Power factor correction is one of the few electrical interventions whose effect on the bill is directly quantifiable. It is also the field in which solutions lacking any technical basis circulate most frequently. Both facts have the same cause: the phenomenon is deterministic and measurable, but insufficiently understood outside a technical audience.

1. Definition of the phenomenon

In an alternating current circuit with a purely resistive load, voltage and current are in phase and the whole of the energy transported is converted into useful work.

Most industrial and commercial loads do not exhibit this behaviour. An electric motor requires energy to maintain its magnetic field. That energy is not consumed but exchanged with the supply within each cycle. The result is an inductive phase displacement, in which current lags voltage.

The consequence is that the network transports a current greater than that corresponding to the active energy consumed. Conductors, transformer and switchgear are sized for this total current. The distributor charges for reactive energy above a contractually defined threshold, since this component occupies transport capacity without producing useful work.

Correction consists of introducing a reactance of opposite sign, sized so that the two components cancel and current returns into phase with voltage. Active energy consumption is unchanged; what is eliminated is the component charged in addition.

Criterion for assessing a proposal: a measurable phase displacement is corrected by a reactance of opposite sign, sized from the measured value. In the absence of a prior measurement and the calculation derived from it, a proposal cannot be assessed on technical grounds.

2. The capacitive component of reactive power

The inductive character of motor loads is generally understood. Increasingly, however, the opposite component is present in modern installations.

Capacitive loads introduce a phase displacement in which current leads voltage. The usual sources are:

  • LED lighting, through the driver units. The individual contribution is small; across several hundred installed luminaires it becomes significant.
  • Variable speed drives, through DC link capacitors and input filters.
  • UPS equipment and switch-mode supplies, particularly at reduced load.
  • Distributed cable capacitance, relevant in extensive networks.
  • Oversized capacitor banks, or banks left connected as inductive load falls — a correction system generating the very deviation it is intended to correct.

An installation may accordingly present an inductive characteristic during production hours and a capacitive one at reduced load, when only lighting and IT equipment remain in service. Both deviations are generally chargeable. A correction addressing only the inductive component resolves part of the problem and may aggravate it at reduced load.

Sizing therefore requires a profile recorded over a minimum of several complete days, including periods of reduced load and non-working days. There are cases in which the conclusion of the audit is that no correction is required and the charged component arises from another cause.

3. Structure of a correction system

A functional correction system comprises the following elements:

  • Stepped capacitor banks, switched in and out according to load measured in real time. Fixed correction is accurate at a single operating point and incorrect elsewhere.
  • A power factor controller, continuously measuring displacement and commanding the steps.
  • Capacitor-duty contactors fitted with pre-charging resistors. Conventional switching devices do not withstand the inrush current when energising a discharged capacitor.
  • Detuning reactors, in installations with harmonic content. In the presence of drives, a capacitor bank without reactors may resonate with the network inductance, amplifying harmonics and accelerating capacitor degradation.
  • Dedicated protection and thermal management, the assembly operating continuously.

In a project delivered by our team, a system of this type corrected three-phase inductive and capacitive loads of up to 35 kVAr. Installation was planned modularly, with a total production line interruption of 65 minutes. Details are set out in the case study.

4. Assessment of plug-in correction devices

Small devices connected to a socket outlet are marketed as consumption reduction solutions. Technical analysis yields the following findings:

Insufficient capacity. The volume of an enclosure of this size accommodates capacitors offering a correction capacity of the order of 0.3-0.5 kVAr. The reactive load of a small industrial unit is measured in tens of kVAr. The device's contribution falls below the threshold of measurement significance.

Absence of regulation. Fixed correction, without a controller and without measurement, is accurate at a single load point. At reduced load it displaces the installation into the capacitive region.

Incorrect location. Correction is applied close to the inductive load or centrally at the board, not at an arbitrary point of connection on a circuit unrelated to the loads producing the displacement.

Constructional risk. Examination of such devices has revealed: undersized capacitors, absence of discharge resistors, inadequate creepage and clearance distances on the printed circuit board, and the absence of an internal fuse or any fault protection. A capacitor failing to short circuit, without dedicated protection and within a plastic enclosure, constitutes a source of fire initiation.

Not applicable to domestic consumers. Domestic customers are not charged for reactive energy. In the absence of a charged component, the device cannot produce any saving.

5. Verification methodology

Assessing whether investment in correction is warranted proceeds in two stages.

Examination of the invoice. A separate line for reactive energy carrying a material value confirms a quantifiable component. In its absence, correction has no object.

Measurement of the load profile. A power quality analyser installed over several days records the actual profile per phase, including periods of reduced load. The recorded data determines the correction required, the number of steps, and whether detuning reactors are necessary given the harmonic content.

The same measurement yields the payback period, which is the decision criterion. Where power factor already falls within contractual limits, the correct recommendation is not to proceed with the investment.

Related services

Reactive energy on your invoice: technical assessment

We install a power quality analyser and determine, from the recorded profile, whether correction is justified and what the payback period is.

Get a Quote