13.3 Power factor correction techniques

Capacitor-based Correction

Power factor correction brings the power factor closer to unity by supplying or absorbing reactive power. A low power factor means more current is drawn for the same real power, which increases losses and wastes system capacity. The techniques here range from simple capacitor banks to sophisticated electronic compensators.

Types of Capacitor Systems

Shunt capacitors are the most common approach. They connect in parallel with the load and supply leading reactive power (vars) to offset the lagging reactive power drawn by inductive loads like motors and transformers. The net effect is a reduction in the reactive component of line current.

Series capacitors serve a different purpose. Installed in series with a transmission line, they cancel a portion of the line's inductive reactance. This lowers the effective impedance of the line, improving power transfer capability and voltage regulation over long distances. Series capacitors are mainly used in transmission systems rather than at the load level.

Capacitor banks group multiple capacitor units in parallel, series, or both to achieve the required kvar rating and voltage class. Banks can be fixed (always connected) or switched in discrete steps to match varying load conditions.

Implementation and Benefits

Strategic placement matters. Capacitors installed close to the load correct the power factor upstream of that point, reducing losses in all conductors and transformers between the source and the capacitor. Placing them at the main bus of a facility is simpler but provides less loss reduction than distributing them near individual loads.

Key benefits of capacitor-based correction:

• Reduced $I^2R$ losses in cables and transformers, since correcting the power factor lowers the current magnitude
• Freed-up system capacity, allowing existing infrastructure to serve additional real power load
• Improved voltage profile, because lower reactive current flow means less voltage drop across line impedances
• Lower utility bills in systems where the utility charges a power factor penalty

Switching can be manual (an operator closes a contactor) or automatic (a controller monitors power factor and switches stages in or out). Automatic switching is standard for loads that vary throughout the day.

Design Considerations

Sizing a capacitor bank requires knowing the existing power factor and the target power factor. The required reactive power is:

$Q_C = P(\tan\phi_1 - \tan\phi_2)$

where $P$ is the real power, $\phi_1$ is the original power factor angle, and $\phi_2$ is the target power factor angle. Overcompensation (pushing the power factor leading) can cause voltage rise and should be avoided.

Harmonic resonance is a serious concern. A capacitor bank and the system inductance form a parallel resonant circuit at a frequency $f_r = f_1 \sqrt{\frac{S_{sc}}{Q_C}}$, where $S_{sc}$ is the short-circuit power at the bus and $Q_C$ is the capacitor bank rating. If this resonant frequency lands near a harmonic produced by non-linear loads (drives, rectifiers), currents and voltages at that harmonic can be amplified dramatically. Solutions include detuning reactors (typically tuned to around 189 Hz or 7th harmonic for 50 Hz systems) placed in series with the capacitors.

Protection for capacitor banks includes fuses or circuit breakers rated for the high inrush currents during energization, as well as unbalance detection relays that trip the bank if individual units fail.

Dynamic Compensation Devices

When loads change rapidly or the system needs continuous, smooth reactive power control, capacitor banks alone may not respond fast enough. Dynamic compensation devices fill this role.

Synchronous Condensers

A synchronous condenser is a synchronous motor running without a mechanical load. By adjusting its field (excitation) current, it can absorb or generate reactive power continuously.

• Overexcited: the machine generates vars (acts like a capacitor), raising the power factor
• Underexcited: the machine absorbs vars (acts like an inductor)

Synchronous condensers respond within a few cycles to voltage disturbances. They also contribute rotational inertia to the grid, which helps stabilize frequency during sudden generation-load imbalances. This inertia benefit is increasingly valuable as grids integrate more inverter-based renewables that provide no natural inertia.

The tradeoff is higher maintenance (bearings, brushes, cooling systems) and slower installation compared to static alternatives.

Static VAR Compensators (SVCs)

SVCs use power electronics to achieve rapid, stepless reactive power control without any rotating parts. A typical SVC combines two elements:

• Thyristor-Controlled Reactors (TCRs): inductors whose current is varied by adjusting the thyristor firing angle. At full conduction the reactor absorbs maximum vars; at minimum conduction it absorbs almost none.
• Thyristor-Switched Capacitors (TSCs): capacitor banks switched in or out by thyristors in discrete steps, providing blocks of capacitive vars.

The TCR provides continuous fine adjustment between the discrete TSC steps. Together, they can swing from full capacitive to full inductive output in milliseconds.

SVCs improve voltage regulation at weak points in the grid, increase the power transfer limit of transmission corridors, and reduce flicker in industrial plants with arc furnaces or large motor starts. One limitation is that TCRs generate odd harmonics (especially 5th and 7th), so SVCs typically include passive harmonic filters tuned to those frequencies.

Comparison and Selection Criteria

Selection depends on the application. A commercial building with predictable loads can use a switched capacitor bank. A steel mill with arc furnaces needs an SVC for flicker control. A remote grid connection point for a wind farm might benefit from a synchronous condenser for both var support and inertia.

Control Systems

Power Factor Correction Controllers

Modern PF correction controllers are microprocessor-based units that measure voltage and current in real time, compute the power factor (and often individual harmonics), and decide which capacitor steps to switch.

The controller compares the measured reactive power or power factor against a setpoint (commonly 0.95 to 0.99 lagging) and determines how many capacitor stages to connect. Most controllers use a C/k algorithm: they define a minimum switching step size and only add or remove a stage when the reactive power error exceeds that threshold. This prevents hunting (rapid on-off cycling).

Typical features include:

• Programmable target power factor and switching thresholds
• Time delays between switching operations (often 30 to 60 seconds) to protect contactors and capacitors
• Alarm outputs for overcurrent, overvoltage, or capacitor failure
• Communication interfaces (Modbus, Ethernet) for integration with SCADA or building management systems
• Data logging of power factor, kvar, and switching events for analysis

Automatic Power Factor Correction (APFC) Systems

An APFC system pairs a controller with a bank of contactors (or thyristor switches) and multiple capacitor stages. The system operates as follows:

1. The controller continuously samples voltage and current waveforms.
2. It calculates the present power factor and the reactive power deficit or surplus relative to the target.
3. If correction is needed and the time delay has elapsed, the controller closes or opens the appropriate contactor to add or remove a capacitor stage.
4. After each switching event, the controller waits for the system to settle, then re-evaluates.
5. Steps repeat as the load changes throughout the day.

For loads with very fast fluctuations, thyristor-switched stages replace mechanical contactors. Thyristors can switch at the voltage zero-crossing, eliminating inrush transients and allowing much faster response.

Advanced Control Strategies

Predictive control uses historical load profiles (time of day, day of week, seasonal patterns) to anticipate reactive power needs and pre-position capacitor stages before a load change occurs. This reduces the number of switching events and keeps the power factor closer to the target during transients.

Adaptive control adjusts its parameters over time based on observed system behavior. If a particular load pattern consistently causes the controller to overshoot, the algorithm tightens its response for that scenario.

Integration with smart grid infrastructure allows a central energy management system to coordinate multiple APFC units across a facility or distribution network. This coordination prevents conflicting actions (one unit adding vars while another removes them) and enables system-wide optimization of voltage and reactive power flow.