Loads
VFD harmonics: what a reactor actually buys you
Six-pulse drives are the largest harmonic source in most plants, and the difference between a bare drive and one with a reactor is measurable, predictable, and cheaper than filtration.
A six-pulse VFD produces characteristic harmonics at orders 6k±1 — the 5th, 7th, 11th and 13th — with current THD typically between 30% and 45% when no input reactor is fitted.
Adding an AC line reactor or DC choke reduces that to 15–35%, at a fraction of the cost of active filtration. It is the first thing to check before specifying a filter.
Why the drive produces harmonics at all
A VFD converts fixed-frequency AC to DC, then inverts it back to variable-frequency AC. The first stage is a rectifier feeding a DC capacitor. That capacitor only accepts charge when the rectified voltage exceeds the voltage already stored on it, so current flows in short pulses near the peak of each half cycle rather than continuously.
The pulsed current is rich in odd harmonics. For a balanced three-phase six-pulse bridge, the symmetry cancels all orders except 6k±1 — that is, 5, 7, 11, 13, 17, 19 and so on. The 5th and 7th typically carry 60–80% of the total harmonic current.
h = 6k ± 1 → 5, 7, 11, 13, 17, 19, 23, 25 …
# magnitude falls roughly as 1/h
I5 ≈ I1 / 5 I7 ≈ I1 / 7 I11 ≈ I1 / 11
The 1/h relationship explains why the 5th harmonic is always the largest, and why filtration effort concentrates there.
The four front-end options, compared
| Front end | Current THD | Cost impact | Also provides |
|---|---|---|---|
| Bare six-pulse diode bridge | 30–45% | baseline | — |
| + DC link choke | 25–35% | +3–6% | Reduced DC ripple, better capacitor life |
| + AC line reactor (3–4%) | 15–25% | +4–8% | Surge protection, reduced inrush, lower drive faults |
| Twelve-pulse with phase-shift transformer | 8–12% | +25–40% | Removes 5th and 7th almost entirely |
| Active front end (AFE) | <5% | +40–70% | Regeneration capability, unity PF, DC bus control |
| Passive harmonic filter at drive | 5–10% | +20–35% | Tuned to 5th, partial 7th |
The AC line reactor row deserves attention. For a modest cost premium it halves the THD and also solves a set of unrelated problems — supply transients, capacitor-charging inrush, and nuisance overvoltage trips. On a plant with a dozen drives and no reactors, fitting them is usually the highest-return action available.
Reactor versus filter, in economics
The choice depends mainly on how many drives are involved, because a central filter shares its cost across all of them while reactors are per-drive.
| Drives on the board | Lower-cost route | Reason |
|---|---|---|
| 1–3 | Line reactors at each drive | Filter has a fixed cost floor that cannot be spread |
| 4–8 | Either, model both | Crossover depends on drive rating and filter size |
| 9+ | Central active filter | One device covers all sources, and handles future additions |
There is a third option that is often overlooked: replacing the oldest, largest drive with an active front end model. A single AFE drive in a group also acts as a harmonic sink for the others if correctly controlled, though this is a design decision rather than a retrofit.
The power factor trap
Drives produce an effect that catches out plants that installed capacitors on the basis of an energy audit.
A six-pulse drive has a displacement power factor close to 0.97 — the fundamental current is almost in phase with the voltage. But its true power factor, which includes distortion, may be 0.70. Utilities that meter true power factor will bill for the difference, and capacitors cannot correct it because the shortfall is not reactive power.
Worse, adding capacitors to a drive-heavy board creates the resonance condition described in why capacitor banks overheat. The bank appears to work for a few months, then fails.
If a plant with drives has a power factor penalty that capacitors did not fix, the answer is filtration or an AFE, not more capacitance.
Frequently asked
How much harmonic distortion does a VFD produce?
A six-pulse drive with no input reactor typically produces 30–45% current THD. Adding a DC link choke brings it to 25–35%, and an AC line reactor to 15–25%. A twelve-pulse configuration reaches 8–12%, and an active front end typically stays under 5%.
Is a DC choke or an AC reactor better for reducing VFD harmonics?
An AC line reactor also protects the drive from supply transients and reduces inrush, so it addresses two problems. A DC choke is somewhat more efficient at reducing current distortion for a given cost and generates less heat. In practice many drives ship with one or the other already fitted, and the specification question is which of the two the vendor has included.
Do VFDs cause power factor penalties?
They can, even though the drive's displacement power factor is close to unity. A diode rectifier front end draws current in pulses, so true power factor is much lower than displacement power factor. Utilities that bill on true power factor will apply a penalty that capacitors cannot correct, because the shortfall is distortion rather than phase shift.
What is the cheapest way to reduce harmonics from existing drives?
Adding AC line reactors to drives that lack them is usually the lowest-cost intervention per point of THD reduction, and it also provides surge protection. Beyond that, a central active filter at the board is more economical than retrofitting reactors to every drive once the number of drives exceeds roughly eight to ten.
Drives pushing you over a harmonic limit?
Send the drive list with ratings, plus a harmonic recording at the main board. We will model reactor retrofit against central filtration and give you the cheaper route with the resulting TDD.