Industrial power quality: when the origin of a shutdown isn't where it seems

A residual-current device trips for no apparent reason. A drive locks up. A PLC restarts and the line stops. Maintenance checks the installation but finds no clear fault. When the incident repeats, it's common to replace equipment, readjust protection settings, or open a discussion with the utility — all without yet having enough data to explain what actually happened.

The problem is that an industrial electrical installation can no longer be analyzed from its internal components alone. The rise of frequency drives, PV inverters, uninterruptible power supplies, rectifiers, and other electronic loads has increased process sensitivity to voltage dips, overvoltages, harmonics, and high-frequency disturbances. These technologies bring control and flexibility, but they also change how the plant responds to the grid.

That's why, faced with an electrical shutdown, the useful question isn't just which device tripped, but what electrical phenomenon caused it to trip. Answering that requires looking at the whole system: supply, protection devices, loads, and the production process.

Grid or plant: a divide that doesn't always explain the problem

On the plant floor, the word "micro-cut" is often used to describe any brief stoppage. Behind that symptom, though, there could be a voltage dip, a short interruption, a transient overvoltage, or a rapid voltage change. The event can originate outside the installation — during a switching operation or a grid incident, for example — but also inside the plant itself, during the start-up of a high-power load, a welding operation, or a fault in another circuit.

There's also a third scenario that's especially hard to diagnose: an external disturbance that, on its own, might not have serious consequences, but that gets amplified when it interacts with existing leakage currents, cable length, power electronics, or poorly coordinated protection. In that case, pinning all the responsibility on the grid or on the installation leads to an incomplete diagnosis.

Spain's supply-quality framework, set out among other regulations by Royal Decree 1955/2000, establishes continuity and quality parameters. But for an industrial site, it's not enough to know whether the supply meets general requirements: it's also necessary to determine whether a specific event is compatible with the actual immunity of its equipment and with the plant's configuration.

When modern loads change how the installation behaves

Many electronic devices include filters to stop the high-frequency signals they generate from spreading through the installation. These filters produce small leakage currents during normal operation, and when many devices accumulate under the same protection device, the sum of those leakages can approach the trip threshold without any insulation fault existing at all.

It then only takes a transient leakage — from an overvoltage, a switching operation, or a piece of equipment starting up — to push that sum past the threshold and trip the protection. This explains why some trips seem random, and why simply swapping the RCD for a more immune one doesn't always solve the underlying problem: the cause is usually in the installation as a whole, not in a single component. Choosing cheaper equipment without considering its class and real implications is, in this area, a saving that tends to cost more in the end: what isn't paid for at purchase ends up being paid for in trips, shutdowns, and replacements.

Harmonics add another layer of complexity. Non-linear loads distort the current, and that distortion can carry over into the voltage, increasing heating in cables and transformers, losses, and equipment wear. Capacitor banks are especially sensitive to this, so they're worth reviewing whenever the installation has a lot of power electronics.

Measuring to distinguish origin, propagation, and impact

A one-off inspection can catch visible faults but will rarely explain a sporadic disturbance. Diagnosis starts by reconstructing the production incident: what time the line stopped, which protection device tripped, what alarms the equipment logged, which loads were switching, and how long the process took to recover. That timeline makes it possible to correlate the electrical event with its industrial consequences.

From there, the measurement campaign is designed. It's not just about placing an analyzer at the main panel — it's about choosing the points that let you test your working hypotheses. When it's necessary to separate an external disturbance from one generated inside the plant, synchronized multi-point measurements allow the phenomenon to be compared at the point of supply and at different levels of the installation, tracking its direction and propagation.

Class A analyzers, based on the measurement methods defined by IEC 61000-4-30, allow comparable recording of parameters such as frequency, voltage magnitude, flicker, temporary overvoltages, dips, interruptions, harmonics, and interharmonics. If interference or higher-frequency disturbances are suspected, the study needs to be completed with specific instrumentation, because measuring the wrong indicator can let the phenomenon go unnoticed.

The value isn't in accumulating records but in interpreting them alongside the process. A voltage dip can be brief and still block a particularly sensitive piece of equipment. Likewise, an external disturbance may not be enough to stop the plant on its own — until it coincides with high leakage currents, a poorly coordinated protection device, or a deficient grounding system. The root cause emerges from relating the electrical data to the plant's actual behavior.

From technical evidence to an industrial decision

When there's disagreement about the origin of an incident, simply stating that "the grid failed" isn't enough. What's needed is traceable data showing what happened, when it occurred, where it was detected, and how it propagated. An accredited measurement builds confidence in the technical competence involved, the equipment, its calibration, and the procedure applied, and it supports a discussion based on evidence with the utility, the engineering firm, the maintenance provider, or the equipment manufacturer.

This doesn't mean an accredited report automatically assigns responsibility to a third party. Its purpose is to replace impressions with technically defensible results. CIRCE holds an ENAC-recognized accreditation scope for electrical testing and carries out on-site tests and power quality campaigns aimed at identifying incidents, characterizing harmonics, analyzing unexpected trips, and proposing mitigation measures.

The right solution depends on the demonstrated cause: sometimes it means passing the results on to the utility, and other times it means acting on the installation itself. Power quality should therefore be built into a broader strategy of operational continuity and energy efficiency and consumption reduction, especially as electrification links industrial processes more closely with smart grids and with new generation and storage assets.

The economic impact of a disturbance isn't limited to the energy that goes unsupplied, either. An event lasting milliseconds can cause hours of downtime if it forces a line restart, a process clean-up, equipment recalibration, or a scrapped batch. That's why identifying the root cause isn't just an electrical maintenance issue — it's a decision about productivity, asset lifespan, and competitiveness.

At CIRCE, we support industry in designing measurement campaigns, analyzing incidents, and defining solutions tailored to each installation. Because knowing which protection device tripped is only the starting point. The real technical decision begins when we can explain why it tripped — and what needs to change so it doesn't happen again.