Comfort limits: rpm, radius and Coriolis

How fast a habitat can spin, and therefore how small it can be, is the single largest driver of station size. The thread's conclusion was that the data are too thin to fix a number, but the most conservative limits (1–2 rpm) are probably too strict.

Key results

Why rpm matters so much

g-level 1 rpm 2 rpm 3 rpm 5 rpm 7 rpm 10 rpm
Earth (1 g) 1800 m 450 m 200 m 70 m 35 m 20 m
Mars (0.38 g) 680 m 170 m 75 m 30 m 15 m 7 m
Moon (0.17 g) 300 m 75 m 30 m 12 m 6 m –
5% g 90 m 20 m 10 m – – –

Published limits and experiments

Source Max rpm
Hill & Schnitzer 1962 4
Gilruth 1969 6 ("optimum" 2)
Gordon & Gervais 1969 6
Stone 1973 6
Cramer 1985 3
  • The mean is about 5 rpm, against SpinCalc's 2 rpm "green" limit. Hall himself accepts an adaptation period like space adaptation syndrome (1–3 days) Paul451, 2018-07-17.
  • Graybiel/Lackner/DiZio (Brandeis rotating room).
  • About 10 rpm is reachable with stepwise acclimatization; adaptation is fast when early head movement is allowed Paul451, 2015-02-21.
  • Restricting head motion harms adaptation, which may explain inconsistent early results Paul451, 2023-04-19. Brandeis rotating-room video Coastal Ron, 2019-04-04.
  • Engle et al. 2017 (AIAA-2017-5139). After ten incremental 25-minute sessions, 7 of 10 subjects tolerated 15 rpm and 4 exceeded 20 rpm Paul451, 2018-07-18. LMT's caveat: restricted head-tilt tests, not normal activity LMT, 2018-07-18.
  • SpinCalc-cited study: at 5.4 rpm only low-susceptibility subjects performed well, and they were nearly symptom-free by day two Habitant, 2026-09-09.
  • Data quality.
  • Rotating-room studies were all done at >1 g total (spin plus Earth gravity), with the plane of rotation perpendicular to gravity Paul451, 2015-02-14; darkmelmet, 2019-03-19.
  • The longest ran about 12 days Paul451, 2015-02-24.
  • Vomit-comet sessions (~30 s) cannot measure adaptation Paul451, 2015-02-22.
  • Paul451 summarized the data as effectively unusable for long-duration sub-1 g design Paul451, 2016-01-18.
  • A deaf participant in early studies had no discomfort, which suggests vestibular involvement Lampyridae, 2016-01-13.

Working numbers the thread converged on

What you actually feel

  • You cannot sense a constant spin rate. Discomfort comes from Coriolis and cross-coupled accelerations when moving, relative to body size. A 1 m radius for small animals is roughly equivalent to 20–40 m for humans Paul451, 2016-01-17.
  • There is no lean at constant rotation. An early "lean angle" table was corrected: Coriolis acts only when moving radially or changing velocity Paul451, 2016-01-13.
  • Walking prograde or retrograde changes felt g. Running spinward at 22% of rim speed changes perceived g by ±50%. With a rim speed of 10 m/s, a 2.2 m/s jog does this mikelepage, 2022-10-23. The human walk-run transition is about 2 m/s, lower in reduced g Lampyridae, 2022-10-24.
  • Views out of windows. lamontagne's simulated views of Earth at 1 rpm were "fast but calmer"; at 3 rpm they were "almost stressful" lamontagne, 2023-01-25; lamontagne, 2023-01-25. See lighting and windows.
  • Room geometry. Walking in VR through a 24 m radius cylinder showed that cuboid buildings feel wrong after a few metres. Walls should be radial, pointing at the axis Habitant, 2026-09-09; Habitant, 2026-09-11. Paul451 found the square-walled version close to nauseating Paul451, 2026-09-12.
  • Sports and daily life. Ping-pong should be played across the spin direction for fair Coriolis. Basketball hoops belong on the sides of a gym to avoid high ceilings Twark_Main, 2023-06-17; Twark_Main, 2023-06-20. Pouring drinks needs care joek, 2023-06-14.

Open questions

  • Is the maximum adaptable rpm related to g-level? For example, is 6 rpm tolerable at 0.38 g but not at 1 g? Paul451, 2015-02-13
  • Does intolerance recur each time a person moves between rotating and non-rotating areas (see transitions)?
  • Long-duration (months) data at 2–10 rpm in real microgravity do not exist. mikelepage asked for 3–6+ rpm tests in orbit mikelepage, 2023-12-30.

