Hello,

Can someone tell me what type of electical length does 802.11 antennas
(lambda/2, lambda/4, etc) have? The 2.4GHz has 12.5cm wavelength but
none of the rubber duck antennas that come with the wirless router have
a length of 12.5cm, 6.25cm, or 3.125cm, etc.

I recently purchased a 9dBi antenna to replace the 2.2dBi stock rubber
ducks. It also does not appear to measure in any n*lambda/4 lengths.
What puzzles me is that an outdoor $200 antenna of 3 foot long and a
$60 antenna of slightly less than 1 ft both have about the same gain.

And before the purchase I have tried to make a few higher gain antennas
out of a wire coat hanger in n*lambda/4 lengths but they don't have
much of an impact in signal strength. Why?

I did an antenna course years ago and remember the gain of a monopole
is related to its radiation resistance, which is proportional to
n*lambda/4 from what I can recall.

Thanks!

welcome to the foggy world of advertised antenna properties. i would guess
that in the 802.11 stuff there is probably just as little control over
quotes of antenna gains, measurement methods, etc as there is an amateur
antennas... this is why for a long time the arrl refused to print antenna
gain numbers.

wrote in message
oups.com...
Hello,

Can someone tell me what type of electical length does 802.11 antennas
(lambda/2, lambda/4, etc) have? The 2.4GHz has 12.5cm wavelength but
none of the rubber duck antennas that come with the wirless router have
a length of 12.5cm, 6.25cm, or 3.125cm, etc.

I recently purchased a 9dBi antenna to replace the 2.2dBi stock rubber
ducks. It also does not appear to measure in any n*lambda/4 lengths.
What puzzles me is that an outdoor $200 antenna of 3 foot long and a
$60 antenna of slightly less than 1 ft both have about the same gain.

And before the purchase I have tried to make a few higher gain antennas
out of a wire coat hanger in n*lambda/4 lengths but they don't have
much of an impact in signal strength. Why?

I did an antenna course years ago and remember the gain of a monopole
is related to its radiation resistance, which is proportional to
n*lambda/4 from what I can recall.

Thanks!

wrote:
Hello,

Can someone tell me what type of electical length does 802.11 antennas
(lambda/2, lambda/4, etc) have? The 2.4GHz has 12.5cm wavelength but
none of the rubber duck antennas that come with the wirless router have
a length of 12.5cm, 6.25cm, or 3.125cm, etc.

I recently purchased a 9dBi antenna to replace the 2.2dBi stock rubber
ducks. It also does not appear to measure in any n*lambda/4 lengths.
What puzzles me is that an outdoor $200 antenna of 3 foot long and a
$60 antenna of slightly less than 1 ft both have about the same gain.

And before the purchase I have tried to make a few higher gain antennas
out of a wire coat hanger in n*lambda/4 lengths but they don't have
much of an impact in signal strength. Why?

I did an antenna course years ago and remember the gain of a monopole
is related to its radiation resistance, which is proportional to
n*lambda/4 from what I can recall.

Thanks!

t
You are correct, the radiation resistance partly depends on, or partly
governs, the dimensions (depending on how you look at it). The size
(length) of the antenna depends not only on the diameter of the wire,
but on the surrounding material too. An antenna can be made shorter
than what you would expect by simply increasing the permittivity (or
dielectric constant) of the material around it. My guess would be that
the antenna is coated with a high permittivity material which (in the
material) makes the wavelength of any frequency appear shorter than it
actually is. So the antenna only actually has to be as long as the
quarter wavelength in the dielectric.

Another possibility would be to introduce a reactance (inductance or
capacitance [I can't remember which at the moment]) to shorten the
electrical length of the antnna, thereby, once again, only requiring a
shorter than expected antenna length.

As far as your tests go, I can't really comment on why you don't see a
change. But... If you'll recall from your antenna course, a lambda/2
dipole has a figure 8 pattern when looked at from the side (E plane).
If you increase the length of the dipole (or monopole with a ground)
you are effectively creating an antenna array and you get interference
from each element (each section of lambda/2). Your figure 8 with a low
gain broad main beam begins to look like a flower with thin petals.
Each petal (lobe) has higher gain but a narrower beamwidth. You may be
getting multipath (a type of interference) due to the extra lobes or
you simply may be looking from an undesireable angle. Perhaps you
could try moving your non homebrew antenna station in the E plane of
your antenna and see if the signal strength changes.

