The fundamental formula for power factor correction to determine the required reactive power compensation ($Q_c$) in kilovolt-amperes reactive (kVAR) is:

Qc = P × (tan(θ1) - tan(θ2))

Where P is the real power in kW, θ1 is the initial phase angle derived from the existing power factor, and θ2 is the target phase angle derived from the desired power factor. Because utility companies and modern electrical codes heavily penalize facilities operating below a 0.85 to 0.90 power factor, accurately calculating this value is the difference between avoiding demand charges and over-correcting into a leading power factor state, which can cause dangerous voltage swells and resonance.

The Core Formula for Power Factor Correction & Symbol Definitions

To use the formula for power factor correction effectively, we must map the trigonometric functions back to the physical power triangle. In an AC circuit, apparent power (S, measured in kVA) is the vector sum of real power (P, measured in kW) and reactive power (Q, measured in kVAR). The power factor (PF) is simply the cosine of the angle (θ) between the real power and apparent power vectors.

Below is the complete spec-sheet table defining every variable in the standard correction equation:

Symbol Parameter Unit Definition & Notes
Qc Required Reactive Compensation kVAR The exact capacitive reactive power needed to bridge the gap between initial and target states.
P Real (Active) Power kW The actual work-producing power. Must be in kW, not kVA or HP.
PF1 Initial Power Factor Unitless The existing ratio of kW to kVA (e.g., 0.75 lagging). Used to find θ1.
PF2 Target Power Factor Unitless The desired ratio (e.g., 0.95 lagging). Used to find θ2.
θ1 Initial Phase Angle Degrees Calculated as cos-1(PF1).
θ2 Target Phase Angle Degrees Calculated as cos-1(PF2).

For deeper theoretical background on how the power triangle dictates these relationships, the All About Circuits textbook chapter on Power Factor provides an excellent foundational breakdown of the vector geometry involved.

Rearranged Forms: Solving for Every Variable

On the jobsite or at the workbench, you rarely have all the variables neatly handed to you. You might know the capacitor bank you have in stock and need to find out what final power factor it will yield. Here are the algebraically rearranged forms of the formula for power factor correction:

  • Solving for Real Power (P):
    P = Qc / (tan(θ1) - tan(θ2))
    Use case: You have a fixed 20 kVAR capacitor bank and know your PF went from 0.80 to 0.92; this tells you the kW load of the circuit.
  • Solving for Target Tangent (tan(θ2)):
    tan(θ2) = tan(θ1) - (Qc / P)
    Use case: The critical intermediate step for finding your new power factor after installing a known capacitor bank.
  • Solving for Target Power Factor (PF2):
    PF2 = cos( tan-1( tan(cos-1(PF1)) - (Qc / P) ) )
    Use case: Direct calculation of the final meter reading based on installed kVAR.
  • Solving for Physical Capacitance (C) in Farads (Single-Phase):
    C = (Qc × 1000) / (2 × π × f × V2)
    Use case: Converting the required kVAR into physical microfarads (μF) to select or build the actual metallized polypropylene capacitor. Note: V is the RMS voltage across the capacitor, and f is the frequency in Hz.

Worked Examples: Sizing Capacitors for PF Correction

Abstract formulas are useless without rigorous unit tracking. Below are two field-realistic scenarios demonstrating the formula for power factor correction in action.

Problem 1: Sizing a Bank for an Industrial Motor

Given: A 3-phase air compressor draws 50 kW of real power at a measured initial power factor (PF1) of 0.75 lagging. The utility mandates a target power factor (PF2) of 0.95 lagging to avoid penalty fees. Find the required Qc.

  1. Find the initial and target angles:
    θ1 = cos-1(0.75) = 41.41°
    θ2 = cos-1(0.95) = 18.19°
  2. Calculate the tangents of those angles:
    tan(41.41°) = 0.8819
    tan(18.19°) = 0.3287
  3. Apply the core formula with unit tracking:
    Qc = P × (tan(θ1) - tan(θ2))
    Qc = 50 [kW] × (0.8819 [unitless] - 0.3287 [unitless])
    Qc = 50 [kW] × 0.5532
    Qc = 27.66 kVAR

Result: You must install a 27.66 kVAR capacitor bank (typically rounded up to a standard 30 kVAR automated bank with stepped contactors).

Problem 2: Finding the New PF After Installing a Fixed Bank

Given: A manufacturing panel has a steady 120 kW load at a PF1 of 0.80 lagging. You have a surplus 15 kVAR capacitor bank in the stockroom and install it. What is the new power factor (PF2)?

