RF Power Amplifier Impedance Matching at Microwave and mmWave Frequencies

Zachariah Peterson
|  Created: September 25, 2019  |  Updated: September 25, 2020

MOSFET transistor for RF amplifier impedance matching

According to MarketWatch, the overall market for RF amplifiers is expected to surpass $27 billion by 2023. So where are all of these RF amplifiers expected to be used? You can thank 5G and the expansion of cellular networks in general for a healthy portion of the expected growth. For the PCB designers out there, RF amplifier impedance matching becomes an important design point, especially with high power amplifiers.

Large-signal RF Amplifier Impedance Matching

RF power integrity folks are probably familiar with the need for good voltage regulators in mobile devices in order to suppress transient signals through an amplifier’s output, especially when dealing with a pulsed RF power amplifiers. The signal integrity folks who may now start working with RF design are probably used to working with S-parameters at low signal levels when analyzing their RF circuits and determining appropriate impedance matching. The use of S-parameters is not appropriate in Class AB and Class C RF amplifier design as these amplifiers are inherently operating in the nonlinear regime.

In terms of power transfer at low signal levels (i.e., in the linear regime), maximum power transfer is assured when the load impedance is matched to the complex conjugate of the amplifier output impedance. However, a power amplifier (normally placed in the RF transmit section) might provide higher gain and efficiency at the rated output power if there is an intentional impedance mismatch.

When operated at high output power, the amplifier’s output impedance/load impedance match/mismatch that produces maximum power transfer to the load may not coincide with match/mismatch that produces maximum efficiency at the desired frequency (this is certainly true for resistive components). So how can you determine the right matched impedance at the load to ensure you see the best performance? Because the impedance seen by the source depends on the amplifier’s input and output power levels, you will need to use load-pull analysis to determine the appropriate impedance seen by the amplifier’s output. You then need to match the load’s impedance to this value.

There is a rather simple way to do load-pull analysis with a simulator and a Smith chart. The idea is to iterate through a large number of load impedance values (remember, impedance is the sum of resistance and reactance) at a specific input power. You then probe the output current/voltage for each combination of load resistance and reactance, allowing you to also calculate gain and efficiency. You then plot output power contours as a function of load impedance at the particular input power.

This is shown in the Smith chart below: each contour shows the set of resistance and reactance values that produce a specific output power (green) and efficiency (blue). The red contour shows the region where these two sets of curves overlap. You can then determine the tradeoff between output power and efficiency for specific output powers where contours intersect. Note that, at a different input power, you will generate a different set of contours.

Smith chart for RF amplifier impedance matching

Example Smith Chart with results from load-pull analysis for RF amplifier impedance matching [Source]

The combination of reactance and resistance you determine from load-pull results will tell you which matching network you should use to set the load impedance. You can then verify this with vector network analyzer measurements with a test coupon. Pay attention to the behavior of your matching network at high frequencies; in addition to self-resonance (see below), the bandwidth of your matching network may create some problems for FMCW chirped radar. Note that, at 77 GHz, the chirp range can reach 4 GHz, so your bandwidth should be relatively flat from 73 to 81 GHz.

RF Amplifier from ICs and Discrete Components

If your desired IC will not meet your needs and you must design a custom amplifier from discrete components, you’ll have a more difficult time at RF frequencies for a number of reasons. In addition to the nonlinear response of these amplifiers at high power, the actual layout can create signal integrity problems due to impedance mismatches between components. Due to the impedance characteristics of different components, you may not be able to match impedance throughout the amplifier design. This is due to the very short wavelengths of mmWave frequencies (see below).

Before getting into some layout points, let’s look at component selection. Components based on GaN are best for emerging areas of RF design where frequencies span 10-100 GHz (e.g., 5G or other mmWave applications). At lower GHz frequencies, components based on a GaAs process are the best choice. Any capacitors and inductors you use for matching will have some self-resonance frequency; make sure that you choose passive components with sufficiently high self-resonance frequency when building one of these circuits.

At microwave frequencies, your signal wavelengths are on the order of cm (e.g., 6 cm in free space at 5 GHz), so you can probably get away with impedance mismatches when your traces between components are short enough. At mmWave frequencies, it is much more likely that every trace will act like a long transmission line, even if you layout your components in your custom RF amplifier as closely as possible. If there is a mismatch between components, standing waves can form along a trace, either at the desired fundamental frequency or at one or more higher order harmonics. When this happens, your traces start to act like antennas and will radiate strongly.

In this situation, a transmission line architecture like coplanar waveguides will be difficult to implement due to the real estate required for copper on the surface layer, and you’ll need to isolate your amplifier portion of the board to ensure signal integrity. Follow best practices for separating digital and analog ground sections in your ground plane below the surface layer. In multilayer boards with high layer count, Rick Hartley (see slide 55 in this older presentation) recommends placing ground planes on every other layer to provide sufficient shielding and isolation between signal layers. You should also place a copper pour around various RF sections and ground it with vias.

Ground pour for RF amplifier impedance matching

Notice the vias scattered on the surface layer

Be sure to follow some best practices with via spacing and sizing the thickness of the copper pour to shift the lowest order resonance frequency above the RF frequency you are working with. To save some headaches with via stub resonance and backdrilling during manufacturing, you could just use through-hole vias to ground your copper pour. In the most extreme cases, you can use a shielding can to isolate RF sections.

RF amplifier impedance matching can be a difficult prospect, especially with power amplifiers that are decisively nonlinear. The layout, simulation, and signal integrity analysis features in Altium Designer can help you determine the best circuit design and layout choices for your RF amplifier circuits and control impedance in your board.

Now you can download a free trial of Altium Designer and learn more about the industry’s best layout, simulation, and production planning tools. Talk to an Altium expert today to learn more.

About Author

About Author

Zachariah Peterson has an extensive technical background in academia and industry. He currently provides research, design, and marketing services to companies in the electronics industry. Prior to working in the PCB industry, he taught at Portland State University and conducted research on random laser theory, materials, and stability. His background in scientific research spans topics in nanoparticle lasers, electronic and optoelectronic semiconductor devices, environmental sensors, and stochastics. His work has been published in over a dozen peer-reviewed journals and conference proceedings, and he has written 2500+ technical articles on PCB design for a number of companies. He is a member of IEEE Photonics Society, IEEE Electronics Packaging Society, American Physical Society, and the Printed Circuit Engineering Association (PCEA). He previously served as a voting member on the INCITS Quantum Computing Technical Advisory Committee working on technical standards for quantum electronics, and he currently serves on the IEEE P3186 Working Group focused on Port Interface Representing Photonic Signals Using SPICE-class Circuit Simulators.

Related Resources

Related Technical Documentation

Back to Home
Thank you, you are now subscribed to updates.