On Sun, 11 Apr 2004 17:51:17 GMT, Irv Finkleman
wrote:
It's in the e-mail! Just fired it off now!

Damn!

I can see the problem now. The crazy thing is calibrated in MC. No
wonder it stopped resonating after the 60s.

Thanx, Irv, for the copies.

73's
Richard Clark, KB7QHC

I have a clear recollection of using tubular twin lead for FM broadcast
dipoles in the early 1950s. 73 Mac N8TT

--
J. Mc Laughlin - Michigan USA
Home:

"Richard Clark" wrote in message
...
On 10 Apr 2004 14:59:24 GMT,


This hard to accept given the timelines offered by those who have
reported the references. For one, the antenna was invented in the
late 40's and patented 26 Dec. 1950, and reported in 1960. ALL such
dates precede the introduction of tubular Twin Lead in the mid 60s.

73's
Richard Clark, KB7QHC

Cecil Moore wrote in message ...
N2EY wrote:
Dipole dimensions:
Innermost section (not a stub; shorted at both ends): 8'
Next section (stub; open at inner end): 6' 11" [resonant at 10 meters?]
Next section (stub, open at inner end): 13' 10" [resonant at 20 meters?]
Outermost section (stub, open at inner end): 27' 5" [resonant at 40 meters?]

Is this just an attempt at a trapped antenna using stubs for traps?

That's exactly what it looks like to me, Cecil. That's also how its
operation is explained in the article. Why a stub that is excited
along its length should behave as a trap is not explained, nor how the
shortened sections act as loading coils (the entire antenna is only
112' 4" long, plus connections).

Note that the antenna works as follows:

10 meters: 6'11" sections act as open circuits so that 8' sections
function as dipole.

20 meters: 13'10" sections act as open circuits so that 8' plus 6'11"
sections function as dipole.

40 meters: 27'5" sections act as open circuits so that 8' plus 6'11"
plus 13'10" sections function as dipole.

80 meters: Entire antenna functions as dipole.

15 meters: 27'5" sections act as open circuits so that 8' plus 6'11"
plus 13'10" sections function as dipole on 3/2 wavelength resonance.

It doesn't seem to model out to be very functional.

Modeling the thing looks like a real challenge for a whole bunch of
reasons. The gaps between sections may be important, for example.

The outermost
stub on 40m needs to have a very high impedance, i.e. 1/4WL shorted.
That works well to resonate the vertical on 40m, but 1/8WL on 80m
makes the antenna resonant at 3 MHz according to EZNEC.

Lattin's results prove that it is *possible*, so your model obviously
needs work to agree with physical reality. But whether the Lattin
antenna is worth all the effort and mechanical troubles is another
story.

The antenna described is a dipole rather than a vertical, but the same
principles apply.

I suspect that a key element to the antenna's operation is the use of
the tubular Twin-Lead, with its 0.8 velocity factor. This stuff is
probably close to being unobtanium these days. Another point is the
extreme narrowness of resonance on most bands.

The W5GI antenna seems to be a variation on the Lattin theme.

For all that trouble, it seems to me that a better choice (if you want
direct coax feed on the non-WARC HF bands) is the classic W3DZZ trap
dipole. With only two traps and mechanically robust construction it is
possible to achieve direct coax feed and low SWR on 80/40/20/15/10,
and the cut-and-try is much easier.

73 de Jim, N2EY

N2EY wrote:
The W5GI antenna seems to be a variation on the Lattin theme.

He says nobody has been able to model or explain why the
antenna works. KA8NCR over on qrz.com says it's lossy
on receive compared to a resonant dipole and about the
same as a trap dipole. I think this is the thread:

http://www.qrz.com/cgi-bin/ikonboard...=6015 8;st=10

Looking at the W5GI antenna, it would seem that reflections
from the ends of the antenna would flow towards the feedpoint
on the coax braid due to skin effect and encounter an open
circuit. If one sent a TDR pulse toward the end of the antenna
from the feedpoint, what do you suppose one would see coming
back?
--
73, Cecil http://www.qsl.net/w5dxp

On 19 Apr 2004 05:14:44 -0700, (N2EY) wrote:

Why a stub that is excited
along its length should behave as a trap is not explained, nor how the
shortened sections act as loading coils (the entire antenna is only
112' 4" long, plus connections).

