
> up about 15 feet into the air and they are pointed directly at each
> other.  They are both on the same channel, 8, and there are ZERO
> obstructions between the two antenanns.  I can put both into AP mode
> and I seem to get a strong enough signal from both so thatthey reach
> each other, but clearly something is amiss.
>
> I am thinking of swapping out the Yagis for 14 or 24 DBI
> omnidirectional antenans.  My hope is that the more focused signal
> will carry a greater distance thus providing a reliable link between
> the components of the network.

sounds logical, but I what you have should work too, so something
else is wrong, which may carry over to this new setup too.


> At best, when the current system is
> running, I can only link at 1 meg with much less throughput (for the
> RF overhead).  Ideally I'd like to connect at 5.5, and 11 meg would
> put me into a nirvana state.
>
> Has anyone deployed the same kind (sort-of) long-haul 802.11 link?
> What kind of antenans did you use?  Does my next step sound like a
> good idea of throwing good money after bad?

I setup in a field each year

http://www.austintek.com/wifi_in_the_wilderness/

and found I got less range with WRT54Gs than with WAP11s.
WAPs are 100yrds apart (the limit of ethernet). Users
get a minimal (presumably 1Mbps) connection at about 100yrd
from the WAP. 300 yrds would need 9x signal
and to go to 11Mbps rather than 1Mbps you'd need 11x signal,
so you'd need approx 100times the signal from a normal
WAP. You have 2x14=28dbi of antennas and the rubber duckies
are quoted at 2.5dbi (I think), so you are 28-5=23db ahead
and you need 20db. So you should be OK.

Without test gear you have to try reasonable but possibly
inconclusive tests.

o how far apart do the antennas have to be to get reliable
11 Mbps connection? 100yrds, 10yrds?

o do you have 2.4GHz grade coax and is it short enough
(coax has losses of upto a db a foot at this freq)
or as short as you can make it (a couple of feet at most).
It should be at least 1/2" thick and quite stiff
(difficult to flex).

o are the antennas behaving? At short range (20-50yrd)
over flat ground, use a laptop as a signal strength meter
and rotate the yagi so it is pointing directly at the laptop
(should be max signal), then 90 deg away (should be unmeasurably
low, at least 20db down), then 180 deg (ie backwards) should
be about 10db down, then 270 (should be immeasurable again).

o you would have to be unlucky for this to happen, but
ground is a little reflective. In this case the receiving
antenna is seeing the transmitting antenna and it's reflection
(an antenna radiating a weaker signal of opposite polarity)
radiating from underground. This produces a diffraction
pattern in the direction of the receiver varying in amplitude
in the elevation direction. You should lower/raise
the height of one of the antennas to make your you're at the max
(and not at a null). You only have to move the antenna 1/2
wavelength up and down (ie about 6") to move to the next lobe.
By reciprocity, when you've got one end at the right height,
the other end is right too.

Joe NA3T






More information about the TriLUG mailing list