Lattin antenna.............more info sources
Lee Carkenord
understand and build a Lattin. As I said in my previous post, I was
never able to get it to work as Mr. Lattin described. I wish that
when I tried to build it, I would of had access to todays "Antenna
Analyzer" test instruments.
It has been patented, filed Feb 13, 1948. Patent # 2,535,298 My
local library had the patent on microfiche, and I was able to get it
copied. I have a copy of his patent, with illustrations/drawings, on
file here.
Lattin described this antenna in QST in December of 1960.
July 1960 issue of "CQ" mag has a related antenna project, by
Richmond, W1CEJ, called "A Portable Dipole"
Vincent Lear, G3TKN has also worked with the Lattin design, and was
published in RSGB.
July 1992 issue of "73" mag has another article "Six-band linear trap
antenna" by Brumbaugh KB4ZGC..........using approx the same idea.
September 1986 issue of "73" mag has Lattin type antenna project
described...."The Texas Tango Antenna" by Garrett K5BTV.
Maybe the above info will stimulate some more experimentation and/or
modelling of the
Lattin. I sure hope so.
Lee Carkenord KA0FPJ
N2EY
> The following is some of the info I gathered in my attempts to
> understand and build a Lattin. As I said in my previous post, I was
> never able to get it to work as Mr. Lattin described. I wish that
> when I tried to build it, I would of had access to todays "Antenna
> Analyzer" test instruments.
>
> It has been patented, filed Feb 13, 1948. Patent # 2,535,298 My
> local library had the patent on microfiche, and I was able to get it
> copied. I have a copy of his patent, with illustrations/drawings, on
> file here.
>
> Lattin described this antenna in QST in December of 1960.
>
> July 1960 issue of "CQ" mag has a related antenna project, by
> Richmond, W1CEJ, called "A Portable Dipole"
>
> Vincent Lear, G3TKN has also worked with the Lattin design, and was
> published in RSGB.
>
> July 1992 issue of "73" mag has another article "Six-band linear trap
> antenna" by Brumbaugh KB4ZGC..........using approx the same idea.
>
> September 1986 issue of "73" mag has Lattin type antenna project
> described...."The Texas Tango Antenna" by Garrett K5BTV.
See also:
>
> Maybe the above info will stimulate some more experimentation and/or
> modelling of the
> Lattin. I sure hope so.
>
It seems to me that there are two major issues with the Lattin/W4JRW
antenna:
1) Mechanically, it can be challenging. Regular Twin Lead with one of
the wires cut is just too flimsy. If you use spreaders and such it's
not simple any more. What's really needed is for somebody to make
"tubular" Twin Lead with, say #16 or even #14 copperweld conductors.
2) The velocity factor of the Twin Lead is a challenge to model
because the stub sections radiate on some frequencies but not others.
The thing is really just another form of trap antenna.
I get the feeling that Lattin either went through serious math pushups
to get his design, or cut up a lot of Twin Lead, or both.
73 de Jim, N2EY
Richard Clark
>I get the feeling that Lattin either went through serious math pushups
>to get his design, or cut up a lot of Twin Lead, or both.
Hi Jim,
I am working on the design, however the premise appears to be
"academic." In other words, there are some who can tell you how it
should work, but not why it doesn't.
Your description above suits the modeling to a T - serious math
pushups. I don't think anyone could stumble onto a 5 band antenna by
simple cut and try.
73's
Richard Clark, KB7QHC
Lee Carkenord
> antenna:> 1) Mechanically, it can be challenging. Regular Twin Lead with one of
> the wires cut is just too flimsy.
When I tried to build a working version of the Lattin, I had 2
partners. The 3 of us, collectively, have a decent amount of
experience with home-brew antennas.
Our first attempt was done using flat TV twin-lead. It was fragile
and hard to work with, mechanically. Electrically, it was just
confusing. Unpredictable. I never felt that we were close to getting
a good workable multi-band antenna.
Some time passed, and we decided to try again. This time we used the
much sturdier ladder line (for traps/radiator) as a starting point.
Mechanically, this was a fairly robust device. Electrically, it was
the same as our previous attempt. We spent a lot of time with it.
Cut up a lot of wire. We finally reluctantly gave up. We _WANTED_
it to work, but we were just not able to get anything accomplished.
Again, we felt that we were never close to having a workable, 50 ohm
coax-fed, multi-band antenna.
