This is an earlier experiment. Read the amendments below before treating its findings as current.
The question
Wind had first been found to stop deliveries somewhere between 6 and 9 m/s, and the spacing setting seemed to do nothing.
What was tested
Wind was rerun at a cruise speed of 12 m/s as well as the default 8. Closest approach was recorded in metres so that spacing settings could be compared.
What was found then
Half of deliveries were gone at a wind of 0.75 times cruise speed, and 95 % at 1.0 times, at both cruise speeds. The limit is a ratio of wind to airspeed. Deliveries nearly stopped (61 trips at 9 m/s, 1 at 15) with zero collisions. The spacing setting only matters above a floor set by the craft's own size (2.5 × its marker), which covers a minority of traffic here. Above 15 m it is costly: going from 15 to 30 m buys only 2.3 m of real clearance, for a third of throughput.
What changed
Closest approach in metres is recorded in every result.
Numbers and source references
- 6 and 9 m/s
- first wind estimateresearch/envelope.md@9d286de
- 12
- second cruise speed, m/sresearch/envelope.md@9d286de
- 8
- default cruise speed, m/sresearch/envelope.md@9d286de
- 0.75 × cruise
- wind at which half of deliveries are goneresearch/envelope.md@9d286de
- 95 %
- deliveries gone at 1.0 × cruiseresearch/envelope.md@9d286de
- 1.0 × cruise
- windresearch/envelope.md@9d286de
- 2.5 × marker
- craft-size floor on separationresearch/envelope.md@9d286de
- 15 m
- separation above which it saturatesresearch/envelope.md@9d286de
- 15 → 30 m
- separation stepresearch/envelope.md@9d286de
- 2.3 m
- real closest approach gainedresearch/envelope.md@9d286de
- a third
- throughput costresearch/envelope.md@9d286de
- 61, 15 and 1
- trips at wind 9, 12, 15 m/sresearch/envelope.md@9d286de (Wind table)
- 9, 12 and 15 m/s
- windsresearch/envelope.md@9d286de (Wind table)