Skip to content

Chapter 9: Thermal Control Systems

Key Concepts in This Chapter

  • Heat transfer mechanisms in the space environment: conduction, radiation, and the absence of convection

  • Orbital thermal environment characterisation including solar, planetary infrared, and albedo heat loads

  • Passive thermal control: multilayer insulation, optical coatings, and heat pipes

  • Active thermal control: heaters, louvers, mechanically pumped fluid loops, and cryocoolers

  • Thermal mathematical modelling methodology and verification by test

  • Special thermal design challenges: cryogenic instruments and battery thermal management

9.1 Introduction and Design Philosophy

The thermal control system (TCS) of a spacecraft is responsible for maintaining all components within their allowable temperature ranges throughout every phase of the mission, from integration and test on the ground to end-of-life on-orbit operations. Unlike terrestrial engineering where convection dominates heat transfer, spacecraft thermal engineers must work exclusively with conduction and radiation, two fundamentally different mechanisms that demand careful systems-level reasoning.

The consequence of thermal failure can be severe. Battery cells that freeze below \(-20\,^{\circ}\)C lose capacity irreversibly; reaction wheels overheat and seize if their lubricant degrades above \(+60\,^{\circ}\)C; optical detector arrays must often be maintained within \(\pm 0.1\,^{\circ}\)C to preserve calibration stability; and electronic components exposed to thermal cycling beyond their qualified range rapidly develop solder joint fatigue. The thermal design engineer must simultaneously satisfy conflicting requirements: electronics generate heat that must be rejected to space, yet sensitive detectors must be shielded from solar and planetary heating, and batteries must be kept warm during eclipse.

This chapter presents the fundamental physics governing spacecraft thermal behaviour, surveys the suite of passive and active thermal control hardware available to the designer, introduces the computational methods used to model and predict temperatures, and describes the test methodologies used to validate thermal models prior to launch (Gilmore 2002).

9.2 Heat Transfer Fundamentals in the Space Environment

9.2.1 Conduction

Heat conduction within solid bodies obeys Fourier’s Law. For steady-state one-dimensional conduction through a slab of cross-sectional area \(A\), thermal conductivity \(k\), and temperature gradient \(dT/dx\):

\[Q_{\mathrm{cond}} = -k \, A \, \frac{dT}{dx}\tag{9.1}\]

For a composite multi-layer structure (e.g., a panel sandwich of aluminium face-sheets and honeycomb core), thermal resistances are combined in series or parallel analogously to electrical resistors. The thermal resistance of a layer of thickness \(\delta\) is:

\[R_{\mathrm{cond}} = \frac{\delta}{k \, A}\tag{9.2}\]

Aluminium alloy Al-6061 has \(k \approx 167\,\mathrm{W\,m^{-1}\,K^{-1}}\), making it the dominant structural material for heat spreading. Titanium (\(k \approx 7\,\mathrm{W\,m^{-1}\,K^{-1}}\)) is used as a thermal isolator, while carbon-fibre reinforced polymer (CFRP) panels used for dimensional stability have \(k\) ranging from \(5\) to \(700\,\mathrm{W\,m^{-1}\,K^{-1}}\) depending on fibre orientation. Thermal interface resistance between mechanical surfaces can dominate the conductive path; this is discussed in Section 9.6.

9.2.2 Thermal Radiation

All surfaces emit thermal radiation according to the Stefan-Boltzmann law. For a surface of emittance \(\varepsilon\), temperature \(T\), and area \(A\):

\[Q_{\mathrm{rad}} = \varepsilon \, \sigma \, T^4 \, A\tag{9.3}\]

where \(\sigma = 5.6704 \times 10^{-8}\,\mathrm{W\,m^{-2}\,K^{-4}}\) is the Stefan-Boltzmann constant. The net radiation exchange between two large parallel plates of emittances \(\varepsilon_1\) and \(\varepsilon_2\) at temperatures \(T_1\) and \(T_2\) is:

\[Q_{\mathrm{net}} = \frac{\sigma (T_1^4 - T_2^4)}{\frac{1}{\varepsilon_1} + \frac{1}{\varepsilon_2} - 1}\tag{9.4}\]

For arbitrarily oriented surfaces, the geometric view factor \(F_{ij}\) represents the fraction of diffuse radiation leaving surface \(i\) that is intercepted by surface \(j\). The view factor integral is:

\[F_{ij} = \frac{1}{\pi A_i} \int\!\!\int_{A_i} \int\!\!\int_{A_j} \frac{\cos\theta_i \, \cos\theta_j}{\pi r^2} \, dA_i \, dA_j\tag{9.5}\]

where \(\theta_i\) and \(\theta_j\) are the angles between the surface normals and the line joining the area elements, and \(r\) is the distance between them. View factors obey the reciprocity relation \(A_i F_{ij} = A_j F_{j i}\) and the enclosure summation rule \(\sum_j F_{ij} = 1\).