Lee Carkenord KA0FPJ Denver CO
N2EY
<kb7...@comcast.net> writes:
>On 9 Apr 2004 13:44:16 -0700, N2...@AOL.COM (N2EY) wrote:
>>I get the feeling that Lattin either went through serious math pushups
>>to get his design, or cut up a lot of Twin Lead, or both.
>
>Hi Jim,
>
>I am working on the design, however the premise appears to be
>"academic." In other words, there are some who can tell you how it
>should work, but not why it doesn't.
I *like* that turn of phrase!
"There are some who can tell you how it should work, but not why it doesn't."
Perfect companion to:
"If it happens, it must be possible."
>Your description above suits the modeling to a T - serious math
>pushups.
Thanks.
>I don't think anyone could stumble onto a 5 band antenna by
>simple cut and try.
I disagree! I think it not only could happen, but probably has happened
already, through a fortuitous combination of many factors.
But being able to come up with such a design that is well-documented and
reproducible is a whole 'nother thing. The Lattin antenna is a perfect example
of that.
73 de Jim, N2EY
N2EY
cark...@juno.com (Lee Carkenord) writes:
That's quite understandable, given that the original Lattin designed used
tubular Twin Lead and countd on a velocity factor of 0.8.
What sort of test gear and design methods did you folks use?
73 de Jim, N2EY
Richard Clark
>>I don't think anyone could stumble onto a 5 band antenna by
>>simple cut and try.
>I disagree! I think it not only could happen, but probably has happened
>already, through a fortuitous combination of many factors.
Name one that works. And by works, is resonant in each band, and not
simply tuneable (as would be a common doublet).
>But being able to come up with such a design that is well-documented and
>reproducible is a whole 'nother thing. The Lattin antenna is a perfect example
>of that.
Hi Jim,
The notion of a trapped antenna on the basis of resonant stubs
constructions is not shown in the data of my work to date, and
certainly not in the Lattin (insofar as the only interpretation
generally available on the net, setting the patent aside that is).
One of the key points goes to this notion of stub action. However,
the stub is not excited across its mouth, but along its length. This
is very distinctly exhibited in the numbers (the lack of correlation
of stub geometry to resonances). The constructions merely appear to
fatten a thin radiator and add capacitances and inductances that are
basically opportunistic - certainly no one has shown any correlations
that fit the geometries to the bands they are presumed to resonate to.
Hence my statement that caught your fancy. The Stub resonances
"should" explain the bands obtainable, and yet there has been
absolutely no supporting evidence to demonstrate that this occurs.
It does hold my interest, however, and I am hardly one to be put off
by failure's of other's theories - not with my more than 300 pages of
fractal data published in the face of fractal fools who confine
themselves to bragging about their science.
Richard Clark
>tubular Twin Lead and countd on a velocity factor of 0.8.
>
>What sort of test gear and design methods did you folks use?
Hi Jim,
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.
N2EY
<kb7...@comcast.net> writes:
>On 10 Apr 2004 14:59:24 GMT, n2...@aol.comSPAMNO (N2EY) wrote:
>>>I don't think anyone could stumble onto a 5 band antenna by
>>>simple cut and try.
>>I disagree! I think it not only could happen, but probably has happened
>>already, through a fortuitous combination of many factors.
>
>Name one that works. And by works, is resonant in each band, and not
>simply tuneable (as would be a common doublet).
Some forms of trap dipole and parallel dipole I have encountered were clearly
the result of cut-and-try rather than analysis and mathematical design.
>
>>But being able to come up with such a design that is well-documented and
>>reproducible is a whole 'nother thing. The Lattin antenna is a perfect
>>example of that.
>
>Hi Jim,
>
>The notion of a trapped antenna on the basis of resonant stubs
>constructions is not shown in the data of my work to date, and
>certainly not in the Lattin (insofar as the only interpretation
>generally available on the net, setting the patent aside that is).
The title of the QST article (December, 1960) is "Multiband Antennas Using
Decoupling Stubs".
>
>One of the key points goes to this notion of stub action. However,
>the stub is not excited across its mouth, but along its length. This
>is very distinctly exhibited in the numbers (the lack of correlation
>of stub geometry to resonances). The constructions merely appear to
>fatten a thin radiator and add capacitances and inductances that are
>basically opportunistic - certainly no one has shown any correlations
>that fit the geometries to the bands they are presumed to resonate to.
