Will the chip throttle? Does the airflow actually reach the fins? Will the snap-fit crack? We simulate your enclosure, heatsink and plastic parts, compare design options side by side, and hand you a report that states the answer first.
Conjugate heat transfer — airflow and solid conduction solved together — to tell you whether the chip overheats and the fan keeps up.
Static FEA with pass/fail against material allowables and a safety factor — strain-based for plastic parts.
Several design options compared at once, delivered as a readable report — conclusion first, then validation and limits.
Three complete studies, published in full. Reports are in Traditional Chinese; English deliverables are available on request.
Q · Does the SoC throttle at full load in the official case with the stock heatsink? How much does a larger heatsink buy?
Stock config hits 81.5 ℃, past the threshold; the larger heatsink drops it to 52.6 ℃ at almost no extra fan load. Heat shifts from 76 % via PCB to 74 % via heatsink.
Q · 15 W chip, 40 mm fan — which inlet/outlet and heatsink layout runs coolest? Four layouts compared.
A simple shroud drops the chip from 64.1 ℃ to 50.3 ℃. Over half the original airflow bypassed the fins; side-in/top-out was worst at 83.4 ℃.
Q · Will a PC/ABS snap-fit crack under repeated assembly? What do fillets, tapering and length each buy?
Sharp-corner and fillet-only designs fail; tapered beam + R0.75 + 11 mm length gives 1.16 % root strain, SF 1.29, 5.3 N assembly force.
Every number in the report is checked against something independent before it reaches you.
What must be answered, what counts as a pass (throttle temp, allowable strain), and what is out of scope.
Parametric model from your STEP or drawings, keeping what matters for flow and heat, simplifying the rest.
Solved with OpenFOAM / CalculiX, design variants run in parallel.
Mesh convergence, energy balance, analytical or hand-calc checks, and published test data where available.
Conclusion, comparison, figures, limits and a test plan — in one report.
STEP or 2D drawings, materials, power dissipation or loads, and the fan model for thermal work. Incomplete data is fine to start — assumptions are listed in the report.
No. It compares options before tooling or prototyping so you test fewer variants. Every report lists its assumptions, limits and a recommended test plan.
Comparing two to four design options usually takes one to two weeks, depending on geometry and validation needs.