Evaluate the "thermal contact resistance" at a joint where two solids are pressed together. Adjust the contact pressure, surface roughness, heat load and interface material to see the contact conductance and interface temperature drop (the temperature jump) update in real time, and find the bottleneck in a heat-flow path.
Parameters
Interface type
Sets the base contact conductance h_c0 (at 1 MPa, 1 µm roughness)
Contact pressure P
MPa
Pressure clamping the joint. Higher pressure crushes asperities and adds contact
Surface roughness Ra
µm
Arithmetic mean roughness of the faces. Smaller means smaller gaps
Heat load Q
W
Heat flowing through the interface
Nominal contact area A
cm²
Apparent contact area (real contact is only 1-2% of this)
Results
—
Contact conductance h_c (W/m²K)
—
Contact resistance R_c (K/W)
—
Interface temp. drop ΔT (K)
—
Contact area A (cm²)
—
Heat flux (W/cm²)
—
Verdict
—
Zoomed contact interface — asperities and heat flow
The two solids touch only at the peaks of their asperity profiles. Heat-flux lines constrict toward the few contact spots, and the temperature jumps by ΔT across the interface.
Contact conductance h_c [W/m²K]. h_c0: base value set by the interface material, P: contact pressure, R_a: surface roughness. References are P_ref=1 MPa and R_a,ref=1 µm. Higher pressure and smaller roughness raise h_c.
$$R_c=\frac{1}{h_c A},\qquad \Delta T=Q\,R_c$$
Contact resistance R_c [K/W] and interface temperature drop ΔT [K]. A: contact area, Q: heat load. The air trapped between the asperities conducts at only about 0.026 W/mK and barely passes heat — which is why interface materials (TIMs) matter.
What is Thermal Contact Resistance?
🙋
Whenever you install a CPU cooler, you always apply thermal grease. The metal surfaces are clamped tight together — so why is the grease even needed?
🎓
Good question. The thing is, "clamped tight together" is only true on the surface. No matter how well you polish a metal face, microscopically it is covered in hills and valleys — asperities. Press two of them together and only about 1-2% of the apparent area is actually touching. The rest is a gap, and normally there is air in it.
🙋
Wait, only 1-2%?! So the heat has no choice but to flow through those few tiny contact spots?
🎓
Exactly. Heat that has spread across a wide face suddenly gets squeezed into a handful of small contact spots at the interface. We call that constriction resistance. Because the flow is pinched, heat passes less easily, and the temperature jumps discontinuously across the interface. That is thermal contact resistance R_c, and the temperature gap is ΔT. In the zoomed interface view on the left, you can see the heat-flux lines bunching toward the contact spots and the temperature stepping at the interface.
🙋
I see. But if there is air in the gap, surely the air carries some heat too?
🎓
Hardly any, honestly. Air conducts at about 0.026 W/mK. Aluminium is around 200 and copper around 400, so air passes less than one thousandth of the heat a metal does. The layer of air trapped between asperities effectively works as an insulator. So you can think of most of the interface as dead, as far as heat conduction goes.
🙋
So that's where the grease comes in. What does the grease actually do?
🎓
Right. A thermal grease or thermal pad fills those air gaps with "something that conducts heat". The grease itself is not as conductive as metal — only a few W/mK — but that is still over a hundred times better than air. Just replacing the air in the gaps with grease bumps the contact conductance h_c up several-fold. Switch the "Interface type" on the left from a dry contact to grease and you will see h_c shoot up and ΔT drop right away.
🙋
Besides grease, are there other ways to reduce thermal contact resistance?
🎓
Yes, two more. One is raising the contact pressure. Press harder and the asperity peaks crush, the real contact area grows, and h_c climbs — in the formula h_c goes with P to the 0.85 power. The other is finishing the surfaces to lower the roughness R_a; shallower bumps mean smaller gaps. In practice the standard play is a three-pronged attack: secure enough pressure, finish the faces, and apply an interface material.
Frequently Asked Questions
Even when two solids are supposedly in perfect contact, every surface is microscopically a field of peaks and valleys (asperities), and only about 1-2% of the nominal area actually touches. The rest is a gap, usually filled with air. Heat must funnel through those few tiny contact spots, so the heat flow constricts at the interface and the temperature jumps discontinuously across it. Thermal contact resistance R_c measures how easily this jump occurs; this tool computes R_c and the interface temperature drop ΔT.
This tool estimates the contact conductance h_c [W/m²K] with the engineering correlation h_c = h_c0·(P/P_ref)^0.85·(R_a,ref/R_a)^0.6, where h_c0 is a base value set by the materials and interface material, P is the contact pressure and R_a is the surface roughness. The contact resistance is R_c = 1/(h_c·A) (A is the contact area) and the interface temperature drop is ΔT = Q·R_c (Q is the heat load). Raising the pressure and lowering the roughness increase h_c and reduce R_c and ΔT.
There are three main levers. (1) Raise the contact pressure to crush the asperity peaks together and increase the real contact area. (2) Finish the surfaces to lower the roughness R_a and shrink the gaps. (3) Apply a thermal interface material (TIM) such as grease or a thermal pad to fill the air gaps with a conductive substance. Switching the preset from a dry contact to grease or a pad raises h_c several-fold and sharply lowers ΔT.