Sources

Forum posts

  • Paul451, 2015-02-13 (reply #272): Radius ∝ period². Post.
  • Paul451, 2015-02-14 (reply #276): Earth rotating rooms >1 g total. Post.
  • Paul451, 2015-02-21 (reply #348, 2 likes): rpm limits: 1, 2–3, 4–6, up to 10 with Lackner & DiZio (2 likes). Post.
  • Paul451, 2015-02-22 (reply #360): Vomit comet can't measure adaptation. Post.
  • Paul451, 2015-02-23 (reply #382): 0.1 g/4 rpm vs 1 g/1 rpm: 160× smaller. Post.
  • Paul451, 2015-02-24 (reply #394): Longest rotating-room studies ~12 days. Post.
  • docmordrid, 2015-03-31 (reply #556, 1 like): UKY centrifuge thesis. Post.
  • Paul451, 2015-03-31 (reply #558, 1 like): Diameter table by g and rpm. Post.
  • Lampyridae, 2016-01-13 (reply #808): ~4 rpm practical agreed limit. Post.
  • Paul451, 2016-01-13 (reply #812): No lean at constant rotation. Post. Attachments: Spin ship - acceleration.jpg.
  • Lampyridae, 2016-01-13 (reply #817): Deaf participant no discomfort; slow spin-up aids adaptation. Post.
  • Paul451, 2016-01-17 (reply #827, 2 likes): Radius uncertainty 3 orders of magnitude. Post.
  • Paul451, 2016-01-18 (reply #830, 1 like): rpm data inconsistent, effectively unusable. Post.
  • Lampyridae, 2018-07-13 (reply #915, 1 like): 5 rpm most adapt; habituation retained ≥30 days. Post. Attachments: ScreenShot_20180713114209.png.
  • Paul451, 2018-07-15 (reply #935): Paul451 roadmap 2018. Post.
  • Paul451, 2018-07-16 (reply #951): Only giant cities if 1 g/1 rpm. Post.
  • Paul451, 2018-07-17 (reply #955): SpinCalc conservative; Hall accepts adaptation period. Post.
  • Paul451, 2018-07-17 (reply #963, 1 like): Hall table of historical rpm limits; mean ~5 rpm. Post.
  • Paul451, 2018-07-18 (reply #971): Engle 2017: 7/10 tolerated 15 rpm, 4 >20 rpm. Post.
  • LMT, 2018-07-18 (reply #977): Restricted head-tilt caveat; 1 g at 3 rpm safer. Post.
  • Paul451, 2018-07-19 (reply #989): 1–6 rpm = 36× scale. Post.
  • darkmelmet, 2019-03-19 (reply #1356): Earth centrifuge differs from space. Post.
  • Coastal Ron, 2019-04-04 (reply #1469, 6 likes): Brandeis rotating room video (6 likes). Post.
  • mikelepage, 2022-10-23 (reply #2180): Resort target on Hall chart: 25–30 m, 3.8 rpm, Mars g. Post. Attachments: gravComfort03-target.jpg.
  • Lampyridae, 2022-10-24 (reply #2190, 1 like): Walk-run transition ~2 m/s. Post.
  • lamontagne, 2023-01-20 (reply #3005): 1 rpm fine for all; most adapt to 10; cheap stations use 3. Post.
  • lamontagne, 2023-01-25 (reply #3082, 3 likes): Window-view videos 1 vs 3 rpm (3 likes). Post.
  • lamontagne, 2023-01-25 (reply #3097, 1 like): 3 rpm view almost stressful. Post.
  • Paul451, 2023-04-19 (reply #3619): Limiting head motion harms adaptation. Post.
  • joek, 2023-06-14 (reply #3864): Tilt and Coriolis when pouring. Post.
  • Twark_Main, 2023-06-17 (reply #3889): Ping-pong orientation and Coriolis. Post.
  • Twark_Main, 2023-06-20 (reply #3895): Gym layout: hoops on sides. Post.
  • mikelepage, 2023-12-30 (reply #4387, 1 like): Want adaptation data at 3–6+ rpm in orbit. Post. Attachments: Bimetal_coil_reacts_to_lighter.gif, dual spiral.png, Dragon SpinG testbed_small.mp4.
  • Habitant, 2026-09-09 (reply #5091, 1 like): 5.4 rpm: low-susceptibility subjects well by day 2; radius table. Post. Attachments: screenshot3.jpg.
  • Habitant, 2026-09-09 (reply #5092, 1 like): VR: cuboid walls mismatch gravity at 24 m radius (2 likes). Post.
  • Habitant, 2026-09-11 (reply #5094, 2 likes): Radial walls, rounded floors (2 likes). Post.
  • Paul451, 2026-09-12 (reply #5096): Square walls off-putting. Post.

External references

  • Hall, T. W. Artificial Gravity and the Architecture of Orbital Habitats (spacefuture.com), including the comfort chart and the table of historical rpm limits.
  • SpinCalc, artificial-gravity.com/sw/SpinCalc/.
  • Engle, J. et al. (2017). AIAA-2017-5139, spacearchitect.org/pubs/AIAA-2017-5139.pdf.
  • Lackner, J. R. & DiZio, P. (2003 and later). Adaptation to rotating artificial gravity environments. J. Vestibular Research.
  • University of Kentucky thesis on short-radius centrifuge blood pooling (uknowledge.uky.edu/cbme_etds/14/) docmordrid, 2015-03-31.