Skip to content
CiderHQ
Search

Equipment

Calibration: instruments as things that need maintaining

How do I know my hydrometer, refractometer and meters are still telling the truth?

In short

A measurement is a comparison against a standard. An instrument that has not been compared against one for a season is producing a number, and a number is not a measurement.

Almost all of the checks are free. Distilled water zeroes a refractometer and tests a hydrometer, an ice bath tests a thermometer, and two buffer solutions calibrate a pH meter.

The other half of the work is correction rather than calibration. A hydrometer read at 8 °C in an unheated store is not reading what it would at the temperature its scale was made for, and that correction is either applied every time or it is not applied at all.

Zeroing a refractometer

The routine is a minute long. Clean the prism and the cover plate, put a few drops of distilled or deionised water on the prism, close the plate so the film is even and free of bubbles, wait for the water and the instrument to reach the same temperature, and adjust the screw — or press the zero function — until the boundary sits exactly on zero. Then wipe dry with a soft cloth and go to work.

The reasons it needs doing are unglamorous. The calibration screw moves with handling and vibration. The prism gets scratched and the cover plate loses its flatness. Above all, sugar dries: a film of juice left on the prism at the end of a session dries to a residue that becomes part of every subsequent reading, and no amount of squinting at the scale reveals it. An instrument that has spent a year in a drawer with last autumn’s juice on it will read confidently and wrongly.

Automatic temperature compensation is often misunderstood. It corrects the instrument’s own optical response to temperature; it does not correct for a sample and a prism that are at different temperatures. Juice straight off a press in September and a refractometer that has been in a cold shed are not the same temperature, and the reading taken in the first two seconds is not the reading taken after twenty. Let the drop sit until the number stops moving.

The remaining discipline is sample handling. Two drops from a stirred, representative sample; not the first juice off a spout, not a drip from a hand, not a sample taken from the top of a tank that has stratified. The instrument is capable of more precision than the sampling usually deserves.

The hydrometer’s two corrections

The first is an offset check. Float the hydrometer in distilled water at its calibration temperature; it should read 1.000. If it reads 1.002 it is not necessarily useless — an offset that is known and constant is entirely workable, because fermentation monitoring measures a difference between two readings and a constant offset cancels out of a difference. What matters is knowing the offset and applying it consistently, and retiring the instrument when the offset starts to change, which usually means the paper scale has slipped inside the stem.

The second is temperature. Every hydrometer carries a calibration temperature printed on the scale — commonly 20 °C, sometimes 15 °C, occasionally something else — and a reading taken away from it needs a correction, because the liquid’s density and the glass’s volume both change with temperature. Published correction tables exist for the purpose and the supplier normally provides one. The correction is small within a few degrees of the calibration point and grows quickly beyond it, and in a cider house at 6 °C in December it is routinely larger than the change in gravity the maker is trying to detect between two readings a week apart.

That is the whole argument for treating this as maintenance rather than as a refinement. A hydrometer used carelessly on cold samples produces a series of readings whose scatter swamps the signal, and the maker concludes the ferment is doing something odd when the instrument is simply being read wrong. Taking the sample’s temperature at the same moment as its gravity, every time, costs nothing and makes the whole series usable.

Two physical checks belong here too. A chipped or cracked stem changes the volume the instrument displaces and therefore everything it reads. A greasy stem changes the shape of the meniscus, so a hydrometer that has been handled with buttery fingers reads differently from one that has been washed — which is a good reason to rinse and dry it after every use rather than dropping it back in the tube wet.

The rest of the bench

The record is part of the instrument

A calibration that nobody wrote down has to be assumed rather than known, and the assumption is always optimistic. The workable practice is a single page or a single file per instrument with a date, what it was checked against, what it read, and what was done about it. Instruments themselves get a label with the date of the last check, so that the question can be answered by looking rather than by remembering.

This is what turns a small producer’s data into something worth having. Bench instruments do not need to agree with a laboratory to be useful; they need to agree with themselves across seasons. A run of gravities, pH values and acidities taken on the same instruments, checked the same way, over five years, says a great deal about a cider house. The same readings taken on whatever was to hand say almost nothing.

The last part of maintenance is retirement. A pH probe that will not hold a slope, a hydrometer whose offset is wandering, a refractometer whose prism is scratched across the field of view — each of these will go on producing numbers indefinitely, which is precisely the problem. Instruments are cheap relative to a season, and an instrument that is being argued with has already answered the question.

Precision and accuracy are not the same thing

An instrument with a known, constant offset is precise but not accurate: it measures change perfectly and cannot be compared with anyone else’s figure. An instrument that has been calibrated but is read carelessly is the opposite. Fermentation monitoring needs precision, and a number that will be quoted to a customer, a laboratory or a revenue authority needs accuracy — which is why the practical bench answers one kind of question and an accredited laboratory answers the other.

Also answered on this page

Questions this page covers, so you can tell at a glance whether it is the one you want.

Related

What people ask next

Questions readers ask about the things this page mentions. Each one goes to the section that answers it rather than to a page written to receive the question.

Sources

What this page rests on. Where a source is marked as registered rather than read, CiderHQ is recording that the body is authoritative on the subject without claiming to have worked through the document itself. See our evidence policy for what each state means.

Where to go next