Power quality problems almost never announce themselves as power quality problems.
They arrive as a drive that keeps tripping, a UPS that alarms at night, a panel running warmer than it should, a control system that resets for no traceable reason.
So they get treated as what they resemble: equipment faults.
Components are replaced, boards are serviced, and the same issue returns a few weeks later with a different serial number attached to it.
The scale of that misdiagnosis is well documented.
A pan-European survey covering 16 industrial sectors across the EU-25 found that losses related to poor power quality exceeded €150 billion, with industry accounting for more than 90% of the total. A CIRED paper by Schneider Electric engineers, drawing on the same body of work, puts it in terms a site can apply directly: in industry, the cost of poor power quality can reach 4% of annual turnover, and an estimated 30–40% of all business downtime is power quality related.
Those aren’t small numbers, and very little of that cost is ever recorded as a power quality problem. It’s recorded as maintenance, as scrap, as unplanned downtime, as a drive that didn’t last.
What power quality means in practical terms
Power quality describes how closely the supply matches the voltage, frequency and waveform the connected equipment expects.
When supply conditions sit within expected limits, equipment behaves as designed.
When they deviate, the effects appear across motors, drives, control systems, UPS systems, transformers, switchgear and the wider LV distribution, often in places that look unrelated to each other.
The disturbances that matter most are voltage dips and sags, swells, interruptions, transients, harmonic distortion, flicker, voltage unbalance, poor power factor and frequency variation.
Of those, one category dominates the losses: the pan-European survey found that voltage sags and short interruptions were responsible for around 50% of power quality related losses, affecting electronic equipment in industrial and service sectors most heavily.
Some of these events last milliseconds. Others develop over months, or only appear when a particular load is running.
That combination of brief, intermittent and load-dependent behaviour is precisely what a periodic inspection is least able to catch.
Why a dip you never saw can stop your plant
This is worth setting out with real figures, because it’s the mechanism behind a large share of “unexplained” stoppages.
A peer-reviewed IET review of the GB distribution system compiled equipment sensitivity data from IEEE research. It found that AC-coil motor contactors can drop out at between 30% and 75% of nominal voltage, in as little as 10 to 80 milliseconds. Adjustable speed drives are affected in the 60–85% range over 10–170 ms. PLCs are more variable again, and factory automation robots more sensitive still.
The review’s conclusion is blunt: a balanced three-phase sag at 50% of nominal lasting 500 ms at a sensitive customer’s terminals will cause major equipment trip or malfunction and will likely disrupt the production process.
An event of that depth and duration is finished before anyone in the building registers a flicker. The contactor drops out, the line stops, and the subsequent investigation finds a contactor that tests perfectly, because it is perfect. It operated correctly at a voltage nobody was measuring.
There is also a reporting gap that keeps this invisible. GB supply reliability is reported through CI and CML indices, which only count interruptions longer than three minutes. Every dip and momentary interruption falls below that threshold. The same IET review found that short interruptions in GB rose 16% between 2010 and 2015 even as the headline reliability metrics improved, and noted that the true figure may be higher, because recording practice for short interruptions isn’t fully developed.
So your DNO’s reliability performance can be genuinely good while the events that actually trip your plant become more frequent underneath it.
Symptoms, and what usually sits behind them
| What you’re seeing | Likely power quality contributor |
|---|---|
| Nuisance tripping, repeated breaker operation | Harmonic distortion, transients, poor discrimination |
| Drive and VFD trips | Voltage dips, voltage instability, switching events |
| PLC and control faults, unexplained resets | Short-duration voltage dips, transients |
| Contactor dropout | Dips below the coil hold-in threshold, often under 100 ms |
| Overheating motors, cables or transformers | Harmonic currents, voltage unbalance, sustained overload |
| UPS alarms or transfers to bypass | Unstable input voltage, frequency variation, upstream dips |
| Shortened equipment life, repeated failures of the same part | Cumulative thermal and electrical stress from distortion or unbalance |
| Flickering lights, process instability | Voltage fluctuation, flicker from cyclic loads |
| Power factor charges or capacity concerns | Poor power factor, reactive demand, correction equipment failed |
| Problems that began after a commissioning | New non-linear load, PV, BESS or EV charging changing the load profile |
Where do the power quality disturbances come from?
