On 14 November 2013 03:50, Jim W7RY <w7ry@centurytel.net> wrote:
>> There's a lot of people creating web pages, claiming all sorts of
>> wierd and wonderful things, but are seriously in error, but seem
>> believable to others. Here is one example
>>
>> http://www.rason.org/Projects/collant/collant.htm
>
>
>
> What's wrong with this site? Lets be specific here! Not just deal in
> generalities.
>
> 73
> Jim W7RY
Jim,
the web page claims a gain of 9 dB for this coaxial colinear antenna.
We have no idea if it is dBi or dBd. I suspect he means dBd, since I
suspect his logic is
1) Gain of a half-wave dipole is 0 dBd
2) Double the number of dipoles to 2 and it's 3 dBd gain.
3) Double the number of dipoles to 4 and it's 6 dBd gain.
3) Double the number of dipoles to 8 and it's 9 dBd gain.
We don't know his logic, but that is my guess.
I simulated that antenna in HFSS, which is a seriously expensive bit
of software (over $100,000) and got a gain of 7 dBi. This was with the
help at a very good support engineer at Ansys. So I suspect my
simulations are reasonable.
Someone else, who I have never met, simulated the same antenna, but
using another expensive 3D EM simulator. His was based on the finite
different time domain method, which is different to what HFSS uses. He
too got a gain of 7 dBi. So two people, using two bits of software
based on an entirely different method, both get a gain of a little
over 7 dBi. The only thing common about this software is they both
give numerical results starting from Maxwell's Equations. That's as
accurate as you can get, although it ignore quantum effects.
So this ham claims "9 dB" gain, but two people, using two pieces of
high end professional software, both simulate a gain of around 7 dBi
or 5 dBd. We both assumed perfect conductors, as it makes the
simulations faster.
So I believe the claims of the performance of that antenna by that
author are invalid.
There is a little "thought experiment" you can do with that antenna.
What would happen if the dielectric in the coax had a permittivity of
1 million? Dielectrics with a permittivity of 1000000 do exist, but
they are no good for RF. But lets assume there is one. Now the lengths
of all the elements get scaled from the free space length by a factor
of sqrt(1000000)=1000.
In free space, a half-wave of 70 cm would be 35 cm (roughly). Scaling
by a factor of 1000, and the half-wave elements would be 0.35 mm long.
8 sections of 0.35 mm long coax give an antenna with a total length of
2.8 mm, or about 1/9th of an inch. Do you seriously believe an antenna
that is 2.8 mm long could have a gain of 9 dBd or dBi at 70 cm?
The web page is junk. I've tried to communicate with the author, but
there is no response.
There are a lot of bad web pages on amateur radio. I think antennas
are about the worst area since
1) It is difficult to measure antenna gain accurately.
2) Antenna theory is really hard.
3) Under the right conditions, you can hear stations on the other side
of the world on very small antennas, so people can believe they have
some great antenna, when in practice they do not. .
Dave, G8WRB
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|