2.0 mm boards + SPLIT ATTACH + 3.2 mm bar — RF unchanged on the thicker substrate (902 MHz, S11 −24.5 dB, 8.95 dBi — no retune) · rear attach SPLIT to ±35 mm on the tie: the d² seesaw arm + shorter cantilever let the rear bar thin 6→3.2 mm, and its FDTD cost collapses to ΔF/B 0.02 dB (6 mm: 0.19; tall blade: 0.40) · struts re-optimised on the coupled flexible-board model → CROSSED hexapod (rear struts run forward, front V's backward, restoring diagonality): board modes 33 Hz / 0.26 mm/g · yaw is now bar-limited at ~33 Hz — a brisk 5 rad/s² yaw moves the Δ-null ~7 m°, 15× inside the 0.1° spec, and 33 Hz is below the prop bands · dipole/director flap OUT of the 2P band (227/284 Hz) · net −76 g vs the 1.6 mm spar+ribs build · determinate = not fail-safe

8-strut braced — redundant / fail-safe, ~371 Hz

green FR4 · orange copper · grey ties cross each boom · tan Ø5 rods splay to the CF frame · cyan wall-diagonals lock YAW, the DF axis: 4 radial rods leave yaw an unconstrained mechanism the 1.6 mm board can't hold (steep rods react ~94% out-of-plane onto a board ~1300× too soft) — the diagonals react each node AXIALLY → stiff in all 6 DOF (~370 Hz, above the 2P prop band). 8 struts = fail-safe. The BEST minimal 6 is an octahedral hexapod on a 3-POINT kinematic attach (1 on the reflector-tie mid-span between the boards, 2 on the dir1/dir2 tie at the boom bonds; base triangle anti-aligned on the plate): a yaw torque puts ZERO out-of-plane force on the board (carried in-plane, the ~90× stiffer direction) and 3-point support can't warp the boards — BUT the reflector rear corners end up ~200 mm from any support and droop 0.66 mm/g flapping at 24 Hz, so the amber G10 REAR SPAR (15×5, full width) + RIBS (12×6) are part of the design: board modes 24→44 Hz, sag →0.04 mm/g, +110 g — then it beats the 8-strut on every stiffness number, at the price of NO redundancy (determinate = collapse on one strut/bond loss). Stiffener PLACEMENT is RF-driven (FDTD @0.60λ): a spar along the reflector line costs 0.40 dB Σ F/B, so it sits at the board REAR EDGE (behind the reflector copper) and the ribs at the OUTER spine edge, ~10 mm off the feed slot — measured cost ΔF/B 0.05 dB. Splitting the rear point into two (4-pt, 6 struts) fixes the droop without a spar but leaks in-plane loads out-of-plane at the un-paired attach points → ~23 Hz coupled sway; V-pairs at ALL four points = the 8-strut. Beam fires +x.