Extraction yield from TDS

This tool turns a refractometer reading (TDS, total dissolved solids) and the mass of your beverage into extraction yield-often called EY-a snapshot of how much of your coffee dose became solubles in the cup. It is invaluable for QC, R&D, competition prep, and training when used with taste, not instead of it. Save this page as your reference for what the numbers mean, what they cannot see, how to sample consistently, how to talk about EY with staff without turning the bar into a spreadsheet cult, and how to pair measured data with the other tools on this site.

Part 1 - How the estimate is calculated

The shorthand is EY% ≈ (TDS × beverage mass) ÷ dose × 100. TDS is entered the way your device reports it-often as a percentage (e.g., 1.35 for 1.35% dissolved solids). Different refractometers apply slightly different solubility assumptions and calibration curves. For competition or R&D, standardize on one meter model, one calibration fluid, and one export setting. Mixing devices mid-study introduces systematic error that no recipe tweak can fix.

The equation is a simplified mass-balance model. It assumes all solubles in the beverage came from the coffee and that beverage mass approximates beverage volume closely enough for the precision of the model. In practice, coffee is 98–99% water; the assumption holds well for filter and reasonably for espresso, though espresso's higher solids concentration and emulsified oils push the limits slightly.

If you want the full formula written out: EY = ((TDS / 100) × beverage mass) / dose × 100. When TDS is entered as a percentage (say 1.35), the division by 100 converts it to a decimal (0.0135). Multiply by beverage mass (say 220 g) to get dissolved solids mass (2.97 g). Divide by dose (say 15 g) and multiply by 100 to arrive at EY (19.8%). The same algebra works whether you think in grams or ounces, as long as you use the same unit for both beverage mass and dose.

Part 1B - Step-by-step measurement protocol

Consistent measurement is not just about owning a refractometer-it is about following a repeatable sampling protocol every single time. Cutting corners on any step adds noise that makes your data less useful than no data at all. Here is the full procedure, start to finish.

1. Brew the coffee. Complete your brew as normal, recording dose, grind setting, water temperature, and total brew time. Do not alter anything just because you plan to measure-the point is to capture what the recipe actually produces.
2. Stir the beverage thoroughly. Coffee stratifies as it brews-the first liquid out is chemically different from the last. Stir the full batch (server, cup, or shot glass) so that the sample represents the entire extraction, not just one layer.
3. Let the sample cool to 20–25 °C. Most refractometers are calibrated for readings in this range. Hot liquid refracts light differently, and even devices with automatic temperature compensation have limits. Draw a small amount (5–10 mL) into a separate vessel and let it cool naturally, or place the vessel on a cold surface to speed things up. Do not blow on it or add ice-either introduces contamination or dilution.
4. For espresso: syringe-filter the sample. Espresso contains emulsified oils, CO₂ micro- bubbles, and fine particles that scatter light and inflate apparent TDS by 0.1–0.5%. Push the cooled sample through a 0.45 µm syringe filter into a clean vessel. Filter coffee usually does not need this step, but if your brew is visibly turbid (e.g., French press), filter it too.
5. Zero the refractometer with distilled water. Place a few drops of distilled (not tap, not mineral) water on the prism, close the lid, and confirm the reading is 0.00%. If it is not, re-zero. This step corrects for ambient temperature drift and residual film on the prism. Skipping the zero makes every subsequent reading suspect.
6. Pipette the sample onto the prism. Use a pipette or syringe to place the cooled, filtered sample onto the prism. Avoid touching the prism surface with your fingers-skin oils change the reading. Cover the prism fully; an air gap or partial coverage gives unreliable results.
7. Close the lid and wait for the reading to stabilize. Digital devices take 1–3 seconds; optical models require you to look through the eyepiece and wait for the boundary line to settle. Do not rush this. On a digital refractometer the number may flicker for a second before locking-take the locked value.
8. Record the TDS%. Write it down immediately, or enter it into your logging system. Memory is less reliable than you think, especially during a busy service.
9. Repeat with a second sample for verification. Clean the prism with distilled water and a lint-free wipe, re-zero if needed, and take a second reading from a fresh draw of the same beverage. Two readings guard against sampling artifacts. If the two readings differ by more than 0.03%, investigate (stir again, re-filter, re-zero) before trusting either.
10. Average the readings. Take the arithmetic mean of your two (or more) readings as your working TDS value. This averaged figure is what you feed into the EY equation.
11. Log alongside dose, yield, grind, water, and time. A TDS reading without context is a number without a story. Record it in a sheet or app that also captures every brew variable. Next month, when you want to know why that Kenyan tasted incredible, you will be glad you logged everything-not just the refractometer number.

