Power quality problems and how to find them

Some electrical problems are not failures of any single component but disturbances in the voltage or current itself. Computers reboot when the elevator starts, LED lights flicker, drives trip on overvoltage at the same time every morning, transformers run hot at half load, neutral conductors overheat, and capacitor banks fail repeatedly. These are power quality problems. Diagnosing them requires understanding the types of disturbance, where they come from, how to measure them with power quality analyzers, and which remedies work. This lesson covers voltage sags and swells, interruptions, transients, harmonics, flicker, imbalance and power factor, along with a practical approach to power quality investigations.

What power quality means

An ideal supply is a perfect sine wave at constant amplitude and frequency, balanced across three phases, with no interruptions. Real supplies deviate from this ideal, and loads also distort the current they draw, which in turn distorts the voltage. Standards such as IEEE 1159 (monitoring power quality), IEEE 519 (harmonic limits), CSA C61000 and IEC 61000 series and EN 50160 describe the types of disturbances and acceptable levels. A power quality problem exists when a disturbance causes equipment to malfunction, fail or wear prematurely. Studies consistently find that most power quality problems originate inside the customer's facility, from its own loads and wiring, rather than from the utility, although utility events such as faults on the distribution system cause many sags.

Voltage sags, swells and interruptions

A sag (dip) is a short reduction in RMS voltage, typically to between 10 % and 90 % of nominal for half a cycle to a minute. Sags are the most common disturbance affecting industrial equipment. They are caused by faults on the utility system or in the facility, often far away, and by starting large motors. Sensitive equipment such as drives, PLCs, contactors and computers can trip or drop out during sags lasting only a few cycles; a contactor coil may drop out when voltage falls below about 70 % for a cycle or two. A swell is a short increase in voltage, often caused by single-line-to-ground faults raising the voltage on unfaulted phases or by switching off large loads. Interruptions are complete losses of voltage, momentary (from utility reclosers clearing temporary faults) or sustained.

Remedies depend on the cause and duration: staggering or soft-starting large motors; ride-through settings in drives (allowing them to continue through short sags using DC bus energy); powering control circuits from uninterruptible power supplies, constant-voltage transformers or DC supplies; and specifying equipment that meets sag immunity standards such as SEMI F47, which requires semiconductor equipment to ride through sags to 50 % for 200 ms, 70 % for 0.5 s and 80 % for 1 s.

Transients

Transients are very short, high-magnitude disturbances lasting microseconds to milliseconds. Impulsive transients come from lightning and from switching inductive loads; oscillatory transients come from switching capacitor banks, which can briefly double the voltage. Transients damage insulation and electronic components, cause equipment to lock up or reset, and degrade surge protective devices over time. Remedies include surge protective devices (SPDs) in a coordinated scheme at the service entrance, distribution panels and sensitive equipment, good grounding and bonding to keep equipment at a common potential, and controlled switching or detuning reactors for capacitor banks. Because transients are so short, they require analyzers with fast sampling to capture.

Harmonics

Non-linear loads, such as VFDs, computers, LED drivers, battery chargers, UPS systems and EV chargers, draw current in pulses rather than smooth sine waves. Mathematically, these distorted currents consist of the fundamental frequency (60 Hz in North America) plus harmonics at integer multiples: 180 Hz (3rd), 300 Hz (5th), 420 Hz (7th) and so on. Total harmonic distortion (THD) measures the combined harmonic content relative to the fundamental:

Harmonic currents cause several problems. They add heating in transformers, conductors and motors beyond what the RMS current suggests, because losses increase with frequency; transformers serving non-linear loads may need derating or K-rated designs. Triplen harmonics (3rd, 9th, 15th) from single-phase non-linear loads add in the neutral of a three-phase four-wire system instead of canceling, so the neutral can carry more current than the phase conductors, up to about 1.7 times phase current in extreme cases, overheating neutrals that were sized for balanced linear loads. Harmonic currents flowing through system impedance distort the voltage, which can upset other equipment. And harmonics can resonate with power factor correction capacitors, amplifying currents and destroying capacitors.

IEEE 519 sets limits on current distortion at the point of common coupling with the utility, depending on the ratio of short-circuit current to load current, and limits on voltage distortion, which for systems up to 1 kV are 8 % total harmonic distortion and 5 % for any individual harmonic. Remedies include line reactors and DC chokes on drives, multi-pulse drives (12- or 18-pulse), active front-end drives, passive harmonic filters, active harmonic filters that inject canceling currents, oversized or separate neutrals, and K-rated or harmonic-mitigating transformers.

