[KL7AA] Cage Dipole

[Jim Wiley]

Mike -


Well, first thing is, there ain't no free lunch.    I did some 
experiments with attempting to broaden the SWR bandwidth of various 
antennas, and while I had some success, I concluded that the gain just 
wasn't worth the effort.   There is a wonderful book out there that 
anyone who is interested in antennas should own, it is titled 
"Reflections".  The author is Walt Maxwell, W2DU.  There are three 
editions, the most recent one being published in 2010.  I strongly 
recommend that you obtain a copy of you don't already have one.   It is 
stocked at Amazon.com and may be available for free, or at little cost 
via the Internet.  Do a g Google search for  "Reflections" and "Maxwell" 
and you will find it immediately.


This book will truly open a person's eyes about the effect of SWR and 
most importantly, why *SWR doesn't matter* for many applications, 
particularly at HF.  What _does_ matter is being able to properly couple 
the transmitter (the source) the the antenna (the load).  Proper 
coupling (matching) allows the transmitter to deliver the maximum amount 
of energy to the antenna, regardless of what load impedance is presented 
by the antenna.  There is way too much information on this subject to be 
handled here, but the short version is that even if there is a high SWR 
in the system, a properly coupled system will radiate all the power if a 
match can be achieved.  In some situations there will be some degree of 
additional loss caused by the signal "bouncing" up and down the 
transmission line, but at HF, these losses are generally negligible.


A correct match can be accomplished with ease by any competent antenna 
coupler, and several are available to the ham market.  Some of the best 
ones are the Drake "MN" series, the MN-2000 and MN-2700 being my 
favorites.  The Drake couplers use a design that is basically a low-pass 
"reversed PI network" that helps with harmonic reduction.  The Drake 
tuners are no longer made, but are readily available on the used 
market.  Most other designs, such as the ones from MFJ, do the matching 
chore well,. but are "high-pass" T-network designs,  and these 
inherently do not reduce harmonic energy.  In these days of solid-state 
transceivers and transmitters, an antenna coupler is almost a necessity 
in any case - in fact many rigs come with one built in.


Remember this one truth if you take way nothing else:  *"There is no 
requirement whatsoever that an antenna must be 'resonant' to be an 
efficient radiator."*    This is singularly most important fact about 
antennas that a person needs to keep in mind.  It is indeed true that 
the antenna must be "matched" if the transmitter is to deliver all of 
its power,  but there is nothing anywhere that says the antenna itself 
must be resonant at the operating frequency.  Keep in mind that this 
applies to "simple" antennas.  Certain antenna designs, such as the 
Yagi, depend on the elements being self-resonant to achieve "gain", but 
even then there are wide tolerances allowed.


Consider this - in the early days of ham radio, most operators used 
antennas that were essentially a doublet (two wires, center fed, often 
mistakenly referred to as a dipole, but that's another discussion), or 
an end-fed wire of some (not specified) length, fed with open-wire 
feeders.  In those days, there was no such thing as coaxial cable (or 
SWR meters).  Somehow they managed to get on the air and make hundreds 
of thousands of contacts, even with systems exhibiting SWRs that could 
range upwards of 50:1 - that's right, FIFTY to ONE!  Open wire feeders 
are VERY low loss, and that means, in simplified terms,  that a signal 
can reflect up and down the line until eventually it all  gets radiated.


The purpose of the antenna coupler it to provide something called a 
"conjugate match", which is a fancy way of saying that the coupler + 
transmitter combination appear as a perfect "one-way mirror" to radio 
signals, allowing signals to flow freely from the transmitter toward the 
antenna, but perfectly reflecting any signal energy that comes back from 
a  distant mis-matched load right back up the feeder for another try.  
Bounce a wave up and down the line a few times, radiating some portion 
at each round trip, and pretty soon it is all heading out to the distant 
station. There are other factors involved of course, but this simplified 
illustration will give you some idea of the process.


A coaxial cable feedline has more loss than an open-wire feeder, but at 
HF, the losses are still essentially negligible if the SWR is below 
perhaps 5 or 6 to one, and again, open-wire is a viable option.  Now, it 
is true that high-SWR operation does cause additional stress to the 
line, in the form of increased peak voltages and currents at certain 
places along the line, but unless you are running more than 1.5 KW, 
standard RG-8/U type cables will not have a problem, and open-wire wont 
have an issue in any case.   This also means that you should avoid using 
small-diameter cables, such as RG-58/U, at power levers above 100 watts 
if you are also planning on having some SWR that will be "tuned out" 
with an antenna coupler.


An absolutely fabulous, and simple, antenna that gives good performance 
across several bands is the square-loop antenna.  If such an antenna is 
constructed so that it is 1-wavelength around the loop at the lowest 
frequency of interest, for example 62 feet on a side, 248 feet around 
the loop,  for 80 and 75-meter operation, and feed with a combination of 
open-wire feed at one corner and an antenna coupler, it will be usable 
on all bands 80 through 10 meters, and give good results on all bands.   
Such an antenna need only be 15 to 30 feet high, and can be supported by 
ropes thrown over nearby trees.   Now, I agree that  such an antenna 
will not be as effective as a multi-element Yagi "beam" antenna on a 
100-foot tower on the higher bands, no question about that - but it will 
work, and you can work lots of DX with such an antenna, and it's a LOT 
cheaper to construct.   The full-wave  loop mounted low is an effective 
NVIS (Near Vertical Incidence Skywave) antenna on 80 and 40 meters, and 
as such is particularly effective for local HF nets, such as the 
Sniper's Net, the Motley Group, or the Bush Net.   All the truly BIG 
signals on those nets come from stations using horizontal loops.   The 
horizontal loop antenna is also a "low-Q" antenna, which means it 
naturally has a wider bandwidth than a simple dipole or inverted vee.   
If you are interested in such a design for your station, I can offer 
additional constructions information that may be of use. 


73


- Jim, KL7CC




Mike Tibor wrote:
> Has anyone in the Anchorage or Valley areas here built a cage dipole?  If 
> so, I'd love to come out and get a look at it, and I have a few questions 
> as well.
>
> I have an 80m off center fed dipole that I'm thinking about converting to 
> an OCF cage dipole to try to flatten the SWR on a couple bands.
>
> Thanks,
> Mike
> KL3HY
>
>