wifi antenna in the mountains
getwilde
I'm only finding parts of the answer. Here's the deal.
I live in a beautiful canyon 4 miles from the valley/city. A peak 600
feet from my house has line of sight to the city where there are WISP
towers. Likewise, I have line of site from my front door to the peak
600 feet away. I can build a biquad antenna and put it on that point
and it receives a signal (tested with my laptop), but I'm stuck after
that.
Obstacle 1. There's no electrical power plug on the peak.
a. I'm not comfortable running power/conduit that distance (and I'm
sure there are ordanances against that)
b. The distance is too far for Power Over Ethernet
Obstacle 2. How to get the signal from the bi-quad to my home:
a. The distance is too far for Cat5 cable, and there's no way to power
any in-line repeaters
b. I could run fiber cable, but that requires power at the antenna
c. I could run old ThickNet 10-base-5 cable, but that requires power at
the antenna
So, a couple exotic ideas I have:
Is it possible to build a bi-quad that captures the signal without
power or only with simple batteries, then use a cantenna at my house
pointed to the bi-quad?
Can I build some sort of reflector on the point such as a mirror or
metal sheet, then locate the bi-quad at my house and point it at the
reflector?
Are there other viable options? Looking for some creative ideas here.
Thanks in advance
Me
seems that power is your problem here. I am interested inseeing what people
come up with.
K
"getwilde" <getw...@gmail.com> wrote in message
news:1132279860.1...@g43g2000cwa.googlegroups.com...
Paul
"Me" <m...@place.com> schreef in bericht
news:qabff.3606$w84.6...@news20.bellglobal.com...
> Why not just get satellite internet? Is this a hobby project? Anyways,
> it seems that power is your problem here. I am interested inseeing what
> people come up with.
>
> K
Paul
youcantoo
Rôgêr
Exactly what I was thinking. getwilde, just take two decent antennas,
grid parabolics would be nice and not necessarily all that expensive,
and hook them up with a bit of coax running from one antenna to the
other. Aim one at your house, one at the WISP. No electronics involved
so no need of elecjuicity.
Derek Broughton
> Why not just get satellite internet? Is this a hobby project? Anyways,
> it
> seems that power is your problem here. I am interested inseeing what
> people come up with.
Ack! As a satellite internet user, I'd jump through more hoops than the OP
is facing to get service from a WISP!
> "getwilde" <getw...@gmail.com> wrote in message
> news:1132279860.1...@g43g2000cwa.googlegroups.com...
>>
>> I live in a beautiful canyon 4 miles from the valley/city. A peak 600
>> feet from my house has line of sight to the city where there are WISP
>> towers. Likewise, I have line of site from my front door to the peak
>> 600 feet away. I can build a biquad antenna and put it on that point
>> and it receives a signal (tested with my laptop), but I'm stuck after
>> that.
>>
>> Obstacle 1. There's no electrical power plug on the peak.
OK, that's an obstacle. You could probably solve it with a solar panel and
a battery - but theft could be a problem. It would still be my choice.
>> Obstacle 2. How to get the signal from the bi-quad to my home:
>> a. The distance is too far for Cat5 cable
No, it isn't. I know the standard specifies 100m, but Jeff will tell you he
runs it 1000'
--
derek
ReginaldP
high spot and a good gain antenna at the home , some simple testing with
the laptop will confirm this suggestion but do remember the passive
needs specific high gain and careful installation , good luck
do...@xrexxwifix.usenet.us.com
> 600 feet away. I can build a biquad antenna and put it on that point
> and it receives a signal (tested with my laptop), but I'm stuck after
> that.
> Obstacle 1. There's no electrical power plug on the peak.
You could do something solar powered on the peak.
Cringely has a writeup about using passive repeaters in the same situation,
with no power required.
http://www.pbs.org/cringely/pulpit/pulpit20010628.html
http://www.pbs.org/cringely/pulpit/pulpit20010712.html
Some people question that he actually accomplished this.
