What horsepower actually measures in an engine
Horsepower is a rate of doing work — historically defined as 550 foot-pounds of work per second, a figure James Watt settled on by benchmarking against how much a strong horse could hoist. In a car, that translates to how much power the engine delivers at a given moment, not how much force (that's torque's job) or how fast the car ultimately goes (that depends on gearing, weight, and aerodynamics too).
Torque and horsepower describe two different things about the same engine: torque is the twisting force available right now, while horsepower folds in engine speed — it's torque multiplied by how fast the engine is spinning. A high-revving small engine and a low-revving big engine can produce identical horsepower through very different torque curves.
The formula, and why 5,252 shows up
Horsepower = (Torque in lb-ft × RPM) ÷ 5,252
Worked example: 300 lb-ft of torque at 6,000 RPM gives (300 × 6,000) ÷ 5,252 ≈ 342.7 HP.
That 5,252 isn't an arbitrary rounding — it falls out of the original horsepower definition (33,000 foot-pounds per minute) divided by 2π, the constant that converts rotational speed into linear work. One tidy consequence: on any engine's dyno chart, the torque curve and the horsepower curve always cross at exactly 5,252 RPM, because at that specific engine speed the two numbers become mathematically identical no matter what the engine is.
Power and torque units, side by side
| Power unit | Equals |
|---|---|
| 1 HP | 0.7457 kW |
| 1 PS (metric HP) | 0.9863 HP |
| BHP | same scale as HP, measured at the output shaft |
| Torque unit | Equals |
|---|---|
| 1 N·m | 0.7376 lb-ft |
| 1 kg·m | 7.233 lb-ft |
PS (Pferdestärke, "horse strength") is the metric horsepower European and Japanese manufacturers historically quoted — it's close enough to imperial HP that the two get used almost interchangeably in casual conversation, but not quite identical.
Brake horsepower vs. wheel horsepower
Brake horsepower (BHP) is measured right at the engine's output shaft, before the power ever reaches a transmission. Wheel horsepower — what a chassis dyno actually measures — is always lower, because the transmission, driveshaft, and differential all soak up some power as friction and heat on the way to the tires.
The size of that loss depends on the drivetrain: a manual rear-wheel-drive car commonly loses somewhere around 15-20% between crank and wheels, while an automatic transmission or an all-wheel-drive system with more rotating hardware often loses closer to 20-25%. These are rough, commonly cited ranges rather than fixed numbers — the actual loss varies by vehicle.
Why a dyno number never quite matches the spec sheet
Factory horsepower ratings come from standardized test procedures — SAE J1349 in North America, DIN in parts of Europe — run under fixed temperature, humidity, and barometric pressure assumptions. Run the same engine on a dyno on a hot, humid day at altitude and the raw numbers will read lower purely because the air is thinner, which is exactly why dyno software applies a correction factor to estimate what the engine would have made under standard conditions.
On top of that, a chassis dyno measures wheel horsepower (after drivetrain loss) while manufacturer specs quote crank or brake horsepower (before it) — two different measurement points being compared as if they were the same thing is the single most common source of "my car doesn't make the horsepower it's supposed to" confusion.
For a broader look at horsepower outside the automotive context — motors, pumps, generic machinery — see the Horsepower Calculator.