Contact wear is the one failure mode that every contactor is guaranteed to reach eventually, and the one that most maintenance programs still diagnose by guesswork. The contacts are sacrificial by design — every operation transfers a little silver, every high-current start erodes the facing — so the real question is never “are the contacts worn?” but “how much, how fast, and is it time to replace the contacts, or the whole device?”
This guide gives the four measurable indicators — contact resistance, tip thickness, overtravel, and temperature rise — with the instruments, the threshold values, and the judgment lines between “replace contacts” and “replace contactor”. The material science behind why certain contacts erode faster is covered separately in the AgSnO2 vs AgNi vs AgCdO contact material comparison; here we stay on measurement.
| Indicator | Instrument | Healthy value (typical IEC frame) | Replace / investigate |
|---|---|---|---|
| Contact resistance | 4-wire micro-ohmmeter (DLRO), 100 A preferred | 50–500 µΩ depending on frame size (small frames lower, large frames higher) | > 2× the commissioning baseline, or > 1 mΩ on a frame measured in the hundreds of µΩ |
| Tip thickness remaining | Depth gauge / caliper on eroded pole | Full silver tip intact | < 1/3 of original tip thickness, or any base-metal exposure |
| Overtravel (wipe) | Feeler gauge / depth rod on manual actuation | Per datasheet (typically 2–4 mm on larger frames) | Near zero — contact springs bottoming out; contact about to fail open |
| Pole temperature rise | IR camera under ≥ 50% load, or thermocouple | Pole-to-pole ΔT < 5–10 K | One pole > 10 K hotter than its neighbors |
Any single indicator past its limit is grounds for replacement; two together means the contactor has already been running degraded and the surrounding checks (coil, mechanism, arc chutes) need to happen before it goes back into service.
Contact resistance is the fastest field check, but it is also the most abused. A normal two-wire multimeter on the milliohm range reads its own lead resistance — which is the same order of magnitude as the contact you are measuring. The measurement is worthless unless the test current bypasses the leads: four-wire (Kelvin) measurement, with the voltage sense points on the contactor terminals themselves.
Field procedure:
The dry-circuit trap: silver oxide and silver sulfide films are semiconductors. At high test current the film frits (breaks down) and the reading drops; at milliamp-level currents the film stays intact and the same contact can read 10–100× higher. A contact that reads 800 µΩ at 100 A and 40 mΩ at 10 mA is not necessarily bad — but a PLC-switched signal path through that same contact will misbehave. Measure at the current class your load actually represents, and compare like with like.
| Reading (100 A test, per pole) | Interpretation | Action |
|---|---|---|
| Within ±20% of commissioning baseline | Normal film and seating | No action; log the value |
| 1.5–2× baseline, all poles equally | Film growth from environment (sulfidation, dust) | Schedule cleaning inspection; check enclosure filtering |
| One pole > 2× baseline, others normal | Localized erosion, contamination, or terminal loosening on that pole | Open and inspect that pole; check terminal torque first — it is the cheap cause |
| Reading unstable / drifting during test | Loose contact, broken spring, or severely eroded seating | Do not return to service before inspection |
Resistance trends late; erosion is visible early. Silver contact tips on IEC frames are typically 1–3 mm thick depending on frame size, brazed or riveted to a copper backing. Erosion concentrates on one or two poles in three-phase devices because the arc does not share equally between breaks.
What to look for when the contacts are accessible (major overhaul, or after a suspected fault):
Overtravel (also called wipe or follow-through) is how far the moving contact travels after touching the stationary one. It exists so that contact spring pressure keeps the pair seated as the tip erodes. Every millimeter of tip erosion consumes overtravel; when the wipe reaches zero, the contacts touch with no spring force — high resistance, bouncing, welding on the next make.
Field check on larger frames (with the contactor isolated):
Overtravel is the leading indicator of the group: it declines linearly with erosion while resistance stays flat until late-stage failure. Programs that track only resistance consistently get surprised by contact failures.
