Choosing Insert Nose Radius: Finish Against Edge Strength

By Bradley Taylor · August 2026

The nose radius is the last digit on the insert box that anybody thinks about, and it is the one doing the most work at the surface of the part. A CNMG 432 and a CNMG 431 are the same insert everywhere except the corner, and that corner is what decides your surface finish, a good share of your tool pressure, and whether the edge survives an interrupted cut. Picking a nose radius is one long trade between finish and strength, and once you can see both sides of it, the choice mostly makes itself.

What the nose radius actually does

In a normal turning pass, the straight cutting edges do the bulk of the material removal, but the rounded corner is the only part of the insert that touches the finished surface. As the tool feeds along, the nose sweeps a series of overlapping arcs into the part, and the little scallops left between those arcs are your surface finish. The finish you measure with a profilometer is, at its core, the geometry of those scallops. A bigger radius leaves flatter, shallower scallops at the same feed. A smaller radius leaves taller, sharper ones.

The corner is also the weakest point on the insert. All the cutting force in the finish zone funnels through that little arc of carbide, and a small radius has less material behind the edge to carry it. That is the whole tradeoff in two sentences. The radius that gives you the nicest theoretical finish is the one most likely to chip when the going gets rough.

The finish side of the trade

The scallop geometry gives you a clean approximate formula, but you have to know which roughness number it produces. Feed per revolution squared, divided by eight times the nose radius, gives the theoretical peak to valley height, the number the metric world calls Rz. American prints almost always call out Ra instead, and Ra is the average, which lands at roughly the peak to valley divided by 3.9. The Ra version of the formula is feed squared divided by 31.2 times the radius. Mixing them up is a two hundred percent error in whichever direction you make it, and a machinist who applies the peak to valley formula to a 32 Ra callout ends up feeding half as fast as the print actually requires. The exact constants matter less than the shape of the relationship. Finish scales with feed squared, so doubling your feed roughly quadruples the roughness, while doubling the nose radius only cuts it in half. Feed is the loud lever and radius is the quiet one, which is why the first fix for a rough finish is usually feed, not a new insert. One caution on that though. Below the built up edge threshold in gummy materials, 304, 316, low carbon steel, soft aluminum, dropping the feed further makes the finish worse, not better, because the edge starts rubbing instead of cutting. The chip thinning article covers that mechanism. In those materials the fix is often more surface speed, not less feed.

The word theoretical is doing honest work here. Real parts usually come out somewhat rougher than the formula because of built up edge, vibration, material tearing, and edge wear, so treat it as the floor for that nose geometry at that feed. A worn flank land burnishing the surface, or a wiper geometry, can measure below the standard corner math, which is exactly what wipers are sold to do. If the print calls for a finish the math says your feed and radius cannot produce, you already know the outcome before you press cycle start. The surface finish calculator on this site shows both numbers, labeled Ra and Rt, for your feed and nose radius, and it is worth a ten second check before chasing a finish callout the hard way.

The strength side of the trade

A larger nose radius means a stronger corner, plain and simple. There is more carbide backing up the edge, the cutting load spreads over a longer arc, and the chip thins out toward the nose instead of loading one small spot. That buys you real advantages in roughing, where you want to push heavy feeds and deep cuts without babying the tool. It buys even more in interrupted cuts, where the corner slams into the work several times a second and a delicate nose chips on the first hit. Heat also has somewhere to go on a big nose, so the edge holds up longer at aggressive parameters. When I am hogging material and the finish pass belongs to a different tool, I reach for the biggest radius the job will tolerate and stop thinking about it.

The catch with big radii

If big radii were free, everything would ship with a 1/16 nose. The bill comes due as tool pressure. A larger radius has more edge in contact with the work at any moment, and more of the cutting force turns radial, pushing the tool away from the part instead of down the chip. On a stout part in a rigid setup, nobody notices. On a slender shaft, that radial push becomes deflection, and the part springs away from the tool and comes out tapered or oversize in the middle. On thin walls and long stickouts, it becomes chatter, and once a big nose starts singing it does not politely stop. The insert that shrugs off an interrupted cut in a chuck can make a 3/8 shaft between centers sound like a banjo.

A big radius also sets a floor on your depth of cut. When the depth drops below the radius, the entire cut rides on the arc, the effective lead angle collapses, and nearly all the force turns radial, which is where deflection and chatter come from on light passes. The usual guidance is to keep depth of cut at or above about two thirds of the nose radius for stable chip formation. A 1/32 nose taking a two thousandths spring pass is working in exactly the zone where it behaves worst.

The other catch is the print. In a straight plunge to a shoulder, the nose radius is the smallest inside corner the tool can leave. If the drawing calls out a 0.015 fillet at a shoulder, a 1/32 nose cannot cut it in one move, and since most fillet callouts are a maximum, the smaller nose is the safe pick. A smaller radius can always profile a larger fillet by interpolating the corner, but a bigger nose can never make a smaller one. Check the corner callouts before you pick the insert, because finding out at first article is the expensive way.

Common sizes and where they land

In inch designations the usual lineup runs 1/64, 1/32, 3/64, and 1/16, which is roughly 0.4, 0.8, 1.2, and 1.6 in millimeter terms. The 1/64 nose is the finishing and detail radius, the pick for fine work, light depths of cut, slender parts, and prints with small corner radii. The 1/32 is the shop default for a reason. It is a reasonable compromise on finish, strong enough for moderate roughing, and it matches a corner callout you see on drawings constantly. If a job gives you no reason to choose otherwise, the 1/32 is rarely a wrong answer. The 3/64 and 1/16 sizes are roughing corners, happiest at heavy feeds and depths where their strength gets used and their tool pressure has a rigid setup to push against.

The half radius rule

There is a working rule that ties the whole thing together. Feed per revolution runs roughly a quarter to a half of the nose radius, toward the low end for general turning and finishing, toward the high end for roughing on a rigid setup. Half the radius is the ceiling the insert makers publish, and past it the scallops get deep fast, because of that squared relationship, and the chip starts forming badly at the corner. For a 1/32 nose that means around 0.008 per rev is a comfortable general purpose feed and something like 0.015 is the roughing ceiling. If the cycle time math demands more feed than half your radius, the fix is a bigger radius, not wishful thinking.

When you genuinely need heavy feed and fine finish in the same pass, that is what wiper inserts are for. A wiper adds a short flat or large secondary radius behind the nose that smears the scallops flat, holding finish at feeds that would embarrass a standard corner. They want rigidity and they raise tool pressure, but they are the honest way to cheat the tradeoff.

The ten second version

Start at 1/32 and let the job argue you off it. Go smaller for slender parts, thin walls, long stickouts, and tight corner callouts. Go bigger for roughing, interrupted cuts, and heavy feeds, as long as the setup is rigid enough to absorb the pressure. Keep feed between about a quarter and half the radius, keep depth of cut above about two thirds of it, and let the calculator confirm the theoretical finish before the profilometer does it for you.

As always, this is general practice, not a spec. Prints, customer requirements, and the governing standard win every argument.