RF PPU for CubeSat and SmallSat Electric Propulsion

The small-spacecraft power and volume reality

A 6U to 16U CubeSat or an ESPA-class SmallSat lives inside hard limits. Solar arrays generate a modest orbit-average power, often a few tens to a few hundred watts, and the propulsion system only gets what is left after the bus, avionics and payload take their share. The power bus is frequently unregulated, so voltage sags and swings with battery state of charge and eclipse. There is no room for a large radiator, so waste heat leaves mostly by conduction into the structure and then by radiation from whatever external area is available.

The RF power processing unit has to take that imperfect input and deliver a clean, stable RF drive to the thruster antenna through a matching network, while fitting in a slot measured in fractions of a U and adding as little mass as possible. Every design decision trades against the payload.

Why efficiency and power density matter more here

Efficiency has two compounding effects on a small platform. First, the power the PPU wastes is power the thruster never sees, which directly cuts thrust and total impulse for a fixed array. Second, every wasted watt becomes heat that has to be conducted away through a structure with very little thermal margin. A unit at 75 percent efficiency and a unit at greater than 90 percent efficiency, both delivering 200 W of RF, differ by more than 40 W of dissipation. On a large satellite that is a radiator sizing note. On a CubeSat it can be the difference between a thermally viable design and one that throttles or shuts down.

• Thrust budget. Lost DC power is lost thrust. Efficiency sets how much of the array actually reaches the plasma.
• Thermal survival. Conduction cooling has a low ceiling. Lower loss means the junction and baseplate temperatures stay in range without a dedicated radiator.
• Volume. A higher-efficiency switch-mode stage needs less heatsinking and fewer bulk components, so power density rises as efficiency rises.
• Duty cycle. A cooler unit can fire longer per orbit, which matters for orbit raising and drag make-up timelines.

This is the main reason small-spacecraft RF PPUs favour gallium nitride power transistors in a Class-E or related switch-mode topology. GaN offers roughly three times the power density of silicon at these frequencies and low switching loss, which is what lets the unit hit a high-efficiency target in a small envelope. The RF generator section of the guide covers the topology in more detail.

Form factor and integration constraints

Mechanically, a small-spacecraft PPU is usually expected to fit a PC104-style board footprint or a slim standalone module that bolts to a structural panel with a defined conduction interface. The connector set has to be small and lightweight, the board stack has to survive launch vibration without potting the whole assembly, and the thermal path from the transistors to the mounting feet has to be short and well-defined. Harness length between the PPU, the matching network and the antenna matters at 13.56 MHz because cable reactance becomes part of the tuned circuit.

The 50 to 500 W RF band

Many CubeSat and SmallSat RF plasma thrusters in development sit in the 50 W to 500 W RF range. Much below that, an RF discharge is hard to sustain efficiently and thrust is very low. Much above it, a small platform usually cannot supply the DC input or reject the heat. A PPU that covers this full band with a single architecture lets an integrator use one qualified unit across a product line rather than requalifying a new design for each mission.

Within the band, the useful capability is smooth throttling. A thruster is rarely run flat out. It is commanded to a thrust level that suits the current manoeuvre, so the PPU needs to hold efficiency and a stable match across a wide output range, not just at one design point. This is where an adaptive matching network earns its mass: it keeps the amplifier seeing a good load as the plasma impedance moves with power level, gas flow and background pressure.

What an integrator should look for

• Efficiency stated across the range, not just a peak number at one operating point.
• Unregulated bus tolerance, with a defined input voltage window that matches your battery chemistry and eclipse behaviour.
• A defined conduction-cooling interface: mounting face, allowable baseplate temperature, and dissipation at each RF level.
• Mass and envelope that fit your slot, including connectors and any separate matching unit.
• A standard data interface and a documented telemetry and command set, so integration does not need bespoke ground-support equipment.
• Autonomous fault handling for reflected power, arc events and over-temperature, so a bad match does not damage the amplifier.
• Export-control clarity on the bill of materials, which is simpler with a fully European supply chain. See the ITAR-free RF PPU page.
• An honest TRL and test evidence. Ask what has been demonstrated and at what level.

MHz Labs is developing its RF power processing unit and adaptive matching network for this class of spacecraft. The unit is on a TRL 4 to 6 roadmap and has no flight heritage yet. Its design targets include an RF output range of 50 to 500 W, 13.56 MHz operation, conduction cooling, and a fully European supply chain. On mass, volume and cost it aims at the EU DEEP-PPU benchmark: roughly 30 to 40 percent below legacy space PPUs on mass and volume, 35 to 50 percent on cost.

Sources and further reading

• Goebel and Katz, Fundamentals of Electric Propulsion, NASA JPL DESCANSO, for power-budget and efficiency context.
• ThrustMe, RF acceleration for small-spacecraft electric propulsion, an example of the small-satellite RF thruster class.
• SatNow, What is a power processing unit (PPU) for thrusters, an introductory overview.