Skip to content
Faultpath
Chapter 0555 min read

Motors & Motor Starters

Three-phase induction motors, contactors, overloads, and why they trip.

  • Slip, and why load shows as current
  • Three protective jobs, three devices
  • Overload trip curves and trip class
  • Single phasing and imbalance
  • Diagnosing a repeat trip

What it is

A three-phase induction motor is the most reliable thing on the plant, and most of the calls you will get about one are not about the motor.

That is the frame worth carrying into this chapter. Motors are rugged, they are simple, and they mostly fail for reasons outside themselves: something is loading them harder than it should, a supply has gone wrong, or a starter component has worn out. A motor that has genuinely failed is a real thing and it is not the way to bet first.

How it turns

Three-phase supply into three windings spaced around the stator produces a magnetic field that rotates at a speed fixed by the supply frequency and the number of poles:

Synchronous speed (rpm) = 120 × frequency ÷ poles

A 4-pole motor on 60 Hz has a synchronous speed of 1800 rpm; on 50 Hz, 1500. The rotor is dragged along by that field but always runs slightly behind it, because it has to cut the field to generate any torque at all. That difference is slip, and it is typically 2–5 %, which is why the nameplate of that 4-pole 60 Hz motor says something like 1750 rpm rather than 1800.

Slip is the useful mental model for the whole chapter: more load means more slip means more current. When somebody tells you a motor is drawing more current than it used to, they have told you something about the load before they have told you anything about the motor.

Three protective jobs, three devices

The word "starter" covers an assembly that does three genuinely different things, and conflating them is the source of most confused fault reports.

| Device | Protects against | Speed | | --- | --- | --- | | Fuse or circuit breaker | Short circuit — hundreds or thousands of amps | Milliseconds | | Contactor | Nothing — it switches | On command | | Overload relay | Sustained modest overcurrent, protecting the motor's thermal life | Seconds to minutes |

A fuse will not protect a motor from running 20 % overloaded for an hour — the current is nowhere near enough to blow it, and the motor cooks. An overload relay will not protect anything from a dead short — it is far too slow. Each covers what the other cannot.

Reading the nameplate

The nameplate is the specification you compare every measurement against, and it is the first thing to photograph on any motor job.

  • Full-load current (FLA/FLC) — what it should draw at rated load. Your clamp meter reading means nothing without it.
  • Voltage and connection — star or delta, and dual-voltage motors have different connections for each. A motor connected for the wrong one is a short and expensive story.
  • Rated speed — reveals the pole count and confirms slip.
  • Service factor — permitted continuous overload margin, often 1.15. Running there continuously shortens life; it is headroom, not a rating.
  • Insulation class — how hot the windings may get. The rule of thumb that matters: roughly every 10 °C above rating halves insulation life.
  • Duty and frame — whether it is rated for continuous running, and what fits.

How it works

The starter, drawn out

POWER CIRCUITL1L2L3Fuses — short-circuit protection onlyK1 main contacts — switchingOverload heaters — thermal model of the motorM3-phase induction motorCONTROL CIRCUITLNSTOPSTARTK1 SEAL-INOL CONTACT (NC)K1The overload does not break the motor current.It opens this contact, which drops out K1.
Fig 5.1 — A direct-on-line starterPower circuit on the left, control circuit on the right. Note where the overload actually acts: its heaters carry motor current, but its contact sits on the control rung. The overload never breaks the motor current itself — it tells the contactor to, which is why a tripped overload leaves the power circuit perfectly intact and why pressing reset restarts the motor without addressing anything.

The control rung is the seal-in circuit from Chapter 3: press start, the coil energises, and its own auxiliary contact holds it in when you let go. Stop, the overload contact, and any interlocks are all in series with that coil, so any one of them opening drops the motor out. Chapter 8 takes this apart properly.

Starting current, and why starting methods exist

A motor at standstill has no back-EMF, so the winding presents almost nothing but its own low resistance. A direct-on-line start therefore draws six to eight times full-load current for the second or two it takes to get up to speed.

