"Sonie" wrote in message
...

" wrote in
message
news:1k%vd.198273$V41.135426@attbi_s52...

"hnkie" wrote in message
...

"art" wrote in message
oups.com...
In the past I mentioned that I was modelling an antenna for 20 metres.
This antenna became a single driven element with five reflectors which
emulate a dish in it's crudish form not only in position of reflectors
but also the elements were not shaped at this time.
The following table follows what one would get for a reflective dish
used on frequencies in the giga hertz range where the bandwidth is
broader than
that obtained with the yagi format PLUS relevant atributes stay more
constant over a frequency range than that obtained with a super gain
type array.
The band width is wider than a typical yagi in that the major portion
of the radiation is to the front this.This band with is expected to
narrow and to increase gain when individual elements are shaped to a
parabolic form and
it is hoped that forward side lobes will not occur as is normal for a
long boom multi element yagi form as used on 440 Mhz.
The driven element is at a height of 76 feet over perfect ground.

Normally one would say that if results seem out of the ordinary one
should question it, so comments are requested regarding my
explanations
which
lead me to think that they really are what one should expect.



Freq 14.0 14.1 14.2 14.3 14.35

Gain dbi 13.86 14.05 13.99 13.9 13.87

F/B 30 25.6 27.47 33.17 32.5

Zr 45.9 44.5 43.8 43.2 42.8

Zi -12.87 -6.52 0.6 7.84 11.54

SWR 1.32 1.2 1.14 1.25 1.34


I do not advocate the replacement of the normal yagi with the above
because of building difficulties tho it is my intention to build it.

Regards
Art

(Modelled using AO PRO with 80 segment/half wave and set up by a
consumate error generator Grin )


F/B looks too high, you should have some big splinters off the
backside.

Why? Do you have splinters in a parabolic dish?

You have individual elements, a parabolic has continous surface. Antennas
always have splinters, with just 5 elements and no blocking you probably
have major lobes on the back side, between elements.

Not so!
Small lobe yes. When you used the term "splinter" I thought it described a
large thin splinter or lobe, which is not the case. Tho the reflectors do
have a space
between them which would have a terrible effect where "reflection "
occurs.In the
case where definitive phase angles occur I doubt what you have referred to
is correct.

There will be places on the back side of your antenna that add in phase
with
two or more elements, which is a lobe or splinter.

see above

Parabolics have a rolloff on the edge to decay the field to minimize the
back lobe.

Yes I am hoping for that effect in conjunction with the narrowing of the
main beam.
Decay however is not the word I would chose. Redirect yes.

(with a dish it can be hard to get 30)
Gain looks to be high for a 1 driven and 5 reflectors.
If you picture a balloon squashed at one end more than
normal would not the gain increase?

What is the capture area? Each element captures RF energy and reflects it
back to the driven element in phase with the others.

Capture area? I don't know ( Now you can see the limits of my expertise)

Your idea has been tried very sucessfully at 450 to microwave, and they
now
use flat antennas with each element phase adjusted for delay (same as
curverature of a parabolic dish) .

Yes I certainly agree, but for some reason 440 seems to be a cut-off
frequency
and I do not know why. I did try using the director in a parabolic shape but
it was unrewarding. The delay thing you referred to may well be why the
reflectors deviated from a true parabolic shape.

One antenna has over a thousand elements

1000 elements could well be connected to what you referred to above with
respect to "blocking" where the spacing between elements must be
commensurate with the frequency used.

Regards
Art



Does model take into account mutual effects of the reflectors?
Yes
Art

Roy
After I mused a while on your post I then accepted it for the
unbiased
analysis that it was rather than a desired reflection of my own
thoughts.
It then drove me to consider chosing gain as the main criteria, after
all my interests
are talking to my mates in the U.K., but at the same time carrying on
with my main thought of moving away from the long boom analogy and its
accompanying gain aproach.
By adding a mast of 30 feet to accommodate additional elements rather
than extending boom length I was able to narrow the beam width to some
where in the 57 degree region but my excellent F/B that I obtained
earlier went way down. This was done by constructing elements in "V"
form with no restriction on alignment with respect to the driven
element other than the center of the elements must be close to the mast
i.e
to take advantage of low torque requirements.
With six elements on the 30 foot mast on the tower not only did the
gain increase
to 16 dbi ( others please take note of prior posts on this thread) but
the 'take off angle'
dropped to pretty close to !0 degrees ! This aproach for somebody who
wants to work DX seems a good way to go.
I now intend to pursue the aproach of letting the elements deviate
from the "V" shape to what I expect to be a parabolic shape, to
determine whether the extra work is worth while, as well as the
"ommission" of the most upper most element for when heavy static
occurrs.
What I really do find interesting is that one CAN get away from the
boom length aproach when seeking gain as well as obtaining a reasonable
impedance.
I also intend to attempt a tri band aproach as I move along, which may
take a while as my program has provision for a limited number variable
dimensions to optimize at one time and the number of dimensions in the
next model will certainly exceed 40 at a minimum, thus modelling cannot
be achieved in a single shot.
My present model, which moves me towards gain, seems to exceed what
performance I can expect compared to high F/B and is certainly opening
fresh avenues as to what type of antenna I will construct after the new
year.