Building this routine into muscle memory takes about a week of deliberate practice. Once it is second nature, the whole process adds less than 90 seconds to a brew cycle.

Part 2 - What extraction yield is (and is not)

Extraction yield is a modeled percentage built from concentration and beverage mass relative to dose. It correlates with how hard you pushed solubles out of the grounds, but it does not know whether those solubles tasted good. Two cups at the same EY can differ wildly in balance if one channelled, one did not, one used softer water, or one sat in a server for 30 minutes.

Use EY to catch drift between shifts, compare brew methods fairly, or bracket a roast-but let cup quality veto the number when they disagree. An EY of 22% that tastes clean and sweet is better than an EY of 20% that tastes muddy-even if 20% is "in the window."

EY vs TDS: two different stories

TDS tells you how concentrated the cup is-how much stuff is dissolved in the liquid. It is primarily controlled by ratio (more coffee per water = higher TDS, all else equal). EY tells you how deeply you extracted the coffee-how much of the available material you pulled out of the grounds. It is primarily controlled by grind, time, temperature, agitation, and water chemistry.

A cup can be strong but under-extracted (high TDS, low EY-tight ratio, coarse grind, short time-sour and intense). A cup can be weak but over-extracted (low TDS, high EY-loose ratio, fine grind, long time-bitter and watery). Separating these two stories is the entire point of measuring both TDS and calculating EY rather than relying on taste alone.

Part 3 - Sampling and measurement hygiene

The number is only as good as the sample. Sloppy sampling turns refractometry into expensive guessing.

• Temperature: Cool hot samples to your refractometer's specified range (usually 20–25 °C). Temperature matters-hot liquids refract differently, and most devices compensate within a narrow window. Sampling immediately off the brew without cooling gives inflated readings.
• Filtration: Filter aggressively per device instructions. Espresso samples almost always need syringe filters-unfiltered espresso contains emulsified oils and micro-fines that scatter light and inflate apparent TDS by 0.1–0.5%. Filter coffee may need settling or paper filtration depending on your device. Fines lie to the prism.
• Sampling point: Stir and sample from a representative point in the cup or server, not only the first drips. Stratification is real: the top of a V60 brew can read differently from the bottom. For batch brew, stir the server before sampling. For espresso, sample the whole shot (stir the cup gently).
• Labeling: Document whether the sample is "beverage as served" or a "lab draw" from a bypass split or a filtered aliquot. If you compare numbers later without knowing the sampling method, the comparison is meaningless.
• Log context: Record dose, yield, time, grind, water batch, and roast date alongside TDS so the number is interpretable next month. A TDS reading without context is a number without a story.

Part 4 - Using EY responsibly on the bar

Taste first. If a cup is clearly sour from under-extraction, fix that story before you chase a target percentage you saw online. The number should confirm what your palate suspects, not override it.

Log everything when you use numbers in training: baristas learn faster when they can connect a percentage to a flavor they remember. "This 1.42% TDS, 21.8% EY cup was the sweet one we all liked" is more powerful than "always aim for 20–22%."

Know the limits of the model: Dissolved solids do not fully describe colloids, oils, or aromatic volatility. Espresso and filter behave differently. Different refractometer models may not intercompare without calibration discipline. Water chemistry affects perceived flavor at the same measured extraction.

Part 5 - Reading EY next to taste

Part 5B - The brewing control chart explained

The classic SCA Brewing Control Chart is a two-axis graph that plots TDS (strength) on the vertical axis against extraction yield on the horizontal axis. It was originally developed in the 1960s by the Coffee Brewing Institute and later adopted by the Specialty Coffee Association. Understanding the chart is one of the most useful things you can learn about coffee measurement, because it turns two abstract percentages into a visual map of where your brew sits.

