Cold-gas propulsion produces thrust by opening a valve and expanding stored propellant through a nozzle. It uses no combustion or catalytic decomposition: the available pressure and enthalpy of the stored fluid provide the exhaust energy.1 Cold gas is therefore a non-reacting, pneumatic form of Spacecraft Propulsion in a mechanism-based ontology.

NASA and ESA technology catalogues nevertheless often place cold gas beside monopropellant and bipropellant systems under a broad chemical propulsion heading, while ECSS applies liquid-propulsion requirements except those specific to combustion.234 These are useful programme and standards groupings. They do not mean that a pure cold-gas thruster releases chemical-reaction energy.

Storage and feed architecture

Propellant can be stored as compressed gas, a saturated liquid that vaporises during discharge, or a solid that sublimes when heated. The system typically includes a pressure vessel, isolation and flow-control valves, a regulator or blowdown feed, lines, filters and one or more nozzles.12 The storage state, rather than the name cold gas, determines whether the tank contains gas, liquid and vapour, or solid propellant.

Compressed gas gives simple single-phase operation but needs pressure-vessel volume. A lower-molecular-mass gas can improve specific impulse while worsening volumetric storage. A self-pressurising saturated liquid is denser and can suit a small spacecraft, but pressure and mass flow then depend on temperature and two-phase behaviour.1

GomX-4B illustrates the latter architecture. ESA’s pre-flight design used two tanks feeding two pairs of nominal 1 mN thrusters. Butane was stored as a liquid and vaporised as it left, providing dense storage without combustion.5 NASA’s current review records that the system subsequently flew in 2018; the pre-flight article itself is evidence for design intent rather than achieved on-orbit performance.1

Warm-gas or electrothermal thrusters sit at the boundary with electric propulsion. They heat the propellant electrically without causing a chemical reaction, raising exhaust performance at the cost of spacecraft power and thermal hardware.1 Their power demand and heated flow should not be attributed to a pure cold-gas system.

Performance and applications

Cold gas is attractive for fine attitude control and other manoeuvres needing a small, repeatable impulse bit. Its simple flow path and potential use of inert, low-toxicity propellants can ease integration with a primary payload. Low specific impulse and limited stored mass usually constrain large orbit changes and total impulse.1

NASA’s current Small Spacecraft survey covers cold-gas devices from 10 µN to 3.6 N and from 40 to 110 seconds specific impulse.1 These figures aggregate unlike products, propellants and operating conditions. They are not universal bounds or a performance promise for one design; pressure, temperature, molecular mass, nozzle, duty cycle and blowdown state must accompany a selected value.

Euclid provides a larger precision-pointing example. Before launch, ESA documented six nitrogen cold-gas microthrusters supplied from four high-pressure tanks containing 70 kg of nitrogen, sized for at least six years of planned operation.6 This establishes the flight configuration and design life, not an already achieved lifetime.

Qualification, failure and operations

Cold-gas verification covers proof and leak testing, regulator and valve behaviour, thrust and impulse-bit repeatability across the pressure and temperature envelope, lifetime cycling, plume effects, vibration and thermal-vacuum operation.2 Blowdown operation needs characterisation across the declining tank pressure; a saturated-liquid system also needs representative heat transfer and phase behaviour.

Removing combustion removes ignition and catalyst risks but leaves stored-pressure hazards, leakage, blockage, contamination and valve failure. NASA’s BioSentinel cold-gas system flew in 2022 with six reaction-control thrusters; one valve failed closed during checkout. Operations used the remaining five thrusters and had accumulated 408 firings by May 2023.1 MarCO-B suffered a tank-to-plenum leak accepted before launch and another leak through a thruster valve in flight, creating a continuous disturbance torque even though the mission completed its relay objective.1

Propellant choice also sets ground and end-of-life hazards. Inert or low-toxicity does not remove pressure-vessel rupture, asphyxiation, refrigerant, leakage or passivation concerns. Redundancy, isolation and fault response should follow the mission consequence rather than the apparent simplicity of the device.12

UK capability

Nammo’s current Westcott facility page lists cold-gas thruster acceptance work in its mini high-altitude facility.7 This is a bounded UK test capability. Acceptance at a facility is not the same as qualification or flight heritage; those claims require the named hardware, configuration, campaign and mission evidence.

References

Footnotes

  1. NASA Small Spacecraft Systems Virtual Institute, In-Space Propulsion. ↩ ↩2 ↩3 ↩4 ↩5 ↩6 ↩7 ↩8 ↩9 ↩10

  2. European Cooperation for Space Standardization, ECSS-E-ST-35-01C: Liquid and electric propulsion for spacecraft. ↩ ↩2 ↩3 ↩4

  3. European Space Agency, Technology Harmonisation: CubeSat Propulsion. ↩

  4. European Space Agency, Anatomy of a spacecraft. ↩

  5. European Space Agency, ESA’s next satellite propelled by butane. ↩

  6. European Space Agency, Euclid fuelled for launch. ↩

  7. Nammo, Westcott. ↩