2804WBR antenna diversity

Is it possible to disable antenna diversity on this router, because I'd like
to connect just one external antenna to it.

Thanks

Hi rozsi,

I have diversity antennas on my Linksys WRT54G router, too.

The two antennas try their best to receive, and electronics inside the
router chooses the best of the two signals. You could call it a
"voter." (Does anyone prefer a different term?)

Also, I think if there is multiplex distortion (reflections causing
interference with the main signal), the electronics can combine the two
signals and make a good undistorted signal out of them. (Can someone
confirm that?) Unfortunately, that is not a trick that can be
accomplished with broadcast FM, but two identical digital streams with
microsecond delays between them should be easy for electronics to line up.

Anyway, if you remove one rubber duckie antenna and replace it with a
"real" antenna, then most likely the voter will always choose the signal
from the big, high performance antenna. This is not a problem for the
voter, or the vestigial rubber duckie that will never again see any
action. It won't break anything.

My question: If you do this, what happens to the router's transmitted
signal? It would be a shame if the power is divided equally, with half
of it going to the vestigial rubber duckie instead of the big antenna.
Maybe in this case, it would have been better to get a wireless router
with only one antenna (if there is such an animal).

CU,
-Neil-

I wrote:

> It would be a shame if the power is divided equally, with half
> of it going to the vestigial rubber duckie instead of the big antenna.
> Maybe in this case, it would have been better to get a wireless router
> with only one antenna (if there is such an animal).

On the other hand, if there are two antennas, they can be used to
advantage. I know you said that you only want to use 1 antenna, but
there are other possibilities.

If one antenna is a rubber duck, it could serve a nearby client within a
few nearby rooms. If the other is a gain antenna that's directional, it
could serve your neighbor next door.

If all of the clients are "far away," you could have two directional
antennas spreading the signal over the desired area, with a little more
control than a single directional antenna, and no wasted transmitter
power going to that vestigial rubber duckie. (I like saying vestigial
rubber duckie.) Or you could point one antenna down the street in one
direction, and the other in the other direction.

Thanks again,
-Neil-

Neil, thanks for the input.

Yes, it would be nice to use both the external antennas, but I feel that it
just doesn't work that way. If I don't connect 2 antennas, it shows 53%
signal on the right side and a lot less on the left, but no packets sent and
also disconnects all the time.
It looks like one is just receiving and the other just transmitting. Maybe
I'm wrong, but I'm 100% sure the 2 outputs aren't equal.
At the moment I'm using it with 2 sector antennas both in the same
direction, but I don't want this...

rozsi

"Flwrite" <(E-Mail Removed)> wrote in message
news:geogb.8427$(E-Mail Removed) t...
> I wrote:
>
> > It would be a shame if the power is divided equally, with half
> > of it going to the vestigial rubber duckie instead of the big antenna.
> > Maybe in this case, it would have been better to get a wireless router
> > with only one antenna (if there is such an animal).
>
> On the other hand, if there are two antennas, they can be used to
> advantage. I know you said that you only want to use 1 antenna, but
> there are other possibilities.
>
> If one antenna is a rubber duck, it could serve a nearby client within a
> few nearby rooms. If the other is a gain antenna that's directional, it
> could serve your neighbor next door.
>
> If all of the clients are "far away," you could have two directional
> antennas spreading the signal over the desired area, with a little more
> control than a single directional antenna, and no wasted transmitter
> power going to that vestigial rubber duckie. (I like saying vestigial
> rubber duckie.) Or you could point one antenna down the street in one
> direction, and the other in the other direction.
>
> Thanks again,
> -Neil-
>

Hi Rozsi,

Thanks for coming back.

News:

www.1stwave.de/produkte/ 1stWAVE_Professional_PC_Card.pdf - "The antenna
function of the PC Card provides a built-in diversity antenna. The
diversity antenna contains two radiating elements: one is used for
transmission, and the DSP selects which of the two to use for reception,
based on signal strength."

Strike one against me.

http://www.dfwwireless.org/article.php?sid=1 - Diversity: Most access
points have two antennas. One of these is used as the primary
transmitting and receiving port, while the other is periodically checked
(polled) to see if it is receiving a stronger signal than the main
antenna. This is called a "diversity" antenna system. It can help to
reduce variations in signal strength as you vary the location of an
access point and a client. While there is nothing to stop you deploying
two good antennas for each access point, one good antenna is always
superior to two ordinary ones.

Strike two.

It raises other questions; do you need to choose the "primary antenna
port" for the big antenna? or are they equally able to become primary?
If the other antenna is selected for receiving, does that also switch
the transmitter's antenna? (Probably not, but my hypothesizing is
getting me into trouble.) If you replace the wrong rubber duckie with
the big antenna, you may improve only your reception, but not transmission.

Linksys threatens to void warranties if you remove your antenna, so it
is not likely they will be forthcoming in answering these hypothetical
questions. Hmmmmm, nothing in their online glossary on diversity.

