Harmonics and Power Factor: Why VFDs Complicate Correction

Variable frequency drives save energy by letting motors run at exactly the speed the load demands, but they introduce a problem that plain capacitor banks cannot solve and can actually make worse. Understanding why requires looking at how a VFD draws current from the supply in the first place.

How VFD Rectifiers Distort the Current Waveform

A standard VFD front end is a six-pulse diode bridge rectifier. It converts AC to a DC bus, and the DC bus capacitors hold that voltage relatively steady. The diodes only conduct when the incoming AC voltage exceeds the DC bus level, which happens twice per cycle near the peaks of each phase. The result is that the drive pulls current in short, high-amplitude pulses rather than in a smooth sine wave.

This pulsed current is non-sinusoidal, and any non-sinusoidal waveform contains harmonic components on top of the fundamental 60 Hz (or 50 Hz) signal. Fourier analysis breaks that distorted waveform into a series of sine waves at integer multiples of the fundamental: the 5th harmonic at 300 Hz, the 7th at 420 Hz, the 11th at 660 Hz, and so on. Six-pulse rectifiers generate harmonics following the pattern 6n ± 1, so the dominant orders are the 5th and 7th, with the 11th and 13th present at lower amplitudes.

The 5th harmonic is particularly problematic because it is the largest in magnitude and it rotates in the negative sequence, meaning it circulates in the opposite direction from the fundamental. Negative-sequence harmonics create braking torque in induction motors and cause extra heating in transformer windings and neutral conductors.

Displacement Power Factor vs. True Power Factor

This is where the distinction between displacement power factor and true power factor becomes critical. See the full breakdown in displacement vs. true power factor.

Displacement power factor describes the phase angle between the fundamental voltage and fundamental current. A VFD with a diode bridge front end often has a displacement power factor of 0.95 or better, because the drive's DC bus capacitors pull the current peaks close to the voltage peaks at the fundamental frequency.

True power factor accounts for all the harmonic content. It is calculated as:

True PF = P / (V_rms × I_rms)

where I_rms includes the energy in every harmonic. When a 100 kW drive draws significant 5th and 7th harmonic current, its I_rms rises well above what the fundamental alone would suggest, and the true power factor drops accordingly. A drive that measures 0.95 on displacement power factor may have a true power factor of 0.75 to 0.85 once harmonic distortion is accounted for.

The distinction matters for utility billing. Some utilities measure true power factor for demand charges; others measure only displacement. Knowing which one applies to a facility changes the economics of any correction strategy.

Why Capacitor Banks Resonate With Harmonics

Traditional power factor correction works by adding capacitance to cancel inductive reactive power. That works cleanly when the current is sinusoidal. When VFDs are present, the capacitors interact with system inductance to form LC resonant circuits at specific frequencies.

Resonance frequency is proportional to 1/√(LC). If that resonant point lands near the 5th or 7th harmonic, the system amplifies those harmonic currents dramatically rather than absorbing them. A 480 V bus with a standard capacitor bank could see harmonic currents five to ten times higher than without the capacitors. Capacitor fuses blow, capacitor cans bulge, and harmonic voltage distortion climbs to levels that disrupt sensitive controls and overheat transformer cores.

Adding capacitor banks to a bus with significant VFD load without harmonic analysis is one of the more reliable ways to cause equipment failures. Details on capacitor sizing without VFDs are covered in sizing power factor correction capacitors.

Mitigation Options

There is no single solution that fits every situation. The right choice depends on the number of drives, the total harmonic distortion before mitigation, utility requirements, and budget. The table below summarizes the main options.

Mitigation MethodTypical THD ReductionNotes
3% AC line reactorModerate (to ~35% THD)Low cost, always worth adding
5% AC line reactorBetter (to ~28% THD)Slightly higher voltage drop
12-pulse rectifierGood (to ~10–12% THD)Requires phase-shifting transformer
Detuned filter (C + reactor)Good at fundamental, harmonic-safeCorrects displacement PF without resonance
Passive harmonic filterVery good (to ~5–8% THD)Tuned to 5th/7th; bulky, load-sensitive
Active harmonic filterExcellent (under 5% THD)Injects cancelling currents; load-flexible
Active front end (AFE) driveExcellent (under 5% THD)Built into drive; regenerative capable

AC line reactors are the cheapest first step. A 3% or 5% impedance reactor placed in series between the supply and the VFD input increases the effective source impedance seen by the rectifier, which smooths out the current pulses and reduces peak harmonic magnitudes. They do not eliminate harmonics, but they reduce them enough to protect the drive and push THD into a range that may be acceptable for smaller facilities.

Detuned filters combine a power factor correction capacitor bank with a series reactor that shifts the resonant frequency below the 5th harmonic, typically to around the 4th. This prevents the bank from amplifying drive harmonics while still delivering the displacement power factor correction the utility sees. Read more about this approach in detuned reactors and capacitor protection.

Active harmonic filters are standalone units that monitor the bus current in real time and inject equal-and-opposite harmonic currents to cancel distortion. They adapt to changing load and can achieve very low THD on buses with mixed linear and nonlinear loads.

Active front end drives replace the diode bridge with a fully controlled IGBT rectifier. The converter draws nearly sinusoidal current at unity power factor and can return energy to the grid during braking. The upfront cost is higher, but for large drives in facilities where regenerated energy has value, the economics often work out.

Practical Starting Points for Facilities With Multiple VFDs

When a facility runs more than two or three VFDs on a common bus, harmonic currents from each drive add together. The combined distortion at the point of common coupling can exceed IEEE 519 limits even if each individual drive is within spec on its own.

A harmonic study using measured current data and the facility's system impedance is the right starting point for anything above about 50 hp of combined VFD load. The study identifies the resonant frequencies that exist in the system, predicts THD at the metering point, and shows which mitigation measures are cost-effective. Guessing is expensive.

For smaller installations, the rule of thumb is simple: add a 3% or 5% line reactor to every VFD that does not already have one, avoid standard fixed capacitor banks on the same bus as drives, and use a detuned filter if displacement power factor correction is needed. This combination handles most cases without a full harmonic study.

Frequently Asked Questions

Does a VFD improve or hurt power factor?

It depends which type is measured. A six-pulse VFD typically improves displacement power factor compared to a motor running across the line at partial load. However, the harmonic current it generates reduces true power factor. From a utility billing perspective, the outcome depends entirely on whether the utility measures displacement or true power factor.

Can I add a standard capacitor bank to correct power factor on a bus with VFDs?

Not safely without analysis. Standard capacitor banks can resonate with system inductance at harmonic frequencies, which amplifies distortion and risks damaging the capacitors and the drives. A detuned filter or active harmonic filter is the correct choice when VFDs share a bus with power factor correction equipment.

What is the difference between a passive harmonic filter and an active harmonic filter?

A passive filter uses fixed inductors and capacitors tuned to specific harmonic frequencies. It works well for stable, predictable loads but loses effectiveness when load or harmonic content changes. An active filter uses power electronics to continuously measure and cancel harmonic currents in real time, making it more flexible but also more expensive.

Does adding a line reactor help with power factor, or only with harmonics?

Primarily harmonics. The line reactor reduces harmonic current magnitude, which improves true power factor by bringing measured I_rms closer to the fundamental-only value. It does not provide reactive power compensation for displacement power factor in the way a capacitor bank does. For displacement correction alongside harmonic mitigation, a detuned filter is a better tool.