9.2.3 Absence of Convection

In the space vacuum (\(p < 10^{-6}\,\mathrm{Pa}\)), there are essentially no gas molecules available to carry heat convectively. This has profound implications: heat generated inside a spacecraft can only leave by radiation from an external surface, with conduction providing the internal transport path. The thermal engineer must therefore create a low-resistance conductive path from every heat-generating component to a radiating surface, while simultaneously insulating the spacecraft from external heat fluxes. In pressurised volumes (crew modules, battery enclosures) convective effects are present and must be accounted for.

9.3 The Orbital Thermal Environment

9.3.1 Solar Irradiance

The solar irradiance (solar constant) at 1 AU from the Sun is:

\[S_0 = 1361 \pm 0.5 \,\mathrm{W/m^2}\tag{9.6}\]

This value varies seasonally by \(\pm 3.5\%\) due to Earth’s orbital eccentricity, reaching a maximum of approximately \(1414\,\mathrm{W/m^2}\) at perihelion (January) and a minimum of \(1322\,\mathrm{W/m^2}\) at aphelion (July). For missions at other distances from the Sun, the irradiance scales as the inverse square:

\[S(r) = S_0 \left(\frac{1\,\mathrm{AU}}{r}\right)^2\tag{9.7}\]

yielding \(S \approx 9090\,\mathrm{W/m^2}\) at Mercury (0.387 AU) and \(S = 50\,\mathrm{W/m^2}\) at Jupiter (5.2 AU).

9.3.2 Earth Infrared Emission and Albedo

The Earth emits thermal infrared radiation at a disk-averaged flux of approximately:

\[Q_{\mathrm{IR}} = 250 \pm 20 \,\mathrm{W/m^2}\tag{9.8}\]

This flux varies with latitude, season, and cloud cover. For thermal design, worst-case hot and cold values must be separately analysed. The Earth also reflects solar radiation with an average albedo factor of:

\[a = 0.30 \pm 0.05\tag{9.9}\]

The albedo heat flux absorbed by a spacecraft surface facing the Earth is:

\[Q_{\mathrm{alb}} = a \cdot S \cdot \alpha_s \cdot A_{\mathrm{proj}} \cdot F_{\mathrm{Earth}}\tag{9.10}\]

where \(\alpha_s\) is the solar absorptance of the surface, \(A_{\mathrm{proj}}\) is the projected area toward the Earth, and \(F_{\mathrm{Earth}}\) is the view factor from the surface to the Earth disk.

9.3.3 Spacecraft Energy Balance

The transient energy balance for a spacecraft thermal node of mass \(M\) and specific heat \(C_p\) is:

\[M \, C_p \, \frac{dT}{dt} = Q_{\mathrm{solar}} \cdot \alpha_s \cdot A_{\mathrm{proj}} + Q_{\mathrm{IR}} \cdot \varepsilon \cdot A_{\mathrm{Earth-facing}} + Q_{\mathrm{alb}} \cdot \alpha_s \cdot A_{\mathrm{proj}} + Q_{\mathrm{int}} - \varepsilon \, \sigma \, T^4 \, A_{\mathrm{rad}}\tag{9.11}\]

In steady-state (\(dT/dt = 0\)), the equilibrium temperature of a simple isothermal spacecraft body is:

\[T_{\mathrm{eq}} = \left[ \frac{\alpha_s \, A_{\mathrm{sun}} \, S + \varepsilon \, Q_{\mathrm{IR}} \, A_{\mathrm{earth}} + Q_{\mathrm{int}}} {\varepsilon \, \sigma \, A_{\mathrm{rad}}} \right]^{1/4}\tag{9.12}\]

The ratio \(\alpha_s/\varepsilon\) is arguably the most important single parameter in spacecraft thermal design. A high ratio causes the spacecraft to run hot; a low ratio causes it to run cold. Passive thermal control is largely the art of selecting surface finishes with appropriate \(\alpha_s/\varepsilon\) ratios.

