Resistive vs Reactive Load Bank: kW, kVAR & Power Factor

A resistive load bank tests real power (kW) at unity power factor, while a reactive load bank adds reactive power (kVAR) to simulate inductive or capacitive loads at a lower power factor. To test a generator to 100% of its nameplate kVA rating at the standard 0.8 power factor, you need a combined resistive plus reactive load bank.

What is the difference between a resistive and reactive load bank?

The difference comes down to which type of power each one loads. A resistive load bank converts electrical energy entirely into heat through resistive elements, drawing real power measured in kilowatts (kW) at a power factor of 1.0 (unity). It answers one question: can the prime mover (the engine or turbine) actually produce its rated real power?

A reactive load bank adds inductance or capacitance to the circuit, drawing reactive power measured in kilovolt-amperes reactive (kVAR). Reactive power does no useful work but circulates between the source and the load, and handling it is the job of the alternator and its automatic voltage regulator (AVR) rather than the engine. A reactive load bank therefore tests a different part of the system entirely.

In short: resistive load banks stress the engine, and reactive load banks stress the alternator and voltage regulator. Most real-world electrical loads contain both real and reactive components, which is why comprehensive testing usually needs both.

• Resistive load bank: loads real power (kW) as heat, power factor 1.0, tests the engine and cooling system.
• Reactive load bank: loads reactive power (kVAR) via inductors (lagging) or capacitors (leading), tests the alternator and AVR.
• Combined resistive plus reactive load bank: loads both simultaneously to reach full kVA at a defined power factor such as 0.8 lagging.

What are kW, kVAR and kVA? Understanding the power triangle

AC power has three components that are linked by simple geometry. Real power (P), measured in kW, is the power that performs useful work such as heat, torque or light. Reactive power (Q), measured in kVAR, is the power that sustains the magnetic and electric fields in motors, transformers and capacitors; it flows back and forth but does no net work. Apparent power (S), measured in kVA, is the total power the source must supply.

These three relate as a right-angled triangle: apparent power squared equals real power squared plus reactive power squared (S squared = P squared + Q squared). Power factor is the ratio of real power to apparent power (pf = kW / kVA), which is also the cosine of the phase angle between voltage and current.

A worked example makes it concrete. A generator rated 1000 kVA at 0.8 power factor delivers 800 kW of real power and 600 kVAR of reactive power at the same time (800 squared plus 600 squared equals 1000 squared). A resistive load bank alone can only load the 800 kW leg of that triangle. To load the full 1000 kVA, you also have to supply the 600 kVAR leg, and only a reactive load bank can do that.

• kW (real power): does useful work; loaded by resistive elements.
• kVAR (reactive power): sustains magnetic or electric fields; loaded by inductors or capacitors.
• kVA (apparent power): the vector sum, S squared = P squared + Q squared; what the alternator is ultimately rated for.
• Power factor = kW / kVA; 0.8 lagging means 80% of the apparent power is real and 20% is reactive.

Why does power factor (0.8 lagging) matter for testing?

Generator alternators are almost universally rated at 0.8 power factor lagging. This is the reference point in the international standards for rotating machines and generating sets (IEC 60034 and ISO 8528), so a set's nameplate kVA is defined at that power factor. A 0.8 rating means the machine is built to deliver 80% of its apparent power as real power and the remaining 20% as reactive power.

This matters because a purely resistive load can never exercise that reactive capability. If you load a 0.8 pf generator with a resistive-only bank, you can reach its kW rating but you top out at roughly 80% of its nameplate kVA. The alternator windings, the excitation system and the AVR that manage reactive current are left largely untested, and that is precisely the equipment most likely to struggle in service.

Real installations are dominated by lagging (inductive) loads: induction motors, transformers, pumps, compressors, fluorescent and LED drivers. When a large motor starts, its inrush current can be six to ten times its running current at a very poor lagging power factor, causing a sudden voltage dip that the AVR must correct instantly. Testing at 0.8 lagging with a reactive load bank proves the alternator and AVR can hold voltage and frequency under exactly that kind of stress before the load ever depends on it.

• Alternators are rated at 0.8 pf lagging per IEC 60034 and ISO 8528, so full-rating testing must occur at that power factor.
• Resistive-only testing reaches 100% kW but only about 80% of nameplate kVA, leaving the alternator and AVR under-tested.
• Lagging loads (motors, transformers) and motor inrush are the real-world conditions that a reactive load bank reproduces.

