Induction Motor Starting Voltage Dip Case Study

Large induction motors can disturb a distribution system during starting. The motor may run efficiently after acceleration, but its locked-rotor current can be several times full-load current. If the source impedance is high enough, that current creates a voltage dip that can drop out contactors, reset controls, trip drives, dim lighting, or disturb other production loads.

This case study follows a 480 V motor-control center where a new pump motor causes nuisance trips when started across the line. The event is realistic rather than tied to one site. The engineering task is to determine whether the voltage dip is credible from the motor-starting current, whether the measured sag matches the electrical model, and which mitigation is technically defensible.

The purpose is to show how an electrical engineer should connect motor kVA, locked-rotor current, transformer impedance, existing load, sensitive-load ride-through, starting torque, protection settings, and validation evidence.

Case Context

A process building adds a 300 kW chilled-water pump motor. The motor is connected to an existing 480 V motor-control center supplied from a 13.8 kV to 480 V transformer. During the first across-the-line start, the pump accelerates, but several control contactors chatter and a packaging line trips on undervoltage. The event lasts less than 6 seconds, so normal steady-state metering does not explain it.

The simplified system data are:

Item	Value
transformer rating	1500\ \text{kVA}
transformer secondary voltage	480\ \text{V} line-to-line
transformer impedance	5.75\%
existing coincident load before start	620\ \text{kVA}
existing-load power factor	0.86 lagging
pump motor shaft output	300\ \text{kW}
motor efficiency at rating	0.94
motor running power factor	0.88 lagging
locked-rotor current	6.0 times full-load current
locked-rotor power factor	0.25 lagging
measured minimum MCC voltage during start	0.88\ \text{pu}
measured sag duration	4.8\ \text{s}
contactor dropout concern	below about 0.80\ \text{pu}
PLC power-supply alarm concern	below about 0.85\ \text{pu}

The values are simplified for a teaching case. A real motor-starting study should include utility source impedance, transformer X/R ratio, feeder impedance, motor locked-rotor code or manufacturer data, load torque-speed curve, acceleration time, protection curves, control-power ride-through, and measured RMS voltage trend.

Event Evidence

The event record shows:

1. the voltage dip begins at the instant the pump contactor closes;
2. the minimum voltage occurs during the first acceleration interval;
3. current decays as the pump accelerates;
4. no downstream fault indication appears;
5. the motor protective relay does not trip;
6. the affected packaging line trips on undervoltage rather than overcurrent.

This evidence points to a starting voltage dip, not a short circuit or transformer differential event. The next question is whether the calculated dip is consistent with the measured 0.88\ \text{pu} minimum.

Motor Rated Apparent Power

Motor output power is not the same as electrical apparent power. Estimate rated input apparent power from shaft power, efficiency, and running power factor:

Substitute:

The corresponding full-load current at 480 V three-phase is:

This current is reasonable for a 300 kW low-voltage motor. It is not the starting current.

Across-the-Line Starting Current

For direct across-the-line starting:

The apparent starting power is:

On the 1500 kVA transformer base:

The existing load is:

The starting event therefore asks a 1500 kVA transformer to support existing load plus a transient motor-starting current that is larger than the transformer rating by itself. That does not automatically mean the transformer is thermally overloaded for a few seconds, but it does mean voltage dip must be checked.

Voltage-Dip Screening

A simple voltage-dip screen uses transformer impedance and per-unit current. If the transformer impedance dominates and the source is stiff:

For motor-starting current alone:

Existing load also produces voltage drop. A simple conservative screen adds the existing-load component:

Total approximate dip:

Estimated bus voltage during the start:

The estimate is close to the measured minimum:

This agreement supports the diagnosis. The voltage dip is credible from motor starting and source impedance. It is not necessary to invent a hidden fault to explain the event.

The calculation is still a screen. A detailed study would use complex impedance, current phase angle, feeder voltage drop, source impedance, motor acceleration current, and time-varying torque-speed behavior.

Sensitive-Load Interpretation

The measured dip did not fall below the approximate contactor dropout concern of 0.80\ \text{pu}, but it did fall below the PLC power-supply alarm concern of 0.85\ \text{pu} only slightly less than the measured minimum margin would suggest. Field devices vary, and some control supplies may be fed from smaller control transformers with additional voltage drop.

The engineering interpretation is:

• the pump start is acceptable for the pump motor only if acceleration and thermal limits pass;
• the same start is not acceptable for the building distribution system if other loads trip;
• steady-state voltage after acceleration does not validate starting performance;
• the mitigation must protect both motor acceleration and sensitive-load ride-through.

Mitigation Options

Three options are reviewed.

Option	Electrical effect	Engineering risk
keep direct start	simple, high starting torque	repeats voltage dip and nuisance trips
soft starter with current limit	reduces line current and voltage dip	reduced starting torque may extend acceleration
variable-frequency drive	lowest electrical disturbance and speed control	harmonics, configuration, cost, bypass, and protection changes

The pump is centrifugal and starts with a partially closed discharge valve. That helps because required starting torque is lower than for a loaded conveyor, crusher, compressor, or positive-displacement pump.

