What is a load bank test?

A load bank applies a known, controllable electrical load to a generator and dissipates the resulting power as heat. It lets you test the set at or near its rated output even when the building's own connected load is too small to do that, verifying the generator can actually carry the load it is rated for.

Why do generators need a load bank if they run on a schedule?

A routine exercise often runs the set at no or light load. That proves it starts and transfers, but not that it can carry rated load. Many standby generators are oversized relative to their normal load, so routine transfers may run the engine at only a small fraction of its rating — which over time can cause wet-stacking. A load bank applies real load to prove capacity and bring the engine to full operating temperature.

What is wet-stacking?

Wet-stacking is the accumulation of unburned fuel and carbon that builds up when a diesel engine runs at no or light load for prolonged periods and never reaches full operating temperature. The fuel does not burn all the way, and deposits collect in the combustion chambers, exhaust manifold, turbocharger, and exhaust piping, where they can foul injectors and degrade performance over time.

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

A resistive load bank applies real power (kilowatts) at unity power factor by converting electricity to heat, which verifies the engine side and the kW rating. A reactive load bank adds inductive load to pull the power factor down toward the generator's rating (many sets are rated at a power factor of about 0.8), which stresses the alternator and voltage regulator like a realistic mixed load and verifies the full kVA rating.

How often should a generator be load tested?

There is no single universal interval, load percentage, or duration. It depends on the governing standard for the facility, the manufacturer's recommendation, and the authority having jurisdiction (AHJ). A routine light-load exercise does not, as a rule, satisfy a periodic load-test obligation. Confirm the requirement against the governing standard for the specific site and the AHJ.

What should a load test record capture?

Date and time, who performed it, the generator's make, model, serial, and kW rating, the load steps and the duration at each, and the readings taken under load — voltage, frequency, current, oil pressure, coolant temperature, and exhaust observations.

What load percentage should a load bank test run at?

No single percentage applies everywhere. The steps and minimums come from the governing standard as adopted in your jurisdiction, the AHJ, and often the job specification, and they differ by system classification and by why the test is being run. In practice, what matters is the record: the kW and percent of nameplate held at each step, for how long, with the readings taken while it was held.

Load Bank Testing Generators: A Practical Guide

Plenty of facilities run their generator on a schedule, watch it start, hear it run, and check the box. The problem is the same one that hides a dead transfer switch: the engine started, but nobody proved it can carry the load it's rated for. A routine exercise often runs the set at no load or light load. That is not the same as proving capacity.

Most standby generators are oversized relative to the load they normally carry, so routine transfers may run the set at only a small fraction of its rating. A load bank is how you close that gap — and how you keep light-load running from quietly damaging the engine.

What a load bank actually is

A load bank is a device that creates an artificial electrical load and dissipates the resulting power as heat, typically through fan-cooled resistive elements. It applies a known, controllable load that the building's own connected load cannot supply, allowing the generator to be tested at or near its rated output.

That is the whole point. You can't prove a generator carries its rated load if the building only ever asks it for a fraction of that. The load bank supplies the rest, on demand, in controlled steps — and turns the power into heat instead of into the building.

Why light-load running is a problem

Running a diesel generator at no load or light load for prolonged periods prevents the engine from reaching full operating temperature. When a diesel never reaches full operating temperature, fuel does not fully burn, and unburned fuel and carbon accumulate in the combustion chambers, exhaust manifold, turbocharger, and exhaust piping.

This accumulation of unburned fuel and carbon from prolonged light-load running is called wet-stacking. Over time it can foul injectors, affect valves and rings, degrade performance, and carbon up the turbocharger and exhaust.

A load bank test loads the engine enough to reach full operating temperature and burn off accumulated unburned fuel and carbon, which can address early wet-stacking. So a load bank does two jobs at once: it proves the set can carry rated load, and it brings the engine up to temperature to undo the damage that light-load running quietly causes.

The point

The exercise schedule that keeps a generator "ready" is often the same schedule that slowly wet-stacks it. Light-load running and full-load testing are not the same activity, and one does not substitute for the other.

Resistive vs. reactive load banks

Not all load banks load the generator the same way. The distinction matters for what the test actually proves.