--
---
Clarence A Dold - Hidden Valley (Lake County) CA USA 38.8,-122.5
Jeff Liebermann
<Re...@Hanky-Bannister.co.uk> wrote:
>You should investigate placing a decent gain passive radiator on the
>high spot and a good gain antenna at the home , some simple testing with
>the laptop will confirm this suggestion but do remember the passive
>needs specific high gain and careful installation , good luck
Do you have any idea how large the passive reflector will need to be
at 600ft in order to make this useable? At 2.4GHz, the highest gain
antenna is about 24dBi with a 7 degree beamwidth. At 600ft, that beam
is 74ft wide. With a 45 degree tilt angle, that's a 103 x 103 ft
diameter reflector. Anything less will result in corresponding loss
of signal. Half the size = 1/4th the area = -12dB loss.
I can work the numbers in detail if you want, but such arrangements
never work at 2.4Ghz. See references under "periscope antennas".
--
Jeff Liebermann je...@comix.santa-cruz.ca.us
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
getwilde
between antennas (but I'm still trying to learn all of the principles
involved). If I were to build a passive setup like Roger explains, do
I decrease range by not having power? The 4 or 5 miles to the Acess
Point worries me a bit.
So, am I thinking about this right...? To increase range: Is adding an
amplifier at my house the best way to increase the outgoing signal,
even though it will be passing through the passive antenna? Likewise,
if the grid parabolic pointed at the WISP is larger, is that the best
way to increase incoming reception? (Assuming I build the whole setup
well)
Incidentally, yes, solar and a car battery are options, but a huge
hassle if I can avoid them (and I would run the risk of having them
stolen at some point). But if I were told that power at the middle
antenna is the only way to insure good range, then I'd definitely
explore that further. I'd even be willing to march up there every few
weeks to replace batteries (skip the solar power thing altogether), if
power requirements were low enough to make a small battery rack an
option.
getwilde
and lower costs. Not having a phone line might also be an issue... but
I'll only explore that further if I can't get this wi-fi approach
working.
Jeff Liebermann
>I live in a beautiful canyon 4 miles from the valley/city. A peak 600
>feet from my house has line of sight to the city where there are WISP
>towers. Likewise, I have line of site from my front door to the peak
>600 feet away. I can build a biquad antenna and put it on that point
>and it receives a signal (tested with my laptop), but I'm stuck after
>that.
This is good news. This should be no problem.
>Obstacle 1. There's no electrical power plug on the peak.
>a. I'm not comfortable running power/conduit that distance (and I'm
>sure there are ordanances against that)
>b. The distance is too far for Power Over Ethernet
Wrong on both counts. I've run all kinds of power through the forest.
As long as you don't cross any major boundaries (roads, property
lines, etc), you're fine. Low voltage power is not considered a fire
hazzard. Xmas tree lamps draw more power and get hotter than a
typical access point.
In addition, you can run up to about 1000ft of CAT5 at 10baseT-HDX
speeds. I've scribbled several postings in this newsgroup on the
topic. You can also use coaxial cable for about the same distances
using RG-6/U and 10base-2 using media converters. Of course, you can
also use fiber, but that's a bit more expensive and overkill.
http://groups.google.com/group/alt.internet.wireless/msg/f1e3a5e2b13d0793
Let's grind the numbers for running power up the hill. I'll use 700ft
because there will probably be some snaking of the wire. I'll assume
you'll use an ethernet bridge radio or one that supports client mode.
I suggest a Linksys WRT54G (NOT v5) with DD-WRT firmware. The nice
part of this device is that it has a very wide range switching
regulator which will run the box on anything between about 4 and
18VDC. This is critical for delivering power. I forgot how much
power it draws but I'll guess about 8 watts max.
Start with some higher voltage at the house, run 600ft of wire, and
want at least 6VDC at the access point, at a current drain of about
1.3 amps (at 6VDC). So, how much voltage does it require?
Well, you can almost do it with CAT5 cable, which has a resistance of
3 ohms per 100ft per conductor. With 2 conductors in parallel, that's
3 ohms/100ft loop resistance. 600ft would have 18 ohms loop
resistance. The approximate current drain is:
8 watts / 6 volts = 1.33Amps
Therefore, the voltage drop across the cable is:
1.33A * 18 ohms = 24VDC.
So, if you start with a 24VDC power supply, and take the loss in the
cable, you get 6VDC at the WRT54G.
Unfortunately, the power dissipated in the cable is too high:
1.33A * (24 - 6V) = 24 watts
It won't melt the cable but it will get quite warm. Bad idea.
