Gravity readings: what a hydrometer tells you, and what it does not

A hydrometer floats higher in a denser liquid and lower in a thinner one, and the scale reads that as specific gravity: the density of the sample relative to water, which is 1.000 by definition. Sugar dissolved in water raises it. Alcohol, which is less dense than water, lowers it.

That is the whole mechanism, and it explains both what the instrument is good for and its central limitation. A single reading tells you the density of a liquid. It does not tell you what is dissolved in it, whether fermentation has finished, or how much alcohol is present — all of which are inferences from two readings taken at different times.

• Use a trial jar, not the fermenter — Floating the instrument in the vessel risks it hitting the side, gives a reading disturbed by whatever is on the surface, and means retrieving it afterwards. Draw a sample instead — and do not return it, because that is a contamination route into the batch.
• Spin it, and wait — Bubbles clinging to the stem lift it and give a reading that is too high. A gentle spin releases them, and letting it settle until it stops bobbing is what makes two readings comparable.
• Read at the surface, at eye level — The liquid climbs slightly where it meets the glass. The convention is to read at the flat of the surface rather than at the top of that curve, and to be consistent about it — consistency matters more here than which convention you follow.
• Mind the temperature — A hydrometer is calibrated at one temperature, marked on it, and a hot sample reads low. This is the commonest reason a reading looks wrong on brew day, and it is corrected by cooling the sample or by applying the correction the manufacturer gives.
• Check it in water — Plain water at the calibration temperature should read 1.000. If it does not, the offset is real and applies to every reading; a hydrometer that has been dropped is a hydrometer to distrust.

The first is taken before yeast is added, once the liquid is thoroughly mixed — an unmixed sample is the second commonest reason a reading looks impossible. That is the starting gravity, and it is the reference everything else is measured against.

The second is taken at the end. The difference between them is how much dissolved sugar has gone, and the alcohol figure brewers quote is arithmetic on that difference rather than a measurement of alcohol. It is a good estimate and it is not a laboratory result, which is worth remembering before it is written on a label.

The genuinely useful use of the instrument is neither of those. It is the pair of readings taken two or three days apart near the end: if they are the same, fermentation has stopped, and if they are still falling it has not. That is a question that cannot be answered by looking at an airlock, because an airlock that has stopped bubbling means only that the vessel has stopped releasing gas — which happens for several reasons, of which finished fermentation is only one.

• Whether the fermentation is healthy — A stalled fermentation and a finished one look identical to a hydrometer. What distinguishes them is where the reading has stopped relative to what was expected, and that is a judgement about the recipe rather than a property of the number.
• What the alcohol content is, exactly — The standard calculation assumes the only significant density changes are sugar leaving and alcohol arriving. That is close enough for a home brewer and it is an estimate, and the estimate degrades at high strengths.
• Whether anything has gone wrong microbiologically — Density says nothing about what is living in the vessel. A reading that continues to fall long past where it should is a signal worth taking seriously, but the instrument is reporting a symptom rather than a cause.
• Anything at all about a liquid it was not designed for — Hydrometers are made for particular ranges. One intended for finished wine will not read a heavy wort, and using the wrong instrument produces a number that is wrong rather than a refusal.