The four quadrants

The chart divides the brew space into four broad zones based on the combination of strength and extraction:

• Strong + Under-extracted (upper left): High TDS, low EY. The cup is intense and concentrated but has not pulled enough solubles from the grounds. Taste is typically sharp, sour, or aggressively bright. This happens with very tight ratios, coarse grinds, or very short brew times.
• Strong + Over-extracted (upper right): High TDS, high EY. The cup is concentrated and has been extracted deeply. Taste can be bitter, harsh, astringent, or woody. This happens when a tight ratio is combined with fine grinds, high temperatures, or long brew times.
• Weak + Under-extracted (lower left): Low TDS, low EY. The cup is watery and has not extracted enough solubles. Taste is typically sour, hollow, and lacking body. This is the result of loose ratios with coarse grinds or insufficient contact time.
• Weak + Over-extracted (lower right): Low TDS, high EY. The cup is dilute but the grounds have been exhausted. Taste is thin, bitter, papery, or drying. This happens with very open ratios, very fine grinds, and long extraction times-the water keeps dissolving solubles but there is too much liquid to produce any body.

The "ideal" zone

The SCA chart marks a central rectangle-roughly 1.15–1.35% TDS and 18–22% EY for filter-as the preferred zone. Historically, tasting panels scored brews in this region highest for acceptability. The zone is a statistical average, not a commandment. Your roast, origin, water, and brewing style may produce exceptional cups outside the box. The chart is a guide, not a law.

Modern high-extraction techniques

Since the chart was first drawn, the specialty industry has pushed the boundaries of what is achievable. Ultra-fine grinding, SSP and flat-burr grinders with tight particle distributions, bypass brewing, high agitation, and advanced water recipes routinely produce filter brews at 23–26% EY that taste clean and sweet. These brews would fall outside the classic chart's "ideal" rectangle, yet they win competitions and delight customers. The chart's zones remain conceptually useful, but the boundaries of "acceptable" have expanded.

Plotting your own brews mentally

You do not need to print the chart or use special software. After calculating TDS and EY, ask two questions: (1) Is TDS above or below my target strength? (2) Is EY above or below my target extraction? The answers place you in one of the four quadrants. From there, the adjustment logic is straightforward- the taste table in Part 5 maps directly onto the chart's quadrants. If you log your brews with both TDS and EY over time, you will develop an intuitive sense of where each recipe lands without needing the physical diagram.

Part 6 - Benchmarks without cult thinking

Many training programs cite a wide ~18–22% EY band for filter as a rough QC screen. The SCA "gold cup" standard describes a narrower preference zone, but it was derived from specific populations and brew methods and is not a law of nature. Espresso is judged in context of ratio, time, and balance-not a single universal EY finish line.

Some roasters and competition teams push EY well above 22% with high-extraction techniques (very fine grind, high temperature, bypass brewing, or agitation). If it tastes good, the number is not "too high." If it tastes bitter or astringent, the number might confirm why-but taste decides, not the benchmark.

If you compete, publish an internal SOP: cooling curve for samples, filter type, number of readings averaged, discard rules for outliers, and device model. Without this discipline, "my EY was 23%" is not reproducible or comparable.

Part 7 - Water chemistry and EY

Water mineral content directly affects extraction efficiency. Higher mineral content (particularly magnesium and calcium) generally increases extraction at the same grind and time. If you change water recipes-switching from soft to hard water, or changing your mineral water formula-expect EY to shift even with identical brew parameters.

This means comparing EY across shops or locations only makes sense when water chemistry is also documented. Two baristas at the same EY on different water are extracting different compounds in different proportions. Log water batch alongside TDS readings to make historical comparisons meaningful.

Part 7B - Comparing EY across brew methods

Different brew methods operate in fundamentally different extraction and concentration spaces. The table below provides rough reference ranges for each common method. These are starting points, not targets-your coffee, water, grinder, and preferences may place you anywhere within (or outside) these ranges.