Table 9.1. Thermo-optical properties of common spacecraft surface finishes (BOL = Beginning of Life). Note that \(\alpha_s\) degrades with radiation dose (UV darkening).
Surface Finish \(\alpha_s\) \(\varepsilon\) \(\alpha_s/\varepsilon\) Notes
Polished aluminium (bare) 0.09 0.03 3.0 Gets very hot, degrades
Anodised aluminium (black) 0.87 0.90 0.97 Good radiator, absorbs solar
Alodine (Iridite) treatment 0.15 0.08 1.88 Common finish, moderate
Optical Solar Reflector (OSR) 0.07 0.80 0.09 Excellent radiator
White paint (S13G-LO) 0.20 0.85 0.24 Stable, widely used
Black paint (Aeroglaze Z306) 0.92 0.89 1.03 Absorbs all wavelengths
Aluminised Kapton (MLI outer) 0.38 0.67 0.57 Flexible film
Aluminised Mylar 0.14 0.02 7.0 MLI inner layers only
Gold plating 0.25 0.02 12.5 Keeps surfaces warm
Second-surface silver mirror 0.07 0.80 0.09 Similar to OSR
ITO-coated Kapton 0.37 0.79 0.47 ESD-safe flexible cover
Carbon composite (CFRP) 0.88 0.85 1.04 Structure material

9.4 Passive Thermal Control

Passive thermal control encompasses all techniques that operate without power consumption: surface coatings, multilayer insulation blankets, heat pipes (which require no electrical power), and structural design for thermal paths.

9.4.1 Multilayer Insulation

Multilayer insulation (MLI) blankets are the workhorse of spacecraft thermal control. A typical blanket consists of 10 to 30 layers of \(6\)–\(25\,\mu\mathrm{m}\) Mylar or Kapton film with a thin vacuum-deposited aluminium coating, separated by Dacron mesh spacers to prevent direct contact between layers. The Dacron spacers have very low area contact fraction (typically \(\sim 2\%\)), minimising conductive bridging.

The effective emittance \(\varepsilon^*\) of an \(N\)-layer MLI blanket, accounting for both radiation and residual conduction through spacers, can be estimated. For a perfect MLI with \(N\) shields of individual emittance \(\varepsilon_s\) (both sides), the radiation component gives:

\[\frac{1}{\varepsilon^*_{\mathrm{rad}}} = \frac{1}{\varepsilon_1} + \frac{N}{\varepsilon_s / 2} + \frac{1}{\varepsilon_2} - (N+1) \approx \frac{N+2}{\varepsilon_s}\tag{9.13}\]

For adjacent layer pairs with emittances \(\varepsilon_a\) and \(\varepsilon_b\), the effective emittance of that pair is:

\[\varepsilon_{\mathrm{pair}} = \frac{\varepsilon_a \, \varepsilon_b}{\varepsilon_a + \varepsilon_b - \varepsilon_a \, \varepsilon_b}\tag{9.14}\]

The total heat flux through an \(N\)-shield MLI in the radiation limit is:

\[Q_{\mathrm{MLI}} = \frac{\sigma (T_1^4 - T_2^4)}{N+2} \cdot A \cdot \varepsilon^*\tag{9.15}\]

In practice, effective emittances \(\varepsilon^* \approx 0.010\)–\(0.025\) are achievable for well-fabricated 20-layer blankets. The actual performance is degraded by edges, seams, penetrations (for cables, tubes, and bolts), and handling damage. A conservative design value of \(\varepsilon^* = 0.03\)–\(0.05\) is often used for budgeting.

Table 9.2. MLI blanket performance data for typical configurations used in spacecraft thermal control.
Configuration Layers \(\varepsilon^*\) Heat Flux (W/m\(^2\)) Application
(typical) (at 300–77 K)
Double Aluminised Mylar (DAM) + Dacron 10 0.030 3.5 Instrument covers
DAM + Dacron 20 0.015 1.8 Spacecraft body
Aluminised Kapton + Dacron 20 0.025 2.9 External (UV stable)
Single-sided aluminised Mylar 30 0.008 1.0 Cryogenic dewar
NRC-2 foil 10 0.050 5.8 High-temperature

9.4.2 Optical Solar Reflectors

Optical solar reflectors (OSRs) consist of thin (\(0.1\)–\(0.5\,\mathrm{mm}\)) fused silica or ceria-stabilised glass tiles with a silver or aluminium metallic backing. The glass transmits solar radiation (\(\lambda < 5\,\mu\mathrm{m}\)) and reflects it from the backing, giving a very low solar absorptance. The glass itself is opaque to thermal infrared (\(\lambda > 5\,\mu\mathrm{m}\)) and radiates with high emittance. The resulting thermo-optical properties are excellent:

  • Solar absorptance: \(\alpha_s = 0.06\)–\(0.08\)

  • Infrared emittance: \(\varepsilon = 0.78\)–\(0.82\)

  • Ratio: \(\alpha_s / \varepsilon \approx 0.08\)–\(0.10\)

  • End-of-life (post-radiation): \(\alpha_s \approx 0.10\)–\(0.13\) (slight UV darkening of glass)

OSRs are radiation-stable compared to paints (which suffer UV darkening and particle radiation damage). They are commonly used on GEO satellite radiator panels where the 15+ year mission duration demands BOL-to-EOL stability. Typical radiator specific rejection capacity with OSR cover is \(250\)–\(300\,\mathrm{W/m^2}\).

