Answers are grounded in this paper and cite their sources.
Human-powered machines are often sold or described with power figures taken from short efforts: a strong rider can produce several hundred watts for a minute, but a machine that must run for a working day lives on a much smaller number. This paper derives a sizing bound from three stated assumptions: a basal metabolic rate of 82 W, a sustainable total metabolic rate of five times basal for heavy labour, and a gross pedalling efficiency of 0.22. While pedalling, the worker delivers 72.2 W of mechanical power. Under a realistic duty pattern of 50 minutes of work and 10 minutes of rest per hour, and after a drivetrain efficiency of 0.90, the average power delivered to a machine is 54.1 W, with a sensitivity range of 33.2 W to 80.0 W across the tested assumptions. A four-hour shift delivers 0.216 kWh and an eight-hour shift 0.433 kWh. A treadle, worked by one leg intermittently, delivers about 29.8 W on the same duty pattern. The sizing rule follows directly: a human-powered machine intended for all-day use must have a continuous load of no more than about 54 W, and any device whose average draw exceeds that figure will run down any buffer and stop. Commonly advertised pedal-generator outputs of 200 W to 750 W as continuous outputs exceed the while-pedalling bound by factors of 2.77 to 10.39 and do not survive the arithmetic as sustained claims.
No takes yet. Share an insight, caveat, or question.
William Alexander Aldred (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: