8 Mechanical Engineering Formulas You'll Actually Use

A torque wrench doesn't measure force directly - it measures the force applied at a known handle length, then reports the product as torque, which is why a bolt's tightening spec is always given in torque units (like N·m or lb·ft) rather than force alone.

Two mechanics using different-length wrenches on the same bolt, each applying whatever force feels right by hand, will almost never produce the same actual clamping force - torque specs exist precisely to standardize that outcome regardless of wrench length or grip strength.

Stress and strain are often taught together because stress is the cause and strain is the material's response, but they're not the same kind of quantity: stress (σ = F/A) measures internal force intensity, with units of pressure, while strain is a dimensionless ratio of length change.

A thin rod and a thick rod under the identical applied force experience very different stress, even though the force itself hasn't changed - exactly why a bolt's diameter, not just its material, determines how much load it can safely carry.

Strain is what actually happens to a material's length under stress - the change in length divided by the original length - and because it's a ratio of two lengths, it has no units at all, even though it's routinely reported as a percentage or in microstrain for readability.

A steel cable stretching 2 mm out of an original 2,000 mm length has a strain of 0.001, or 0.1% - a number meaningful on its own, but only half the picture without knowing the stress that produced it, which is exactly what Young's modulus connects it to.

Stress, strain, and Young's modulus form a closed loop: E is defined as the ratio between the two, so knowing any two of the three quantities always determines the third for a material still in its elastic range.

That's also what makes Young's modulus a genuine material property rather than a part-specific number - it doesn't change with a part's size or shape the way stress and strain do, which is why it's tabulated once per material and reused across every part made from it.

Belt speed is the linear speed at any point on a moving belt, and it depends only on the pulley's pitch diameter and how fast that pulley is spinning - a 200 mm pulley turning at 1,750 RPM produces a belt speed of about 18.3 m/s, which is why belt speed and pulley speed are really the same fact expressed two different ways.

Running a belt faster than its rated range doesn't just wear it out sooner - at high speed, centrifugal force pulls the belt away from the pulley face, reducing the grip needed to actually transmit power, so an overdriven belt can end up transmitting less power than the same belt running at its rated speed, despite moving faster.

Gear ratio is set by tooth count, not gear diameter directly - for two meshing gears (which must share the same tooth pitch to mesh at all), it's the number of teeth each one engages per revolution that actually sets the speed ratio, with diameter simply following along as a consequence of tooth count and pitch rather than being the independent variable.

That's also why gear ratios come out as exact fractions in a way belt-and-pulley ratios don't: teeth are discrete, whole-number features, so a real gearset either hits a ratio exactly or it doesn't - there's no equivalent of belt slip quietly shifting the result.

A belt-and-pulley drive uses diameter instead of tooth count to set its ratio - the larger pulley always turns slower and the smaller one faster, in direct proportion to their diameters, the same inverse relationship gears use, just measured a different way.

Unlike a gear pair, a belt drive isn't a rigid mechanical link: normal operating slip and the gap between a pulley's outside diameter and its effective pitch diameter both introduce small real-world deviations from the ratio this formula predicts, which is why belt-driven equipment gets periodically checked against its actual output speed rather than trusted to the nominal ratio indefinitely.

Horsepower didn't originate as an SI or even a purely scientific unit - James Watt coined it in the 1780s to compare his steam engines against the horses they were replacing, defining one horsepower as 33,000 foot-pounds of work per minute, a figure that's stuck around in engineering ever since.

That historical definition is also where the constant 5,252 in the imperial torque-to-horsepower formula comes from (33,000 divided by 2π) - which is why torque and horsepower curves for any engine cross at exactly 5,252 RPM, the one rotational speed at which the two numbers, expressed in their usual units, must be numerically equal by definition.