9.4.3 Heat Pipes

Heat pipes are passive two-phase heat transport devices exploiting the latent heat of vaporisation of a working fluid. The principle of operation involves:

  1. Evaporator: heat input vaporises the working fluid, raising local vapour pressure.

  2. Vapour transport: pressure difference drives vapour flow through the central vapour core from evaporator to condenser.

  3. Condenser: vapour condenses, releasing latent heat to the heat sink, and the liquid is returned to the evaporator.

  4. Wick: a porous capillary structure (screen mesh, sintered powder, grooves) pumps liquid by capillary action back to the evaporator, requiring no mechanical pump.

The maximum heat transport capability (limited by capillary pumping) is:

\[Q_{\max} = \frac{r_e A_v \rho_v h_{fg}}{\mu_v L_{\mathrm{eff}}} \left[ \frac{2\sigma_l}{r_c} - \rho_l g L \sin\phi \right]\tag{9.16}\]

where \(r_e\) is the effective pore radius, \(A_v\) is the vapour space cross-sectional area, \(\rho_v\) is the vapour density, \(h_{fg}\) is the latent heat of vaporisation, \(\mu_v\) is the vapour dynamic viscosity, \(L_{\mathrm{eff}}\) is the effective length, \(\sigma_l\) is the liquid surface tension, \(r_c\) is the capillary pore radius, and \(\phi\) is the tilt angle relative to horizontal.

The most common spacecraft heat pipe working fluid is ammonia (NH\(_3\)), which has an excellent figure of merit \(N_l = \sigma_l \rho_l h_{fg} / \mu_l\) in the temperature range \(-60\) to \(+60\,^{\circ}\)C covering the typical spacecraft operating range. Propylene (\(-50\) to \(+45\,^{\circ}\)C) and ethane (\(-120\) to \(-10\,^{\circ}\)C) serve lower temperature applications.

A variable conductance heat pipe (VCHP) contains a reservoir of non-condensable gas (NCG, typically nitrogen or helium) connected to the condenser end. As the heat source temperature rises, the working fluid vapour pressure increases and compresses the NCG, exposing more condenser area and increasing the effective thermal conductance. This provides passive temperature control without any power input. The conductance varies as:

\[G_{\mathrm{eff}} = G_{\max} \cdot F_{\mathrm{condenser}}(T_{\mathrm{source}})\tag{9.17}\]

Loop heat pipes (LHPs) use the same two-phase cycle but with a mechanically separate evaporator/compensation chamber and remote condenser connected by flexible transport lines, enabling heat transport over distances of \(10\)–\(30\,\mathrm{m}\). LHPs are used on the GOES-R weather satellite series and ESA’s MetOp spacecraft.

9.5 Active Thermal Control

9.5.1 Electrical Heaters

Electrical heaters are used to maintain components above their minimum allowable temperature (MAT). Two distinct classes are employed:

Survival heaters are powered from the battery during eclipse or emergency power mode. They maintain non-operating components (e.g., propulsion lines, battery cells during launch, pyrotechnic initiators) above their survival lower limit (SLL). Typical temperature set-point: \(-20\,^{\circ}\)C.

Operational heaters are thermostatically controlled (via bimetal thermostats or software-commanded solid-state relays) during normal operations. They compensate for variations in internal dissipation and orbital environment. Heater sizing requires:

\[P_{\mathrm{heater}} = Q_{\mathrm{loss}} - Q_{\mathrm{int,\min}} = (\varepsilon \sigma T^4 A + G \Delta T) - Q_{\mathrm{int,\min}}\tag{9.18}\]

Resistive foil heaters (Kapton-encapsulated etched-foil elements) are applied directly to component surfaces or structural panels. Typical power densities are \(0.5\)–\(3\,\mathrm{W/cm^2}\).

9.5.2 Louvers

Louver assemblies (also called jalousie shutters) provide variable effective emittance by opening and closing bi-metallic actuated blades over a radiator surface. As temperature rises, the blades open, exposing the high-emittance radiator coating; as temperature falls, blades close and the low-emittance blade back-surface reduces radiative loss.

