Using outdoor antenna with AP - should I leave take off in..

G

Guest

Guest
Archived from groups: alt.internet.wireless (More info?)

I've got a 3Com 7250 802.11g access point, which comes with two small
5" antennas that screw into the unit.

I also bought an 8 dBi omnidirectional outdoor antenna and 20'
low-loss cable so I could mount the antenna outdoors. This
antenna/cable replace one of the two original 5" antennas.

Should I leave the 2nd small antenna attached, or remove it? Why or
why not?

Thank you.


Also, while I'm here, one other question, about antenna gain and
range.

My understanding is that a 3 dBi increase indicates a doubling of
power, but effective power drops as the square of distance. To double
range you need 4x the power at the antenna, or a 6 dBi gain. Is this
correct?
 
G

Guest

Guest
Archived from groups: alt.internet.wireless (More info?)

On 15 Sep 2004 04:47:48 -0700, monkeyomen@nym.hush.com (MonkeyOmen)
wrote:

>I've got a 3Com 7250 802.11g access point, which comes with two small
>5" antennas that screw into the unit.
>
>I also bought an 8 dBi omnidirectional outdoor antenna and 20'
>low-loss cable so I could mount the antenna outdoors. This
>antenna/cable replace one of the two original 5" antennas.
>
>Should I leave the 2nd small antenna attached, or remove it? Why or
>why not?

If you have no need for indoor coverage, then remove the 2nd antenna.
If you need coverage in an area best covered by the 2nd antenna, leave
it connected.
Most current access point software has a setting to enable or disable
one antenna.
A few access points are really weird with how they handle the
diversity antennas. Some will receive on both antennas but only
transmit on one. If you start getting weird results, switch to the
other antenna port.

>Also, while I'm here, one other question, about antenna gain and
>range.
>
>My understanding is that a 3 dBi increase indicates a doubling of
>power, but effective power drops as the square of distance. To double
>range you need 4x the power at the antenna, or a 6 dBi gain. Is this
>correct?

Correct. 3dB increase in radiated power will double the coverage area
(area of the circle in square feet), but only increase the range
(radius) 1.4 times. 6dB increase in radiate power will give 4x the
coverage area and double the range. Think:
Area=Pi*Radius^2

Drivel: Note that for every increase in antenna gain in one
direction, you take away signal from some other direction. Antennas
to not manufacture additional RF. Only transmitters do that. The
"taken away" part can be in the up and down direction, which is
usually useless, so there's an obvious benifit. However, if you
happen to be a user in these areas, you're got a problem. This is
what happens when someone puts a very high gain omni antenna on the
roof, and finds that they can no longer connect from directly below
the antenna. All the signal is going towards the horizon.


--
Jeff Liebermann jeffl@comix.santa-cruz.ca.us
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 AE6KS 831-336-2558
 
G

Guest

Guest
Archived from groups: alt.internet.wireless (More info?)

Jeff Liebermann <jeffl@comix.santa-cruz.ca.us> wrote in message news:<9qngk0pi04gafmvcpgmgm4i7aosi8elak7@4ax.com>...
> [snip response]

Thanks for the response, and for devoting so much of your time to
answering questions on this newsgroup. Your posts are very
informative.
 
G

Guest

Guest
Archived from groups: alt.internet.wireless (More info?)

On 15 Sep 2004 19:42:28 -0700, monkeyomen@nym.hush.com (MonkeyOmen)
wrote:

>Jeff Liebermann <jeffl@comix.santa-cruz.ca.us> wrote in message news:<9qngk0pi04gafmvcpgmgm4i7aosi8elak7@4ax.com>...
>> [snip response]

>Thanks for the response, and for devoting so much of your time to
>answering questions on this newsgroup. Your posts are very
>informative.

Thanks. Sometimes, my answers are amazing accurate.

I managed to forget one obscure point. If you're using WDS (wireless
distribution something) to make your access point act as a store and
forward repeater, do NOT use both antennas. Disconnect one of them.
The problem is that the diversity antenna juggling algorithm
apparently works on the basis of the last successfully received
packet. With WDS, the packets can alternate between antennas as it
receives on one antenna and forwards on the other. The result is the
worlds worst performance because of the intentional delay between
switching antennas.