Hi Jim,

Simply because the so-called explanations are more wish than reality.
I've not slackened my study of this design. I've done thirty or forty
measurements that have interest in their own right, but show the
"theory" is sheer fantasy. These notions of tuned stubs as traps are
so far off the mark in practice that no one notices they don't make
sense electrically either.


Modeling the thing looks like a real challenge for a whole bunch of
reasons. The gaps between sections may be important, for example.

....

I suspect that a key element to the antenna's operation is the use of
the tubular Twin-Lead, with its 0.8 velocity factor. This stuff is
probably close to being unobtanium these days. Another point is the
extreme narrowness of resonance on most bands.

These apologies ring false. The issue of gaps is desperate and the
selection of tubular has no basis in special characteristics. All
such considerations MUST yield to simple scaling. For instance, if
you need 0.8 and have 0.9, there is no magic formula beyond
proportions necessary to achieve "what should be."

The difficulty lies in the supposed explanations.

The W5GI antenna seems to be a variation on the Lattin theme.

And it is even more gauche. It approaches antenna design as an allied
art of Palmistry.

73's
Richard Clark, KB7QHC

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

"N2EY" == N2EY writes:

N2EY For all that trouble, it seems to me that a better choice (if
N2EY you want direct coax feed on the non-WARC HF bands) is the
N2EY classic W3DZZ trap dipole. With only two traps and mechanically
N2EY robust construction it is possible to achieve direct coax feed
N2EY and low SWR on 80/40/20/15/10, and the cut-and-try is much
N2EY easier.

Wow. This might work in my limited space, providing that it deals
well with the short height above ground. Now I just have to learn how
to make traps. Any recommended resources for this sort of thing?

N2EY 73 de Jim, N2EY

Jack.
(always looking for something new)
- --
Jack Twilley
jmt at twilley dot org
http colon slash slash www dot twilley dot org slash tilde jmt slash
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (FreeBSD)

iD8DBQFAhBqiGPFSfAB/ezgRAnVwAJ9zNi7yN9R7ItWYFOy7qPgUwCU5eQCg9vNN
GOthFgo/2rYNy6izXP1vWEg=
=1Fwb
-----END PGP SIGNATURE-----

Richard Clark wrote in message . ..
On 19 Apr 2004 05:14:44 -0700, (N2EY) wrote:

Why a stub that is excited
along its length should behave as a trap is not explained, nor how the
shortened sections act as loading coils (the entire antenna is only
112' 4" long, plus connections).

Hi Jim,

Simply because the so-called explanations are more wish than reality.
I've not slackened my study of this design. I've done thirty or forty
measurements that have interest in their own right, but show the
"theory" is sheer fantasy. These notions of tuned stubs as traps are
so far off the mark in practice that no one notices they don't make
sense electrically either.
Modeling the thing looks like a real challenge for a whole bunch of
reasons. The gaps between sections may be important, for example.

...

I suspect that a key element to the antenna's operation is the use of
the tubular Twin-Lead, with its 0.8 velocity factor. This stuff is
probably close to being unobtanium these days. Another point is the
extreme narrowness of resonance on most bands.

These apologies ring false. The issue of gaps is desperate and the
selection of tubular has no basis in special characteristics. All
such considerations MUST yield to simple scaling. For instance, if
you need 0.8 and have 0.9, there is no magic formula beyond
proportions necessary to achieve "what should be."

I disagree! In the QST article, Lattin describes an 80/40 dipole using
his method. It has wires dangling from the stub junctions to get 40
meter resonance.

The difficulty lies in the supposed explanations.

Agreed. They are oversimplified at best and just plain wrong at worst.
However, I take it as a given that W4JRW got the results he claimed
from the antennas he built.

The fact that people report so much trouble duplicating and modeling
the W4JRW antennas indicates to me that there is more to it than meets
the eye - and maybe more than met W4JRW's eye, as well.

Most important to me is that the antenna offers no real advantages
over, say, a conventional trap dipole. Yet it offers many
disadvantages, such as mechanical frailty and difficulty of
duplication.


The W5GI antenna seems to be a variation on the Lattin theme.

And it is even more gauche. It approaches antenna design as an allied
art of Palmistry.

HAW!

And again - what advantages does it have over, say, a W3DZZ trap
dipole?

