Why Blanket High-SWR Balun Derating Rules Fall Apart

The balun discussion that needs answering is about Gary Baker, K7EMF’s matched-side advice—not a mysterious SWR threshold. In Fabricating A Balanced ATU, Gary places the balun on the matched 50 Ω side of the tuner and says that low SWR “eliminates core heating” within the balun’s power-handling specifications. He also links that placement to balanced operation and the absence of core saturation.

I agree that keeping a long coax run matched can be a sound system choice. I disagree with treating the SWR reading as the physical explanation for everything a balun does. Low SWR does not measure current balance or magnetic loss. Nor can an SWR-only rule tell us the same power reduction for every device called a balun.

The useful distinction is this: a 1:1 current choke impedes a common-mode path while allowing the differential transmission-line mode to pass. A matching transformer may deliberately excite magnetic flux to establish a voltage or impedance ratio. Confusing those jobs produces the wrong explanation—and sometimes the wrong cure.

That is an argument for using the right model, not for ignoring a component limit. True Ladder Line’s Model 4116T description, for example, conditions its rating on the load and warns of reduced handling under high SWR. An assembly-specific warning belongs to that assembly. It is not a universal thermal law, and this article is not permission to exceed it.

The category error: one label, different circuits

Balun describes a balanced-to-unbalanced function; it does not identify one circuit. Unun likewise describes an unbalanced-to-unbalanced function. The internal topology determines how the device establishes current or voltage relationships, transforms impedance and controls common-mode current.

A “current balun” is not automatically a simple 1:1 choke. Guanella networks can be current-balancing transmission-line transformers with an impedance ratio. Conversely, a device sold as a voltage balun may not provide enough common-mode impedance for the installed antenna. Ask for the schematic, winding interconnection and measured port behaviour before assigning a category.

This is why I reject a universal verdict based on the label. Even “1:1” specifies a ratio, not the complete magnetic circuit. A voltage-type device can have a 1:1 ratio, while a current-balancing Guanella network can also transform impedance. The mechanism has to come before the derating explanation.

The SWR number does not identify the stress

For a line with real characteristic impedance Z0, the magnitude of the load reflection coefficient is related to SWR by:

SWR gives the magnitude of reflection. It does not give forward power, absolute voltage or current, the reflection phase at a particular device, waveform, duty cycle, cable loss, ambient temperature or common-mode current. Moving a device along an ideal lossless mismatched line changes the phase at its terminals and therefore its local voltage/current condition even though the line SWR is unchanged. On a lossy line, reflection magnitude changes with position too.

For a lossless line, sinusoidal RMS quantities and declared forward power Pf, the largest possible differential voltage and current magnitudes along the line are:

Voltage and current maxima occur at different positions. The accepted-power expression applies at the declared plane under the lossless-line model; a real transmitter, tuner and lossy line interact with the reflected wave and may fold back or dissipate part of the power.

As a bounded example, 1 kW forward power on an ideal 50 Ω line at 4:1 SWR gives |Γ| = 0.6, a possible Vmax ≈ 358 V RMS and Imax ≈ 7.16 A RMS. Those maxima are not co-located, and they are not a rating for any component. At fixed accepted load power, the required forward power rises with reflection magnitude unless the source reduces output, so a rating cannot assume that “1 kW” always means the same quantity.

There is an even simpler way to see the missing information. At the load plane of an ideal 50 Ω line, both a 200 Ω resistor and a 12.5 Ω resistor give 4:1 SWR. Deliver 100 W to either: the first needs about 141 V RMS and 0.707 A RMS, while the second needs 35.4 V RMS and 2.83 A RMS. Same SWR, four times the voltage in one case and four times the current in the other. These are circuit examples, not hardware ratings.

What makes a 1:1 current choke different

In an ideal coaxial or two-conductor differential mode, equal and opposite currents produce substantial magnetic-flux cancellation in a 1:1 common-mode choke. That is why the wanted transmission-line current need not create the same core excitation as common-mode current. That is the real advantage: the full wanted differential current need not drive the ferrite in the same way as a flux-coupled matching winding. In the ideal limit of purely differential excitation, standing waves do not by themselves create net common-mode core excitation. A suitable choke can therefore pass a mismatched differential mode while still providing common-mode isolation.

That advantage has a boundary. The cable or paired conductors still carry the full local differential current. Their dielectric and connectors still withstand the local differential voltage. A current maximum can increase conductor and contact heating; a voltage maximum can reduce dielectric margin or produce arcing at terminals. Winding leakage, imperfect symmetry, line-to-environment capacitance and finite coupling can also convert part of the differential excitation into common mode.

