In a previous issue of the B-VARC Bulletin, I wrote about the DX benefits 
and construction of the full-size, 1 wavelengh long, vertically oriented,
corner fed, vertically polarized, Delta Loop.  Deployment of a Delta Loop 
requires only a single high point making it very beneficial to those of us 
who live on small city lots.  After building and testing the full size loop, 
I decided to experiment and determine if I could improve it in some way.  
I thought that I could make it smaller and therefore make deployment easier.
But, could it be made smaller and still maintain the great performance of 
the full size version?

The Theory
The full size DX loop highlighted in a previous article provided the 
mechanical benefits of a single support point and corner feed.  However, 
I noticed that on the standing wave current distribution plot (Fig. 1) of 
a loop fed 1/4 wavelength from the top that the horizontal (baseline)
portion of the full size loop consisted of half and half opposing polarity
current distribution.  These currents effectively cancel and therefore
eliminate any horizontally polarized farfield emissions.  In light of
this, I decided to try and shorten the loop by applying loading within this
horizontal portion of the loop.

By placing the loading elements within the horizontal portion of the loop, 
I could keep the vertical portions of the loop full size and in the 
identical orientation as the full size loop.  (The vertical sides are the 
part of the loop that radiates the low angle radiation.)  By starting the 
baseline right at the feed point and making the baseline the same physical 
length as the other 2 sides, the triangle would remain equilateral.  I felt 
that this configuration should allow the shortened loop to provide similar 
vertically polarized low angle radiation performance to that of the full 
size loop.  However, I did have concern over the loss of the vertical 
portion of the loop from below the feed point to the corner as this is 
an area of high antenna standing wave current.

I researched some ham radio magazine articles dealing with shortened Delta 
Loops (bibliography is given at the end).  These articles gave me a 
starting point for my design.

The Design
If I was to have an equilateral triangle with 2 sides equal to 1/4 wavelength
then the 3rd side (the baseline) would also have to be the same physical
length. From this, the total electrical length of the 3 sides of the loop
is 3/4 wavelength, a non-resonant length.  To provide a total electrical
length of 1 wavelength, the 1/4 wavelength (physical length) baseline
would have to be "loaded" to obtain 1/2 wavelength (electrical length).
(Loading is that act of replacing the physical element length with the
appropriate value of inductors or capacitors of various forms.)

Implementation
I took this baseline reduction and testing in a few steps. (Fig. 2a through 
2e shows the progression from a full size loop fed 1/4 wavelength from the
top to a baseline loaded short loop.)

1.   I folded the loops lower 1/2 wavelength into a baseline with a
     center 1/4 wavelength inductive stub (Fig. 2b).  This method was
     straightforward, but had deployment problems in trying to figure
     out in which direction to run the 8.5 ft. stub.

2.   I then folded the stub using the efficient linear loading 
     method (Fig. 2c).  This worked fine performance-wise, but 
     was not a straightforward deployment due to the increased 
     complication of hanging multiple uneven length wires under
     the baseline.

3.   I decided to try capacitive loading at the voltage loop 
     (high voltage point) (Fig. 2d), but again, the physical 
     layout was not stable with the hanging wires, etc.

All of the above stub, linear and capacitive loading techniques measured 
well and on-the-air tests proved their performance but mechanically, they 
lacked ease of implementation and size reduction.  For example, the 
capacitive loaded 40m loop only decreased the vertical height by 5 ft., 
not the 10 ft. for which I was hoping.

Finally, I tried loading the baseline with lump inductive reactance (coils).
By using coils, it would provide a single wire baseline and maximize the 
loop height reduction.  With the use of coils, I could efficiently load 
the baseline and keep the mechanical implementation simple and within close 
proximity to the baseline wire.

Loading Coil Values
The value of the inductors required was figured simply by using the chart 
for "determining coil inductance values for off center loaded dipoles" 
found in the ARRL Antenna Handbook Chapter on HF Antennas for Limited Space.
Since the loops lower 1/2 wavelength current distribution was the equivalent
of an inside out 1/2 wavelength dipole (Fig. 1), I could apply dipole
shortening techniques. I was reducing the 1/2 wavelength into a 1/4 
wavelength space, therefore the antenna size reduction was 50%.

As seen on this chart, inductor values are determined by their location on 
the antenna.  Mounting them at the feed point and at the opposite corner, 
"the high current points," the inductive reactance (XL) values needed are 
lowest.  Moving the coil locations away from the high current portions of 
the antenna leg requires an increase in the inductance of the coil.

I chose a midpoint location for the inductors based on the good performance
of center loaded mobile whips.  XL at this point from the chart is
approximately 950 ohms.  Applying the formula XL=2(pi)fL, the inductor 
value works out to be 5 microhenrys (per side) for the 40m version.

Using the Radio Amateurs Handbook section on inductance, I calculated the 
physical dimensions for 3" diameter "airwound" coil (3" was the cardboard 
tube size I had available).  I wound the coils from 18 turns of #12 soft 
uninsulated house wire.  The final coils would be supported within the 
baseline on "Thompson's Water Seal treated" broomstick handles (Fig. 3).

Construction
Follow the instructions for building the full size loop in the previous 
issue, but make the sides 1/4 wavelength configure the baseline.
Construction is straightforward and the comments in the previous issue
should suffice.

Feeding
Although the feed impedance of the shortened loop is lower than the full 
size loop, the same 1/4 wavelength series section transformer in the
previous article can be used.

Performance
I have the loop hung from a 42' push up pole.  The loop baseline is 12' 
from the ground and conveniently is just above my house roof.  The 
triangle is oriented in a North/South plane and therefore broadside 
East/West.  On-the-air tests on 40m over the last 9 months with Mel-
KB5ION and Maurie-VK3CWB in Australia, have shown this smaller delta 
loop antenna to be as good a DX performer as the full sized version. 
It even keeps up with Mel's 3 element Bob-tail Curtain (most of the time, 
anyway).

Summary
If you don't have room for the full size vertical loop, this reduced size 
version could provide you with an antenna that will fit your available 
space.  This loop, as with the full size loop, can be scaled for other 
bands.  Remember, that when converting the loading inductors inductive 
reactance to Henrys, that the formula is influenced by the frequency. 
Build your coils accordingly.  The series section transformer length 
must be adjusted also.  Turn on your soldering iron and have fun.

If you would like a full size copy of this article with drawings and 
additional charts or you need to borrow any of the references listed, 
please give me a call during the day at 388-4466.

References
Low Band DXing-ON4UN-John Devoldere-Section 2.8 Loops
ARRL Antenna Handbook-HF Antennas for Limited Space
Corner Fed Loop Antenna-Ham Radio Mag. 1976
Top Loaded Delta Loop Antenna-Ham Radio Mag. 1978
The Reduced Size Delta Loop-Ham Radio Mag. 1985
the B-VARC Bulletin, Sept. 1995, Antenna Loading-KF5NU