So, we use heavier cable. I suggest outdoor low voltage UV proof
rubberized cable used for garden flood lights and outdoor low voltage
wiring. It's typically about #14AWG which is about 0.25 ohms/100ft
per conductor.
Grinding the same 600ft length at 1.33A, I get:
1.33A * 3 ohms = 4 volts
So, we start with 10VDC, lose 4 volts in the cable, and end up with 6
volts at the WRT54G. Power dissipation is:
1.33A * (10 - 4v) = 8 watts
which is still a bit high, but workable.
In reality, starting with 12VDC makes more sense. The current drain
is somewhat less than 1.33A. 8 watts over 600ft is a bit too high so
I would use some heavier wire if available. The important point is
that it will work by running a power cable and either CAT5 or RG-6/u
if the radio has a wide range of acceptable power supply voltages.
I can answer the RF questions below if running power is unacceptable.
>Obstacle 2. How to get the signal from the bi-quad to my home:
>a. The distance is too far for Cat5 cable, and there's no way to power
>any in-line repeaters
>b. I could run fiber cable, but that requires power at the antenna
>c. I could run old ThickNet 10-base-5 cable, but that requires power at
>the antenna
>
>So, a couple exotic ideas I have:
>Is it possible to build a bi-quad that captures the signal without
>power or only with simple batteries, then use a cantenna at my house
>pointed to the bi-quad?
>Can I build some sort of reflector on the point such as a mirror or
>metal sheet, then locate the bi-quad at my house and point it at the
>reflector?
>
>Are there other viable options? Looking for some creative ideas here.
>
>Thanks in advance
getwilde
detailed response. It's obvious that you're very knowledgable in this
area.
Regarding passive antennas: A bit more research shows that they're
commonly referred to as "Passive Reflectors". Google reveals all sorts
of information about them. With guidance from XReXXwifiX, I've located
Cringely's claim to success
(http://www.pbs.org/cringely/pulpit/pulpit20020207.html). However,
Jeff, your response about the size area of the signal after travelling
600 feet makes a lot of sense, and really brings into question
Cringely's credibility. Unfortunately. :(
Jeff, thank you for crunching those numbers regarding power cables.
Sounds like a more viable option than I originally thought it to be.
Maybe this is the route to go, combined with running the signal across
cat-5 at half-duplex and lower speed (10baseT-HDX). (I may still be
missing something??? -- Still need to give this more though and
research.)
However, Jeff, you have me extremely curious now with your comment "I
can answer the RF questions below if running power is unacceptable."
If you would be willing to explain this, I would really appreciate it!
It probably opens up options I've erroneously ruled out or don't even
know exist.
Thanks in advance.
getwilde
getwilde wrote:
> Regarding passive antennas: A bit more research shows that they're
> commonly referred to as "Passive Reflectors". Google reveals all sorts
> of information about them.
Sorry, I meant to type "Passive Repeaters".
do...@xrexxwifix.usenet.us.com
> Thanks to everyone for the great ideas. Jeff, thank you for your
> detailed response. It's obvious that you're very knowledgable in this
> area.
Everybody thinks so at first.
> (http://www.pbs.org/cringely/pulpit/pulpit20020207.html). However,
> Jeff, your response about the size area of the signal after travelling
> 600 feet makes a lot of sense, and really brings into question
> Cringely's credibility. Unfortunately. :(
It does seem a little awkward and non-reproducible. But these are
antennas. Jeff was talking about how large a flat passive reflector would
need to be. That's a different thing.
> Jeff, thank you for crunching those numbers regarding power cables.
> Sounds like a more viable option than I originally thought it to be.
I missed the fact that it is only 600 feet. I was thinking of my own
environment, where there's a handy peak, but it's about 1000 feet as the
crow flies, across a road, and on someone else's property.
Would it still be on your property?
getwilde
> I missed the fact that it is only 600 feet. I was thinking of my own
> environment, where there's a handy peak, but it's about 1000 feet as the
> crow flies, across a road, and on someone else's property.
>
> Would it still be on your property?
The peak is not my personal property, but part of the community share
that I pay into annually. The land is unused -- just a bunch of scrub
oak, and I'm quite certain the board will grant me permission to place
an antenna (or small-ish reflector) up there if it's not visible from
the road. I'm less certain that I can convince them that a low-voltage
power wire is safe to run up there, but I'm definitely ready to discuss
it with them if it turns out to be the best approach!
d...@XReXXwifiX.usenet.us.com wrote:
> It does seem a little awkward and non-reproducible. But these are
> antennas. Jeff was talking about how large a flat passive reflector would
> need to be. That's a different thing.
