Correctly sized does not necessarily mean correctly performing

(See motec at MachineBuilding.Live, 14 October 2026, on stand 248)

The symptoms are familiar: vibration, excessive settling time, following errors, audible noise or an axis that becomes unstable as the control gains are increased. At that point, it is tempting to revisit the motor selection. But torque and speed calculations may already show that the motor can do the job. The problem is that motor sizing establishes capacity; it does not establish dynamic performance.

A servo axis is a closed-loop system comprising the motor, drive, feedback device, transmission and machine structure. Each element influences the performance that can actually be achieved.

Example 1: The calculation can be right while the machine is wrong

Calculating peak and RMS torque requirements is important. Peak torque determines whether the motor can meet acceleration and load requirements, while RMS torque establishes whether the motor can operate thermally within its duty cycle.

However, these properties do not define how quickly or cleanly the axis can respond on their own. Consider a high-inertia load driven through a flexible coupling. The motor may have more than enough torque to accelerate the load, but the coupling and mechanical structure introduce compliance between the motor and the load. Leading to an increase in oscillation, settling time, and a difference in motor and load position. All completely independent of the motor properties.

Before changing the motor, consider if mechanical properties of the system need adjusting, or maybe if a load-mounted encoder could help with end point control and accuracy.

Example 2: When tuning exposes a mechanical problem

An engineer increases the proportional or velocity gain to improve response. The axis becomes more dynamic, but vibration increases, particularly at one speed. Increase the gain further and the vibration becomes more pronounced, and the axis begins to oscillate.

At this point, many engineers endlessly try to adjust the control loops further, hoping for a better outcome, but there comes a point where mechanical resonance cannot be overcome through software adjustment alone. Sometimes a call must be made between mechanical re-design or accepting less dynamic motion for a reduction in resonance.

Example 3: Think about the control system as a whole

The motor cannot be considered independently from its drive and controller. Feedback compatibility, current capacity, regenerative requirements, control-loop bandwidth, communications, safety functions and the required motion-control functions all influence drive selection.

The controller and communications architecture also become important where multiple axes need to operate synchronously. Electronic gearing, camming, registration and coordinated motion can place requirements on timing and synchronisation that are not apparent from the motor's torque-speed characteristics.

Two motor, drive and controller combinations may therefore be electrically compatible while producing very different machine performance.

Example 4: Getting the requirements right. What is really going to be your limiting factor?

Positioning accuracy is commonly stated as a headline requirement, but in a high-speed machine the more useful question may be: how quickly does the axis reach that position and become stable?

An axis can reach its commanded position but still be unsuitable for the application if it takes too long to settle before the next operation can begin. This is where performance bandwidth, resonance and mechanical stiffness all become critical. A high-resolution encoder cannot compensate for a mechanically flexible structure. Equally, a high-performance servo drive cannot create stiffness that is not present in the machine.

The required cycle time should therefore be considered alongside positional accuracy when designing the axis. A specification that only states the final position tolerance can hide the dynamic requirement that actually determines whether the machine will achieve its intended throughput. Dig down into customer requirements wherever possible.

Conclusion: Design the axis, not just the motor

When designing an axis to achieve its intended performance, the motor is rarely the only consideration. The useful sequence is to look at the axis as a whole system:

• Is the torque-speed requirement correct for the actual duty cycle?
• Is the inertia relationship appropriate? (load inertia 5x motor inertia is typically the limit for highly dynamic applications)
• Is the transmission introducing excessive compliance or backlash?
• Is there anything that is likely to mechanically resonate?
• Is the feedback device precise enough and measuring the position that actually matters?
• Does the drive have sufficient bandwidth and the appropriate control functions?
• Can the machine structure support the required dynamic response?

At motec, motion-control applications are assessed across the motor, drive, feedback, transmission and machine dynamics rather than treating individual component selection as isolated calculations. That system-level approach is particularly valuable where the requirement is not simply to move a load, but to move it accurately, repeatedly and within a short cycle time.

A correctly sized servo establishes that the motor can do the work. The engineering challenge is ensuring the complete axis can perform the motion.