Why the Overload Relay Keeps Tripping

The overload has tripped. The motor is stopped. The operator has reset the relay and pressed START, and the motor ran for forty minutes before the relay tripped again. That is a different problem than a motor that trips on startup. It is a different problem than a relay that trips and never resets. The sequence of events — when the trip happens, how long the motor ran, what the load was doing, what the ambient conditions were — is the diagnostic evidence. The relay is not the problem. The relay is the messenger.

A technician who resets the overload and walks away has fixed nothing. A technician who resets the overload four times in a shift has created a thermal damage pattern inside the motor windings that no amount of resetting will reverse. The relay trips because current through the heaters exceeded the thermal model for long enough to actuate the mechanism. Everything else is interpretation — and the interpretation is the job.

What the Overload Relay Measures

A bimetallic thermal overload relay does not measure current directly. It measures heat. Three heater elements sit in series with the motor windings in the power circuit. Motor current passes through these heaters, warming bimetallic strips that deflect as temperature rises. When the cumulative deflection reaches a calibrated threshold, a mechanical trip mechanism opens the 95-96 normally-closed contact in the control circuit. The coil loses excitation. The contactor drops out. The motor stops.

The trip is a thermal integral — current magnitude multiplied by time. A motor drawing 115% of its full-load amperes may run for minutes before the relay trips. A motor drawing 600% at locked rotor trips in seconds. The relay does not know why the current is high. It does not distinguish between a seized bearing and a voltage sag. It accumulates heat and trips when the accumulation crosses the line.

This matters because the first question after a trip is never "is the relay bad?" The first question is: "what thermal condition did the relay respond to?" The answer lives in the motor circuit, the supply, the load, and the environment — not in the relay.

Reading the Trip Pattern

Before opening a tool bag, read the pattern. A single trip after a known event — a jammed conveyor, a process upset, a momentary brown-out — may need nothing more than a reset and a watch. A recurring trip is a diagnostic case. The interval between trips, the load condition at the time of trip, and the thermal history of the motor all shape what to test first.

Trips on startup, every startup. The motor draws locked-rotor current for longer than the relay's trip class allows. Possible causes: seized bearing, locked load, excessive load inertia for the motor and starter combination, wrong trip class on the relay, or a motor winding fault that draws starting current but never transitions to running current.

Trips after sustained running. The motor starts and runs, then the motor trips on overload minutes or hours later. The running current is close enough to the trip threshold that thermal accumulation eventually crosses the line. Possible causes: overloaded driven equipment, voltage sag, phase imbalance, high ambient temperature, blocked ventilation on the motor, or a relay heater sized too tight for the application.

Trips only on hot days. The relay's bimetallic mechanism starts closer to its trip point when the ambient temperature inside the panel is higher. A relay that holds at 25°C ambient may trip at 40°C with the same motor current. Possible causes: panel in direct sun, failed enclosure ventilation fan, heat from adjacent equipment, or the relay is compensated for a narrower ambient range than the installation requires.

Motor hums and fails to start, then trips. A phase is missing. One of the three supply conductors is open — a blown fuse, a pitted contactor tip, a broken wire, or a utility-side fault. The motor sees voltage on two phases only. With the rotor stalled, the current on the two surviving phases can reach locked-rotor levels — far beyond what the windings can sustain. If the motor is already running when a phase drops, the situation is different but still dangerous: the two active windings carry roughly 1.73× their normal current trying to maintain torque, enough to overheat the motor within seconds. A standard bimetallic overload relay may not trip fast enough in either case, because the elevated current on two phases is partially offset by zero current on the third. Single-phasing is one of the leading causes of motor winding failure and warrants dedicated phase-loss protection on critical drives.

Trips irregularly, no obvious pattern. Intermittent mechanical load spikes, intermittent voltage events, or a relay with a marginal heater connection that creates its own local heating. These are harder. They require data.

The First Three Checks

Diagnostic order matters. Testing the relay first is tempting and almost always wrong. A relay that has been in service and tripping correctly for months did not spontaneously become defective on the day it tripped again. Start upstream of the relay — at the supply and the motor — and work toward the relay only after eliminating the conditions the relay is designed to detect.

1. Measure Supply Voltage — All Three Phases

Put a meter across L1-L2, L2-L3, and L1-L3 at the starter's line terminals with the motor running. Not at the disconnect. Not at the MCC bus. At the starter, where the voltage drop through the feeder and the disconnect contacts is included in the measurement.