That inrush is why alternatives exist: star-delta starting (windings in star for the start, giving about a third of the DOL current and a third of the torque), soft starters (ramping voltage), and VFDs (ramping frequency, Chapter 6). Each trades starting torque or cost against inrush.

The overload is a thermal model, not a threshold

This is the idea that makes overload faults diagnosable.

1000 s100 s10 s1 s1×2×4×6×8×10×Current, as a multiple of the motor’s full-load currentTime to tripCLASS 20CLASS 10DOL START6× for ~2 s — passes underneath1.15× sustained — trips, but only after minutesAn overload is a thermal model of the motor, not a current threshold.
Fig 5.2 — Overload trip curvesDeliberately slow at small overcurrents and fast at large ones, because that is how heat actually accumulates in a winding. A DOL start at six times full-load current for two seconds passes underneath the curve without tripping. A sustained 15 % overload trips — but only after minutes, because that is how long it takes to do thermal damage.

An overload relay does not ask "is the current above a number". It approximates how hot the windings are getting, which depends on how much current for how long. That is why the curve slopes: 6× for two seconds is harmless and 1.15× for ten minutes is not.

Trip class names the curve. A class 10 overload trips within 10 seconds at 6× FLA; class 20 within 20 seconds; class 30 within 30. Motors with long run-ups — big fans, loaded conveyors, high-inertia loads — need a slower class, or they trip every time they start.

Two kinds you will meet:

  • Thermal (bimetallic) — heaters warm a bimetal strip that snaps a contact. Cheap and robust. Sensitive to panel ambient temperature unless it is ambient-compensated, which is why a starter in a hot panel in July can trip on a load it handled in January.
  • Electronic — measures current directly and models the thermal behaviour in firmware. Usually adds phase-loss and phase-imbalance protection, which matters for the failure below.

The setting goes to the motor's nameplate FLA. Not to the cable rating, not to whatever stops it tripping.

Single phasing

If one of the three supply phases is lost — a blown fuse, a burnt contact, a loose terminal — a running motor does not stop. It keeps turning on the remaining two, with reduced torque and badly increased current in those two phases, typically around 1.7 times normal. It gets hot fast.

A motor that is stopped when a phase is lost will not start at all: it hums, draws heavy current, and does not turn.

Whether the overload catches this depends on what it is. A three-element thermal overload usually will, eventually. A single-element or poorly matched one may not, and this is precisely why electronic overloads with explicit phase-loss detection are worth having. The tell on the clamp meter is unmistakable once you look for it: two phases high and roughly equal, one at zero.

Phase imbalance

Even with all three phases present, unequal voltages cause disproportionate current imbalance — a rough rule is that a small percentage of voltage imbalance produces several times that percentage of current imbalance, and the extra current is heat.

So when you clamp all three phases, the number you actually care about is not any single reading, it is how close they are to each other. Balanced and slightly high is a loaded motor. Unbalanced is a supply or connection problem, and you are now looking upstream rather than at the motor.

What normally fails

Symptom
Overload trips after the machine has been running for a while, resets fine, then trips again
Likely cause
A genuine mechanical overload — the driven load has got harder, not the motor weaker
How common
Very common

A seized or dry bearing downstream, a jammed or overfilled conveyor, a blocked fan or filter, product build-up, a belt over-tensioned at the last service. The overload is doing its job correctly and reporting a real thermal condition. Turning it by hand with the power off tells you more in ten seconds than any measurement.

Symptom
Motor hums but will not turn, or runs weakly and gets very hot
Likely cause
Single phasing — one supply phase lost at a fuse, contact or terminal
How common
Common

Two phases reading high and roughly equal with the third at zero is the signature, and it is unmistakable once you clamp all three. Left running, this destroys a motor in minutes rather than hours. Find the lost phase upstream — fuse, contactor contact, terminal, or the isolator itself.

Symptom
Contactor chatters, buzzes, or drops out under vibration
Likely cause
Low or unstable control voltage, or a failing coil — not usually the contacts
How common
Common

A coil that is not getting its rated voltage cannot hold the armature firmly, so it buzzes and the contacts arc themselves to destruction. Measure the coil voltage while it is pulled in and the machine is loaded, because a supply that reads fine unloaded can sag when everything energises together.

