Electric Motors News
How to size medium voltage industrial motors for high-inertia loads
Size medium voltage industrial motors for high-inertia loads with confidence. Learn how to check torque, acceleration, thermal limits and starting methods to avoid costly failures.

How to size medium voltage industrial motors for high-inertia loads

Sizing medium voltage industrial motors for high-inertia loads requires more than matching nameplate power to steady-state demand. A motor may run a large fan comfortably once it reaches speed, yet overheat during acceleration or fail to accelerate through a weak point in its torque-speed curve.

The selection must satisfy three conditions together: enough torque throughout acceleration, sufficient thermal capacity for the starting duty, and acceptable electrical demand on the supply. Increasing the kilowatt rating can help, but it can also increase starting current and rotor inertia. The better decision starts with the driven equipment—not the motor catalogue.

Define the starting condition before choosing the rating

Large fans, centrifuges, mills and flywheel-equipped machinery can all present high inertia, but they do not impose the same starting duty. A fan may have relatively low aerodynamic torque at low speed, while a loaded mill can demand substantial breakaway torque. High inertia describes resistance to a change in speed; it does not describe the resisting torque.

Request the load torque-speed curve for the actual startup arrangement. Damper position, process pressure, material loading and lubrication condition can change the result. An unloaded-start assumption is only useful if the process controls reliably enforce it.

The initial equipment data should include:

  • Required shaft power and speed across normal and maximum operating conditions.
  • Driven-equipment inertia, its reference shaft and its units.
  • Breakaway torque and resisting torque throughout acceleration.
  • Gear ratio, transmission efficiency, coupling arrangement and any disengaged components.
  • Permitted acceleration time, starts per hour and restart sequence after a trip.
  • Site voltage, frequency, ambient temperature, altitude and cooling constraints.

Establish the continuous power requirement separately. For shaft torque T in N·m and speed n in rpm, P = T × n / 9550 gives power in kW. Apply justified allowances for operating variation and site derating, rather than stacking unrelated margins. Neither a larger nameplate nor a service-factor allowance, where specified, proves that the motor can handle an extended start.

Put every inertia value on the motor shaft

Acceleration calculations require total inertia referred to one shaft. For a geared load, the inertia reflected to the motor is:

Jload,ref = Jload × (ωload / ωmotor)2

Add the motor rotor inertia and the appropriately reflected inertia of couplings, gears and other rotating parts. Reflect resisting torque separately, accounting for transmission losses. A gearbox therefore changes both the torque requirement and the inertia seen by the motor; treating it only as a speed conversion misses half the problem.

Check the notation before calculating. Supplier documents may use J, WK² or GD², with conventions that differ between sources. Do not assume they are interchangeable. Ask for an explicitly defined mass moment of inertia in kg·m², or a documented conversion.

This step also makes motor selection iterative. A larger candidate can have greater rotor inertia, so the acceleration calculation must be updated when the proposed motor changes.

Compare complete torque-speed curves, not isolated torque figures

The torque available to accelerate the system is the difference between motor torque and load torque:

Tacc(ω) = Tmotor(ω) − Tload(ω)

For fixed total inertia, acceleration time is calculated over the starting trajectory:

tacc = ∫ Jtotal / [Tmotor(ω) − Tload(ω)] dω

In practice, this is often evaluated in speed increments using manufacturer curves. The endpoint should be the defined completion of acceleration near the operating speed, not an assumed synchronous speed for an induction motor.

The important feature is the smallest accelerating-torque margin. Strong locked-rotor torque does not guarantee a successful start if pull-up torque falls too close to the load curve later. Where the curves nearly touch, the motor can spend a long time drawing substantial current with little increase in speed. If motor torque falls below resisting torque, it cannot accelerate through that region under those conditions.

A simplified calculation shows the scale. Assume, purely for illustration, a total inertia of 1,000 kg·m² and a constant net accelerating torque of 5,000 N·m up to 1,500 rpm. With a speed change of approximately 157 rad/s, acceleration takes about 31 seconds. This is not a motor-selection result: actual net torque varies with speed, voltage and starting method.

Use minimum expected motor-terminal voltage when checking a line-started motor. At fixed frequency, induction-motor torque approximately follows voltage squared over comparable operating conditions. A voltage dip can therefore extend acceleration materially. Rated-voltage curves alone may give a misleadingly comfortable result.

