The symptoms are familiar in any building full of drives, UPS units, LED lighting and IT equipment. A transformer runs hotter than its load suggests. The neutral carries current it should not. The utility bill shows a power factor penalty — even though a capacitor bank was installed precisely to prevent one.
The usual response is to add more capacitance. Sometimes that works. Often it makes things worse, because the problem was never only reactive power. This guide shows how to tell the difference with measurements, and what evidence to collect before anyone specifies a filter, a detuned bank or a larger neutral.
The question to answer
Is the problem reactive loading, waveform distortion, or both? Those are different physical problems with different remedies, and a single dashboard number — “PF 0.82” or “THD 28%” — cannot tell them apart.
Two power factors, not one
A facility with nonlinear loads has two power factors, and confusing them is the most common reason a capacitor bank “doesn’t work”.
- Displacement power factor describes the phase shift between the fundamental voltage and current. It is what capacitor banks correct.
- True power factor also includes the effect of harmonic currents. It is what the supply actually experiences.
If displacement PF is already close to 1 but true PF is low, the shortfall is distortion — and adding capacitors cannot fix it. Worse, capacitors interact with the system inductance and can create a resonance that amplifies particular harmonics. That is why the analyzer must report both values, side by side, at the same time.
Why THD percentages mislead
Total harmonic distortion is a ratio: harmonic content divided by the fundamental. When the fundamental changes, the percentage changes even if the harmonics do not.
Worked example: the denominator trap
- Total harmonic current stays at 10 A throughout the day.
- At full load the fundamental is 100 A, so current THD reads 10%.
- At light load the fundamental falls to 50 A, so current THD reads 20%.
- Nothing got worse. The harmonic amperes are identical — only the denominator moved.
So review harmonic components in amperes as well as percentages, and always keep the fundamental current and the operating state alongside the THD figure. A night-shift THD alarm on a lightly loaded feeder is often this effect, not a new problem.
The measurement plan
- Compare the incomer with representative nonlinear-load feeders — drive panels, UPS inputs, IT distribution — during similar production or occupancy states.
- Keep the quantities together: harmonic amperes per order, phase and neutral currents, voltage distortion, active and reactive power, true and displacement power factor.
- Log capacitor-bank switching and major load changes so a step in the data can be explained.
- Add temperature where it matters. With a temperature sensor on the transformer or busbar, compare temperature rise above ambient against electrical loading — not temperature alone, which mostly tracks the weather.
- Record more than one load state, including light load. Many harmonic problems only appear when the mix of loads changes.
Inspector V4 measures harmonic amplitude and phase, interharmonics, THD, K-factor, unbalance and both power factors on two isolated channels, so the incomer and a feeder can be compared on one timeline. The Portable V4 does the same for a survey that moves from panel to panel.
Reading the evidence
- Rank feeders by contribution, not by percentage. The feeder with the most harmonic amperes matters more than the one with the highest THD.
- Check whether voltage distortion changes with the operating combination of drives, UPS units and compensation. If voltage THD rises when the capacitor bank switches in, suspect resonance.
- Look at the neutral. Triplen harmonics (3rd, 9th, 15th) from single-phase electronic loads add in the neutral rather than cancelling. A neutral carrying more current than the phases is a classic signature.
- Treat heat as a lead, not a verdict. Harmonics can contribute to heating, but the actual thermal cause — loading, a poor connection, ventilation — still needs inspection.
Compliance needs the right denominator. Current THD and demand-based distortion metrics (such as total demand distortion) are not interchangeable. Any compliance statement must use the limit that applies at the correct measurement point, with the specified denominator — avoid presenting one generic THD number as a universal pass mark.
Choosing the remedy
Give the power quality specialist ranked evidence rather than a conclusion. Depending on what the data shows, the right action may be:
- re-balancing or re-allocating single-phase loads,
- resizing the neutral or the transformer for the actual K-factor,
- replacing a standard capacitor bank with a detuned one,
- active or passive harmonic filtering at the dominant feeder,
- or simply changing when certain loads run together.
Do not add capacitance solely because the dashboard reports a low power factor. Confirm first that the shortfall is displacement, not distortion.
Proving the fix worked
Repeat the same measurements across the same load states — including light load — after the intervention. Track harmonic amperes, current imbalance, temperature rise above ambient and both power factor values. A credible result is a measured reduction at the point that mattered, under comparable conditions, not a single good screenshot.
Related reading
- What a power quality analyzer actually tells you
- Building a defensible power quality baseline
- What did the UPS really deliver?
Paying a power factor penalty you can’t explain?
Tell us what the bill and the capacitor bank are doing. In 20 minutes an engineer will tell you what we would measure first. Book a meeting
