CNC Thread Milling Basics: How the Helical Path Cuts a Thread

• 1
  CNC Thread Milling Basics: What Happens at the Cutting Edge
• 2
  Where Thread Milling Fits and Where It Does Not
• 3
  Speeds, Feeds, and the Numbers That Decide the Finish
• 4
  What the CAM Post Has to Get Right
• 5
  Setup Details That Decide Whether the Thread Passes Gauge
• 6
  Thread Milling vs Tapping: Which Process Fits the Job

A thread mill is a small-diameter cutter with a profile that matches the thread form. The tool spins on its own axis and the machine moves it along a helical path, so the tip enters the bore off-center, sweeps around the wall, and rises by exactly one pitch per revolution. Each pass removes a thin chip. Nothing is pushed into the material.

Tapping is the opposite idea. A tap is a form tool driven into a pilot hole, and material has to deform or be cut away around its full perimeter at once. That works well in short holes and ductile material. In hardened steel, in titanium, or in a hole deeper than three diameters, the torque climbs fast and the tap becomes a liability.

The difference shows up in the forces. A tap loads the spindle axially and radially along the whole thread length at the same moment. A thread mill loads one small contact zone at a time, so the cutting force stays low even when the thread is long. That is why the same machine can cut an M30 × 3.5 thread in 4140 with a cutter that fits in a ER20 collet.

Thread milling also splits the geometry from the tool. The pitch comes from the machine's interpolation, not from the cutter's size. One 1.5 mm pitch insert can produce M10 × 1.5, M12 × 1.5, or any other diameter at that pitch. A tap cannot do that. Each thread size needs its own tap.

The trade-off is cycle time and tool cost. A single-point mill needs a full orbit per thread, often two or three passes plus a spring pass. A tap cuts the whole thread in one plunge. For a 6 mm deep M6 hole in aluminium, tapping wins on time every day of the week. Pick the process by hole count, material, and thread depth, not by habit.

Thread milling earns its place when the thread is large, deep, or close to a wall that a tap would rub against. A 1/2-13 thread 40 mm deep in 316L stainless is a tap-breaking job. The same thread milled with a helical path takes a few extra seconds and no risk of a stuck tool.

Blind holes are another clear case. A tap needs run-out clearance at the bottom, and a bottoming tap still leaves chips packed in the last few turns. A thread mill can cut to within 0.5 mm of the drilled depth and the chips fall free, because the cutter is smaller than the hole and the chips have somewhere to go.

Hard material changes the arithmetic. Above roughly 35 HRC, taps wear quickly and break without warning. Thread mills in carbide or coated HSS handle 45–55 HRC workpieces at reduced surface speed, and a broken mill is a simple tool change instead of a salvage operation.

It is the wrong choice for very small threads in high volume. Below about M4 or #8, the cutter gets too slender to survive long, and the orbit time per part kills the cycle. A form tap or a thread roll in a screw machine will beat it on cost.

Same story for soft gummy aluminium at high volume. A roll tap forms a stronger thread and runs in one pass. Thread milling is the better answer only when the part is complex, the batch is small, or the thread sits in a location that a tap can never reach.

The cutting speed for a thread mill is chosen from the workpiece material and the cutter coating, not from the thread size. In 6061 aluminium, carbide mills run comfortably at 150–250 m/min. In 4140 steel, expect 80–120 m/min with a coated carbide tool. In 316L stainless, drop to 50–80 m/min. Titanium Ti-6Al-4V sits lower again, around 30–50 m/min.

Feed per tooth is where most shops get into trouble. Because the cutter is small and the engagement is a narrow arc, the chip load per tooth is tiny. A typical starting point is 0.02–0.05 mm per tooth for a single-point mill. Multi-flute mills and full-profile cutters can take 0.05–0.12 mm per tooth in aluminium.

The feed rate the machine actually needs is the linear feed along the helical path, not the feed at the tool center. If the control uses a center feed, the real chip load at the cutting edge is larger by the ratio of the orbit diameter to the cutter diameter. Always check which convention your CAM system posts.

Radial depth of cut per pass usually runs 0.1–0.3 mm for a single-point tool. That is the radial step into the thread flank, not the axial depth. Two or three passes with a spring pass at zero radial step is normal practice for a clean thread form.

Coolant selection matters less than chip evacuation. Through-spindle air or high-pressure coolant clears the helical groove. Flood coolant alone can leave chips in a deep blind hole, and the next orbit recuts them. Recutting is the fastest way to chip a thread mill.

A thread milling cycle is a helical interpolation: a circular move in XY paired with a linear move in Z, usually written as a G02 or G03 with a Z endpoint. The pitch is the Z advance per full revolution. Get the sign of that Z move wrong and the control cuts a left-hand thread where you wanted a right-hand one.

The lead-in and lead-out arcs matter more than most programmers expect. A full-profile cutter needs a tangential arc entry of at least a quarter of the cutter diameter so the first tooth does not slam into the wall. A single-point tool can enter more directly, but a ramped approach leaves a better surface at the thread start.

Cutter compensation is the practical way to control the thread diameter. Program the theoretical pitch diameter, then offset the tool radius to hit the go/no-go gauge. This lets you adjust for tool wear without reposting the file. Keep a log of the offset per tool per material.

Retract clearance is the classic crash point. Single-point tools are the only ones that can retract radially at the bottom of a blind hole. Full-profile and multi-flute mills must exit the way they came in, or they will clip the last thread turn. Check the clearance in the post before the first run.

Simulation catches most of these errors, but verify the stock model. A simulator that only shows the toolpath will not warn you that the shank rubs the top of the hole. On a deep thread, the shank diameter is often larger than the thread minor diameter, and that interference is real.

Start with the pilot hole. Thread milling needs a hole that is at least the minor diameter plus clearance for the cutter. A hole drilled 0.1 mm undersize will rub the mill's shank and leave a torn thread. Drill, then measure the hole before you commit to the cycle.

Rigidity is the next limit. A small thread mill in a long holder will chatter, and chatter shows up as a 60-degree pitch error on the gauge. Keep the tool as short as the geometry allows and use a shrink-fit or hydraulic holder instead of a collet extension when you can.

Run-out at the cutting edge should stay under 0.01 mm. A thread mill with 0.03 mm of run-out cuts one flank deeper than the other, and the go gauge will not enter. Indicate every tool at the start of the job, not once a month.

For external threads, the same helical idea applies but the feed direction reverses. The cutter orbits the outside of a boss and steps down in Z. Support the workpiece well, because the interrupted cut on a thin boss will deflect it and produce a tapered thread.

Finally, gauge the thread with the actual mating part where possible. Thread ring and plug gauges check form and pitch diameter. They do not tell you whether the thread will hold torque in service. On safety-critical parts, that check belongs in the test plan.