Introduction to

Off-Center Fed Dipole

Antennas
(OCFD)
Also know as
Windom
 My Goal
• Need 80m & 40m antennas for CQP (my favorite
HF contest)
• Contest bands are 80m, 40m, 20m, 15m, 10m
• Need to repair my previous 80m OCFD
• Need to be able to handle 1000 watts
• Both antennas should be up ¼ wavelength or
slightly more for NVIS into California
• Additional bands are a huge plus as long as
primary bands work well
• 40m antenna is a nice size for many other uses.
 Key Items:
• 1:1 SWR is NOT important
– the dip at resonance is important
• The correct balun MUST be used
– Dual-core Guanella
• A Common Mode Choke is almost always
necessary (why leave it out). 1:1 Guanella
• Should not be tuned on the lowest frequency
band
• The bands are not strictly harmonically related
I’m not an expert
• Anything for an antenna
will work
• Better antennas improve:
– Signal Strength
– Band Openings
– Number of contacts
– Enjoyment
• The OCFD is not the best
antenna for everyone, or
every time
• There is no best antenna
 What is Off-Center Fed?

• I thought the antenna was simple until I

started learning enough to repair my
previous antenna
 OCFD Antennas have a bad
Reputation
• Don’t work all bands as often advertised
– Marketing driven
• Do not have more gain than a dipole
• Squirrely
• Need a tuner to operate some bands
– tells you something isn’t quite right
• Might not work properly on those bands that “need a tuner”
• RF in the shack
• Cook supplied Baluns
• [Carolina Windom
– Feedline designed to radiate] not normal
Loren Windom ( 1905 – 1988 )
Windom antenna was an off-center-fed
resonant dipole with a single-wire feedline
of any length-fed against earth and
intended to be used on one frequency
only. Its main advantages were its single-
wire feedline —would not be of any value
anymore today because of propensity to
cause RFI— and the fact that it could be
easily matched to a valve amplifier.
 Why a OCFD?
• I wanted an antenna capable of several bands
– Don’t we all?
• A wire antenna is easier for temporary locations
• A wire antenna is almost a necessity for lower
bands
• The OCFD advertisements seemed to promise
that.
 OCFD CANNOT Work All Bands
• Only operates properly on designed bands
– Baluns & tuners can be easily damaged even on
resonant bands when not used properly!
• Any non-designed band is the same as tuning
a random-length wire
• Radiation pattern is unpredictable for non-
resonant bands
• They can work most bands IF you know the
limitations, and build and install it properly.
 A Dipole Antenna Can Be Fed at
Any Point
• Center
• End Fed – a J-Pole is an example

• Center has low Impedance ~ 50 ohm

• Ends have high impedance ~ 4000 ohm
• Impedance between center and ends is
somewhere between!
• Almost any antenna is resonant on multiple
bands, just find the spot to feed it.
 Let’s Build One
• The usual split (ratio) is 33% - 67%
• The usual Balun is 4:1
– Off-center has a higher feed impedance and 200 to
300 ohms is a closer match than 50 ohm.

• 33% split is a high impedance point for 15m

– Many manufacturers say you can use the antenna on
15m with a tuner or adjusting feed line length.
– DON’T DO THIS!
• There are other splits that are high impedance
points for other bands
– Be aware of the limits of your antenna
All bands plotted as if current with a 40m dipole
Looking for mid-point currents for the desired bands - I picked 40.2% for my 40m OCFD
40.2% split Goal: 40m, 20m, 15m, 10m resonant
 Some reasons nothing is perfect
• End Effects
• Height above earth in proportion of wavelengths
• Bands not truly harmonically related
3.500 3.600 3.750 4.000
7.000 7.200 7.500 8.000
10.500 10.800 11.250 12.000

14.000 14.400 15.000 16.000

17.500 18.000 18.750 20.000

21.000 21.600 22.500 24.000

24.500 25.200 26.250 28.000

28.000 28.800 30.000 32.000

Picked 33.3%
Goal: 80m, 40m, 20m, 10m
 The following split ratio relates to a 80m OCF dipole
(~135 feet total length)
• 50%, lose 40m, 20m, 15m, and 10m bands (high impedance), but should do
very well on 80m, 30m, 17m, 12m, and 6m bands. This is a normal dipole,
where you only get odd-harmonics. "Lose," means the antenna feedpoint
will have very high impedance.
• 41.6%, you don't lose any ham band (80m-6m). Although the SWR will be a
little higher on 17m and 12m.
• 36%, you don't lose any ham band (80m-6m). Although the SWR will be a
little higher on 30m.
• 33% (1/3:2/3) you lose 30m, 15m, and 6m bands (high impedance).
• 30%, you don't lose any ham band (80m-6m). Although the SWR will be a
little higher on 12m.
• 28.89%, you lose 12m band (high impedance).
• 25%, you lose 20m and 10m bands (high impedance).
• 20% (20/80), you lose 17m and 6m bands (high impedance).
• 16.6%, you lose 15m band (high impedance).
• 8.9%, you don't lose any ham band (80m-6m) (no in ham band sine nulls).
Although the SWR will be higher on 80m.
• No “best spot”, just different bands from different splits.
 Multiple half-wavelengths Or 40m on 40
produce multiple radiation lobes
• Each band is at a different
Wavelength-Related Height
• Feed Impedance changes
with relative height
•Radiation angle changes with
relative height
•With 2 OCFD antennas, 40m
will have different patterns
20m @ 70’

