And, consider: even then, It might not work as it does on another
frequency. Have a friend (this years ago), who modeled the Komosko-
Johnson yagi for 2 meters (a 13 el, antenna , that was emperically
designed), long before modeling programs, and then antenna range
measured. This was a staple of the old VHF handbooks, and the
ARRL handbooks for years, and seemed to deliver exceptional results.
He had pull at his local company (not to name them)! and had an
EXACT copy made for 1296, and then entered it into a antenna contest-
it failed misserably! (this not only was element length/Dia, spaceing,
,but BOOM diameter/length)!! It was (and is) a work of ART,
but it wont make his retirement check (or Contest Score)
any higher! For those that
wish to try it- look for those names, or for W2NLY-W6QKI in the
ARRL handbooks mentioned. But it must leave to the builder one
of the following conclusions: 1) It was great on 144, but a dog
on 1296, or: 2) It wasn't the greatest design , for even 144 (tho
it compairs favorable with the range patterns that were shown
in the VHF handbook, for 144)! As info, Jim NN7K

David wrote:
Ian,

Thanks for that.

I did previously look at your site and bookmarked it. I am going to make
up the element cutting jig and bending jig (for when I want to
experiment with folded dipoles).

Great site.

Regards

David

Ian White GM3SEK wrote:

David wrote:

Hi,

I have a design for a Yagi antenna at 436MHz and would like to scale
it to 921 MHz. The calculations I used from the understanding I have
obtained from reading various articles appears below.

Would someone be kind enough to advise me if this procedure and
values look correct ?


[...]


Regards

David Huisman




The basic principle of scaling antennas according to wavelength is
that you must make an *exact* scale model. That means scaling every
dimension, and not adding any new or different 'features'.

Often that isn't practical for mechanical engineering reasons, so then
you need to apply corrections. However, you should still aim to keep
those corrections as small as possible. For example, simple scaling is
likely to produce odd, unavailable values for the element diameter, so
it's OK to use corrections to change to the nearest commercially
available diameter. But if you also decide to double the diameter, the
correction is more risky - you're entering the territory of a
completely new design.

Sorry that I don't have time to check your particular dimensions in
detail, David, but you can find resources for scaling and
constructional techniques on my 'VHF/UHF Long Yagi Workshop' pages.
There is a downloadable program which leads you through the relevant
questions, and also a link to an online Javascript version.

(From about 1200utc I will not have access to this newsgroup for the
next few days, so I hope the above will keep you going.)

Jim,

Thanks for your input of caution.

I'm sure that along with all the horror stories there are a multitude of
success stories.

My belief is that antennas have been around for years and millions have
been constructed on almost every frequency imaginable.

In my case I am new to antenna design and green when it comes to
determining the parameters that would best suit my applications.

My requirements are not critical with regard to front/back ratio or the
last drop of gain available for a specific design.

I find there are always such a range of responses to a request.
Some of these are the theorists that love to explain how black holes in
the universe some 10 million light years away might effect free
electrons floating on the end of a copper wire.

Then we have the ones that are not particularly experienced but like to
have a say on every topic. They would normally ask why you want it in
the first place. "Oh, you want blue, why not have red".

Between these extremes are a bunch of very kind people who have "played"
with antennas and gained practical experience that they are prepared to
pass on to others.

Modeling programs have come a long way and my understanding is that as
long as you model the unit correctly, your actual results should not be
too far off. (I don't know how to model yet so hence the reason to ask
for help).

I imagine there are still some disasters when engineers build new cars
but I believe this would mainly occur when something is overlooked or
they are trying to achieve something new.

I am not trying to break any new ground here, just looking for around 7
dBi gain from a Yagi at 921MHz.

Regards

David Huisman


Jim - NN7K wrote:
And, consider: even then, It might not work as it does on another
frequency. Have a friend (this years ago), who modeled the Komosko-
Johnson yagi for 2 meters (a 13 el, antenna , that was emperically
designed), long before modeling programs, and then antenna range
measured. This was a staple of the old VHF handbooks, and the
ARRL handbooks for years, and seemed to deliver exceptional results.
He had pull at his local company (not to name them)! and had an
EXACT copy made for 1296, and then entered it into a antenna contest-
it failed misserably! (this not only was element length/Dia, spaceing,
,but BOOM diameter/length)!! It was (and is) a work of ART,
but it wont make his retirement check (or Contest Score)
any higher! For those that
wish to try it- look for those names, or for W2NLY-W6QKI in the
ARRL handbooks mentioned. But it must leave to the builder one
of the following conclusions: 1) It was great on 144, but a dog
on 1296, or: 2) It wasn't the greatest design , for even 144 (tho
it compairs favorable with the range patterns that were shown
in the VHF handbook, for 144)! As info, Jim NN7K

David wrote:

Ian,

Thanks for that.

I did previously look at your site and bookmarked it. I am going to
make up the element cutting jig and bending jig (for when I want to
experiment with folded dipoles).

