Machinability Ratings Explained

By Bradley Taylor · August 2026

Look up almost any metal in a materials handbook and you will find a machinability rating, usually a percent. 12L14 sits up around 170 percent or better. 304 stainless sits down around 45. The number looks official, and it is genuinely useful, but it only makes sense once you know what it is being compared to and what the old test actually measured. It also lies to you in a few specific ways, and knowing where it lies is half the value.

Where the percent comes from

A machinability rating is a comparison, not a measurement of the material by itself. The baseline is AISI B1112, a free machining steel with both sulfur and phosphorus added, made by the old acid Bessemer process the B in the name refers to, specified at 160 Brinell, and defined by the industry as exactly 100 percent. Every other material gets rated against it. A material at 200 percent machines about twice as easily as B1112, and a material at 40 percent machines less than half as easily.

The historical basis was turning tests, and two different conventions are floating around out there, which explains some of the disagreement between published tables. The original AISI tests ran B1112 at a fixed 180 surface feet per minute with the high speed steel tooling of the era and scored candidate materials on a weighted mix of tool life, finish, and speed. The other convention defines the rating through tool life directly: find the speed at which a tool lasts 60 minutes in the candidate material, divide by the speed that gives 60 minutes in B1112, commonly cited around 100 surface feet per minute for high speed steel, and that ratio is the rating. The two schemes usually land in the same neighborhood but they are not the same test, different labs ran them differently, and the published numbers for the same alloy do not always agree. Treat any rating as an approximate figure with soft edges, not a certified property like yield strength.

What the number means at the machine

The practical use is simple. The rating is a first pass scaling knob for cutting speed. If you know a comfortable SFM for one material, a material rated twice as high will usually tolerate somewhere near twice the speed, and a material rated half as high wants somewhere near half. It will not land you on the perfect number, but it gets you into the right neighborhood on the first part instead of the fifth.

The number is also an expectation setter. When the router calls out something rated 200 percent, you can expect the material to cut like it wants to be cut. Chips break, tools last, and the spindle load meter barely notices. When you see 40 percent, plan for the opposite. Slower speeds, shorter tool life, higher cutting pressure, and a material that fights back the whole way. Neither situation is a problem if you saw it coming. Most of the ugly surprises in this trade come from treating a 40 percent material like a 100 percent material for the first twenty minutes.

Where the single number lies

One number cannot describe everything about how a material cuts, and the rating quietly ignores several things you will care about at the machine.

It says nothing about chip control. Plain low carbon steel like 1018 carries a respectable rating, and the tools do last, but the material is gummy and the chips come off in long stringers that wrap the part into a bird nest if the feed and chipbreaker are not right. The rating told you tool life would be fine. It never promised you would not spend the afternoon fishing steel wool out of the chuck.

It says nothing about work hardening. Austenitic stainless like 304 is meaner than its number. The rating captures the fact that it is slow going, but not the fact that a dull tool or a timid cut hardens the surface ahead of the edge and makes the next pass worse than the last one. A material can carry the same rating as some annealed alloy steel and still punish hesitation in a way the alloy steel never will.

It says nothing about abrasiveness. Glass filled plastics and sandy castings can rate fine on paper, because in a short tool life test they cut easily. Run them for a shift and the filler or the scale grinds the edge off your tooling like a lapping compound. The rating measured how hard the material resists cutting, not how fast it sands your insert down to a nub.

And condition matters enormously. A rating gets published for a material name, but 4140 annealed and 4140 prehard are different animals wearing the same name tag. Heat treat condition, cold work, and even which mill poured the heat can move real world machinability a long way from the handbook figure. When the cert says prehard and the handbook number was taken annealed, believe the cert.

Do the old ratings still matter

The reference test was built around tooling that no longer exists on most shop floors. Coated carbide moved the absolute speeds up several fold from what the old high speed steel testing assumed, so nobody should read the historical 180 SFM baseline as a modern recommendation for anything. What survived is the relative ranking, but hold it loosely. The order of materials from friendly to hostile is still right, easy steels still beat stainless and stainless still beats superalloys. The exact ratios did not survive as cleanly, because coatings, tougher carbide grades, and high pressure coolant closed the gaps unevenly, helping the hard materials more than the easy ones. So use the rating to rank materials and to scale a first guess, and expect the real spread between two materials on modern tooling to be smaller than the 1950 numbers imply. The ratings aged the way a good map ages. The terrain is in the same places, but do not trust the mileage table.

How to actually use a rating

Here is how I use them. Start from a material you know cold. If you run a lot of cold drawn 1018 and you know exactly where it likes to live on your machine, its published rating is around 78 percent, and hot rolled runs closer to 70, which is a live example of how much condition moves the number. When a new job shows up in a material rated 60, scale your known speed by the ratio of the ratings and call that your starting SFM. Take a cut, then adjust by ear and by insert wear. The sound of the cut, the color of the chips, and the condition of the edge after the first few parts will tell you more than any table. The rating gets you to a safe starting point. The machine tells you where to go from there.

The material library on this site lists ratings and starting speeds for the common shop materials so you have a baseline without digging through handbooks, and the speeds and feeds calculator turns the SFM into RPM and feed for your actual tool diameter. Between the two you can walk up to an unfamiliar material with a defensible first guess instead of a shrug.

Used that way, the machinability rating is one of the more honest numbers in the handbook. It admits it is a comparison, it gets you close on speed, and it warns you when a material is going to be a fight. Just remember what it does not say, watch the chips, and keep one hand near the feed hold the first time you cut something new.

As always, this is general practice, not a spec. Prints, customer requirements, and the governing standard win every argument. Standards referenced: historical AISI machinability test data (B1112 baseline).