Ahhh. Didn't notice the word "reflector". I figured we were talking
about 103 ft diameter dish or something! I've found comments that
"corrugated metal sheets" (with their sine wave-shaped crosscut) work
well as reflectors. Makes me wonder if I could somehow reduce the 600
ft distance (by running power and an antenna even further out on my own
property), then mount a smaller sheet. Also makes me wonder if the
radio/cell towers located 1/3 mile away from the house have passive
reflectors already mounted that I could use. (Not likely based on the
size issue Jeff brought up. There's also the question of legality.)
do...@xrexxwifix.usenet.us.com
> property), then mount a smaller sheet. Also makes me wonder if the
> radio/cell towers located 1/3 mile away from the house have passive
> reflectors already mounted that I could use. (Not likely based on the
> size issue Jeff brought up. There's also the question of legality.)
Lots of problems with "sheets". They are ugly, easily spotted, and catch
a lot of wind, so they need to be well mounted. Yagi's present themselves
as ideal bird pirtches, a different loading problem.
I think the large telco flat reflectors are, or look like, buildings.
Pointing a directional antenna around randomly to see what signal you get
is what some people say made Cringley's setup work, that he wasn't even
pointing at his own repeater, that he was actually picking up some
naturally echoed signal from a canyon wall in a slightly different
direction.
Jeff Liebermann
>Regarding passive antennas: A bit more research shows that they're
>commonly referred to as "Passive Reflectors". Google reveals all sorts
>of information about them. With guidance from XReXXwifiX, I've located
>Cringely's claim to success
>(http://www.pbs.org/cringely/pulpit/pulpit20020207.html).
Yech. I finally got around to reading the above article. Cringly
used two yagi antennas, back to back. That's very different from a
"passive repeater", "passive reflector", or "periscope antenna". The
math is simple enough to prove that such a thing will never work.
Calcs follow at bottom.
>However,
>Jeff, your response about the size area of the signal after travelling
>600 feet makes a lot of sense, and really brings into question
>Cringely's credibility. Unfortunately. :(
I'm from an engineering background. I practice the cult of
calculation before construction. When Cringly supplies reproducible
numbers and measurements, I'll believe his claims. However, I am
always impressed with his ability to support marginal technology,
marginal construction techniques, and dubious practices. As a former
hacker, I find his writings inspirational. Well, maybe not quite
former hacker.
>Jeff, thank you for crunching those numbers regarding power cables.
>Sounds like a more viable option than I originally thought it to be.
Reminder: The real trick is that the recommended access point or
router can handle a wide range of voltages. It can be done with other
routers, but the WRT54G, WAP54G, and such are quite ideal.
Incidentally, I forgot to mumble that before I figured it out, I was
using DC-to-DC converters to supply regulated 5VDC and sometimes 12VDC
down extended power lines, usually up a tower or on an inaccessible
rooftop. These worked just fine. One of them is running 3ea WAP11
access points on a rooftop. However, with the WRT54G, it's not
necessary to use the DC to DC converter. Also, I had a solar powered
access point in a tree, but someone shot a hole in it, so it was
removed.
>Maybe this is the route to go, combined with running the signal across
>cat-5 at half-duplex and lower speed (10baseT-HDX). (I may still be
>missing something??? -- Still need to give this more though and
>research.)
Forcing the access point or router to 10baseT-HDX (half duplex) is the
major problem. The easiest way is to use an old 10baseT hub in the
circuit, or to use a media converter. Most managed switches will let
you set the interface speed and protocol. Now that I think of it, I
suspect (not sure) that the WRT54G with DD-WRT firmware will allow you
to force the speed on any port. I'll look (when I got time).
>However, Jeff, you have me extremely curious now with your comment "I
>can answer the RF questions below if running power is unacceptable."
That's because the [deleted expletive] telephone would not stop
ringing this morning. I just hate it when paying customers interrupt
my screwing around. Actually, I was just being lazy/bizzee, as I
didn't want to get into the RF aspects if you were just going to do
the long power extension cord method.
>If you would be willing to explain this, I would really appreciate it!
