Why AC Contactors Fail: Chatter, Welded Contacts, Burnt Coils

Three Symptoms, Three Different Fingerprints

An AC contactor is a simple device: an electromagnet, a return spring and three sets of silver contacts. That simplicity is why it fails in a few predictable ways, and each failure mode leaves a different fingerprint on the hardware. A chattering contactor points at the coil circuit or the magnetic system. Welded contacts point at the load and the duty the contacts are being asked to perform. A burnt coil points at the control voltage or a mechanical obstruction.

Reading these fingerprints matters because in our experience the majority of contactors returned as "defective" were never defective at all — they were applied to a duty their rating was never designed for, or installed with a fault somewhere else in the panel. A five-minute diagnosis usually separates the two, and it starts by knowing what each symptom looks like.

Chatter and Buzzing: the Magnet Cannot Seat

When an AC contactor closes, the armature must land flush against the magnetic core. If it does not, the magnetic circuit has an air gap, the coil impedance stays low, and the coil keeps drawing inrush current while the armature vibrates — the familiar buzz or chatter. Four things cause it:

1. Control voltage below the pull-in threshold. Long, thin control wires and supplies shared with large starting loads can drop the coil terminal voltage enough that the armature never seats firmly. Measure the voltage at the coil terminals, not at the supply, while the contactor is energised.

2. Dirty or corroded pole faces. Dust in the panel, or light rust after storage in a humid port, keeps the armature from seating even at correct voltage. Cleaning the pole faces with a dry cloth restores quiet operation.

3. A cracked shading ring. The shading ring is what holds the magnetic force above zero at the current zero-crossing; when it cracks, the contactor buzzes loudly at twice the supply frequency even though everything else is healthy. This one needs a replacement contactor.

4. Supply frequency mismatch. A 50 Hz coil fed from a 60 Hz system pulls in but holds weakly; the reverse overheats the coil. If you are importing into a 60 Hz market, confirm the coil frequency before ordering rather than after.

Chatter is not just noise. Every bounce of the armature is a make-and-break at the main contacts, and each bounce strikes a small arc. A contactor that chatters for weeks will often end its life with welded contacts — so treat the noise as an early warning, not an annoyance.

Welded Contacts: Almost Always an AC-4 Duty Problem

The single most misunderstood figure on a contactor datasheet is the rated working current, because it changes with the utilization category. Under AC-3 — the normal running duty of a motor that starts once and then runs — the contacts only interrupt the modest running current. Under AC-4 — plugging, jogging and frequent reversing — the contacts must make and break the full motor starting current on every single operation.

Look at what that does to a real frame. Our CJX2-25 is rated 25 A under AC-3, but only 8.5 A under AC-4. On the larger CJX2-115, the electrical life is 2,000,000 operations at AC-3 and about 150,000 at AC-4 — the same device, roughly thirteen times shorter contact life. If an operator jogs a hoist into position, inches a mixer forward and back, or reverses a conveyor often, the duty is AC-4 and a contactor chosen by its AC-3 number alone will weld its contacts far sooner than expected.

Welding shows on the contact face: a raised spike on one contact and a matching crater on the other, where the metal melted on make, flowed, and solidified on break. Uniform grey wear across the whole face, by contrast, is normal erosion — it means the contactor simply reached the end of a long life. When a welded contactor is found with a motor that kept running after an emergency stop was pressed, the sequence is worth reconstructing: what was the load doing at that moment?

There is a second way contacts weld that has nothing to do with motors: closing onto a discharged capacitor. The inrush from a shunt capacitor stage can reach tens of times the rated current, far beyond what any standard contactor is built to make. That failure mode is common enough in power factor correction panels that it deserves its own discussion below.

