A three-dimensional biomechanical model of the ankle was developed and then was used to determine the muscle and joint forces which occur at the ankle during the stance phase of running. The passive moments and forces that occur at the ankle during stance were also determined. The forces predicted by this three-dimensional model and a similar two-dimensional model were compared, and the effects of changes in some of the major assumptions of the three-dimensional model were examined.

At each instant the foot was assumed to be a rigid body in equilibrium with the external ground reaction forces and moments and the internal forces and moments exerted by tendons and other structures which cross the ankle joint. The application points and lines of action of these tendon forces were established from the averaged measurements from five dissected cadavers. The number of unknown tendon forces were reduced to five by grouping together muscles which tend to act together. Only two equivalent muscle groups were assumed to be active at any instant. This model yields the minimum joint forces possible for equilibrium, since any additional muscle activity would produce greater joint forces. The joint forces calculated were assumed to act at a point mid-way between the tips of the medial and lateral maleoli.

The passive internal moments exerted on the ankle by the stretching of tissues were experimentally determined for different ankle angles from a group of seven subjects. A series of cubic equations was derived from this data relating joint angles to passive ankle moments. These curves were used to estimate joint forces which occurred due to passive stretch.

This model was used to calculate the ankle joint and muscle forces of three subjects running at a six-minute-mile pace. External forces and moments acting on the foot were measured by a force platform. Maximum absolute joint forces were found to range between 9.0 to 13.3 times body weight. The plantar flexion muscle group was found to exert the highest maximum forces, ranging from 5.3 to 10.0 times body weight. Passive forces were found to contribute approximately 10 percent to the compression forces at the joint.

A two-dimensional model based on the same anatomical data was used to predict muscle and joint forces for one of the subjects. The two-dimensional model predicted joint forces during stance which were slightly lower than those predicted by the three-dimensional model.

Some major assumptions in the original model were modified to determine the effect of each on the forces predicted by the model. It was found that the grouping of muscles into functional groups and the method of distributing force among the individual muscles of each group made little difference in the joint forces predicted by this model. However, variations in relative body dimensions among subjects and variations in the assumed position of the center of the joint forces were found to affect the joint forces predicted by this model to a great extent.