Torque is equal to current when it comes to electric motors. When sizing the motor and drive circuit, the discussion usually revolves (pun intended) around the torque of acceleration. This is because the torque required to accelerate the load is generally the largest component of the load requirement. However, there are two aspects to the torque requirement that should be considered.
First, the total torque required is really made up of three components. Torque of acceleration, frictional torque and torque required to overcome the inertia at constant speed. In most cases the frictional torque is small and is often ignored. The torque required to keep the inertia mass moving at constant speed is often a fraction of the torque of acceleration and is taken into account in servo sizing software, so it is not considered separately.
Second, what makes the torque of acceleration so important is that the formula is divided by time. So as the time allowed for the load to move decreases, which is usually what we’re trying to do in motion control, the torque required to accelerate goes up arithmetically. This is why the torque of acceleration dominates the discussion when evaluating motor requirements.
Everything in the control system is oriented as a PID controlled velocity loop, and the other major control loop, current, is being ignored from the programming standpoint. Of course current regulation is performed in the drive circuit between the power electronics and the motor winding. This is required to prevent damage to the drive circuit. But current control has no place in the programming of the trajectory. This is an oversight that needs correction.
The immediate problem is that we don’t have a good rule base to apply current control to trajectory planning. This is however, a great opportunity to improve how motion control is done by the entire industry. Some simple rules come to mind that might demonstrate beneficial results.
What would be the impact of knowing that the trajectory can be divided by the sign of the acceleration? Simply knowing that acceleration is positive, negative or zero would permit better regulation of the load. Knowing that the acceleration is increasing or decreasing has similar potential benefit. If the acceleration of the load leads to a period of constant velocity, then as the acceleration is performed, there is an inflection point where the acceleration force starts decreasing to reach the torque that is required to maintain constant velocity. This approach suggests that acceleration could be dynamically controlled through current and achieve a move profile with little or no overshoot using no gains whatsoever in the control system.
The control system of the future will achieve superior performance because the control model makes use of both speed and an torque to move the load. A more complete model should lead to more realistic control with better performance. That’s what this industry is all about.