Autotransformer vs Ruthroff Transformer in UNUNs and BALUNs

Tapped autotransformers and Ruthroff transmission-line transformers are often drawn with a common terminal, a tap and a high-voltage terminal. That visual similarity encourages us to treat them as the same circuit—or to invent an isolated “secondary” in the Ruthroff drawing. I separate them by asking how power crosses the network, which conductors are galvanically continuous and which mode each piece of magnetic material is controlling.

A Tapped Autotransformer Shares One Winding

A conventional autotransformer has one continuous winding with at least one tap. The low-voltage port spans part of the winding; the high-voltage port spans more of it. The two ports therefore share conductor and are galvanically connected.

Power reaches the load by two coupled mechanisms. Part is conducted directly through the common winding section, and part is transferred through magnetic flux linking the series and common portions. In the ideal lumped model, voltage follows the declared turn counts:

A two-to-one turns ratio therefore gives an ideal four-to-one impedance ratio; a three-to-one turns ratio gives an ideal nine-to-one ratio. Reversing the ports reverses the step-up or step-down direction. These equations describe an ideal relationship under a defined termination, not a 50-ohm guarantee, bandwidth claim or power rating.

At HF, the winding also has leakage inductance, conductor resistance, inter-turn capacitance and distributed electrical length. Core complex permeability and loss change with frequency, field and temperature. A physically compact tapped winding may behave like a useful lumped transformer over one range and depart sharply from that model elsewhere.

A Ruthroff Transformer Uses Transmission-Line Voltage Addition

Clyde Ruthroff's 1959 work describes a family of broadband transformer and hybrid circuits built from closely coupled conductors that act as transmission lines. In the classic 4:1 Ruthroff connection, the intended line mode transfers a voltage with the required polarity and that voltage is added in series with a directly connected port voltage. Ideally the high-voltage port then sees twice the low-port voltage and four times its impedance.

The winding may look autotransformer-like when drawn as a circuit, but the intended RF transfer includes propagation on the two-conductor line. Characteristic impedance, termination and electrical delay are therefore central. The magnetic material helps the structure support the required common-mode or magnetising impedance; it is not merely a core coupling two isolated windings.

Many classic Ruthroff connections have a direct conductive path between input and output. They do not contain an isolated secondary and do not provide galvanic isolation. If a specific circuit does provide isolation through another arrangement, that must come from its exact schematic and insulation system—not from the word Ruthroff.

The Two Circuits Can Share a Ratio Without Sharing Behaviour

The table is not a winner's scorecard. A well-executed autotransformer can be compact and efficient for a defined transformation. A Ruthroff design can maintain its voltage relationship over a broad useful range. Either can fail when the complex load, electrical length, parasitics or common-mode path fall outside the measured design window.

Nominal Ratio Is Not an Antenna Prescription

A nominal 4:1 transformer maps an impedance by approximately four only while its own errors remain small. It does not establish that an off-centre dipole is 200 Ω, that an end-fed wire is 200 Ω, or that the transformed result is 50+j0 Ω. The same warning applies to 9:1, 49:1 and 64:1 labels.

The antenna presents a frequency-dependent complex load at a specific reference plane. Height, geometry, soil, counterpoise or radial system, feedline route, nearby conductors and weather can all change it. The transformer then adds its own loss and reactance. Select the nominal ratio from a measured load region, not from the antenna's nickname.

A tuner can transform the radio-side impedance after either device. That does not prove low transformer loss, low feedline loss, correct current balance or acceptable internal voltage. Keep the measurement planes explicit and separate the tuner's successful match from the transformer's power balance.

Continuity Is Not the Same as RF Return Control

Both circuit families can provide DC continuity between ports. That may be operationally useful, but it is not galvanic isolation and it does not tell us where RF return current flows. The source shield, enclosure, counterpoise, radials, coax exterior, mast, earth coupling, bonding and station wiring can all participate.

An UNUN has intentionally unbalanced external ports. It still needs a declared return-current structure. If the coax exterior is part of that structure until a choke, its route and length become part of the antenna. If it is not intended to carry antenna current, the common-mode boundary may belong close to the transformer. No universal placement works for both cases.