Symptom
Contactor main contacts are pitted, welded, or the motor will not stop
Likely cause
Ordinary switching wear, accelerated by frequent starting or a chattering coil
How common
Common

Contacts erode every time they break load current, and a welded contact is the parallel-fault character from Chapter 3 — the machine will not stop when it should. Check contact condition on any starter that has been jogged or frequently cycled.

Symptom
Overload trips on starting, every time, but the motor runs happily once up to speed
Likely cause
Trip class too fast for the load's run-up time, or an overload set below nameplate FLA
How common
Common

A high-inertia load takes longer to accelerate than a class 10 overload will tolerate. The fix is the correct trip class for the run-up, not a higher setting — the distinction matters, because one preserves protection and the other removes it.

Symptom
A starter that has been reliable for years starts nuisance-tripping in hot weather
Likely cause
Panel ambient temperature rising, on a thermal overload that is not ambient-compensated
How common
Occasional

The overload is modelling motor temperature using a bimetal strip that is itself sitting in a hot enclosure. Check panel ventilation and filters before concluding anything about the motor. Electronic overloads largely remove this.

Symptom
Motor fails to earth, trips protection, or fails an insulation test
Likely cause
Winding insulation breakdown from heat, moisture, contamination or age
How common
Occasional

The genuine motor failure, and the least common item on this list. Heat is usually the underlying cause, which means something on this page caused it — the overload that was turned up, the phase that was lost, the bearing that was dragging.

How to troubleshoot it

  1. Establish which device actually operated

    Safety

    Blown fuse, tripped breaker, tripped overload, or a drive fault — four different reports and four different directions. Look at the devices rather than accepting "it tripped". A tripped overload with intact fuses is a thermal story; a blown fuse is a short-circuit story.

  2. Read the nameplate and compare it with the overload setting

    FLA on the plate against the number set on the overload, and the trip class against what the load needs to accelerate. This is two minutes and it resolves a surprising share of nuisance-trip calls outright.

  3. Find out when it trips

    On starting, after a few minutes, after an hour, only when the machine is full, only in the afternoon. Trips on starting point at trip class or run-up time. Trips after a period of running point at a genuine thermal condition. Trips that correlate with ambient point at the panel.

  4. Turn the load by hand — with the machine isolated

    Safety

    Isolate and prove dead first (Chapter 2). Then turn the shaft. A load that is stiff, notchy, or will not turn has told you the answer before you have taken a single electrical measurement, and this is the step most often skipped because it is not electrical.

  5. Clamp all three phases under load, and compare them

    Safety

    The absolute value matters against the nameplate; the balance matters more. Two high and one at zero is single phasing. All three high and equal is a loaded motor. Unequal is a supply or connection fault upstream. Note the machine's actual duty at the time, since the numbers are meaningless without it.

  6. Check supply voltage at the motor terminals, all three phases

    Safety

    Balance again, measured under load rather than at rest. Small voltage imbalances produce disproportionately large current imbalances, and the extra current is heat in the windings.

  7. Inspect the starter itself

    Main contacts for pitting and welding, terminals for discolouration and looseness, the coil voltage while energised and loaded. A voltage-drop measurement across each main contact while the motor runs finds a high-resistance contact that looks perfectly serviceable.

  8. Insulation test the motor and its cable — isolated

    Safety

    Only once the simpler explanations are eliminated, and following Chapter 4: electronics disconnected, appropriate test voltage, and the result judged against the previous reading rather than against an absolute figure.

  9. Answer why it failed before you restore it

    An overload trip is a report about a thermal condition, and a burnt contact is a consequence of something. Replacing the component without answering the question buys a shift. Record the currents you measured and the machine's duty so the next person has a baseline rather than an anecdote.

Common technician mistakes

  • Resetting the overload and walking away

    Why

    It restores production in five seconds and the machine runs, which is a powerful combination when a supervisor is waiting. But the overload is not a fault, it is a report, and resetting it clears the report rather than the condition. The second and third trips are the same fault becoming more expensive.