Disturbances originate both inside and outside the site boundary, and the split matters commercially.
Internally: non-linear loads, variable speed drives, large motor starts, switching events, poor load balance, transformer loading, and harmonics generated by site equipment. In industry, motors are often 60–80% of the load, which is why voltage and current unbalance deserve routine attention rather than occasional investigation.
Externally: upstream supply issues and network events. Notably, the IET review found that although only around 19% of GB distribution faults originate in the HV network, those faults account for 70% of customer interruptions and 60% of customer minutes lost, and that around 80% of overhead line incidents are transient in nature.
There’s a second-order effect there that’s rarely discussed. Network operators have improved their headline reliability figures partly by replacing spur fuses with auto-sectionalisers. That reduces long interruptions, but it means more customers on the main and parallel feeders now experience short interruptions and voltage sags instead. The regulated metric improved. For some sites, the disturbances that trip equipment did not.
The pressure is also increasing from inside. A site that ran reliably for fifteen years can start behaving differently once new automation, solar PV, battery storage or EV charging is added to a distribution system that was designed for none of it.
If faults began after a commissioning, that timing is the most useful clue you have.
What poor power quality costs in capacity you’ve already paid for
One consequence gets consistently overlooked: harmonic distortion and reactive power don’t just stress equipment, they consume rated capacity.
In a power quality and load study carried out on a client site over December 2025 and January 2026, current harmonic distortion exceeded 20%, generated by internal loads. Total demand distortion ran between roughly 20% and 35%, against the 15% guideline in IEEE 519. The consequence, once derating for harmonic pollution and reactive power was applied: the effective available capacity of a 1,000 kVA distribution transformer was reduced to 42%.
That site had already paid for 1,000 kVA. It could usefully draw on less than half.
The same measurement period captured two voltage sags, on 23 December and 3 January, that forced the UPS to transfer to bypass. The UPS protected what was behind it. Everything else on site was exposed.
For any team facing a capacity constraint and a quotation for a larger transformer or an upgraded supply, that’s a question worth asking before committing the capital: is the capacity genuinely insufficient, or is a substantial share of it being consumed by distortion?
When to monitor power quality
The trigger points are reasonably clear.
Power quality monitoring is worth doing when:
- Electrical issues are recurring, hard to explain
- Affecting operational reliability, particularly with repeated nuisance tripping, unexplained equipment resets, UPS instability, control faults, overheating, process disruption
- Faults that return after maintenance has already been carried out.
- Before a capacity decision, and after commissioning significant new load.
What matters is capturing enough evidence to separate an isolated equipment fault from a system condition: disturbance records, voltage and current data, event logs and enough history to establish whether there’s a pattern. That evidence is what turns a recurring problem into a specification for corrective action: filtering, power factor correction, load balancing, protection review, or a conversation with the DNO backed by data.
Acteniq delivers this through power quality monitoring, event capture and analysis, technical interpretation, and the corrective engineering that follows, then verifies the outcome against the original baseline. We look at the system, not the device.
Sources cited in this piece
- Targosz & Manson, Pan-European Power Quality Survey, CIRED 2007 / Leonardo Energy, European Power Quality Survey Report, 2008, as reported in Vegunta et al., IET Generation, Transmission & Distribution, 2019
- Ignatova, Lafort & Bilic (Schneider Electric), Power Quality Management Methodology, CIRED 2015, paper 0726: cost as share of turnover, share of business downtime, motor share of industrial load
- Vegunta, Watts, Milanović, Djokic & Higginson, IET Generation, Transmission & Distribution, 2019: equipment sag tolerance (Table 1, rows sourced to Djokic et al., IEEE Transactions on Power Delivery, 2004–2005; EPRI 2002), CI/CML threshold, 16% rise in short interruptions, HV fault contribution, auto-sectionaliser effect
- IEEE 519: harmonic distortion guideline
- Power Quality and Load Research, Fortop Automation & Energy Control for Acteniq, anonymised, measurement period 12 December 2025 – 4 January 2026