Check thermal capability against the whole operating sequence

“It reaches speed” is not an adequate acceptance criterion. The motor must do so without exceeding its rotor or stator thermal limits, including when it is already hot.

Request allowable acceleration time, hot and cold starting capability, thermal damage curves and the permitted starting sequence. Ask whether stated limits apply to locked-rotor conditions or to the proposed accelerating duty. A locked-rotor withstand time is not automatically the allowable duration of every start: current, rotor frequency and heat distribution change during acceleration.

Repeated starts deserve particular attention. A process may require one normal start, an aborted attempt, then a restart before the rotor has cooled. That sequence can be more demanding than the nominal starts-per-hour figure suggests. Confirm minimum intervals between starts and recovery requirements after unsuccessful acceleration.

Cooling also matters. Shaft-mounted ventilation is limited at low speed, and a variable-speed application may require independent ventilation for sustained low-speed operation. Stator temperature detectors are useful, but they do not directly establish rotor thermal safety during a long start. Protection should combine suitable thermal modelling with the manufacturer's limits.

Choose the starting method around the torque margin

The starting method changes the motor-sizing problem. It should be evaluated alongside the motor, not selected after the rating has been fixed.

Direct-on-line starting

Direct-on-line starting provides the motor's full-voltage torque characteristic, subject to supply voltage depression. It can be appropriate where the network supports the starting current and the mechanical train tolerates the transient. Verify the complete acceleration duty; a robust supply does not remove rotor-heating limits.

Reduced-voltage starters and soft starters

Reducing voltage lowers current but also reduces available torque. This trade-off can work for a load with modest starting resistance and sufficient torque margin. It can fail on a high-inertia load when current limitation leaves too little torque to accelerate within the thermal limit.

A soft starter is not a variable-frequency drive. It does not maintain controlled motor flux by reducing frequency with voltage. Require a coordinated motor-and-starter calculation showing acceleration time, current profile, starting-duty capability and bypass transition where applicable.

Variable-frequency starting

A medium-voltage variable-frequency drive can provide controlled torque over a programmed ramp without conventional direct-on-line inrush. For difficult starting duties, this may avoid an unnecessarily large motor. However, available acceleration torque remains limited by the drive's current and overload capability, motor characteristics and cooling.

Evaluate the complete drive package: converter-duty insulation requirements, output waveform, cable length, bearing-current mitigation where needed, input harmonics and low-speed ventilation. If rapid stopping is required, calculate stored kinetic energy, E = ½Jω², and specify how it will be absorbed or returned. A system that accelerates satisfactorily may still need separate braking provisions.

Check the electrical network and mechanical train together

A motor cannot be approved in isolation from its supply. The starting study should include transformer impedance, upstream source strength, feeder impedance and other connected loads. Calculate voltage at the motor terminals as well as at sensitive buses. Starting from a generator or a weak utility connection may require more detailed dynamic assessment.

Protection settings must permit legitimate acceleration while remaining coordinated with motor thermal limits. Simply increasing a trip delay to accommodate a slow start can leave a stalled motor inadequately protected.

On the mechanical side, verify coupling and gearbox transient torque limits. Flexible shafts and gear trains can require torsional analysis, particularly where starting transients or drive torque pulsations may excite resonances. A single lumped-inertia calculation predicts overall acceleration; it does not establish shaft-system integrity.

Make the supplier prove the proposed combination

Before approving medium voltage industrial motors for high-inertia duty, request a coordinated submission—not separate catalogue sheets. It should contain the proposed motor's torque-speed and current-speed curves, rotor inertia, the agreed load curve, acceleration results at specified voltage conditions, and confirmation of hot and cold starting capability.

Include the actual starter or drive settings, restart restrictions, thermal-protection coordination and mechanical limits in that review. Applicable IEC 60034 or NEMA MG 1 requirements should be identified by relevant edition and scope, but standards compliance alone does not demonstrate suitability for a particular high-inertia start.

For the Global Power & Electrical Grid Matrix (GPEGM), this is where motor, power-electronics and grid intelligence intersect: the useful comparison is the complete operating duty, not nameplate efficiency or kilowatts alone. If a proposal depends on unloaded starting, a minimum bus voltage or a cooling interval, put that condition into the operating specification. The selected motor is only as reliable as the assumptions the plant can actually maintain.

Related News