Bifilar vs. Coax: Which Winding Is Best for Chokes?
Bifilar vs. Coax: Which Winding Is Best for Chokes?
Both can work. The better winding is the one that carries the wanted differential mode cleanly while presenting enough measured impedance to the unwanted common mode.
“Coax for coax and bifilar for balanced line” is a useful starting instinct, but it is not the physics. A ferrite common-mode choke acts on the mode, while the winding also acts as a short section of transmission line. Coaxial cable and a properly designed two-wire line can both perform those two jobs. The choice depends on the ports, differential impedance, common-mode impedance, voltage, current, power and frequency range.
The Core Does Not Know Whether the Winding Is Coax
In the wanted differential mode, current goes toward the load on one conductor and returns on the other. Their magnetic flux in the core subtracts. In common mode, current has components flowing in the same longitudinal direction on both conductors, so the flux adds and the ferrite presents impedance.
ZCM(f) = RCM(f) + jXCM(f)
The impedance is frequency-dependent and is not a fixed ideal inductance. Core material, core dimensions, turns, winding capacitance and layout shape both its resistive and reactive parts. More turns increase low-frequency inductance, but also increase inter-turn capacitance and can move the high-frequency resonance downward.
Preserve the line impedance, keep insertion loss low and withstand the differential current and voltage.
The core should see flux addition and present useful ZCM over the required bands.
Insulation, bend radius, connectors, temperature and common-mode voltage all matter at power.
Why Coax Is Usually the Natural Choice on a Coax Feedline
In coaxial differential mode, the wanted currents flow on the centre conductor and the inside surface of the shield. Their external field is largely confined by the cable structure. Any longitudinal current on the outside of the shield belongs to the external common mode.
Passing the complete coax through ferrite therefore leaves the intended transmission-line mode largely undisturbed while the outside-shield current sees the choke impedance. Coax also preserves its specified characteristic impedance and shielding through the winding, making construction repeatable.
That makes coax the default for a 50 Ω coax-feedline choke—but not “better by far” under every condition. A coax winding still has:
- turn-to-turn and winding-to-enclosure capacitance;
- a minimum bend radius and possible shield deformation;
- dielectric, conductor and connector loss;
- voltage and current limits that change with SWR; and
- a finite common-mode impedance that must suit the installation.
A coax choke is also a 1:1 current balun when used at an unbalanced-to-balanced transition. It is routinely used between coax and a balanced antenna such as a dipole. It does not transform 50 Ω into 300 or 600 Ω, but its common-mode impedance helps prevent the feedline exterior from becoming the unwanted third conductor.
When a Bifilar or Two-Wire Winding Makes Sense
“Bifilar” simply means two conductors are wound together. Depending on spacing, insulation and twist, they form a two-wire transmission line. If their characteristic impedance and loss suit the circuit, they can make an effective 1:1 Guanella current balun or common-mode choke.
This is useful when:
- the ports are naturally two-terminal rather than coaxial;
- a balanced transmission line must pass through the choke;
- the required voltage calls for specific wire insulation and spacing;
- a compact two-wire winding fits the chosen core better; or
- the intended transmission-line impedance can be designed and controlled.
Two wires laid tightly together do not automatically make a 50 Ω line, a 300 Ω line or a good balun. Conductor diameter, centre-to-centre spacing, insulation permittivity, twisting and proximity to the core determine differential characteristic impedance and loss. A random pair may still appear to “work” when electrically short, but that is not the same as a controlled design.
Bifilar does not enforce balance by itself. Balance comes from the complete circuit and sufficiently high common-mode impedance relative to the external common-mode path. A beautifully symmetrical winding cannot correct an asymmetric antenna and environment with an inadequate choke.
Balanced Line and the Tuner-Output Question
A balanced line can be choked by routing both conductors together through the same ferrite structure so differential flux cancels and common-mode flux adds. The winding geometry should preserve the line’s differential impedance as well as practical.
An unbalanced tuner feeding open-wire line often needs a current-balancing strategy, but “always put a bifilar 1:1 choke directly on the tuner output” is too absolute. At that point, the balun may face high or low transformed impedance, large reactive current and high common-mode voltage. A device comfortable at 50 Ω may overheat or arc there.
