MV Motor Starting Methods Compared: Calculation Guide

Motor starting methods are the techniques used to energize a medium-voltage (MV) induction or synchronous motor while controlling the inrush current and mechanical torque transient that occur between standstill and rated speed. The right starting method protects the supply network from excessive voltage dip, protects the motor windings and coupled load from thermal and mechanical stress, and determines the sizing of the upstream switchgear, cables, and transformer. This guide walks through the governing formulas, a fully worked numerical example, and a side-by-side comparison of direct-on-line (DOL), reactor, autotransformer, soft-starter, and variable frequency drive (VFD) starting — relevant to utilities, EPC contractors, and industrial plants across the Middle East and Gulf (including Egypt and Saudi Arabia), North Africa, CIS, and Sub-Saharan Africa, where large pumps, compressors, and fans are frequently fed from 6.6 kV, 11 kV, and 13.8 kV networks.

Why motor starting method selection matters

When an induction motor is started, its slip is near 100% and its rotor impedance is at a minimum, so the stator draws a current many times higher than its full-load current (FLC) — commonly in the range of several times rated current for DOL starting of MV motors, depending on design and rotor type. This locked-rotor current is drawn at a near-zero power factor, which causes:

• A voltage dip at the point of common coupling (PCC), which can disturb other loads, relays, and contactors on the same bus.
• High mechanical torque pulsation on the shaft, coupling, and gearbox.
• Thermal stress on stator windings if starting is prolonged or repeated.

Reduced-voltage and electronic starting methods exist specifically to limit this inrush while still delivering enough torque to accelerate the driven load (pump, fan, compressor, mill) within an acceptable time.

The core formulas

1. Rated (full-load) current of the motor

I_FLC = P / (√3 × V × pf × η)

where P is the mechanical shaft power, V is line-to-line voltage, pf is power factor, and η is efficiency. Equivalently, if the apparent power rating S (kVA) is known:

I_FLC = S / (√3 × V)

2. Starting (locked-rotor) current

I_start = k × I_FLC

where k is the locked-rotor current multiple of the motor, a design parameter published on the motor nameplate or datasheet.

3. Starting torque vs. voltage

T_start ∝ V²

Because induction motor torque is proportional to the square of the applied stator voltage, any reduced-voltage starting method (autotransformer, reactor, soft starter) that reduces voltage to a fraction x of rated voltage reduces available starting torque to x² of the DOL value — this quadratic relationship is the central trade-off in starting-method selection.

4. Voltage drop at the point of common coupling

ΔV (per unit) ≈ I_start × X_system / V_system

or, using fault level: X_system = V² / S_fault, so a higher upstream fault level (stiffer source) produces a smaller starting voltage dip for the same starting current.

Worked example: DOL starting current and voltage dip

Consider an 11 kV, 1000 kW MV induction motor with an assumed combined power factor and efficiency of 0.85, fed from a source with a fault level of 500 MVA at the motor bus.

Step 1 — Full-load current

S = P / pf ≈ 1000 kW / 0.85 ≈ 1176 kVA

I_FLC = S / (√3 × V) = 1176 / (1.732 × 11) ≈ 61.7 A

Step 2 — Starting current

Assume a locked-rotor current multiple of 6 × FLC (a commonly used planning figure for MV squirrel-cage motors during DOL starting):

I_start = 6 × 61.7 ≈ 370 A

Step 3 — System impedance from fault level

X_system = V² / S_fault = 11² / 500 = 121 / 500 ≈ 0.242 Ω

Step 4 — Approximate voltage dip during starting

ΔV ≈ (√3 × I_start × X_system) / V_system ΔV ≈ (1.732 × 370 × 0.242) / 11,000 ≈ 155 / 11,000 ≈ 1.4%

In this illustrative case, a 500 MVA fault level is stiff enough that DOL starting produces only a modest ~1.4% dip — acceptable on most networks. If the same motor were started from a weaker 100 MVA source, the system impedance would be five times larger (≈1.21 Ω), pushing the dip to roughly 7%, which could require a reduced-voltage starting method instead. This is the calculation every MV motor starting study must repeat with site-specific fault-level and motor data before selecting a method.

Comparison of MV motor starting methods

Table values are generic, order-of-magnitude engineering relationships intended for first-pass comparison; always confirm with motor manufacturer data, driven-load torque curves, and a dedicated motor-starting study before final selection.

Selection criteria and common pitfalls

• Match torque to load, not just current. A centrifugal pump or fan has a torque curve that rises with speed squared, so even a reduced-voltage method that drops torque to 40–60% of DOL may still accelerate the load successfully — but a positive-displacement compressor with high breakaway torque may stall under the same reduced voltage.
• Check the upstream fault level, not just the motor nameplate. As the worked example shows, the same motor can be started DOL on a stiff grid but may require a reactor, autotransformer, or soft starter on a weak feeder.
• Verify thermal withstand for repeated starts. Frequent starting or extended acceleration time at reduced voltage increases rotor and stator I²t heating; motor data sheets specify permissible starts per hour and maximum locked-rotor time.
• Coordinate protection settings. Starting current transients must be distinguished from genuine fault currents by the protection relay‘s thermal/starting-current discrimination settings — see our protection relay guidance for coordination principles.
• Size upstream switchgear and cables for starting current, not just running current. Circuit breakers, contactors, and cable thermal ratings in the MV switchgear feeding the motor must withstand the selected starting method’s peak current and duration.
• Consider transformer impact. If the motor is fed through a dedicated MV transformer, its impedance adds directly to system impedance in the voltage-dip calculation and may itself need to be sized or tapped to limit starting dip.
• VFD harmonics and bypass. VFDs virtually eliminate starting current and torque transients but introduce harmonic distortion and require filtering/isolation considerations plus a bypass strategy for drive maintenance.

Standards and references

Motor starting performance and ratings are addressed in IEC 60034-1 (rotating electrical machines — rating and performance) and IEC 60034-12 (starting performance of single-speed three-phase cage induction motors). Variable-speed electronic starting and drive equipment above 1000 V is covered by IEC 61800-4 (adjustable speed electrical power drive systems — general requirements for AC drives above 1000 V a.c.). System short-circuit and fault-level calculations referenced in the voltage-dip method above follow the general approach of IEC 60909. These standards should be consulted alongside motor manufacturer data for any formal motor-starting study.

Related guides

See also our guides on medium voltage switchgear, gas-insulated switchgear and ring main units.