In brief
No grinder makes particles of a single size. At any setting it produces a spread, from dust-like fines to larger chunks called boulders. Grind size distribution describes that spread: how much of the dose sits at each size.
This is why a single number like "600 microns" never fully describes a grind. Honest Coffee Guide puts it plainly: "A single value cannot be an accurate description of every particle." The micron figures on grind charts describe where most of the grounds should fall, not every particle.
For a buyer, distribution is the closest thing to an objective measure of what a grinder does. It is also easy to over-read. The measurements are real, but how they map to flavor is still being worked out.
On this page · 6 sections
How is grind size distribution measured?
The studies cited here used dynamic image analysis, where cameras photograph particles as they fall through a chute. Kaffeemacher and Coffee ad Astra used a Retsch Camsizer X2 for their study of 24 espresso grinders. Imaging could in principle capture particle shape too, though Kaffeemacher focuses on size.
The output is a curve. The horizontal axis is particle size in microns, and the vertical axis is how much of the dose, by volume, sits at each size. Reading a grinder's character is mostly a matter of reading the shape of that curve.
How do you read a distribution curve?
Look for the main peak, the fines peak, and the median. Kaffeemacher defines the median, written x50, as the size where "50% of particles are smaller than x₅₀ and 50% are larger." For espresso, Kaffeemacher gives an example median of roughly 200–300 µm.
| Term | What it tells you |
|---|---|
| Main peak | The size most of the dose clusters around at that setting |
| Fines peak | A smaller hump of particles under about 100 µm |
| x50 (median) | Half the volume is finer than this size, half coarser |
| Peak width | How tightly the main population clusters; narrower is more uniform |
| Extent of boulders | How far the curve trails into large particles |
Kaffeemacher measures peak width as the size range holding 60% of the particles above 100 µm. Coffee ad Astra uses "extent of boulders," the half-width on the coarse side of the main peak, in microns. Larger values mean more boulders.
What do bimodal and unimodal mean?
A bimodal distribution has two peaks, one for fines and one for the main grind. A unimodal one has a single main peak and few fines. Kaffeemacher notes that most espresso grinds are bimodal.
The 2024 Scientific Reports study by Smrke and colleagues found that the balance between the two peaks depends on average grind size, but is also "indicative of the grinding technology." That is why two grinders set for the same median can still differ in how much of the dose sits in the fines peak.
Coffee ad Astra found that, contrary to a popular assumption, "more unimodal grinders appear to also be more uniform on average." Grinders with fewer fines also tended to have tighter main peaks.
Are flat burrs more uniform than conical burrs?
The data leans toward flat burrs, but with a large overlap, and the taste side is contested. In the 24-grinder analysis, conical burrs were "on average less unimodal and less uniform" than flat burrs. The author adds that a specific flat burr can easily measure worse than a specific conical one.
Grinders in that dataset included the Niche Zero, Baratza Forté and Sette 270, Eureka Mignon Specialita and Mahlkönig X54. Burr shape is one input. Burr design, alignment and build all move the curve.
Does a tighter distribution taste better?
It might, and the evidence points both ways. Coffee ad Astra's earlier work argues that a wide distribution mixes "lots of different flavor profiles." It says uneven grinds tend to force a lower average extraction, around 20–21%, to avoid bitterness, while uniform grinds allow 22–24%.
The 24-grinder study is more cautious. It compares distributions with blind tasting only qualitatively, and the best-rated grinders were "somewhat balanced" rather than extreme. A poorly rated grinder, the G-Iota Probarista, measured close to the top performers, which the author suggested might come down to something a distribution cannot show, such as clumping or heating of the grounds.
The author's own summary is the fairest one: "we still need to map these parameters out in terms of how they affect taste, and that will not be easy."
How should this change which grinder you buy?
Use distribution data to rule out grinders that are clearly uneven, then decide on fit, workflow and price. A curve is measured at one setting, with one coffee, so it does not tell you how the grinder behaves across your whole range of brew methods.
Two grinders with similar curves can still need very different dial numbers to reach the same grind. That is a separate problem, which BrewMark's grind converter handles by mapping each grinder onto a shared scale. For the micron ranges per brew method, see the grind size chart. When you are down to two models, compare them side by side.
Sources
- Kaffeemacher: Coffee particle distribution decoded: x50, main peak and more
- Coffee ad Astra: What I learned from analyzing 300 particle size distributions for 24 espresso grinders
- Coffee ad Astra: The dynamics of coffee extraction
- Smrke, Eiermann and Yeretzian (2024), The role of fines in espresso extraction dynamics, Scientific Reports
- Honest Coffee Guide: Coffee grind size chart
By Jeff Gluck · Checked Oct 21, 2026