Run-Hours First: A Single-Variable Funnel for Perkins 4000 vs Caterpillar C32 at 750 kW
A sizing framework for the band where both ranges overlap · current to 2026-06
There are a dozen variables you could weigh between a Perkins 4000 and a Caterpillar C32. Most of them don't actually change the answer. One does — and if you sort by it first, every other comparison falls into place behind it.
That variable is annual run-hours. Not load, not enclosure, not control brand — those are real, but they're downstream. At a shared 750 kW (inside the Perkins 4000 band of 600–1800 kW, at the lower edge of the C32 band of 830–1000 kW), the run-hours number tells you which machine you're really buying before any other spec gets a vote. This framework pours the decision through that single funnel and watches where it comes out.
The funnel: pour everything through run-hours
Picture a vertical filter. At the top you drop your whole requirement. The first and widest screen is one question: how many hours a year will this set actually run under load? Everything below it is shaped by where you land on that screen.
Here fuel is a rounding error. You are buying starting reliability, load acceptance, and instrumentation you can trust the day the utility drops. Caterpillar generator publishes its standby rating as available for the duration of a normal-source interruption at an average 70% load, and pairs the C32 with EMCP control built for mission-critical duty. At 750 kW the C32 also loafs well below its 1000 kW ceiling, so it starts cold and takes block load with thermal margin to spare. The fuel curve never gets a turn at this run level.
Two 750 kW sets, both run ~120 hours a year for tests and the odd outage. The lifetime fuel difference between them is trivial — a few hundred litres, lost in the noise. What isn't trivial is whether the set starts and holds frequency on a 110 kW chiller step at 3 a.m. after sitting idle for six weeks. Decision input: at this run level, throw the bsfc sheets away and compare warranted ISO 8528-5 step class and starting reliability. The funnel has already told you fuel won't pay back any premium — so don't buy a fuel story you'll never cash.
This is the band where the answer genuinely depends on the rest of your site, because fuel has started to matter but hasn't taken over. Now load fraction enters: at 750 kW the Perkins 4000 sits high in its own range and near the productive part of its bsfc curve, while the C32 at the same kW runs a lower load fraction and deeper into its inefficient zone. The wider the gap between your average load and 750 kW, the more this screen tilts toward the engine that sits higher on its own curve.
Run the set 800 hours a year averaging 480 kW. On the Perkins generator that's ~64% load — efficient territory, and Perkins markets the 4000 for prime-power economy. On a C32 sized to 1000 kW, 480 kW is ~48% load. A few percent bsfc gap over 800 hours is real money but not yet decisive; what makes or breaks it is whether you'll drift toward Screen 1 or Screen 3 over the asset's life. Decision input: in the middle band, don't decide on today's hours — decide on the trend. If run-hours are climbing year over year, pre-commit to the Perkins economy case; if they're flat or falling, the C32's standby pedigree holds.
Past this line fuel dominates total cost so thoroughly that almost nothing else can outvote it. Every percent of bsfc advantage at your load point, multiplied by thousands of hours, compounds into a number larger than the entire purchase-price gap. This is precisely the duty Perkins designed the 4000 family's economy tuning for, and at 750 kW the set is operating where its curve is healthiest.
At 3,000 prime hours a year averaging 550 kW (~73% load on the Perkins), a sustained bsfc edge is worth a tank of diesel every week or two (illustrative duty). Over ten years that buries any reasonable difference in sticker price or service-contract cost. Decision input: if you cleared Screen 3, stop comparing controls and pedigree — get each vendor's bsfc at your average kW and let the fuel curve decide. The funnel has concentrated the entire decision into one number, and that number is litres per kWh at your duty point.
Where the funnel deposits you
| Annual run-hours | Dominant cost | Lands on |
|---|---|---|
| Under ~200 h | Reliability & step load | Caterpillar C32 (EMCP, loafing margin) |
| ~200–1,500 h | Mixed — read the trend | Whichever the trend points to |
| Over ~1,500 h | Fuel | Perkins 4000 (economy tuning) |
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.
