[email protected] +39 02 8700 4500
Mon-Fri: 8:00 - 18:00 CET Get a Free Sample

Reading the Rating Plate Honestly: Kohler-SDMO D830 vs a Caterpillar 750 kW-Class Set

Industrial diesel · teardown by where the numbers come from

Reading the Rating Plate Honestly: Kohler-SDMO D830 vs a Caterpillar 750 kW-Class Set

Most genset comparisons argue about whose kW is bigger. This one argues about whose kW you can trust — because at the ~660 kW overlap, the D830 and a Cat 750 kW-class set are close enough that the deciding question becomes which rating was measured under conditions that match your install.

Pull the rating plates side by side and the first temptation is to subtract. The Kohler-SDMO D830 is published at 750 kVA prime / 825 kVA standby — roughly 600 kW prime, 660 kW standby at 0.8 power factor. Caterpillar generator's diesel range brackets this from both sides: the C18-class reaches up here and the C32 is published at 830–1000 kW. So a fair fight pairs the D830 against a Cat set rated near 660–750 kW, not against a 1 MW C32. But the rating number itself is the least interesting thing on the plate. What matters is the basis behind it — and that basis is what we tear down here, dimension by dimension.

Dimension 1

Prime vs standby — the same set wears two numbers

Mechanism. A single genset carries multiple published ratings because the rating describes a duty contract, not a fixed capability. Standby permits the higher output but only for the duration of a utility interruption, at a limited average load. Prime permits unlimited hours but at a lower continuous ceiling. Caterpillar states this explicitly: its standby rating is available for the duration of a normal-source interruption "at an average load of 70% of the standby rating." Kohler-SDMO generator publishes the D830 with the same two-number structure — 750 kVA prime, 825 kVA standby.

Worked consequence

If you compare the D830's 825 kVA standby figure against a Cat prime figure, you will "discover" a gap that is an artefact of reading two different contracts. Size a continuous-process load off a standby number and you will overload the set within a season, because the 70%-average ceiling forbids the duty you bought it for. Decision it drives: before you compare two sets, normalise both to the same rating class for your actual duty — if your load runs unlimited hours, compare prime-to-prime (≈600 kW for the D830) and ignore the standby column entirely.

When this reverses: for a genuine emergency set that runs only during outages, the standby column is the honest number, and comparing prime figures would now understate both sets equally.

Dimension 2

The ambient and altitude asterisk

Mechanism. Every published kW sits on top of an unstated reference condition — a reference ambient temperature, altitude, and intake restriction. The engine's real ceiling falls as air thins or warms, because the turbocharger ingests less oxygen mass per stroke and the cooling system rejects less heat into hotter air. Two sets quoting the same kW at different reference conditions are not quoting the same machine.

Worked consequence

Suppose the Cat set is rated at a cool reference ambient and your D830 comparison is read at a different one. On a site at, say, an illustrative 40 °C and moderate altitude, one set might derate a few percent more than the other purely on its reference basis — turning a paper tie into a real 20–30 kW difference at the worst hour. Decision it drives: demand both sets' derate curves and re-rate both to your site ambient and altitude before signing. The set that holds more usable kW at your conditions wins, even if the plate numbers were identical.

When this reverses: in a climate-controlled plant room near sea level, every set runs at or above its reference condition, the derate vanishes, and this dimension stops separating them.

Dimension 3

Heat rejection — the number that isn't on the plate at all

Mechanism. The rating plate shows electrical output; it does not show the thermal load you must remove to sustain it. Total heat to reject is jacket-water heat plus charge-air (after-cooler) heat plus radiator and airflow losses, with alternator losses on top. At ~660 kW, that thermal budget is large, and it is set by combustion and cooling design — not by "power density," which is marketing shorthand that hides the actual airflow you must provide.

Worked consequence

If the Cat set rejects more charge-air heat into its radiator loop, your plant-room ventilation and duct sizing must grow to match — a cost that lands on the building, not the genset quote. Conversely, if the D830's soundproofed enclosure restricts intake airflow to hit its acoustic target, its cooling margin shrinks unless the radiator and fan were sized for that restriction from the start. Decision it drives: ask each vendor for the full heat-rejection table and required cooling airflow, then check it against the room you actually have. A set that "fits" electrically but starves for air will trip on high coolant temperature long before it reaches its plate rating.

When this reverses: an open-skid set in a large, naturally-ventilated hall has so much airflow headroom that the heat-rejection delta becomes academic, and the plate numbers govern again.

Dimension 4

Who measured the control data — APM403 vs EMCP 4

Mechanism. The numbers you live with day to day — voltage, frequency, load, fault history — come from the controller, and the two platforms present them differently. Kohler-SDMO's larger units use the APM403 panel; Caterpillar uses EMCP 4-series boards that consolidate metering and diagnostics into one interface. The provenance question here is not which has more screens, but which exposes the parameter you will actually use to make operating decisions.

Worked consequence

If your operations team needs paralleling, remote start/stop and protective-relay data surfaced cleanly, the controller's reporting depth becomes a recurring cost or saving every shift. A panel that hides load-history behind menus turns each capacity-planning question into a site visit. Decision it drives: have each vendor demonstrate the specific telemetry you depend on — block-load recovery trace, load profile, fault log — on the actual controller, not a brochure screenshot. Buy the platform whose data you will trust without re-instrumenting the set.

When this reverses: if you already standardise on one SCADA layer that ingests Modbus from any panel, the native UI differences flatten out and this dimension carries little weight.

Summary

DimensionWhat it really measuresDecides
Prime vs standbyDuty contract, not capabilityWhich column to compare
Ambient/altitudeReference condition behind kWUsable kW at your site
Heat rejectionAirflow & ventilation costWhether it fits your room
Control provenanceTrust in daily telemetryOperating-decision cost
The closing rule

Never compare two rating plates until both numbers describe the same contract under your conditions. Concretely: normalise both sets to the same rating class for your duty, re-rate both to your site ambient and altitude, and confirm the heat-rejection airflow against the room you have. If after that the two usable-kW figures land within about 5% of each other, the electrical comparison is a tie — and the decision should move entirely to controller telemetry and local service reach, not to whichever plate printed the larger standby number.

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

Leave a Reply