Surface Finish Calculator
Drag the feed and watch the surface change. Theoretical Ra and Rt for turning and ball-nose milling — with the scallops drawn live, so the number stops being abstract.
Max feed for a target Ra
| Ra 3.2 µm(N8) | ≤ 0.283 mm/rev |
| Ra 1.6 µm(N7) | ≤ 0.200 mm/rev |
| Ra 0.8 µm(N6) | ≤ 0.141 mm/rev |
| Ra 0.4 µm(N5) | ≤ 0.100 mm/rev |
The surface, up close
cross-section · 6 passes0.90
Ra, µm
35 µin
3.52
Rt, µm
peak to valley
N7
ISO grade
good machining
Where it lands
Chips show the Ra ceiling (µm) of each ISO 1302 grade. Your theoretical finish qualifies for the highlighted grade — if the process behaves.
Finish is geometry before it's anything else
Every pass of a round edge leaves an arc-shaped groove, and the next pass leaves another one a feed-width away. The ridges between them — the scallops — areyour surface finish. That's why the two levers in this calculator are the only two in the formula: pitch between passes (feed or stepover) and the radius doing the cutting. Feed enters as the square, which is the practical takeaway — halving the feed quarters the theoretical Ra, while doubling the nose radius only halves it.
The catch is the fine print on that big nose radius: more edge in contact means more radial force, and more radial force means chatter — which ruins a surface faster than any scallop. I wrote about attacking that problem from the tool-geometry side in why I put variable flute and index geometry on an endmill. And since feed is also your cycle time, finish requirements are exactly where auditing the rest of the program pays off — reclaim the air time, and you can afford the finishing feed the drawing actually needs. When it's time to measure the result, mind the uncertainty.
Frequently asked questions
How is theoretical Ra calculated from feed and nose radius?+
A round cutting edge stepping across the surface at a fixed pitch leaves circular-arc scallops. The peak-to-valley height is Rt ≈ f²/(8·r) and the arithmetic-mean roughness is Ra ≈ f²/(31.2·r), with feed f and nose radius r in millimetres. The same geometry applies to ball-nose milling with stepover as the pitch and ball radius as r.
Why does my real part measure rougher than the calculator says?+
The formula is pure geometry — a perfect edge on a perfectly rigid machine. Real surfaces pick up vibration, built-up edge, tool wear, and material effects, and typically come out 1.5–3× rougher than theoretical. Use the number as the floor you cannot beat, and leave margin against the drawing's spec.
What's the difference between Ra and Rt (or Rz)?+
Ra is the arithmetic average deviation of the profile — a smoothed, single-number summary and the value most drawings call out. Rt is the total peak-to-valley height of the profile, and Rz is a mean of several peak-to-valley samples. For the ideal scallop profile, Rt is roughly four times Ra, which is why a part can 'pass on Ra' and still feel ridged.
Does a bigger nose radius always give a better finish?+
Geometrically yes — Ra falls linearly as the radius grows. But a bigger nose radius increases radial cutting force and the tendency to chatter, especially on slender parts or long overhangs, and chatter destroys a finish far faster than scallop geometry improves it. In practice you pick the largest radius the setup tolerates without vibration.
How do I convert Ra in µm to µin?+
Multiply by 39.37. So Ra 0.8 µm is about 31 µin, and Ra 3.2 µm is about 126 µin. The calculator shows both, since North American drawings often spec µin (e.g. 32 or 63 µin) while ISO drawings use µm.
Chasing a finish spec you can't hit?
Tool geometry, parameters, and process — that's my day job.