🧑🎓
What exactly is a "thermal bridge"? I've heard it makes buildings less efficient, but I don't get the physics.
🎓
Basically, it's a localized spot in a building envelope where heat escapes much faster than the surrounding area. In practice, it happens where materials with high thermal conductivity, like steel or concrete, create a shortcut for heat to flow from the warm inside to the cold outside. Try selecting the "Steel Bridge" preset in the simulator above—you'll instantly see a bright red "hot spot" of heat flux piercing through the wall.
🧑🎓
Wait, really? So the 2D temperature map isn't uniform? Why can't we just use a simple 1D calculation?
🎓
Great question! A 1D calculation assumes heat only flows straight through layers. But at corners, window frames, or structural connections, heat flows in two or three dimensions, creating complex patterns. For instance, in the simulator, slide the "Concrete k₁" and "Insulation k₂" conductivities far apart. You'll see the isotherms (lines of equal temperature) bend sharply at the material interface—that's the 2D effect a 1D model completely misses, leading to underestimated heat loss.
🧑🎓
So how do engineers quantify this "extra" heat loss? Is there a standard number to look for?
🎓
Yes! They use a metric called the **linear thermal transmittance**, or psi (ψ). It measures the *additional* heat loss per meter of bridge length. The simulator calculates it using the formula ψ = Q₂D/ΔT − U₁D×L. A common target, like the Passive House standard, requires ψ < 0.01 W/(m·K). Try changing the "Steel Bridge k₃" to a very high value and watch the calculated ψ value skyrocket, showing how bad the bridge is.
The core physics is governed by the steady-state heat conduction equation. For 2D, with no internal heat generation, this simplifies to Laplace's equation for temperature.
$$\nabla^2 T = \frac{\partial^2 T}{\partial x^2}+ \frac{\partial^2 T}{\partial y^2}= 0$$
This equation states that at any point (except boundaries), the net flow of heat into a tiny volume is zero. The simulator solves this numerically using the Finite Difference Method across the grid of materials you define.
Once the temperature field T(x,y) is known, we calculate the heat flux (heat flow per unit area) using Fourier's Law, and from that, derive key performance metrics.
$$
\begin{aligned}\text{Heat Flux: }&\mathbf{q}= -k\,\nabla T \\[5pt]
\text{Linear Transmittance: }&\psi = \frac{Q_{2D}}{\Delta T}- U_{1D}\cdot L \\[5pt]
\text{Temp. Factor: }&f_{Rsi}= \frac{T_{si,min} - T_e}{T_i - T_e}\end{aligned}
$$
q is the heat flux vector (W/m²), k is thermal conductivity (W/m·K). ψ quantifies the bridge's severity. f_Rsi predicts surface condensation risk; a value below 0.7 is often a problem.
Common Misunderstandings and Points to Note
When you start using this simulator, there are a few points you should be careful about. First, "the thermal conductivity value is not determined solely by the material name." For example, even something broadly called "concrete" can have a thermal conductivity that varies greatly depending on its density and moisture content. The tool's default values are merely representative. For real projects, the golden rule is to check and input the catalog values or standard values (like JIS A) for the specific material you're using. Using rough values can sometimes lead to psi-values differing from reality by as much as 30%.
Next, understand the limitations of 2D analysis. Since this tool looks at a cross-section, it cannot fully capture the complete picture of "three-dimensional thermal bridges" like corners or column heads/bases. For instance, a concrete corner becomes a "3D thermal bridge" where heat escapes concentrated from three directions: the two inner surfaces and the ceiling surface. Even if the 2D analysis shows a high surface temperature, there might still be a condensation risk at the actual corner. For particularly critical areas, it's best to verify with a 3D simulation.
Finally, boundary condition setting errors. The internal and external surface heat transfer resistances (Rsi, Rse) are often set to ISO standard values by default, but this assumes "normal natural convection." For example, if you place large furniture in front of an external wall obstructing airflow, or if there's forced ventilation indoors, the actual heat transfer changes. Don't over-rely on simulation results; get into the habit of always clearly stating the premise: "given these conditions, the result is this."
Related Engineering Fields
The technology behind this 2D heat conduction simulation actually forms the foundation for various engineering fields beyond just building insulation evaluation. Starting with a closely related field: "Thermal design (thermal management) for electronic devices." How do you efficiently dissipate heat from a CPU chip generating heat on a smartphone's circuit board? When designing the heat pathways (heat sinks, thermal plates), the exact same "steady-state heat conduction analysis" used by this tool is applied. The process of setting material thermal conductivities (silicon, copper, resin) and observing the temperature distribution is completely identical.
Another is "the design of buried pipes for ground-source heat pumps." Pipes are buried in the ground for heat exchange, and the temperature distribution created in the surrounding ground (soil) can be solved as a heat conduction problem with the soil's thermal conductivity and the pipe surface temperature as boundary conditions. Here, "Material k₃" becomes the soil.
Broadening the perspective further, the Laplace equation $∇^2 T = 0$ governing heat conduction is also used to describe entirely different physical phenomena like "electrostatic fields (electric potential distribution)", "potential flow of ideal fluids", and even "deflection of elastic bodies." In other words, the concept you're learning here—"calculating a field by giving boundary conditions"—builds foundational skills applicable across a very wide range of CAE.
For Further Learning
If you want to dive deeper, try taking the following steps. First, the mathematical background. The "Finite Difference Method" at the core of this tool is the simplest numerical method for approximately solving differential equations. The next step is to learn the "Finite Element Method (FEM)." FEM solves problems by dividing complex shapes into small triangles or quadrilaterals (elements), and it's used by most commercial, full-fledged CAE software. Understanding the "grid point" concept in the Finite Difference Method should make the FEM concepts of "elements and nodes" much easier to grasp.
Next, challenge yourself with "unsteady (transient) heat conduction." This tool only looks at the final stable state (steady-state), but actual buildings experience temperature changes from morning to night and across seasons. How the temperature within a wall changes over time is described by the following equation: $$ρ c_p \frac{\partial T}{\partial t} = ∇ \cdot (k ∇ T)$$ where $ρ$ is density, $c_p$ is specific heat, and $t$ is time. Adding this term means the effectiveness of insulation relates not only to "resistance to heat flow" but also to "the ability to store heat (thermal capacity)." Considering this time-dependent change is crucial in energy-saving calculations.
For learning directly connected to practical work, try using this simulator to create models according to standards while referencing "the Building Energy Efficiency Act" or "ISO 10211 (the standard for thermal bridge calculation)." For example, reproduce standard thermal bridge detail drawings defined by the standards and compare the calculation results. This will help you see not just "what the tool is calculating," but the path to "how it's used in practice."