These ranges overlap because the brew variables (ratio, grind, time, temperature) are the real drivers-not the label on the brewer. A coarse-grind, open-ratio AeroPress can land in French-press territory, while a fine-grind, tight-ratio pour-over can approach espresso-like concentration. The table is a starting orientation, not a rulebook.

Part 8 - Refractometer care and calibration

• Zero calibration: Zero your device with distilled water before each session. Some devices drift with temperature and humidity. If the zero is wrong, every reading is wrong.
• Cleaning: Clean the prism between samples with distilled water and a lint-free wipe. Coffee oils film the prism and accumulate; a dirty prism reads high.
• Calibration fluid: Use the manufacturer's calibration fluid periodically (weekly for heavy use, monthly for light use) to verify your device is within spec. Document calibration dates.
• Battery and firmware: Low battery can affect measurement accuracy on digital devices. Keep firmware updated if your model supports it.

Part 9 - Common mistakes

• Comparing EY across shops without aligning brew protocols and measurement devices.
• Optimizing only EY while strength (TDS) wanders into unpleasant territory.
• Ignoring grinder maintenance-particle distribution moves EY more than many recipe tweaks.
• Forgetting that water chemistry shifts perceived flavor at the same measured extraction.
• Sampling hot coffee directly onto the prism without cooling or filtering.
• Treating the refractometer number as more authoritative than consistent palate feedback.
• Using one reading from one sample as gospel-average at least two readings from a stirred sample.
• Not zeroing the device between sessions or after ambient temperature changes.

Part 10 - Limits of the model

Real liquids carry dissolved gas, oils, and fines that the equation does not fully describe. The model assumes a simple soluble-solids mass balance. In reality, coffee extraction produces hundreds of compounds with different solubility profiles, volatility, and sensory impact. Two cups at identical EY percentages can taste different because they extracted different types of solubles in different proportions-especially when grind distribution, temperature profile, or agitation differ.

Use EY as a planning and training aid, a QC screen, and a diagnostic tool. It is not a certificate of absolute extraction quality. When numbers and taste disagree, trust taste-then investigate why the numbers did not predict what you experienced.

Part 10B - When to use a refractometer vs when to just taste

A refractometer is a powerful tool, but it is not always the right tool. Knowing when to reach for it-and when to leave it in the drawer-saves time and keeps the focus on cup quality.

When to measure

• QC spot-checks: Pick a random brew each shift and measure it. Over time, this builds a control chart of your bar's consistency. If TDS drifts beyond your target window, investigate before customers notice.
• Competition prep: When you are refining a routine, you need objective data to separate real improvements from palate fatigue. Measure every iteration, log it, and review trends over sessions rather than judging each cup in isolation.
• New coffee onboarding (bracketing a roast): When a new lot arrives, brew three to five recipes across a range of grind settings and ratios. Measure each, taste each, and identify where the sweet spot is. This structured approach finds a good recipe faster than random tasting.
• Resolving team disagreements about extraction: When two baristas disagree about whether a coffee is under-extracted or just bright, a TDS reading provides a shared reference point. It does not settle the aesthetic question-but it grounds the conversation in data rather than opinion.
• Training and calibration: Using a refractometer during cupping or training sessions helps baristas connect numbers to flavors. "This cup at 19.5% EY tastes sour to all of us" is a powerful learning moment that builds shared vocabulary and palate calibration.

When NOT to measure

• Do not use the refractometer as the sole decision-maker. Numbers inform decisions-they do not make them. If TDS says 1.35% but the cup tastes flat and lifeless, the cup is flat and lifeless. Fix the brew, not the spreadsheet.
• Do not measure every single cup during service. It slows workflow, adds stress, and produces more data than you can meaningfully act on. Spot-check systematically instead. If every cup must be measured, your recipe or grinder needs attention-not more readings.
• Do not measure when prep is obviously inconsistent. If your grinder is retention- clogged, your scale is bouncing, or you spilled half the bloom water, fix the obvious problem first. Measuring a broken brew just adds noise to your log.

The refractometer is a diagnostic instrument, like a thermometer for a patient. You use it when you need to confirm or investigate-not to replace clinical judgment.