Great site.

Regards

David

Ian White GM3SEK wrote:

David wrote:

Hi,

I have a design for a Yagi antenna at 436MHz and would like to scale
it to 921 MHz. The calculations I used from the understanding I have
obtained from reading various articles appears below.

Would someone be kind enough to advise me if this procedure and
values look correct ?



[...]


Regards

David Huisman





The basic principle of scaling antennas according to wavelength is
that you must make an *exact* scale model. That means scaling every
dimension, and not adding any new or different 'features'.

Often that isn't practical for mechanical engineering reasons, so
then you need to apply corrections. However, you should still aim to
keep those corrections as small as possible. For example, simple
scaling is likely to produce odd, unavailable values for the element
diameter, so it's OK to use corrections to change to the nearest
commercially available diameter. But if you also decide to double the
diameter, the correction is more risky - you're entering the
territory of a completely new design.

Sorry that I don't have time to check your particular dimensions in
detail, David, but you can find resources for scaling and
constructional techniques on my 'VHF/UHF Long Yagi Workshop' pages.
There is a downloadable program which leads you through the relevant
questions, and also a link to an online Javascript version.

(From about 1200utc I will not have access to this newsgroup for the
next few days, so I hope the above will keep you going.)

Jim - NN7K wrote:

And, consider: even then, It might not work as it does on another
frequency. Have a friend (this years ago), who modeled the Komosko-
Johnson yagi for 2 meters (a 13 el, antenna , that was emperically
snip
in the VHF handbook, for 144)! As info, Jim NN7K


When properly done, I have never had a scaled design fail to perform as
expected. I have done several dozen. The designs were 144, 220, 222,
and 432 as source and destination.

I don't expect this one will surprise me as to how it performs,
especially as it is insulated above the boom.

tom
K0TAR

That's what I expected from people who have practical experience and
understand the principles (much better than I current do).

I managed to track down thinner aluminum rod that is much closer to
scaled values of elements in the original design.

The values should be 3.78mm and I can get 3.175mm
For the 4.73mm , I can get 4.763mm.

This should now be close enough to only require small trim in length
to compensate for small change in diameter.

Tom Ring wrote:
Jim - NN7K wrote:

And, consider: even then, It might not work as it does on another
frequency. Have a friend (this years ago), who modeled the Komosko-
Johnson yagi for 2 meters (a 13 el, antenna , that was emperically

snip

in the VHF handbook, for 144)! As info, Jim NN7K


When properly done, I have never had a scaled design fail to perform as
expected. I have done several dozen. The designs were 144, 220, 222,
and 432 as source and destination.

I don't expect this one will surprise me as to how it performs,
especially as it is insulated above the boom.

tom
K0TAR

David wrote:

That's what I expected from people who have practical experience and
understand the principles (much better than I current do).

I managed to track down thinner aluminum rod that is much closer to
scaled values of elements in the original design.

The values should be 3.78mm and I can get 3.175mm
For the 4.73mm , I can get 4.763mm.

This should now be close enough to only require small trim in length
to compensate for small change in diameter.

The diameter ratio would be fine given the first diameter you listed.
The others would be fine also. Just specify what you want to use.

tom
K0TAR

Tom,

I have heaps of 6mm in stock. The smaller diameter has to be ordered
specially and is more expensive.

So:

Frequency 921 MHz. (20MHz BW or more would be good).

Gain = 7dBi

F/B ratio not critical.

4 Element Yagi - input impedance of around 29 Ohms (can easily match
with 2 par 75 Ohm 1/4 wave section).

Unit to be operated vertically orientated at height greater than 2m.

Element diameters 6mm

Boom is 12mm square aluminium. Elements mounted 3-5mm above boom on
nylon insulator block.

Just need accurate lengths and spacing now.

Tom Ring wrote:
David wrote:

That's what I expected from people who have practical experience and
understand the principles (much better than I current do).

I managed to track down thinner aluminum rod that is much closer to
scaled values of elements in the original design.

The values should be 3.78mm and I can get 3.175mm
For the 4.73mm , I can get 4.763mm.

This should now be close enough to only require small trim in length
to compensate for small change in diameter.


The diameter ratio would be fine given the first diameter you listed.
The others would be fine also. Just specify what you want to use.

tom
K0TAR

Jim - NN7K wrote:

wish to try it- look for those names, or for W2NLY-W6QKI in the
ARRL handbooks mentioned. But it must leave to the builder one
of the following conclusions: 1) It was great on 144, but a dog
on 1296, or: 2) It wasn't the greatest design , for even 144 (tho
it compairs favorable with the range patterns that were shown
in the VHF handbook, for 144)! As info, Jim NN7K

And, except for the K1FO designs, the VHF handbook contains one of the
worst collections of yagis ever assembled. They are at least as bad as
the Cushcraft VHF/UHF yagis that had the dual reflectors.