Not quite correct

 Imbalance – Cause of squirrely
• Imbalance can also be caused by situations where
the two sides of the antenna are not symmetrical:
Coupling to nearby objects, the tilt relative to the
ground or slight differences in the lengths of the
antenna halves. Another cause of imbalance current
is induction. If the feed line is not exactly placed at a
right angle to the antenna, a net current is induced
into it by the antenna field. This current is called
Common Mode Current.
• If there is common mode current with transmit, your
feedline is also acting as part of your receive
antenna.
Imbalance is difficult to avoid, and will affect all antennas. With an OCFD, the overall
imbalances are compounded by the off-center feedpoint. This adds to the squirrely
nature of the antenna, but can be overcome by using proper components and making an
effort at best placement.
 A choke needs to be added at
the feedline / antenna
connection (1:1 balun)
Bad – needs a choke
 Components
• Don’t be cheap
• Need a good 4:1 Dual core, Dual Winding
Current Balun – Guanella
– I use a Balun Designs #4115 (about $97 )
• #4114 is Single Core
• Need common-mode choke
– Should be a 1:1 Guanella Choke
– I got away with a Maxwell Choke (beads over coax)
(Wireman # 8231 (about $29 ))
• I had ideal locations for these antennas
Very Bad

The balun must be a Dual-Core balun

Dual Core Balun
Common-mode chokes

Guanella
http://www.karinya.net/
g3txq/chokes/
Antenna Environment affects operation
• The more radical the split (feedpoint further from
the middle), the more squirrely the antenna
becomes.
• Feedline needs to run at 90° from the antenna
for as far as possible.
• Need good common-mode choke
– Guanella 1:1 is recommended
• If touching a antenna connector causes a
change in SWR, your body has become part of
the antenna, so you have some amount of
common mode.
How to Remove Coax from SWR

• If you have some analyzer that can record the

measurements in an external file, and b) you have
Microsoft Excel, you can use the free Zplots Excel
app. Zplots has the "Add/Subtract" feature.
• More details and a download link:
• http://ac6la.com/zplots1.html
• For info on the Add/Subtract feature see the third
and fourth bullets and the first appendix at the
above link.
• http://www.arrg.us/pages/Loss-Calc.htm
 Adjusting OCFD
• A graphing antenna analyzer is highly
recommended
– Borrow one – Borrow a friend with one!
• The older single-frequency tunable SWR
meters could work, but would be more work
than most are willing to do.
– Every 50 KHz over the band and ± 500 KHz
needed. Then a detailed graph needs to be done
for each band.
• A radio and SWR meter will NOT work.
– Cannot look at what is outside the Ham Bands
 Adjustments
• Cut both legs a little longer, fold back at the
end insulators and adjust for the desired split
ratio.
• Measure the SWR curves, then prune,
maintaining the split ratio.
• Do not adjust for the lowest band, but all
bands
• An inverted V OCFD should have the center
highest, and the balun is not at the center.
 Common Mode Impedance (CMI) of
Baluns & Chokes
• CMI of BALUNS depends on many things:
• - Type of ferrite,
• - size of ferrite,
• - number of turns of wire,
• - design circuit
• ALL contribute to the CMI of a balun, but THE CMI
IS NOT CONSTANT across the hf spectrum. It
varies with frequency. What works fine on one band
may not have enough CMI on lower bands.
Goal: 40m, 20m, 15m, 10m
Goal: 80m, 40m, 20m, 10m
 Reference Material
• =ABS(SIN(A3*2*PI()/180)) where A3 = angle (draw sine wave)
• windom_antenna Yahoo Group
• http://hamwaves.com/cl-ocfd/en/index.html
• http://www.aerial-51.com/
• http://www.balundesigns.com/ Model 4115
• http://thewireman.com/ #8231
• http://www.dj0ip.de/
• http://www.karinya.net/g3txq/chokes/
• QST, July 2000, Everything Works, Tom Schiller
• A Ham's Guide to RFI, Ferrites, Baluns, and Audio
Interfacing http://k9yc.com/RFI-Ham.pdf
– Exactly as the title says, and it might take several readings to
understand, but well worth looking at.
 Summary
• The correct balun MUST be used
– Dual-core Guanella
• A Common Mode Choke is necessary
– Guanella 1:1 is recommended
• A split for the desired bands needs to be selected.
• Should not be tuned on the fundamental frequency
band
– A graphing antenna analyzer is almost a necessity
– The bands are not strictly harmonically related
– Keep the desired split while adjusting
• 1:1 SWR is NOT important
– the dip at resonance is important
• A tuner is always a good idea to protect your radio