Research note / Cycle 021
The floor beneath the witness
A proposed independent reference route is reviewed on paper; five constructed cases show how shared floor motion and a small angle can still produce an apparent error.
A reference needs somewhere to stand
The Space Elevator Initiative still has two mounted measurement channels on the KIS-61/KB-2 fixture and no validated independent witness. On February 8 Reed and Voss reviewed on paper where a third reference could stand. The first route, mounting it on the same base as the fixture, was kept only as a negative control because it was already rejected. The second was a free-standing floor datum beyond the occupied adjacent-bench clamp envelope. It was drawn, not built.
For the second route they recorded the sightline to the point of interest, the access path, trip and load hazards, how the support would be labelled and held, and two questions they could not answer: how much the floor under it moves, and how its alignment would be checked. The review was a pencil floor-plan and sightline walk of the existing footprint. No gauge, stand or specimen was installed and no motion was measured.
A separate stand can still move
A short software calculation makes the concern concrete. Apparent error is taken as the uncoupled share of base motion plus the reference offset times a yaw angle: (1 − coupling) × base motion + offset × yaw. For a 4 µm base motion, a 100 mm offset and 20 µrad of yaw, a stand that does not follow the base shows 6 µm of apparent error. If the stand fully follows the base, the base term vanishes but 2 µm remains from the angle. With no angle and full coupling the error is zero.
All five inputs were chosen for the calculation. They are not measurements of the floor, a stand, the fixture or a ribbon, and the formula is a small-angle sensitivity illustration, not a structural model. Dover Motion’s explanation of Abbé error describes why an offset between the point being observed and the reference introduces an angular term; it says nothing about the actual motion or calibration of Meridian equipment.
What the next step must supply
No installation was approved. Before a reference stand is mounted the team needs a clearance drawing, an assessment of how the support moves, a calibration and uncertainty plan that includes correlated motion, and a safety review. NQ-17 remains returned and no strip is loaded.
The earlier A witness with nowhere to stand removed one tempting location. This review keeps the question open for a better one without treating a drawing as a reference.
Results at a glance
| Constructed case | Base coupling | Base motion (µm) | Offset (mm) | Yaw (µrad) | Apparent error (µm) |
|---|---|---|---|---|---|
| isolated stand, drifting base | 0 | 4 | 100 | 20 | 6 |
| shared base, angular term only | 1 | 4 | 100 | 20 | 2 |
| shared base, zero angle | 1 | 4 | 100 | 0 | 0 |
| separated, zero yaw | 0 | 4 | 100 | 0 | 4 |
| half coupled with yaw | 0.5 | 4 | 100 | 20 | 4 |
External research and methods
- Abbé Error · Dover Motion · Motion Control Handbook, Abbé Error section (Undated online handbook; retrieved 2026-10-26). Abbé error as angular error times reference offset.