.png)
Industrial Power Quality Monitoring for Troubleshooting Voltage Events and Harmonics
A variable speed drive trips at 2:17 am, a moulding line stops, and by the time the maintenance team arrives everything looks normal. No blown fuse. No clear fault. No obvious supply failure. This is where guesswork becomes expensive.
Power quality problems often appear as short events, repeated stress, or a pattern that only shows up when certain loads run together. A loose connection, voltage dip, high harmonic current, poor power factor, undersized neutral, capacitor switching event or utility disturbance can all create similar symptoms at the equipment level.
The practical answer is a measurement-led investigation. Capture what happened close to the affected equipment, compare it with upstream data, then decide whether the issue came from the supply, the installation or a particular load.
Start with the symptom, then choose the measurement point
The best investigation does not begin at the main switchboard by default. It begins with the symptom.
If one drive trips, measure near that drive. If a whole production area drops out, measure at the distribution board feeding that area. If multiple buildings report problems at the same time, include the main incoming supply.
That first measurement point matters because voltage disturbances can look very different at different points in the installation. A voltage dip at the main incomer may worsen at the end of a long submain. A harmonic problem may only become obvious on the bus where several non-linear loads operate together. A poor connection may create local voltage unbalance that upstream instruments barely see.
A useful first pass records:
Phase-to-phase and phase-to-neutral voltage
Current on each phase and neutral where relevant
Frequency
Real power, reactive power and apparent power
Power factor and displacement power factor where available
Voltage dips, swells, interruptions and transients
Harmonic voltage and harmonic current
Total harmonic distortion
Demand trends over time
For industrial power quality monitoring, the goal is not to collect data for its own sake. The goal is to connect electrical behaviour with plant behaviour. A recording is far more useful when the team can match it to drive alarms, PLC events, operator logs and production changes.
Separate supply events from installation events
When a voltage event occurs, the first question is usually simple: did it come from outside the site or from inside it?
The answer needs comparison. One monitor at the affected load may prove that the equipment saw a dip, but it cannot always prove where the dip started. A second monitor upstream, even temporarily installed at the main switchboard or a major distribution board, helps narrow the cause.
If both instruments record the same voltage dip at the same time, with similar depth and duration, the event may have arrived from the supply network or from a large upstream load. If the downstream instrument records a much deeper event than the upstream instrument, the site installation is likely contributing. Cable impedance, transformer loading, poor joints, undersized conductors or large local motor starts may be involved.
A simple comparison can reveal a lot.
Measurement result | Likely direction of investigation |
Dip appears at main incomer and affected load at the same time | Check supply records, large site loads and upstream switching |
Dip is mild upstream but severe near the load | Check submain length, conductor size, terminations and local load starts |
One phase drops more than the others | Check phase loading, loose connections, single-phase loads and supply unbalance |
Voltage is normal but current spikes before a trip | Check the equipment, drive settings, motor condition and mechanical load |
Events occur when capacitors switch | Check capacitor banks, detuning reactors and transient effects |
This is also where time synchronisation helps. Instruments should use the same clock reference as far as practical. Even a few seconds of mismatch can make event comparison harder when machines start and stop rapidly.
Capture voltage dips, swells and interruptions properly
Voltage events are often short. A standard multimeter will miss them. A panel meter that only shows live values may also miss the event unless it has logging and event capture.
Voltage dips can be caused by motor starting, transformer energisation, faults, welding equipment, large compressors or external network events. Swells may occur during load rejection, switching or regulation issues. Short interruptions can come from protection operation, transfer switching or unstable connections.
The investigation should record both the depth and the duration of each event. A dip to 85 per cent of nominal voltage for several cycles is different from a dip to 40 per cent for half a second. Equipment sensitivity also varies. Some contactors, power supplies, drives and PLCs ride through events better than others.
The most useful voltage event record includes:
Event start time and end time
Minimum or maximum voltage on each phase
Duration
A waveform or high-speed record where available
Load current before, during and after the event
Any matching equipment alarm
For drives, compare the electrical record with the drive fault history. A trip labelled undervoltage may not always mean the incoming supply failed. It may reflect DC bus behaviour inside the drive, wiring conditions, braking energy, acceleration settings, or a voltage dip amplified by the local installation.
Treat harmonics as a load and system issue
Harmonics rarely come from one mysterious source. They are usually produced by non-linear loads such as variable speed drives, rectifiers, UPS systems, DC power supplies, LED drivers and some welding equipment. These loads draw current in pulses rather than smooth sine waves.
The current distortion can create voltage distortion as it flows through the impedance of the electrical system. That distinction matters. A load may draw harmonic current, but the wider installation determines how much harmonic voltage distortion appears at sensitive equipment.
High harmonics can contribute to:
Overheated transformers
Hot neutral conductors
Nuisance tripping
Capacitor bank stress
Motor heating and vibration
Reduced equipment life
Interference with control and communications circuits
The investigation should measure harmonic current and harmonic voltage at the same time. Current harmonics point towards the sources. Voltage harmonics show the effect on the supply at that point.
Total harmonic distortion is useful, but it is not the whole story. Individual harmonic orders can matter more than the total number. For example, a strong 5th or 7th harmonic pattern often points towards six-pulse rectifier loads. Triplen harmonics, such as the 3rd, can be important in systems with many single-phase non-linear loads and a shared neutral.
