Three Sites, One Rating: Perkins 4000 vs Caterpillar C15 at 500 kW
The same nameplate behaves differently on three real sites — we prove it case by case · current to 2026-06
500 kW is the ceiling of the Caterpillar generator C15 (320–500 kW) and the floor of the Perkins 4000 series (600–1800 kW reaches down to this neighbourhood in its smaller variants). At that shared rating, "which is better" has no single answer — it has three, one per site. We'll prove the point by working three cases and watching the verdict move.
The dimension that drives all three cases: where the rating sits in each range
A C15 at 500 kW is at the top of its band — little headroom above. A 4000-series unit reaching 500 kW is near the bottom of its band — lots of engine above the duty. That single geometric fact ripples through cooling, fuel, and step load differently depending on what the site demands. Hold the rating fixed; change the site; watch the better choice change.
Case A — Cold-storage warehouse, steady prime, mild climate
Refrigeration compressors cycling around a 350 kW average on a 500 kW set; runs prime, thousands of hours a year; moderate ambient; outdoor enclosure with airflow to spare.
At 350/500 = 70% average load, fuel burn (load × bsfc) is the dominant lifetime cost. Perkins generator markets the 4000 for prime-power fuel economy; a unit with the duty sitting low in its range often runs cool and efficient. A C15 pinned near its 500 kW ceiling has less thermal slack and is working harder for the same output. Over thousands of prime hours, a few percent of bsfc is a freight container of diesel a year. Buying decision: for steady prime, the Perkins 4000 wins Case A on the fuel curve. Verify with the bsfc at 70% load, not 100%.
Case B — Port crane standby, rare runs, heavy block loads
Emergency backup for crane and hoist motors; runs only during grid loss; when it runs, big DOL motors slam on; under ~150 hours a year.
Fuel is irrelevant at 150 h/yr. What matters is whether the set holds frequency when a large motor lands in one step (ISO 8528-5). Both a C15 with EMCP controls and a common-rail Perkins 4000 can be tuned for this — but the engine with more headroom above the rating (the low-in-range 4000) often has torque reserve for the step, while a C15 at its ceiling has less to give. Buying decision: require the warranted single-step kW from both; in Case B the set with the larger warranted step wins — and the extra engine above the duty frequently makes that the Perkins. But confirm it on the actual quote, because a well-governed C15 with a generous alternator can still clear it.
Case C — Hospital wing, tight hot plant room, mixed load
Life-safety standby in a basement plant room: fixed louvre area, ambient creeping toward 42 °C in summer (illustrative), moderate run-hours for testing plus real outages.
Here neither fuel nor pure step load dominates — airflow does. Heat rejection splits across jacket-water, charge-air and radiator, and the room can only carry so much. The deciding number is required cooling airflow at 42 °C for each set. A C15 worked near its ceiling rejects heat hard; a 4000 loafing low in its range may need less airflow for the same 500 kW — or the C15's compact single-bank package may simply fit the small room better. Buying decision: get heat-rejection-to-air at your ambient for both, and confirm physical fit. In Case C the winner is whichever set's cooling demand fits the louvres you already have — and that can go either way, which is the whole point.
What the three cases prove
| Site | Deciding physics | Leans |
|---|---|---|
| A · cold store prime | Part-load bsfc | Perkins 4000 |
| B · port standby, hard steps | ISO 8528-5 step (warranted) | Larger warranted step (often Perkins) |
| C · hot tight room | Cooling airflow & fit | Whichever fits the louvres |
Topology/standards per the cited standards; all product ratings are manufacturer-stated values from the cited datasheets, current to 2026-06; derived/illustrative figures are labelled as such. This is not an independent head-to-head test. Perkins is a brand affiliated with this site; competitor names are used for identification only.
