Eliminating resonance in stepper motor applications

Engineers are increasingly turning to stepper motors for a simple and cost-effective positioning solution. However, due to the design of these motors and the way they are driven, there is an inherent risk of resonance under certain operating conditions. This can manifest as vibrations, which impact on the synchronisation between the command and the actual rotor position – ultimately affecting accuracy. Daniel Muller discusses the importance of preventative measures to avoid the resonance that can lead to these problems

Stepper motors are a popular option in a host of positioning applications, as they can be simply driven, step-by-step, without the need for an encoder or additional device to provide position feedback information. Commutation from an external electronic driver moves the rotor from one stable position to the next, and as long as one phase is energised without switching to the next phase, the rotor holds a stable position.

If we consider the operating mechanism more closely, however, we can see the potential for problems. If the rotor moves slightly ahead of the target position the motor will develop a negative torque to pull the rotor back to the target position. On the other hand, when the rotor has not yet reached the target position, a positive torque will draw the rotor towards it. In these conditions, oscillation can easily occur.

What this means in practice is that if the rotor speed and/or the inertia from the load causes overshoot from the target position, the rotor will then start oscillating around the target position, with a natural frequency that is a factor of the holding torque, the number of poles and the total moment of inertia. The magnitude of the oscillation will reduce over time with the inherent damping of the system, but if the oscillation is not over before the next step is commanded there is a risk of resonance, with the mechanical system responding with oscillations of greater amplitude.

The problems that can be caused by resonance should not be underestimated, potentially leading to a loss of steps, possible changes in direction of rotation and general erratic motion.

Measures to prevent resonance

Resonance typically occurs when the commutation frequency is close to the mechanical system’s natural frequency of vibration. It is not always possible to keep the commutation frequency away from the resonance frequency, and thus avoid it. It may be an option instead to shift the system’s natural frequency either up or down. This can be done by working on two of the parameters that affect it: holding torque and the total inertia of the system.

Holding torque depends on the rated current of the motor. Using a higher current to increase the holding torque has the unwanted effects of increasing joule losses, leading to increased coil temperature. However, it is possible to use a lower current to shift the natural frequency down if the lower torque still meets the application requirements.

Looking at inertia, the mechanical system’s moment of inertia is the sum of the motor’s rotor inertia plus the load inertia. The design engineer could select a motor with different specifications to change to the rotor inertia. Or it may be possible to adjust the load inertia to shift the natural frequency up or down, if this doesn’t impact on the motor’s performance in the application.

Another measure to prevent resonance is to operate the motor in microstepping mode rather than driving the motor with full steps. The smaller step angle requires less energy to move from one stable position to the next, so the resulting overshoot and magnitude of oscillation are smaller. In addition, microstepping generally offers lower noise, less vibration and a smoother operation.

Preventing resonance with damping

As discussed previously, the motor system’s inherent damping will steadily decrease the magnitude of oscillation in many applications and prevent resonance occurring. So where resonance is a problem, can we increase damping to eliminate the problem? The answer is yes, and there are a number of mechanisms available to do so.

Mechanical friction provides a braking torque that is constant and independent of speed. Increasing load friction or motor bearing friction might be an option to increase damping in some applications, although because it acts on the motor at all speeds it is important to ensure that the motor’s performance will not be compromised.

A better option is generally to add viscous friction; this also provides a braking torque, but its magnitude is dependent on motor speed (being higher at higher speeds). Thus, it provides strong braking while the oscillation amplitude is great and only very light braking once the oscillation is smaller.

There are a number of different phenomena that can bring viscous friction to a system. Eddy-currents generated in the iron of the stator act as a braking torque, but these levels of these iron losses vary between different motor technologies. Disk magnet motors usually have only limited iron losses; this enables them to achieve high speeds, but means eddy-currents alone should not be relied upon to prevent resonance.

The back-EMF (voltage) induced in the coil provides a current that also creates a braking torque that will dampen oscillation and can also be considered viscous friction. The effectiveness depends on the motor drive – chopper drivers typically don’t enable this type of damping since the current is kept constant despite back-EMF variations.

The design engineer might also want to consider other electronic damping solutions, driving the motor in a particular way without changing any mechanical parameters within the system. Or there may be the option to add an external mechanical damper to absorb some of the vibration energy to prevent resonance.

In any stepper motor application where resonance is a problem, it can often be attributed to a number of conditions. Sometimes acting on just one of these conditions can be sufficient to eliminate resonance. In any application, it is always worth engaging with a knowledgeable supplier to determine what frequency ranges are likely to cause a resonance problem, and to look at the possible solutions to eliminate it.

Daniel Muller is a Portescap Application Engineer.