Burnt Coils: Overheating That Builds Over Weeks

A burnt coil is the control-circuit fingerprint. The coil is a solenoid designed to run cool once the armature is seated, because a seated armature closes the magnetic path and drops the coil current to a holding fraction of the pull-in value. Anything that prevents that seating cooks the coil:

1. Sustained undervoltage with chatter — the coil sits near pull-in current for hours. This is the same root cause as the buzzing in the section above, and the coil burn is just the end result.

2. A mechanically stuck armature. Swarf, a dropped screw, or a bent DIN rail clip can hold the armature a few millimetres open. The coil then runs at full inrush current and burns out in minutes, usually during commissioning — which is why a "new contactor, coil dead on arrival" report deserves a look inside the housing before a claim.

3. Ambient heat. CJX2 contactors are rated for operation between -5 and +40 °C. A contactor mounted above a VFD heatsink or inside a sealed outdoor cabinet in a tropical climate can sit permanently above that window, and the insulation ages quietly until it fails.

The burn pattern helps distinguish the cause. Uniform darkening of the winding points to chronic overheating over weeks. A local melted spot points to shorted turns from a voltage event. Either way, when replacing the coil, also solve the condition that killed it, or the new coil follows the old one.

Overheating Terminals: the Cheapest Fix in the Panel

Not every failure is inside the contactor. Loose terminal screws — loosened further by transport vibration and thermal cycling — create a high-resistance joint that heats the whole terminal block. Undersized cable does the same: each CJX2 frame specifies a wiring ability, from 2.5 mm² on the CJX2-18 up to 35 mm² on the CJX2-95, and a conductor smaller than the frame expects becomes the hotspot instead of the contact.

The inspection is quick: look for discoloured insulation or melted shrouding around the load terminals, then re-torque every screw on the frame to specification. A thermal camera makes this a ten-second check across the whole panel, and it is the single highest-value maintenance action on a contactor installation, because it catches the fault before the contact ever sees it.

Capacitor Panels Need a Contactor Built for Inrush

If welded contacts appear only inside automatic power factor correction cabinets, the contactor was probably never the problem — the application was. A standard AC-3 contactor is not designed to close onto a discharged shunt capacitor, and no amount of upsizing fully solves it, because the inrush waveform is what the contacts cannot survive.

That is what the CJ19 capacitor-duty series is for. Its pre-charge contacts close first through damping resistors, limiting the closing inrush to about 20 times rated current, and the main contacts follow once the surge has passed. The CJ19-43, for example, is rated 29 A under AC-6b and switches shunt capacitor stages up to 25 kvar at 380/400 V, with an electrical life of 1,000,000 operations at switching rates up to 1,200 per hour. Panels that switch capacitors with ordinary contactors should be converted to a capacitor-duty frame before the next weld.

A Five-Minute Diagnostic Checklist

1. De-energise the panel, lock it off, and discharge any capacitor stages before touching anything.

2. Read the contact faces. Crater-and-spike patterns point to duty problems; uniform grey erosion points to an honest end of service life.

3. Measure the coil voltage at the coil terminals while energised, and compare it against the printed coil rating.

4. Inspect the pole faces and shading rings for dirt, rust or cracks.

5. Re-torque the main terminal screws and check the conductor size against the frame's wiring ability.

6. Ask what the load actually does every day. Constant jogging, reversing or capacitor switching changes which contactor should be in that slot.

7. If the contactor pairs with a thermal overload relay, confirm the relay's setting range still matches the motor full-load current — a common secondary fault after a contactor event, covered in our thermal overload relay selection guide.

When the diagnosis does call for a replacement, match the new frame to the motor power rather than to the old part number alone, since installed bases often contain frames that were undersized from day one. Our AC contactor catalogue lists the AC-3 and AC-4 ratings, coil options and auxiliary contact layouts for every frame from 9 A to 150 A, and the contactor selection guide walks through that matching process step by step.

Finally, when sending an enquiry, include two details alongside the quantity: what the contactor will switch, and how many times per hour it switches. Those two facts let us confirm the utilization category and frame size before the order is placed — and they are the same two facts that would have prevented most of the failures described above.