When people talk about heating or cooling a room, they almost always think in terms of air: a thermostat measuring a temperature, a system raising or lowering it. But there is a second, often more influential channel through which the human body exchanges heat with its surroundings: thermal radiation. It’s the same physical principle that lets sunlight warm your skin on a cold day, and the reason a cold wall or ceiling can make a room feel chilly even when the air thermometer reads a perfectly acceptable temperature. Understanding how radiation works isn’t an academic exercise — it’s the key to designing and evaluating heating and cooling systems that are more efficient and more comfortable than air temperature control alone can deliver.
The three heat transfer mechanisms: conduction, convection, radiation
Heat moves from a warmer body to a cooler one through three distinct mechanisms, which in real buildings almost always operate together, mixed in varying proportions.
| Mechanism | How it works | Typical example |
|---|---|---|
| Conduction | Energy transfer through direct molecular contact, with no movement of material | Heat passing through a wall, or a floor felt underfoot |
| Convection | Heat carried by the movement of a fluid (air or water) that warms or cools on contact with a surface | Warm air rising above a traditional heat emitter |
| Radiation | Electromagnetic energy emitted by a surface, which travels even through a vacuum and is absorbed by another surface with no medium required | The warmth of the sun felt on skin, or the heat of a hot surface felt at a distance |
The most important practical difference is that conduction and convection both need a physical medium (a solid, a fluid), while radiation also travels through air without directly heating it: it acts on the surfaces and bodies it reaches, not on the volume of air it passes through. That single characteristic is what makes radiation a physical principle in its own right, with specific implications for both heating and cooling design.
What thermal radiation actually is: the physics in brief
Every body above absolute zero (-273.15 °C) emits energy as electromagnetic radiation, typically in the infrared band at the temperatures found in occupied buildings. The amount of energy radiated by a surface is described by the Stefan-Boltzmann law:
P = ε · σ · A · T⁴
where P is the radiated power (in watts), ε is the surface emissivity (a number between 0 and 1 describing how efficiently a surface radiates compared to an ideal black body), σ is the Stefan-Boltzmann constant (5.67 × 10⁻⁸ W/m²K⁴), A is the surface area, and T is the absolute temperature in kelvin.
Two features of this formula have very concrete consequences:
- The fourth-power dependence on temperature means small changes in surface temperature produce much larger changes in the net radiative exchange between two surfaces. It isn’t the linear relationship we’re used to when thinking about heat loss through a wall.
- Emissivity varies enormously between materials. Human skin, plaster, wood, matte paint and even glass all have high emissivity, close to 0.9-0.98, regardless of their visible colour. Polished or bare metals, by contrast, have very low emissivity — sometimes below 0.1 — which is why a shiny metal surface radiates, and absorbs, far less radiant heat at the same temperature.
When two surfaces at different temperatures “see” each other, the net radiant energy exchanged between them depends on the difference between their absolute temperatures raised to the fourth power, on their emissivities, and on how much of each surface actually faces the other (the so-called view factor). It’s the same principle, always, that governs both heat emission from a warm surface and heat absorption by a cold one — only the direction of the net energy flow changes.
Why a room can feel cold even when the air temperature is right
The human body exchanges heat with its surroundings through all three mechanisms at once, but radiation typically accounts for a substantial share — often comparable to, or larger than, convection under typical indoor conditions. That explains a very common experience: a room with air at 20 °C but a large cold window, or poorly insulated perimeter walls, can still feel cold, because the body loses more heat by radiation toward those cold surfaces than it gains back. The air thermometer, on its own, doesn’t tell the whole story.
That’s why thermal comfort design increasingly relies on mean radiant temperature (the “equivalent” uniform surface temperature that would exchange the same amount of radiant heat with the body as all the real surrounding surfaces combined) and on operative temperature, which combines air temperature and mean radiant temperature into a single value that is much closer to what a person actually perceives. In many cases, adjusting the radiant temperature of surfaces is a more effective — and more energy-efficient — lever than simply raising the air temperature further.
The same principle, in reverse: radiant cooling
If a warm surface radiates energy toward the cooler objects and people around it, a surface kept below room temperature does exactly the opposite: it absorbs the radiant energy emitted by bodies, furniture and occupants, removing heat without moving large volumes of air. This is the physical principle behind radiant cooling systems — chilled ceiling, wall or floor surfaces fed with cooled water. They don’t cool the air through forced convection the way a fan-driven terminal unit does; instead, they lower the mean radiant temperature perceived by occupants, often allowing acceptable comfort to be maintained even with air temperatures slightly higher than a purely air-based system would require.
Radiant cooling, however, comes with a physical constraint that heating doesn’t share: the surface temperature can never drop below the dew point of the room air, or condensation will form on the surface itself. This keeps the cooled-water temperature relatively “high” by cooling standards, sized with a safety margin calculated directly from the room’s relative humidity, and often requires pairing the system with dedicated ventilation or dehumidification whenever indoor humidity is significant.
