Jeff Liebermann wrote:
On Sun, 21 Feb 2010 21:44:56 -0800 (PST), Art Unwin
wrote:

On Feb 21, 11:18 pm, Jeff Liebermann wrote:
On Sun, 21 Feb 2010 19:45:03 -0800 (PST), Art Unwin

wrote:
Model of 2 element multi polarization.
http://users.sdsc.edu/~unwin/art/Picture6.png
http://users.sdsc.edu/~unwin/art/Picture7.png
http://users.sdsc.edu/~unwin/art/Picture10.png
http://users.sdsc.edu/~unwin/art/Picture11.png
I don't see a model. There is no .EZ, .GAA, .N4W or .NEC file for me
to play with and tear apart. I can't even tell what the antenna looks
like from what you've posted.

You have two elements each located with x,y and z co ordinates at
each end, so you make your own file to suit the program you intend to
use or is available.

I do not have two elements. I can't located them in x, y, or z
because you didn't specify any such coordinates. I can make a
suitable NEC deck, but *YOU* need to supply the numbers.


Jeff

I hate to support Art, but he did give the endpoint data in picture6. I
modeled it with eznec+ 5 and the pattern and gain are reasonably close
to what he shows in the other "pictures". Close considering he appears
to be using mininec and if he includes conductor loss and real ground
it's off a fair amount (I show 9.36 dBi gain). I didn't run circular,
and doubt that it has much in the way of circularity, which is kind of
obvious from the elements. Even though they are skewed in a way which
he probably patented.

tom
K0TAR

On Feb 23, 10:35*pm, tom wrote:
Jeff Liebermann wrote:
On Sun, 21 Feb 2010 21:44:56 -0800 (PST), Art Unwin
wrote:

On Feb 21, 11:18 pm, Jeff Liebermann wrote:
On Sun, 21 Feb 2010 19:45:03 -0800 (PST), Art Unwin

wrote:
Model of 2 element multi polarization.
http://users.sdsc.edu/~unwin/art/Picture6.png
http://users.sdsc.edu/~unwin/art/Picture7.png
http://users.sdsc.edu/~unwin/art/Picture10.png
http://users.sdsc.edu/~unwin/art/Picture11.png
I don't see a model. *There is no .EZ, .GAA, .N4W or .NEC file for me
to play with and tear apart. *I can't even tell what the antenna looks
like from what you've posted. *

You have two elements each located with x,y and z *co ordinates at
each end, so you make your own file to suit the program you intend to
use or is available.

I do not have two elements. *I can't located them in x, y, or z
because you didn't specify any such coordinates. *I can make a
suitable NEC deck, but *YOU* need to supply the numbers.

Jeff

I hate to support Art, but he did give the endpoint data in picture6. *I
modeled it with eznec+ 5 and the pattern and gain are reasonably close
to what he shows in the other "pictures". *Close considering he appears
to be using mininec and if he includes conductor loss and real ground
it's off a fair amount (I show 9.36 dBi gain). *I didn't run circular,
and doubt that it has much in the way of circularity, which is kind of
obvious from the elements. *Even though they are skewed in a way which
he probably patented.

tom
K0TAR

I got the same gain as you using lower segments but it went up when I
doubled the segments so I held to that figure. I thought I provided
the cp and it was something like 3 dbi down which is much better than
20 dbi or more down if the array was not sensitive to cp. The same
segment problem occured again when modeling the
Beverage in circular form as it requires a tremendous amount of
segments to get total accuracy but the actual antenna in practice
showed conformality to make me comfortable. The same problem occurs
again using wire mesh and it would cost me near $1000 to acomplish
that but in practice it functioned very close to the circular wound
Beverage. The best antenna in the trials appeared to be long dual
wire mesh curtains which I checked out on top band tho conditions were
not consistent to hang my hat on. I intent to check out the "tilt"
action during the coming year using "garbage can" shaped mesh and at
ground level.
The whole point of the modeling was to show two FW elements that were
not planar or parallel provided an array that was sensitive to more
than a yagi built for horizontal gain alone. This to my mind makes it
a more efficient antenna for conversion of signals that hit it.
Another interesting point showed up with the circular beverage which
clearly merges the two vector radiations instead of the saucer shape
radiation from the earths rotation vector
and the gravity vertical column radiation that is added to the above
to form a planar yagi radiation form.