Do not jump straight to filters. First confirm:
Which feeders carry the highest harmonic current
Whether distortion rises with certain production loads
Whether capacitor banks are installed and how they are tuned
Whether transformers or neutral conductors show heating
Whether sensitive equipment is connected near high-distortion loads
Acrel power monitoring devices and power quality equipment can form part of this process when selected and installed to suit the switchboard, voltage level and reporting needs. The key is to use instruments that reveal trends and events, not only instantaneous values.
Power factor is not only a billing issue
Power factor often gets discussed because of network charges or capacity. In troubleshooting, it tells a broader story about how the site uses apparent power.
Poor displacement power factor usually comes from inductive loads such as motors and transformers. Low true power factor can also result from harmonic distortion. Correcting one without understanding the other can create new problems.
For example, a capacitor bank may improve displacement power factor, but it can also interact with harmonic-producing loads. If the system has significant harmonic content, correction equipment may need detuning reactors or a different approach. Installing standard capacitors into a distorted system without checking resonance risk can lead to overheating, blown fuses or repeated switching faults.
A measurement-led power factor review should look at:
kW, kvar and kVA trends
Power factor by feeder and by time of day
Harmonic voltage and current levels
Capacitor switching times
Load profiles during production and idle periods
Transformer loading and spare capacity
The aim is to distinguish between a capacity issue, a correction issue and a distortion issue. They overlap, but they are not the same.
Build an investigation plan that avoids guesswork
Good troubleshooting follows a sequence. It does not start by replacing equipment, changing drive parameters or adding filters without evidence.
A practical plan looks like this.
Record the operating context
Document what was running when the problem occurred. Include shift patterns, machine starts, large motor operation, welding cycles, compressor loading, capacitor switching and any process changes.
Operator notes can be as valuable as logger data. A comment such as “the trip happens when line two and the chiller start together” gives the electrical team a strong lead.
Install monitoring close to the affected equipment
Place the first instrument as close as practical to the load or distribution board involved. Use qualified personnel and follow site isolation and safety procedures. For fixed monitoring, select meters and current transformers suited to the installation.
For temporary monitoring, leave the recorder in place long enough to capture normal operation and the fault condition. A few hours may be enough for frequent events. Intermittent issues may need days or weeks.
Add upstream comparison points
If the first recording shows events, add an upstream point. This can separate local installation effects from broader supply disturbances.
A main incomer monitor is useful, but it should not be the only source if the symptom is local. Comparing main switchboard data with data near the affected equipment gives a clearer picture.
Match electrical events to equipment behaviour
A voltage dip at 10:42:16 means more when the drive trip log shows an undervoltage fault at the same time. If the power quality event appears after the trip, it may be a result rather than the cause.
This order of events matters. A motor fault can cause a current surge that pulls voltage down. A supply dip can cause a drive to trip and unload. The waveform and timing help separate cause from effect.
Inspect what the data points to
Measurements should guide physical inspection. If one phase behaves differently, inspect terminations, fuses, contactors and cable joints on that phase. If neutral current is high, inspect neutral conductors and single-phase load distribution. If harmonics concentrate on one feeder, inspect the loads on that feeder.
Thermal imaging, insulation testing, torque checks and load current checks may all belong in the follow-up, depending on the findings and site rules.
Choose corrective equipment after the evidence is clear
Corrective equipment can be very effective, but it should match the fault mechanism.
Common measures include:
Supply-side or load-side reactors for drives
Line filters for specific equipment
Active harmonic filters for changing harmonic profiles
Passive filters for known harmonic conditions
Detuned power factor correction banks
Static var generators for dynamic reactive power support
UPS or ride-through solutions for sensitive controls
Segregation of sensitive loads from high-disturbance loads
Cable, transformer or switchboard upgrades where impedance or loading is the issue
The right answer may be operational rather than hardware-based. Staggering large motor starts, changing acceleration ramps, moving a sensitive control supply, repairing a poor termination, or balancing phases can solve problems that no filter would fix well.
When applying devices from the Acrel range or any comparable range, match the product to the measured issue. A meter helps reveal the problem. A harmonic filter addresses distortion. A reactive power device addresses changing kvar demand. They are related, but they are not interchangeable.
Make monitoring part of normal plant reliability
A one-off investigation can solve the immediate fault, but ongoing monitoring gives the maintenance team a baseline. That baseline makes future faults easier to spot.
For critical boards, permanent monitoring can track trends in voltage, demand, harmonics, power factor and event history. It also supports energy management and capacity planning. If a new production line increases harmonic current or pushes a transformer closer to its limit, the change becomes visible before failures start.
Useful long-term practices include:
Set practical alarm thresholds for voltage and current events
Review harmonic trends after adding major drives or rectifier loads
Compare power factor during production and low-load periods
Keep records of capacitor bank switching and maintenance
Correlate electrical trends with equipment trips and downtime
Review data after supply changes or plant expansions
Alarm settings deserve care. If alarms are too sensitive, people ignore them. If they are too loose, important changes go unnoticed. Start with a baseline, then adjust thresholds around the site’s real behaviour and equipment tolerance.
The practical takeaway
Power quality faults are frustrating because they often disappear before anyone arrives. The way to solve them is to stop relying on memory, alarms and assumptions.
Measure close to the affected equipment. Compare upstream and downstream data. Record voltage events, harmonics, current, power factor and timing. Then match the electrical evidence to what the plant was doing at the moment of failure.
That approach gives the electrical team a fair chance of finding the real cause, whether it sits in the supply, the installation or a particular load. It also prevents the common mistake of fitting corrective equipment before the problem is properly understood.