The construction determines where that stress appears. In a coax-wound choke, centre-to-shield voltage mainly stresses the coax dielectric and terminations, while exterior-shield voltage relative to the surrounding system belongs to the common-mode circuit. In a bifilar or multi-line device, inter-conductor, inter-winding, winding-to-core and winding-to-enclosure voltages can all differ. A single SWR value cannot represent those electric fields.

Placement therefore matters twice: it sets the differential standing-wave condition at the device, and it changes the surrounding common-mode circuit. A choke near a differential current maximum may be copper- or connector-limited. The same choke at another point may face greater dielectric stress. Either position can also be a common-mode current or voltage maximum depending on antenna asymmetry, feed-line route and nearby conductors.

A low-SWR location is not a core-temperature measurement

Gary’s matched-side placement can reduce differential mismatch at that port. But a balun’s common-mode voltage is measured against its surrounding return path, not solely between the two terminals used for the differential SWR reading. An antenna, tuner enclosure, mast or control cable can provide part of that path. A low differential SWR can coexist with common-mode excitation and core loss.

Current balance is also not the same as imposing equal voltages relative to ground. Unequal load impedances can draw unequal currents from equal-and-opposite voltages. A current choke acts on the unwanted sum current; whether it suppresses that current adequately depends on its impedance in the installed common-mode circuit.

Andrew Roos, ZS1AN’s antenna-tuner balun analysis makes the topology dependence explicit. Its 1:1 current-balun, voltage-balun and hybrid models do not have identical current balance or winding excitation under the same load conditions. The finite common-mode winding impedance must be considered relative to the load and its reference to ground. A high differential load impedance can challenge a real choke’s balance even when ideal flux cancellation is the starting model.

That is why “put it at a low-SWR point so it can do its job” is incomplete advice. On a uniform lossless line, moving along the standing wave does not create a lower SWR; a matching network changes the conditions between its ports. Choose placement to control the actual common-mode path and keep voltage, current and temperature within limits. Do not choose it merely to obtain a reassuring meter reading.

Common-mode heating has its own cause

The choke’s magnetic heating follows the common-mode excitation and its resistive impedance. Characterise the finished complex impedance rather than quoting only its magnitude:

The power and voltage expressions are first-order sinusoidal estimates using a consistent choke-port current definition. Real loss can be distributed unevenly, and ferrite impedance may change with temperature and drive. Fair-Rite’s suppression guidance explicitly identifies frequency, temperature and bias as material-performance variables; it does not turn an SWR reading into a ferrite temperature rating. A resistive choke can damp a common-mode resonance effectively yet heat when installed current is appreciable. A highly reactive choke can carry substantial RF voltage without a large temperature rise.

High differential SWR does not mathematically require common-mode current. It also does not prove that common mode is absent. Load imbalance, feed-point geometry, parasitic capacitance, unequal conductor coupling, mast and cable routing can create an exterior-current path. Measure installed exterior current at several positions and bands; one clamp-probe reading may coincide with a current minimum.

Transformation adds a different set of burdens

This is the other half of my disagreement. Do not take a warning derived for a flux-coupled matching winding and apply its explanation unchanged to a 1:1 choke. Equally, do not assume that every impedance-transforming device works by the same mechanism.

A Guanella ratio transformer uses transmission-line sections and can provide current balance as well as an impedance ratio. Each section still needs suitable characteristic impedance, electrical length, insulation, conductor capacity and common-mode winding impedance. The series/parallel connection can expose different sections to different voltages relative to the environment.

A Ruthroff transformer shares energy through an autotransformer-like connection. Its useful bandwidth depends on transmission-line behaviour at the high-frequency end and magnetising impedance at the low-frequency end. Load asymmetry can produce unequal currents even when terminal voltages have the intended relationship. Flux, copper loss, winding voltage and common-mode suppression must be evaluated separately.

Flux-coupled transformers add a direct volts-per-turn check. For a sinusoidal winding, a first estimate is Bpk ≈ Vrms/(4.44 fNAe) when the correct excited turns and effective core area are used. Core loss and temperature can become limiting before classical saturation, while reactive loads and non-sinusoidal waveforms require the actual volt-seconds and current waveform rather than a nameplate watt value.

In all three families, parasitic capacitance and leakage inductance can create resonances, redistribute voltage and degrade balance. Manufacturer small-signal ferrite or component curves do not establish a completed assembly’s high-power mismatch rating.