On Feb 24, 5:16*am, Art Unwin wrote:
On Feb 23, 10:35*pm, tom wrote:



Jeff Liebermann wrote:
On Sun, 21 Feb 2010 21:44:56 -0800 (PST), Art Unwin
wrote:

On Feb 21, 11:18 pm, Jeff Liebermann wrote:
On Sun, 21 Feb 2010 19:45:03 -0800 (PST), Art Unwin

wrote:
Model of 2 element multi polarization.
http://users.sdsc.edu/~unwin/art/Picture6.png
http://users.sdsc.edu/~unwin/art/Picture7.png
http://users.sdsc.edu/~unwin/art/Picture10.png
http://users.sdsc.edu/~unwin/art/Picture11.png
I don't see a model. *There is no .EZ, .GAA, .N4W or .NEC file for me
to play with and tear apart. *I can't even tell what the antenna looks
like from what you've posted. *

You have two elements each located with x,y and z *co ordinates at
each end, so you make your own file to suit the program you intend to
use or is available.

I do not have two elements. *I can't located them in x, y, or z
because you didn't specify any such coordinates. *I can make a
suitable NEC deck, but *YOU* need to supply the numbers.

Jeff

I hate to support Art, but he did give the endpoint data in picture6. *I
modeled it with eznec+ 5 and the pattern and gain are reasonably close
to what he shows in the other "pictures". *Close considering he appears
to be using mininec and if he includes conductor loss and real ground
it's off a fair amount (I show 9.36 dBi gain). *I didn't run circular,
and doubt that it has much in the way of circularity, which is kind of
obvious from the elements. *Even though they are skewed in a way which
he probably patented.

tom
K0TAR

I got the same gain as you using lower segments but it went up when I
doubled the segments so I held to that figure. I thought I provided
the cp and it was something like 3 dbi down which is much better than
20 dbi or more down if the array was not sensitive to cp. The same
segment problem occured again when modeling the
*Beverage in circular form as it requires a tremendous amount of
segments to get total accuracy but the actual antenna in practice
showed conformality to make me comfortable. The same problem occurs
again using wire mesh and it would cost me near $1000 to acomplish
that but in practice it functioned very close to the circular wound
Beverage. The best antenna in the trials appeared to be *long dual
wire mesh curtains which I checked out on top band tho conditions were
not consistent to hang my hat on. I intent to check out the "tilt"
action during the coming year using "garbage can" shaped mesh and at
ground level.
The whole point of the modeling was to show two FW elements that were
not planar or parallel provided an array that was sensitive to more
than a yagi built for horizontal gain alone. This to my mind makes it
a more efficient antenna for conversion of signals that hit it.
Another interesting point showed up with the circular beverage which
clearly merges the two vector radiations instead of the saucer shape
radiation from the earths rotation vector
and the gravity vertical column radiation that is added to the above
to form a planar yagi radiation form.

if you wind a wire into a coil it is NOT a Beverage antenna. And note
it is properly Beverage with a capital B since it is named after it's
inventor. go invent something useful and make up your own name,
don't try to distort a perfectly good antenna that has well know
characteristics.

tom wrote:

Jeff

I hate to support Art, but he did give the endpoint data in picture6. I
modeled it with eznec+ 5 and the pattern and gain are reasonably close
to what he shows in the other "pictures". Close considering he appears
to be using mininec and if he includes conductor loss and real ground
it's off a fair amount (I show 9.36 dBi gain). I didn't run circular,
and doubt that it has much in the way of circularity, which is kind of
obvious from the elements. Even though they are skewed in a way which
he probably patented.

tom
K0TAR

Picture6 data show equal horizontal and total gain, which means it's
purely horizontally polarized in the direction of the analysis. That's
also consistent with equal CW and CCW (or right and left hand) circular
components, which mean polarization (again in the direction of the
analysis) is purely linear.

Roy Lewallen, W7EL

Roy Lewallen wrote:
tom wrote:

Jeff

I hate to support Art, but he did give the endpoint data in picture6.
I modeled it with eznec+ 5 and the pattern and gain are reasonably
close to what he shows in the other "pictures". Close considering he
appears to be using mininec and if he includes conductor loss and real
ground it's off a fair amount (I show 9.36 dBi gain). I didn't run
circular, and doubt that it has much in the way of circularity, which
is kind of obvious from the elements. Even though they are skewed in
a way which he probably patented.

tom
K0TAR

Picture6 data show equal horizontal and total gain, which means it's
purely horizontally polarized in the direction of the analysis. That's
also consistent with equal CW and CCW (or right and left hand) circular
components, which mean polarization (again in the direction of the
analysis) is purely linear.