Part 11 - Frequently asked questions

Is this the same as brew ratio? No-ratio is liquid mass vs dose (a strength frame); EY estimates how much of the coffee solids made it into that liquid (an extraction frame). They answer different questions and should be used together.

Do I need a refractometer to use this page? Yes, for real data. Without a TDS reading, the calculator has nothing to compute. If you do not own a refractometer, use the taste-based dial-in helper and brew ratio tools instead-they work without lab equipment.

Which refractometer should I buy? The VST CoffeeTool and DiFluid R2 Extract are popular in specialty coffee. Any device designed for coffee TDS measurement works; consistency matters more than brand. Pick one and stick with it for internal comparisons.

Can I compare filter EY to espresso EY? With caution. The extraction dynamics are so different that the same percentage means different things sensorially. Use EY for within-method comparisons (filter vs filter, espresso vs espresso) rather than cross-method battles.

What if my EY is above 24%? It is not automatically bad. Some modern techniques push extraction higher with good results. If it tastes good, log it and move on. If it tastes harsh, the number confirms you extracted too aggressively-back off grind or contact time.

How much does a refractometer cost? Entry-level digital coffee refractometers (like the DiFluid R2 Extract) start around $100–150 USD. Mid-range devices are $200–400. The VST CoffeeTool LAB III sits around $700+. Optical (analog) Brix refractometers cost $20–50 but are not suitable for coffee TDS measurement-they lack the resolution and calibration curves needed. Budget for distilled water, syringe filters, and lint-free wipes as ongoing consumables.

How long does a reading take? Once you have a cooled, filtered sample, the actual measurement takes 5–10 seconds on a digital device. The full protocol-cooling, filtering (if espresso), zeroing, measuring twice, cleaning-adds about 60–90 seconds to a brew. With practice it becomes fast enough to do mid-service without disrupting workflow.

Can I use a wine/beer refractometer? Generally no. Wine and beer refractometers measure Brix (sugar content) and use calibration curves designed for sucrose solutions. Coffee solubles are not sugar-they include acids, melanoidins, lipids, and hundreds of other compounds that refract light differently. A Brix refractometer will give you a number, but it will not be an accurate coffee TDS reading. Use a refractometer calibrated specifically for coffee, or at minimum one that reports in refractive index (nD) so you can apply the correct conversion.

What is the difference between Brix and TDS? Brix is a scale that expresses the concentration of a solution as if all dissolved solids were sucrose. TDS (total dissolved solids) is a broader term for all dissolved material in the liquid. In coffee, refractometers read the refractive index and apply a coffee-specific calibration curve to convert it to TDS%. If your device only reports Brix, it is using a sucrose curve-close but not identical to coffee TDS. For casual use, Brix and coffee TDS track similarly; for precision work (competition, R&D), use a device that applies a coffee- specific curve.

My readings are inconsistent-what is wrong? Work through this checklist: (1) Are you zeroing with distilled water before each session? (2) Is the sample cooled to 20–25 °C? (3) For espresso, are you syringe-filtering? (4) Are you covering the prism fully with no air gaps? (5) Is the prism clean-no oily film? (6) Is the battery charged (for digital devices)? (7) Are you stirring the beverage before sampling? If all of these are correct and readings still vary by more than 0.05% between samples from the same brew, the device may need calibration-fluid verification or service.

Should I publish EY on my menu? Almost certainly not. EY is a production metric, not a customer-facing one. Most customers do not know what extraction yield means, and publishing it invites misinterpretation ("higher is better" or "this coffee is only 20% coffee"). If you want to signal quality, describe the flavor. If you want to signal precision, describe your process. Keep EY in the back of house where it belongs-as a tool for the people making the coffee, not a marketing badge.

Part 12 - Workflow pairing

Pair this calculator with the brew ratio tool to separate strength changes from extraction changes, and with the espresso tools when you need dose/yield context for shots. For dialing without lab gear, use taste tags in the dial-in helper instead-this page is for when you choose to add measured data. The water mineral calculator helps you control one of the biggest hidden variables that moves extraction without touching the grinder.