Power and Thermal Control Beyond Earth

Abstract

Electrical power and thermal control are one architecture. Every watt used by avionics, life support, communications, propulsion, or industry eventually becomes heat that must be stored or rejected. Solar generation falls with distance from the Sun and vanishes during eclipse. Batteries bridge time but do not create energy. Radioisotope and fission systems provide solar-independent output while adding shielding, conversion, and heat-rejection obligations. I develop a tiered power system with independently isolatable sources, sectional distribution, protected storage, load shedding, and black start. I then close the corresponding thermal balance through passive surfaces, pumped loops, heat storage, and segmented radiators. The design is organized around the worst credible energy interval and the worst credible heat-rejection state, not the average day. It treats unattended survival, measurement, and repair as primary operating modes.

Every watt has two lives

Electrical engineers describe generation, conversion, distribution, storage, and loads. Thermal engineers describe absorption, conduction, convection inside occupied volumes, radiation, and heat rejection. In space, these are the same energy ledger viewed at different points.

For a steady spacecraft,

$$P_{generated}=P_{useful}+P_{loss}+P_{stored},$$

while the thermal system must ultimately reject absorbed environmental heat plus internal dissipation. If electrical power rises by 100 kilowatts for propulsion, the radiator does not care that the energy performed useful work before becoming heat. Power growth without a thermal path is fictional capacity.

I classify loads before choosing sources.

Tier	Examples	Interruption policy
Life critical	Atmosphere, thermal survival, fire detection, command	Independent feeds and no planned shedding
Vehicle critical	Guidance, navigation, communication, attitude control	Brief ride-through and prioritized restoration
Mission critical	Science, mobility charging, processing	Controlled curtailment with state preservation
Deferrable	Bulk refining, discretionary experiments, comfort loads	Scheduled around generation and thermal margin

This hierarchy lets the grid fail deliberately. A poorly designed outpost may protect every load until voltage collapses and all loads are lost. A deliberate grid sheds refining, then charging, then science, while preserving survival and command.

Size energy to the interval

Peak power sizes converters and cables. Energy during darkness, fault, or low generation sizes storage. For a load profile $P_L(t)$ and available generation $P_G(t)$, required stored energy is at least

$$E_{req}=\frac{1}{\eta_s}\int_{t_0}^{t_1}\max[0,P_L(t)-P_G(t)]\,dt,$$

with storage efficiency $\eta_s$, plus reserve and degradation.

If protected loads require $25\ \mathrm{kW}$ for 18 hours and no source is available, an ideal system needs $450\ \mathrm{kWh}$. At 85 percent delivered efficiency and 80 percent allowable depth of discharge,

$$E_{nameplate}=\frac{450}{0.85(0.80)}\approx662\ \mathrm{kWh}.$$

Battery specific energy alone does not close the design. Temperature conditioning, containment, switching, fire control, degradation, radiation, charge power, and replacement logistics matter. Regenerative fuel cells or thermal storage may serve long intervals better, while batteries handle fast transients.

Solar flux varies approximately as inverse distance squared:

$$G(r)=G_1\left(\frac{1\ \mathrm{AU}}{r}\right)^2.$$

At Mars near 1.52 AU, incident flux is about 43 percent of the value at 1 AU before atmosphere, dust, temperature, pointing, and degradation. At the lunar poles, low Sun angle and terrain shadows dominate. Generation claims must use the actual trajectory or site illumination history.

Sources form a portfolio

Solar arrays are modular, mature, and productive near the Sun. Their architecture includes deployment, pointing, dust response, radiation degradation, micrometeoroid tolerance, and cable protection. Batteries stabilize the bus and bridge gaps. Fuel cells can provide energy when reactants and product management fit the mission.

Radioisotope power systems convert decay heat into modest, persistent electrical power and provide useful heat in dark, distant environments. Fission systems can reach much higher continuous power. NASA’s fission surface power work targets a 40-kilowatt class system for long lunar operation, illustrating the scale at which a reactor can become a utility rather than an instrument source.

I would separate sources physically and electrically. A reactor fault should not damage the solar field. One impact should not cut every feeder. Storage should be distributed near critical load islands. The system needs a protected black-start source that can energize controls, communication, and one generation path after total bus loss.

GenerateIndependent source fields

Solar, fission, and auxiliary sources fail and maintain separately.

StabilizeDistributed storage

Batteries and other stores ride through transients and local isolation.

ProtectSectional DC distribution

Fast isolation limits arc and cable faults to one electrical zone.

RecoverBlack-start hierarchy

A protected source restores command, sensing, and generation in sequence.

Radiation is the only external heat sink

In vacuum, a radiator rejects heat approximately according to

$$Q=\epsilon\sigma A\left(T_r^4-T_{env}^4\right),$$

where $\epsilon$ is emissivity, $\sigma$ is the Stefan-Boltzmann constant, $A$ is area, and $T_r$ is radiator temperature. Ignoring environmental backload, a surface at 350 K with emissivity 0.9 rejects roughly $765\ \mathrm{W/m^2}$. Rejecting $100\ \mathrm{kW}$ would then require about $131\ \mathrm{m^2}$ before view factors, degradation, plumbing, shadows, and reserve.

Higher radiator temperature reduces area strongly, but electronics, crew spaces, batteries, and fluids limit allowable temperature. A high-temperature reactor loop may reject heat efficiently while a habitat loop needs larger area. I keep loops separated so one leak or contaminant does not cross every thermal domain.

Passive design comes first: surface optical properties, multilayer insulation, sunshades, conductive isolation, placement, and attitude. Pumped loops transport unavoidable heat. Phase-change storage absorbs short peaks. Louvers or variable conductance manage changing conditions. Heaters protect equipment from becoming too cold during low-power survival.

Design for the coupled worst cases

The worst electrical state and worst thermal state may differ. A cold eclipse demands heater power while generation is absent. A high-power communication or propulsion event may create peak rejection demand in full sunlight. A dust-covered surface array can lose generation while the same dust changes radiator performance. A safe-mode attitude may preserve communication and worsen thermal view.

ForecastGeneration and environment ScheduleDeferrable work ReserveStorage and thermal capacity IsolateElectrical or fluid fault RestoreBlack start and load return

I require a joint set of cases: maximum generation and maximum load, eclipse with degraded battery, failed source during crew occupancy, failed radiator section, pump loss, stuck heater, safe-mode attitude, dust event, cable fault, and cold restart after long dormancy. Controls must prevent oscillation as sources and loads reconnect.

Maintain the utility, not only the equipment

Power electronics, connectors, pumps, valves, and radiator segments should be replaceable without de-energizing the whole outpost. Measurement needs enough accuracy to find losses and predict state of charge. Thermal loops need leak localization and bypass. Source and load controllers need interoperable command, timing, and protection behavior.

The first deep-space utility will not be impressive because of peak megawatts. It will be impressive when it can lose a feeder, isolate the fault, keep life support alive, explain its state to an operator, accept a replacement module, and black start after months unattended. That is the difference between a power subsystem and infrastructure.