Roy Lewallen, W7EL

Thanks Roy. I was rushed at the time and also didn't care too much,
since it was obvious that even though the elements weren't parallel or
on quite the same centerline, they were very close to it. He picks at
nits when he says his antennas are different. I could get similar
results from a good windstorm.

tom
K0TAR

tom wrote:

Thanks Roy. I was rushed at the time and also didn't care too much,
since it was obvious that even though the elements weren't parallel or
on quite the same centerline, they were very close to it. He picks at
nits when he says his antennas are different. I could get similar
results from a good windstorm.

tom
K0TAR


Ok, revisiting things now that I have time, I discovered something,
which The Art may (ok, will) get off into an incomprehensible rant about.

I did a simple normalization. I calculated the lengths of the elements
and the location of their centerpoints. I then calculated the distance
between the centerpoints. I then put it all in a nice level and square
2 element array. I left the diameters alone.

New endpoints -

-37.455 -219.135 707 -37.455 219.135 707 1
37.455 -195.45 707 37.455 195.45 707 1.25


Results at 14.175 using EZNEC+ 5

Gain F/B
Original 9.87 dBi 7.84
Normalized 10.29 dBi 8.58

The impedance curves were not different enough to be of note and were
matchable to 50 ohms with good efficiency.

I'll take the one with better gain and F/B that's also nice looking with
elements that are easier to mount because they're at 90 degree angles,
thank you. Yes, run-on sentence.

tom
K0TAR

On Wed, 24 Feb 2010 19:50:47 -0600, tom wrote:

Ok, revisiting things now that I have time, I discovered something,
which The Art may (ok, will) get off into an incomprehensible rant about.

This outrage is largely due in part to revealing an inferior design:

Gain F/B
Original 9.87 dBi 7.84
Normalized 10.29 dBi 8.58

I wonder whatever happened to the golden city shining on the top of
the hill that Art called "length efficiency?" A back of the envelope
calculation of Art's 20M antenna reveals elements that are
conventional half wave. This raises the more immediate
head-scratcher: whatever happened to the golden calf of his designs
being in equalithium?

73's
Richard Clark, KB7QHC

Richard Clark wrote:

I wonder whatever happened to the golden city shining on the top of
the hill that Art called "length efficiency?" A back of the envelope
calculation of Art's 20M antenna reveals elements that are
conventional half wave. This raises the more immediate
head-scratcher: whatever happened to the golden calf of his designs
being in equalithium?

73's
Richard Clark, KB7QHC

I believe his designs are the result of lackolithium.

tom
K0TAR

On Feb 24, 8:14*pm, Richard Clark wrote:
On Wed, 24 Feb 2010 19:50:47 -0600, tom wrote:
Ok, revisiting things now that I have time, I discovered something,
which The Art may (ok, will) get off into an incomprehensible rant about..

This outrage is largely due in part to revealing an inferior design:

* * * * * * *Gain * * * F/B
Original * * 9.87 dBi * 7.84
Normalized *10.29 dBi * 8.58

I wonder whatever happened to the golden city shining on the top of
the hill that Art called "length efficiency?" *A back of the envelope
calculation of Art's 20M antenna reveals elements that are
conventional half wave. *This raises the more immediate
head-scratcher: whatever happened to the golden calf of his designs
being in equalithium?

73's
Richard Clark, KB7QHC

Richard,
I am sharing results provided by computer optimizers where an array
must be in equilibrium. I am not party to how these programs were
generated or how closely they adhere to maxwells equations.If they
provide results that are different then that seller has to answer for
it.
Obviously clever people such as yourself will guide hams on the
correct path to follow
as well as to explain why Maxwells equations do not favor the yagi.
Hopefully you will supply technical details and not just wordy fluff.
I would point out that Tom provided a planar form, which you copied,
for a different frequency to the one that I supplied which provided f/
b/r that greatly exceeded his design. If your idea is to be a pied
piper to lead ham radio away from
examining change then such tactics will certainly do the job. But
first to justify your dreams as a leader you have to provide reasons
that make optimizer programs suspect.
A point in your favor would be that Roy declares that his optimizer do
NOT supply the results that others do so that experts such as you can
direct your talents where they would be more profitable. Since you are
not wired the same as I am I will not respond to you any more so that
guilt by association may not be applied. But I will mention to all
that you are in total disagreement with the association of a static
field being made dynamic which is a staple in physics around which
this discussion rests upon. Nobody, nobody, has come forward to
provide academic reasons why this staple should not be held to or why
people such as you should trash pursuit of that what I disclose. What
I have provided is that computer programs side lines yagis in favor of
arrays where elements are not parallel but all are resonant and where
the array as a whole is in equilibrium. They do this because Maxwell
supports the physics staple I proffer which you declare as NOT being
correct but without supporting data other than you said so.

On Feb 24, 9:24*pm, Art Unwin wrote:

Hopefully you will supply technical details and not just wordy fluff.

Talk about spray painting the kettle flat black... :/