Stirling engine: designed, analysed and machined by hand
A complete Stirling engine taken from SolidWorks model to a full set of manufacturing drawings, stress-checked for a space mission case study, then machined part by part and assembled in the university workshop.
Key numbers
The project
A Stirling engine turns a temperature difference into motion: heating the gas in a sealed chamber drives a displacer and piston, which spin a flywheel. The brief was to design one completely in CAD, document it to a manufacturable standard, and then make it for real.
I machined every part myself in the university workshop and assembled the engine by hand, turning my own drawings into a working mechanism.
What I did
- CAD: modelled every component and the full assembly in SolidWorks, from the top and bottom plates to the crank wheels, piston, displacer and fasteners.
- Drawings: produced a 22-sheet drawing pack with dimensions, sections and detail views ready for the workshop.
- Thermal FEA: simulated the M4 bolts at 85 °C in steel, aluminium, titanium and nylon to compare expansion against weight.
- Machining: made each part by hand in the workshop, then assembled and set up the finished engine.
Stress case study
In a team of five we re-engineered the engine to survive on the James Webb Space Telescope, with a hot side at 85 °C and a cold side at −233 °C. Each of us analysed one component to a factor of safety of at least 3.
- My part, the piston con rod: peak shear stress of 112.4 MPa at the end loops under a 320 N load.
- Buckling: critical load of 28.3 kN from Euler's formula, far above the working load.
- Material: selected Ti-8Al-1Mo-1V titanium (910 MPa yield, 2.5 g part), for a factor of safety of 8.1.
The finished engine
coming soon
coming soon
coming soon
Gallery
Looking back
What went well
Seeing a design go from SolidWorks to a real object I had machined myself. Producing a full set of 22 manufacturing drawings forced me to think about tolerances and fits, and the stress and buckling hand calculations gave clear, defensible safety factors.
Even better if
I'd design more closely around the workshop's real capabilities, as some features were harder to machine than they looked in CAD. I'd also back the hand calculations with more FEA and test the finished engine, for example measuring flywheel speed against temperature difference.