Flicker, imbalance and power factor

Flicker is the visible fluctuation of lighting caused by rapid voltage variations, from loads such as arc furnaces, welders, large motors starting frequently or fluctuating PV output. It is measured with flickermeters as short-term (Pst) and long-term (Plt) severity indices. LED lamps vary greatly in their sensitivity to voltage fluctuations, so flicker complaints sometimes increase after lighting retrofits. Voltage imbalance, discussed in Lesson 6 for motors, arises from unevenly distributed single-phase loads, high-resistance connections on one phase, or utility conditions. Power factor is the ratio of real power to apparent power; low displacement power factor from induction motors increases current and losses and may incur utility penalties, while distortion from harmonics also lowers true power factor. Power factor correction capacitors improve displacement power factor but must be applied carefully where harmonics are present, often with detuning reactors.

Investigating power quality problems

A structured investigation follows the troubleshooting method of Lesson 1:

• Define the problem precisely: which equipment malfunctions, how, when and how often.
• Inspect the installation: grounding and bonding, neutral sizing, loose connections, shared circuits between sensitive and disturbing loads.
• Measure: install a power quality analyzer meeting IEC 61000-4-30 Class A or S at the affected equipment and, if needed, at the service, recording voltage and current on all phases and neutral for at least a full business cycle, typically a week, with event triggers set to capture sags, swells and transients.
• Correlate: compare recorded events with the times of equipment problems; check whether disturbances originate inside or outside (for example, a sag accompanied by a current increase on the monitored circuit originates downstream; one with no current change likely originates upstream).
• Remedy and verify: apply the most targeted, cost-effective fix and monitor again to confirm improvement.

Analyzers must be installed safely, by qualified people with appropriate PPE, using voltage leads with fused connections and current probes of suitable range. Recording neutral-to-ground voltage alongside phase voltages helps identify wiring and grounding problems.

Grounding and wiring problems that look like power quality issues

Many complaints attributed to power quality are actually wiring and grounding problems. Loose neutral connections cause voltage fluctuations and imbalance; shared neutrals between circuits carry unexpected current; improper neutral-to-ground bonds create objectionable current and noise on grounding conductors; and isolated ground receptacles installed incorrectly can create ground loops. Data and communication equipment is particularly sensitive to differences in ground potential between devices connected by data cables. Measuring neutral-to-ground voltage at sensitive loads is a useful first check: values above a few volts under load suggest neutral problems, undersized neutrals or excessive neutral current. Correcting the wiring often solves the problem without any specialized power conditioning equipment.

Surge protection in practice

Surge protective devices are installed in a coordinated scheme: a Type 1 or Type 2 device at the service entrance diverts large external surges, Type 2 devices at distribution and branch panels limit residual and internally generated surges, and point-of-use devices protect individual sensitive equipment. SPDs are most effective with short, straight connecting leads, because lead inductance adds voltage during fast surges; every 30 cm of lead can add tens to hundreds of volts to the let-through voltage. SPDs wear out as they absorb surges, so their status indicators should be checked during maintenance, and failed units replaced. Grounding and bonding must be sound for SPDs to work, because they divert surge current to ground.

Monitoring permanently

Facilities with critical loads, such as data centers, hospitals and manufacturing plants, increasingly install permanent power quality monitoring at the service and key distribution points. Permanent monitors capture events as they happen, so there is no need to wait for a problem to recur after installing a portable analyzer, and they build a long-term record of supply performance useful in discussions with the utility. Software can correlate events across many meters, showing where a disturbance originated and how it propagated through the system.

Working with the utility

When measurements show that disturbances originate on the utility supply, such as frequent sags from faults on the distribution system or sustained voltage outside the permitted range, the facility should share the data with the utility. Utilities can sometimes identify the cause, such as a failing component, tree contact or a nearby customer's equipment, and correct it. Utility voltage ranges are defined in standards and supply agreements: in Canada, CSA C235 sets preferred voltage ranges, and in the United States, ANSI C84.1 does. Data recorded with a Class A analyzer carries more weight in such discussions.

Recording a baseline

As with other troubleshooting, a baseline helps. Recording power quality at the service and key panels when a facility is commissioned, or before major new loads are added, provides reference values for voltage, harmonics, imbalance and flicker. When complaints arise later, new measurements can be compared with the baseline to see what has changed, such as harmonic distortion increasing after a large number of drives or EV chargers were installed, and to plan corrective measures before problems become severe.

Power quality work also benefits from teamwork between electricians, engineers and equipment suppliers. Electricians see symptoms first and install the instruments; engineers analyze data and design remedies; suppliers know how their equipment responds to disturbances and which settings improve ride-through. Bringing them together early shortens investigations.