On 7 Dec 2006 20:27:37 -0800, "
wrote:

Actually, maybe it doesn't have gain compared to a halo... I'm not
really sure... I have an inconclusive 6m Big Wheel EZNEC model with an
average gain problem... though at this point it's only showing a -0.5dB
average gain or so, and impedance and gain seem constant with
segmentation changes, so it's probably not so bad.

What average gain did you expect? In case you had in mind that average
gain would be a big positive number, an antenna doesn't create power,
average gain (ie gain averaged over a sphere) MUST be less than 0dB if
it has ANY loss, mustn't it?

Owen
--

"Owen Duffy" wrote in message
...
On 7 Dec 2006 20:27:37 -0800, "
wrote:

Actually, maybe it doesn't have gain compared to a halo... I'm not
really sure... I have an inconclusive 6m Big Wheel EZNEC model with an
average gain problem... though at this point it's only showing a -0.5dB
average gain or so, and impedance and gain seem constant with
segmentation changes, so it's probably not so bad.

What average gain did you expect? In case you had in mind that average
gain would be a big positive number, an antenna doesn't create power,
average gain (ie gain averaged over a sphere) MUST be less than 0dB if
it has ANY loss, mustn't it?

Owen
--

The way to get gain is to stack them (with phasing) in order to put as much
power as
possible into low angle transmission.

I have 4 stacked 2 meter big wheels originally manufactured by Cushcraft.
Best Regards,
Ed, N5EI

What average gain did you expect?

Three megaBels ;-) Seriously, though, most antennas I model end up
being within 0.05dB of zero. You're right about the spacing, I was
making it wide for no particularly good reason ... with the original
segmentation and a more realistic spacing of the transmission-line
wires (0.6 inches or something), I was getting an average gain in free
space with zero wire loss of about -3dB... wider spacing got me closer
to zero.

I was changing the wrong things to fix the average gain discrepancy.
I'll play around with it more as an educational experience... I posted
it mostly because the wire geometry is tricky to get in the first
place, lots of creating loops and deleting segments and so forth, but
it's easy to modify once you've got it, so I figured it would be of
some use.

Now that you've fixed the segmentation, maybe we can put the spacing
back and have a working model... I'll have to look into it.

Thanks Owen..

Dan

On 8 Dec 2006 15:08:43 -0800, "
wrote:

....
Now that you've fixed the segmentation, maybe we can put the spacing
back and have a working model... I'll have to look into it.
....

It looks an interesting thing to model.

There are certainly widely ranging claims for its gain in articles on
the 'net.

I have worked up the parameters for a 3D cubic Bezier curve that would
be a good model of half of one of the loops, so I will write some code
to generate a NEC model from it.

Owen
--