  • Turning the overload setting up

    Why

    The logic is seductive: it trips at 12 A, the motor obviously wants more than 12 A, so set it to 15. What has actually happened is that the thermal protection has been removed from a motor that is running hot for a reason nobody established. This is the single most damaging habit in this chapter.

  • Diagnosing a contactor coil with Ohm's law

    Why

    Chapter 3's callout, met in the wild. An AC contactor coil draws a large inrush while the armature is open and the air gap is dominating its impedance, then settles to a fraction of that once it has pulled in. Comparing a measured current against the DC winding resistance gives you a number that means nothing. Compare against the manufacturer's figure.

  • Concluding 'bad motor' from a motor that hums and will not start

    Why

    The symptom is dramatic and the motor is the thing making the noise, so attention lands there. But a stopped motor that hums, draws heavily and does not turn is the classic single-phasing signature, and the fault is upstream at a fuse, a contact or a terminal. Clamping all three phases separates the two in under a minute.

  • Ignoring a small phase imbalance

    Why

    A few percent looks like measurement noise and gets rounded away. It is not noise — voltage imbalance produces several times as much current imbalance, and that current is heat in a winding whose insulation life roughly halves every 10 °C. A motor that fails "for no reason" a year later often has an imbalance in its history.

  • Replacing the contactor when the problem is its supply

    Why

    A chattering, buzzing contactor is a visibly faulty-looking component and swapping it is satisfying. If the cause is a control supply sagging under load, the new contactor chatters too — and its contacts are now being destroyed by the same arcing. Measure the coil voltage while it is held in and the machine is running.

Hands-on challenge

Scenario

Extract fan EF-7 — trips its overload every afternoon

Extract fan EF-7 has tripped its overload three times this week, always in the afternoon, never in the morning. Each time it resets and runs.

What you have established:

  • Motor nameplate: 400 V, 8.2 A FLA, 4-pole, 1440 rpm, service factor 1.15.
  • Overload: thermal, class 10, set to 8.2 A, not ambient-compensated.
  • Clamp readings on all three phases under normal running: 9.1 A, 9.1 A, 9.0 A.
  • Supply voltage at the motor terminals, under load: 398 / 399 / 397 V.
  • The fan's inlet filter was last changed four months ago, on a three-monthly schedule.
  • Panel temperature in the afternoon is noticeably higher than in the morning.

Write down: what the balanced currents rule out, what the current level tells you about the load, which two explanations the afternoon pattern is consistent with, how you would separate them, and what you would say to somebody who suggests setting the overload to 10 A.

Show how to approach it

Two of the given facts are the whole diagnosis, and one of them is not electrical. Work out which readings are consistent with which explanations.

  1. The currents are balanced. That eliminates single phasing, a lost phase, and supply imbalance in one measurement — the three explanations that would otherwise be top of the list.
  2. They are also above nameplate. 9.1 A against an 8.2 A plate is roughly 11 % over, sustained. That is a real thermal condition, so the overload is reporting accurately and this is not a nuisance trip.
  3. Balanced and high means the load, not the supply. More load means more slip means more current. Something is making this fan harder to turn than it used to be.
  4. The afternoon pattern has two candidate explanations and you can separate them: rising panel ambient acting on a non-compensated thermal overload, or rising process load. The filter-change record and the panel temperature distinguish them.
  5. The blocked filter is the mechanism to check first — it is free to look at, it explains a gradual current rise over weeks, and a fan working against a restriction is the commonest version of this fault.
  6. The overload setting is correct at 8.2 A and must stay there. Raising it to 10 A stops the trips and lets a motor run 11 % overloaded indefinitely.

Knowledge check

Five questions. Each is answerable by reasoning from readings rather than by recalling a specification, which is the point of the whole course.

Question 1 of 5

A running conveyor motor is hot and losing speed. You clamp all three phases and read 18 A, 17 A and 0 A against an 11 A nameplate. What has happened?