Off-Center-Fed Dipole ≠ “Unbalanced Antenna”
Off-Center-Fed Dipole ≠ “Unbalanced Antenna”
One arm is longer. That does not tell the whole electrical story. For the imperfect installations we actually build, my starting point is clear: choose the impedance transformation, then give common-mode control its own properly specified choke.
RF.Guru working definition: Common-mode current is the non-cancelling phasor-sum current in a specified set of conductors, evaluated at a defined cross-section and using a declared current-direction convention. In the intended differential transmission-line mode, the outgoing and return currents are equal and opposite, so their phasor sum is zero. When they do not cancel, the remaining current must close through another reference or return path—such as the outside of a coax shield, a mast, equipment chassis, station wiring, nearby structures, earth, the operator, or distributed coupling through the environment.
This broader working definition is especially useful in practical antenna systems. On transmit, non-cancelling current on the outside of the coax can make the feedline and connected structures part of the radiating antenna system unless that path is intentional, clearly defined and properly controlled—for example by providing the required return path and placing a suitable common-mode choke at the correct boundary.
“One leg is longer, therefore the antenna is unbalanced.” I hear that shortcut often, and it mixes the shape of an antenna with the behaviour of its feed system. An off-centre-fed dipole, or OCFD, has unequal arms. It can still be driven differentially at a two-terminal port—but in a real garden, those arms do not see the same surroundings, and the feedline can become an unwanted third path.
After roughly two decades working with HF baluns and UNUNs, I do not start by assuming textbook balance will survive the tree, roof, mast and coax route. My practical default at RF.Guru is to separate the jobs: a suitable UNUN for impedance transformation, often 4:1 where the load calls for it, and a dedicated 1:1 current choke for common-mode control. That is a design preference for real installations, not a claim that every OCFD is 200 Ω or that a good current balun cannot work.
Mark takes up the same distinction
In Is an Off Center Fed Dipole “UNBALANCED?”, Mark the Ham Florida Man reads from this RF.Guru article, written by me, Joeri Van Dooren, ON6URE. The essential point is that “physical symmetry is not the same as electrical balance”. He also follows the practical arrangement: a 4:1 UNUN and a strong current choke, rather than expecting the transformation ratio to suppress outside-coax current.
The distinction needs its electrical boundary: equal-and-opposite currents at an ideal port do not prove equal voltages to the surroundings, nor zero external current in the installed antenna. Mark also discusses feeding an OCFD with ladder line. A balanced line still needs sensible routing and an appropriate matching interface.
His closing preference for a lower-ratio OCFD feed over a wide-band-span EFHW points toward a useful design choice: transforming a moderate load need not involve the same voltage ratio as a kilohm-class end feed. The actual load, transformer and band set still decide loss; “4:1” alone is not an efficiency measurement.
What do we mean by balance?
- Is the geometry symmetric? An OCFD is not: the two wire lengths and their positions differ.
- Is the intended port differential? A two-terminal feed can drive current out of one terminal and back through the other.
- Are the terminal impedances and voltages symmetric to the surroundings? Equal-and-opposite feed currents alone do not establish that stronger sense of electrical balance.
- Does the structure convert differential excitation into common mode? Unequal arm impedances to the environment, feed-device parasitics and routing can create that conversion.
- Is the installed common-mode current acceptably small? That is a measured system result, not a conclusion from the drawing.
The geometry does not answer the other questions for us. A centre-fed dipole can acquire common mode through an asymmetric environment or feedline route. An off-centre-fed dipole can have low external feedline current when its feed and return paths are controlled. Neither outcome should be assumed from appearance.
The useful contrast with a monopole is the intended current system. An OCFD has two wire arms intended to radiate; the shorter arm does not automatically become an earth reference. A mobile whip deliberately works against a vehicle body, and a vertical may work against radials or another counterpoise. Both types still need a return path and both interact with their surroundings. Calling one “balanced” does not remove that interaction.
The Two-Terminal Statement Has a Boundary
At an ideal isolated two-terminal port, define both currents as entering the load. Kirchhoff’s current law gives:
That equality is important, but it applies at the named port. It does not say the currents have equal magnitude at corresponding positions along unequal arms, guarantee equal-and-opposite terminal voltages relative to the environment, or produce a symmetric radiation pattern.
When the outside of a coax shield, mast, support, equipment chassis, control cable or stray capacitance supplies another RF path, the practical feed junction is no longer a two-current problem. With all currents defined into the junction:
A non-zero Iexternal is the evidence that the intended two-conductor mode is sharing current with another path. That is the useful installed definition of common-mode trouble. The third path must exist before it can carry current; the off-centre coordinate alone is not that path.
What Moving the Feedpoint Really Changes
A wire antenna has a distributed current and charge solution. Moving the driving point samples a different part of that solution and changes the driving-point impedance. On a multiband wire, the current nodes, antinodes and pattern can change substantially from band to band.
There is no universal OCFD impedance. The complex value at the terminals depends on:
- frequency, total electrical length and feedpoint fraction;
- wire diameter, insulation, bends and element angle;
- height, ground properties and nearby conductors;
- loss in wire, connections and any loading components; and
- the feed device, feedline route and any outside-conductor current included in the measurement.