Thermal comfort: how much can a surface be “off” before it becomes uncomfortable
Not all radiant asymmetries are perceived equally. Thermal comfort research — the basis for standards such as ISO 7730 — has identified indicative thresholds beyond which the difference between one surface’s temperature and the surrounding surfaces becomes a source of discomfort, even when the air temperature and overall mean radiant temperature stay constant:
| Configuration | Indicative temperature difference tolerated before discomfort |
|---|---|
| Warm ceiling (overhead) | About 5 °C |
| Cool wall (to the side) | About 10 °C |
| Cool ceiling | About 14 °C |
| Warm wall (to the side) | About 23 °C |
These indicative values, drawn from comfort research that technical standards later adopt with their own margins, explain design choices that would otherwise seem counter-intuitive: a radiant heating ceiling can feel uncomfortable at surface temperatures a warm wall would tolerate without issue, while a radiant cooling ceiling has more headroom than a cold side wall before causing discomfort. This is also why well-designed radiant systems keep the surface temperature of active surfaces relatively close to room temperature, spreading heat exchange across a large area rather than concentrating it on a few square metres with a large temperature difference.
A related, more practical constraint governs heated floors: EN 1264 limits floor surface temperature to roughly 9 °C above the design air temperature in occupied zones (so, typically, around 29 °C with 20 °C room air), rising to about 15 °C above air temperature in perimeter strips under large glazed areas, where the heat load to compensate is higher.
What this means in practice for system design
Understanding radiation isn’t just basic physics — it has direct, measurable consequences for equipment and system choices.
- Large surfaces, small temperature differences. Heat exchange dominated by radiation works best when the exchanging surface is large (floor, wall, ceiling) rather than concentrated, because delivering the same power requires a much smaller temperature difference relative to the room.
- Lower flow temperatures for heating. A system that exchanges heat efficiently through radiation can run on lower supply water temperatures (often 30-40 °C, versus 60-70 °C for a system sized mainly around convective exchange). For a heat pump, this translates into a significantly higher COP, since the machine has to produce a smaller temperature lift.
- Higher chilled water temperatures for cooling. For the same reason, a radiant cooling system can operate with less extreme chilled water temperatures (typically 15-18 °C) compared with a direct-expansion or fan-coil system, with similar efficiency benefits for the chiller, plus the absence of forced air recirculation and its associated noise.
- Thermal inertia to manage. Floor-based radiant systems, because of the thermal mass involved, respond more slowly to load changes than an air-based system or a lightweight metal radiant ceiling panel, which reacts much faster. The choice between technologies therefore also depends on how the space is used and how quickly it needs to respond to changing conditions.
- Comfort per unit of energy. Because the human body responds to operative temperature (air plus radiation), not air temperature alone, adjusting the radiant component often makes it possible to maintain the same comfort level with slightly cooler air in winter or slightly warmer air in summer — saving energy for the same perceived comfort.
Turning these principles into a real system almost always calls for a project built to measure, especially when a building’s constraints, room geometry or efficiency targets fall outside standard catalogue solutions. Here’s a closer look at how a truly custom heating and cooling system comes together, from initial requirements through to technical design.
Frequently asked questions
Do traditional radiators actually work mainly by radiation? Less than the name suggests. A typical panel or column radiator releases most of its output through convection — air warming on contact with the fins or panels and rising by natural draft — with a radiant share that usually falls somewhere between 20% and 30% of total output, depending on the geometry and surface area of the emitter. Systems designed to maximise the radiant component — low-temperature flat surfaces in floors, walls or ceilings — flip that ratio, favouring radiative exchange over convective exchange.
What’s the difference between air temperature and perceived temperature? The air temperature read by a thermometer is only one of the variables that determine how warm or cold a space feels. Operative temperature — which combines air temperature with the mean radiant temperature of surrounding surfaces — is a much closer proxy for actual perception, and is the reference parameter used in comfort design under current technical standards.
Can radiant cooling fully replace an air-based system? It depends on the latent load involved. Radiant cooling is very effective at handling the sensible load — the component that raises temperature — but it doesn’t dehumidify the air. In climates or spaces with significant moisture loads (high occupancy, kitchens, humid environments) it normally needs to be paired with dedicated ventilation and dehumidification, if only to provide fresh air and keep humidity below the dew point of the radiant surfaces.
Why doesn’t a radiant cooling surface “sweat” the way a cold glass of water does? Because it’s designed and controlled specifically to avoid it: the chilled water temperature, and therefore the surface temperature, is kept with a safety margin above the dew point calculated from the room’s actual relative humidity, often using a humidity sensor that automatically adjusts the system so the surface never drops below that threshold.
Does a surface’s colour affect thermal radiation? For long-wavelength infrared radiation — the kind relevant at room temperatures, as opposed to visible sunlight — visible colour has minimal influence: nearly all non-metallic building materials, whether light or dark plaster, wood, or fabric, share a high, similar thermal emissivity regardless of how they look to the eye. The picture changes for short-wave solar radiation, where colour and surface finish have a much bigger effect on absorption.
Understanding the physics of radiation makes it easier to read heating and cooling equipment specifications with a more critical eye — looking not only at the declared output, but at how that output is actually exchanged with the space and the people occupying it. When a project’s requirements move beyond standard catalogue options — unconventional geometries, space constraints, a specific match with a heat pump — the radiant component is often the first parameter worth engineering around: explore our approach to custom heating and cooling systems.