Celsius, Fahrenheit and Kelvin: Formulas, Mental Shortcuts and When Precision Matters
Why Temperature Is Different From Other Units
Most unit conversions are a single multiplication: feet times 0.3048 gives meters. Temperature is not, because the Celsius and Fahrenheit scales put zero in different places. Converting requires a multiplication and an offset, and the order matters. That is also why temperature is the category where people make the most avoidable mistakes.
The Exact Formulas
- Celsius to Fahrenheit: °F = °C × 9/5 + 32 (9/5 = 1.8)
- Fahrenheit to Celsius: °C = (°F − 32) × 5/9
- Celsius to Kelvin: K = °C + 273.15
- Kelvin to Celsius: °C = K − 273.15
- Fahrenheit to Rankine: °R = °F + 459.67
Worked examples
- 25 °C to °F: 25 × 1.8 = 45; 45 + 32 = 77 °F.
- 350 °F to °C: 350 − 32 = 318; 318 × 5/9 = 176.7 °C (recipes round to 180 °C).
- −15 °C to °F: −15 × 1.8 = −27; −27 + 32 = 5 °F.
- −4 °F to °C: −4 − 32 = −36; −36 × 5/9 = −20 °C.
- 300 K to °C: 300 − 273.15 = 26.85 °C, about 80 °F.
With negative values, keep the sign through every step. A common slip is subtracting 32 from −4 and writing 28 instead of −36.
Mental Shortcuts and How Wrong They Are
The popular shortcut for Celsius to Fahrenheit is double it and add 30. The reverse is subtract 30 and halve it. The shortcut uses 2 instead of 1.8 and 30 instead of 32, so it is exact at 10 °C (50 °F) and drifts further away the farther you go from there.
| °C | Exact °F | Double + 30 | Error |
|---|---|---|---|
| −20 | −4 | −10 | 6 °F too low |
| −10 | 14 | 10 | 4 °F too low |
| 0 | 32 | 30 | 2 °F too low |
| 10 | 50 | 50 | exact |
| 20 | 68 | 70 | 2 °F too high |
| 30 | 86 | 90 | 4 °F too high |
| 40 | 104 | 110 | 6 °F too high |
| 100 | 212 | 230 | 18 °F too high |
| 180 | 356 | 390 | 34 °F too high |
For weather, between about −10 °C and 30 °C, the shortcut is within 4 °F, which is good enough to decide on a jacket. For cooking it is not: 180 °C comes out 34 °F too hot. A better mental method for oven temperatures is "double, subtract 10%, add 32": 180 × 2 = 360, minus 36 = 324, plus 32 = 356 °F, which is exact because doubling and removing 10% is the same as multiplying by 1.8.
Another trick for weather: memorize pairs where the digits flip. 16 °C ≈ 61 °F and 28 °C ≈ 82 °F (exact values are 60.8 and 82.4).
Reference Points
| Reference | °C | °F | K |
|---|---|---|---|
| Absolute zero | −273.15 | −459.67 | 0 |
| Where Celsius and Fahrenheit meet | −40 | −40 | 233.15 |
| Water freezes (at standard pressure) | 0 | 32 | 273.15 |
| Typical room temperature | 20 to 22 | 68 to 72 | about 293 to 295 |
| Commonly cited normal body temperature | 37 | 98.6 | 310.15 |
| Water boils (at sea level) | 100 | 212 | 373.15 |
The −40 crossover is a useful check: if your formula does not turn −40 into −40, it is wrong. The 98.6 °F figure is a 19th-century average. Normal body temperature varies by person, time of day and where it is measured, so treat it as a convention rather than a precise threshold. Water's boiling point also falls with altitude, by roughly 1 °C for every 300 m of elevation, which is why high-altitude recipes need adjustments.
Temperature Differences vs Absolute Temperatures
This is the classic error. A temperature change of 10 °C is not 50 °F. Because a change has no starting point, the +32 offset does not apply. Only the scale factor does:
- A difference of 10 °C = 10 × 1.8 = 18 °F (a difference).
- A difference of 10 K = 10 °C, because the kelvin and the degree Celsius are the same size.
- A difference of 9 °F = 5 °C.
Example: a forecast says tomorrow will be 5 °C warmer than today. That is 9 °F warmer, not 41 °F warmer. Climate figures, thermostat setpoint changes, and engineering tolerances ("±2 °C") are all differences. Write them as differences, and convert them with the scale factor only. Most converters, including QuickToolApp's, convert absolute temperatures, so do not put a difference through them.
Kelvin and Rankine
Kelvin is the SI base unit of temperature, and science and engineering calculations that involve ratios of temperatures (gas laws, radiation, efficiency) need an absolute scale where zero means no thermal energy. It is written without a degree sign: 300 K, not 300 °K. Since 2019, the kelvin has been defined by fixing the value of the Boltzmann constant, as described by NIST.
Rankine is the absolute scale that uses Fahrenheit-sized degrees. It still turns up in some US engineering fields such as thermodynamics and HVAC calculations. 0 °R is absolute zero, and 491.67 °R is the freezing point of water. To get from kelvin to Rankine, multiply by 1.8.
Rounding and Precision in Reports
- Match precision to the measurement. A thermometer read to the nearest degree does not support a converted value of 77.36 °F. Report 77 °F.
- Round once, at the end. Converting 176.67 °C back after rounding to 177 gives 350.6 °F, not 350 °F.
- Keep two decimals for kelvin in science contexts (273.15), but it is fine to use 273 for rough work if you say so.
- State whether a value is a temperature or a difference in tables and charts.
These follow the same rules as the unit conversion best practices guide: keep full precision internally and document your rounding.
Converting in QuickToolApp
Open the temperature converter with 37 °C to °F, or type "180 c to f" in the main box. The converter supports Celsius, Fahrenheit and Kelvin; for Rankine, convert to Fahrenheit and add 459.67. Batch Convert handles a column of readings in one pass, and the multi-target table lists the same value in the other scales, which is a quick way to sanity-check a value against the reference points above.