Amplitude-Compensated Combiner
Unequal Wilkinson Combiner Equations
For a compensated combiner splitting power by a ratio of K² (where P2 / P3 = K²), the physical impedances must be scaled relative to the system impedance (Z0 = 50 Ω).
| Component | Equation (Normalized to Z0) | Example: 2:1 Ratio (K² = 2) |
|---|---|---|
| Arm 1 Impedance (Z2) | Z0 · √(K · (1 + K²)) | Z2 = 50 · √(1.414 · 3) = 103.0 Ω |
| Arm 2 Impedance (Z3) | Z0 · √((1 + K²) / K) | Z3 = 50 · √(3 / 1.414) = 72.8 Ω |
| Isolation Resistor (Riso) | Z0 · (K + 1/K) | Riso = 50 · (1.414 + 0.707) = 106.1 Ω |
| Output Matching (Zout) | Usually requires a transformer | Must match to Z0 |
Ploss = (V1 − V2)² / Riso
If V1 ≠ V2, power is dissipated. The compensated combiner forces V1 = V2 by altering the transmission line transformation ratio on each arm.
Frequently Asked Questions
Why not use a standard Wilkinson?
Standard Wilkinson combiners assume equal amplitude. If fed with unequal signals, a voltage differential forms across the isolation resistor. Current flows through it, turning RF power into heat. This ruins efficiency and often burns out the resistor.
How does it compensate?
By scaling the characteristic impedance of the arms. The arm carrying more power is designed with a lower impedance (wider microstrip trace), and the arm carrying less power is designed with a higher impedance (narrower trace). This forces the voltages at the isolation resistor to be equal, meaning no current flows and no power is lost.
Where are they used?
Primarily in asymmetric Doherty amplifiers (where peaking and carrier stages have different power ratings) and in corporate feed networks for phased arrays utilizing amplitude tapering (where edge elements transmit less power than center elements to reduce sidelobes).