Loads. The pre-tensioned sail pulls on each boom tip through two edge lines at 45° to the boom, so the boom is a column in compression C = 2 T cos45° = √2 T. A thin-walled tube has I = π r³ t.
Euler buckling. Pcr = π²EI / (K L)² with K = 2 (free tip), 1 (pinned) or 0.7 (tip held by tensioned lines). While deploying, the unsupported length grows, so Pcr falls as 1/L². The boom buckles at the length where Pcr(L) = C, i.e. Lcrit = (π/K) √(EI/C). That is the deployment-failure mode the page shows (the boom turns red and the unrolling stops).
Imperfections. A real boom is never straight and the sail pushes sideways. The bow grows as δ = δ0 / (1 − C/Pcr) (Southwell/beam-column amplification), with δ0 = F L³/(3EI) from the quarter of the sail force on one boom plus the initial crookedness. The bow is drawn exaggerated by the chosen factor; a buckled boom is drawn at true size.
Sunlight. Radiation pressure P = (1+ρ) S₀/(c r²) (normal incidence, ρ = 0.9 reflected), force F = P A cos²α, lightness number β = Fsrp/Fgrav.
Simplifications. Linear elastic isotropic tube; no local tube-wall (shell) buckling, no hub compliance, no thermal bending, no sail billowing load, no deployment dynamics (the boom unrolls at constant speed, quasi-static loads). Real booms (e.g. coilable composite or tape-spring types) also fail by local flattening and torsion, which this model does not include. Numbers are illustrative and not tied to any specific mission.
Running numerical checks…