Celsius to Fahrenheit Calculator

What separates the two scales

Celsius and Fahrenheit disagree about two things at once, and that is the whole reason conversion needs two operations instead of one. They disagree about where zero sits, and they disagree about how big a degree is.

Celsius puts zero at the freezing point of water and 100 at its boiling point under one standard atmosphere, so a Celsius degree is one hundredth of that interval. Fahrenheit places the same two points at 32 and 212, which stretches the same interval across 180 degrees. Divide 180 by 100 and you get 1.8, or 9/5: a Fahrenheit degree is five ninths the size of a Celsius degree. That ratio is why a 10 °C swing in a forecast is an 18 °F swing, and why a 1 °C rise in a fever is nearly two full Fahrenheit degrees.

The offset is the second half. Because Celsius zero lands at Fahrenheit 32, you scale first and then shift: multiply by 1.8, add 32. Reverse the operations in reverse order to come back — subtract 32, then multiply by 5/9. Getting that order wrong is the single most common error, and it is easy to catch: any conversion you can check against 0 °C = 32 °F and 100 °C = 212 °F is either right or obviously wrong.

Kelvin and Rankine remove the offset problem entirely by starting at absolute zero, the temperature at which the thermodynamic energy available for extraction as work is zero. Kelvin keeps the Celsius degree size, so K = °C + 273.15. Rankine keeps the Fahrenheit degree size, so °R = °F + 459.67. Kelvin has been an SI base unit since 1954 and, since the 2019 revision of the SI, is defined by fixing the Boltzmann constant rather than by a property of water.

Why the formula has the shape it has

Any two linear temperature scales are related by y = mx + b, and you only need two shared reference points to pin down m and b. Take the ice point and the steam point.

At the ice point, 0 °C corresponds to 32 °F, so b = 32 immediately. At the steam point, 100 °C corresponds to 212 °F, so 212 = 100m + 32, giving m = 180/100 = 1.8. That is the entire derivation: F = 1.8C + 32. Nothing about it is arbitrary once you accept the two anchor points.

The crossing point falls out of the same algebra. Set F = C and solve: C = 1.8C + 32, so −0.8C = 32 and C = −40. That is why −40 °C and −40 °F are the same physical temperature, and it is the only value for which that is true.

One subtlety trips up engineers rather than travellers: a temperature difference converts differently from a temperature reading. A difference of 10 °C is a difference of 18 °F, not 50 °F, because the offset cancels when you subtract two readings. Whenever a specification says "a rise of" or "a delta of", apply the 1.8 factor alone and leave the 32 out. This matters in heat-transfer work, where you will also meet the same distinction in pressure units on our psi to bar calculator, which separates gauge readings from absolute ones for exactly the same reason.

Modern national laboratories do not realise these scales from ice and steam any more. Practical thermometry follows the International Temperature Scale of 1990 (ITS-90), which defines a set of fixed points — the triple point of hydrogen, the freezing point of silver, and so on — and interpolating instruments between them. ITS-90 changes nothing about the arithmetic here; it changes how accurately a laboratory can say what the true temperature is in the first place.

Neither scale is defined by water any more, which is why those two anchor points are descriptions rather than definitions. Since 2019 the SI, as published in the BIPM brochure, defines the kelvin by fixing the Boltzmann constant at 1.380649 × 10⁻²³ J/K, and Celsius follows from it by the exact offset t/°C = T/K − 273.15. Practical thermometers are calibrated against ITS-90, the International Temperature Scale of 1990, on which the realised boiling point of water at one standard atmosphere is 99.974 °C rather than a round 100. None of that disturbs the arithmetic on this page — the 1.8 ratio and the 32-degree offset are exact by definition — but it is why a calibration certificate cites ITS-90 and not a kettle.

How to read the number you get

Context decides how much precision you need, and the honest answer is usually less than the calculator gives you.

Weather. One Celsius degree is the smallest difference a forecast meaningfully resolves, so quoting a converted air temperature to two decimals is false precision. Round to whole degrees. The bands worth memorising are 0 °C = 32 °F (ice), 10 °C = 50 °F (cool), 20 °C = 68 °F (mild) and 30 °C = 86 °F (hot).

Clinical. Here tenths matter. The conventional normal of 37.0 °C is 98.6 °F, and the usual fever threshold of 38.0 °C is 100.4 °F. Notice that a rise of exactly 1 °C moves the Fahrenheit reading by 1.8 °F, so a Fahrenheit thermometer reading in whole degrees is coarser than a Celsius one reading in tenths.