Metals conduct heat at tens to hundreds of W/mK, while air conducts at only about 0.026 W/mK — less than one thousandth of a metal. The air trapped between asperities barely conducts heat, so heat is forced through the contact spots alone. In effect, most of the interface acts as an insulator. That is exactly why a grease or pad that displaces the air with a conductive material is so effective, and why it is essential for cooling CPUs and power semiconductors.
Real-World Applications
Electronics cooling (CPUs, GPUs, power semiconductors): The path that carries heat away from a hot chip through the heat spreader, heat sink and chassis always contains solid-solid contact interfaces. A large thermal contact resistance creates a big temperature jump there, and the chip stays hot no matter how powerful the heat sink. Applying the right amount of thermal grease (TIM) to fill the air gaps is done precisely to lower this contact resistance. Switching between a dry contact and grease in this tool shows that effect as an interface temperature drop.
Bolted joints in mechanical structures: Flange joints, machine mounting feet and other bolted metal-to-metal contacts are also heat paths. If the clamping force (the contact pressure) loosens, the contact resistance rises and causes unexpected temperature rise and thermal distortion. Lowering the contact pressure P in this tool drops h_c and raises ΔT, showing that fastener management is directly tied to thermal design.
Tooling, moulding and heat-treatment processes: In processes where heat is exchanged through contact — an injection mould and the moulded part, a hot plate and a workpiece — the contact resistance governs the cooling or heating rate. Scatter in roughness or pressure can become a source of quality scatter, so managing the interface condition is key to stabilising the process.
Setting boundary conditions in CAE thermal analysis: Finite-element thermal analyses apply a contact conductance (GAPCON, TCC, etc.) as a boundary condition at the contact interface. Treating it as an ideal "perfect contact" is a classic mistake that predicts temperatures lower than the real hardware. Knowing a reasonable order of magnitude for h_c from a correlation like this lets you tune the interface settings in the analysis model to realistic values.
Common Misconceptions and Pitfalls
The biggest misconception is assuming heat flows freely as long as the metals are in contact. In CAE thermal analysis, too, the contact is often joined by simply "sharing nodes" (perfect contact). In reality the contact interface is a "half-insulating wall" with only 1-2% real contact area, and ignoring it badly underestimates the temperature downstream of the joint. In high-power-density electronics, overlooking a single contact resistance can shift the temperature prediction by tens of degrees. Always assign a finite contact conductance to a contact interface.
Next, the belief that "more thermal grease is always better". The grease only fills the air gaps; its own conductivity is far below a metal's (often a few W/mK). Applying it too thickly separates even the contact spots — where metal should touch metal directly — with a grease layer, which actually raises the resistance. The optimum is a thin application that fills only the asperity valleys while leaving metal contact at the peaks. The h_c0 values in this tool assume a proper application, so the effect of over- or under-application must be considered separately.
Finally, treating thermal contact resistance as a constant. The correlation this tool uses is an engineering approximation; the real contact conductance depends on many factors — material hardness, yield strength, surface waviness, oxide films, temperature and time (growth of contact spots by creep). The same parts give a different value each time they are assembled, and the value also drifts with use. Use this tool to grasp the order of magnitude early in design and to decide the direction of countermeasures (more pressure, surface finishing, an interface material), and combine measurements or vendor data for the final guarantee.
How to Use
Enter contact pressure (0.1–100 MPa) using pNum slider; higher pressure increases surface conformity and reduces air gaps at the interface.
Set surface roughness (Ra: 0.4–6.4 micrometers) via raNum; smoother surfaces achieve lower contact resistance.
Input heat flux (qNum: 1–500 W/cm²) and contact area (areaNum: 1–1000 cm²) to calculate interface temperature drop ΔT = q·R_c, where R_c depends on pressure, roughness, and material pair properties.
Review output: contact conductance h_c (W/m²K), contact resistance R_c (K/W), and interface temperature drop to validate thermal performance.
Worked Example
Aluminum–copper joint under 10 MPa clamping pressure, Ra=1.6 micrometers, heat flux 150 W/cm², contact area 25 cm². Simulator yields h_c≈8500 W/m²K, R_c≈4.7×10⁻⁵ K/W, and interface drop ΔT≈0.71 K across the 25 cm² footprint. At 50 MPa pressure with identical roughness, contact conductance jumps to 15200 W/m²K (interface drop reduces to 0.40 K), illustrating pressure-dependent conformity gain in aerospace heat-sink assemblies.
Practical Notes
Thermal grease (silicone-based, k≈3–5 W/mK) applied at 50–100 micrometers fills surface asperities; simulator alone ignores grease—use h_c multiplier ~2–4× for coated joints in CPU cooler mounts.
Pressure saturation occurs above 30 MPa for rough surfaces; further increases yield diminishing returns, critical when bolt preload is limited in composite structures.
Verification: cross-check R_c against tabulated microcontact models (e.g., Greenwood–Williamson); discrepancies >15% suggest nonlinear material plasticity or oxidation layers.