On Mar 17, 1:22 am, Ian White GM3SEK wrote:
Owen Duffy wrote:
"ve2pid" wrote in news:1174098010.745176.35010
:

Hi to all,

I read somewhere that, in the case of Yagis, ''in the range of 25-35
Ohm you get the
best balance between gain, pattern, bandwidth and element currents.'''

Is that true? And if so, I would like to have the theoretical
explanation behind this.

I don't understand why there is such a relationship, if there is.

I don't think there is such a relationship either. The feedpoint
impedance is part of the results from the design process, along with the
gain, pattern and beamwidth; but it does not determine any of those
other properties.

The so-called "best balance" between all these properties will depend
entirely on the personal priorities of the designer or the user.
(Antenna optimization software demands very clear instructions about
this, and forces you to think very hard about what you really mean by
"best".)

Hmmm. Interesting comments considering below.


Many good designs do have a feedpoint impedance in that 25-35 ohms
region; but so do some real lemons, too, so it isn't a reliable
indicator.

As Roy says, yagis with much lower feedpoint impedances tend to have
high internal (I^2*R) losses, which increase rapidly as the impedance
falls and element currents rise. Those can be classified as poor
designs, simply because there are plenty of better alternatives.

Also, it is possible with many designs to increase the feedpoint
impedance towards 50 ohms by adding a director at very close spacing
(about 0.05 wl). That director has relatively little effect on other
performance parameters, so it can be added fairly late in the design
process as a means of matching. (After construction, that close-spaced
director also allows final adjustment of the matching, by bending the
ends towards or away from the driven element.)

Yagis are often designed for a feedpoint impedance that is relatively
easily transformed to 50 ohms for the main transmission line.

A quarter wave transformer from 28 ohms to 50 ohms is relatively easily
made from two parallel 75 ohm lines. Perhaps that is the attraction to
the impedance range you mention.

That is a very reasonable strategy: if the impedance comes somewhere
close to a convenient value for matching, then optimize it to exactly
that value.

For example, DK7ZB has developed a range of yagi designs optimized for
28 ohms.

And there he says: "For the VHF-Bands (50-50.5MHz, 144-146MHz,
430-440MHz) a radiation resistance of 25-30Ohm has the best balance
for gain, back- and sidelobes, bandwidth and SWR at tenable losses."

We have come full-circle. [g]

Also, DK7ZB in describing what is clearly an unbalanced connection of
parallel lengths of coax ("Classic" match) says, "2. This line is a
simplified coaxial sleeve balun to avoid sleeve-waves on the braid of
the cable running to the station.", which it is clearly not.


There are links to these and several other designs from:http://www.ifwtech.co.uk/g3sek/diy-yagi/index.htm

--

73 from Ian GM3SEK 'In Practice' columnist for RadCom (RSGB)http://www.ifwtech.co.uk/g3sek

N7WS

Wes wrote:
A quarter wave transformer from 28 ohms to 50 ohms is relatively easily
made from two parallel 75 ohm lines. Perhaps that is the attraction to
the impedance range you mention.

That is a very reasonable strategy: if the impedance comes somewhere
close to a convenient value for matching, then optimize it to exactly
that value.

For example, DK7ZB has developed a range of yagi designs optimized for
28 ohms.

And there he says: "For the VHF-Bands (50-50.5MHz, 144-146MHz,
430-440MHz) a radiation resistance of 25-30Ohm has the best balance for
gain, back- and sidelobes, bandwidth and SWR at tenable losses."

We have come full-circle. [g]

Why, so we have...

At the risk of going round again, I definitely wouldn't agree with that
statement as written. However, it is valuable to point out that 28 ohms
is one of those "convenient" impedances, which might otherwise have been
overlooked.


Also, DK7ZB in describing what is clearly an unbalanced connection of
parallel lengths of coax ("Classic" match) says, "2. This line is a
simplified coaxial sleeve balun to avoid sleeve-waves on the braid of
the cable running to the station.", which it is clearly not.

Agreed. However, there is an alternative for 50MHz because the
paralleled quarter-wave sections are a convenient length to be coiled up
to make a small, neat choke.

I had a 3-element 50MHz beam which had been thrown together using
existing gamma match parts, but it was picking up a lot of noise and
crud - on surprising on thinking about it, because the coax shield, the
boom and the mast were all connected together and acting as an antenna
for local noise. Since the feedpoint impedance happened to be about 28
ohms, it was very easy to convert it to a fully balanced feed with a
choke, and all the noise problems went away.


There are links to these and several other designs from:http://www.ifwtech.co.uk/g3sek/diy-yagi/index.htm


--

73 from Ian GM3SEK 'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek

Ian White GM3SEK wrote:
I had a 3-element 50MHz beam which had been thrown together using
existing gamma match parts, but it was picking up a lot of noise and
crud - on surprising on thinking about it, because the coax shield, the
boom and the mast were all connected together and acting as an antenna
for local noise.

Sorry, made a t6po. That should read: "NOT surprising, on thinking
about it."


--

73 from Ian GM3SEK