Reverse Dilution

Two questions come up again and again around proofing. The first is asked after the fact — the water is already in, and you want to know what strength you actually have. The second is asked before it — how much water does this jar need to reach the target? Both are the same physics run in opposite directions, and both need the contraction accounted for, or the answer comes out wrong by more than a hydrometer can hide.

Take the volumes at 20 °C, once everything is at room temperature.
The strength of the spirit before the water went in.
The volume of water you actually added.
Resulting ABV—
Naive ABV (volumes add, wrong)—
Final volume—
Volume lost to contraction—

The same physics, run in both directions

"What did I get?" starts from the volumes you mixed and works out the strength of the blend. Each volume is turned into a mass through the density table — the spirit at its starting strength, the water at 0.998 g/mL — and the ethanol mass is the spirit's mass times its mass fraction of alcohol. Divide the ethanol mass by the total mass and you have the final mass fraction, which the density table converts back to an ABV. The final volume is the total mass divided by the density at that strength. The naive answer just divides the starting alcohol by the sum of the two volumes, which understates the strength, because the denominator it uses — the two volumes added together — is bigger than the volume the mixture actually ends up occupying.

"What do I add?" runs the same balance in reverse. The ethanol alone sets the final volume: the alcohol does not change, so the final volume is the starting volume scaled by the ratio of the two strengths. The water needed is then the final mass minus the starting mass, with both masses read from the density table. This is exactly the calculation the forward Dilution tool performs, and the two agree to the last digit — which is the point of having both directions here: take the water figure from one mode, feed it into the other, and you should land back on the strength you started from. If the round trip does not close, one of the two inputs was mistyped.

The classic case shows why the correction matters. Pour 1 L of water into 1 L of 80% spirit and the naive arithmetic says 2 L at 40%. What you actually get is about 1.96 L at 40.8% — nearly a full point stronger than intended, and 38 mL short of the expected volume. A sharper version of the same trap comes from pharmacy practice: to make 100 mL of 70% alcohol from 96% stock, the naive water figure is 27.08 mL, but adding that much leaves about 97.5 mL at 71.8% instead of the intended 100 mL at 70%. The correct water volume is 29.74 mL, and the shortfall is pure contraction.

As with every dilution, the calculator's answer is only as good as the strength you typed in, so check the finished blend with an alcoholmeter once it has cooled and rested — mixing releases heat, and a warm blend reads larger and weaker than it really is. If the reading disagrees with the calculation by more than a few tenths of a point, the starting strength was probably off: a hydrometer reading taken at the wrong temperature, or a spirit carrying dissolved sugar, will both throw the figures. The temperature correction tool will put a warm or cool reading right before you type it in here.

See also: Dilution & Proofing Calculator, Dilution by Weight, Blending (Pearson Square), Temperature Correction, Using an Alcoholmeter, Using a Hydrometer.