PPU Fundamentals
Why 13.56 MHz? The ISM Band in Plasma Propulsion
Almost every RF plasma thruster in development runs its discharge at 13.56 MHz. The reason is not physics alone. It is a regulatory allocation that created a deep, cheap component ecosystem, and that ecosystem now shapes how RF power processing units are built.
On this page
What the ISM bands are
ISM stands for industrial, scientific and medical. The ISM bands are slices of the radio spectrum that the International Telecommunication Union set aside for uses other than communication: heating, drying, welding, plasma generation, medical diathermy, and laboratory equipment. Inside an ISM band, equipment is allowed to radiate far more energy than a communication transmitter would ever be permitted, because the band is understood to be noisy by design. Devices that operate there have to tolerate interference and must not complain about it.
The commonly used ISM centre frequencies include 13.56 MHz, 27.12 MHz, 40.68 MHz, 433.92 MHz, 915 MHz in some regions, 2.45 GHz, and several higher microwave points. Each has a defined tolerance band around it. The two that matter for plasma work are 13.56 MHz in the high-frequency range and 2.45 GHz in the microwave range. Regional rules differ, so a flight programme still confirms the allocation and emission limits for its operating and ground-test locations, but 13.56 MHz is allocated almost everywhere.
Why 13.56 MHz specifically
Three things pushed RF plasma sources toward 13.56 MHz, and they reinforce each other.
- A mature component ecosystem. The semiconductor industry has driven 13.56 MHz plasma etch and deposition tools for decades. That built a supply base of RF power transistors, matching capacitors, ferrite cores, directional couplers, and test equipment all optimised for this exact frequency. Industrial generator vendors such as MKS Instruments and Advanced Energy sell 13.56 MHz units by the thousand. A propulsion team designing a power stage at this frequency is working with parts that already exist rather than parts it has to characterise from scratch.
- A generous emission allowance. Because 13.56 MHz is an ISM centre, a thruster and its power electronics can leak some RF without breaching a communication-band limit. That relaxes shielding mass and filtering on the power processing unit, though the spacecraft still has its own electromagnetic compatibility budget to protect its radios and payload.
- Good coupling at this wavelength. 13.56 MHz has a free-space wavelength near 22 metres. That is long compared with any thruster antenna, so the antenna behaves as a compact lumped element rather than a distributed structure. The physics of driving current into a low-pressure plasma through an inductive coil or a helicon antenna works well in the low tens of megahertz.
Coupling to inductive and helicon discharges
RF thrusters generally use one of two coupling schemes, and both suit the HF range. In an inductively coupled plasma the RF current flows in a coil wrapped around a dielectric tube. The changing magnetic field drives an azimuthal electric field in the gas that sustains the discharge. In a helicon source a shaped antenna launches a bounded whistler wave along a static magnetic field, which deposits power deep in the plasma column and reaches higher density for the same input.
In both cases the plasma presents the generator with a load that is mostly reactive with a small, variable resistive part representing the power actually absorbed. At 13.56 MHz the coil or antenna inductance and the tuning capacitors land in comfortable, buildable value ranges: tens to hundreds of nanohenries and tens to hundreds of picofarads. Push the frequency far up and those values shrink until stray inductance and parasitic capacitance dominate. Push it far down and the components grow large and heavy. The matching network is where this plays out most directly.