  • Fully open effective emittance: \(\varepsilon_{\mathrm{open}} \approx 0.60\)–\(0.70\)

  • Fully closed effective emittance: \(\varepsilon_{\mathrm{closed}} \approx 0.05\)–\(0.10\)

  • Temperature control range: \(\pm 5\)–\(10\,^{\circ}\)C

  • Control range (open to closed): \(15\)–\(30\,^{\circ}\)C

Louvers require no power and represent a well-proven heritage technology (used since early US and Soviet satellites). Their disadvantages are mass, complexity, potential for jamming, and susceptibility to contamination of sliding surfaces.

9.5.3 Mechanically Pumped Fluid Loops

For high-power spacecraft (\(>5\,\mathrm{kW}\) total dissipation), mechanically pumped single-phase fluid loops offer efficient heat collection from distributed sources and rejection at centralised radiators. The heat transport capacity is:

\[Q = \dot{m} \, C_p \, \Delta T = \rho \, \dot{V} \, C_p \, (T_{\mathrm{out}} - T_{\mathrm{in}})\tag{9.19}\]

The International Space Station (ISS) uses two independent mechanically pumped single-phase ammonia loops in the External Active Thermal Control System (EATCS), which together can reject up to about \(70\,\mathrm{kW}\) through large deployable radiator arrays. Commercial GEO satellites such as Boeing 702 and Airbus Eurostar E3000 use MPFL with Novec 7100 (methoxy-nonafluorobutane, \(\rho = 1528\,\mathrm{kg/m^3}\), \(C_p = 1183\,\mathrm{J\,kg^{-1}\,K^{-1}}\)) as the working fluid, which is non-toxic, non-flammable, and compatible with all spacecraft materials.

9.5.4 Cryocoolers

Cryocoolers are mechanical refrigerators that maintain sensors and detectors at temperatures from \(4\,\mathrm{K}\) to \(150\,\mathrm{K}\) by actively pumping heat from the cold stage to a warm rejection point. The thermodynamic efficiency is characterised by the coefficient of performance (COP):

\[\mathrm{COP} = \frac{Q_{\mathrm{cold}}}{W_{\mathrm{input}}} = \frac{Q_{\mathrm{cold}}}{Q_{\mathrm{hot}} - Q_{\mathrm{cold}}}\tag{9.20}\]

The Carnot COP provides the theoretical maximum:

\[\mathrm{COP}_{\mathrm{Carnot}} = \frac{T_{\mathrm{cold}}}{T_{\mathrm{hot}} - T_{\mathrm{cold}}}\tag{9.21}\]

Practical cryocoolers achieve \(10\)–\(30\%\) of Carnot COP. Major types used in space:

  • Stirling cryocoolers: linear piston compressor + displacer piston, \(20\)–\(100\,\mathrm{K}\) cooling, with heritage on numerous military and Earth-observation infrared instruments.

  • Pulse tube cryocoolers: no moving displacer (replaced by inertance tube and reservoir), lower vibration than Stirling, \(30\)–\(100\,\mathrm{K}\), used on, for example, the GOES-R Advanced Baseline Imager.

  • Joule-Thomson (JT) cryocoolers: expansion of pre-cooled gas through throttling valve, \(<20\,\mathrm{K}\) achievable, used as a final stage, no moving parts in cold head.

  • Adiabatic demagnetisation refrigerators (ADR): paramagnetic salt demagnetised to reach \(<1\,\mathrm{K}\), used on the Astro-H (Hitomi) Soft X-ray Spectrometer to reach \(50\,\mathrm{mK}\).

  • Dilution refrigerators: \(^3\)He/\(^4\)He dilution, sub-millikelvin temperatures, planned for future quantum sensor missions.

The James Webb Space Telescope (JWST) Mid-Infrared Instrument (MIRI) operates at \(6.7\,\mathrm{K}\), achieved by a dedicated cryocooler (a pulse-tube precooler with a Joule-Thomson stage) combined with the telescope’s passive cooling to \(40\,\mathrm{K}\) provided by its L2 orbit and multi-layer sunshield.