You wrote:

> If I don't connect 2 antennas, it shows 53% signal on the right side
> and a lot less on the left, but no packets sent and
> also disconnects all the time.

Questions: What is "it" that is showing 53% signal. Do you mean RF
power? If you have some kind of power-output indication on your router,
then I am jealous.

If I disconnect one rubber duckie from my Linksys WRT54G, it stops
completely. One antenna isn't enough. I have to put the antenna back
on, and then hit the little Reboot button in the back of the router. Or
maybe I have to unplug it for a few seconds... One mistuned, broken, or
missing antenna shuts down my router. This could be parallel to you
getting "disconnected." Maybe it was dedicated for transmitting, and
removing the other antenna would not shut the router down?

> It looks like one is just receiving and the other just transmitting.

What looks like it? One of the articles suggest that one may be
transmitting, but both share reception, as needed.

> Maybe I'm wrong, but I'm 100% sure the 2 outputs aren't equal.

How did you notice this? Do you have measurement equipment? (Inquiring
minds want to know, and by the way, can you measure SWR, too?)

There are some quick resistance checks you can to with your antennas.
The connector on your end of the coax should have infinite resistance
from center-pin to shield, else there is a short. There should also be
low resistance from the center-pin on one end of the coax to the antenna
on the other end (and even from the shield on one end to the shield on
the other), or else you have an open in one of the conductors. I don't
imagine you should get over 5 ohms resistance from one end of the coax
to the other, on either conductor.

Special case: If the antenna is 5/8 waves type, then you should measure
a short from center to shield, because the coax terminates in a small
transformer, which gives the ohmmeter a DC-Short. Not to worry, the
transformer isn't a short at microwave frequencies.

If you have two similar antennas, you can compare readings. Something
is screwy if they're not similar. This would confirm your belief that
the antennas are not doing things equally.

It would be nice to check the SWR of our antennas, but that's going to
take a few thousand dollars of microwave measurement equipment. All we
can hope is that our antennas were designed properly to cover our
frequencies.

> At the moment I'm using it with 2 sector antennas both in the same
> direction, but I don't want this...

Yeh, but, is it working? Or is this the setup that is giving you packet
loss and disconnects? It would be nice to get this working before
reverting one of the big antennas to a rubber duckie. Then, the only
question will be "which one works better?" That will also be the one
that the transmitter is tied to.

If that's the way it works. The two articles above had slightly
different definitions for "diversity," so who knows.

CU,
-Neil-

http://www.smallnetbuilder.com/FAQ-50-Antennas-16.php (This website has
fantastic content, but a slow server.)

http://www.personaltelco.net/index.cgi/DiversityAntenna

http://lists.bawug.org/pipermail/wir...st/002145.html

Hi again, Neil,

thank you for those links, cleared up many things and I'll try to be
more specific now about my findings...oohh, my poor English...

> Strike two.
>
> It raises other questions; do you need to choose the "primary antenna
> port" for the big antenna? or are they equally able to become primary?

No, I don't think so, at least not the Smc2804.

> If the other antenna is selected for receiving, does that also switch
> the transmitter's antenna? (Probably not, but my hypothesizing is
> getting me into trouble.)

Hmmm...

> If you replace the wrong rubber duckie with
> the big antenna, you may improve only your reception, but not
transmission.

Yes, I think...

> Questions: What is "it" that is showing 53% signal. Do you mean RF
> power? If you have some kind of power-output indication on your router,
> then I am jealous.

No, no... I played with an SMC 2802W Pci adapter, and this was the
signal strength reading from that. It has a link quality indicator as well,
but it shows exactly the same figure as signal all the time, so I don't
believe it's working, it's just fake.
Maybe in the next driver... Hey SMC, do you read this?

> If I disconnect one rubber duckie from my Linksys WRT54G, it stops
> completely. One antenna isn't enough. I have to put the antenna back
> on, and then hit the little Reboot button in the back of the router. Or
> maybe I have to unplug it for a few seconds... One mistuned, broken, or
> missing antenna shuts down my router. This could be parallel to you
> getting "disconnected." Maybe it was dedicated for transmitting, and
> removing the other antenna would not shut the router down?

Yes, this is similar to my stuff...

> > It looks like one is just receiving and the other just transmitting.
> What looks like it? One of the articles suggest that one may be
> transmitting, but both share reception, as needed.

If this was true, then 1 antenna would be enough, wouldn't it?

> > Maybe I'm wrong, but I'm 100% sure the 2 outputs aren't equal.
> How did you notice this? Do you have measurement equipment? (Inquiring
> minds want to know, and by the way, can you measure SWR, too?)

No, unfortunately I don't have any of those equipments... As I said, I
played with a pci card and I was talking about its signal readings...
I know its not precise, but at least it gives some ideas...

> If you have two similar antennas, you can compare readings. Something
> is screwy if they're not similar. This would confirm your belief that
> the antennas are not doing things equally.

It was not the antennas, because the readings were very different for the
2 sectors as well as for the 2 original rubbers. But now I believe, the
2nd connector doesn't transmit at all.