Depending on the system, the better architecture may be:
- a balanced matching network;
- a suitable current balun on a defined 50 Ω side;
- an output current balun explicitly designed for the expected impedance and voltage range; or
- a remote balanced tuner close to the line transition.
The correct location follows from the actual common-mode path and the stress at that point—not from the word “ladder line.”
The “Mirrored Bifilar” Warning Needs a Circuit Diagram
A mirrored physical layout does not “suck common-mode energy into the system.” That phrase has no general circuit meaning. What matters is winding sense, conductor identification and how the four terminals are connected.
A wiring error can make differential flux add in the core, causing unwanted insertion loss and heating. Another connection can make common-mode flux cancel, destroying the choking action. But a split, crossed or physically mirrored layout can also be perfectly valid when its electrical polarity is correct.
Practical rule: mark the start and finish of both conductors, draw the complete four-terminal connection, and verify differential transmission plus common-mode impedance. Judge the circuit, not the photograph.
What to Compare Instead of Winding Labels
| Design question | Coax winding | Two-wire/bifilar winding |
|---|---|---|
| Differential line impedance | Specified by the cable and highly repeatable | Set by wire diameter, spacing, insulation and core proximity |
| Shielding | Maintained through the winding | Normally unshielded |
| Natural ports | Coax feedline; also a 1:1 current balun to a balanced load | Balanced/two-terminal circuits and custom transmission-line transformers |
| Voltage design | Limited by cable dielectric, connectors and SWR | Can select spacing and insulation, but must control geometry |
| High-frequency limit | Usually set by winding capacitance, electrical length, core behaviour and layout—not the name of the conductor | |
| Proof of performance | Measured common-mode impedance plus differential return loss/insertion loss over frequency | |
Power Handling: Two Different Heating Mechanisms
Wanted differential power ideally produces little net core flux, but it still heats conductors, dielectric, connectors and contacts. Common-mode current does magnetise the core and can create ferrite loss. The relevant dissipation is associated with the resistive part of the common-mode impedance:
Pcore-related ≈ ICM,rms² × RCM
This is why “handles 1.5 kW” is incomplete without load, SWR, duty cycle, frequency and existing common-mode current. A choke that stays cool in a matched dummy-load test may heat in an asymmetric antenna system because the common-mode voltage and current are different.
A Better Selection Workflow
- Identify the wanted transmission-line mode. Coaxial 50 Ω, balanced 300/450/600 Ω, or a custom two-wire line are different tasks.
- Identify the unwanted common-mode path. Include the feedline exterior, tuner chassis, station wiring, mast, antenna and environment.
- Set the band and impedance target. Choose core material, size and turns for useful ZCM across that range.
- Check differential integrity. Measure return loss and insertion loss with the intended source and load impedance.
- Measure common mode. Use an appropriate fixture to obtain RCM and XCM, not only a single magnitude at one frequency.
- Test at realistic stress. Include maximum power, SWR, duty cycle, enclosure and ambient temperature.
- Recheck in the installation. Clamp-current measurements along the line reveal whether the external mode is actually controlled.
The better rule of thumb: use coax when preserving a coaxial line is the simplest solution; use a controlled two-wire line when the ports or voltage demand it. In both cases, trust the modal circuit and measurements—not “coax versus bifilar” as a universal ranking.
Mini-FAQ
- Must a choke on coax be wound with coax? No, but coax normally preserves the feedline impedance and shielding most predictably.
- Can a coax choke feed a balanced antenna? Yes. Used as a 1:1 current balun, it can suppress the unwanted external mode at a coax-to-balanced transition.
- Is bifilar automatically balanced? No. It is a two-wire geometry; balance depends on connection, common-mode impedance, load and environment.
- Are mirrored bifilar windings always wrong? No. Electrical polarity and terminal connections determine flux addition or cancellation.
- What measurements matter? Common-mode R and X, differential return loss and insertion loss, followed by realistic thermal and installation-current checks.