>
>Hence my statement that caught your fancy. The Stub resonances
>"should" explain the bands obtainable, and yet there has been
>absolutely no supporting evidence to demonstrate that this occurs.
>
The claims of the QST article are that the stubs work as traps. It also
explains that the velocity factor is important in the whole design. The article
refers repeatedly to "tubular Twin Lead" as the optimum material for
construction because of its velocity factor of 0.8.
>It does hold my interest, however, and I am hardly one to be put off
>by failure's of other's theories - not with my more than 300 pages of
>fractal data published in the face of fractal fools who confine
>themselves to bragging about their science.
>
(grin)
73 de Jim, N2EY
Lee Carkenord
> tubular Twin Lead and countd on a velocity factor of 0.8.
Yes...we were aware of the differences in velocity factor. And now I
have dug up more notes from those experiments. I had forgotten about
a _THIRD_ atempt!! We also tried using heavy 2-conductor speaker
wire as trap/radiator material. This was also a sturdy design,
mechanically. And again, we took into account the different v.
factor.
But it didn't produce the antenna we had hoped for.
>>> What sort of test gear and design methods did you folks use?
Pretty fundamental test equipment. The ordinary items that are in
just about all shacks. We did have a good SWR meter.......a Siltronix
FS-310. We also had a good grid dipper, the military AN-PRM/10 unit.
There was also a Palomar Noise bridge in my possession during that
era, but I don't see anything in my notes that indicate that we used
it in our Lattin experiments.
We cut it longer, and we cut it shorter. We diddled with the traps
_IN_ the circuit, and we diddled with 'em removed from the circuit.
We used up a lot of wire, and a lot of notebook paper.
This was back in 1990 and 91. I still wonder what we would have
accomplished if we had owned an Antenna Analyzer? And what if 1 or
more of our trio was fleunt in antenna computer modelling? Would we
have "cracked the nut"......or would we have just wasted more wire,
paper, and time? We never once felt that we were close to getting a
good multiband antenna. We sure tried, though. It was kinda fun,
really........
N2EY
<kb7...@comcast.net> writes:
Hello again!
Direct quotes from the 1960 QST article:
"If open line with a velocity factor near unity is used for the stubs, the
over-all lenght for a two-band antenna would be nearly a full free-space
wavelength at the higher frequency and and the whole antenna would resonate at
something less than half that frequency. Very fortunately, the the velocity
factor of 300-ohm tubular Twin-Lead (0.8) gives such lenghts for the stubs that
that, in most cases, the adding the stub makes the antenna resonate at just
half the original frequency."
"Fig. 3 shows how tubular Twin-Lead can be used for the antenna itself as well
as the stubs and includes dimensions for 10- and 20-meter operation. The
foam-filled type of Twin-Lead is recommended to keep out moisture."
So it must have existed when W4JRW wrote the article, some time before the
December 1960 QST went to press.
In addition, my 1953-54 RESCO (Radio Electronic Service Company, once a great
source of all things radio and TV here in the Philly area) catalog lists
Amphenol "flat and tubular Twin Lead". The tubular stuff is number 14-271 and
was 5 cents a foot - less in quantities of 100, 500 or 1000 feet. And it wasn't
a new item.
All of this supports both your experiences and those of others. If the exact
characteristics of tubular Twin Lead are important to the design, it's no
wonder that the reported results cannot be obtained.
One more point. The SWR graphs shown for the 5 band version show quite narrow
SWR bandwidths. If anyone is interested I can report them in a future post.
73 de Jim, N2EY
Cecil Moore
> "Fig. 3 shows how tubular Twin-Lead can be used for the antenna itself as well
> as the stubs and includes dimensions for 10- and 20-meter operation. The
> foam-filled type of Twin-Lead is recommended to keep out moisture."
Jim, I haven't been following very closely. Is it possible to describe
this antenna with an ASCII diagram. Apparently on one band, the stub
is installed at the 1/50 conductance circle to SWR circle intersection.
--
73, Cecil, W5DXP
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----
Cecil Moore
> Jim, I haven't been following very closely. Is it possible to describe
> this antenna with an ASCII diagram. Apparently on one band, the stub
> is installed at the 1/50 conductance circle to SWR circle intersection.
