How Power Factor Affects Cable Sizing — Free Practical Guide

Power factor is usually discussed in the context of electricity bills and utility penalties. Engineers learn to install capacitor banks to avoid surcharges for reactive power consumption. But there is a far more expensive consequence of poor power factor that rarely gets the attention it deserves: the impact on cable sizing, cable losses, and conductor costs.

A motor running at 0.7 power factor draws 43% more current than the same motor would at unity power factor — for exactly the same shaft power output. That 43% increase in current means you need a significantly larger cable, more copper, bigger switchboard busbars, and larger protective devices. The cost of this additional infrastructure frequently exceeds the cost of the reactive energy charges on the electricity bill.

This article focuses specifically on what power factor does to cables — the component most engineers size without thinking carefully about the PF of the load they are supplying.

In an AC circuit, voltage and current are both sinusoidal, but they may not be in phase. The phase angle between them, denoted φ, determines the power factor:

Power Factor (PF) = cos(φ) — (Eq. 1)

This phase difference creates three types of power:

• Real Power (P) — measured in watts (W). This does actual work: turning motors, generating heat, producing light.
• Reactive Power (Q) — measured in volt-amperes reactive (var). This sustains magnetic fields in inductors and electric fields in capacitors. It flows back and forth without doing useful work.
• Apparent Power (S) — measured in volt-amperes (VA). This is the vector sum of real and reactive power, and it represents the total current the supply must deliver.

S = √(P² + Q²) — (Eq. 2)

The key insight for cable engineers: cables must carry the current corresponding to apparent power S, not real power P. The cable does not know or care whether the current is doing useful work. Every ampere generates I²R losses regardless of the load’s power factor.

The current drawn from the supply for a given real power P is:

I = P / (√3 × V_LL × PF) — three-phase — (Eq. 3)

As PF decreases, current increases in direct inverse proportion. For a fixed 100 kW three-phase load at 415 V:

At PF = 0.70, the current is nearly 200 A instead of 139 A. The cable must be sized for 200 A, not 139 A.

A factory has a 100 kW three-phase load at 415 V. Compare the cable requirements for two scenarios.

Scenario A: PF = 1.0

I = 100,000 / (1.732 × 415 × 1.0) = 139 A — (Eq. 4)

From AS/NZS 3008.1.1:2017, Table 13, Column 17 (single-core cables in trefoil on cable tray, 75°C thermoplastic), a 35 mm² copper cable is rated at 158 A. This is sufficient.

Scenario B: PF = 0.70

I = 100,000 / (1.732 × 415 × 0.70) = 199 A — (Eq. 5)

A 35 mm² cable (158 A rating) is no longer sufficient. Stepping up: 50 mm² is rated at 192 A — still not enough. 70 mm² at 246 A provides adequate capacity.

The cost impact: The cable size jumps from 35 mm² to 70 mm² — doubling the cross-sectional area. For a 100-metre run, three phases plus neutral:

• 35 mm² cable: approximately $8.50/m × 400 m = $3,400
• 70 mm² cable: approximately $15.50/m × 400 m = $6,200

The poor power factor adds $2,800 in cable costs alone for a single 100-metre circuit. For a large industrial facility with dozens of circuits, this multiplies to tens of thousands of dollars in additional copper.

Beyond the capital cost of larger cables, poor power factor increases I²R losses for the entire operating life of the installation. Using the same 100 kW load over 100 metres of cable:

Scenario A: PF = 1.0, 35 mm² cable

Scenario B: PF = 0.70, 70 mm² cable

Despite using a cable twice the size (and twice the cost), Scenario B still has higher losses because the current is so much greater. Without upsizing, a 35 mm² cable carrying 199 A would dissipate 7,936 W — nearly 8% of the useful 100 kW load — and would also be operating above its current rating and overheating.

