Four standards use the same numbers and mean different things by them. A stone marked #1000 can carry particles of 4.5 micrometres or 18.3, depending on which standard the maker used. The listing rarely says which.
| Grit | FEPA-F | ANSI | JIS R 6001 | FEPA-P |
|---|---|---|---|---|
| #400 | 17.3 ±1 | 22.5 | 40 | 35.0 ±1.5 |
| #500 | 12.8 ±1 | 18 | 34 | 30.2 ±1.5 |
| #600 | 9.3 ±1 | 14 | 29–30 | 25.8 ±1 |
| #800 | 6.5 ±1 | 12 | 20 | 21.8 ±1 |
| #1000 | 4.5 ±0.8 | 7.8 | 14–16 | 18.3 ±1 |
| #1200 | 3.0 ±0.5 | 5.5 | 12–13 | 15.3 ±1 |
| #1500 | 2.0 ±0.4 | — | 10 | 12.6 ±1 |
| #2000 | 1.2 ±0.3 | — | 8–9 | 10.3 ±0.8 |
| #2500 | — | — | — | 8.4 ±0.5 |
| #3000 | — | — | 5 | 7 |
| #4000 | — | — | 3 | — |
| #5000 | — | — | — | 5 |
| #6000 | — | — | 2 | — |
| #8000 | — | — | 1–1.2 | — |
Reuse this. The table is CC BY 4.0 — take it, with attribution. Download as CSV, or cite the page:
“What a whetstone grit number actually measures”, Kogei.
https://kogei.pages.dev/grit
If you are rebuilding this table for your own readers, read the limits at the bottom of this page first. The figures are not from the standards themselves.
Open the number printed on your stone. If the listing named a standard, read that row of the table above instead — most listings do not.
The coarsest reading is 2.3 times the finest.
Nothing on the stone tells you which standard was used. Until the maker names one, #400 is not a figure you can compare against another maker's #400.
The coarsest reading is 2.7 times the finest.
Nothing on the stone tells you which standard was used. Until the maker names one, #500 is not a figure you can compare against another maker's #500.
The coarsest reading is 3.2 times the finest.
Nothing on the stone tells you which standard was used. Until the maker names one, #600 is not a figure you can compare against another maker's #600.
The coarsest reading is 3.4 times the finest.
Nothing on the stone tells you which standard was used. Until the maker names one, #800 is not a figure you can compare against another maker's #800.
The coarsest reading is 4.1 times the finest.
Nothing on the stone tells you which standard was used. Until the maker names one, #1000 is not a figure you can compare against another maker's #1000.
The coarsest reading is 5.1 times the finest.
Nothing on the stone tells you which standard was used. Until the maker names one, #1200 is not a figure you can compare against another maker's #1200.
The coarsest reading is 6.3 times the finest.
Nothing on the stone tells you which standard was used. Until the maker names one, #1500 is not a figure you can compare against another maker's #1500.
The coarsest reading is 8.6 times the finest.
Nothing on the stone tells you which standard was used. Until the maker names one, #2000 is not a figure you can compare against another maker's #2000.
Nothing on the stone tells you which standard was used. Until the maker names one, #2500 is not a figure you can compare against another maker's #2500.
Nothing on the stone tells you which standard was used. Until the maker names one, #3000 is not a figure you can compare against another maker's #3000.
Nothing on the stone tells you which standard was used. Until the maker names one, #4000 is not a figure you can compare against another maker's #4000.
Nothing on the stone tells you which standard was used. Until the maker names one, #5000 is not a figure you can compare against another maker's #5000.
Nothing on the stone tells you which standard was used. Until the maker names one, #6000 is not a figure you can compare against another maker's #6000.
Nothing on the stone tells you which standard was used. Until the maker names one, #8000 is not a figure you can compare against another maker's #8000.
FEPA separates grit for coated abrasives from grit for bonded abrasives. Sharpening stones fall under the second, FEPA-F. Sandpaper uses FEPA-P. A maker quoting a number without saying which one has told you almost nothing, and the two scales are furthest apart exactly where sharpening stones live.
JIS divides the same way FEPA does. JIS R 6001 covers bonded abrasives — grinding wheels and sharpening stones — and states that it does not apply to coated abrasives. JIS R 6010 is the coated one: sandpaper. A stone is a bonded abrasive, so R 6001 is the series that should govern the number stamped on it.
Two things stop that from settling the question. Japanese makers are reported to label with the
abrasive-paper scale as often as the bonded one, and they do not print which. And the current
bonded standard does not use bare numbers at all: JIS R 6001-2:2017 designates microgrits
F230 to F2000, so a stone marked simply “#1000” is not quoting
it directly either way.
This is the same problem as FEPA-F against FEPA-P, one layer further in. The number on the stone is not wrong. It is unsourced — and no one selling you the stone is obliged to source it.
Charts that place an ANSI number beside a JIS number are lining up the nearest numbers, not equal particle sizes. Reading across a row and treating the result as a conversion is the most common way these charts are misused.
Two stones with the same particle size can cut differently. The spread of sizes around the average matters, and so does the shape of the particles, how much of each the binder exposes, whether they fracture under pressure, their hardness, and what they are made of. A tight distribution is what a good stone is selling.
Japanese natural stones are graded as ara-to, naka-to and shiage-to. The classes are broad and do not convert to a grit number.
The figures come from published conversion tables from European abrasives suppliers, cross-checked against Japanese sharpening references. The two European tables agree exactly, which may mean they share a common origin rather than confirming each other independently. The JIS column also matches Japanese sources to within rounding, which makes it the best-supported column here. The JIS and FEPA standards themselves are paid documents and have not been read directly. Ranges are given wherever sources round differently. Figures for diamond and CBN abrasives follow a different standard, JIS B 4130, and are deliberately excluded.
One limit is worth stating on its own, because it is about the column and not the figures. The JIS column is labelled R 6001 because that is the bonded-abrasive series, and a stone is a bonded abrasive. Whether any given maker actually grades to it is a separate question this page cannot answer: makers are reported to use the coated-abrasive series too, they do not print which, and the current bonded standard designates microgrits F230 to F2000 rather than as bare numbers. Read the column as what published tables label JIS, not as a claim about which series your stone was graded under.