Simplified beta-type visualization (the real engine is alpha-type — see post) · Working fluid: air · Hot reservoir: Prestidigitation warm envelope · Cold reservoir: Prestidigitation chill envelope · Output: friction-heated iron
What's happening, step by step
- Heating (0°→90°): The displacer sits near the cold (right) end, so most of the air is in the hot (left) chamber bathed by 40°C water. The air warms, expands, and pressure rises — pushing the power piston outward to the right. This is the power stroke; mechanical work leaves through the connecting rod.
- Transfer to cold (90°→180°): The crankshaft carries the displacer leftward, shuttling air past it (around its loose-fitting edges) into the cold chamber on the right.
- Cooling and compression (180°→270°): Air now sitting against the 5°C wall loses heat and contracts; pressure falls. The flywheel's stored momentum pushes the power piston back inward at this lower pressure — so it costs less work to compress than was extracted during expansion. The net work over the cycle is the difference.
- Transfer to hot (270°→360°): Displacer moves rightward again, shuttling the now-cool air back into the hot chamber. The cycle repeats.
The 90° phase offset is the whole trick
The displacer and power piston are both linked to the same crankshaft, but their crank pins are 90° apart. That offset means whenever the gas is hot (high pressure), the power piston is in the middle of moving outward; whenever the gas is cold (low pressure), the piston is moving inward. Pressure pushes outward more than it resists being pushed back in — net positive work per cycle.
From flywheel to iron
The flywheel transmits its rotation through a drive belt to a small friction wheel pressed against a ~60 g iron target (2 cm cube). Friction converts mechanical work directly into heat, concentrated in a much smaller mass than the warm reservoir that supplied the original energy. That's how you get the temperature "amplification": you trade total energy (most stays in the warm envelope as waste heat) for energy density in the target.
About the numbers: Carnot efficiency at this 35 K spread is about 11.2%, with realistic small-engine performance around ~7% (roughly 60% of Carnot, consistent with a well-built Mark I). The energy delivered per cycle in this animation is intentionally accelerated so you can watch the iron heat up in seconds rather than hours — a real desk-sized Stirling running on a 35 K gradient delivers fractions of a joule per cycle, not the ~150 J shown here. The shape of the cycle and the relationships between parts are accurate.