Space hardware: qualifying for a mission with no repair option

Testing consumes more schedule than building

A flight unit spends far more of its programme in test than on the bench. Qualification typically means vibration and acoustic exposure, thermal cycling under vacuum, shock, electromagnetic compatibility and functional performance verified at every stage, on a dedicated article built to flight standard. Each anomaly triggers investigation, disposition and often partial retest. Scheduling is consequently dominated by chamber and shaker availability rather than by machining or assembly hours. Programmes that plan backwards from a launch date without protecting test slots discover that the facility, not the factory, holds their critical path.

Constellations forced a move to serial production

Building a handful of exquisite satellites and building hundreds of similar ones are different manufacturing problems. Serial constellation work brings flow lines, standardised harnesses, automated test sequences and statistical process control that traditional space programmes never needed, while accepting a different reliability philosophy: redundancy across the fleet instead of exhaustive qualification of every unit. That shift changes supplier expectations too, since a components vendor used to small orders now faces schedule commitments. Firms that have made the transition describe it less as scaling up and more as rebuilding the engineering approach around repeatability.

Component screening substitutes for field experience

Electronics destined for orbit face radiation, thermal extremes and no possibility of replacement, and there is no field population to learn from. The industry compensates by screening: lot testing, burn-in, radiation characterisation, construction analysis and traceability of every device back to its wafer lot. Using commercially available parts is possible and increasingly common, but only with an upfront screening and derating programme the buyer funds. The commercial trap lies in assuming a cheap component stays cheap; once screened, characterised and documented, its delivered cost bears little resemblance to the catalogue price.

Facilities are a qualification asset, not overhead

Clean assembly areas with controlled particle counts, thermal vacuum chambers large enough for the article, vibration tables rated for its mass, and anechoic space for radio testing are capital items with long build times. They also carry recurring cost regardless of workload, since environmental control cannot be switched off between programmes. Smaller firms rent capacity, which works until schedules collide and no slot exists when a programme needs one. Facility strategy is a genuine commercial decision, and buyers should probe it: a supplier without controlled access to test capacity is quoting someone else's queue.

Contract structure decides who pays for a rework loop

Space contracts allocate the cost of failure differently depending on maturity. Development work is often reimbursed against effort, which shelters the supplier while transferring schedule risk to the customer. Firm-price production contracts reverse that, and a single rework loop triggered by a test anomaly can absorb the programme's entire margin. Between the two sit milestone-linked incentive structures. The practical guidance is unromantic: before accepting a fixed price, be explicit about how many test failures that price assumes, who owns the investigation, and whether retest of unaffected articles is chargeable. Suppliers who leave those points unspoken tend to discover them during the first anomaly review.

Frequently asked questions

Why must qualification testing be repeated for each build standard?

Because the qualification argument applies to a specific configuration, and small changes can alter structural response, thermal behaviour or electromagnetic characteristics in ways analysis alone cannot settle. Different harness routing changes resonance; a substituted component changes thermal load. Programmes handle this by defining a qualification baseline and controlling deviations tightly, accepting analysis or reduced testing only where similarity is demonstrable. Uncontrolled minor changes are how a schedule quietly acquires an extra test campaign nobody budgeted for.

Are commercially available components acceptable in space hardware?

Frequently yes, and their use has grown considerably alongside constellation programmes. The condition is that the buyer takes on characterisation work the component's maker never did: radiation testing, lot traceability, screening, derating and construction analysis. That effort must be repeated whenever the vendor changes a process or a die. The approach suits missions where fleet redundancy absorbs individual failures, and suits poorly where one unit has to survive alone for a long duration.

What limits how quickly a satellite production line can scale?

Test capacity usually binds first, followed by qualified component supply and then by trained integration staff. Chambers and shakers are expensive and slow to install, and screened parts carry lead times that lengthen whenever several programmes expand together. Cleanroom floor area is the other constraint, since integration bays cannot safely be compressed. Realistic scaling plans commit to facility and component procurement well ahead of the order book, which is uncomfortable but difficult to avoid.