Abstract Objective. Investigation of the effect of visual feedback on effective hand rim wheelchair force production and the subsequent effect on gross mechanical efficiency. Design. Ten subjects in an experimental group and 10 subjects in a control group practised three weeks ( 3· wk −1 , i.e., a pre-test and 8 trials) on a computer-controlled wheelchair ergometer. Every trial consisted of two blocks of 4 min at 0.15 and 0.25 W· kg −1 at 1.11 m.s −1 . On three trials an additional block at 0.40 W· kg −1 was performed. The experimental group practised with and the control group practised without visual feedback on the effectiveness of force production. Background. In mechanical terms, the low gross mechanical efficiency of hand rim wheelchair propulsion may be the result of ineffective force production. Methods. During all trials oxygen uptake, power output, forces and torque on the hand rims were measured. Results. In comparison with the control group, the experimental group at trial 8 had a significantly more effective force production compared to the control group (90–97% vs. 79–83%, respectively), but showed a significantly lower mechanical efficiency (5.5–8.5% vs. 5.9–9.9%, respectively). Conclusion. Findings indicate that the most effective force production from a mechanical point of view is not necessarily the most efficient way – in terms of energy cost – from a biological point of view and that force direction is based on an optimization of cost and effect. Relevance Learning a more effective force production by visual feedback is not useful for increasing the mechanical efficiency of hand rim wheelchair propulsion.