Simulation & Design Solutions
Structural, thermal, CFD, modal, impact and multi-physics analysis with optimisation.
Read more
Steady-state, transient, conjugate heat transfer, thermal stress and cooling design.
Most machines fail thermally before they fail structurally. A motor that runs ten degrees hotter than design loses insulation life exponentially, and nothing about that shows up in a stress plot.
The six analysis types on the poster are the ones that matter in practice: steady-state thermal, transient thermal, conjugate heat transfer, thermal stress, heat sink and cooling analysis, and design optimisation. Which of them you need depends on the question — a continuously rated machine is a steady-state problem, a duty-cycle one is not.
The exploded motor assembly on the poster illustrates the point. Heat is generated in the stator windings and the bearings, conducted through the housing, carried away by the cooling fins through convection, and radiated to the surroundings. The temperature distribution, heat flux, thermal stress and total deformation shown alongside it are the four results a thermal study is usually asked for.
The parameters listed on the artwork — maximum temperature, heat flux distribution, thermal gradient, thermal stress in von Mises terms, total deformation, heat transfer coefficient and factor of safety — are what we report as standard, because between them they answer both 'does it get too hot' and 'does the heat break it'.
Six analysis types from the poster, and the parameters each one yields.
In thermal work the boundary conditions dominate the result to an extent that surprises people coming from structural analysis. A convective heat transfer coefficient assumed rather than derived can change the peak temperature by tens of degrees, and there is no mesh refinement that will fix a wrong coefficient.
We derive convection coefficients from correlations appropriate to the geometry and flow regime, or from a CFD run where the geometry is complicated enough to make correlations unreliable. Either way the source is documented. The same applies to material properties: thermal conductivity is temperature-dependent in most materials, and using a room-temperature value for a component running at 150 degrees is a real error.
Contact resistance between parts is the other commonly missed effect. Two bolted surfaces do not conduct as though they were one solid, and modelling them as bonded will underpredict temperature. Where the joint matters we model the resistance and state the value used.
Operating conditions, duty cycle, ambient environment and the temperature limits the design has to respect.
Geometry cleaned for thermal analysis, with fins, gaps and thin walls treated appropriately rather than defeatured away.
Sized for thermal gradients, refined at interfaces and boundary layers where the temperature changes fastest.
Heat sources, convection, radiation and contact resistance, each with its source documented.
Steady state or transient as the question requires, with convergence and energy balance checked.
Temperature, flux, gradient, thermal stress and deformation, plus what to change if the design does not pass.
Heat causes expansion, expansion causes stress, and in a constrained assembly that stress can be larger than anything the mechanical load produces. A thermal analysis that stops at temperature has answered only half the question for any part that is bolted, welded or press-fitted into something else.
We run the coupled analysis where it matters: temperature field from the thermal solve mapped onto the structural model, with the mechanical loads applied simultaneously. The result is a combined stress state and a factor of safety that accounts for both. For assemblies of dissimilar materials this is essential, because differential expansion is often the governing load case.
The output is practical. If the design fails, the report says where, why, and which of the available changes — more fin area, a higher-conductivity interface material, a different flow rate, a design change to relieve constraint — gives the most margin for the least cost.
| What you receive |
|---|
| Thermal model with documented boundary conditions |
| Temperature distribution and heat flux contours |
| Thermal gradient and hot-spot identification |
| Thermal stress and deformation results |
| Heat transfer coefficient derivation and sources |
| Transient temperature history where applicable |
| Parametric or optimisation study |
| Report with recommendations and all native files |
Steady state if the machine runs continuously at a fixed load and you want the final temperature. Transient if the load varies, if you need the time to reach a limit, or if the thermal mass matters. Many studies need both: steady state to size the cooling, transient to check the worst part of the cycle.
From published correlations chosen for the geometry and flow regime, from a CFD run where the geometry is too complex for a correlation, or from measurement if you have it. Whichever it is, the source and the value are stated in the report, because it is the assumption that most affects your result.
Yes — board level, component level and enclosure level, in Icepak or Fluent. Junction temperature, airflow paths, fan sizing and heat sink selection are all standard parts of that work.
If the fluid side is simple and a correlation applies, a thermal solve with derived coefficients is enough and much faster. If flow distribution is uneven, recirculation matters, or the coefficient varies strongly over the surface, you need conjugate heat transfer. We tell you which case you are in before quoting the work.
Yes. Given the heat load, the ambient conditions and the temperature limit, we work backwards to the required thermal resistance and then to fin geometry, airflow and fan curve. You get a sizing calculation you can check by hand as well as the simulation that confirms it.
Structural, thermal, CFD, modal, impact and multi-physics analysis with optimisation.
Read more
Conceptual design, CAD, FEA, thermal and fluid analysis, optimisation and validation.
Read more
Theoretical modelling, experimental data interpretation, computation and validation.
Read moreSend your topic, your dataset or one draft chapter. We will tell you honestly what it needs — before you pay anything. The first consultation is free.