SI Units Explained: Why Base Factors Matter

The SI system defines exactly seven base units: metre (length), kilogram (mass), second (time), ampere (electric current), kelvin (temperature), mole (amount of substance) and candela (luminous intensity). Every other unit — newton, joule, pascal, watt — is a derived combination of these.

Since the 2019 redefinition, all seven are anchored to fixed numerical values of physical constants rather than artefacts. The kilogram is now defined through the Planck constant instead of a platinum cylinder in Sèvres.

A naive converter stores a factor for every unit pair. With 20 units that is 380 factors to maintain, and every one is an opportunity for a typo. A correct converter stores one factor per unit — its value in the SI base unit — and pivots.

value_target = value_source × factor_source ÷ factor_target. Converting 5 miles to nautical miles becomes 5 × 1609.344 ÷ 1852 = 4.3444 — two exact factors, no rounding chain.

Some conversions are exact by definition: 1 inch = 0.0254 m, 1 pound = 0.45359237 kg, 1 calorie (thermochemical) = 4.184 J. Others are measured and carry uncertainty, such as the astronomical unit or the electronvolt.

Using 2.54 cm/inch instead of a rounded 2.5 changes a 100-inch measurement by 4 cm. In machining tolerances that is a scrapped part; in structural steel it is a failed fit-up.

Temperature is not a simple ratio scale. Converting Celsius to Fahrenheit requires an offset as well as a factor: °F = °C × 9/5 + 32. Kelvin is the base unit, and only kelvin and rankine intervals scale multiplicatively.

Gauge pressure behaves the same way relative to absolute pressure — always confirm whether a pressure figure is gauge (barg, psig) or absolute (bara, psia) before converting.

Floating-point arithmetic introduces representation error at roughly 1 part in 10^16. That is irrelevant for engineering but visible if you display 15 digits. Round the output to the significant figures the input justifies — three input digits do not become ten output digits.