1. The ship that had to look straight at the Sun

Alex Garland’s Sunshine returns as the recalled work because the Icarus II is, at its core, a power-management problem wrapped in a psychological thriller. A shield the size of a city must stay pointed at the Sun so that a small disc of collectors behind it can harvest enough energy to restart a star. The geometry is implausible at that scale, but the emotional truth is familiar: if the panels stop looking at the Sun, everything else stops soon after.

This entry wonders how the desktop should arrange and steer its own solar collectors without a human watching them.

2. Body-mounted, deployable, or something stranger

The reading arc established three coarse architectures for space solar arrays: body-mounted, deployable rigid panels, and flexible roll-out blankets. The wondering begins with a question that sounds like furniture arrangement but is actually mission-critical: how many axes of freedom should the collectors have?

A body-mounted panel is the couch that cannot move. It is simple, stiff, and thermally well integrated, but it only sees the Sun when the whole spacecraft happens to face the right way. That can work if the desktop’s primary attitude is Sun-pointing, or if the power budget is generous enough to tolerate long off-axis intervals.

A single-axis deployable wing is the standing desk: one degree of freedom, better than nothing, still limited. It can track the Sun across one plane while the bus does something else, which is useful if the payload wants nadir and the power system wants the Sun.

A two-axis gimballed array is the full ergonomic setup. It can keep the panel normal to the Sun vector almost continuously, maximizing collection and simplifying the thermal picture because the illuminated face is predictable. The cost is mechanism mass, cable flexure, lubrication or dry-bearing lifetime, and the need for the attitude computer to know where the Sun is and where the panels are pointing.

There is also the exotic option: concentrators, inflatables, or arrays mounted on the thermal shield itself. These are interesting because the desktop already has a large surface that must face the ram direction at some phases of flight. If that shield could also host photovoltaic collectors, the spacecraft would carry less dedicated array structure. The catch is that the shield’s job is to reject heat, not collect it, and its orientation is driven by aerodynamics and thermal survival, not by solar geometry.

3. Tracking without losing the plot

Sun tracking sounds easy because the Sun is bright. In practice the desktop must solve several sub-problems in a loop:

  • Where is the Sun? A coarse sun sensor gives a vector. An orbit propagator predicts it. Neither is perfect; together they are usually good enough.
  • Where are the panels? The array drive electronics report angle. The kinematic chain from bus attitude to panel normal must be calibrated.
  • What is in the way? Earth, the spacecraft body, antennas, payloads, and the thermal shield can shadow the array. The tracker must know when to expect occlusion and not chase a ghost target.
  • How fast can we move? Gimbal rates are limited. If the bus is slewing rapidly, the array may not keep up. The autonomy must plan slews that do not outrun the power system.
  • What if the mechanism jams? A stuck array is worse than a fixed one because it becomes an asymmetric disturbance and a thermal surprise. The controller should detect stall, current anomalies, or encoder disagreement and fall back to a safe configuration.

A useful design principle emerges: track the Sun not as an end in itself, but as one term in a value function that also includes payload pointing, communications, momentum, and thermal state. Sometimes the best move is to accept lower power collection because the array must be stowed for a maneuver or because the bus must point elsewhere to downlink critical data.

4. The geometry of eclipse and edge effects

LEO eclipses are predictable but brutal. The array goes from full sun to full shadow in seconds, and the battery must take over instantly. But there are also partial-shadow regimes: Earth’s penumbra, structural shadowing, and the shallow angles near sunrise and sunset. At low Sun angles the effective area of the array drops by the cosine of the incidence angle and the cell current drops with it.

An autonomous power manager should therefore forecast array output using the same orbit geometry it uses for pointing. The forecast feeds the battery state-of-charge predictor and the load-shedding policy. A panel that is about to spend twenty minutes in shadow is a different asset from a panel that will be in full sun.

5. What this changes

  • The desktop’s solar geometry is a coupled design decision involving power, attitude, thermal, and mechanical disciplines.
  • Body-mounted panels are simple but constraining; gimballed arrays add power and flexibility at the cost of mechanism complexity.
  • Sun tracking must be aware of occlusion, rate limits, and fallback modes.
  • The next entry will wonder about autonomous battery management and load forecasting.