Convert a temperature difference (interval) between kelvin, Celsius, Fahrenheit and Rankine degrees. For absolute temperatures, use the Temperature Converter.
Heat • 5 units
All 5 units on the Temperature Interval Converter are defined against the Kelvin interval (K), so each result is one conversion factor away from a single reference rather than the end of a chain of roundings.
The conversion this page is most often opened for is ready before you type anything: 1 K = 1.8 °F. Change either side and every row in the table recalculates with it.
Everything runs in your browser — no account, no uploads, nothing leaves your device.
An interval of one kelvin is 1.8 Fahrenheit degrees, so a 5 K rise is a 9 °F rise and the familiar HVAC design figure of a 20 °F drop across a coil is an 11.1 °C drop. The table also contains an exact identity that is easy to overlook: Rankine and Fahrenheit degrees are the same size, differing only in where each scale places its zero, so an interval converts between them unchanged.
Almost every quantity written with a temperature in its denominator takes these conversions rather than the absolute ones. Thermal transmittance in W/(m²·K), specific heat in J/(kg·K), a resistor's temperature coefficient in ppm/°C, the ΔT across a heat exchanger and heating or cooling degree days are all intervals. The tell is the word 'per': as soon as a temperature appears per something, the offset between scales cancels out and only the ratio of degree sizes survives.
The millikelvin row belongs to stability specifications rather than to everyday work. An oven-controlled oscillator or a precision reference may be held to a few millikelvin, and that figure is a tolerance band — an interval — not a temperature. Converting it into Fahrenheit produces thousandths of a degree in a unit nobody writes, which is a reliable signal that the number should have stayed in SI. Where a specification is this tight, converting it at all usually costs more clarity than it buys.
Quick reference — 1 Kelvin interval (K) is equal to:
| Celsius degree | °C | 1 |
| Fahrenheit degree | °F | 1.800000 |
| Rankine degree | °R | 1.800000 |
| Millikelvin | mK | 1,000 |
5 units of temperature interval, each a fixed multiple of the K. The table spans 1,000:1, from mK (0.001 K) to °C (1 K). Conversion is one multiplication into the base unit and one division out of it, with no lookup table and no approximation.
value_to = value_from × (factor_from ÷ factor_to)1 K = 1 K1 °C = 1 K1 °F = 0.555556 K1 °R = 0.555556 K1 mK = 0.001 Kwhere:
Assumptions: Factors are exact definitions. Full double precision is carried internally and rounding is applied only for display.
Result25 K = 45 °F
This converter handles differences in temperature, not temperatures themselves, and the distinction changes the arithmetic completely. Converting 20 °C to Fahrenheit requires the offset — 68 °F — but converting a rise of 20 °C does not: it is a rise of 36 °F.
Getting this wrong is one of the most common errors in engineering calculation, because the two operations look identical on the page. Any quantity written "per degree" — specific heat, thermal expansion, temperature coefficient of resistance — takes the interval conversion, never the absolute one. A useful check: kelvin and Celsius intervals are always numerically equal, so if a conversion between them changes the number, it is the wrong conversion.