☕ Key takeaways
- TDS (Total Dissolved Solids) measures coffee concentration as a percentage of soluble mass in water; EY (Extraction Yield) calculates the percentage of the coffee actually extracted — two complementary indicators.
- The SCA Gold Cup table defines target windows: TDS 1.15–1.35% for filter coffee, EY 18–22% for balanced extraction — outside these zones, the cup is either too concentrated or too diluted.
- The EY formula: EY = (TDS × beverage weight in g) / (dry coffee weight in g) — calculable with a coffee refractometer (Atago, DiFluid) and a precision scale.
TDS and EY Extraction Guide: Refractometer, Calculations, Golden Cup
3 key takeaways
- Two numbers summarise the state of a coffee extraction: TDS (Total Dissolved Solids) and EY (Extraction Yield). Popularised by the SCA and practitioners like Scott Rao and Matt…
- TDS measures the concentration of the cup, not the quality of the extraction. A high TDS simply means a stronger, more concentrated drink. A ristretto espresso might have a TDS of…
- A refractometer measures the refractive index of a liquid — how light bends as it passes through. The more concentrated the solution, the greater the bending. Coffee-calibrated…
Two numbers summarise the state of a coffee extraction: TDS (Total Dissolved Solids) and EY (Extraction Yield). Popularised by the SCA and practitioners like Scott Rao and Matt Perger, these metrics allow you to move from subjective tasting impressions to objective analysis. This guide explains what each value measures, how to calculate them, how to use a refractometer, and how to interpret the results to dial in your brewing.
TDS: What Are We Actually Measuring?
TDS (Total Dissolved Solids) expresses the proportion of dissolved matter in the cup relative to the total liquid mass. A TDS of 1.25% means that in 100 g of brewed coffee, 1.25 g is dissolved solids (sugars, acids, emulsified lipids, caffeine, aromatic compounds) and 98.75 g is water.
TDS measures the concentration of the cup, not the quality of the extraction. A high TDS simply means a stronger, more concentrated drink. A ristretto espresso might have a TDS of 8–12%, a filter coffee 1.2%, a cold brew concentrate 2–4%. Each style has its own norms.
For filter coffee, the SCA established a preference zone known as the "Gold Cup Standard": 1.15–1.35% TDS. This emerged from consumer preference studies in the 1950s–60s by E.E. Lockhart, refined since by the SCA. It's a statistical reference, not an absolute truth — many professionals and enthusiasts prefer 1.40–1.50% (stronger) or around 1.05% (lighter, more tea-like).
EY: The Extraction Yield Explained
EY (Extraction Yield) measures what proportion of the dry coffee mass has been extracted and ended up in the cup. An EY of 20% means that from 20 g of ground coffee, 4 g of soluble matter was dissolved into the liquid. The remaining 16 g stays in the puck or grounds (insoluble fibres, cellulose, unextracted proteins).
Why can't you extract 100%? Because a large portion of coffee is insoluble (cellulose, some proteins), and soluble compounds don't all dissolve at the same rate or temperature. Extraction is selective: acids extract first (within the first seconds), sugars next, bitter and astringent compounds last. Under-extraction (EY < 18%) produces sour, salty cups; over-extraction (EY > 22%) produces bitter, astringent ones.
The EY Calculation Formula
The standard formula used by the SCA and most brewing software:
EY (%) = (TDS% × brewed beverage weight in g) / (coffee dose in g × 100)
Practical worked example:
- Dose: 20 g ground coffee
- Brewed volume: 320 ml (weighed: 320 g, density ≈ 1)
- Measured TDS: 1.30%
- EY = (1.30 × 320) / (20 × 100) = 416 / 2000 = 20.8%
20.8% sits comfortably in the target window (18–22%). This cup should be balanced, with good sugar development and controlled acidity.
