Doherty Combiner

The combiner incorporates an impedance inverter, typically a quarter-wave transmission line (λ/4), on the carrier amplifier's output path. This network transforms the load impedance seen by the carrier device. As the peaking amplifier injects current, the impedance at the combining node decreases. The inverter translates this decreasing node impedance into an increasing impedance presented to the carrier transistor's intrinsic current source, enabling active load-pull.

• Characteristic impedance: Z₀ = R_opt (optimal load for carrier at peak power)
• At 6-dB back-off: Carrier sees 2 × R_opt, maximizing efficiency
• At peak power: Carrier sees R_opt, delivering maximum output

The combiner enables active load modulation through current injection from the peaking amplifier. Unlike passive matching networks that present a fixed impedance, the Doherty combiner creates a dynamic impedance environment. The peaking amplifier acts as a controlled current source that actively pulls the load impedance seen by the carrier.

• Low power: Peaking off (high-Z state), carrier sees modulated high impedance
• Transition region: Peaking begins conducting, impedance starts shifting
• Peak power: Both amplifiers contribute equally, carrier sees optimal R_opt
• The ratio of peaking-to-carrier current determines the instantaneous modulation depth

Proper power combining demands precise phase coherence between the carrier and peaking branches at the combining node. The combiner network must account for the 90° phase shift introduced by the quarter-wave impedance inverter on the carrier path. A corresponding offset line is typically added to the peaking amplifier's output to ensure both signal paths arrive in-phase at the summation point.

• Carrier path: Includes λ/4 inverter (+90° phase shift)
• Peaking path: Requires compensating offset line (+90° equivalent)
• Phase error < 5° required for minimal combining loss
• Misalignment causes efficiency degradation and AM-AM distortion

Modern Doherty combiners often integrate post-matching networks that absorb the parasitic output capacitances of the transistors and present the required harmonic terminations. The combiner is no longer a simple λ/4 line but a multi-stage network that simultaneously performs impedance inversion, harmonic control, and broadband matching.

• Fundamental matching: Presents optimal R_opt at carrier frequency
• 2nd harmonic termination: Short-circuit for Class-F⁻¹ operation
• 3rd harmonic termination: Open-circuit for voltage peaking
• Absorbs transistor C_ds into the network design

In asymmetric Doherty configurations where the peaking amplifier has higher power capability than the carrier, the combiner's characteristic impedance must be scaled accordingly. The impedance inverter's Z₀ is set to R_opt_carrier (not a simple 50Ω), and the power ratio determines the combining node impedance transformation.

• 1:1 symmetric: Both amplifiers equal, inverter Z₀ = R_opt
• 1:2 asymmetric: Peaking 2× carrier power, extended back-off range
• Combiner must handle unequal current contributions
• Enables efficiency enhancement beyond 9-dB back-off

Conventional λ/4 impedance inverters are inherently narrowband, limiting Doherty operation to ~10-15% fractional bandwidth. Advanced combiner topologies address this through multi-section transformers, Klopfenstein tapers, or coupled-line structures that maintain the required impedance inversion and phase characteristics over wider frequency ranges.

• Multi-section λ/4 transformers: Chebyshev or binomial responses
• Continuously tapered lines: Klopfenstein for optimal broadband matching
• Coupled-line combiners: Compact implementation with inherent phase shift
• Post-matching Doherty: Separates matching from combining for bandwidth extension

The fundamental operating principle enabled by the Doherty combiner. As the peaking amplifier activates and injects current into the combining node, the impedance inverter dynamically transforms the load impedance seen by the carrier amplifier. This active load-pull effect maintains the carrier at peak voltage swing and high efficiency over a wide range of output power back-off levels. Without precise impedance transformation within the combiner, load modulation fails, collapsing the amplifier's back-off efficiency.

The core functional block within the Doherty combiner, typically realized as a quarter-wave transmission line (90-degree electrical length at the center frequency). Its defining characteristic is the inversion of impedance: a high impedance at one port is transformed to a low impedance at the other. This property is critical for converting the peaking amplifier's increasing current into a decreasing load impedance for the carrier amplifier, enabling the active load-pull mechanism central to Doherty operation.

The critical calibration of electrical path lengths at the input and output of the carrier and peaking branches to ensure constructive in-phase power combining at the Doherty combiner output. A phase mismatch at the combining node results in power cancellation rather than summation, directly degrading power-added efficiency (PAE) and output power. Input-side phase alignment is equally critical to ensure both amplifiers receive coherent drive signals, preventing AM-AM and AM-PM distortion from being compounded by timing errors.

The deviation from the ideal gain ratio between the carrier and peaking amplifier paths. In a symmetric Doherty, the peaking amplifier should deliver current equal to the carrier at peak power. Gain mismatch causes suboptimal load modulation: if the peaking gain is too low, the carrier load line is not fully modulated, reducing back-off efficiency. If too high, the carrier is over-driven, causing premature gain compression and severe nonlinearity that burdens the digital predistortion system.

The intentional presentation of specific short-circuit or open-circuit impedances at harmonic frequencies (2f₀, 3f₀) to the transistor's intrinsic current source. While primarily a function of the matching networks, the Doherty combiner's frequency response influences harmonic impedances seen by both amplifiers. Proper harmonic termination shapes voltage and current waveforms to minimize overlap, reducing dissipated power and enabling Class-F or inverse Class-F operation for enhanced efficiency beyond the theoretical Doherty limit.

The large-signal output reflection coefficient of a power amplifier measured under nominal drive conditions. Unlike the small-signal S22 parameter, Hot S22 captures the impedance presented by the amplifier under realistic operating conditions where nonlinearities and self-heating effects are present. Accurate characterization of the carrier and peaking amplifiers' Hot S22 is essential for designing the Doherty combiner, as the impedance inverter must transform these dynamic, power-dependent impedances to achieve optimal load modulation across the entire power range.