What chip load per tooth actually measures
Chip load is the thickness of material one cutting edge removes on one pass through the cut. It is the only feed number that describes what the tool experiences. A feed rate of 60 in/min means nothing on its own — it is punishing on a two-flute cutter at 1,200 rpm and trivially light on a ten-insert face mill at 4,000 rpm. Divide it down to the edge and you get a number you can compare against a tool maker's rating.
Carbide fails in two directions and chip load sets which one you get. Too much and the edge chips or the tool snaps; too little and the edge cannot get under the material, so it rubs, generates heat instead of chips, work-hardens the surface ahead of itself and wears on the flank until it stops cutting entirely. The rubbing failure is the one that surprises people, because the machine sounds fine and the finish looks acceptable right up to the point the tool dies.
The chip also carries the heat away. A properly sized chip leaves the cut hot and takes the energy with it. A thin, smeared chip leaves the energy in the tool and the workpiece, which is why light-feed cuts in stainless burn edges that heavy-feed cuts survive.
The formula and why radial engagement changes it
Start with the identity. In one minute the spindle turns n times and each revolution presents Z edges to the work, so n × Z teeth pass through the cut. The table advances F inches in that minute, so each tooth advances the work by F / (n × Z). That quotient is the chip load. Rearranged the other way it gives you the feed rate for a target chip load, which is what the milling feed rate calculator does.
The complication is that feed per tooth is not the same as chip thickness. A milling cutter sweeps an arc, and the chip it forms is a comma: zero thickness where the tooth enters, maximum somewhere in the arc, zero again where it leaves. The maximum chip thickness equals the feed per tooth only when the tooth actually passes through the point where its motion is perpendicular to the feed direction — that is, when radial engagement reaches half the cutter diameter.
Below half-diameter engagement the tooth leaves the cut before it reaches that point, and the thickest chip it forms is thinner than the feed per tooth by the factor 2√(ae/D − (ae/D)²). At 20% stepover that factor is 0.8; at 10% it is 0.6; at 5% it is 0.436. Dividing by it gives the chip thinning factor, the multiplier you apply to the programmed feed per tooth to get the chip thickness back to what the catalogue assumed. This is why high-efficiency toolpaths that run at 8–15% stepover can carry feed rates that look impossible: the chips are being thinned, and the feed is compensating.
Worked example: 1/2 in four-flute carbide at 6,000 rpm
You are side-milling with a 0.500 in four-flute carbide end mill. The program calls 6,000 rpm and 60 in/min, stepping over 0.100 in with 0.500 in of axial depth. What is each edge actually taking?
- Teeth per minute. 6,000 rev/min × 4 teeth = 24,000 teeth pass through the cut every minute.
- Chip load. 60 in/min ÷ 24,000 = 0.0025 in per tooth. Feed per revolution is 60 ÷ 6,000 = 0.010 in/rev, which is the same 0.0025 × 4.
- Radial engagement ratio. 0.100 ÷ 0.500 = 0.20, well under half, so chip thinning applies.
- Thickness ratio. 2√(0.20 − 0.20²) = 2√(0.20 − 0.04) = 2√0.16 = 2 × 0.4 = 0.80.
- Actual chip thickness. 0.0025 × 0.80 = 0.0020 in. The chip thinning factor is 1 ÷ 0.80 = 1.25.
- Chip volume per tooth. 0.0025 × 0.100 × 0.500 = 0.000125 in³, and the removal rate is 0.100 × 0.500 × 60 = 3.0 in³/min.
Now suppose the tool data sheet calls for 0.0025 in of chip thickness, not feed per tooth. Multiply: 0.0025 × 1.25 = 0.003125 in per tooth, and the feed becomes 0.003125 × 4 × 6,000 = 75 in/min. Same tool, same depth, 25% more metal per minute, because you corrected for a geometric effect rather than guessing.
How to read the number you get
Compare it against the tool maker's published feed per tooth for your diameter and material, and check which quantity they publish. Some catalogues list feed per tooth expecting you to apply thinning yourself; others list chip thickness and expect the correction to be built into your feed. The calculator gives you both so you can match whichever convention the data sheet uses.