SCA Gold Cup Table by Brewing Method
| Method | TDS target (%) | EY target (%) | Typical coffee:water ratio |
|---|---|---|---|
| Filter (V60, Chemex, batch brew) | 1.15 – 1.35 | 18 – 22 | 1:15 to 1:17 |
| AeroPress (diluted recipe) | 1.15 – 1.35 | 18 – 22 | 1:13 to 1:16 |
| Espresso (classic double shot) | 8 – 12 | 18 – 22 | 1:2 to 1:2.5 |
| Moka pot | 2.0 – 3.5 | 15 – 20 | 1:7 to 1:10 |
| Cold brew concentrate | 2.0 – 4.0 | 12 – 18 | 1:4 to 1:8 (then diluted) |
| Lungo espresso (1:3) | 5 – 8 | 20 – 25 | 1:3 |
Using a Refractometer: Step-by-Step Protocol
A refractometer measures the refractive index of a liquid — how light bends as it passes through. The more concentrated the solution, the greater the bending. Coffee-calibrated refractometers convert this refractive index directly into TDS.
Recommended Devices
- Atago PAL-COFFEE (BX/RI): professional reference, ±0.01% Brix precision, compact and robust. Around €200–250. Ideal for professional baristas and serious enthusiasts.
- Difluid R2 Extract: newer, more affordable alternative (~€80–100), precision comparable to the Atago according to independent tests (Barista Hustle, Lance Hedrick). Bluetooth app for logging.
- VST Coffee Tools (VST LAB Coffee III): the World Barista Championship reference tool, exceptional precision (~€500). For professionals and competitors.
Measurement Protocol
- Let the sample cool: digital refractometers compensate for temperature, but for stable readings, wait until the coffee reaches 20–25°C. Pour a small amount into a glass or tablespoon.
- Calibrate with distilled water: place 2–3 drops of deionised water on the prism, verify the reading shows 0.00%. Use the calibration function if not.
- Apply the sample: 2–3 drops of cooled coffee on the prism. Avoid air bubbles.
- Read and record: note the displayed TDS%. For reliability, take 2–3 readings and average them.
- Clean between readings: wipe the prism with a clean, damp microfibre cloth.
Interpreting Results and Adjusting
Once you have TDS and EY calculated, four quadrants are possible:
- Low EY + Low TDS: under-extracted and under-concentrated. Grind finer, increase temperature, extend contact time, or increase coffee:water ratio.
- Low EY + High TDS: under-extracted but concentrated (rare). High dose, short extraction. Sour but intense. Adjust ratio or extraction time.
- High EY + Low TDS: over-extracted and diluted. Too much water for the dose. Reduce water volume or increase dose.
- High EY + High TDS: over-extracted and concentrated. Bitter, astringent, heavy. Grind coarser or reduce contact time.
The refractometer doesn't replace tasting — it complements it. Some coffees are delicious at 23% EY (light Ethiopian roasts), others become flat beyond 20%. Data guides, the palate decides.
Calculating EY from TDS: the formula and its assumptions
The relationship between TDS (total dissolved solids, as a percentage) and EY (extraction yield, as a percentage) is mathematical rather than mysterious, but the formula contains assumptions that every coffee measurer should understand before trusting their calculated EY numbers.
The standard EY formula is: EY (%) = (Beverage Weight × TDS%) / Dose Weight. For a filter brew using 15 grams of coffee to produce 225 grams of beverage at a TDS of 1.4%, the calculation is (225 × 0.014) / 15 = 0.21 = 21% EY. This is a mass balance calculation: it estimates what percentage of the original dry coffee mass ended up dissolved in the beverage. The formula assumes that the measured TDS is accurate, that the beverage weight is measured after extraction (not during), and that the dose weight is dry weight measured before grinding.
The limitations of this calculation are important. First, TDS measurement by refractometer has an uncertainty of approximately ±0.02–0.05% under good technique conditions — translating to approximately ±0.2–0.5% EY uncertainty. This is small but not negligible; EY differences of 0.5% are at the threshold of sensory perceptibility for most palates in controlled conditions. Second, the formula assumes a fixed moisture content in the green and roasted coffee — if the coffee has absorbed atmospheric moisture (old coffee in poor storage conditions), the effective dry mass is higher than the scale weight, and the calculated EY will be slightly understated. Third, the formula does not account for water retention in the coffee grounds — spent grounds retain approximately 2.0 grams of water per gram of coffee, meaning the actual beverage yield is less than the water input, and this retained water contains dissolved coffee solids that are not captured in the beverage TDS measurement.