With the panel live and carrying at least half its normal load, an IR scan of the contactor tells you which pole is working hardest. A pole running more than about 10 K above its neighbors is developing a hot spot — typically the terminal connection (loose lug, wrong torque) or an eroded contact interface raising I²R loss at that pole.
| IR finding | Most likely cause | First action |
|---|---|---|
| Hot spot at the cable lug, not the contactor body | Terminal torque wrong or lug sized wrong | Re-torque to spec (cool), re-scan |
| Whole pole warm, symmetric line/load sides | Contact interface resistance rising | Confirm with 4-wire resistance on next shutdown |
| Coil area hot | Sealed coil current higher than expected — check coil voltage and mechanical friction | See the contactor coil test procedure |
| Everything uniformly hot | Load above rating, or enclosure thermal problem | Check current with clamp meter; review contactor derating for the enclosure |
Spare contact sets exist for the larger frames precisely because tip replacement is legitimate maintenance — but only when the rest of the device is sound. The decision line:
| Condition found | Contacts only | Whole contactor |
|---|---|---|
| Erosion past 2/3 tip thickness, mechanism and coil healthy, no fault history | ✔ Replace contact set; verify overtravel after reassembly | — |
| Contacts welded after a fault event | — | ✔ The mechanism, coil and chutes saw the same current |
| Any base metal exposed on a tip | Only if the other poles are virgin and the event was minor | ✔ Safer default on frames where contacts are riveted |
| Housing / arc chute carbon tracking or cracking | — | ✔ Chute condition is not field-restorable |
| Small frames (≤ 32 A, economy class) | — | ✔ Contact set cost vs device cost does not justify the labor |
The last row is the one that saves the most debate: on compact frames, the contact set is a third of the device cost and the labor exceeds the saving. Contact replacement is an economic decision above roughly 40 A frames, and a false economy below.
Thresholds belong in the maintenance record, not in memory. For each contactor in the plan, log at commissioning: contact resistance per pole at a fixed test current, overtravel (if measurable), and a photo of the contact faces. Then set:
AC-4 applications (plugging, jogging, reversing) erode contacts 10–20× faster than AC-3; the same inspection interval that is generous on a pump starter is negligent on a hoist. The category definitions that set this ratio are in the AC-1 to AC-4 guide.
| Mistake | Consequence | Correct practice |
|---|---|---|
| Measuring contact resistance with a 2-wire multimeter | Reading dominated by lead resistance; meaningless trend | 4-wire DLRO with sense points on the terminals, fixed test current |
| Filing or sanding silver contacts “to clean” them | Removes tip material; new sharp edges increase bounce; shortened life | Leave the film; track resistance instead |
| Replacing only the welded pole’s contact after a fault | Asymmetric erosion and resistance; the unwelded poles carry the same fault history | Replace all poles as a set; audit the whole device |
| Ignoring overtravel because it is not on the checklist | Zero-wipe failure arrives without any resistance warning | Add wipe to the inspection sheet for frames ≥ 40 A |
| Trending resistance across different test currents | Film fritting makes 10 A and 100 A readings incomparable | Fix the test current; record it with each reading |
Contact wear is measurable long before it fails: resistance for the electrical trend, thickness and overtravel for the mechanical trend, IR for the live cross-check. Track all four with fixed instruments and baselines, replace tips as a set on large frames while the mechanism is still healthy, and treat any welded or base-metal-exposed contact as a whole-device event rather than a consumable swap. For the erosion physics behind the patterns above, see the contact material comparison; for the internal construction these tips mount on, see inside an AC contactor; and for replacement contact sets across CJX2/CJX4 and CJ35 frames, browse the contactor contacts range or the wider AC contactor category.
Baseline annually for AC-3 duty, every 6 months for AC-4 or dusty environments, and physically (contacts out of the device) every 100,000 operations or per the duty-class maintenance table. Electrical resistance checks can be more frequent since they are non-invasive.
Small frames typically measure tens to low hundreds of micro-ohms; large 200 A+ frames run into the hundreds of µΩ. The absolute number matters less than stability against your own commissioning baseline at the same test current.
No. Abrading removes silver tip material and creates sharp edges that bounce on make, accelerating erosion. A dark oxide film with stable resistance is normal; file nothing, replace by measurement.
After any welded-contact fault event, when arc chutes are cracked or carbon-tracked, when base metal is exposed on the tips, and on small frames where contact-set cost plus labor exceeds device cost.
Three-phase arcs do not distribute symmetrically between breaks, and make-bounce concentrates transfer on specific pole pairs. This is why wear limits apply per-pole and why sets are replaced together.
Indirectly. Rising contact resistance adds heat into the pole circuit and, in severe single-pole erosion cases, contributes to current imbalance that the differential-amplified thermal relay may respond to. Severe wear should be corrected before it reaches the protection chain described in the thermal overload relay guide.