Never mind, I found the information at:
http://www.g3ycc.karoo.net/lattin.htm
Richard Clark
>All of this supports both your experiences and those of others. If the exact
>characteristics of tubular Twin Lead are important to the design, it's no
>wonder that the reported results cannot be obtained.
>
>One more point. The SWR graphs shown for the 5 band version show quite narrow
>SWR bandwidths. If anyone is interested I can report them in a future post.
Hi Jim,
I stand corrected on the tubular twin lead (must've died in the
mid-60s then).
It offers nothing substantially different to the physics of design
however, just a different velocity factor which is scalable to suit
any implementation.
However, my data to this point confounds the "theory" of it all by
exhibiting stubs that do NOT resonate at their computed length
(complete with correction for ANY velocity factor commonly observed).
More data is always appreciated.
N2EY
<kb7...@comcast.net> writes:
>On 10 Apr 2004 18:25:04 GMT, n2...@aol.comSPAMNO (N2EY) wrote:
>>All of this supports both your experiences and those of others. If the exact
>>characteristics of tubular Twin Lead are important to the design, it's no
>>wonder that the reported results cannot be obtained.
>>
>>One more point. The SWR graphs shown for the 5 band version show quite
>narrow
>>SWR bandwidths. If anyone is interested I can report them in a future post.
>
>Hi Jim,
>
>I stand corrected on the tubular twin lead (must've died in the
>mid-60s then).
No biggie. I did not expect to find it in the 1953 catalog, but there it was,
and not a new item either.
Probably unobtanium now.
>
>It offers nothing substantially different to the physics of design
>however, just a different velocity factor which is scalable to suit
>any implementation.
>
Lattin seemed to think it was important. Do you have the 1960 QST article?
>However, my data to this point confounds the "theory" of it all by
>exhibiting stubs that do NOT resonate at their computed length
>(complete with correction for ANY velocity factor commonly observed).
>
>More data is always appreciated.
>
Well, here's the info from the 1960 QST article.
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?]
Entire antenna 57' 2" per side (plus connections)
SWR, 50 ohm coax, coax length not given (all numbers guesstimated from graph):
80 meters:
Minimum SWR: 3725 kHz - 1.1:1
2:1 SWR points: 3625 & 3800 kHz (175 kHz)
SWR at 3500: 5:1
SWR at 4000: 4.5:1
40 meters:
Minimum SWR: 7225 kHz - 1.6:1
2:1 SWR points: 7150 & 7275 kHz (125 kHz)
SWR at 7000: Off end of scale (5:1 at 7075 kHz)
SWR at 7300: 3:1
20 meters:
Minimum SWR: 14250 kHz - 1.3:1
2:1 SWR points: 14100 & 14375 kHz (275 kHz)
SWR at 14000: 3.5:1
SWR at 14350: 1.8:1
10 meters:
Minimum SWR: 28600 - 1.6:1
2:1 SWR points: 28500 & 28750 kHz (250 kHz)
SWR at 28900: 3:1
Chart for 10 meters covers 28500 to 28900 only. No chart for 15 meters but text
says it will work there on 3/2 wavelength resonance of 40 meter section and SWR
of not less than 3:1.
Note how tight the 2:1 SWR points are, even on the higher bands.
--
Also described is an 80/40 dipole. Inner section is 28' and shorted both ends,
outer stub section is 27' 5". Interesting feature of this one is that in order
to achieve 40 meter resonance there are pieces of wire 2' 6" long hung from the
junction of the two sections. The text and diagram say the outer section
resonates on 40 but the inner section has to be kept at 28 feet or the 80 meter
resonance will be too low.
--
Seems to me that the classic W3DZZ trap dipole would be a lot less aggravation
to model and get working, plus more flexible in choice of minimum SWR points.
Particularly if additional "resonance wires" were added if needed. Certainly
easier to make mechanically strong traps than stubs, and adjusting single wire
lengths with Burndys is a lot easier than fooling with tubular Twin-Lead.
73 de Jim, N2EY
Irv Finkleman
I have the original article here. If you want it I can scan it and
send it to those requesting.
This has become an interesting thread. I went through hell with mine,
using flat twin lead and not knowing the velocity factor. I had pieces of wire
hanging from the ends of each trap to try and 'trim' them. I also had the problem
of support -- strung a string through the line through holes poked in the twinlead
with a soldering pencil. It came close, but it was quite a while ago. I mentioned
Bob, VE6TE (now SK) who had the full model -- probably the tubular twin lead too --
he never threw anything away! The tubular was a precursor to foam filled (I would
imagine).
