Quick answer: A particle analyzer measures how big the pieces of coffee are and how that size is spread out — and because particle size controls how fast coffee gives up its flavor, reading those numbers tells you a lot about what the cup will taste like. There are three common ways to measure it: sieve analysis (stacked screens), laser diffraction (the lab workhorse, standardized by ISO 13320), and dynamic image analysis (a camera that captures size and shape). Each answers a different question. This guide explains what they measure, how to read the numbers (the Dv10 / Dv50 / Dv90 and "span" you'll see on a report), and how the same particle thinking carries from brewed grounds to instant powder to supercritical-CO₂ extraction.
Why particle size decides flavor before you taste anything
Coffee flavor is an extraction problem. Hot water pulls acids, sugars, and aromatic compounds out of coffee solids, and one of the biggest levers on how fast that happens is surface area. Break a coffee bean into smaller pieces and you expose dramatically more surface for water to work on, so, all else equal, smaller particles tend to extract faster than larger ones — though very fine, tightly packed beds can also slow water flow and extract unevenly.
That's why grind size maps so directly onto taste. Push too fine and the tiny particles — the "fines" — extract very fast and can tip into over-extraction: harsh, bitter, drying. Go too coarse and the big particles — the "boulders" — can stay relatively under-extracted in a fixed brew time, leaving under-extraction: sour, thin, hollow. The sweet spot is a grind whose particles extract evenly, without a big fraction racing ahead or lagging behind.
Notice the word distribution. No grinder produces particles of one single size — it produces a range, and the shape of that range matters as much as the average. A grind can have a perfectly reasonable average size and still taste muddy if it also carries a big cloud of fines. Two grinds with the same average but different spreads can taste completely different. That is exactly the thing a particle analyzer is built to show you — and the reason "what grind setting?" is a weaker question than "what does the distribution look like?"
How to read a particle-size report: Dv10, Dv50, Dv90 and span
When a particle analyzer hands back a result, you don't get one number — you get a curve and a few summary statistics. Learn four of them and you can read almost any report.
- Dv50 (the median). Half the sample's volume sits below this size and half above. It's the volume median (not the arithmetic mean), and it's the number most people mean when they talk about how fine or coarse a grind is.
- Dv10. 10% of the volume is below this size. A lower Dv10 means the fine tail reaches down to smaller sizes — the over-extracting, bitterness-prone end.
- Dv90. 90% of the volume is below this size. A higher Dv90 means the coarse tail reaches up to larger sizes — the under-extracting, sour-prone end.
- Span. A measure of how wide the distribution is, calculated as (Dv90 − Dv10) ÷ Dv50. A small span means a tighter, more uniform grind; a large span means a wider spread that tends to extract unevenly. Span is a width metric, so on its own it won't reveal a double-peaked grind — for that you read the full curve.
One subtlety worth knowing: laser-diffraction instruments usually report a volume-weighted distribution, where a few large particles carry a lot of weight because volume scales with the cube of diameter. A number-weighted view of the same sample would make the many tiny fines look far more dominant. Neither is "wrong" — they answer different questions — but it's why a report can say the grind is mostly coarse by volume while a microscope still shows clouds of dust. Knowing which weighting you're looking at keeps you from misreading the cup.
The three ways to actually measure coffee particles
Different tools see particles in different ways. Here's what each one does, and what it's good and bad at.
1. Sieve (mesh) analysis — simple and cheap
A stack of screens with progressively smaller openings is loaded with coffee and shaken; each screen catches the particles too big to pass through, and you weigh what lands on each layer. It's inexpensive, intuitive, and great for a quick grind check. The limits: it reports the mass caught in a handful of size bins rather than a smooth curve, it struggles with very fine, static-prone particles, and it tells you nothing about particle shape.
2. Laser diffraction — the lab workhorse (ISO 13320)
Shine a laser through a dispersed coffee sample and the particles scatter the light — small particles scatter at wide angles, large particles at narrow ones. Software turns that scattering pattern into a full volume-based distribution in seconds, across a huge size range. It's fast and the method is standardized internationally as ISO 13320; it commonly reports the Dv10/Dv50/Dv90 numbers above (those percentile terms apply to any weighted distribution, not only laser diffraction). Its main caveat: laser diffraction models every particle as an "equivalent sphere," and coffee particles are jagged and irregular — so the numbers are excellent for comparison and control, but should be read as a consistent model rather than a literal ruler.
3. Dynamic image analysis — size and shape
A camera photographs thousands of moving particles and software measures each one, adding what diffraction can't: shape. Metrics like circularity and aspect ratio reveal whether particles are blocky or flaky, whether particles are clumping into visible agglomerates, and how uniform the population really is. It's the right tool when morphology — not just size — is the question, which matters a great deal for instant-coffee powders (next section).
Which analyzer should you use?
