How do I design the output matching network of a power amplifier for maximum power transfer?
Output Matching Network Design
The output matching network is the most critical component in PA design because it determines the power, efficiency, and linearity simultaneously. The optimum load impedance for maximum power is not the conjugate match of the transistor's output impedance; it is determined by the loadline analysis or load-pull measurement.
| Parameter | LNA | Driver | Power Amplifier |
|---|---|---|---|
| Noise Figure | 0.3-2.0 dB | 3-8 dB | 5-15 dB (not specified) |
| Gain | 10-25 dB | 10-20 dB | 8-15 dB |
| P1dB | -10 to +10 dBm | +15 to +25 dBm | +30 to +50 dBm |
| OIP3 | +5 to +25 dBm | +25 to +40 dBm | +40 to +55 dBm |
| DC Power | 10-100 mW | 0.5-5 W | 5-500 W |
Bias and Operating Point
Load-pull data provides the contours of constant output power, efficiency, and ACPR on the Smith Chart. The matching network must present the load impedance at the intersection of the desired power and efficiency contours to the transistor's drain (or collector) reference plane. The optimum impedances for power and efficiency are usually close but not identical; the designer selects a compromise point.
- Performance verification: confirm specifications against the application requirements before finalizing the design
- Environmental factors: temperature range, humidity, and vibration affect long-term reliability and parameter drift
- Cost vs. performance: evaluate whether the application demands premium components or standard commercial grades
- Interface compatibility: verify impedance, connector type, and mechanical form factor match the system architecture
- Margin allocation: include sufficient design margin to account for manufacturing tolerances and aging effects
Stability Considerations
For narrowband applications (< 10% bandwidth), a two-element (L-section) matching network suffices. For moderate bandwidth (10-30%), a three-element pi or T-network provides the needed impedance transformation with acceptable loss. For wideband PAs (> 30%), distributed matching using quarter-wave transformers, Chebyshev multi-section transformers, or real-frequency technique designs are needed.
Frequently Asked Questions
Lumped or distributed matching?
Below 3 GHz: lumped elements (chip capacitors and inductors or bond-wire inductors) provide compact, narrowband matching. Above 3 GHz: distributed elements (microstrip stubs and transformers) are preferred due to lumped element parasitics. At mmWave: MOM capacitors and thin-film inductors integrated on the MMIC are standard.
What about harmonic termination?
Controlling the impedance at the second and third harmonics (2f0, 3f0) significantly affects efficiency. Class F amplifiers require a short at 2f0 and an open at 3f0; inverse Class F reverses these. Adding harmonic traps (series LC resonators) to the output matching network provides the required harmonic terminations.
How sensitive is the PA to load mismatch?
Very sensitive. A load VSWR of 2:1 can change the output power by ±1 dB, the efficiency by ±5%, and the ACPR by ±3 dB. Worse, certain load phases can cause device failure by exceeding the breakdown voltage. Isolators or circulators between the PA and antenna protect against load mismatch in base station applications.