Cooking. Recipes round aggressively, and you should follow the rounding rather than the arithmetic. 180 °C converts to 356 °F but every English-language recipe calls it 350 °F; 200 °C converts to 392 °F and is written as 400 °F. Ovens rarely hold better than ±10 °F anyway, so the rounding costs you nothing. Our oven temperature conversion calculator handles gas marks and the conventional rounded pairs, and the convection oven conversion calculator applies the fan-oven reduction on top. Where precision genuinely does matter in the kitchen is low-temperature cooking, which is why the sous vide cooking time calculator works in whole-tenth Celsius setpoints.

Engineering. Use kelvin or Rankine. Any formula containing a temperature ratio, a gas law, a radiation term or an efficiency limit needs an absolute scale, because a ratio of Celsius values is meaningless — 20 °C is not twice as hot as 10 °C in any physical sense, whereas 293.15 K genuinely carries 1.035 times the thermodynamic temperature of 283.15 K.

Reference points across all four scales

Ice and steam points are quoted at one standard atmosphere (101.325 kPa). At altitude water boils lower; that is why canning and baking instructions change with elevation.

Mistakes that produce a wrong temperature

• Reversing the operation order. Going Fahrenheit to Celsius, subtract 32 before multiplying by 5/9. Multiplying first gives an answer roughly 18 degrees too low.
• Converting a difference like a reading. A 5 °C temperature rise is a 9 °F rise, not 41 °F. Drop the 32 whenever the number describes a change, a tolerance band or a gradient.
• Using Celsius in a thermodynamic formula. Gas laws, radiation terms and Carnot efficiency all need kelvin or Rankine. Substituting Celsius silently produces a nonsense ratio.
• Quoting more precision than the instrument has. A domestic oven thermostat and a household weather station are not accurate to a tenth of a degree, so a converted value with two decimals overstates what you know.
• Trusting the ‘double it and add 30’ shortcut outside its range. It is within about two degrees near room temperature and drifts badly at oven or freezer temperatures.
• Writing ‘°K’. Kelvin takes no degree sign and is never pluralised as ‘degrees kelvin’ in SI style; 295.15 K is simply ‘295.15 kelvin’.

Where each scale is still used, and why

Celsius is the civil standard almost everywhere and the working scale of science outside thermodynamics. Fahrenheit survives in general use in the United States and a handful of territories, and it survives for a defensible reason in weather reporting: its finer degree means whole-number forecasts carry slightly more resolution about how the day will feel.

Kelvin is the SI base unit and the only scale that can appear in a ratio. Since 20 May 2019 it has been defined by fixing the Boltzmann constant at exactly 1.380649 × 10⁻²³ J K⁻¹, which detached the definition from the triple point of water and from any particular sample of water at all.

Rankine persists in US mechanical and aerospace engineering, where steam tables, psychrometric charts and gas-turbine cycle calculations are still worked in British units. If you are reading a chart whose entropy column is in Btu/(lb·°R), you are in Rankine territory, and the energy figures on that same chart convert cleanly with our BTU to kWh calculator.

Two historical scales occasionally appear and are worth recognising rather than converting by guesswork. Réaumur, with 0 at ice and 80 at steam, still surfaces in old European confectionery and cheese-making texts. Delisle, running backwards from 0 at steam, appears in eighteenth-century Russian records. Neither is in current use, and neither should be assumed when a document simply says ‘degrees’.

Finally, remember that a temperature reading is only as good as where the sensor sits. An oven thermometer on the middle shelf, a meat probe in the thickest part of the joint and a thermostat on the wall all report genuinely different numbers about the same appliance, and no amount of conversion arithmetic fixes a badly placed probe. If you are baking at altitude, the boiling point itself moves, which the high altitude baking adjustment calculator accounts for directly.

Key terms

Absolute zero

The lower limit of thermodynamic temperature, 0 K = −273.15 °C = −459.67 °F. It is a limit, not a reachable state.

Ice point

The equilibrium temperature of pure water and ice at one standard atmosphere: 0 °C, 32 °F. Not identical to the triple point, which is 0.01 °C.

ITS-90

The International Temperature Scale of 1990, the practical realisation of thermodynamic temperature through defined fixed points and interpolating thermometers.

Temperature difference

A change or interval, converted with the 9/5 degree-size ratio alone. The 32-degree offset applies only to readings.