Three balanced phases on a 208V system should read within 1–2% of each other. NEMA MG 1 (Section 14.36) defines voltage unbalance as the maximum deviation from the average divided by the average. A 2% voltage unbalance can produce current unbalance six to ten times larger in the motor windings, because the motor's negative-sequence impedance is much lower than its positive-sequence impedance. At 5% voltage unbalance, NEMA MG 1 calls for derating the motor to roughly 75% of nameplate — meaning it cannot carry full load without exceeding its thermal limits — and most motor manufacturers void the warranty.

A 208V system reading 208–205–198 across three phase pairs looks close enough at a glance. The average is 203.7V. The largest deviation is 5.7V (the 198V phase). That is 2.8% voltage unbalance — above the 1% threshold where NEMA MG 1 says motors should operate successfully at rated load, and well into the range where current unbalance accelerates winding heating.

Check for complete phase loss while the meter is out. With the motor running, if one phase-to-phase reading is near zero — or if two readings are abnormally low while the third is near normal — a phase is open. The motor may still be turning on two phases, drawing dangerously elevated current on the surviving windings. A standard bimetallic OL relay without differential sensing may not respond fast enough, particularly at partial load.

Voltage sag under load is different from voltage unbalance. If all three phases drop together when the motor starts, the issue is upstream impedance — undersized feeder, long run, loose connection, or a utility supply that sags under the facility's demand profile. The motor draws more current to produce the same torque at lower voltage, and the relay responds to that current.

If the line-side readings look normal but trips persist, repeat the measurement at the motor terminals — across T1-T2, T2-T3, and T1-T3. Voltage drop through the contactor poles, overload heaters, and cable run can introduce enough imbalance at the motor that the line-side readings miss.

Record all three readings. Record them with the motor running under load, not at no-load. Record the time of day — utility voltage profiles shift between morning and peak afternoon demand.

2. Measure Motor Current — All Three Phases

Clamp an ammeter around each of the three load conductors at T1, T2, and T3. With the motor running under its normal load, all three phase currents should be reasonably close to each other and at or below the motor's nameplate full-load amperes (FLA).

Take readings twice — once under load and once at no-load. The difference tells you where the current is going. If the motor draws 18A under load and 6A unloaded, the 12A difference is the load component. If it draws 15A under load and 14A unloaded, the motor is consuming most of that current itself — bearing drag, winding degradation, or a rotor issue. A motor whose no-load current is unusually high relative to its loaded current — compared to manufacturer data or a baseline reading taken when the motor was healthy — warrants further investigation even if the loaded reading is below nameplate.

The nameplate is the reference. Not the relay setting. Not the NEC table. The motor's own nameplate FLA, at the motor's own nameplate voltage and frequency, is the thermal design point. NEC 430.6(A)(1) says motor branch-circuit and feeder calculations use NEC Table 430.250, but the overload relay is sized to protect the specific motor — and the motor's nameplate FLA is the anchor for that sizing per NEC 430.32.

What you are looking for:

All three phases high and roughly equal. The motor is overloaded. The driven equipment is drawing more torque than the motor is rated to deliver continuously. Check the mechanical load: coupling alignment, bearing condition, belt tension, conveyor or pump blockage, or a change in the process since the motor was originally sized. Mechanical overload is the single most common cause of recurring OL trips.

One phase significantly higher than the other two. Current unbalance. Trace it back to the supply voltage measurement. If the voltages are balanced but the currents are not, the problem may be inside the motor — a winding fault, a high-resistance connection at a motor terminal, or a partially shorted turn that changes the impedance of one phase.

One phase reads zero, the other two are elevated. Single-phasing. A supply conductor is open. The motor is running on two phases, and the two active windings are carrying approximately 1.73× their normal share of the current. This condition can burn out windings in seconds. Find the open — blown fuse, pitted contactor tip, broken conductor, or loose terminal.

All three phases at or below FLA, but the relay trips anyway. The relay heater may be undersized for the motor. Or the relay's ambient temperature compensation is not adequate for the enclosure temperature. Or the relay itself has a defect — a cracked heater, a corroded connection at the heater terminal, or a trip mechanism that has drifted from calibration. This is where the relay becomes a legitimate suspect.

3. Check the Relay Sizing

The overload relay heater must match the motor. Not the circuit. Not the wire. The motor.