>It probably opens up options I've erroneously ruled out or don't even
>know exist.
Ok, but later.
Why Cringly's back to back yagi's wont work.
When you put a passive repeater (back to back yagi's or dish's) in the
path, you essentially split the path loss equation in two parts. Let's
try a real example using your 600ft numbers and situation.
First the assumptions and guesswork. I'll assume that the remote
access point is 1 mile away and has a typical +15dBm access point, 4
dB coax loss, and 8dBi outside omni antenna. The two yagis should
have about +12dBi gain each. Your wireless client radio is also a
typical +15dbm xmitter with yet a 3rd +12dBi yagi. I'll do the
numbers at 5.5Mbits/sec.
Pluging the quesswork into:
http://www.terabeam.com/support/calculations/som.php
Path between remote hot spot and the first yagi antenna.
TX power = +15dBm
TX coax loss = 4dB (3ft LMR-240 plus a mess of connectors)
TX ant gain = 8dB (omni)
Distance = 1 mile
RX ant gain = +12dBi (yagi)
RX coax loss = 1dB (mostly connectors)
RX signal = Unknown
Grinding the numbers, the RX signal at the yagi's antenna connector
will be -74dBm.
Now, we take that -74dBm signal from the first yagi and plug it into
the 2nd yagi antenna which is pointed at your client radio. Grinding
the numbers:
TX power = -74dBm (from first yagi)
TX coax loss = 0dB (connector loss take care of in the first yagi)
TX ant gain = 12dB (yagi)
Distance = 0.114 miles (600ft)
RX ant gain = +12dBi (yagi)
RX coax loss = 4dB (3ft LMR-240 plus connectos)
RX signal = Unknown
Grinding the numbers, the RX signal at the client radio receiver will
be -139dBm. That *WAY* below the required -85dBm (rx sensitivity)
plus another 20dB of fade margin. In other words, the path is 74dB
short of working. That's a HUGE amount which can never be (legally)
fixed by juggling antenna gains and tx power levels. If I increase
all 3 yagi antennas to +24dBi dish antenns, for an additional 36dB
system gain, we're still 38dB short of being useable. This isn't
gonna work.
I just notice that Cringly said he used a 21dBi Cushcraft dish and
cranked up his tx power to +20dBm, but that will help little with a
74dB path gain shortfall. I don't know what Cringly did to make his
work, but my calcs show that it can't work.
getwilde
fully, but I get the gist.
If/when you have a chance to tell me about other RF options, please do.
In the meantime, this weekend, I'm gonna' go ahead and build a
parabolic antenna and see what signals I can find (if any) bouncing
down that canyon.
Best regards.
Jeff Liebermann
do...@XReXXwifiX.usenet.us.com wrote:
>Lots of problems with "sheets". They are ugly, easily spotted, and catch
>a lot of wind, so they need to be well mounted. Yagi's present themselves
>as ideal bird pirtches, a different loading problem.
Corregated sheets don't work as reflectors. The signal bounces all
over the place, with little going in the desired direction. It might
work if the corregations were less than about 1/10th of a wavelength
apart for 2.4Ghz, but that's 1.2cm, which is impractical.
>I think the large telco flat reflectors are, or look like, buildings.
Billboards. I'm talking about a periscope antenna. See:
| http://www.w1ghz.org/antbook/chap8.pdf
Paul Wade calls them "flyswatter" reflectors. Most people call them
eyesores.
| http://www.valmont.com/asp/communication/communication6.shtml
Ah, found the catalog:
| http://www.valmont.com/asp/communication/specialty_structures/pdf/Catalog161A.pdf
Ugh. 7MBytes and 118 pages. Lots of nice pictures.
Notice the overkill rigid mounting structure. There's a good reason
for to be so beefy. Visualize how far off the beam would be if the
structure rotated even a few fractions of a degree. Can you say
critical?
Incidentally, the FCC refuses to licence point to point commercial
microwave links using periscope antennas because to chronic alignment
errors causing interference to other users on the same frequency.
This looks kinda cool. (I haven't tried it yet).
| http://www.rfcafe.com/business/software/link_planning_tool/link_planning_tool_excel.htm
It will do passive repeaters.
do...@xrexxwifix.usenet.us.com
Oops, posted the wrong Cringely link earlier.
