Horsepower from torque and rpm, or estimated from quarter-mile trap speed or elapsed time, with kW and PS alongside.
Torque is rotational force — how hard the crankshaft twists. Power is the rate of doing work, which is torque multiplied by how fast that twisting is happening. The relationship is:
hp = torque (lb-ft) × rpm ÷ 5252
The 5252 is not arbitrary. One horsepower was defined by James Watt as 33,000 foot-pounds per minute. Converting rotational work to linear gives a factor of 2π, and 33,000 ÷ (2π) = 5252.113. That is the whole derivation.
It has a consequence you can see on any dyno chart: the horsepower and torque curves always cross at exactly 5252 rpm, when both are plotted in these units. Below 5252 the torque figure is higher; above it, horsepower is. If a published dyno graph shows them crossing anywhere else, the axes have been scaled independently.
The old line is that torque gets you moving and horsepower keeps you going, which is roughly right but hides the real answer: power is what determines acceleration, because power is the rate at which energy can be added to the vehicle. Torque only matters through the power it produces at a given engine speed.
Gearing is what makes this concrete. A gearbox trades speed for torque, so a high-revving engine with modest torque can multiply it through gearing and out-accelerate a low-revving engine with more torque but less power. This is why a superbike engine producing 80 lb-ft outperforms a truck engine producing 400 — it does so at 12,000 rpm rather than 1,800, and gearing does the rest.
Two long-standing empirical formulas, attributed to Roger Huntington, estimate power from quarter-mile performance:
hp = weight × (mph ÷ 234)³
hp = weight ÷ (ET ÷ 5.825)³
The trap-speed version is the more reliable of the two, because terminal speed depends mostly on power and weight. Elapsed time is heavily influenced by launch quality, traction and gearing, so a car that hooks up well will show a better ET than its power justifies, and a powerful car that spins its tyres will show worse.
Both estimate flywheel horsepower, and both are approximations fitted to typical cars — aerodynamics, drivetrain losses and track conditions all move the answer. Treat them as ballpark figures, not measurements.
| Unit | Equals | Used in |
|---|---|---|
| Mechanical hp | 745.7 W | US, UK |
| Metric hp (PS, ch, pk) | 735.5 W | Europe, Japan, Korea |
| Kilowatt | 1000 W | Official SI, Australia, South Africa |
Metric horsepower is about 1.4% smaller than mechanical, so a car quoted at 300 PS is roughly 296 hp. It is a small difference but it explains why the same car is advertised with slightly different numbers in different markets. German manufacturers quoting PS is the usual source of confusion.
Manufacturers quote power at the flywheel. A chassis dyno measures at the wheels, after the drivetrain has taken its cut. Losses run roughly 10 to 15% for a manual rear-wheel-drive car and 15 to 25% for an automatic or all-wheel-drive one.
So a car rated 300 hp at the crank might read 255 at the wheels, and that is normal rather than a sign of underperformance. Comparing a wheel figure to a manufacturer's flywheel figure without accounting for this is the most common error in interpreting dyno results.
Because one horsepower was defined as 33,000 foot-pounds of work per minute, and converting rotation to linear work introduces a factor of 2π. Dividing 33,000 by 2π gives 5252.113. It follows that horsepower and torque curves always intersect at exactly 5252 rpm when plotted in lb-ft and hp — if a dyno graph shows them crossing anywhere else, the two axes have been scaled independently.
Power determines acceleration, because it is the rate at which energy can be added to the vehicle. Torque matters only through the power it produces at a given engine speed. Gearing is what reconciles the two — a gearbox trades speed for torque, so a high-revving engine with modest torque can out-accelerate a low-revving one with much more. A superbike engine making 80 lb-ft beats a truck engine making 400 for exactly this reason.
Reasonable as ballpark figures, not as measurements. The trap-speed formula is the more reliable of the two, because terminal speed depends mostly on power and weight. Elapsed time is heavily affected by launch quality, traction and gearing — a car that hooks up well shows a better ET than its power justifies. Both estimate flywheel power and both are empirical fits to typical cars, so aerodynamics and drivetrain differences move the answer.
Mechanical horsepower is 745.7 watts; metric horsepower — written PS in Germany, ch in France, pk in the Netherlands — is 735.5 watts, about 1.4% smaller. So 300 PS is roughly 296 hp. The gap is small but it is why the same car appears with slightly different power figures in different markets, and German manufacturers quoting PS is the usual source of the confusion.
Because manufacturers quote power at the flywheel and a chassis dyno measures at the wheels, after drivetrain losses. Expect roughly 10 to 15% loss for a manual rear-wheel-drive car and 15 to 25% for an automatic or all-wheel-drive one. A 300 hp car reading 255 at the wheels is behaving normally. Dyno readings also vary between machines and with temperature, humidity and correction factor, so comparing figures across different dynos is unreliable.
Related: Speed Calculator · Tire Size Calculator · Gas Mileage Calculator · Unit Converter