Soft-Starter Current-Limit Check

Assume a soft-starter current limit of 3.5 times full-load current.

Starting current:

Starting apparent power:

Per-unit on the transformer base:

Voltage dip from the soft-starter start current:

Add the same existing-load component:

Estimated bus voltage:

This is a substantial improvement over 0.887\ \text{pu} and should provide much better ride-through margin for ordinary control loads.

Starting Torque Check

Reducing voltage or current reduces motor torque. A soft starter is acceptable only if the motor still accelerates the load.

Assume direct-start locked-rotor torque is 1.8 per unit of rated torque. Approximate torque scales with the square of current ratio for this screening:

The pump breakaway torque with the discharge valve partly closed is estimated at:

Torque margin at breakaway:

The soft starter appears feasible for this pump, but the margin is not universal. A high-inertia load, loaded conveyor, compressor, or pump started against an open discharge valve could fail to accelerate with the same current limit.

VFD Comparison

A variable-frequency drive can start the motor with controlled frequency and voltage. Suppose the drive limits line current to about 1.2 times motor full-load current while accelerating.

Approximate VFD starting kVA:

Per-unit current on transformer base:

Approximate voltage dip including existing load:

Estimated bus voltage:

The VFD gives the best voltage-dip result and may also save energy if the pump often runs at reduced flow. It also adds harmonic, grounding, cable, cooling, control, bypass, cybersecurity, and maintenance considerations. If the only problem is occasional starting sag, a soft starter may be the better proportional mitigation. If flow control is valuable, the VFD may be justified.

Failure Modes

The event exposes several failure modes that are easy to miss in steady-state design.

Failure mode	Consequence	Evidence or control
starting current underestimated	bus sag larger than expected	motor data sheet and measured start trend
transformer impedance ignored	weak 480 V bus during acceleration	per-unit voltage-dip screen
sensitive load not included	unrelated equipment trips	ride-through inventory and event logs
current limit set too low	motor stalls or overheats	acceleration test and thermal model
soft starter bypass closes too early	current step returns	commissioning trend
VFD selected without harmonics review	power-quality problem moves elsewhere	harmonic and grounding review
protection settings ignore acceleration	nuisance trip or inadequate stall protection	relay curve and start-time verification

The case is therefore not only a motor problem. It is a distribution-system operating case.

Corrective Decision

The engineering recommendation is:

Replace direct across-the-line starting with a soft starter initially limited to 3.5I_{FL}, start the pump with the discharge valve in the specified position, verify acceleration time and bus-voltage trend, and hold a VFD as the preferred option if speed control or repeated low-flow operation justifies the additional system changes.

The line should not be released on calculation alone. Commissioning must prove:

1. minimum MCC voltage remains above the ride-through criterion during start;
2. affected control supplies, contactors, PLCs, drives, and relays do not trip;
3. motor accelerates within allowed time;
4. motor thermal model remains within limit;
5. soft-starter bypass transition does not create a second voltage dip;
6. protection settings tolerate normal acceleration but still protect against stall;
7. start trend is retained as baseline evidence.

Validation Plan

A practical validation test should record:

• three-phase RMS voltage at the MCC bus;
• motor current during acceleration;
• acceleration time from start command to running state;
• soft-starter current limit and ramp settings;
• pump discharge valve position;
• affected control-power voltage;
• event logs from the motor relay, PLC, and packaging line;
• upstream feeder current and voltage trend if available.

Acceptance criteria for this case:

Quantity	Acceptance value
minimum MCC voltage during start	\ge0.90\ \text{pu}
acceleration time	\le8\ \text{s}
control-power undervoltage alarms	none
motor thermal utilization during start	below relay alarm limit
repeated starts	inside manufacturer and relay limits
upstream feeder trip or alarm	none

These criteria are project-specific. A hospital, data center, islanded microgrid, mine ventilation fan, fire pump, or safety-critical process may require stricter ride-through, redundancy, or starting method requirements.

Lessons for Engineers

Motor starting is an operating case, not an afterthought. A motor that satisfies running kW can still disturb the distribution system while it accelerates. The correct review connects motor data, transformer impedance, existing load, sensitive-load ride-through, mechanical torque, protection, and commissioning evidence.

Transferable lessons:

• Use apparent power and current, not only shaft kW, when checking motor starting.
• Put starting current on the same per-unit base as the transformer or source.
• Compare calculated sag with measured RMS voltage trends before changing equipment.
• Reducing starting current also reduces starting torque.
• Sensitive loads can trip even when the starting motor itself is healthy.
• VFDs reduce starting disturbance but introduce their own system requirements.
• A successful mitigation requires a recorded start trend, not only a revised setting.

The engineering question after the fix is simple:

Can the system start the motor, keep other loads alive, protect the motor, and prove all of that with measured evidence?

If any part is missing, the motor-starting study is not complete.