Grind Size and Particle Distribution

Surface area is what water actually sees. How grind size, fines and grinder design shape extraction.

Water can only dissolve what it touches. Grinding turns a bean with a few square centimetres of surface into thousands of particles with hundreds of times that area, and the size of those particles is the main way you control extraction rate¹.

Why finer extracts more

Extraction happens in two stages: the surface of a particle is washed almost instantly, then water has to diffuse into the particle and dissolved coffee has to diffuse back out¹. Diffusion time scales with the square of distance, so a particle half the size extracts in roughly a quarter of the time; a 2015 multiscale model of coffee extraction, validated against real brews, reproduces exactly this fast-surface, slow-interior behaviour¹. Finer grind means:

more surface area per gram — faster initial extraction;
shorter diffusion paths — the inside of each particle gets extracted too;
in a percolation brew, slower flow — longer contact time on top of the above².

All three push extraction up. That is why "grind finer" is the fix for sour coffee and "grind coarser" is the fix for bitter coffee.

Typical settings by method

Method	Grind	Approx. particle size
Turkish	powder	< 200 µm
Espresso	very fine	200–400 µm
Moka pot	fine	400–600 µm
AeroPress	fine-medium	500–800 µm
Pour over	medium	600–1000 µm
French press / cold brew	coarse	1000–1400 µm

These ranges are indicative practice, compiled from practitioner measurements rather than a standard³⁴; espresso grinds in the low hundreds of microns are consistent with laser-diffraction measurements in the literature⁵⁶. Grinders differ, and roast level, freshness and bean density all shift the sweet spot. The calculator uses a 1–10 scale (1 = finest, 10 = coarsest) as a common language across grinders.

Fines and boulders

No grinder produces a single particle size. Every grind is a distribution, and the shape of that distribution matters as much as the average. Laser-diffraction studies consistently find that distribution to be bimodal: a main mode set by the grind setting plus a second population of fines far below it⁵. A 2026 study at the Technical University of Munich fitted two-component models to grinds from several grinders and found that the grinder type and setting, more than the roast, decided the shape of that distribution⁶.

Fines (particles under ~100 µm) extract almost instantly and then keep giving up bitter compounds. They also clog filters and slow drawdown³. Some fines are essential for body, but excessive fines are what make a cup taste harsh and under-extracted at the same time — the fines over-extract while the boulders under-extract.
Boulders (oversized particles) contribute mostly their surface and stay sour and under-extracted inside¹.

A tighter distribution lets you push extraction higher before bitterness shows up. The figures often quoted — good grinders allowing 22–23 % EY while cheap ones top out around 19–20 % — are practitioner observations⁴³, not a measured law; the mechanism is what the modelling supports.

InteractiveWhat your grinder actually produces

No setting produces one particle size. Change the grinder and setting to see the distribution water meets — and why the fines hump matters more than the average.

Grinder

Setting (1 fine … 10 coarse)6

Median size

603 µm

Fines < 100 µm

13 %

Spread d90 / d10

15.0×

Usable EY ceiling

22–23 %

before bitterness shows

Moderate fines: good body and a forgiving cup, but the wide coarse mode means you'll hit bitterness a point or two of EY earlier than a tight flat-burr grind.

How this is computed: Illustrative two-mode log-normal: the coarse mode's median runs from ~250 µm (setting 1) to ~1300 µm (setting 10); a fines mode centred on ~40 µm holds 8–25 % of the mass depending on burr type, rising at finer settings. Widths and fines shares are chosen to match the qualitative findings that real grinds are bimodal, that conical burrs and blades broaden the coarse mode, and that fines (<100 µm) drive early over-extraction. They are not measurements of any specific grinder. Sources: [5] [6] [2] [8]

Burrs, blades and alignment

Blade grinders chop randomly and produce a wide distribution. Usable for French press; frustrating for anything else³.
Conical burrs crush along a cone. Slightly wider distribution with more fines; produce body and forgiving cups, popular for espresso⁷.
Flat burrs shear between two parallel discs. Tighter distribution, higher clarity, more sensitive to grind adjustments⁷.
Burr alignment matters: even a tenth of a millimetre of wobble widens the distribution noticeably. This is Cup Engineering's summary of practitioner experience⁴, not a measured figure.

Two further findings from the physics literature are worth knowing. Cold beans grind finer and more uniformly: a 2016 study found particle size distributions narrowed as bean temperature dropped from room temperature toward −196 °C, with the effect visible already at freezer temperatures⁵. And grinding generates static that clumps grounds and scatters the distribution; a 2024 study led by Hendon's group showed that a small spray of water on the beans before grinding suppresses that charge, reducing clumping and, for espresso, raising extraction and consistency⁸.

Practical rules

The rules below are Cup Engineering practice, consistent with the sources above but not individually measured.

Change grind in small steps and change one thing at a time.
Dark roasts are more brittle and produce more fines — grind a step coarser than you would for a light roast.
Denser, high-grown beans need a finer setting than soft, low-grown beans to hit the same extraction.
A pour over that stalls or drains in under two minutes is telling you about grind before it tells you about anything else.
Purge the grinder when switching coffees; retained grounds from the last setting will muddy your first cup.