--- title: "MCCB keeps tripping: how to find out why" description: "Time the trip first — it halves the search. Then work through overload, short circuit, earth fault and the causes that are not faults at all." date: "2026-04-12" author: "Divakar B" source: "https://energycalchq.com/blog/mccb-keeps-tripping" license: "© 2026 Divakar B. Quote with attribution to https://energycalchq.com/blog/mccb-keeps-tripping" --- A breaker that has tripped tells you almost nothing by itself. The handle is in the same place whether the cause was a slow overload building over twenty minutes or a dead short that cleared in eight milliseconds, and those two need completely different things done about them. So before touching anything, get one piece of information: **how long was it running before it went?** That single answer eliminates most of the list below. ## Time the trip, and read the curve A thermal-magnetic MCCB has two independent trip elements with completely different jobs. ![Time current curve of a thermal magnetic MCCB showing the inverse-time thermal band and the vertical magnetic trip at ten times rating](/blog/mccb-trip-curve.svg "Two elements, two failure modes. The one that operated tells you which half of the problem you have.") The **thermal** element is a bimetallic strip responding to heat. It takes minutes at a modest overload and seconds at several times rating. If the load ran normally for a while and then tripped, this is what operated, and you are looking for an overload — real or apparent. The **magnetic** element is a solenoid with a fixed or adjustable pickup, often around 10 × In. It operates in under 20 ms. If the trip happened at the instant something switched on, this is what operated, and you are looking for a short circuit or an inrush. Many MCCBs have a trip indicator or a separate earth-fault flag. Look before you reset — resetting destroys the evidence. ## If it took minutes: the overload family **The load has genuinely grown.** The commonest cause and the least interesting. Clamp each phase under normal running and compare against the setting. Plant accumulates load — a compressor added, a heater fitted, a motor replaced with a larger one — and nobody revisits the breaker. **The setting is wrong for the cable, not the load.** On a motor circuit the overload relay in the starter protects the motor; the breaker upstream is doing short-circuit duty. Confusing the two is how a breaker ends up set at the motor's full load current, tripping on every long run-up. Check what each device is actually protecting with the [breaker sizing calculator](/tools/breaker-sizing). **The panel is hot.** Thermal elements are calibrated at a reference ambient — 30 °C for MCBs to IEC 60898, 40 °C for MCCBs to IEC 60947-2 — and derate above it. A 100 A MCCB in a 55 °C panel is closer to an 85 A device. The tell is seasonal: it trips in May and behaves in December. Fix the ventilation, not the breaker. **Phase imbalance.** A three-phase breaker trips on whichever pole sees the excess. Single-phase loads unevenly distributed across phases can put one pole 20 % above the others while the total looks fine. Clamp all three. **Harmonics, on a device that reads true RMS.** Distorted current has a higher RMS value than its fundamental, and a plant full of drives can run a breaker above its rating while a cheap clamp meter reading average-responding says otherwise. The [harmonic loss calculator](/tools/thd-losses) converts a measured spectrum into the real RMS current. **A loose connection.** A terminal that is not tight runs hot, and that heat conducts into the breaker's thermal element. The breaker trips because it believes it is overloaded, and it half is. Thermal-image the panel at load, or at minimum check terminal temperatures by hand after a run. This one also gets worse over time, which makes it look like a slowly growing load. ## If it went instantly: the short-circuit family **A genuine fault.** Megger the circuit with the load disconnected before assuming anything else. A cable damaged by a screw, a motor with a winding failure, water in a junction box. **Inrush the breaker was never meant to pass.** Transformers, capacitor banks, LED driver arrays and DOL motors all draw a surge far above running current for a few cycles. If the trip happens only at switch-on and never during running, this is it. The fix is the curve, not the rating: a C-curve device trips magnetically at 5–10 × In, a D-curve at 10–20 ×. Motors usually want D, or better, an MPCB whose magnetic element sits around 13 × In with a thermal element set to the motor. **Capacitor switching.** An APFC step energising onto an already-charged capacitor draws an enormous transient. If the trip correlates with the bank switching rather than with production, look at the discharge resistors and the contactor type — capacitor duty contactors exist precisely for this. **Cumulative starts.** A motor started repeatedly does not let the thermal element cool between attempts. Each start is survivable and the fourth in five minutes is not. Class 10 relays trip in 10 s at 6 × setting; frequent starting needs class 20 or 30. ## If the earth-fault flag is up That is a different fault entirely and none of the above applies. Insulation breakdown, moisture, a damaged cable, or a VFD leaking common-mode current to earth through its filter capacitors — the last one is a common cause of apparently random trips on drive circuits, and it needs a Type B residual device, not a smaller setting. ## What not to do **Do not fit a larger breaker.** If the cable was sized for the original device, a larger one no longer protects it — the cable can then sit above its rating indefinitely with nothing tripping. `Ib ≤ In ≤ Iz` is the rule, and the Iz is the cable's capacity *after* derating, not the table figure. **Do not raise the setting to make it stop.** Same problem, no paperwork. **Do not reset repeatedly into a fault.** Each operation on a fault erodes the contacts, and a breaker that has interrupted near its Icu may no longer be fit for service — which is what the separate Ics rating exists to describe. ## Quick reference | What you saw | Element | Look at first | |---|---|---| | Ran fine for 20 min, then went | Thermal | Load growth, panel ambient, loose terminal | | Trips only in summer | Thermal | Ambient derating — 100 A at 55 °C is nearer 85 A | | Went the instant something started | Magnetic | Inrush, wrong trip curve, DOL motor | | Went with a bang or a smell | Magnetic | Genuine fault — do not reset, inspect | | Trips on the fourth start in five minutes | Thermal | Cumulative starting, relay class too low | | Random, on a drive circuit | Earth fault | Common-mode leakage — needs a Type B RCD | | Gets steadily worse over months | Thermal | Loose connection heating the element | | One phase warm, others cool | Thermal | Imbalance across single-phase loads | The last two are the ones people chase hardest, because neither looks like a fault on a meter. A joint that has loosened conducts its own heat into the breaker, so the device trips believing it is overloaded — and it half is. ## A sequence that works 1. **Read the indicator** before resetting. Thermal, magnetic or earth fault. 2. **Time it.** Minutes, or instant? 3. **Clamp all three phases** under normal running, and compare with the setting and with each other. 4. **Take the panel's internal temperature**, not the room's. 5. **Check terminal torque** on that breaker and its cable. 6. **Correlate with events** — does it coincide with a motor start, a capacitor step, a shift change? 7. **Megger the circuit** if it is instantaneous and no inrush explains it. 8. Only then look at the device selection: rating, curve, and whether the cable supports the change. If step 3 shows the load genuinely exceeds the design, the answer is a bigger cable *and* a bigger breaker, in that order — and everything downstream needs rechecking, including [fault level](/tools/short-circuit-current) and the [busbar it lands on](/tools/busbar-sizing). ## When to stop and call someone A trip you cannot explain after the sequence above is a reason to stop resetting. So is any trip accompanied by a bang, a smell, or discoloured terminals — that is arc damage, and the breaker and its terminations need inspecting before the circuit is re-energised. Repeated earth-fault trips on a circuit people can touch are not something to live with while you investigate.