9.5.5 Thermoelectric Coolers (Peltier Devices)

Thermoelectric coolers (TECs) use the Peltier effect: when current flows through a junction of two dissimilar semiconductor materials (bismuth telluride Bi\(_2\)Te\(_3\) alloys), heat is pumped from the cold face to the hot face. The heat pumped and electrical power consumed are:

\[Q_{\mathrm{cold}} = \alpha_P \, I \, T_{\mathrm{cold}} - \frac{1}{2} I^2 R - K (T_{\mathrm{hot}} - T_{\mathrm{cold}})\tag{9.22}\]

where \(\alpha_P\) is the Seebeck coefficient of the thermocouple pair, \(I\) is the current, \(R\) is the electrical resistance, and \(K\) is the thermal conductance of the module. TECs are used for small precise temperature control applications: focal plane array (FPA) temperature stabilisation (\(\pm 0.01\,^{\circ}\)C), laser diode temperature control, and electronics boxes where \(\Delta T < 40\,^{\circ}\)C is needed.

9.6 Thermal Interface Materials

Even highly polished metal surfaces in intimate contact leave microscopic air gaps that dominate thermal resistance at the interface. Thermal interface materials (TIMs) fill these gaps and dramatically reduce contact resistance. The interface thermal resistance is:

\[R_{\mathrm{int}} = \frac{1}{h_c \, A_{\mathrm{contact}}}\tag{9.23}\]

where \(h_c\) is the contact conductance (\(\mathrm{W\,m^{-2}\,K^{-1}}\)). Table 9.3 summarises common TIM options.

Table 9.3. Thermal interface material (TIM) properties and typical applications in spacecraft.
TIM Type Conductivity Contact Conductance Notes
\(k\) (W m\(^{-1}\) K\(^{-1}\)) \(h_c\) (W m\(^{-2}\) K\(^{-1}\))
Indium foil (0.1 mm) 81 \(5000\)–\(15000\) Reusable, soft metal
Thermal grease (Apiezon N) 0.2 \(3000\)–\(8000\) Low temp, pumpout risk
Thermal grease (Dowsil TC-5026) 4.0 \(8000\)–\(20000\) High conductivity
Graphite pad (Grafoil) 5 (in-plane), 3 (thru) \(4000\)–\(12000\) Compressible, EMC too
Phase change material (PCM) 0.7 \(5000\)–\(18000\) Reflows at 52C
Bare Al-Al joint (bolted) — \(500\)–\(3000\) Highly bolt-torque dependent

The contact conductance between two bolted aluminium surfaces increases approximately as:

\[h_c \approx h_{c,0} \left( \frac{P}{P_0} \right)^{0.7}\tag{9.24}\]

where \(P\) is the contact pressure (MPa) and \(P_0\) is a reference pressure. The strong dependence on bolt torque means that the TCS thermal model must track the bolt torque specification and any relaxation due to vibration or thermal cycling.

9.7 Thermal Mathematical Modelling

9.7.1 Geometric Math Model and View Factor Calculation

The thermal mathematical model (TMM) consists of a network of thermal nodes connected by conductors, implemented in commercial software such as ESATAN-TMS (European Standard Analysis Tool for Thermal Networks), Thermal Desktop/SINDA (Cullimore & Ring), or MSC Nastran thermal. The modelling process begins with a Geometric Math Model (GMM) that captures the spacecraft geometry for view factor and absorbed flux computation.

View factor calculation for complex geometries uses the Monte Carlo ray-tracing method: \(N_{\mathrm{rays}}\) rays are launched from surface \(i\) in random directions weighted by the cosine law; the fraction intercepting surface \(j\) gives \(F_{ij}\). The statistical uncertainty is:

\[\sigma_{F_{ij}} = \sqrt{\frac{F_{ij}(1-F_{ij})}{N_{\mathrm{rays}}}}\tag{9.25}\]

Typically \(N_{\mathrm{rays}} = 10^6\) to \(10^8\) are used to achieve \(\sigma < 0.001\).

9.7.2 Node Network and Conductors

The TMM discretises the spacecraft into a set of thermal nodes. For a network of \(n\) nodes, the thermal state is described by the vector \(\mathbf{T} \in \mathbb{R}^n\). The governing equations are:

\[C_i \frac{dT_i}{dt} = Q_{\mathrm{ext},i} + Q_{\mathrm{int},i} + \sum_{j \neq i} G_{ij}^{\mathrm{cond}} (T_j - T_i) + \sum_{j \neq i} G_{ij}^{\mathrm{rad}} (T_j^4 - T_i^4)\tag{9.26}\]

where \(C_i = M_i C_{p,i}\) is the thermal capacitance of node \(i\), \(G_{ij}^{\mathrm{cond}}\) is the linear conductive conductor, \(G_{ij}^{\mathrm{rad}} = \varepsilon^* \sigma A_i F_{ij}\) is the radiative conductor (radiation exchange factor, REF), and \(Q_{\mathrm{ext},i}\) includes solar, IR, and albedo inputs computed by the GMM for each orbital position.