These limitations do not invalidate the measurement — they contextualise its precision. EY calculated from beverage TDS is an estimate, not an exact measurement of extraction completeness. Used consistently with the same technique, equipment, and calculation method, it provides reliable comparative data for understanding how parameter changes affect extraction outcomes. Used as an absolute measurement of extraction completeness, it overstates precision that the underlying method cannot actually deliver.
Beyond the golden cup: when to ignore the targets
The SCA golden cup standards — TDS 1.15–1.55%, EY 18–22% for filter coffee — are population-level averages derived from consumer preference studies. They are useful reference points, but the most interesting and often most delicious coffees frequently fall outside these ranges, and understanding when to trust your palate over the chart is a sign of brewing maturity rather than carelessness.
Very light-roasted coffees with high green density — Ethiopian Washed at 2000+ metres altitude, Rwanda Washed Bourbon — often extract best at EY values above 22%. These coffees are physically denser than lower-altitude arabicas, and their high-altitude concentration of organic acids and sugars means that extraction compounds dissolve more slowly. Stopping at 22% EY with these coffees often leaves sweetness and complexity undeveloped; extending to 23–24% EY may produce a richer, more satisfying cup that a strict reading of the SCA guidelines would classify as "over-extracted." The appropriate response is to taste the 23% cup before declaring it out of bounds.
Natural-processed coffees — particularly Ethiopian naturals with intense fruit fermentation character — sometimes produce their best expression at EY values below 18%. The fermentation-derived esters and higher alcohol compounds in natural-processed coffees are more soluble than the terroir compounds in washed coffees; extraction progresses faster, and the "sweet spot" may occur before the theoretical minimum EY threshold. A natural-processed Guji Ethiopian that tastes overripe and alcoholic at 20% EY but clean, vibrant, and fruity at 17% EY is telling you something about its specific chemistry that the golden cup standard cannot capture.
The practical conclusion is to use TDS and EY measurements as tools for understanding and replicating, not as evaluative standards. A coffee that measures 23.5% EY and tastes extraordinary is an extraordinary coffee at 23.5% EY — not an over-extracted mistake that needs to be corrected to comply with a published standard. Developing the confidence to trust taste over numbers, while using numbers to understand and reproduce the taste, is the mark of a brewer who has moved beyond recipe-following into genuine coffee understanding.
Equipment comparison: refractometer options for different budgets and needs
The market for coffee refractometers has expanded considerably as specialty coffee's precision culture has grown, creating a range of options that vary in accuracy, ease of use, and price. Understanding what you actually need for home or café use prevents both under-buying (a tool too imprecise to provide useful data) and over-buying (laboratory-grade precision at a cost that home brewing cannot justify).
Entry-level optical refractometers — a scaled prism in a handheld device, read by eye — cost €15–40 and provide TDS measurements to ±0.05–0.10% accuracy. This level of accuracy is adequate for understanding gross extraction differences (significantly under- vs. appropriately extracted) but insufficient for the ±0.02% precision needed to detect subtle parameter changes. These instruments also require accurate sample temperature management because their automatic temperature compensation is limited — a drawback for home use where brewing temperature and room temperature are often substantially different.
Digital coffee refractometers — the VST LAB Coffee III being the industry reference — cost €250–350 and provide accuracy to ±0.01% with ATC (automatic temperature compensation) over a 5–50 °C range. This accuracy level is genuinely useful for systematic brewing optimisation and is the standard in competition and café contexts. The VST instrument's companion iOS app automates the EY calculation from inputted dose, yield, and TDS values, and stores brewing records — features that transform individual measurements into a longitudinal brewing database. For home enthusiasts who brew analytically and want reliable data, the VST is the recommended instrument despite its cost.
Mid-range digital refractometers in the €80–150 range — several Chinese-manufactured instruments marketed specifically at the coffee community — provide accuracy around ±0.02–0.03% with reasonable ATC performance. For home use where absolute accuracy matters less than consistent comparative measurement, these instruments offer a practical middle ground. The critical evaluation criterion is measurement-to-measurement reproducibility: measuring the same sample five times should give readings within ±0.02% of each other. A refractometer that produces variable readings for the same sample is measuring instrument noise rather than coffee properties, and no amount of calculation sophistication can rescue data quality from an imprecise instrument.