Reaction torque measures of six different nutrunners were correlated w
ith EMG measures and subjective ratings of perceived exertion for diff
erent work surface orientations, joint stiffnesses, rpm levels, air pr
essure levels and handle configurations. Using an underpowered tool or
using a softer joint resulted in larger impulses and worse ratings. A
pulse-type shut-off mechanism produced the lowest reaction torque and
the best rating. Overall, ratings correlated directly with the impuls
e level. The subjects were good judges of stress levels, and subjectiv
e ratings were valid measures of operator stress.