Rock Tumbling Grit Schedule
Every grit step in a working schedule is a 4x drop. The step into polish is 19x, which is where dull stones come from.
Every step is a 4× drop, except the last one
Each stage has to remove the scratches the last one left, and scratch depth tracks the size of the abrasive particle. So what decides whether a schedule works is not the grit numbers but the ratio between them. In the standard four-stage run:
| Step | Particle size | Drop |
|---|---|---|
| 60/90 grit (coarse) → 220 grit (medium) | 250 µm → 63 µm | 4.0× |
| 220 grit (medium) → 500 grit (fine) | 63 µm → 19 µm | 3.3× |
| 500 grit (fine) → Aluminium oxide polish | 19 µm → 1 µm | 19.0× |
The grit-to-grit steps top out at 4.0×. The step into polish is 19× — five times bigger than anything else in the run. That is where "the stones came out dull" comes from: polish is being asked to remove scratches far deeper than any earlier stage ever had to.
Why polish gets a different rule
Polishing is not grinding. A polish works partly by flow and chemical action on the surface rather than purely by cutting, so it tolerates a bigger step than an abrasive stage does — which is why this page allows up to 12× into polish but only 4× between grits. One threshold cannot cover both operations, and using the grinding limit everywhere would flag the schedule this page itself recommends.
Adding a pre-polish stage at 1000 grit takes the worst step from 19× down to 10×, and the new step it introduces is only 1.9×. One extra week, and it is the single most effective change most people can make to a schedule — which is why it is on by default here rather than offered as an extra.
The shortcuts, measured
| Shortcut | Drop | |
|---|---|---|
| Skip the medium stage — 60 straight to 500 | 13× | Three times past the limit. |
| Skip the fine stage — 220 straight to polish | 63× | Nothing can recover this. |
| Coarse straight to polish | 250× | A month wasted to find out. |
The cruelty of all three is that the batch looks fine at every stage until the last one. A surface that is uniformly matte after the fine stage is what you want; a surface that is matte in some places and shiny in others means an earlier stage was cut short, and the time to go back is then rather than after the polish.
Grit sizes
Nominal mid-range particle size. Grading gives a distribution rather than one exact figure, which is the right precision for comparing one stage against the next.
| Abrasive | Particle size |
|---|---|
| 46 grit | 350 µm |
| 60/90 grit (coarse) | 250 µm |
| 80 grit | 180 µm |
| 100 grit | 125 µm |
| 120 grit | 102 µm |
| 150 grit | 82 µm |
| 220 grit (medium) | 63 µm |
| 320 grit | 36 µm |
| 400 grit | 23 µm |
| 500 grit (fine) | 19 µm |
| 600 grit | 16 µm |
| 800 grit | 12 µm |
| 1000 grit (pre-polish) | 10 µm |
| 1200 grit | 9 µm |
| Cerium oxide polish | 1.5 µm |
| Aluminium oxide polish | 1 µm |
| Tin oxide polish | 1 µm |
How to use
- Pick your rock hardness and tumbler type.
- Keep the pre-polish stage — it halves the biggest jump in the run.
- Give the coarse stage the time it asks for; nothing later reshapes anything.
- Move on only when the whole batch is uniformly matte.
Frequently asked questions
Why did my rocks come out dull?
Almost always because the step into polish was too big. Each stage removes the scratches the last one left, and scratch depth tracks abrasive particle size — so the ratio between stages is what matters. In the standard four-stage schedule every grit-to-grit step is a drop of about 4 times, and the step from 500 grit into polish is 19 times. Polish is being asked to remove scratches five times deeper than anything else in the run ever had to.
What fixes that?
A pre-polish stage at 1000 grit. It takes the worst step from 19 times down to about 10, and the new step it introduces — 500 to 1000 — is only 1.9 times. One extra week, and it is the single most effective change most people can make to a schedule, which is why it is included here by default rather than offered as an extra.
How big a grit jump is safe?
About four times, which is not a chosen figure but the largest step in schedules that are known to work: 60/90 coarse at 250 microns into 220 medium at 63 microns is 4.0. Beyond that the next stage cannot reach the scratches in any reasonable time, and you find out weeks later at the polish stage.
Why does polish get a bigger allowance than grit?
Because polishing is not grinding. A polish works partly through flow and chemical action on the surface rather than purely by cutting, so it tolerates a larger step than an abrasive stage. That is why 10 times into polish is accepted practice while 10 times between two grits is not — one threshold cannot describe both operations.
How bad are the common shortcuts?
Worse than they look. Skipping the medium stage and going 60 straight to 500 is a 13 times drop; skipping fine and going 220 to polish is 63 times; coarse straight to polish is 250 times. None can work, and each one costs a full month before the failure is visible.
Which stage matters most?
The coarse one, and it is the one people cut short. All the shaping happens there — later stages remove scratches rather than material, so a flat spot or a sharp edge still present at the end of the coarse stage is there permanently. Hard rock can need two or three weeks of it.
How do I know a stage is finished?
The whole batch should look uniformly matte, with no shiny patches. Shiny areas mean the abrasive has not reached them, which usually means the load is wrong or the stage was short. Moving on with an uneven surface guarantees an uneven finish, because nothing later has the cutting power to catch up.
Is a vibratory tumbler faster?
Considerably, roughly a third of the time, but it does something different. A vibratory tumbler preserves shape rather than rounding it, so it will not turn a sharp rock into a smooth pebble however long you run it. Start with material that is already close to the shape you want, or do the coarse stage in a rotary and finish in the vibratory.
Why does hardness change only the coarse stage here?
Because hardness governs how long shaping takes, and shaping only happens in the coarse stage. The later stages are removing scratches of a known depth, which takes about the same time whatever the rock, so lengthening them for hard material buys nothing.
Can I tumble different rocks together?
Only if they are within about a point of hardness of one another. A softer stone in with harder ones gets ground away and bruised while the hard ones are still being shaped, so you finish with a ruined soft batch and an unfinished hard one. Sorting by hardness is worth the trouble.
What are the micron figures based on?
Nominal mid-range particle sizes for silicon carbide grading. Grading gives a distribution rather than one exact size, and the standards differ slightly between systems, so these are approximate — which is the right precision for comparing one stage against the next rather than for specifying an abrasive.
Do the day counts have to be exact?
No. They are a starting point, and longer never hurts in the grit stages — the abrasive breaks down and simply stops cutting. What does hurt is stopping early, since every later stage assumes the previous one finished. When in doubt, run it another few days and check for a uniform matte surface.
🔒 This tool runs entirely in your browser. Nothing you enter is uploaded, logged, or stored.