Elite cyclists often train on roller cylinders because the low inertial load allows them to fine time their muscular coordination. This observation suggests that effective crank inertial load (and hence crank load dynamics) affects pedaling coordination. However, this possibility has not been adequately investigated. Experimental studies have not thoroughly investigated pedaling under different inertial loads. Similarly, computer modeling studies often model the limb dynamics explicitly while only accounting for the frictional aspect of the load. Thus, a potentially important element has been omitted from the biomechanical foundation necessary for studying muscular coordination during pedaling.

This dissertation investigates how to model crank load dynamics and the significance of such models to steady-state pedaling biomechanics. The investigation proceeded in three stages. First, the characteristics of the crank load during steady-state pedaling were determined. A one degree-of-freedom crank load model was developed to emulate the crank load dynamics of a standard ergometer (low inertial load) and a road bicycle (high inertial load). Based on this model, a laboratory ergometer was modified to emulate the crank load dynamics of a specific road-riding situation. A dynamic optimization was then used to evaluate the model’s ability to simulate experimental trajectories from emulated ergometer pedaling and emulated road riding. A one degree-of-freedom model was adequate to emulate the crank load dynamics of a standard ergometer. However, due to the frequency response characteristics of the load, a more complex two degree-of-freedom model was needed to emulate the crank load dynamics of a road bicycle.

Next, the influence of the crank load on steady-state pedaling biomechanics was investigated. Experimental data was collected from ten male subjects to compare the biomechanics of low and high inertia pedaling. Inertial load was found to have only a small influence on some biomechanical quantities (i.e., driving pedal force, crank torque, and net muscle joint torques) but a profound influence on less commonly reported quantities (i.e., cadence variability, driving pedal force variability, and crank angle variations). Consequently, though the biomechanics of standard ergometer pedaling appears to be similar to that of road riding, the neuromuscular control of the two tasks may be somewhat different. Finally, the usefulness of the crank load to a mechanical power analysis of how net muscle joint torques produce pedaling movements was demonstrated. The power analysis was derived from a dynamical model of seated, two-legged pedaling that included the dynamics of the crank load. This new power methodology allows one to compute how each net muscle joint torque contributes to crank and limb mechanical power. To develop the necessary inputs to the analysis, a dynamic optimization was used to adjust net muscle joint torque estimates from inverse dynamics only enough to produce a two-legged simulation that tracked experimental trajectories.

The power analysis revealed that net ankle and hip extensor joint torques function synergistically to deliver power to the crank during the downstroke. To do this, the net hip extensor joint torque generates power to the limb, while the net ankle extensor joint torque transfers this power from the limb to the crank. In contrast, net knee extensor and flexor joint torques function independently to deliver power to the crank through the top and bottom of the stroke, respectively, which prevents freewheel decoupling. During the upstroke, net ankle extensor joint torque transfers power from the crank to the limb to restore the potential energy of the limb. In both halves of the crank cycle, gravity forces augment power transfers performed by net ankle extensor joint torque between the crank and the limb.

These experimental and computer modeling results lay the biomechanical foundation needed for future studies which will investigate the coordination of individual muscles during pedaling.