From Motion.
To Control.

Source and status. These findings review the supplied Detumbler arm v2: design, loads and simulation report, its CAD source and simulation files. This is an internal design-and-simulation package, not a validated publication, physical demonstration or flight result. The website uses the qualifications below where the report overstates the evidence.

Stability times are not an efficacy result. Selected 180-second MuJoCo runs reach the software’s stability threshold after approximately 20–39 seconds. “Stable” means an estimated angular rate below 1 degree per second for more than 2 seconds, not zero motion. The report also describes a magnet-off bearing-friction baseline stopping in approximately 60 seconds. These model outputs do not isolate a magnetic braking benefit; matched baseline runs and physical measurements are needed. Simply subtracting stop times is insufficient.

Overlap samples are not impact events. The reported counts of 18, 1,332 and 165 are geometric safety-overlap samples. Physical collision response is disabled on the target hull, ring and yoke. In case A_roll40_s12, the raw result records 1,081 gimbal overlap samples and 251 arm-link/hull overlap samples. The report’s claim that every overlap was with the stand is incorrect: only the magnet head remained clear of the hull. A clear head alone does not establish a safe arm trajectory.

CAD checking is preliminary. The static checker classifies any fastener-to-part overlap as thread engagement, including overlaps that need separate review. Its zero-interference summary does not certify clearance, assembly fit or manufacturability. HJ2 actuator bolt patterns and cable exits must be confirmed against vendor drawings before fabrication.

Reconcile mass before relying on loads. The selected magnet’s catalog mass is 574.1 g; the CAD accounts for two 245.81 g halves, or 491.62 g. Rotor, head and moving-mass totals, inertias and derived load margins require reconciliation. CAD estimates are not measured assembly properties.

Materials need measurement. Materials are selected to reduce parasitic magnetic interaction. Titanium, carbon composites, stainless pins and brass ballast are not all electrically insulating; the claim that the field only brakes the hull is not established. Characterize coupling in the assembled head and stand.

Model limits. Rotor reaction torque is disabled in the supplied simulation. The applied electromagnetic approximation has no finite-element field validation or measured calibration in this package, and some coefficients are assumed. The replay demonstrates a software run, not validated coupled hardware dynamics.

Integration remains open. Two-camera fusion and gimbal-ring occlusion handling are unfinished. The supplied hardware interface covers serial/simulation paths; a working EtherCAT adapter is still to be implemented and verified. The gimbal’s geometry and balance also limit what can be inferred about free-flight motion.