The system of equations can be written in matrix form:

\[\mathbf{C} \dot{\mathbf{T}} = \mathbf{Q}_{\mathrm{ext}} + \mathbf{Q}_{\mathrm{int}} + \mathbf{G}^{\mathrm{cond}} (\mathbf{T}_j - \mathbf{T}_i) + \mathbf{G}^{\mathrm{rad}} (\mathbf{T}_j^4 - \mathbf{T}_i^4)\tag{9.27}\]

9.7.3 Thermal Analysis Cases

Four primary analysis cases bound the temperature range:

  1. Hot case: maximum solar (perihelion), maximum internal dissipation (payload fully on), orbit with maximum sunlit time, BOL surface properties (clean surfaces, low \(\alpha_s\)).

  2. Cold case: minimum solar (aphelion), minimum internal dissipation (safe mode, minimum equipment on), maximum eclipse fraction, EOL degraded surface properties (UV-darkened, high \(\alpha_s\)).

  3. Eclipse start/end transient: rapid temperature change as spacecraft enters/exits Earth’s shadow, driving thermal cycling fatigue analysis.

  4. Transfer orbit: applicable to GEO satellites during the apogee engine firing phase when attitude is constrained by the engine pointing requirement.

9.8 Thermal Testing and Verification

9.8.1 Thermal Balance Test

The thermal balance test (TBT) is conducted in a large thermal vacuum (TV) chamber at pressures below \(10^{-4}\,\mathrm{Pa}\). The spacecraft is placed on a support structure surrounded by a cold shroud (liquid nitrogen cooled, \(T_{\mathrm{shroud}} \approx 100\,\mathrm{K}\)) to simulate the cold space sink. Solar simulation is provided by infrared heater arrays or, for higher fidelity, quartz lamp arrays. The test sequence includes:

  1. Pump down to vacuum (\(< 10^{-4}\,\mathrm{Pa}\))

  2. Cold soak stabilisation (typically 12–24 h)

  3. Hot case stabilisation at maximum internal dissipation

  4. Eclipse simulation (heater arrays off)

  5. Additional cases per test plan

Correlation between test-measured temperatures and TMM-predicted temperatures is performed. Acceptance criteria typically require:

\[|T_{\mathrm{predicted}} - T_{\mathrm{measured}}| < 3\,^{\circ}\mathrm{C} \quad \text{for critical nodes}\tag{9.28}\]

Parameters adjusted during correlation include conductances, heater powers, and surface properties, following a systematic sensitivity analysis.

9.8.2 Thermal Cycling

Thermal cycling tests subject the spacecraft or unit to repeated temperature excursions between \(T_{\min} = -40\,^{\circ}\)C and \(T_{\max} = +70\,^{\circ}\)C (qualification) or \(-25\) to \(+55\,^{\circ}\)C (acceptance), with dwell times sufficient to achieve temperature stabilisation within \(\pm 2\,^{\circ}\)C. Typical cycle counts: 8 (acceptance), 100 (qualification). Cycling is performed in ambient air for unit-level tests (air cooling compensates for lack of radiation) and in vacuum for spacecraft-level tests.

9.9 Special Thermal Design Challenges

9.9.1 Cryogenic Instrument Thermal Design

Cryogenic infrared and X-ray instruments present the most demanding thermal design challenges. The key principle is to minimise parasitic heat loads to the coldest stage through:

  • Passive cooling: L2, lunar trailing, or Sun-synchronous (with appropriate baffle) orbits allow large passive radiators to reach \(40\)–\(80\,\mathrm{K}\).

  • V-groove radiators: JWST uses three deployable V-groove radiators to achieve passive cooling of the telescope structure to \(\sim40\,\mathrm{K}\).

  • Thermal straps: braided aluminium or copper straps with low stiffness connect the cold stage to the cryocooler cold finger or passive radiator.

  • Multilayer insulation: cryogenic MLI with 30+ layers provides effective emittance \(\varepsilon^* < 0.005\).

9.9.2 Battery Thermal Management

Lithium-ion battery cells exhibit strong temperature sensitivity:

  • Optimal charge/discharge temperature: \(0\) to \(+40\,^{\circ}\)C

  • Charging below \(0\,^{\circ}\)C: lithium plating causes irreversible capacity loss

  • Storage at elevated temperature accelerates calendar aging (capacity fade)

  • Thermal runaway threshold: \(\sim 130\,^{\circ}\)C

  • Capacity recovery at low temperature: cells deliver \(\sim 50\%\) capacity at \(-20\,^{\circ}\)C vs. \(+20\,^{\circ}\)C

Battery thermal design must therefore maintain cells within \(0\) to \(+30\,^{\circ}\)C throughout the mission. This requires survival heaters (typically \(10\)–\(30\,\mathrm{W}\) per battery) for eclipse cold-soak and heat spreading/rejection in hot orbit.

