I built this antenna expecting better performance from my WiFi
setup. Stock antennas are 1/2 wave verticals at the back of the
router. New antenna is a 16 element coaxial collinear @32' agl.
Total transmission system losses work out to 5.299db @2.412
ghz (includes feedline + N connectors).

So, why does this antenna only have about (rough measurement)
1-3 db gain over the stock antenna? I was meticulous in my work,
measuring the antenna parts with a digital caliper and checking
the connections with a DVM along the way.

Maybe my design is a fault? I built the antenna just like this:

http://wireless.gumph.org/content/4/...s-antenna.html

Except in my version, I've got 16 elements made from RG58. I also
made 2 four element antennas for the back of my thinkpad and they
seem to be working about twice as far as the antennas they replaced
(inverted V's).

Any ideas?

73's de Ken KG0WX - Kadiddlehopper #11808,
Flying Pigs #-1055, Grid EM17io,
Elecraft K2 #4913, XG2, 4SQRP Tenna Dipper,
Heath GD-1B, MP-1(X)antenna

On Sun, 9 Apr 2006 12:45:06 -0500, "Ken Bessler"
wrote:

New antenna is a 16 element coaxial collinear @32' agl.
Total transmission system losses work out to 5.299db @2.412
ghz (includes feedline + N connectors).

So, why does this antenna only have about (rough measurement)
1-3 db gain over the stock antenna? I was meticulous in my work,
measuring the antenna parts with a digital caliper and checking
the connections with a DVM along the way.

Maybe my design is a fault?

Hi Ken,

What did you compute as being the Velocity Factor for the shield being
covered with thick PVC?

The problem with this modified Franklin Array is that you are cutting
for half-wave dimensions once, for two Velocity Factors along the same
length. Only one is going to be correct, and it is going to impact
the theoretical application of the other.

You might want to try again using bare wire, twin lead fashion.

73's
Richard Clark, KB7QHC

"Richard Clark" wrote in message
...
On Sun, 9 Apr 2006 12:45:06 -0500, "Ken Bessler"
wrote:

New antenna is a 16 element coaxial collinear @32' agl.
Total transmission system losses work out to 5.299db @2.412
ghz (includes feedline + N connectors).

So, why does this antenna only have about (rough measurement)
1-3 db gain over the stock antenna? I was meticulous in my work,
measuring the antenna parts with a digital caliper and checking
the connections with a DVM along the way.

Maybe my design is a fault?

Hi Ken,

What did you compute as being the Velocity Factor for the shield being
covered with thick PVC?

The problem with this modified Franklin Array is that you are cutting
for half-wave dimensions once, for two Velocity Factors along the same
length. Only one is going to be correct, and it is going to impact
the theoretical application of the other.

You might want to try again using bare wire, twin lead fashion.

73's
Richard Clark, KB7QHC

To Ken

Hi Ken, I'd always take Richard's advice before mine. But, I wonder if
you realize how difficult your task is. It is a very difficult project to
develop the design of any array of radiators thats about 5 wavelengths long.
And, it *is* a project, not just a matter of precise measuring and cutting.
I'd think that 3 dB improvement over the antenna that was designed for
optimum performance shows that you are doing something right.
This project is a great learning opportunity for you.

I'd suggest that you get a directional coupler and get some idea of the
match of the array while it is being made longer. If you can manage to keep
the VSWR under 2:1 with 6 or 8 elements, at your operating frequencies, you
will be way ahead of me.

Jerry KD6JDJ

"Jerry Martes" wrote in message
news:qCc_f.2174$wH1.1885@trnddc03...

"Richard Clark" wrote in message
...
Hi Ken,

What did you compute as being the Velocity Factor for the shield being
covered with thick PVC?

The problem with this modified Franklin Array is that you are cutting
for half-wave dimensions once, for two Velocity Factors along the same
length. Only one is going to be correct, and it is going to impact
the theoretical application of the other.

You might want to try again using bare wire, twin lead fashion.

73's
Richard Clark, KB7QHC

To Ken

Hi Ken, I'd always take Richard's advice before mine. But, I wonder if
you realize how difficult your task is. It is a very difficult project
to
develop the design of any array of radiators thats about 5 wavelengths
long.
And, it *is* a project, not just a matter of precise measuring and
cutting.
I'd think that 3 dB improvement over the antenna that was designed for
optimum performance shows that you are doing something right.
This project is a great learning opportunity for you.

I'd suggest that you get a directional coupler and get some idea of the
match of the array while it is being made longer. If you can manage to
keep
the VSWR under 2:1 with 6 or 8 elements, at your operating frequencies,
you
will be way ahead of me.