When do you need a combined resistive plus reactive load bank?

You need a combined resistive plus reactive load bank whenever the acceptance criterion is the full nameplate kVA rating at a specified power factor, rather than just the kW rating. This is the norm for commissioning and warranty acceptance of standby and prime-power generator sets, where proving the machine can deliver 100% of its kVA at 0.8 lagging is a contractual requirement.

A combined unit contains both resistive elements and inductive elements that switch independently, so a single machine can apply resistive-only, reactive-only, or any blend of the two. To create a 0.8 lagging load, the inductive rating is sized at 75% of the resistive rating: for every 100 kW of resistance you add 75 kVAR of inductance, which produces 125 kVA at 0.8 pf. Independent switching also lets operators sweep across power factors to map alternator and AVR behaviour.

For UPS systems, data centres and installations with significant electronic or power-factor-correction loads, the concern can instead be leading power factor. Here a capacitive (leading) reactive load bank is used, often with a resistive bank, to verify voltage-regulator stability under leading conditions that resistive tests never expose. Choosing between inductive and capacitive reactive loading depends on whether the real-world load is dominated by motors and transformers (lagging) or by modern switch-mode electronics and correction capacitors (leading).

• Use combined resistive plus reactive testing to prove full kVA at 0.8 lagging during generator commissioning and warranty acceptance.
• Sizing rule of thumb: 75 kVAR of inductance per 100 kW of resistance yields a 0.8 lagging power factor (125 kVA total).
• Use capacitive (leading) reactive loading for UPS and data-centre systems where electronic loads and correction capacitors create a leading power factor.

When is a resistive-only load bank enough?

A resistive-only load bank is sufficient when the goal is to exercise the engine, clear wet-stacking and confirm real-power output rather than to certify full kVA. Regular maintenance runs of standby generators, which exist mainly to burn off unburned fuel and carbon deposits that build up during light idling, are commonly performed with resistive load alone because the priority is heat and load on the engine.

Resistive banks are also the right tool for many battery, UPS and photovoltaic tests where the load is genuinely resistive, and for any application specified purely in kW. They are simpler, lighter, cheaper and easier to move than combined units, so where a reactive component is not required they are the pragmatic choice.

The limitation to keep in mind is the one described above: a resistive-only test cannot verify the alternator's reactive capability and will not reach full nameplate kVA on a 0.8 pf machine. If the specification calls for kVA acceptance at a stated power factor, resistive-only is not enough.

• Good for routine exercise runs, wet-stacking removal and any test specified in kW only.
• Appropriate for genuinely resistive loads such as many UPS, battery and PV acceptance tests.
• Cannot certify full kVA at 0.8 pf or validate alternator and AVR reactive performance.

How to choose the right load bank

Start from the acceptance criterion, not the hardware. If the requirement is expressed in kW, a resistive load bank sized to the rated kW (with margin for altitude and temperature derating) will do. If the requirement is expressed in kVA at a power factor, you need reactive capability matched to that power factor and a resistive component large enough to reach the target kW at the same time.

Match the reactive type to the real load. Predominantly motor and transformer loads call for inductive (lagging) reactive loading at 0.8 pf; predominantly electronic, UPS or capacitor-corrected loads call for capacitive (leading) loading. Then confirm the electrical basics: voltage and frequency (for example 400 V/50 Hz or 480 V/60 Hz), phase configuration, connection method, and the standards the unit is built and marked to (IEC, UL and CE as applicable).

Load banks span a wide range of families for different jobs, including AC resistive, combined resistive plus reactive, DC, liquid-cooled, data-centre and server-emulator designs. As an international load bank manufacturer with four decades of group experience, Ashford Energy UK builds and supplies these across all of those categories worldwide; if you are unsure which configuration matches your acceptance test, specifying the target kVA, power factor, voltage and frequency is the quickest way to get the right unit.

• Define the acceptance criterion first (kW versus kVA at a stated power factor), then size the load bank to it.
• Match reactive type to load: inductive/lagging for motors and transformers, capacitive/leading for UPS and electronics.
• Confirm voltage, frequency, phase, connection and relevant standards (IEC, UL, CE) before selecting a unit.

Frequently asked questions