There's no single "best" instrument — only the right tool for the question you're asking. Use this quick router.
- Quick grind check on a budget → sieve analysis. Good enough to confirm a grinder is in the right ballpark.
- Precise, repeatable distribution for QC or dialing-in → laser diffraction. When sampling and method are controlled, it gives repeatable Dv50 and span values run to run.
- A question about shape, agglomeration, or fines → dynamic image analysis. When "how big" isn't enough and you need "what shape."
Beyond the brew basket: instant powder and SFE feedstock
Particle science doesn't stop at the grinder. Two coffee contexts most drinkers never think about live or die by it — and both are where a brand actually controls quality.
Instant coffee is a particle product through and through. Once the brewed extract is dried into a soluble powder or granule, its particle size and morphology drive how it behaves in your cup: whether it wets, sinks, disperses, and dissolves cleanly, or clumps and leaves a powdery edge. Fine spray-dried particles dissolve fast but can dust and clump; porous freeze-dried or agglomerated granules tend to wet, sink and disperse more readily and resist dusting and clumping; the perceived texture also depends on the formulation and how completely it dissolves. Here a shape tool like image analysis earns its keep, because how a powder reconstitutes is partly a morphology story — though mouthfeel also depends on composition, soluble solids, and how the drink is prepared.
Supercritical-CO₂ (SFE) extraction is the other place size rules. In SFE, pressurized carbon dioxide acts as the extraction solvent, flowing through a bed of milled material to pull out target compounds — often in place of conventional organic solvents such as hexane or ethanol. How finely that feedstock is milled sets the surface area the CO₂ can reach and the distance it has to diffuse: too coarse can slow extraction (less surface area, longer diffusion paths); too fine can reduce the bed's permeability, raise the pressure drop, and encourage compaction or channeling. Dialing particle size is a core part of running the process well. CafeBank's instant 3-in-1 coffee leans on this approach — its plant ingredients are extracted with supercritical CO₂ — and the same particle-size principles apply — though the finished instant powder's behavior is set mainly later, by how the extract is dried and agglomerated. (If you want the extraction method on its own, see our sensory analysis guide for how these process choices show up in the cup.)
How this topic fits into Coffee Sensory
Particle size is the upstream cause of many of the sensory effects covered elsewhere in this series. If a grind's distribution skews fine, you'll often read it downstream as harshness and lost clarity; if it skews coarse, as thin body and sourness. To connect the measurement to the taste vocabulary, pair this guide with what is coffee body, understanding good acidity in coffee, and what creates coffee aroma. Then put it into practice with how to practice tasting coffee flavors.
FAQ
What does a coffee particle analyzer actually measure?
It measures the size of the coffee particles and how that size is distributed across the sample — usually as a volume-based curve summarized by Dv10, Dv50 (the median) and Dv90, plus a "span" that describes how wide the spread is. Some analyzers (dynamic image analysis) also measure particle shape.
What do Dv10, Dv50 and Dv90 mean on a grind report?
They are volume percentiles. Dv50 is the median size (half the sample's volume is smaller, half larger). Dv10 marks the fine end (10% of volume below it) and Dv90 the coarse end (90% below it). A low Dv10 means lots of fines; a high Dv90 means boulders. The gap between them, normalized by Dv50, is the span.
Is laser diffraction or sieve analysis better for coffee?
Neither is universally better. Sieve analysis is cheap and fine for a quick grind check but low-resolution and shape-blind. Laser diffraction (ISO 13320) gives a fast, precise, repeatable full distribution and is the lab workhorse, though it models particles as equivalent spheres. Choose based on whether you need a rough check or trustworthy run-to-run numbers.
Why do fines make coffee taste bitter?
Fines are very small particles with a lot of surface area, so they extract extremely fast and over-extract before the rest of the grind catches up, releasing harsh, bitter compounds. A grind with too many fines (a low Dv10 and wide span) tends to taste bitter and muddy rather than clean.
Does particle size matter for instant coffee?
Yes. In instant coffee the particle and agglomerate size and shape control how the powder wets, sinks, disperses and dissolves, which shapes mouthfeel and how cleanly it mixes. It also matters upstream in supercritical-CO₂ extraction, where feedstock particle size affects how efficiently the target compounds are extracted.
Related Coffee Sensory reading
- Sensory Analysis of Coffee: A Comprehensive Report
- What is Coffee Body?
- Understanding Good Acidity in Coffee
- What Creates Coffee Aroma?
- How to Practice Tasting Coffee Flavors
References & further reading
- International Organization for Standardization. ISO 13320: Particle size analysis — Laser diffraction methods. https://www.iso.org/standard/69111.html
- Malvern Panalytical. A basic guide to particle characterization. malvernpanalytical.com
- Rao, Scott. The Professional Barista's Handbook. (On grind, even extraction, and the sensory impact of fines.)