Model. A square-section truss (4 longerons, battens, one diagonal per face per bay), clamped to a station at one end. A 3D pin-jointed bar finite-element model solves K u = F (direct stiffness, Cholesky). Loads are superposed: a sideways tip force, the inertial load −m a of a thruster burn acting on every node mass, and a thermal gradient (sunlit side hotter by ΔT) that adds E A α θ to each bar.
Closed forms. Equivalent beam EI = E Al w²; tip load δ = P L³/(3EI) + N P d /(2 E Ad sin²φ) (bending plus bay shear); a thermal gradient bends the truss with curvature κ = α ΔT / w so δ = κ L²/2; first cantilever frequency ω1 = 1.875² √(EI/(μ L&sup4;)).
Assembly. A bay of mass m arriving at speed v and captured at lever arm r makes the station recoil with V = m v/(M+m) and spin ω = μ r v / (I + μ r²) with reduced mass μ = M m/(M+m).
Struts. Compression members are checked against Euler buckling Pcr = π² E It / a² for a thin tube (2 mm wall); the readout shows the smallest margin.
Honest simplifications. Linear elastic pin-jointed bars (joint stiffness, gaps and slack ignored); thermal gradient is steady and linear; no damping or dynamics beyond the first-mode estimate (Rayleigh quotient on the static bending shape); thruster burn is quasi-static; capture is a perfectly inelastic point-mass collision. Real structures need much more analysis.
Independent checks (run on load):
| running… |