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831.336.2558 voice http://www.LearnByDestroying.com
# jeffl@comix.santa-cruz.ca.us
# 831.421.6491 digital_pager jeffl@cruzio.com AE6KS
 

Mike

Splendid
Apr 1, 2004
3,865
0
22,780
Archived from groups: alt.internet.wireless (More info?)

"Jeff Liebermann" <jeffl@comix.santa-cruz.ca.us> wrote in message
news:up1ik0p9ckhs34bigvllfsdst2671ch7a4@4ax.com...
> On 15 Sep 2004 19:42:28 -0700, monkeyomen@nym.hush.com (MonkeyOmen)
> wrote:
>
>>Jeff Liebermann <jeffl@comix.santa-cruz.ca.us> wrote in message
>>news:<9qngk0pi04gafmvcpgmgm4i7aosi8elak7@4ax.com>...
>>> [snip response]
>
>>Thanks for the response, and for devoting so much of your time to
>>answering questions on this newsgroup. Your posts are very
>>informative.
>
> Thanks. Sometimes, my answers are amazing accurate.
>
> I managed to forget one obscure point. If you're using WDS (wireless
> distribution something) to make your access point act as a store and
> forward repeater, do NOT use both antennas. Disconnect one of them.
> The problem is that the diversity antenna juggling algorithm
> apparently works on the basis of the last successfully received
> packet. With WDS, the packets can alternate between antennas as it
> receives on one antenna and forwards on the other. The result is the
> worlds worst performance because of the intentional delay between
> switching antennas.
>
> --
> # Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
> # 831.336.2558 voice http://www.LearnByDestroying.com
> # jeffl@comix.santa-cruz.ca.us
> # 831.421.6491 digital_pager jeffl@cruzio.com AE6KS

Hi Jeff,

Thanks for the great info! I have a Cisco AP350 with a T1 coming into my
house. I would like to provide access for my neighbors. What do you
recommend for an outdoor antenna? My house sits on a high spot, but the
problem is a lot of tall tree and thick woods between houses.

TIA,
Mike




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----
 
G

Guest

Guest
Archived from groups: alt.internet.wireless (More info?)

On Fri, 17 Sep 2004 09:22:49 -0400, "Mike" <mikebegin@hotmail.com>
wrote:

>I have a Cisco AP350 with a T1 coming into my
>house. I would like to provide access for my neighbors. What do you
>recommend for an outdoor antenna? My house sits on a high spot, but the
>problem is a lot of tall tree and thick woods between houses.

Dunno. The best I can do is offer some general clues and rhetorical
questions.

Mini rant: No numbers = no specific answers.

What are the distances involved?
How much difference in vertical elevation?
Is this going to be a neighborhood LAN (everyone joins in) or are you
going to do this point to point, one house at a time?
Is price an issue?
Are there wired alternatives?

Trees and woods are a serious problem. 2.4Ghz does not go through
anything with water inside. If you have a view of the *TOPS* of the
trees, you can possible mount the client radios (for the neighbors) in
the tops of the trees, and run PoE (power over ethernet) to the
treetop. The major problem with this is maintenance. Climbing a
100ft tree, in the middle of a storm, to remove an ant colony, was not
my idea of fun. Basically, you need to get line of sight somehow.
Perhaps this would be a good excuse to do some tree thinning or
logging.

I helped with a neighborhood LAN similar to yours, where the view from
the central hilltop was nothing but tree tops. Looking from above,
that's all you see. However, trees tend to not have branches at
ground level, so there was a few between the trunks at ground level.
The access points were moved to ground level and located so there was
some semblence of line of sight between the tree trunks. I had to use
multiple access points on the central hub to cover all the neighbors,
but it did work.

Another way of going through trees is to use 900MHz. We had quite a
bit of experience with Metricom in the area. 900MHz is blocked by
trees, but nowhere as bad as 2.4GHz. The problem is that 900MHz
equipment is difficult to find. The available speeds are also limited
to about 1.5-3.0Mbits/sec depending on manufacturer. The band is also
full of interference from other devices. The only currently
manufactured 900MHz hardware I know about is Motorola Canopy system.

One thing you shouldn't do is install a very high gain omnidirectional
antenna. The problem is that the vertical radiation angle is very
narrow. From even a modest hill, you will be sending the signal over
everyone's head. Think multiple access points, sectored antennas,
panels, dishes, or anything directional.