Trade-offs against kHz and 2.45 GHz
| Aspect | Low frequency (kHz to low MHz) | 13.56 MHz ISM | 2.45 GHz ISM |
|---|---|---|---|
| Passive component size | Large inductors and capacitors, heavier | Moderate, buildable lumped elements | Small, but layout parasitics dominate |
| Power semiconductor choice | Wide, mature | Wide, GaN and LDMOS both usable | Narrower, specialised microwave devices |
| Plasma coupling | Capacitive coupling favoured, lower density | Efficient inductive and helicon coupling | Electron cyclotron and microwave coupling, needs waveguide or applicator |
| Matching network | Bulky, slow to retune | Compact, fast adaptive tuning practical | Distributed structures, tuners are mechanically complex |
| Switch-mode amplifier efficiency | Very high, easy | High, Class-E family works well | Falls off, hard switching losses rise |
| Component ecosystem | General industrial | Deep, semiconductor-fab driven | Deep, but microwave-oven and radar driven |
Lower drive frequencies make the amplifier easy but the passives and the antenna large, and capacitive discharges tend to run at lower density with more energetic ion bombardment of internal surfaces. Microwave drive at 2.45 GHz gives small hardware and is a good match for electron cyclotron resonance sources, but the power stage moves into specialised microwave devices, efficiency of a high-power switch-mode amplifier drops, and the matching hardware becomes distributed and mechanically awkward. 13.56 MHz sits in the band where a high-efficiency switch-mode power stage, compact lumped matching, and efficient plasma coupling all coexist.
Harmonics are the main cost of the choice. A switch-mode stage at 13.56 MHz produces energy at 27.12 MHz, 40.68 MHz and beyond. Those fall on higher ISM points, which helps, but the PPU still needs a low-pass or harmonic-trap filter between the amplifier and the antenna to keep the radiated spectrum clean.
What it means for PPU design
Choosing 13.56 MHz sets several downstream decisions for the RF generator and the rest of the unit.
- Amplifier topology. The Class-E and related switch-mode families reach very high efficiency at this frequency with gallium nitride transistors, which is why an efficiency-driven space design lands there.
- Matching network scale. Tuning elements are physically small enough that a real-time adaptive network, with electronically variable reactances, fits within a spacecraft mass and volume budget.
- Filtering. A compact harmonic filter is needed, but its components are standard HF parts.
- Parts availability and cost. Transistors, capacitors, cores and connectors are commodity items, which supports a lower unit cost and a fully European bill of materials.
- Test equipment. Vector network analysers, RF power meters and directional couplers for 13.56 MHz are standard bench gear, so characterisation and acceptance testing do not need custom instrumentation.
MHz Labs designs its RF power processing unit around 13.56 MHz operation for exactly these reasons. The unit is in development along 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, and an integrated real-time adaptive matching network.
Frequently asked questions
Is 13.56 MHz a legal requirement for a plasma thruster?
No. It is a practical convergence, not a rule. A thruster could run at another frequency, but operating on an ISM centre simplifies the emission budget and gives access to a mature component base. Regional spectrum rules and the spacecraft's own electromagnetic compatibility limits still apply.
Why not use 2.45 GHz like a microwave oven?
2.45 GHz suits electron cyclotron resonance sources and keeps hardware small, but the high-power amplifier moves to specialised microwave devices, switch-mode efficiency drops, and the matching and feed hardware becomes distributed and mechanically complex. For inductive and helicon thrusters, 13.56 MHz couples better.
Where does the deep component ecosystem come from?
Semiconductor fabrication. Plasma etch and deposition tools have driven 13.56 MHz generators, matching capacitors and RF transistors for decades, so those parts are commodity items rather than custom builds.
Does running at an ISM frequency mean the thruster can be electrically noisy?
It relaxes the external regulatory limit, but not the engineering. The PPU still needs a harmonic filter and shielding so the thruster does not interfere with the spacecraft's radios, GNSS receiver and payload.
How does the frequency choice affect the matching network?
At 13.56 MHz the tuning inductors and capacitors are small lumped components in buildable value ranges, which makes a compact, electronically tunable adaptive matching network feasible within a spacecraft mass budget.
Sources and further reading
- ITU, International Telecommunication Union, custodian of the international radio regulations that define the ISM allocations.
- MKS Instruments, ELITE 13.56 MHz RF plasma generators, an example of the industrial component and generator base.
- Takahashi, Helicon-type radiofrequency plasma thrusters and magnetic plasma nozzles, a review of RF coupling in plasma propulsion.
Designing a 13.56 MHz power stage for a thruster?
MHz Labs builds the RF power processing unit and adaptive matching network so your propulsion team can focus on the thruster. Ask for the current design targets.
Get in touch