Comments solicited from all........
and happy hollidays

Art

Cecil Moore wrote:

Dave, Your 2+ dBi gain is for a dipole in free space Vs an isotropic in
free space. Roy chose to compare the dipole over perfect ground to the
isotropic in free space. The ~3dB of reflections from the perfect ground
cause the additional gain due to constructive interference.

Art chose to compare his antenna to an isotropic antenna in free space
-- he reported the gain in dBi. So I compared a dipole at the same
height to an isotropic antenna in free space. This combination allows
comparison of Art's antenna in its position over ground to a dipole at
the same height above ground.

Roy Lewallen, W7EL

The elevation angle at which radiation is maximum (the "takeoff angle")
can be influenced by narrowing the free-space elevation pattern of the
antenna. A Yagi provides considerable directivity in the horizontal
direction, but not very much in the vertical direction. The elevation
pattern of the forward lobe of a Yagi doesn't look much different from
the elevation pattern of a dipole. Therefore, the elevation patterns of
a Yagi and dipole are very similar (except for very long Yagis, which do
begin to have some noticeable directivity in the elevation pattern). One
consequence of this similarity is that the gain of a Yagi over a dipole
in free space is about the same as the gain of a Yagi over a dipole at
any elevation angle when the two are mounted at the same height -- a
single number is adequate to describe the gain of a Yagi over a dipole,
no matter where the two are mounted (as long as they're at the same height).

There are some antennas that do compress the vertical pattern. One
notable antenna in this category is the W8JK. The result is that the
"takeoff angle" is lower for the W8JK than for a dipole at the same
height, and the gain of a W8JK compared to a dipole is different at
different elevation angles. I haven't looked at one in a long time, but
believe the "Lazy H" antenna also compresses the vertical pattern
somewhat, resulting in a lower "takeoff angle" than a dipole when
mounted over ground.

It's not too surprising that Art's antenna has the ability to compress
the vertical pattern somewhat, for the same general effect as the W8JK,
due to its physical geometry. This general characteristic of a lower
"takeoff angle" can be an advantage for working DX, so it might make the
constructional difficulties worthwhile for some people. The reduction in
F/B when maximizing gain also isn't surprising. I don't believe there's
any fundamental reason for this to always happen, but it's all too often
the case. Most arrays, both parasitic and driven, have relatively poor
F/B when adjusted for maximum gain. It becomes up to the user to decide
on the optimum tradeoff for his particular application.

Roy Lewallen, W7EL

art wrote:
Roy
After I mused a while on your post I then accepted it for the
unbiased
analysis that it was rather than a desired reflection of my own
thoughts.
It then drove me to consider chosing gain as the main criteria, after
all my interests
are talking to my mates in the U.K., but at the same time carrying on
with my main thought of moving away from the long boom analogy and its
accompanying gain aproach.
By adding a mast of 30 feet to accommodate additional elements rather
than extending boom length I was able to narrow the beam width to some
where in the 57 degree region but my excellent F/B that I obtained
earlier went way down. This was done by constructing elements in "V"
form with no restriction on alignment with respect to the driven
element other than the center of the elements must be close to the mast
i.e
to take advantage of low torque requirements.
With six elements on the 30 foot mast on the tower not only did the
gain increase
to 16 dbi ( others please take note of prior posts on this thread) but
the 'take off angle'
dropped to pretty close to !0 degrees ! This aproach for somebody who
wants to work DX seems a good way to go.
I now intend to pursue the aproach of letting the elements deviate
from the "V" shape to what I expect to be a parabolic shape, to
determine whether the extra work is worth while, as well as the
"ommission" of the most upper most element for when heavy static
occurrs.
What I really do find interesting is that one CAN get away from the
boom length aproach when seeking gain as well as obtaining a reasonable
impedance.
I also intend to attempt a tri band aproach as I move along, which may
take a while as my program has provision for a limited number variable
dimensions to optimize at one time and the number of dimensions in the
next model will certainly exceed 40 at a minimum, thus modelling cannot
be achieved in a single shot.
My present model, which moves me towards gain, seems to exceed what
performance I can expect compared to high F/B and is certainly opening
fresh avenues as to what type of antenna I will construct after the new
year.

Comments solicited from all........
and happy hollidays

Art

Roy,
Thanks for the pertinent points raised in your post.
I might add that since the element centers are close to the mast
I have avoided the installation of a boom at the top of the mast
as one would do to lower the take off angle by the use of a stacked
beam. This 'top beam' provides major instability when using a single
mast. With me using my normal fishing pole elements the arrangement
will help quite a bit in mast mechanical terms. Since it is element
height over the feed element that, in the main, is creating a
compressed lobe (as I believe your post infers) I will take a different
tac with respect to the driven element, where it will be the only
element shaped in parabolic form and all others will
remain in a "V" configuration( straight portion for clamping at the
center). The parabolic shaped driver would then hopefully allow the
addition of staggered elements for other bands. The mast used will be
a 2 inch diameter fibre glass to prevent vertical radiation mutations.
Regards
Art