Pull the motor nameplate FLA. Pull the heater element table from the relay manufacturer — Allen-Bradley, Eaton, Square D, Siemens, and ABB all publish selection tables that cross-reference motor FLA against heater catalog numbers. Compare the installed heater to the table. If the plant has swapped the motor since the original installation, or if the motor has been rewound, the heater may no longer be correct.

NEMA starters with bimetallic overload relays use replaceable heater elements. The trip class — Class 10, Class 20, or Class 30 — defines how long the relay allows the motor to draw locked-rotor current before tripping from a cold start. Class 10 trips within 10 seconds at 600% of FLA. Class 20 trips within 20 seconds. A motor driving a high-inertia load that needs 15 seconds to accelerate through starting current will trip a Class 10 relay every startup, regardless of the heater size.

Electronic overload relays replace the bimetallic mechanism with current transformers and a microprocessor. The FLA is set by a dial or a digital parameter, and the trip class is selectable. These relays also offer phase-loss detection, phase-unbalance detection, ground fault sensing, and configurable alarm contacts — protections that a bimetallic relay cannot provide. On an OL trip diagnosis, an electronic relay often shortens the investigation: most display the trip cause (overcurrent, phase loss, ground fault, stall), per-phase current, and thermal capacity used. Check the relay's fault register before reaching for a meter — it may already contain the answer.

The relay heater check is third, not first, because it answers a narrow question. A correctly sized heater on a motor drawing excessive current will trip exactly as designed. The heater is not wrong — the current is wrong. Finding the heater mismatch only matters after the supply and motor current measurements have established that the current is within normal range for the motor and load.

When the First Three Checks Are Inconclusive

Supply voltage is balanced. Motor current is at or below FLA on all three phases. The heater matches the nameplate. The relay still trips. Now the investigation moves deeper.

Insulation resistance (megger test). A motor with degraded winding insulation can draw excess leakage current that generates heat inside the winding without showing up as a measurable current spike on a clamp meter. Measure insulation resistance phase-to-phase and phase-to-ground with the motor disconnected from the starter, using 500V DC for motors rated under 1000V. Per IEEE 43-2013, the minimum acceptable insulation resistance for modern low-voltage motors is 5 megohms (corrected to 40°C). Compare all three phases — a reading significantly lower on one phase indicates localized insulation breakdown.

Winding resistance comparison. Measure the DC resistance of each phase winding at the motor terminals with the motor disconnected. All three should be within 1–2% of each other. A shorted turn in one winding lowers its resistance and increases its share of the current — enough to trip the OL over time without showing up as gross current unbalance on a clamp meter.

Mechanical inspection. Spin the shaft by hand with the motor de-energized. Listen for bearing noise. Check coupling alignment with a straightedge or dial indicator. Feel for axial play. A dragging bearing adds mechanical load that increases motor current just enough to push past the relay's trip curve — particularly in a high-ambient enclosure where the relay's thermal margin is already reduced.

High-resistance connections. A loose lug or corroded terminal introduces resistance that dissipates heat and can cause voltage drop at the motor terminals. An infrared scan of the starter, disconnect, and motor junction box under load is the fastest way to find these. Alternatively, measure voltage drop across each connection point — anything above a few millivolts at rated current warrants investigation.

Documenting the Evidence

A trip diagnosis is not complete when the motor restarts. It is complete when the reason for the trip is recorded and the corrective action addresses the cause.

The forensic habit is a written record: date and time of trip, ambient temperature, load condition, all three supply voltages, all three motor currents, relay heater part number, motor nameplate FLA, and any mechanical observations (noise, vibration, heat, smell). That record is evidence. If the motor trips again next week, the record from today becomes the baseline that separates "same problem" from "new problem."

Maintenance management systems, CMMS work orders, and even a logbook in the MCC room serve this function. The format matters less than the discipline. A technician who documents the electrical measurements alongside the trip event is building a thermal history of the motor that compounds in value over time. A technician who resets and walks away is starting from zero on every call.

The overload relay cannot tell you why it tripped. It can tell you that it tripped. The diagnostic reasoning — what was measured, what was ruled out, what was found — belongs to the technician, not the device.

Hands-On Practice

Reading is one thing — wiring it yourself is another. Open the interactive trainer and build this circuit from scratch.

Practice OL trip diagnosis in the trainer →