Hi Bruno,
About 18 months ago we tried pushing this exact combination to the limits as
an experiment.
Solar Powered RG1000 at each end running TC point to point, 5m LMR400 cable,
25db parabolic grid, no amplifier.
We achieved 74Km BUT there is a few catches as stated below:-
The test was across the ocean (perfect fresnel clearence) , from 1100ft
hilltop to 1300ft hilltop. (Havelock North to Wairoa, New Zealand)
Fade margin was very average but it worked through a storm !
Throughput was in the region of 140KB/sec (HTTP) - FTP is lower.
RG's in 'Turbocell - no basestations' mode seem to have better throughput
than any other setting for PtP. The other option is 'Turbocell - Master' and
'Turbocell - Slave' but for some reason its slower... due to polling maybe ?
I wouldnt roll this combination out to a commercial customer due to the low
fade margin, but if you can get away with a small amp/bigger antenna at each
end it would work wonders.
The slightest bit of interference would otherwise knock it over.
My advice would also be to set both antennas up as horizontal polarity as
this would give you 20db rejection from many other signals in the 2.4 band
that are 'normally' polarised vertically.
-Sam
----- Original Message -----
From: "Bruno Lopes F. Cabral" <bruno@openline.com.br>
To: <Karlnet@WISPNotes.com>
Sent: Wednesday, May 14, 2003 5:27 AM
Subject: [Karlnet] TC on 70km link?
> Hi there
>
> this is my first post to this list, so please be gentle
>
> I dig all the list archives and didn't find any thread on this
> subject. My question is how TC would behave on a 70km
> link, because 802.11b has timing issues that limit it
> to about 35km (*)
>
> Does any of you have a working long link like this one?
>
> My setup: two 1.5 meters solid parabolic 31dBi antennaes,
> 105m high on one side, 660m on the other, about 70km
> between the points. each orinoco card is connected to
> a pigtail and directly to each antennae cable (no cable
> long runs to loose signal)
>
> two 802.11b machines (running linux in ad-hoc mode
> with orinoco cards) gave a poor throughput so I borrowed
> two WP-II (PtP) and they gave a lot better throughput but
> not perfect, yet, so my wonders in the use of Karlnet's TC.
>
> I plan to test this using two RG1x00 with Satelite firmwares
> in PtP mode. I'm aware that the RG1x00 will have it's
> bottleneck at 3Mb max (because of lack of processor power)
> but I need to convince ppl here that it works before spend
> several hundred dollars on AP1000 or WPBASE+ with
> 2 cards (RG1x00 is cheap and easily found here in Brazil)
>
> Any thoughts on this would be highly apreciated
>
> Cheers
> !3runo
>
> (*) 802.11b expects that each sent packet receive its
> "ack" within a certain time that, calculated based on
> light speed, translates to about 35km. It's the "DTIME".
> With a link longer than that, the transmitter will start
> retransmitting packets which could collide with the
> previous transmitted acks and so on until the link becames
> nearly unusable.
>
> _______________________________________________
> Karlnet mailing list
> Karlnet@WISPNotes.com
> http://lists.wispnotes.com/mailman/listinfo/karlnet
>
|