If I were ever to go near this antenna again, I'd build a two band model which
might be simpler, and of course now I have a MFJ-259B which might help a bit. It
lies rolled up in a corner of the shack gathering dust -- I had so much hell with it
that I'm reluctant to ever look at it again. One alternative would be to make a
vertical model operating against a good ground -- it would be simpler!
The only reason I built it was because of VE6TE -- and my desire to find the
'ultimate' cheap and simple multiband antenna! Dream on!
Irv VE6BP
--
--------------------------------------
Diagnosed Type II Diabetes March 5 2001
Beating it with diet and exercise!
297/215/210 (to be revised lower)
58"/43"(!)/44" (already lower too!)
--------------------------------------
Visit my HomePage at http://members.shaw.ca/finkirv/
Visit my very special website at http://members.shaw.ca/finkirv4/
Visit my CFSRS/CFIOG ONLINE OLDTIMERS website at http://members.shaw.ca/finkirv5/
--------------------
Irv Finkleman,
Grampa/Ex-Navy/Old Fart/Ham Radio VE6BP
Calgary, Alberta, Canada
Cecil Moore
> Innermost section (not a stub; shorted at both ends): 8'
No wonder it didn't work on EZNEC. The web page I referenced
had that as a stub.
Richard Clark
wrote:
>Hey Guys,
Hi Irv,
> I have the original article here. If you want it I can scan it and
>send it to those requesting.
Put me on your distribution list.
>
> If I were ever to go near this antenna again, I'd build a two band model which
>might be simpler, and of course now I have a MFJ-259B which might help a bit.
I've got two and three band models right now. Just not the right
bands is all.
>One alternative would be to make a
>vertical model operating against a good ground -- it would be simpler!
That's exactly how I start. It gives me a faster running model so I
can go through the iterations quicker.
Richard Clark
>Lattin seemed to think it was important. Do you have the 1960 QST article?
>
Hi Jim,
No, but hopefully Irv will rectify that.
>
>Well, here's the info from the 1960 QST article.
I will keep those numbers close by, thanx.
Cecil Moore
> 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?
It doesn't seem to model out to be very functional. 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.
Irv Finkleman
>
> On Sun, 11 Apr 2004 02:07:31 GMT, Irv Finkleman <fin...@shaw.ca>
> wrote:
>
> >Hey Guys,
>
> Hi Irv,
>
> > I have the original article here. If you want it I can scan it and
> >send it to those requesting.
>
> Put me on your distribution list.
It's in the e-mail! Just fired it off now!
Irv
Richard Clark
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.
J. McLaughlin
dipoles in the early 1950s. 73 Mac N8TT
--
J. Mc Laughlin - Michigan USA
Home: J...@Power-Net.Net
"Richard Clark" <kb7...@comcast.net> wrote in message
news:3l5g70t7l8an9p5io...@4ax.com...
> On 10 Apr 2004 14:59:24 GMT,
>
N2EY
> 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
Cecil Moore
> 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.cgi?s=4083e6f2738bffff;act=ST;f=5;t=60158;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
Richard Clark
>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.
Jack Twilley
Hash: SHA1
>>>>> "N2EY" == N2EY <N2...@aol.com> 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-----
N2EY
> On 19 Apr 2004 05:14:44 -0700, N2...@AOL.COM (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
N2EY
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> >>>>> "N2EY" == N2EY <N2...@aol.com> 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 are:
- 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
Richard Clark
>> 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.
N2EY
> On 20 Apr 2004 09:06:35 -0700, N2...@AOL.COM (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
Roy Lewallen
be a little trickier than meets the eye.
Consider, for example, scaling a piece of TV twinlead to twice the
frequency so it'll behave exactly the same (both as a transmission line
and as a radiator) at the new frequency.
The wire diameters have to be reduced by a factor of two.
The spacing between the wires has to be reduced by a factor of two.
Luckily, if the scale model and the original are both in free space, then
The dielectric constant of the insulator remains unchanged.
And, almost always overlooked,
The wire conductivity has to be increased by a factor of two.
The dielectric conductivity has to be increased by a factor of two.