Derating is real; a universal explanation is not

A manufacturer can legitimately specify a conservative maximum SWR or mismatch derating for a particular finished device. Follow that limit. My objection is to treating that application rule as proof that every balun core responds to mismatch in the same way.

A valid application limit must state the reference condition from which any reduction is made. A cable manufacturer, for example, may publish a specific VSWR correction for a defined cable family because its thermal and voltage model is known. That correction must not be transferred to an unrelated balun or transformer.

A defensible rating record identifies:

• exact circuit, winding interconnection, ferrite parts, conductors, connectors, insulation and enclosure;
• frequency range and complete complex source/load range at declared reference planes;
• SWR plus reflection phase or equivalent complex impedance at the device location;
• forward, reflected, accepted or delivered power—named explicitly;
• modulation, waveform, crest factor, average power and duty cycle;
• differential voltage/current and common-mode voltage/current limits;
• ambient temperature, mounting, cooling, enclosure state and thermal-equilibrium criterion;
• permitted temperature rise, electrical drift, insulation stress and safety margin; and
• measurement uncertainty, sample variation and post-stress acceptance checks.

SWR can be part of that envelope, but it cannot replace it. Two installations with the same SWR can put the device at different voltage/current phases and have completely different common-mode paths, temperatures and duty cycles.

Turn the argument into a useful operating limit

1. Identify the exact topology. Draw the connections and name the wanted differential path, impedance transformation and common-mode-control function separately.
2. Declare the planes and power terms. Put calibration planes at the device ports where practical and distinguish forward, reflected, accepted and delivered power.
3. Map the complex load envelope. Use complex impedance or reflection coefficient, not SWR magnitude alone. Include line lengths or phase states that place high voltage and high current near the device.
4. Measure small-signal behaviour. Record differential match and insertion loss, balance or mode conversion where applicable, and complex common-mode impedance across and beyond the required bands.
5. Calculate first-order stress. Estimate conductor current, dielectric voltage, magnetising current, flux and common-mode loss/voltage for each operating corner.
6. Run controlled power tests. Use representative complex loads, frequency, waveform and duty; increase power in safe steps while instrumenting core, winding, cable, connectors and enclosure.
7. Wait for thermal equilibrium. Record ambient and cooling. Stop on rapid temperature rise, arcing, odour, intermittent match, unexpected common-mode current or electrical drift.
8. Verify the installation. Repeat feed impedance, exterior-current and temperature measurements with the real antenna, feed-line route, mast, earth and nearby wiring.
9. Recheck after stress. Confirm that impedance, transfer, balance, insulation and visual condition remain within the declared acceptance limits.

The better rule: protect the mechanism that can fail

I want a 1:1 current choke selected for the common-mode job and built to carry the actual differential voltage and current. I want a ratio transformer selected for the impedance and voltage range it must transform. If an assembly performs both jobs, both sets of limits count. That is a more useful design rule than reducing every device to one SWR slogan.

For a choke with adequate isolation and adequate conductor, dielectric and thermal margin, a mismatched differential load is not by itself proof of excessive ferrite heating. If the conductors or insulation approach their limits, reduce power or change the arrangement. If common-mode excitation is the problem, improve the isolation and return-path arrangement; making the transmitter’s SWR meter look better does not necessarily remove that excitation.

Source discussion and engineering references

• Gary Baker, K7EMF, Fabricating A Balanced ATU: the matched-side, balance and core-heating explanation addressed here.
• True Ladder Line, Model 4116T description: an example of an assembly-specific mismatch restriction, not a universal device model.
• C. L. Ruthroff, “Some Broad-Band Transformers,” Proceedings of the IRE, 1959: original transmission-line-transformer circuits, frequency limits and balanced/unbalanced configurations.
• Gustav Guanella, US2470307A, “High-frequency matching transformer,” filed 1945: primary series/parallel transmission-line-transformer topology.
• Andrew Roos, ZS1AN, “A Better Antenna-Tuner Balun,” QEX, September/October 2005: circuit analysis of a 1:1 Guanella current balun and a 4:1 Ruthroff voltage balun under balanced and unbalanced loads.
• Würth Elektronik ANP146: measured common- and differential-mode equivalent circuits, leakage, loss, parasitic capacitance and resonance.
• Fair-Rite suppression design considerations: complex permeability and its dependence on frequency, temperature and bias.
• Times Microwave Systems high-power coaxial-cable guidance: cable-specific thermal, voltage, ambient, altitude and VSWR power-rating corrections.
• Keysight reflection-measurement guidance: reflection coefficient, return loss, VSWR and calibration-plane relationships.

Questions, antenna-factor records or height trials to share? Contact RF.Guru.