Across a true thermal resistance, heat flux can remain continuous while temperature jumps by q″R″. The key question is whether the specified resistance represents the physical interface you intended.
Problem: temperature is discontinuous at a solid-solid or fluid-solid interface
Symptoms include a sudden temperature jump across nominally touching solids, one component running much hotter than expected, heat flux that appears blocked, or energy balance that changes dramatically when contact resistance is enabled.
First decide whether the physical interface should be ideal, bonded, coated, gapped or imperfect. The correct temperature behavior depends on that answer.
Relate temperature jump to thermal resistance
For an area-specific thermal contact resistance R″, the temperature jump is approximately ΔT = q″R″ under steady one-dimensional conditions. Thermal conductance is the inverse quantity. Make sure the software expects total resistance, area-specific resistance or conductance before entering a value.
Ideal bonded contact
Temperature should be continuous apart from numerical discretization; heat flux crosses directly.
Finite contact resistance
A real temperature jump is expected while heat flux remains consistent across the interface.
Thin physical layer
Model explicitly or convert thickness and conductivity to an equivalent resistance if assumptions permit.
Gap / contact loss
Conduction may be weak and radiation or gap-fluid heat transfer may become important.
Diagnose the interface systematically
- Check units and resistance type.
m²·K/W, K/W and W/(m²·K) are not interchangeable.
- Check material conductivity and thickness.
An incorrect solid conductivity can look like a contact problem.
- Plot heat flux on both sides.
For a steady interface without storage, the normal heat flow should reconcile.
- Check interface connectivity.
Confirm the two regions or surfaces are actually coupled and not accidentally treated as adiabatic walls.
- Check duplicated resistance.
Do not model a thermal interface material explicitly and add its full resistance again as a contact.
Use energy balance to separate local and global errors
A local temperature jump can be correct even when the global energy balance is excellent. Conversely, a visually smooth temperature field can still be wrong if the heat source, interface flux or outlet enthalpy does not balance.
Build a heat-flow ledger from source to solids, contacts, fluid boundaries and outlets. This reveals whether the resistance is merely redistributing temperature or actually blocking or creating energy numerically.
Cross-check the full thermal model
Use the existing CHT troubleshooting case to trace source-to-outlet energy flow.
Check mesh resolution without over-refining the contact
If the resistance is represented as a boundary condition, the temperature jump occurs at the interface and does not require meshing an artificial thickness. If a real thin layer is modelled explicitly, enough cells are needed through the thickness to resolve its gradient without extreme aspect-ratio or non-orthogonality problems.
Common mistakes
- Entering conductance where the solver expects resistance, or vice versa.
- Ignoring area when converting between K/W and m²·K/W.
- Adding contact resistance to an interface that already includes an explicit thermal interface material.
- Assuming any temperature discontinuity is numerical error.
- Checking temperature only and not the heat-flux or energy balance.
Related Abecator resources
Need help reviewing a thermal interface model?
Submit the materials, contact definition, heat source, expected heat path and non-confidential temperature and heat-flux plots.