Fortunately, these last factors are usually unimportant. If the original
is made from copper, it isn't possible to scale to a much higher
frequency. But it's something to be kept in mind if loss is significant
and an accurate assessment of loss is necessary.
Permeability, incidentally, remains unchanged with frequency when scaling.
But even if you can neglect the conductivity scaling, you wouldn't be
able to run down to the store and buy a piece of the scaled twinlead to
use in your antenna for another band.
(*) Antennas are often scaled to higher frequencies for testing because
the scale model is a more convenient size. When I was involved in the
development of very high-speed sampling circuits, we often made scale
models of various structures (for example, coax connector to microstrip
transitions) at *lower* frequencies, so they'd be large enough to
measure and physically adjust.
Roy Lewallen, W7EL
Richard Clark
>> 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?
Hi Jim,
You should take care with my perverse generalizations, especially when
you follow it with:
>The main attraction of the Lattin is its claim to multiband operation.
>Otherwise one might as well go with a plain dipole.
A plain dipole is already mutibanded. This particular claim is easy
to achieve. It may not fall within interesting bands, but history
already suggests one dipole often serves more than one ham band if one
can accept the short comings (wrong directionality being key).
This is more a matter of the dichotomy of need: match and launch.
There may be more matching options than actual application of where it
is launched (too many lobes in the wrong directions).
In this sense, the trap serves to shorten the antenna to RESTRICT its
physical to electrical wavelength ratio to a quarter or less, thus
guaranteeing a mediocre performance from precious air space and
real-estate. If Art hadn't been nailed for a design 17dB below a
simple vertical, he would be here howling efficiency per unit length
(but, unfortunately in the wrong aspect - that's how you lose 17dB
behind the sofa).
Frankly, the Lattin has yet to prove to me that the notion of a
longitudinal stub as "trap" really holds any water. I've seen the
same "theory" applied to diametrically opposed designs. It is
pleasing to the arm-chair designer to mutter these ideas, but these
so-called streamers needed to make it work just yell foul on every
street corner. It is an ad-hoc design draped with academic mumblings
to lend it the appearance of legitimacy.
To extend my quote above, I have added yet another dozen measurements
to have nailed down patterns that emerged with the basis of a
consistent building paradigm (yeah, I know, gobbldygook). I seek to
generalize such claims as the Lattin makes and reduce them to a
practical minimum that are robust and repeatable. This is not to say
entirely useful, nor optimal.
I did the same thing with the fractal with 300 or more measurements
and reduced that junk science to a simple observation: you can push
more resonances into a length of wire, the more you kink it. Useful?
The test of time has shown that no one has made any money from those
same published 300 pages - why would I expect the Lattin to emerge
from 5 decades of neglect to eclipse that record? Frankly, the Lattin
simply confirms this simple observation, but is more controlled. It
may mature to a more repeatable design, but I doubt its inventor would
recognize it.
David J Windisch
Here's some anecdata (sic) ... :
OUAT, after reading something some where about the Lattin Labs antenna, I
put an 8-ft stub on the side of a 33-ft verticaloverlotsaradials.
I got a "new" indication of low-Z near 10M which wasn't there B4, to go
along with the 40M low-Z reading.
Received noise on 10M went up, from a well-defined nearby source, suggesting
that the el pattern might have come down some.
I moved the stub nearer the top, and found what I remember as a 3/4-wave
low-Z indication on 10M.
I think the KT-XX series of "diametrically opposed designs" use this method
of multibanding, as well.
Might even try this anecdata in Roy's Toy3, to see if its gui-ed
algorithm-ized academic mumblings prove this scandalous anecdata.
Shucks, if it can't be modeled, it can't be made to happen ;o)
Right?
73, Dave, N3HE
"Richard Clark" <kb7...@comcast.net> wrote in message
news:mbog80l60r8kkgsbb...@4ax.com...
> On 22 Apr 2004 07:19:37 -0700, N2...@AOL.COM (N2EY) wrote:
>
> >> 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?
>
> Hi Jim,
>
> You should take care with my perverse generalizations, especially when
> you follow it with:
BIGSNIP
Richard Clark
<dav...@copper.net> wrote:
>Shucks, if it can't be modeled, it can't be made to happen ;o)
>
>Right?
Hi Dave,
More the issue is if it can be made to happen, and it can't be modeled
- what happened? [After that we can then stare at our navels and ask
"what is IT?"]