Page 43 - Handbook of Biomechatronics

Actuator Technologies 37

The capabilities of many actuators can be visualized as a subset of this
interplay between force and motion. For example, ideal electric motors
(e.g., without friction or inertia) have a relationship between torque and
velocity that is independent of position or acceleration. This relationship
can be visualized as an envelope of the maximum capabilities of the actuator
in terms of torque and speed (Sensinger, 2010a) (e.g., Fig. 3A). The demands
of the task and nonidealized portions of the actuator (e.g., friction and iner-
tia) can be calculated over time, from which the net torque, position, veloc-
ity, and acceleration can be calculated. Net torques associated with this
motion often include the inertial torque of the motor, gear, and load caused
by the acceleration, the viscous force of the transmission, and the gravita-
tional force of load. The task can then be overlaid on the actuator envelope
as a parametric function of torque vs speed (since both total task torque and
speed were calculated as functions of time). If the profile of the task falls
within the envelope of the motor, then the motor is capable of performing
the task (e.g., see Fig. 3B). This visualization between forces and motions
represents the most accurate understanding of the ability of an actuator to
perform a task, but it is a fairly involved calculation, and is task specific.
Often, designers wish to use a proxy for performance that conveys a gen-
eral sense of whether or not an actuator will be capable of performing a given
task. These proxies often fail to convey important information relevant to
biomedical tasks. For example, most conventional actuators run at constant
speed, whereas most biomechatronic actuators start and stop at rest, with
substantial acceleration/deceleration in between. Designers often look to
proxies that are either particular to their specific applications, or that best
generalize across the many desired attributes. These proxies are a good
way to quickly compare different actuators, but an envelope technique
should often be used in the final stages of verification that takes into account
the dynamical properties of the task. Proxies can either be given as a final
value, or as a normalized value (e.g., density), depending on the type of
comparison being made. Several useful metrics for describing actuators will
be discussed below.

2.2.1 Stall Torque and No-Load Speed Density
The maximum torque, and the maximum speed, that a motor can produce
are both often-considered metrics. For many actuators, maximum torque
occurs when the actuator is not moving, and the maximum speed occurs
when there is no applied torque or acceleration. Because of this, designers
often look at stall torque (the torque when no motion is occurring), or