9.9.3 Thermal Control for Different Orbit Regimes

Table 9.4. Thermal environment comparison for different orbit/mission types. Peak heat flux values are surface-area-specific assuming worst-case orientation.
Mission Type Eclipse Solar Flux Earth IR Dominant Challenge
Fraction (W/m\(^2\)) (W/m\(^2\))
LEO (500 km) 35% \(1361\) \(250\) Thermal cycling, eclipse heaters
LEO (ISS, 400 km) 38% \(1361\) \(250\) High internal power
SSO (700 km, dawn-dusk) 0% (near) \(1361\) \(250\) Continuous solar, no eclipse
MEO (GPS, 20200 km) 10% \(1361\) \(30\) Low Earth IR, radiation belts
GEO (35786 km) 1% \(1361\) \(7\) Long eclipse equinox periods
L2 (1.5M km) 0% \(1362\) \(\sim 0\) Passive cooling, no eclipse
Mars orbit 20% \(590\) \(110\) Low solar, CO\(_2\) atmosphere
Jupiter orbit variable \(50\) \(5\) Very low solar, nuclear power

9.10 Design Example: LEO Earth Observation Satellite Thermal Control

Consider a \(200\,\mathrm{kg}\) LEO Earth observation satellite in a \(600\,\mathrm{km}\) Sun-synchronous orbit. The satellite bus dissipates \(500\,\mathrm{W}\) in operational mode and \(100\,\mathrm{W}\) in safe mode. The radiator area required to reject the operational dissipation with a maximum panel temperature of \(+30\,^{\circ}\)C and an OSR cover (\(\varepsilon = 0.80\), \(\alpha_s = 0.07\)) is:

\[A_{\mathrm{rad}} = \frac{Q_{\mathrm{int}}}{\varepsilon \sigma T_{\mathrm{max}}^4 - \alpha_s S / 4} = \frac{500}{0.80 \times 5.67\times10^{-8} \times 303^4 - 0.07 \times 1361 / 4} \approx 1.35\,\mathrm{m^2}\tag{9.29}\]

This calculation ignores Earth IR and albedo inputs for brevity. The survival heater power to maintain the spacecraft above \(-15\,^{\circ}\)C during a \(35\%\) eclipse fraction with all equipment off (safe mode):

\[P_{\mathrm{survival}} = \varepsilon \sigma T_{\mathrm{survival}}^4 A_{\mathrm{rad}} - Q_{\mathrm{int,safe}} \approx 0.80 \times 5.67\times10^{-8} \times 258^4 \times 1.35 - 100 \approx 55\,\mathrm{W}\tag{9.30}\]

9.11 Summary

The thermal control system is a critical enabling subsystem that must be designed concurrently with the spacecraft structure, power, and attitude control from the earliest mission concept phase. The choice between passive-dominant and active-dominant thermal architectures is driven by the total internal dissipation, the mission orbit, and the temperature requirements of the payload.

Key design parameters are the thermo-optical properties \(\alpha_s\) and \(\varepsilon\) of all external surfaces, the effective emittance \(\varepsilon^*\) of MLI blankets, the heat pipe conductance sizing, and the heater power budget. The thermal mathematical model must be validated against thermal balance test data before flight predictions can be trusted.

Future trends include the increasing use of mechanically pumped loops and deployable radiators for high-power communications satellites (\(>20\,\mathrm{kW}\)), advanced variable conductance technologies for wide-range autonomy, and the development of sub-Kelvin refrigerators for next-generation quantum sensor and far-infrared astronomy missions (Gilmore 2002).

Further Reading

  • Gilmore, D. G. (Ed.): Spacecraft Thermal Control Handbook, Vol. 1: Fundamental Technologies, 2nd edn., Aerospace Press, El Segundo (2002).

  • Karam, R. D.: Satellite Thermal Control for Systems Engineers, AIAA Education Series (1998).

  • Meseguer, J., Pérez-Grande, I., Sanz-Andrés, A.: Spacecraft Thermal Control, Woodhead Publishing (2012).

  • Incropera, F. P., DeWitt, D. P., Bergman, T. L., Lavine, A. S.: Fundamentals of Heat and Mass Transfer, 7th edn., Wiley (2011).

  • ESA-ESTEC: ECSS-E-HB-31-01 Parts 1–16: Thermal Design Handbook, ECSS Secretariat, Noordwijk.