73 de Jim, N2EY

Jack Twilley wrote in message ...
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

"N2EY" == N2EY writes:

N2EY For all that trouble, it seems to me that a better choice (if
N2EY you want direct coax feed on the non-WARC HF bands) is the
N2EY classic W3DZZ trap dipole. With only two traps and mechanically
N2EY robust construction it is possible to achieve direct coax feed
N2EY and low SWR on 80/40/20/15/10, and the cut-and-try is much
N2EY easier.

Wow. This might work in my limited space, providing that it deals
well with the short height above ground.

How short?

There's nothing magic about a trap dipole of any flavor, be it W3DZZ
or W9INN or whatever. At their very best they're *almost as good* as a
plain vanilla half-wave coax-fed dipole. Their main advantages a

- they can be made mechanically rugged

- they can be directly fed with coax and yield reasonable SWR

- they can be built and adjusted with relatively simple tools and
materials

Now I just have to learn how
to make traps. Any recommended resources for this sort of thing?

W4RNL's site, of course. Some time back I did posts describing
coaxial-cable traps which you can google up.

The main trouble with trap construction is the materials. Ideally we
would use fixed vacuum capacitors and big Miniductor-like coils of
heavy wire or tubing. In practice we often settle for something
less...

1) Conventional LC traps - These are usually made from transmitting
capacitors like the 850 series and Miniductor coils. An alternative to
the Miniductor is to wind heavy wire on a piece of PVC pipe, which can
also serve as an insulator. Main problems with this construction is
getting theparts and weatherproofing.

2) Coaxial-cable traps - These are made from a piece of coax (RG-58 is
probably best) wound on a piece of PVC. Articles by N4UU (QST December
1984, IIRC) and W8NX describe various flavors of coaxial-cable traps,
as do the posts I made here some years back. The main problem with
this method is that the results are highly variable, depending on the
coax used. Depending on the loss of the outer coax jacket, one builder
may have much more loss in a trap than another. Adjusting these traps
isn't easy. Their main advantages are ruggedness and
low cost.

3) Bifilar or "twintraps" - These are similar to coax-cable traps, but
are made by bifilar-winding a pair of wires and connecting the ends
series-aiding. The internal distributed capacitance makes the C and
gthe coil is the L. If wire with low loss insulation is used, these
should be better than coax-cable traps.

Note that not all trap dipoles are "W3DZZ" trap dipoles. That
designation belongs to a specific design, which works as follows:

Each dipole half consists of an outer section (about 22 feet), a
parallel LC trap resonant on 40 meters, and an inner section (about 33
feet) which is also resonant on 40 meters. Direct coax feed or 1:1
balun.

The "trick" is in the LC ratio of the traps.

On 80 meters, the traps are below resonance, and act as loading coils
so that the ~110 foot long antenna is resonant on 80

On 40 meters, the traps resonate and are essentially open circuits.
Thus the middle 66 feet operates as a 40 meter dipole.

On 20 meters, the traps are above resonance and act as capacitors so
that the entire antenna is 3/2 wave resonant.

On 15 meters, the traps are above resonance and act as capacitors so
that the entire antenna is 5/2 wave resonant.

On 10 meters, the traps are above resonance and act as capacitors so
that the entire antenna is 7/2 wave resonant.

The problem is that there are 5 resonance points to hit but only 3
variables to play with (inner section length, outer section length, LC
ratio of trap). So in real life the antenna can only be truly
optimized on 3 bands unless you are very very lucky, or use methods
such as extra "dangle wires" to establish resonance on certain bands,
as is sometimes recommended for getting a 40 meter dipole to work on
15.

If I had my druthers, I'd just use a plain dipole fed with ladder line
(the real stuff, not "Twin Lead with holes" and a true balanced tuner.
But mechanical considerations at my present QTH make coax feed a must,
so I make do with a homebrew trapper.


73 de Jim, N2EY

On 20 Apr 2004 09:06:35 -0700, (N2EY) wrote:
These apologies ring false. The issue of gaps is desperate and the
selection of tubular has no basis in special characteristics. All
such considerations MUST yield to simple scaling. For instance, if
you need 0.8 and have 0.9, there is no magic formula beyond
proportions necessary to achieve "what should be."