That is interesting, though. If he got his link via a visible neighbor
working in July of 2001, why is he hanging Yagi's in trees in 2002?
I just wandered back through the Cringely Archives.
He ought to be required to present a scorecard of previous successes and
failures after a year.
What about Fusion Lighting killing 802.11b? The web site is gone.
do...@xrexxwifix.usenet.us.com
> do...@XReXXwifiX.usenet.us.com wrote:
> >I think the large telco flat reflectors are, or look like, buildings.
> Billboards. I'm talking about a periscope antenna. See:
> | http://www.w1ghz.org/antbook/chap8.pdf
I think we have those in Middletown, CA, on the 707-987 CO.
I was thinking of big, flat, oddly positioned buildings on the side of
hills in the middle of nowhere. I assumed that these were telco flat
reflectors, but I can't remember why, now ;-)
Jeff Liebermann
>If/when you have a chance to tell me about other RF options, please do.
> In the meantime, this weekend, I'm gonna' go ahead and build a
>parabolic antenna and see what signals I can find (if any) bouncing
>down that canyon.
Ok, here's a possible option for you. Run RG-6/u coax up the hill and
install a dish antenna pointed at wherever. 4 miles is a long shot,
but it might work. Satellite grade RG-6/u coax is about 11dB/100ft.
Therefore, your attenuation will be 66dB. Let's grind the numbers
again.
http://www.terabeam.com/support/calculations/som.php
Starting at the access point end:
TX power = +15dBm
TX coax loss = 4dB (3ft LMR-240 plus a mess of connectors)
TX ant gain = 8dB (omni)
Distance = 4 miles
RX ant gain = +24dBi (dish antenna)
RX coax loss = 66dB (600ft of RG-6/u)
RX signal = Unknown
Yields a receive signal of -139dBm. The desired receive signal level
is 20dB of fade margin above the -85dBm receive level needed for
5.5Mbits/sec. No way is that going to work.
So, let's see what happens if you run better coax up the hill. LMR-400
is 6.7dB/100ft loss or 40dB loss. That yields -113dBm rx signal level
which also won't work.
I really shouldn't post ideas that won't work, but I thought the
exercise might be of interest. Basically, the radio needs to be on
top of the hill.
Much more interesting is what equipment needs to be installed on the
hill in order to connect to an access point that's 4 miles away. If
you install the big +24dBi dish with an attached radio, the link
budget looks like:
Starting at the access point end:
TX power = +15dBm
TX coax loss = 4dB (3ft LMR-240 plus some connectors)
TX ant gain = 8dB (omni)
Distance = 4 miles
RX ant gain = +24dBi (dish antenna)
RX coax loss = 4dB (3ft LMR-240 plus some connectors)
RX signal = -85dBm (at 5.5Mbits/sec)
That yields a fade margin of 7.8dB which isn't going to work. You
need 20dB of fade margin for a reliable connection. That's about a
12dB shortfall, which is not going to be made up with a bigger antenna
or less cable loss. Your only hope of this thing working is if the
service provider has a better antenna system than my insipid 8dBi omni
guess. It might be a higher gain sector antenna which has a better
chance.
You stated that you could barely make a connection with a biquad. Well
a typical biquad is about 9dBi gain less coax loss. If you replace it
with a 24dBi dish antenna, you'll pickup 15dB of gain, which should
make the link useable. It really depends on the antenna configuration
of the central access point. Can you supply any details of their
antenna and radio system.
Jeff Liebermann
do...@XReXXwifiX.usenet.us.com wrote:
>What about Fusion Lighting killing 802.11b? The web site is gone.
Fusion Lighting's RF lamps have been off the market since 1999.
http://195.178.164.205/IAEEL/iaeel/newsl/1998/tva1998/LiMa_a_2_98.html
ReginaldP
> On Sat, 19 Nov 2005 00:41:07 +1000, ReginaldP
> <Re...@Hanky-Bannister.co.uk> wrote:
>
>
>>You should investigate placing a decent gain passive radiator on the
>>high spot and a good gain antenna at the home , some simple testing with
>>the laptop will confirm this suggestion but do remember the passive
>>needs specific high gain and careful installation , good luck
>
>
> Do you have any idea how large the passive reflector will need to be
> at 600ft in order to make this useable? At 2.4GHz, the highest gain
> antenna is about 24dBi with a 7 degree beamwidth. At 600ft, that beam
> is 74ft wide. With a 45 degree tilt angle, that's a 103 x 103 ft
> diameter reflector. Anything less will result in corresponding loss
> of signal. Half the size = 1/4th the area = -12dB loss.
