Structures and mechanisms
DraftA spacecraft structure has two jobs that pull in opposite directions. It must survive a few minutes of the most violent environment it will ever experience, and then be as light as possible for the years that follow, during which almost no load acts on it at all.
Launch loads
| Load | Magnitude | Frequency | Source |
|---|---|---|---|
| Quasi-static | 3 to 8 g axial, 1 to 3 g lateral | Steady | Vehicle acceleration, peaking at maximum dynamic pressure and before staging |
| Sine vibration | Up to a few g | 5 to 100 Hz | Vehicle structural modes, engine transients, pogo |
| Acoustic | 140 to 145 dB overall sound pressure level | 20 Hz to 10 kHz | Engine noise reflected from the pad at liftoff, and aerodynamic noise transonically |
| Random vibration | 10 to 15 g RMS | 20 to 2000 Hz | Acoustic energy transmitted through the structure |
| Shock | Up to several thousand g in the shock response spectrum | 100 Hz to 10 kHz | Pyrotechnic separation of stages, fairing and spacecraft |
The distinction between the acoustic and random vibration entries matters. Acoustic loading acts on area, so large lightweight panels such as solar arrays, antenna reflectors and thermal blankets are driven by it. Random vibration acts through the mount, so dense compact units such as batteries and electronics boxes are driven by it. The same launch produces both, and a component is qualified against whichever is worse for its form.
Pyroshock is a distinctive problem because it is high-frequency and high-amplitude but carries little energy. It rarely damages structure and routinely damages brittle components: crystal oscillators, relays, ceramic capacitors and optics mounts. Shock qualification is separate from vibration qualification for this reason.
Stiffness, not strength
The requirement that most often drives structural design is not strength but frequency. Launch vehicle user guides specify a minimum first natural frequency for the payload, typically above 10 to 15 Hz laterally and 25 to 35 Hz axially, depending on the vehicle.
The reason is dynamic coupling. If a spacecraft mode falls near a launch vehicle mode, the two resonate together and loads amplify far beyond the input. Keeping the payload’s fundamental frequency above the vehicle’s excitation band decouples them, so the spacecraft rides quasi-statically rather than resonating.
This is a genuine design driver rather than a formality. A structure that is strong enough but too flexible fails the requirement and must be stiffened, which adds mass without adding strength. Since stiffness scales with the cube of depth for a panel in bending while mass scales linearly, the standard answer is a sandwich construction: thin strong facesheets separated by a light core.
| Material | Use | Property that earns it |
|---|---|---|
| Aluminium 6061 and 7075 | Primary structure, fittings | Cheap, well characterised, easy to machine |
| Aluminium-lithium | Tanks, primary structure | Roughly 5% lower density and higher stiffness than conventional alloys |
| Carbon fibre composite | Optical benches, tubes, panels | High specific stiffness, and a tailorable near-zero coefficient of thermal expansion |
| Titanium | Fittings, tanks, fasteners | Strength at temperature, and low thermal conductivity where isolation is wanted |
| Aluminium honeycomb | Sandwich cores | Very high stiffness per unit mass |
| Invar | Optical mounts | Extremely low thermal expansion |
The near-zero expansion property of composites is what makes precision instruments possible. An optical bench that changes length with temperature defocuses the instrument, so the structure is laid up with fibre orientations chosen to cancel thermal expansion over the operating range. That is a structural solution to an optical problem, and it is why structures and payload design cannot be separated on precision missions.
Structure typically accounts for 20 to 25% of spacecraft dry mass, which makes it one of the largest single allocations in the mass budget.
Deployables
Almost every spacecraft folds to fit a fairing and unfolds once in orbit. Solar arrays, antenna reflectors, booms, sunshields and radiators all deploy, and they deploy once, unattended, after months of storage and a violent launch.
The reliability problem is unusual: a mechanism that has to work a single time cannot be tested in its flight configuration without consuming that single operation. The standard resolution is to test a qualification unit to destruction, test the flight unit in gravity-offloaded conditions that approximate zero g, and accept that the true flight event is unrepeated.
| Release device | Mechanism | Character |
|---|---|---|
| Pyrotechnic bolt cutter or separation nut | Explosive charge severs or releases | Very reliable, high shock, single use, hazardous ground handling |
| Shape memory alloy actuator | Heated alloy element changes shape and releases | Low shock, resettable in ground test, slower |
| Split spool | Redundant heaters release a wound restraint | Low shock, resettable |
| Burn wire | Current melts a restraining line | Simple and cheap, common on CubeSats |
The shift toward non-explosive actuators is driven by the shock column. A pyrotechnic device solves the release problem and creates a shock problem for everything nearby, so removing the shock at source is often cheaper than qualifying the neighbours to survive it.
JWST is the extreme case of deployment risk: 178 release mechanisms and 344 single-point failures, meaning 344 events any one of which would have ended the mission. All of them worked. That programme is simultaneously the strongest argument for careful mechanism engineering and the strongest argument for in-space assembly, which would have avoided the need for most of them.
Verification
The coupled loads analysis step is a formal exchange between the spacecraft team and the launch provider. The spacecraft’s reduced finite element model is integrated into the launch vehicle model and the combined system is run through the flight event, producing the actual loads the spacecraft will see. Those are usually different, and sometimes much higher, than the envelope in the user guide, because they depend on how the two structures interact.
Model correlation appears here for the same reason it appears in thermal control: a model that has not been matched to measured data is a prediction, not a verification, and measured structural frequencies routinely differ from predicted ones by 10% or more because joints and interfaces are hard to model.
References
- Sarafin, T. P. (ed.) Spacecraft Structures and Mechanisms: From Concept to Launch, Microcosm Press, 1995. The standard reference.
- Wijker, J. J. Spacecraft Structures, Springer, 2008.
- NASA-STD-5001, Structural Design and Test Factors of Safety for Spaceflight Hardware, and NASA-STD-7003, Pyroshock Test Criteria.
- ECSS-E-ST-32C, Space engineering: Structural general requirements.
- Menzel, M. et al. “The Design, Verification, and Performance of the James Webb Space Telescope”, Publications of the Astronomical Society of the Pacific, 135:058002, 2023.