Define It Before Measuring It
Still efficiency here means alcohol recovery: the pure alcohol you collected, divided by the pure alcohol that was in the charge, as a percentage. It is not the same as purity, speed or energy use. A still can be highly efficient and still make rough spirit, or beautifully clean spirit at poor recovery. Measuring recovery isolates one honest question: of the ethanol you made in fermentation, how much reached your jars?
The Four-Measurement Protocol
1. Charge volume. Measure the wash going into the boiler, in litres, as precisely as your vessel allows.
2. Charge strength. Take the wash ABV. The practical route is original and final gravity through the standard Balling-derived calculation; a lab-grade alternative is measuring the density of a small distilled sample. Record the method, since it sets your error bars.
3. Collected volume. Total everything you kept and everything you set aside, foreshots included. Recovery counts all alcohol that left the still, and jar-by-jar collection makes this automatic.
4. Collected strength. Alcoholmeter each jar, correct every reading to 20°C, and record per jar. Alcoholometric corrections follow the OIML tables that every proper conversion tool uses.
Then: recovery = Σ(jar volume × jar ABV) ÷ (charge volume × charge ABV).
A Log Template That Makes It Automatic
The protocol only produces knowledge if it survives more than one run. This is the complete record for a run, one row per jar plus a header block; copy it into a notebook or spreadsheet and the calculation falls out for free.
| Field | Example entry | Why it is there |
|---|---|---|
| Date / run number | 2026-08-12, run 14 | Trends need a timeline |
| Charge volume | 25.0 L | Denominator, half of it |
| Wash OG / FG | 1.062 / 0.998 | Charge ABV via the Balling relationship |
| Charge ABV | 8.4% | Denominator, other half |
| Jar rows: volume, raw ABV, temp, corrected ABV | 0.90 L, 74%, 26°C, 72.2% | The numerator, honestly measured |
| Collection stop point | Stopped at 20% jar ABV | Makes runs comparable |
| Boiler heel estimate | ~2 L left | Explains where missing alcohol went |
| Recovery | 91.8% | The number you are tracking |
After five runs, compute your mean and range. That range is your still's honest signature, and it is worth more than any figure quoted online, including the planning ranges in our own yield calculator, because it belongs to your equipment, your cuts and your habits.
A Worked Example
Say the charge was 25.0 L of wash at 8.0% ABV. Absolute alcohol in: 25.0 × 0.080 = 2.00 L.
The run produced, after temperature correction: 0.15 L at 78%, 0.90 L at 72%, 1.10 L at 65%, 0.60 L at 45% and 0.40 L at 25%. Alcohol out: (0.15×0.78)+(0.90×0.72)+(1.10×0.65)+(0.60×0.45)+(0.40×0.25) = 0.117+0.648+0.715+0.270+0.100 = 1.85 L.
Recovery: 1.85 ÷ 2.00 = 92.5%. The missing 7.5% is alcohol still in the boiler when you shut down, vapour losses, and measurement error. Run the same protocol on your next three runs and you will see your still’s real range rather than a single lucky number.
Publish your numbers to yourself: a simple table per run, charge, jars, recovery, where you stopped collecting. Patterns appear fast, and a sudden drop in recovery is an early warning of leaks or a change in technique.
What Actually Moves the Number
Once you have a baseline, deviations become diagnostic. The levers, roughly in order of effect:
Where you stop collecting. The tails hold real alcohol at ever-lower strength. Chasing them to 10% jar ABV versus stopping at 25% can move total recovery by several points while making the spirit worse. Decide a standard stop point and hold it.
The boiler heel. Shutting down early for time reasons leaves alcohol behind in the pot. A quick heel measurement, volume and a hydrometer reading once cooled, converts mystery loss into an explained line item.
Leaks and vapour losses. A drop in recovery with no change in habits is the classic signature of a vapour leak at a joint or a condenser running too warm. The efficiency log catches this weeks before your nose does.
Charge strength itself. Very low-ABV charges spend proportionally more of their alcohol in the vapour equilibrium and the heel. Stripping thin washes and re-running combined low wines is partly an efficiency play, not just a purity one.
Where the Error Hides
- Temperature correction skipped. Warm spirit reads several points low on an alcoholmeter. Correct every jar or the whole calculation inherits the bias.
- Wash ABV guessed. The denominator is half the calculation. A wash strength estimated by feel makes the result decorative.
- Stopping collection early counted as loss. Deciding not to chase the last tails is a choice, not inefficiency. Note where you stopped so runs compare fairly.
- Mixed units. Litres and percent throughout, or gallons and proof throughout. Never both.
Compare your measured recovery against the planning estimate for your wash and still type.
The measuring pair. An accurate reading needs a proper test cylinder: a 250 mL cylinder sized for an alcoholmeter makes per-jar measurement quick enough that you will actually do it.
Frequently Asked Questions
Measured honestly, most home pot stills recover a large majority of the charge alcohol, with the exact figure depending on how far into tails you collect and how much you leave in the boiler. The useful number is your own, tracked across runs.
Reflux changes purity far more than recovery. Recovery is mostly about how much alcohol you leave behind in the boiler and how far you collect, which applies to both designs.
Yes, for recovery purposes. They contain alcohol that left the still. You discard them for quality and safety, but the efficiency measure counts everything collected.
Different stopping points in tails, different boiler heels, temperature-correction habits and honest measurement noise. Standardise the protocol and the spread narrows quickly.
References
Primary and peer-reviewed sources for the technical claims on this page.
- Hall, M.L., Ph.D. (Los Alamos National Laboratory), Zymurgy, Summer 1995, American Homebrewers Association. Brew By the Numbers: Add Up What’s in Your Beer, Zymurgy Vol. 18, No. 2. Cited for: The Balling-derived relationship between original and final gravity and alcohol content used to compute wash alcohol.
- International Organization of Legal Metrology (OIML). International Recommendation R 22: International Alcoholometric Tables. Cited for: The internationally adopted density-to-alcoholic-strength relationship for ethanol and water mixtures at 20°C.