Three rough bands are worth memorising for solid carbide end mills. Below roughly 0.0004 in per tooth the edge preparation on a typical carbide tool is comparable to the chip, and it ploughs rather than cuts — this is the rubbing regime. Between about 0.001 in and 4% of the cutter diameter is the ordinary working range for most metals. Above 5% of diameter you are into heavy roughing territory where the tool maker has to say it is allowed.
Small tools are governed by their own geometry, not by the material. A 1/8 in two-flute end mill at 0.004 in per tooth is taking a chip 3.2% of its diameter, which is aggressive but survivable; the same 0.004 in on a 1/32 in cutter is 12.8% of diameter and will break it. Scale chip load with diameter, then check the result against the material.
Chip load also interacts with the surface speed you chose. If the number here looks fine but the tool still burns, verify the spindle speed with the cutting speed calculator before touching the feed — excessive surface speed and insufficient chip load produce the same symptom.
Chip thinning factors by radial engagement
| Stepover as % of diameter | Stepover on a 1/2 in cutter | Chip thickness ratio | Feed multiplier |
|---|---|---|---|
| 5% | 0.025 in | 0.436 | 2.294 |
| 10% | 0.050 in | 0.600 | 1.667 |
| 15% | 0.075 in | 0.714 | 1.400 |
| 20% | 0.100 in | 0.800 | 1.250 |
| 25% | 0.125 in | 0.866 | 1.155 |
| 30% | 0.150 in | 0.917 | 1.091 |
| 40% | 0.200 in | 0.980 | 1.021 |
| 50% or more | 0.250 in or more | 1.000 | 1.000 |
The ratio is the geometry of the arc the tooth sweeps, not a property of the material or the tool. It applies to carbide, HSS and router bits identically.
Mistakes that produce a wrong chip load
- Counting flutes instead of counting cutting edges. A ball nose cutting at very shallow depth may have only part of each flute engaged, and some finishing tools have unequal flute spacing. Count what touches the work.
- Forgetting the feed override. The chip load the tool sees comes from the feed the machine is executing. A 60% override on a 100 in/min program is a 60 in/min cut.
- Applying chip thinning at full slot width. The correction only exists below half-diameter radial engagement. Applying it while slotting multiplies an already-maximum chip and breaks tools.
- Ignoring axial chip thinning on round inserts and ball noses. A round insert at shallow axial depth thins the chip in the axial direction by a similar mechanism; this calculator handles the radial case only.
- Treating chip load as a material constant. It scales with tool diameter and with how rigid the setup is. The same alloy takes a different chip load on a 1/8 in cutter than on a 1 in cutter.
Where chip load sits among the other cutting parameters
Feeds and speeds decompose into three independent choices, and chip load is only one of them. Surface speed sets the temperature at the cutting edge and comes from the material and the tool grade; that is what the spindle RPM calculator converts into a spindle speed for your diameter. Chip load sets the mechanical load per edge. Depth of cut, radial and axial, sets how much of the tool is doing work and therefore how much it deflects. Change one and you have usually changed the safe range of the others.
Metal removal rate is the product of all three, and it is the number that pays for the machine — the material removal rate calculator works it out for milling, turning and drilling. The productive move is almost never to raise chip load alone. Modern high-efficiency roughing gets its removal rate from a deep axial cut at a light stepover, with the feed raised to compensate for chip thinning, precisely because that combination spreads wear along the whole flute length rather than concentrating it in the bottom 0.1 in.
Drilling uses the same vocabulary with a different geometry: a twist drill has two cutting lips and its chip load is quoted as feed per revolution divided by two, which the drilling speed and feed calculator handles alongside the torque the hole will demand. Machinery's Handbook remains the standard reference for the milling conventions used here, including the definition of feed per tooth and the arc geometry behind the chip thinning correction.
Key terms
- Chip load (f<sub>z</sub>)
- Feed per tooth — the distance the work advances between one tooth entering the cut and the next. Also written IPT (inches per tooth) or mm/tooth.
- Radial depth of cut (a<sub>e</sub>)
- The stepover, measured perpendicular to the tool axis and to the feed direction. Equal to the cutter diameter when slotting.
- Axial depth of cut (a<sub>p</sub>)
- The depth of engagement along the tool axis, sometimes called depth of cut or DOC.
- Radial chip thinning
- The reduction in maximum chip thickness below the programmed feed per tooth that occurs when radial engagement is less than half the cutter diameter.