> > At the moment I'm using it with 2 sector antennas both in the same
> > direction, but I don't want this...
>
> Yeh, but, is it working? Or is this the setup that is giving you packet
> loss and disconnects? It would be nice to get this working before
> reverting one of the big antennas to a rubber duckie.

And I should add, that I'm not 100% satisfied with it... Good signal, but
disconnects (less, but still...) and also the throughput should be better.
When I tested it, I left the other connector empty. Maybe I should've left
one of the rubbies on?

> Then, the only
> question will be "which one works better?" That will also be the one
> that the transmitter is tied to.

My router is now on a 80m tall tower and it's not that easy just to "jump"
up there, therefore I need some real ideas before I climb up again...

> If that's the way it works. The two articles above had slightly
> different definitions for "diversity," so who knows.

Yes, it wasn't clear for me either. Now I know there's separate diversity
for the receiver and the transmitter. My router has a kind of receiver
diversity, which doesn't work properly with just 1 antenna. Why?

Regards,

r)zsi

Quote 1:
access point and a client. While there is nothing to stop you deploying
two good antennas for each access point, one good antenna is always
superior to two ordinary ones.

Quote 2:
A potential alternative solution could be in using the 'second' receiving
only antenna, that most Wlan equipment have. This 'receiving only' antenna
is used for diversity reception. Despite that such an antenna has been
designed for receiving only, it has usually has a -20dB leakage during
transmitting. That means that a +15 dBm output level at the 'first'
transmitting antenna generates a -5 dBm output level transmitting level at
the 'second' receiving antenna. And that is exactly the level that's needed
for the 8,5 km legal limit link of example nr 3. So, basically you can
connect a 24 dBi antenna to the receiving antenna connector in order to
achieve a legal limit long range link.

Extra advantage is, that you still can use the existing local
transmitting/receiving antenna in order to connect to local wlan equipment.

Quote 3:
But whenever local and long-distance clients are active at the same time,
the stronger signal would tend to be selected, possibly locking out one or
the other type of connection.

Quote 4:
> Nearly every 802.11 radio I've seen supports only receive diversity,
> if diversity is supported at all. When transmitting, only one antenna
> us used.

Quote 5:
"One caution with diversity, it is not designed for using two antennas,
covering two different coverage cells. The problem in using it this way, is
that if antenna #1 is communicating to device #1, while device #2 (which is
in the antenna #2 cell) tries to communicate, antenna #2 is not connected
(due to the position of the switch), and the communication fails. Diversity
antennas should cover the same area, from only a slightly different
location."

Quote 6:
...."The AP doesn't always get to control the antenna switches. Some designs
just leave the switch set to where it was during the last reception. "


What a mixture! Interesting, isn't it... ?


r.

Dear Rozsi,

Thanks for extracting quotes from those articles.

It seems clear that if you hook a single, big antenna to the connector that
includes the transmitter, then you should be On The Air. If you're 80
meters up a tower, then perhaps it's not worth installing a 2nd big antenna
to point at the same location, just as you said.

I'm still wondering if the router will complain if no antenna is connected
to the 2nd connector. At 20 dBs down from the transmitting antenna, that
means 1/100 of the transmitted power is going to the 2nd connector. It is
difficult to image that causing a problem, but who knows.

For receiving, the problem with leaving a rubber duckie antenna on the 2nd
connector is if someone brings a Wi-Fi to your tower location. When they
transmit, their signal would be picked up more strongly by the rubber duckie
than the big directional antenna, and they might interrupt the traffic on
the big antenna. This may not be a big problem. It could be a problem if
there are hoodlum delinquents in your area.

However, a microwave "dummy load / terminator / termination" could be
attached to the 2nd connector, to render it inactive. A dummy load provides
the proper impedance to the transmitter, but doesn't radiate or receive
[much]. Then, the transmitter would be certainly happy, and [almost] no
signal would be received at the 2nd connector.

Here's one. I never heard of the company, but the pictures are pretty:

http://www.e-meca.com/rf-termination.htm

I would select a terminator that can handle twice the expected power coming
out of your transmitter, and is rated to 3 GHz. Or, you can get advice
directly from the company. They will want to know if the ambient
temperature will ever get hot; what is the expected power output and
frequency of your transmitter; etc., and let them select the proper
terminator for you.

After leaving the terminator connected for an hour, it would be OK if the
terminator feels warm. But you should be able to hold if with your bare
hands indefinitely without getting burned.

For the best chance of not having to climb the tower again, I would put a
terminator on the unused connector first. The vestigial rubber duckie
sounds like a source of problems, since it is active but provides no desired
function.

Well Roszsi, it took me a long time to write back to you. I don't speak a
2nd language (ugly American), so thanks for doing English. Maybe you
already went back up the tower and put a rubber duckie on the 2nd connector.
If everything is working well, then there is no reason to worry about having
a rubber duckie instead of a terminator.

I hope your router is not locking up anymore.

Best luck,
-Neil-


---
..
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.525 / Virus Database: 322 - Release Date: 10/9/03