+49 69 8700 4500 | [email protected]
Download Technical Datasheet
Blog Wednesday 17th of June 2026

Four Half-Truths a Procurement Lead Carried Into a 800 kW Genset Tender — and the Numbers That Corrected Them (Caterpillar C32 vs Perkins 4000)

author avatar
Jane Smith I’m Jane Smith, a senior content writer with over 15 years of experience in the packaging and printing industry. I specialize in writing about the latest trends, technologies, and best practices in packaging design, sustainability, and printing techniques. My goal is to help businesses understand complex printing processes and design solutions that enhance both product packaging and brand visibility.
Myth vs Reality · ~800 kW Industrial Diesel · Every myth is a real tradeoff in disguise

Four Half-Truths a Procurement Lead Carried Into a 800 kW Genset Tender — and the Numbers That Corrected Them (Caterpillar C32 vs Perkins 4000)

Like-for-like at roughly 750–900 kW: a Caterpillar C32 (published 830–1000 kW) against a Perkins 4000-series set (engine family 600–1800 kW). Each myth below is not simply false — it hides a real, quantifiable tradeoff. Naming the tradeoff is what fixes the spec.

A procurement lead specifying standby for a continuous-process coating line walked in with four reasonable-sounding beliefs. None was pure nonsense — each was a half-truth that pointed at a real engineering tradeoff and then drew the wrong conclusion from it. Below, each belief is stated as the buyer held it, then corrected not with an opinion but with the tradeoff it was hiding — and the number that makes the tradeoff visible.

1
The half-truth

"Both are rated ~850 kW, so a kVA of one equals a kVA of the other."

What's true in it. At steady state, a kW is a kW: 850 kW of resistive load draws the same from either set. The belief only fails the moment the load moves.

The tradeoff it hides. Equal nameplate kW does not mean equal transient capability. How big a step each set absorbs without breaching ISO 8528-5 voltage and frequency limits is set by the alternator's excitation ceiling and the engine's torque rise — neither of which appears in the headline kVA. Two 850 kW sets can differ sharply in the step they survive.

The number that corrects it. On a coating line the worst event is a large drive or a vacuum pump restarting across-the-line — a reactive inrush near five-to-six times running current. Express the worst start as a percentage of the set's kVA; if it lands above roughly 30–40%, the two "equal" sets diverge, and the dip-and-recovery curve, not the nameplate, decides which holds the line. Buy action: require that curve for your exact worst start from both the C32 and the Perkins 4000 set before treating their kVA figures as interchangeable.
2
The half-truth

"The more efficient engine will always cost less to run."

What's true in it. Fuel is roughly load times brake-specific fuel consumption (bsfc), so a lower-bsfc engine does burn less per kWh — at the load point where that bsfc was measured.

The tradeoff it hides. bsfc is not a single number; it is a curve. Diesels are thirstiest per kWh at low part-load and best near 70–85%. An engine that wins the efficiency contest at 100% can lose it at 40%. So "more efficient" is meaningless until you say at what load, and your real duty cycle decides which engine is actually cheaper.

Reality — the tradeoff quantifiedIf a standby set runs only a brief monthly exercise plus rare short outages, lifetime fuel is small and efficiency barely moves the total — acquisition cost and service reach dominate, and a well-supported Perkins 4000 set may simply be cheaper to own. If instead the coating line runs the set several hundred prime hours a year averaging, say, 75% load, a couple of percent of bsfc at that point compounds into a real annual diesel line. Buy action: Caterpillar publishes C32 builds tuned for low fuel consumption or low emissions, and Perkins offers mechanical or common-rail variants — choose the efficiency build only when your annual hours clear roughly 300, and demand the fuel table at your load point, not at 100%.
3
The half-truth

"Higher power density means a tighter, cheaper plant room."

What's true in it. A physically smaller set does take less floor — the footprint claim is real on its own terms.

The tradeoff it hides. Density does not reduce the heat the set rejects; it concentrates it. A diesel sheds waste heat through jacket water, the charge-air cooler and radiator airflow, plus alternator losses — and that heat must still leave the room. Pack the same ~850 kW of heat rejection into a smaller enclosure and you have made the airflow and inlet-temperature problem harder, not easier. Rated output only holds if the package can dump that heat at the room's real inlet temperature against the static pressure the ducting imposes.

The number that corrects it. Compare the two sets at the room's measured worst-case inlet — say 45 °C with a partly fouled louvre — not at a 25 °C test cell. The set whose cooling package was specified for that ambient holds rating; the one specified to the test cell derates and, on a long hot outage, trips on high coolant temperature. Buy action: for both the C32 and the Perkins 4000 set, get the derate-versus-ambient curve in writing and size the room to the package, not the package to a room you wish you had. A tighter footprint that chokes airflow is a false economy paid back as a summer trip.
4
The half-truth

"Standby rating is the number I size to — it's the bigger one."

What's true in it. The standby figure is indeed larger, and for a true emergency duty it is the correct one. The error is assuming it always applies.

The tradeoff it hides. Caterpillar generator's standby rating is defined for the duration of a utility interruption at an average load near 70% of that rating — it is not a continuous figure. Perkins generator sets are likewise published at prime and standby ratings, the prime number lower. A duty that must run for hours or days at high load is prime work, and sizing it on the standby number quietly overloads the machine.

Reality — the tradeoff quantifiedThe decisive variable is run-duration, not which number is bigger. If the coating line must keep curing through a multi-hour or multi-day outage, size on the prime rating — a larger frame for the same load: a C32 higher in its 830–1000 kW band, or the appropriate larger Perkins 4000 build. Buy action: set a threshold — if any single outage can plausibly exceed a few hours at over 70% load, specify on prime and ignore the tempting standby figure. This is the most expensive myth to discover late, because it is found at full load on the worst night.
The half-truthThe tradeoff it hidesThe number that settles it
Equal kW = equal setTransient capability isn't in the nameplateWorst start as % of kVA; demand dip curve
Efficient engine always cheaperbsfc is a curve, not a pointAnnual run-hours vs ~300 hr break-even
Density = cheaper roomHeat is concentrated, not reducedDerate at measured inlet, not test cell
Always size to standbyStandby ≠ continuous dutyOutage duration & load vs prime threshold

The 45 °C inlet, 75% load, inrush ratios and the 300-hour and 30–40% step thresholds are illustrative, labelled as such; published power bands, prime/standby ratings, the 70%-standby-load definition and engine/control options are manufacturer-stated.

The thread through all four.

Every myth here is a real tradeoff that got rounded to a slogan. The fix is the same each time: refuse the headline number and ask for the curve behind it — the dip-and-recovery curve for transient, the bsfc curve for fuel, the derate curve for heat, and the prime-versus-standby duration that governs them all. Do that on both the Caterpillar C32 and the Perkins 4000 set and the tender stops being a contest of slogans. Concretely: if outages can exceed a few hours above 70% load, size on prime; if annual hours clear ~300, weigh the low-fuel build; and never let a tighter footprint cost you the airflow that keeps the rating real.

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. Caterpillar is a brand affiliated with this site; competitor names are used for identification only.

Leave a Reply

Your email address will not be published. Required fields are marked *

Recent Articles