Thermal Simulation: Copper, Airflow and Junction Temperature

A thermal simulation predicts the temperature of a component from its dissipation and its surroundings. Where the prediction is wrong, the usual cause is one of three inputs: the copper, the airflow or the interface.

What the Simulation Needs

The dissipation of each component, the copper area connected to it, the board’s thermal conductivity and the boundary conditions at the edges of the model.

The copper is the dominant path in most designs, and it is also the input that is most often simplified. A copper area modelled as a uniform block ignores the fact that it is connected to the rest of the board through narrow necks and through vias.

The boundary condition is what the board is connected to. A board fixed to a metal chassis loses heat through the mountings, while one held on plastic standoffs loses it only through the air. The two cases differ by a large factor. Our thermal design notes describe how the copper is planned.

Conduction, Convection and Radiation

Conduction through copper and through the laminate is the dominant mechanism inside the board. Its magnitude depends on the cross section and the thermal conductivity of the material.

Convection from the surface depends on the airflow and on the surface area. Natural convection in a sealed enclosure is weak, and a small amount of airflow changes the result substantially.

Radiation matters at higher temperatures and is usually small compared with the other two at moderate temperatures. A thermal camera measures radiation and converts it to a temperature using an assumed emissivity, which is why the setting matters. Our current capacity notes describe the related heating in conductors.

Modelling the Copper Properly

The copper should be modelled with its actual shape, including the necks and the via arrays. A thermal via array is a strong local path and should not be lumped into the surrounding copper.

The laminate’s thermal conductivity is much lower than copper, so the heat spreads laterally in the copper and moves vertically through the laminate mainly at vias. This anisotropy is why a simple isotropic model overestimates spreading.

Where the design has several heat sources, the interaction matters. Two components that each run warm in isolation may run hot when placed close together. Our switching regulator notes describe the layout considerations for a dissipating part.

Airflow and Enclosure

The airflow inside an enclosure is not uniform, and a component in a stagnant corner runs hotter than the average. The model should use the local velocity rather than the fan’s specification.

Where the airflow is complex, a computational fluid dynamics model is needed. Where the enclosure is simple, a measurement of the air velocity at the component’s location is often enough to calibrate a simpler model.

The enclosure also sets the ambient, which is higher than the outside temperature by the amount the other components raise it.

Interface and Contact Resistance

Where a component is mounted to a heatsink or to the chassis, the interface resistance is a significant part of the path and it is often the largest unknown.

The interface depends on the flatness, the pressure and the material, and its value varies by a factor of several between a good and a poor assembly.

The simulation should use a range rather than a single value, and the measurement should be used to confirm which end of the range applies. Our quality notes describe the assembly measurements that are recorded.

Validating the Model

The validation is a measurement of the actual temperatures on the assembled board, at the design load and the worst case ambient, compared against the prediction.

The comparison should be made at several points and not only at the hottest, because a model that is right at one point can be wrong elsewhere.

A model that predicts within a few degrees at several points can be trusted for a design change. One that is fitted to a single point cannot.

When the Prediction Is too Optimistic

The common causes are an airflow assumption that the enclosure does not deliver, a copper area that is interrupted by more necks than the model assumed, and an interface resistance that is at the poor end of its range.

The remedy is to measure rather than to refine the model, since the measurement reveals which assumption failed.

Where the margin is small, the design should be changed rather than the model. Adding copper, moving the component or improving the interface are all more reliable than a better simulation.

Additional Considerations for This Build

Practical attention to junction temperature pays for itself here, because it decides whether the finished board behaves as the drawing intended. Where the requirement is not stated on the fabrication drawing or in the assembly notes, the shop has to assume a default, and that default is rarely the value the design was simulated with. Stating junction temperature explicitly, together with the tolerance that applies, removes the assumption and keeps the result predictable from batch to batch.

Process Control and Verification

On a design of this kind, junction temperature is the item that decides how the rest of the board is arranged. Documenting the assumption is part of the design work, and a short note on the drawing prevents a question that would otherwise arrive a day later and cost a day of schedule. The process window is set by the narrowest step in the flow, so an improvement anywhere else shows up as margin rather than as yield until that step is addressed.

A short note on the drawing about handling, storage or packaging is often worth more than an extra decimal place on a tolerance. Where a value sits close to a process limit, the drawing should say so, since the shop can then open the process window rather than working to a nominal figure that carries no tolerance.

A stack-up that is drawn rather than described removes most of the ambiguity from a quotation, and it lets the fabricator price the board against the dielectric and copper weights that will actually be used.

Process Control and Verification

On a design of this kind, junction temperature is the item that decides how the rest of the board is arranged. Documenting the assumption is part of the design work, and a short note on the drawing prevents a question that would otherwise arrive a day later and cost a day of schedule. The process window is set by the narrowest step in the flow, so an improvement anywhere else shows up as margin rather than as yield until that step is addressed.

Process Control and Verification

On a design of this kind, junction temperature is the item that decides how the rest of the board is arranged. Documenting the assumption is part of the design work, and a short note on the drawing prevents a question that would otherwise arrive a day later and cost a day of schedule. The process window is set by the narrowest step in the flow, so an improvement anywhere else shows up as margin rather than as yield until that step is addressed.

A first article check confirms that the process and the drawing agree on the points listed above, and that the coupon data supports the values used in the design.

Where a measurement falls outside the expected window, the sample is retained so that the cause can be established before the balance of the batch is released.

FAQ

How accurate can a thermal simulation be? Within a few degrees where the inputs are measured and the geometry is modelled in detail, and much worse where they are assumed.

Is a thermal camera enough? It gives the surface temperature of what it can see, and it needs the emissivity set correctly. It cannot see under a component.

What does gopcb provide for thermal design? We provide thermal modelling with the actual copper geometry, airflow measurement inside the enclosure, interface resistance characterisation, and temperature measurements on the assembled board at the load and ambient extremes.