“A few hundred ohms” may describe one design on selected bands. It is not a specification for every OCFD or every harmonic. Measure or model the installed complex impedance on every intended band before choosing a transformation ratio.
Coax Supports an Intended Mode and an External Mode
Inside coax, the wanted transmission-line current flows on the centre conductor and the inner shield surface with equal magnitude and opposite direction. The shield exterior can carry another current relative to the antenna, mast, earth and station. That external mode sees a completely different impedance and electromagnetic environment.
At the OCFD feed, unequal coupling of the arms to ground and nearby objects can drive differential-to-common-mode conversion. The transformer’s winding capacitance and conductive connections can add another conversion path. Coax that runs close and parallel to one arm can be excited again even after a choke.
That is why feedline radiation, a moving SWR, RF in station wiring and changed receive noise are useful warning signs. They are not calibrated current measurements, and their absence is not proof of zero common mode.
My default: transform the impedance, then choke the external path
For the coax-fed OCFD installations discussed here, I prefer a transformer-and-choke assembly whose two functions are explicit. The UNUN handles the differential impedance transformation. The dedicated 1:1 choke adds impedance to the external common-mode path. The choke is doing real work; a UNUN on its own is not a substitute for a balun function.
Why separate them? A tree beside one arm, a gutter under the other and a coax run pulled toward the house create unequal coupling that a neat antenna drawing does not show. The transformed differential load and the unwanted common-mode loop are different circuits. I want to select the transformation for the first circuit and the choke response for the second, rather than assume that a box marked “4:1 balun” has satisfied both requirements.
The practical advantage is control over those two requirements. If the transformation suits the antenna but the installation still drives exterior current, the common-mode part can be improved or repositioned without treating a different impedance ratio as the cure. If the match is wrong, a bigger choke is not a replacement for the right transformation. The completed assembly still has parasitics and coupling; specifying the jobs separately does not make them physically independent.
Where the measured antenna load is near 200 Ω resistive, an ideal 4:1 impedance step presents about 50 Ω. That is why 4:1 is a useful starting point for many OCFD designs. A real multiband load is complex and moves with frequency, height and surroundings: 4:1 is not an instruction to ignore the other bands, nor a promise of a tuner-free antenna.
For most of the practical, asymmetrically installed OCFDs I work with, this is my starting architecture: the two antenna arms feed the impedance transformer; a separately specified 1:1 choke establishes the boundary to the station-side coax. For this two-wire radiator, I normally want that boundary close to the feed assembly so I am not deliberately adding a length of coax exterior as a third radiating leg. An end-fed design that deliberately uses a coax section as its return is a different arrangement.
This approach is not limited to poorly balanced surroundings. A UNUN-plus-choke pair can form a useful hybrid interface for a balanced load too, when the connection, isolation, parasitics and stress are designed and verified as a complete assembly. Two arbitrary boxes connected together do not automatically make that hybrid.
Why this is a preference, not a ban on current baluns
A properly designed current balun can work with an OCFD, including a load that is not symmetric to its environment. Current-balun topology, isolation and voltage capability matter; the word “current” and the ratio on the case are not sufficient specifications. My preference for the separate functions is therefore not “all current baluns fail.” It is that I want common-mode control to remain an explicit design decision in an installation where ideal symmetry is unlikely.
Tom Rauch, W8JI, makes the important supporting distinction: low SWR and equal feed currents do not, by themselves, prove the required balance and isolation. His balun tests exercise different load references, and his OCF example explicitly allows a good 4:1 current balun. His work supports examining the actual circuit and common-mode path; the UNUN-plus-separate-choke default stated here is mine.
The aim is a more controlled feed system: less unintended coax participation, a match less vulnerable to unrelated cable movement, and reduced pickup through an unwanted receive path where that path was contributing noise. Those benefits follow from effective isolation in the installed system—not from a guaranteed dB figure or an assumption that the antenna’s intended pattern becomes symmetric.
Specify the Differential Job
For every intended band, record the antenna’s complex terminal impedance at the feed-device plane. Check transformation accuracy, return loss, insertion loss and voltage/current stress with representative complex loads—not only a nominal resistor. The coax and tuner must tolerate the transformed impedance and mismatch across their actual lengths.
Specify the Common-Mode Job
Draw the complete external loop: shield exterior, mast, station wiring, bonds, earth and stray capacitance. In a simplified loop, the residual current follows:
The choke impedance is complex, frequency-dependent and finite. Parasitic capacitance, resonance, winding loss and the return-path impedance all matter. There is no universal “enough ohms” threshold and no fixed number of turns that proves performance across several bands.
Place the Choke at the Path You Intend to Interrupt
A choke at the feedpoint can keep the first coax section out of the antenna system. A choke farther down the line defines a different external-conductor segment. A second choke is useful only when it interrupts a demonstrated path without creating unacceptable voltage, loss or resonance. Placement is a circuit decision, not a ritual distance.