>
> I can work the numbers in detail if you want, but such arrangements
> never work at 2.4Ghz. See references under "periscope antennas".
>
which should do the job , I have built something very similar with
success and it is certainly worth looking at , do the figures if you want .
ReginaldP
Rethinking it I'm wondering is running some 2.4 mm cable twin up there
and feeding 24ac to a ap at the top is feasible , I guess we wont really
know unless some one does a walk over and posts some measurements of rx
sigs at the hill top .
Jeff Liebermann
<Re...@Hanky-Bannister.co.uk> wrote:
>Rethinking it I'm wondering is running some 2.4 mm cable twin up there
>and feeding 24ac to a ap at the top is feasible , I guess we wont really
>know unless some one does a walk over and posts some measurements of rx
>sigs at the hill top .
It might be a problem because the distance to the access point is 4
miles (6.44km) distant and we have no clue as to what type of antenna
system is at the access point end. I ran the numbers in:
| http://groups.google.com/group/alt.internet.wireless/msg/747ce1170e173e8b?fwc=1
near the bottom starting with "Much more interesting is..."
Just ignore my dumb idea of running 600ft of coax cable.
My guess(tm) is that unless the access point has a directional
(sector) antenna system or much higher gain omni, a 4 mile link may be
functional, but not very reliable. However, the fact that it almost
works with a biquad antenna leads me to suspect that a bigger antenna
may do the trick.
--
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
je...@comix.santa-cruz.ca.us je...@cruzio.com
ReginaldP
> On Sat, 19 Nov 2005 13:55:18 +1000, ReginaldP
> <Re...@Hanky-Bannister.co.uk> wrote:
>
>
>>Rethinking it I'm wondering is running some 2.4 mm cable twin up there
>>and feeding 24ac to a ap at the top is feasible , I guess we wont really
>>know unless some one does a walk over and posts some measurements of rx
>>sigs at the hill top .
>
>
> It might be a problem because the distance to the access point is 4
> miles (6.44km) distant and we have no clue as to what type of antenna
> system is at the access point end. I ran the numbers in:
> | http://groups.google.com/group/alt.internet.wireless/msg/747ce1170e173e8b?fwc=1
> near the bottom starting with "Much more interesting is..."
> Just ignore my dumb idea of running 600ft of coax cable.
>
> My guess(tm) is that unless the access point has a directional
> (sector) antenna system or much higher gain omni, a 4 mile link may be
> functional, but not very reliable. However, the fact that it almost
> works with a biquad antenna leads me to suspect that a bigger antenna
> may do the trick.
>
Exactly my point , of course you made some guesses but if the gent takes
a walk and makes some measurements with what he has ...
Jonathan
I was just reading this thread, it's pretty interesting! I'm curious
how you end up solving your problem.
Anyway, a group call SoCalFreenet, down in San Diego, is *really*
creative when it comes to building a network of relay points for their
backhaul network. Poke around their site a bit and see what they're
building. One project to peek at is the Low-budget Outdoor Solar
Wireless (http://socalfreenet.org/lowbudgetsolar). They had good photos
of one of their solar-powered backhaul nodes previously, but I couldn't
find it just now. Very cool stuff, but you have to be handy enough with
PCs, Linux, antennas and the like to put it all together.
good luck!!
-Jonathan
Jeff Liebermann
wrote:
>Anyway, a group call SoCalFreenet, down in San Diego, is *really*
>creative when it comes to building a network of relay points for their
>backhaul network. Poke around their site a bit and see what they're
>building. One project to peek at is the Low-budget Outdoor Solar
>Wireless (http://socalfreenet.org/lowbudgetsolar). They had good photos
>of one of their solar-powered backhaul nodes previously, but I couldn't
>find it just now. Very cool stuff, but you have to be handy enough with
>PCs, Linux, antennas and the like to put it all together.
Photos. Found by punching "solar" into the searth thingus.
http://socalfreenet.org/node/286
http://socalfreenet.org/node/287
http://socalfreenet.org/solarnode
http://socalfreenet.org/image/tid/20