PLE-SBS · Slidewire crew-escape study, paper and simulator · May to Jun 2026 · Paper dated 10 June 2026
In the hours before a crewed launch, astronauts and close-out technicians work 40 to 115 m above a fully fuelled rocket. If something goes wrong they need a way down that is faster than the lifts, works with the power off and keeps working in strong wind. NASA's slidewire baskets do this, but each has been engineered for one pad.
PLE-SBS is a slidewire basket system and a method for sizing one at any pad. It combines a hazard model that turns any rocket's propellant into a safe distance, a taut-cable model in which every pod's weight changes the slope under every other pod, and a Monte Carlo evacuation simulator, all in one dependency-free physics file with sixteen unit tests. Without tuning, it reproduces the Shuttle-era baskets' reported top speed: 24.0 m/s predicted, about 24.6 m/s reported.
The worked example is a generic 42 m pad with nine people to clear at danger level 4. Five design iterations, each run deterministically, at both wind corners and 1,000 times in Monte Carlo, took the design from a copy of the Artemis layout that failed 93% of runs to one that fails none.
The 95th-percentile time until everyone is safe for the worked example: 42 m pad, nine people, danger level 4, 1,000 Monte Carlo runs, no failures. The first design, a copy of the Artemis layout on this low pad, gave 231 s with 93% of runs failing: from 42 m a 400 m wire ends almost flat, and no passive brake setting satisfies both wind corners. Geometry fixed it, a shorter, steeper wire into a shallow pit; then a bunker instead of a vehicle; then the human factors.
The central finding is general. A passive magnetic brake strong enough to stop a full pod in a tailwind can strand a lightly loaded pod in a headwind, because its force grows with speed and a light pod is braked harder. Passive braking therefore works only in a narrow diagonal band of geometry: for crew access heights of 30 to 120 m, the span should be about 100 m + 4 × the height.
The existing pads sit where their braking choices predict: the Shuttle pad on the band, with its catch nets and drag chains; the Artemis tower on the steep side, where it uses magnetic brakes along the line; SLC-41 on the flat side, where the riders control their own brakes.
Between pods on one wire with the switch yard, against 37 s with a single dock. My add-on moves the braking off the shared wire. At the terminus the rope hands each trolley onto a ground rail, and self-indexing points, a railway turnout advanced one position by a ratchet that each passing pod trips, send pod 1 to bay 1, pod 2 to bay 2, and so on. Each bay has its own 60 m magnetic brake, drag-chain bed and net. The release lever is interlocked by a block marker 60 m down the wire, so if the pod ahead ever stalls the next stays at the platform. Nothing needs power.
24 people in six pods on one wire with the switch yard, in a design headwind: 286 s with a single dock, 111 s with six separate wires.
a full pod whose magnetic brake has failed, stopped in its bay's drag-chain bed (1.76 g with a tailwind). The catch net of D4 gave 3.8 g, above the 3 g double-fault limit.
safety factor of the 36 mm galvanised steel rope on breaking strength. Steel keeps about half its strength near 500 °C; polymer ropes melt at 145 to 260 °C.
Peak pod speed predicted for the Shuttle-era LC-39B baskets with no tuning, against about 55 mph (24.6 m/s) reported: within 3%. The same model says a tall tower needs magnetic braking over a large part of the descent, not only at the end, which is what the Artemis system does.
All results are model predictions from public data and stated assumptions, not a certified safety analysis. The reference launcher and pad are generic; the danger levels and the capsule exit time are my estimates; debris, basket sway, rope stretch and the loss of eddy-current force at high speed are not modelled.
The scale model's pictures are renders of the part files. The timed runs that fit rolling resistance and drag area to measurement, through the simulator's scale-model tab, are the next step.
Where it came from, and how it was built
In September 2025 I designed roller-coaster tracks around the limits of the human body, for a CBSE state-level modelling competition: the thrill and the limit turned out to be the same number, and a track is a machine for giving people forces safely. I presented that model at ISRO in March 2026, and the scientists there raised a different question: a crew works a hundred metres up a tower beside a fuelled rocket; if the alarm sounds, how do they get down? PLE-SBS is the answer, carried from the hazard model to a simulator, a seven-page paper and a 1:50 model printed on a Bambu Lab P1S.
