Kelvin to Fahrenheit (K to F)

Kelvin to Fahrenheit is F = K x 9/5 - 459.67. Written the way it computes, that is 9/5 times the Kelvin figure minus 459.67, and the two parts do different jobs: the 9/5 accounts for the two scales having different degree sizes, and the 459.67 accounts for them having different zero points.

That second term is what makes temperature different from every other conversion. Length, mass and volume units all share a zero (zero centimeters is zero inches), so converting one is a single multiplication. Kelvin and Fahrenheit do not, so the conversion is affine rather than scalar and needs an offset as well as a ratio. It is the reason a temperature and a temperature difference convert by different rules: a rise of one degree Kelvin is a rise of 1.8 degrees Fahrenheit, because an interval scales with the ratio and takes no offset at all.

Worked through at a value people look up: 273.15 K x 9/5 - 459.67 = 32 °F.

Going back, Fahrenheit to Kelvin is 5/9 times the Fahrenheit figure plus 255.37222. Note what does not happen here: with a scalar conversion the two factors are reciprocals and knowing one gives you the other, but an affine conversion has two constants and the reverse offset is not the negative of the forward one unless the slope is 1.

For mental arithmetic: Multiply by 1.8 and subtract 460.

The direction that turns up when a datasheet or a physics result in kelvin has to be read by somebody who thinks in Fahrenheit. Both parts of the formula do work here: the 9/5 shrinks the kelvin degree to a Fahrenheit one, and the 459.67 moves the zero from absolute zero up to the Fahrenheit zero. That zero is itself an arbitrary point, roughly the freezing temperature of a brine solution Fahrenheit could reproduce in 1724. The useful landmarks going this way are 255.37 K at 0 F, 273.15 K at 32 F, and 373.15 K at 212 F. If the figure came from a light bulb's color temperature rather than a thermometer, see Kelvin to Celsius: it is not a temperature.

On precision: the 273.15 is exact by definition: the kelvin and the degree Celsius were fixed against each other, so nothing in this conversion is a measurement. What it cannot do is add precision the input never had. A weather reading of 273.15 K is good to about a degree, and reporting 32 °F from it implies a thermometer nobody owns. The converter scales its decimals to the magnitude of the input for that reason.