I live in an overgrown redwood/fir/oak forest. We have a crude 2.4GHz
neighborhood LAN that barely works through the trees. However, we
also have a buried coax and fiber distribution system that works well
enough. You can go about 1000ft point to point with CAT5 or coax
between two switched ports at 10baseT-HDX. Some drivel I wrote on the
subject:
http://www.google.com/groups?selm=39isj0995dp2gsk180n1hd22ff620tmda8%404ax.com
http://www.google.com/groups?selm=onatj01td5ilomf5dd436e50v7r521f2k7%404ax.com


--
Jeff Liebermann jeffl@comix.santa-cruz.ca.us
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 AE6KS 831-336-2558
 

Mike

Splendid
Apr 1, 2004
3,865
0
22,780
Archived from groups: alt.internet.wireless (More info?)

"Jeff Liebermann" <jeffl@comix.santa-cruz.ca.us> wrote in message
news:ve0mk05fl9mfgkd7hrtssgb7o5dk0bkb12@4ax.com...
> On Fri, 17 Sep 2004 09:22:49 -0400, "Mike" <mikebegin@hotmail.com>
> wrote:
>
>>I have a Cisco AP350 with a T1 coming into my
>>house. I would like to provide access for my neighbors. What do you
>>recommend for an outdoor antenna? My house sits on a high spot, but the
>>problem is a lot of tall tree and thick woods between houses.
>
> Dunno. The best I can do is offer some general clues and rhetorical
> questions.
>
> Mini rant: No numbers = no specific answers.
>
> What are the distances involved? --- within 1 mile
> How much difference in vertical elevation? --- not much, within 50 ft
> Is this going to be a neighborhood LAN (everyone joins in) or are you
> going to do this point to point, one house at a time? --- Neighborhood
> LAN, not PTP (unless I have to).
> Is price an issue? --- Yes, I do not want to spend big $$ if DSL is coming
> down the road soon.
> Are there wired alternatives? --- Not really, unless we run our own wire?
>

Thanks Jeff for the excellent response. Sorry for not providing more info.
Please see above.

Thanks again.

Mike




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----
 
G

Guest

Guest
Archived from groups: alt.internet.wireless (More info?)

On Fri, 17 Sep 2004 14:02:40 -0400, "Mike" <mikebegin@hotmail.com>
wrote:


>> What are the distances involved? --- within 1 mile
>> How much difference in vertical elevation? --- not much, within 50 ft
>> Is this going to be a neighborhood LAN (everyone joins in) or are you
>> going to do this point to point, one house at a time? --- Neighborhood
>> LAN, not PTP (unless I have to).
>> Is price an issue? --- Yes, I do not want to spend big $$ if DSL is coming
>> down the road soon.
>> Are there wired alternatives? --- Not really, unless we run our own wire?
>>
>
>Thanks Jeff for the excellent response. Sorry for not providing more info.
>Please see above.

Please don't reply to posting like that. Put your answers on seperate
lines so I don't have to surgically seperate my drivel from your
answers.

One mile is a bit rough though the trees. See:
http://trevormarshall.com/lapierre.htm
for how one person did an 1100ft link using high gain antennas through
the trees. Despite the increased gain, he still was only able to get
33% signal strength.

50 ft of vertical elevation is not going to make much difference at
5,280 ft. Therefore, you don't have to worry about downtilt or
beamwidth.

A neighborhood LAN usually requires an omni antenna or multiple radios
with panel (patch or sector) antennas. However, at one mile, you're
close to the limits of what can be done with cheap commodity routers.
I suggest a 12dBi omni antenna on your roof, your access point (not
router) mounted at the antenna to eliminate lossy coax, and panel or
patch antennas with 8dBi gain or more at the neighbors. However, if
there are thick trees in the way, and no line of sight, forget it. It
won't work and if it does work initially, it won't stay working when
the trees get wet or grow leafs.

If DSL is coming, you're going to have a difficult time getting the
neighbors to pay for the hardware. Little things add up quickly with
wireless. We had the same problem in my area, where SBC kept
promising DSL, but letting the delivery date slide. We could sell
WISP service, but only to those that didn't know that DSL was coming.

Running your own wire is an alternative. It's not that difficult in
rural areas. Forget it in the cities.


--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831.336.2558 voice http://www.LearnByDestroying.com
# jeffl@comix.santa-cruz.ca.us
# 831.421.6491 digital_pager jeffl@cruzio.com AE6KS