Jerry KD6JDJ


Thanks to both of you for your help. I took the antenna down and by-
passed the 40' of LMR 400. Still, my Rssi with the OEM antennas was
only 3db better on the big antenna. Thinking maybe the radome was an
issue, I took it off and saw about another 2db improvement.

I had fun building the antenna and I learned a lot. I just bought a pair
of OEM high performance (9dbi gain) antennas off of eBay. Someone
suggested running the router up my tower in an enclosure. Neat idea...

Ken KG0WX

"Ken Bessler" wrote in message
news:h5d_f.557$B42.127@dukeread05...

"Jerry Martes" wrote in message
news:qCc_f.2174$wH1.1885@trnddc03...

"Richard Clark" wrote in message
...
Hi Ken,

What did you compute as being the Velocity Factor for the shield being
covered with thick PVC?

The problem with this modified Franklin Array is that you are cutting
for half-wave dimensions once, for two Velocity Factors along the same
length. Only one is going to be correct, and it is going to impact
the theoretical application of the other.

You might want to try again using bare wire, twin lead fashion.

73's
Richard Clark, KB7QHC

To Ken

Hi Ken, I'd always take Richard's advice before mine. But, I wonder if
you realize how difficult your task is. It is a very difficult project
to
develop the design of any array of radiators thats about 5 wavelengths
long.
And, it *is* a project, not just a matter of precise measuring and
cutting.
I'd think that 3 dB improvement over the antenna that was designed for
optimum performance shows that you are doing something right.
This project is a great learning opportunity for you.

I'd suggest that you get a directional coupler and get some idea of the
match of the array while it is being made longer. If you can manage to
keep
the VSWR under 2:1 with 6 or 8 elements, at your operating frequencies,
you
will be way ahead of me.

Jerry KD6JDJ


Thanks to both of you for your help. I took the antenna down and by-
passed the 40' of LMR 400. Still, my Rssi with the OEM antennas was
only 3db better on the big antenna. Thinking maybe the radome was an
issue, I took it off and saw about another 2db improvement.

I had fun building the antenna and I learned a lot. I just bought a pair
of OEM high performance (9dbi gain) antennas off of eBay. Someone
suggested running the router up my tower in an enclosure. Neat idea...

Ken KG0WX

Hi Ken

You might want to look at an "Acess point" to locate up at the antenna so
you can still use the other ports in your router.

Jerry

Hi Ken

Further to my last about a narrow vertical beamwidth.

http://members.cox.net/vk2yqa/pattern2.png

Is a quick and rough modeling of a 16 el collinear. Note the half power
beamwidth of only 4 degrees...

Cheers Bob

"Bob Bob" wrote in message
...
Hi Ken

Further to my last about a narrow vertical beamwidth.

http://members.cox.net/vk2yqa/pattern2.png

Is a quick and rough modeling of a 16 el collinear. Note the half power
beamwidth of only 4 degrees...

Cheers Bob

Wow - that's sharp! It explains why I lost the signal in my front yard.

How about this version?

http://martybugs.net/wireless/collinear.cgi

I just built it out of a coat hanger, soldered to the old coax
feedpoint from the last "experiment". It performs pretty good
but I calculate 2.14db of losses in the line. Something my
back of the set antennas don't have to deal with.

What about expanding the above antenna?

Ken KG0WX

Hi Ken

Note when I modeled the 16 element I did so as a series of dipoles fed
in phase rather than use the coax/crossover design.

I felt like checking your statement about being unable to receive in the
front yard. At around 45 degrees down from the antenna the NEC2 output
says that you'll get a lobe between 20 and 30dB down from an isotropic
(or 35-45dB down from the max gain) I'll admit my near close in signal
theory doesnt really exist but a path of 800m has about 106dB of loss.
If you knock that back to 12 metres (6dB every time you halve) you have
about 70dB path loss. If you then take a WAP kind of power level of
around 10dBm the RX signal at 12m with a halve wave is going to be
around (10-70-25) -85dBm. This kind of corelates with the lowest usable
signal for WiFi! There ya go!

I like the URL you sent as being more predictable/easier to construct
that using bits of coax. Keep in mind that progressively longer and
longer colinears are more and more critical to build and the
construction method that is being used is not really optimal in the area
of spacing between elements. You could probably pick a much better
design. I think however that you need to sit back and consider what you
are trying to achieve. Are you setting up a personal WiFi LAN for others
to use or just for specific contacts? If you are wanting to work in one
direction (or use a rotator) it might be more effective to use a corner
reflector and maybe 4 elements in phase. This will be much less critical
in construction. Remoting the WAP will make a lot of sense too as
you'll lose the line loss! Comments?

Cheers Bob


Ken Bessler wrote:



Wow - that's sharp! It explains why I lost the signal in my front yard.

How about this version?

http://martybugs.net/wireless/collinear.cgi

I just built it out of a coat hanger, soldered to the old coax
feedpoint from the last "experiment". It performs pretty good
but I calculate 2.14db of losses in the line. Something my
back of the set antennas don't have to deal with.

What about expanding the above antenna?

Ken KG0WX

"Bob Bob" wrote in message
...
Hi Ken

Note when I modeled the 16 element I did so as a series of dipoles fed
in phase rather than use the coax/crossover design.

I felt like checking your statement about being unable to receive in the
front yard. At around 45 degrees down from the antenna the NEC2 output
says that you'll get a lobe between 20 and 30dB down from an isotropic
(or 35-45dB down from the max gain) I'll admit my near close in signal
theory doesnt really exist but a path of 800m has about 106dB of loss.
If you knock that back to 12 metres (6dB every time you halve) you have
about 70dB path loss. If you then take a WAP kind of power level of
around 10dBm the RX signal at 12m with a halve wave is going to be
around (10-70-25) -85dBm. This kind of corelates with the lowest usable
signal for WiFi! There ya go!

I like the URL you sent as being more predictable/easier to construct
that using bits of coax. Keep in mind that progressively longer and
longer colinears are more and more critical to build and the
construction method that is being used is not really optimal in the area
of spacing between elements. You could probably pick a much better
design. I think however that you need to sit back and consider what you
are trying to achieve. Are you setting up a personal WiFi LAN for others
to use or just for specific contacts? If you are wanting to work in one
direction (or use a rotator) it might be more effective to use a corner
reflector and maybe 4 elements in phase. This will be much less critical
in construction. Remoting the WAP will make a lot of sense too as
you'll lose the line loss! Comments?

Cheers Bob


That's great info, Bob - btw my router puts out +20dbm and the
background noise level in the receiver @ 2.422ghz is -97dbm +/-
2-3db....

As far as my purpose, I'm trying to get access when I go to a park
to go /p on 40m. I want to be able to access the DX cluster &
HF Pack users as well as check e-mail.

I did a little research and the antennas I bought on eBay are 7dbi
gain with, of course no feedline losses. No matter how much I
crunch the numbers, I just can't find a more cost effective system.

Ken KG0WX

Hi Ken

Well if you are going to a specific location (the park) you may as well
use a directive antenna like a corner reflector/gridpack etc. In fact
you could be real smart and use a 2m handheld and a low data rate link
to swing the antenna in whatever direction you choose (eg packet talking
to a home PC, PC connected to rotator. You could even make it as simple
as a two state tone decoder for CW or CCW the antenna whilst watching
the WiFi signal) There are some quite broad reflector based designs out
there that give good gain on 2.4GHz. For example a 900x700mm gridpack
will give you about 25dBi.

I guess you are using one of the amateur radio 2.4GHz channels? You can
use a higher EIRP than the normal WiFi is limited to. (from memory 30dBm
EIRP is the legal max for unlicensed use) You do need to identify
though. (A simple text based ping might be enough)

I am surprised at the -97dBm sensitivity. The links I worked on in the
past we used -87 for 11MB 802.11b predictions. I guess your figure would
be at the lowest data rate and thus effective bandwidth. You can't beat
Boltzmann! The Ethernet microwave data radios I work with nowadays need
about 20-25dB s/n for a 50MB/sec channel that covers about 10MHz b/w.
Thermal noise in 10MHz is about -103dBm so we need better than -83 for a
good path. Our radios dont change speeds/bandwidth on the fly though.
Interference will be your greatest enemy though...

One of you other posters suggested remoting the router box to remove the
cable loss problem. It is the place where most of your problems lie.
Even moving it part the distance will help. You might also look at a
better coax. We use to use LMR400 extensively. It was much cheaper than
RG213 and lower loss as well.

Oh and be careful with the WAP etc setup as regards "maximum distance".
We had a problem early on where we left it at the default for a 10km
path and the ACKs (or something) kept crashing and kept the speed down.
The distance number introduces some kind of ACK delay..

Good luck!

Cheers Bob VK2YQA

Ken Bessler wrote:


That's great info, Bob - btw my router puts out +20dbm and the
background noise level in the receiver @ 2.422ghz is -97dbm +/-
2-3db....