I disagree! In the QST article, Lattin describes an 80/40 dipole using
his method. It has wires dangling from the stub junctions to get 40
meter resonance.

Hi Jim,

OK, you disagree, but with what? Scaling will ALWAYS answer
everything but the mystical apologies.


Most important to me is that the antenna offers no real advantages
over, say, a conventional trap dipole. Yet it offers many
disadvantages, such as mechanical frailty and difficulty of
duplication.

This is more pilot error than design error (which has its own
problems, of course).

And again - what advantages does it have over, say, a W3DZZ trap
dipole?

I see no such issues if the theory were hammered out. It is plainly
these readings of tea leaves that frustrate construction, because when
a design is described, it is most clear and concise - it just doesn't
work is all.

Like I said, I've done some measures and added a dozen more since.
The results are interesting. I can come up with a four band antenna
without too much trouble; however, getting those bands into Ham
regions (all of them) is another matter. I can do this with a simple
run of twin lead, and one strategically placed short between them.

This antenna (usefully resonant or otherwise) is no worse than any
wire strung between poles - just two wires instead of one, hardly what
I would call fragile. If it has an advantage over your W3DZZ trap
dipole, I leave that strictly in the eye of the beholder as I have
full faith it won't be any worse.

Any way, such work offers a step towards an antenna with MORE gain
(and more wire, a third one) by constructing a Franklin Array style of
antenna. True, not a multi bander, but I am not particularly nailed
to the floor over that.

73's
Richard Clark, KB7QHC

Richard Clark wrote in message . ..
On 20 Apr 2004 09:06:35 -0700, (N2EY) wrote:
These apologies ring false. The issue of gaps is desperate and the
selection of tubular has no basis in special characteristics. All
such considerations MUST yield to simple scaling. For instance, if
you need 0.8 and have 0.9, there is no magic formula beyond
proportions necessary to achieve "what should be."

I disagree! In the QST article, Lattin describes an 80/40 dipole using
his method. It has wires dangling from the stub junctions to get 40
meter resonance.

Hi Jim,

OK, you disagree, but with what?

With the idea that scaling answers all questions.

Scaling will ALWAYS answer
everything but the mystical apologies.

The two-band 80/40 dipole in the QST article has extra wires at the
junctions of the 80 and 40 sections because (according to the author)
the velocity factor of the tubular Twin Lead makes it necessary. Those
wires might or might not be required with a unity velocity factor.

Most important to me is that the antenna offers no real advantages
over, say, a conventional trap dipole. Yet it offers many
disadvantages, such as mechanical frailty and difficulty of
duplication.

This is more pilot error than design error (which has its own
problems, of course).

In a perfect world, maybe. But in the real world of ham radio, most
hams have limited materials, test equipment, time and space. An
antenna made out of unobtainable materials, which requires
unobtainable tools and test equipment to build and adjust is only of
academic interest to a ham.

And again - what advantages does it have over, say, a W3DZZ trap
dipole?

I see no such issues if the theory were hammered out. It is plainly
these readings of tea leaves that frustrate construction, because when
a design is described, it is most clear and concise - it just doesn't
work is all.

Exactly! If it cannot be easily duplicated by a ham with typical
resources, what good is it?


Like I said, I've done some measures and added a dozen more since.
The results are interesting. I can come up with a four band antenna
without too much trouble; however, getting those bands into Ham
regions (all of them) is another matter. I can do this with a simple
run of twin lead, and one strategically placed short between them.

Which is not what Lattin did at all. Your design sounds far superior.
Is it on the web anywhere?

This antenna (usefully resonant or otherwise) is no worse than any
wire strung between poles - just two wires instead of one, hardly what
I would call fragile. If it has an advantage over your W3DZZ trap
dipole, I leave that strictly in the eye of the beholder as I have
full faith it won't be any worse.

All depends on the wires. I use recycled #12 house wire, which stands
up under ice loading and high winds here in EPA. Yet it is hardly
noticed by the neighbors.

Any way, such work offers a step towards an antenna with MORE gain
(and more wire, a third one) by constructing a Franklin Array style of
antenna. True, not a multi bander, but I am not particularly nailed
to the floor over that.

The main attraction of the Lattin is its claim to multiband operation.
Otherwise one might as well go with a plain dipole.

73 de Jim, N2EY