Heat-Sink Thermal Sim (ME 322)

· medium · #eng#class#simulation

Finite-difference 2D conduction solver for an extruded aluminum heat sink, comparing fin spacings against measured bench data.

Coursework project that turned into something I actually liked. Wrote a small Python solver for steady-state 2D conduction across a heat-sink cross-section, then ran a sweep on fin pitch.

Approach

  • Discretized the cross-section on a uniform grid; convective BC on exposed faces, fixed-temperature base.
  • Validated against an analytical 1D fin solution before trusting the 2D output.
  • Compared simulated ΔT to bench measurements with a 5W resistor and four thermocouples.

Results

Simulation tracked measurements within ~8% across three fin pitches, with the gap widening at the tightest spacing — about what you’d expect when the ideal-uniform-h assumption breaks down.