  1. Find initial angle and tangent:
    θ1 = cos-1(0.80) = 36.87°
    tan(θ1) = 0.7500
  2. Calculate the new tangent using the rearranged formula:
    tan(θ2) = tan(θ1) - (Qc / P)
    tan(θ2) = 0.7500 - (15 [kVAR] / 120 [kW])
    tan(θ2) = 0.7500 - 0.1250 = 0.6250
  3. Convert back to an angle and then to Power Factor:
    θ2 = tan-1(0.6250) = 32.01°
    PF2 = cos(32.01°) = 0.848 lagging

Result: The 15 kVAR bank improved the PF from 0.80 to 0.848. If the utility threshold is 0.85, this bank is slightly undersized, and you would need to add another small 2.5 kVAR stage.

Assumptions, Unit Traps, and Realistic Magnitudes

Blindly plugging numbers into the formula for power factor correction will yield dangerous results if you ignore the underlying physics and common calculation traps. According to Fluke's power quality guidelines, ignoring harmonic distortion is the leading cause of capacitor bank failures in modern facilities.

Assumptions & Limits of the Formula:
  • Sinusoidal Steady-State: The formula assumes linear loads (like standard induction motors). It completely breaks down if your facility uses heavy Variable Frequency Drives (VFDs), LED lighting arrays, or UPS systems that generate high Total Harmonic Distortion (THD).
  • The Harmonic Trap: If THD exceeds 15-20%, standard capacitors will act as low-impedance sinks for high-frequency harmonics, leading to catastrophic dielectric failure. In these cases, you must use the formula to size the kVAR, but then install detuned reactors (typically tuned to 189 Hz / 7% detuning for 50Hz systems, or 228 Hz / 7% for 60Hz systems) in series with the capacitors.
Critical Unit Mistakes That Break the Math:
  • Mixing kW and kVA: The multiplier P must be Real Power in kW. If you use Apparent Power (kVA) in the base equation, your Qc result will be artificially inflated, leading you to buy an oversized bank that pushes the system into a leading power factor.
  • Degrees vs. Radians: Ensure your calculator or spreadsheet software is set to Degrees when using cos-1 and tan. Using Radians will yield negative or nonsensical kVAR values.
  • Horsepower to kW conversion: If your motor nameplate only lists HP, you must convert it using: kW = (HP × 0.746) / Motor_Efficiency. Forgetting to divide by efficiency (e.g., 0.92) will underestimate the real power draw.

What does a realistic answer magnitude look like?
In practical electrical engineering, you almost never correct to a PF of 1.0 (unity). A target PF2 of 0.95 to 0.98 is the industry standard. Correcting to exactly 1.0 leaves zero margin for error; if the motor load drops slightly while the capacitor bank remains online, the system becomes net-capacitive (leading PF). This causes voltage magnification across the supply impedance, potentially tripping overvoltage relays or damaging sensitive PLCs. For a standard 100 kW industrial load, realistic Qc magnitudes usually fall between 15 kVAR and 45 kVAR, depending on how poor the initial motor loading is.

Frequently Asked Questions

How do I apply the formula for power factor correction to a 3-phase system?

The core formula Qc = P × (tan(θ1) - tan(θ2)) remains exactly the same for 3-phase systems, provided that P represents the total 3-phase real power in kW. The result is the total 3-phase kVAR required. The difference arises only when selecting the physical capacitors. If you are wiring the capacitor bank in a Delta configuration (standard for medium voltage), the kVAR per phase is simply Qc / 3, and the voltage used for the capacitance (Farads) calculation is the Line-to-Line voltage. If wired in Wye, the voltage used is the Line-to-Neutral voltage.

Why does the formula for power factor correction use tangent instead of sine or cosine?

This is dictated by the geometry of the power triangle. Real Power (P) forms the horizontal adjacent side, and Reactive Power (Q) forms the vertical opposite side. In trigonometry, the ratio of the opposite side to the adjacent side is the tangent of the angle (θ). Therefore, Q / P = tan(θ), which rearranges to Q = P × tan(θ). Since we want to find the difference in reactive power before and after correction, we subtract the target reactive power from the initial reactive power: Qinitial - Qtarget = (P × tan(θ1)) - (P × tan(θ2)). Factoring out P gives us the standard formula.

What happens if my calculated kVAR is negative using the power factor correction formula?

A negative Qc result means your target power factor angle is larger than your initial power factor angle, or your initial load is already leading (capacitive). In industrial settings, this usually happens if you have massive underground cable runs (which have high inherent capacitance) or if a previous capacitor bank was left online while the inductive motor loads were turned off. If your calculation yields a negative number, adding more capacitance will worsen the leading power factor and cause severe overvoltage conditions. To correct a leading power factor, you must install inductive compensation (shunt reactors), and the math requires swapping the trigonometric signs to account for leading vs. lagging vectors.