Resistive load bank

A resistive load bank applies real power (kilowatts) at unity power factor by converting electricity to heat through resistive elements. Resistive-only testing primarily verifies the engine and the kW rating.

Reactive load bank

A reactive load bank adds inductive load, lowering the power factor toward the generator's rating to stress the alternator and voltage regulator like a realistic mixed load. Many standby generators are rated at a power factor below 1.0, commonly 0.8.

That rating exists because real building loads include motors and transformers, which create inductive load and pull power factor below unity. Resistive plus reactive testing verifies the full kVA rating, including alternator and voltage regulator behavior — closer to what the set actually sees in service.

What a load test verifies

A load test verifies, under load, the generator's voltage, frequency, current, oil pressure, coolant temperature, exhaust observations, and overall stability. The reason these numbers matter is that they tell you whether the set holds together when it's actually working.

Under load, voltage should stay within tolerance, frequency should stay steady as the governor holds engine speed, oil pressure should stay in range, and coolant temperature should reach and hold normal operating temperature.

Failures such as a governor unable to hold frequency, a cooling system unable to shed heat, a voltage regulator allowing sag, or an engine unable to make rated power generally appear only under load, not at no load. That is the entire reason a load test exists: the problems that strand a building during an outage are the ones that only show up when the set is carrying real load.

Exercise vs. load test

These two activities get conflated constantly, and they are not interchangeable.

Routine operational exercise

A routine operational exercise — often weekly, often automatic — starts the set, operates the transfer switch, checks alarms, and runs at no or light load. It proves the system will start and transfer, but it does not prove capacity.

Periodic load test

A periodic load test deliberately applies substantial load for a defined period and records the set's behavior, proving capacity and bringing the engine to operating temperature.

The practical upshot: a routine light-load exercise generally does not satisfy a periodic load-test obligation. The exercise keeps the system warm and confirms it will start; the load test is what proves it can actually do the job.

How often — and how much load

There is no single universal load-test interval, load percentage, or duration. These depend on the governing standard for the facility, the manufacturer's recommendation, and the authority having jurisdiction (AHJ).

Standards governing emergency and standby power generally require periodic load testing in addition to routine exercise, and some frameworks require supplemental load testing for sets that do not reach a minimum load during normal operation. Life-safety and regulated facilities — such as hospitals, data centers, high-rises, and wastewater plants — require periodic load testing because routine exercise does not prove capacity.

Confirm the exact interval, load, and duration against the governing standard for the specific site, the manufacturer's recommendation, and the AHJ. Do not assume one facility's schedule applies to the next.

For a service shop, the practical move is to write a default into the agreement: which load test the contract includes, how often, resistive or resistive plus reactive, and the step profile you run. Take the default from what the governing edition and the AHJ accept for that class of site, quote the reactive option as its own line item, and let the site file override the default where an inspector or a spec demands more. A default in writing keeps the schedule out of dispute; a default in someone's head does not survive an audit.

The record is the deliverable

A load test you ran but cannot prove you ran is worth very little when an auditor or AHJ asks. The record is the deliverable, not the run.

A load test record should capture:

Date and time.
Who performed it.
The generator's make, model, serial, and kW rating.
The load steps and the duration at each.
The readings taken under load — voltage, frequency, current, oil pressure, coolant temperature, and exhaust observations.

If you want a ready-made format, the free load bank test log at the end of this page captures exactly those fields — planned load profile, readings under load, observations, and sign-offs — as a printable PDF and an Excel workbook.

Load testing belongs inside the preventive-maintenance scope for the standby system, alongside battery checks, transfer testing, and inspection of the engine and cooling system. It is part of the system, not a separate errand.

Where PowerOps fits

This is exactly the accountability gap PowerOps is built to close. PowerOps captures a load bank result as a structured record attached to the specific generator, committed by a named authenticated technician with a server-stamped time and held immutable after sign-off.

Field data — readings, photos, serials — is captured in the field rather than retyped at the office, and the canonical data (customer, site, the equipment list, prior findings) auto-flows so the tech isn't re-entering it every visit. The result is a load bank result that functions as a permanent, attributed, time-stamped equipment record an auditor or inspector can review.

Load Bank Testing Generators: What It Is, Why It Matters, and What a Defensible Record Looks Like