After So Many Years of Machining, Do You Know Trochoidal Milling?

A conventional slot cut drives the cutter straight down the middle. The tool enters with a full 180° wrap, so half the flute length is buried in the material at the same time. Heat has nowhere to go and the chip has nowhere to curl. Most broken 6 mm end mills in aluminium or 304 stainless die in exactly that cut.

The trochoidal path changes the engagement. The tool travels along a small circle while the circle itself advances down the slot. The cutter sweeps an arc of 20° to 60° of the workpiece at any moment, never the full slot width. Each tooth takes a short, thick chip and then leaves the cut for most of the revolution.

The trade shows up the moment you watch the numbers. Radial engagement stays low, so heat leaves with the chip instead of soaking into the tool and the part. Axial depth of cut goes the other way. A cutter that would take 2 mm in a straight slot can run 1×D or more in a trochoidal pass.

Programmers who came up on straight-line pocketing often skip this path because it looks slower on screen. The arc moves are longer, and the feed rate looks low next to a full-width cut. On the machine the picture flips. The tool survives, the spindle load stays flat, and the part comes off with less bow.

Tool choice drives everything here. A variable-helix, variable-pitch carbide end mill with a relieved neck is the usual pick. The neck clearance lets the tool reach deeper without rubbing the wall behind the flutes, which is the failure mode in a deep trochoidal pass. Coatings help in steel and stainless; AlTiN and AlCrN hold up at the surface speeds this path produces.

Radial engagement sets the chip load. Keep it between 8% and 15% of the cutter diameter for most work, and let the CAM step-over control the arc width. Feed per tooth then rises well above a conventional slot value. In 6061 aluminium, a 10 mm cutter might run 0.10 mm per tooth at 12% radial engagement; in 304 stainless the same cutter drops to roughly 0.04 mm per tooth.

Axial depth is where the method earns its keep. Running 1×D to 2×D axial depth with a light radial bite spreads wear along the flute instead of concentrating it at the tip. The result is a more even tool life and fewer mid-run changes. Coolant should reach the cut, but through-spindle air blast often works better than flood in aluminium because it clears chips from the deep pocket.

Rigidity still matters. A long, thin cutter in a weak holder will chatter no matter how clever the path is. Shorten the gauge length where you can, and check runout at the tool tip before the first part. Runout above 0.01 mm turns one flute into the only flute that cuts.

Deep slots in hard material are the clear win. A 40 mm deep, 12 mm wide slot in 17-4PH stainless is painful with a straight plunge. With a trochoidal path the cutter works its way down in light arcs and the load stays steady. The same logic applies to narrow ribs and deep pockets where chip evacuation is the real limit.

Thin-walled parts also benefit. Low radial engagement means lower cutting force pushing against the wall, so the part deflects less and finishes closer to nominal. On a 1.5 mm wall in aluminium, that difference shows up as a flat wall instead of a bowed one.

The method loses when the slot is shallow and open. Cutting a 3 mm deep groove 20 mm wide with arcs adds travel distance for no thermal benefit. A straight pass with a larger cutter is faster. The same goes for roughing a big open pocket where a face mill or a high-feed cutter moves more metal per minute.

Hardened tool steel above 50 HRC is a gray area. The path works, but cutter cost climbs and the arc moves need a rigid machine to hold tolerance. For one-off parts in that range, conventional roughing plus a finishing pass is often the cheaper route.

Short runs of simple geometry rarely justify the programming time. A CAM template for trochoidal roughing takes a while to tune. Once it is tuned, though, it gets reused on every deep-slot job in the shop.

Tolerance on a trochoidal pass depends more on the machine and the wall stiffness than on the path itself. Our 5-axis centers hold ±0.005 mm on finishing passes, and the light radial load makes that easier to repeat than a heavy slot cut. Surface finish on a trochoidal roughing pass lands around Ra 1.6–3.2 μm. A separate finishing pass brings it to Ra 0.8–1.6 μm, or Ra 0.2–0.8 μm on medical and optical work.

Chip evacuation is the hidden variable. Recutting chips is what kills tools in deep pockets. Air blast, through-tool coolant, or a short peck cycle every few arcs all help. Watch the chip shape: a proper trochoidal chip is a comma, not dust.

Machine choice follows the part. A 4,000 × 400 × 150 mm travel machine handles long extruded profiles where the slot runs the full length. Compact 500 × 500 × 450 mm centers are fine for smaller pockets. A Ø400 mm rotary table lets a 4-axis setup cut arcs on several faces without re-fixturing, which cuts the error stack.

Programming matters as much as hardware. Most CAM packages now include a trochoidal or dynamic roughing option. The default step-over is usually too aggressive for small cutters. Reduce it, raise the feed per tooth, and test on scrap before the first production part.