Many daily activities, from basic grooming to employment tasks, require adequate unsupported arm endurance (UAE). We developed an electromechanical device to measure UAE endurance. The purpose of this study was to standardize the instrument for two rates of arm motion, moderate and slow, in 18 normal adult subjects (FEVI = 3.7L ± .78, FVC = 4.2L ± .74, FEV1 FVC = 1.1 ± .08). Exercise endurance limits, and the following metabolic, ventilatory, and sensation responses were determined at rest prior to exercise and at end-exercise limits for both rates of UAE: minute ventilation (Ve), tidal volume (VT), respiratory rate (RR), duty cycle (Ti/Ttot), oxygen uptake (V̇O2), carbon dioxide production (V̇CO2), inspiratory flow (VT/Ti), heart rate (HR), and visual analog scale measurements (VAS) of dyspnea (D), respiratory effort (RE), and arm fatigue (AF). Significance increases from baseline rest were shown at the endurance limits for both rates of UAE in: V̇O2, V̇CO2, Ve, VT, RR, VT/Ti, HR, VAS-D, VAS-RE, and VAS-AF. There were no changes in TiT/tot and SaO2 with UAE. Peak V̇O2, RR, Ve, VT/Ti, and VAS-D with moderate were significantly greater than slow UAE; and there was a trend increase in peak HR for moderate as opposed to slow rate UAE. Despite these differences, the endurance time between the two rates of UAE were similar. These data provide standards against which UAE in COPD can be evaluated. © 1993.