I hope this sheds some light!

Dario

Hi Dario,

Thanks for the information on the permittivity factor allowing a
shorter antenna due to a shorter wavelength, and for pointing out the
array part for a longer monopole. I actually did quite a bit of
analysis on arrays in multiple dimensions but never thought a longer
antenna element would be equivalent to stacking up monopoles. So
thanks for pointing that out.

But part of the question remains, I guess then is how would be possible
to increase gain of a monopole without going over lambda/2, while
maintaining the dounut shape radiation pattern of a monopole.

Also, out of curiosity, why would any one wants to pay $200 for a 3 ft
long antenna vs $60 for a less than 1 ft long antenna of the same gain,
only because the $200 is an outdoor one?

See P.T. Barnum - something about "A sucker born every minute"...

wrote in message
oups.com...

Also, out of curiosity, why would any one wants to pay $200 for a 3 ft
long antenna vs $60 for a less than 1 ft long antenna of the same gain,
only because the $200 is an outdoor one?

Cost to manufacture/distribute is often the calculation starting point
when determining final price..

Prices tend to be set at whatever the market will bear.... Also note
that those with flashy paint jobs and go-fast stripes will always cost
more! grin

Antennas designed/built long ago may have to amortise higher design
costs. Then again more sales over the years should have bought the price
down.

Quality? Longevity? Gain should not be the sole method for making a
choice of what antenna to buy. I personally would also look at sidelobe
performance (for reduction of possible interference). One also has to
calculate the path requirements and signal margins to do the job
properly. Allowing for cable loss is another biggy. Many big spread
spectrum suppliers (eg Cisco) have a spreadsheet you can download for
calculating the link budget for an LOS path.

Re increasing the gain w/out and incerase in size, generally not. If you
load the elements in such a way that they are smaller to make use of
phasing (eg a collinear) then losses are introduced as well as the
aperture getting smaller. My opinion is that you'd have better
directivity but no/little increase of gain in the intended direction.
The donut would be flatter by the way.

Would you care to explain your exact requirements? Like are you trying
for point to point links or just a larger hot spot coverage. They have
diferent antenna requirments.

Cheers Bob W5/VK2YQA

wrote:
..

Also, out of curiosity, why would any one wants to pay $200 for a 3 ft
long antenna vs $60 for a less than 1 ft long antenna of the same gain,
only because the $200 is an outdoor one?

Does the antenna you bought show a squashed donut shape for its
pattern? Does it show a pattern at all? Telling you 9dBi doesn't say
very much on how the antenna performs. It may be (unlikely though)
that the 9dBi is only good through a few degrees of space (a pencil
beam pattern). Knowing something about antennas obviously helps.

You posed a good question. It made me do a little research and I came
up with the following. As you may also recall, if you space your
elements in a an array lambda/2 apart. You get higher gain in the
broadside direction. Say you stacked many monopoles (end to end) and
made sure their phase centers (feed points in many cases) were lambda/2
apart the donut shape would be slightly preserved, squashing it down to
a flatter donut, thereby increasing the gain. (Higher gain equates to
narrower beamwidth). You would ultimately end up with a donut with
ripples as you moved from the "hole" to the circumference of the donut.
The points farthest away from the center would be the 9dBi gain. And
as you would expect, you would only get that 9 dBi through a few
degrees from the center of the antenna. The tricky part in this is
feeding the many monopoles inside the plastic shaft in which your
antenna array is encased. You want to be able to feed the monopoles
without having your feed network radiate and throw off your desired
pattern. This is where smarter people than I come in...

As for your question on the price... As Bob Bob said, gain isn't the
only characteristic antennas have. The price difference could
represent a differene in phase linearity from one point on the pattern
to another. The matching network, beamwidth, bandwidth, side lobe
levels, and a few others also make one antenna more desirable than
others.

Regards,

Dario