Protective earth and lightning bonding are separate safety systems. A transformer or choke does not replace them, and their conductors must not be removed to alter RF behaviour. Conversely, a DC-continuity check through a winding does not prove an adequate low-impedance lightning path.

Neither Topology Guarantees Common-Mode Suppression

A common-mode choke adds a complex impedance to an in-phase current path while passing the wanted differential transmission-line mode. A tapped autotransformer does not inherently do that. A Ruthroff winding may present impedance to some common-mode excitation, but the amount and location depend on its interconnection, core, frequency, enclosure and attached conductors.

This is why I prefer to specify transformation and choking separately when the installation does not preserve textbook symmetry. A measured-load UNUN handles the differential impedance step. A separately measured 1:1 choke defines the exterior-current boundary. Each device can then be selected, placed and tested for one clear job.

That arrangement is not a declaration that current baluns are wrong. A genuinely balanced installed load can be fed through a suitable current balun whose transformation, balance, common-mode impedance and stress performance are all verified. An integrated transformer-plus-choke assembly is equally valid when its differential and common-mode functions remain adequate together.

Parasitics Decide the Useful Load Domain

Load domain means more than a resistance range. It includes reactance, frequency, reference plane, source impedance and the resulting internal voltage and current. A transformer optimized into one resistor may behave very differently into a high-Q reactive antenna.

For the tapped autotransformer, magnetising current and core loss can dominate at the low edge, while leakage inductance and winding capacitance distort the high edge. The shared winding can expose sections to large algebraic current even when port power seems modest.

For the Ruthroff circuit, the line should see a termination compatible with its characteristic impedance and intended voltage relationship. Electrical delay becomes significant as the line length approaches a meaningful fraction of wavelength. Phase error, response peaks or nulls, unequal conductor environment and interconnection inductance can then spoil the nominal ratio.

No general statement that an autotransformer has fewer losses or that a Ruthroff transformer is broader can survive all these variables. Compare finished devices at the same source, complex loads, frequency, power, duty cycle and temperature.

Voltage, Current and Heat Need Separate Limits

At the high-impedance port, voltage tends to rise for a given transferred power. At the low-impedance port, current tends to rise. Reactive loads and standing waves can increase internal stress beyond the ideal resistive-ratio estimate. A common winding, tap, line crossover, connector or enclosure can become the limiting point.

Ferrite loss is frequency-, field- and temperature-dependent. Differential current heats the conductors and connections; common-mode current can drive magnetic loss; parasitic capacitance concentrates electric-field stress. A small-signal S-parameter sweep cannot establish a power rating.

Powered testing must declare accepted power, waveform, duty cycle, complex load, mismatch, ambient, enclosure, cooling, duration and allowed temperature rise. Measure impedance drift as the device warms and stop for rapid heating, arcing, odour or unstable results.

Measure the Functions Instead of the Label

• Draw the exact circuit. Include every tap, winding start, line conductor, chassis, shield and external return path.
• Declare the logical ports. State which are balanced or unbalanced and where voltage, current and impedance are referenced.
• Measure the antenna load. Capture complex impedance at the intended transformer plane over all bands and relevant geometries.
• Measure differential transformation. Save complex input/output ratio, return loss, insertion loss and phase into representative resistive and reactive loads.
• Test balance and mode conversion. Use calibrated multiport or equivalent measurements to distinguish differential transfer from differential-to-common conversion.
• Characterise the common-mode path. Measure full complex choke impedance with a validated fixture and map installed exterior current at repeatable positions.
• Test the cascade. If transformer and choke are separate, include their jumper, order, enclosure and parasitic coupling in the final measurement.
• Prove stress margin. Run controlled electrical and thermal tests at the real waveform, load, duty and ambient conditions.

The right transformer is the one whose measured circuit behaviour fits the measured load. The right choke is the one that controls the intended common-mode boundary. Sometimes those are one assembly; often they are easier to prove as two.

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