Routing Changes Coupling, Not Antenna Category
Routing coax away from the wire rather than alongside one arm usually reduces direct coupling. Height, mast position, gutters, trees, support ropes, lightning/bonding conductors and cables entering the building can still break environmental symmetry.
There is no universal clean angle or first-several-metres rule. Model or measure the route that will be installed. If moving one section changes outside current, impedance or receive noise, that section is part of the electromagnetic system and should be treated as such.
Measure the Installed Modes
A useful qualification sequence names its reference planes and keeps the antenna, feed device and feedline contributions distinguishable:
Disable transmission and isolate the transmitter’s power before changing antenna connections, chokes, cables or measurement fixtures. Follow the equipment manufacturer’s shutdown and stored-energy discharge instructions. Reconnect and clear the working area before resuming RF tests.
- Model all relevant conductors. Include unequal arms, ground, mast and a representation of the external feedline path; an antenna-only model cannot predict feedline common mode.
- Measure complex differential impedance. Calibrate at the antenna or feed-device plane and repeat on every intended band with the final height and geometry.
- Characterise the feed device. Use a suitable multiport or mode-conversion fixture to measure differential match/transfer, common-mode impedance and differential-to-common conversion. De-embed adapters and fixture delay.
- Measure outside-coax current. A calibrated RF current probe clamped around the complete coax responds to net external current. Scan several positions because standing waves can place a null at one convenient point.
- Compare arm currents carefully. Use matched probes, consistent direction and declared planes; amplitude without phase is incomplete evidence.
- Run controlled A/B/A trials. Add or move one choke or one cable section at a time, restore the starting state, and record current, impedance, receive noise and field/pattern evidence under stable conditions.
- Repeat at operating power. Verify RF voltage, current, temperature rise, insulation and connector limits with suitably isolated instrumentation. Do not use touch as a temperature or RF-voltage test.
A low transmitter SWR proves only an acceptable reflection at that reference plane. It does not prove low outside-coax current, stable pattern, low receive-noise pickup or safe feed-device stress.
The distinction leads to a practical choice
An off-centre-fed dipole has unequal geometry; that is not a complete diagnosis of its current modes. In the imperfect installations I normally encounter, I start with a load-appropriate UNUN—often 4:1—and a dedicated 1:1 choke, arranged to keep the station-side coax out of the intended radiator. The reason is practical: solve the impedance problem and the external-current problem explicitly, rather than expect one label to solve both.
The antenna does not need to win an argument about whether it looks balanced. It needs a usable match, controlled current paths and a feed assembly that survives the intended bands and power. Keep the two jobs visible, and the choices become much easier to defend.
Primary and authoritative references
- Tom Rauch, W8JI — Properly Testing Baluns
- Tom Rauch, W8JI — Windom and off-centre-fed dipoles
- Tom Rauch, W8JI — Common-mode current and the installed return path
- Fischer and Hesselbarth — Off-Centre Fed Dipole Suppressing Feed-Line Radiation
- Bockelman and Eisenstadt — Combined Differential and Common-Mode Scattering Parameters
- Roy Lewallen, W7EL — Baluns: What They Do and How They Do It
- C. L. Ruthroff — Some Broad-Band Transformers
- Keysight — Balanced and mixed-mode network measurements
- Lawrence Livermore National Laboratory — Numerical Electromagnetics Code capabilities
- Fair-Rite — Common-mode ferrite selection and impedance-measurement guidance
Mini-FAQ
- Is an OCFD unbalanced because its legs have different lengths? It is geometrically asymmetric. That alone does not quantify common-mode current; the complete feed device, feedline, return paths and environment determine mode conversion.
- Must the two feedpoint currents be equal and opposite? At an ideal isolated two-terminal port, yes. If an external conductor or capacitive path carries current, the practical junction has a third current and must be analysed as a larger system.
- Does every OCFD have a few-hundred-ohm feedpoint? No. Complex impedance depends on frequency, dimensions, feed fraction, height, ground, surroundings and feedline interaction. Measure or model the installed antenna on every intended band.
- Does every OCFD need a 4:1 transformer? No. Select ratio and topology from measured complex impedances, loss and stress across the intended bands. Common-mode control is a separate requirement.
- What is my practical feedpoint default? For the asymmetrical coax-fed OCFD installations I normally encounter, a load-appropriate UNUN—often 4:1—plus a dedicated 1:1 choke. The benefit is explicit control of transformation and the common-mode boundary, not a guarantee that every 4:1 pair will work.
- Does that mean a current balun cannot work on an OCFD? No. A suitably designed current balun can work. A UNUN-plus-choke pair can also serve a balanced load when deliberately arranged and verified as a hybrid; neither result follows from labels alone.
- Does a choke force perfect current balance? No. It adds finite complex impedance to a named common-mode path. Residual current depends on drive, the complete return loop, placement, parasitics and frequency.
- How do I verify feedline common mode? Measure outside-coax current at several positions with a calibrated probe, define current directions and planes, and combine that result with impedance, mode-conversion and controlled A/B/A checks.