Annual energy cost of the extra losses (4,000 operating hours/year at $0.15/kWh): Scenario A costs $2,324/year, Scenario B costs $2,437/year. The $113/year difference per circuit multiplies across a facility with 20 similar circuits to $2,260/year — over a 25-year installation life, that is $56,500.

This is the insight that changes how engineers think about power factor correction:

The traditional cost-benefit analysis for PFC only considers energy savings. “Install a capacitor bank, reduce reactive energy charges, payback in 2–3 years.” This analysis is correct but incomplete.

The complete analysis should include:

1. Capital savings on cables: Smaller cables for the same real power
2. Capital savings on switchgear: Lower-rated circuit breakers, busbars, cable glands
3. Capital savings on transformers: Smaller transformer for the same real power delivery
4. Reduced cable losses: Lower I²R over the installation life
5. Reduced voltage drop: Lower current means less voltage drop, potentially avoiding cable upsizing for voltage drop compliance
6. Energy savings on the electricity bill: The traditional PFC benefit

When all six factors are included, the payback period for power factor correction equipment typically drops from 2–3 years to 6–12 months — and in new installations, the reduced cable and switchgear costs mean the PFC equipment essentially pays for itself before the building is even energised.

Poor power factor compounds the cable sizing problem when voltage drop is the limiting constraint. The voltage drop in a cable depends on both resistance and reactance:

ΔV = I × (R × cos(φ) + X × sin(φ)) × L — (Eq. 6)

At low power factor, the reactive component (X × sin(φ)) becomes significant. For a cable with non-negligible reactance (large sizes, wide spacing), the voltage drop increases faster than the current increase alone would suggest.

For the 100 kW / 100 m example:

Scenario A (PF = 1.0): ΔV = 139 × (0.668 × 1.0 + 0.08 × 0.0) × 0.1 = 9.29 V per phase (3.87%)

Scenario B (PF = 0.70): ΔV = 199 × (0.342 × 0.70 + 0.08 × 0.714) × 0.1 = 5.90 V per phase (2.46%)

In this case, the larger cable (70 mm²) keeps the voltage drop acceptable. But if the run were longer or the allowable voltage drop more restrictive (say 2% for motor circuits per AS/NZS 3000:2018, Section 3.6), the poor power factor could force yet another cable size increase beyond what current rating alone requires.

Understanding which loads cause poor power factor helps engineers anticipate cable sizing challenges:

The worst offenders are lightly loaded induction motors and welding equipment. A factory with many motors running at partial load can have an overall power factor as low as 0.65, requiring cable infrastructure sized for 54% more current than the real power consumption demands.

The cable impact of poor power factor cascades upstream. Every cable segment between the load and the supply — submain, main switchboard feeder, transformer cable — carries the inflated current. The largest cables in the installation (the most expensive ones) see the aggregated effect of all downstream loads’ power factors.

This is why the optimal location for power factor correction is as close to the load as possible. Capacitors at the motor terminals reduce current in every cable from that point back to the supply. Centralised PFC at the main switchboard only reduces current upstream of the switchboard — all the final circuits and submains still carry the inflated current.

1. A load at PF = 0.70 draws 43% more current than the same load at PF = 1.0 — the cable must be sized for apparent power, not real power.
2. For a 100 kW three-phase load at 415 V, poor power factor (0.70) forces the cable size from 35 mm² to 70 mm² — doubling the copper cost for the same useful power delivery.
3. Even after upsizing the cable, the I²R losses remain higher because the current is higher. There is no way to avoid this penalty without improving the power factor.
4. The traditional PFC cost-benefit analysis that only considers energy savings dramatically underestimates the true benefit. Including cable, switchgear, and transformer savings typically halves the payback period.
5. Correcting power factor at the load (not centrally) reduces current in every cable back to the supply — maximising the infrastructure savings across the entire installation.

Standards referenced: AS/NZS 3008.1.1:2017, Table 13 and Section 4.4; AS/NZS 3000:2018, Section 3.6; BS 7671:2018+A2, Appendix 4; IEC 60364-5-52:2009, Annex G.