Chip Chad
The end mill question comes up constantly: 2-flute or 4-flute, carbide or HSS, coated or uncoated. The catalog doesn't help — every tool looks reasonable until you try to decide. This is the decision process that actually works, starting from the material you're cutting.
Start with the material
The material determines the cutting geometry you need. Tooling follows from that, not the other way around.
- Aluminum wants sharp, highly-polished flutes, large chip clearance, and a rake angle that lets chips evacuate without re-welding. Single-flute (O-flute) or 2-flute carbide; uncoated, ZrN, or DLC — see Coatings below.
- Steel (mild to medium hardness) wants a tougher edge, more flutes, and a coating that handles heat. 4-flute carbide with TiAlN.
- Stainless and titanium are in their own category — heat-sensitive, work-hardening materials that destroy tools fast. 4-flute variable-helix carbide with TiAlN or AlTiN, run conservatively.
- Wood and plywood want maximum chip clearance and a sharp edge to prevent tear-out. Single-flute or 2-flute upcut spiral carbide or sharp HSS.
- Plastics need sharp, polished edges and enough chip load to avoid melting. Single-flute or 2-flute carbide or sharp HSS.
Flute count: the first decision
Single-flute (O-flute) has one cutting edge and maximum chip clearance. It is the most underrated tool in hobby CNC and worth its own callout below.
2-flute end mills have generous chip room. This matters in aluminum, plastics, and wood where chip evacuation is the limiting factor. At the same feed rate as a 3-flute, each tooth takes a heavier cut — which actually helps in these materials by staying above the minimum chip load.
3-flute is a good compromise for aluminum at higher feed rates (where a 2-flute starts to struggle to clear chips) and for softer steels.
4-flute and above are for steel, stainless, and hardened materials. More flutes means more edge contact at any moment, which distributes load and produces better finish — but only if you have the feed rate to keep chip load above the minimum. Running a 4-flute at the same feed as a 2-flute means each tooth takes half as much material per pass. In aluminum, that's often below minimum chip load. In steel, it's fine.
Don't sleep on O-flute
O-flute (single-flute) is the open secret of hobby CNC for aluminum — and it works better than most people expect in wood too.
The reason comes down to chip load math. Chip load equals feed rate divided by RPM times flute count. Hobby spindles are happiest toward the top of their RPM range — trim routers like the Makita RT0701 (10,000–30,000 RPM) or DeWalt 611 (16,000–27,000 RPM) deliver more consistent torque at higher speeds. With a 2-flute at 18,000 RPM, hitting a 0.001" (0.025 mm) chip load in aluminum requires 36 IPM (915 mm/min). That's fast for a machine that's also fighting rigidity limits. Drop to a single flute and the same 18,000 RPM needs only 18 IPM (460 mm/min) for the same chip load. The machine runs where its spindle is happy without demanding feed rates the frame can't hold cleanly.
There's a surface finish argument too. If a 2-flute at the speeds you can actually run puts you below minimum chip load, the tool is rubbing rather than cutting — which builds heat, wears the edge, and produces worse finish than the math suggests. An O-flute operating in its correct chip load window will cut cleaner in practice even if its theoretical finish number is lower.
O-flute also works well in wood and plastics. The chip clearance is generous enough to stay out of trouble, and it keeps the operating point inside the envelope your machine can actually reach. The reputation for being "an aluminum tool" undersells it.
Where O-flute has no advantage: steel, stainless, and hardened materials. Steel doesn't share aluminum's chip evacuation problem, and the single cutting edge carries all the load per revolution — you get no benefit and shorter tool life compared to a 4-flute. Use 4-flute in steel.
Starting parameters for 6061 aluminum
Conservative slotting starting points — run these, listen, and dial in from there. All assume a 0.001" (0.025 mm) chip load target.
| Tool | RPM | Feed rate | DOC | WOC |
|---|---|---|---|---|
| 1/8" (3 mm) O-flute | 18,000–24,000 | 18–24 IPM (460–610 mm/min) | 0.062" (1.6 mm) | 0.062" (1.6 mm) |
| 1/8" (3 mm) 2-flute | 18,000–24,000 | 36–48 IPM (915–1220 mm/min) | 0.062" (1.6 mm) | 0.062" (1.6 mm) |
| 1/4" (6 mm) O-flute | 18,000–24,000 | 18–24 IPM (460–610 mm/min) | 0.125" (3.2 mm) | 0.125" (3.2 mm) |
| 1/4" (6 mm) 2-flute | 18,000–24,000 | 36–48 IPM (915–1220 mm/min) | 0.125" (3.2 mm) | 0.125" (3.2 mm) |
The 1/4" (6 mm) 2-flute row requires 36+ IPM to stay above minimum chip load. Many hobby machines can handle this with adaptive/trochoidal toolpaths and reduced WOC; full-width slotting at that feed rate is harder to sustain cleanly. If your machine can't hold 36 IPM confidently, a 1/4" (6 mm) O-flute — or stepping down to a 1/8" (3 mm) tool — keeps you inside the operating envelope without sacrificing the cut.
Spiral direction: upcut, downcut, and compression
The helix direction of the flutes determines where chips go and which face of the workpiece gets the cleaner edge.
Upcut spiral pulls chips up and out of the slot. The bottom of the cut is clean; the top surface can show tearout or fraying as the tool exits upward through the material. This is the correct choice for most aluminum work — chip evacuation in aluminum is critical, and any re-cutting of chips causes built-up edge and tool failure fast. In wood, upcut is fine when the top surface isn't a finish face (parts being flipped, spoilboard operations, through-cuts where you're trimming to size anyway).
Downcut spiral pushes chips downward. The top surface is clean and crisp; chips pack into the slot rather than evacuating. In wood pockets and dadoes where the top surface finish matters — cabinet faces, inlays, anything that will be seen — downcut gives you a cleaner result without sanding. The tradeoff is heat in deep cuts: chips that can't escape get recut and generate excess wear. Keep depths conservative and clear chips frequently.
In aluminum, downcut has almost no use case. The one scenario that comes to mind: boring an existing through-hole to final size where the geometry means you want chips driven down through the hole rather than lifted up past the entry face. Outside of that, use upcut in aluminum.
Compression spiral is upcut at the tip and downcut at the top of the flute, designed to leave clean faces on both sides of the material simultaneously. The theory is good for sheet goods — plywood, MDF, laminates — where both the top and bottom face need to be clean. The practical limitation for hobby machines: the upcut section at the tip (typically 3–5 mm) must fully clear the material on the first pass for the compression geometry to do its job. If your first-pass DOC is shallower than that, you effectively have an upcut bit. Machines that can't take that first-pass depth are better served by a downcut for pockets and an upcut for through-profile cuts — most of the same result with less complexity.
One note on top-surface burr in aluminum: a climb cut on the finishing pass significantly reduces entry-side fraying, though a light pass with a deburring tool or a few strokes of a file will clean up what remains quickly. Climb cutting on hobby machines has its own caveats — see Tuning Your Hobby CNC Router for when it's appropriate.
End mill geometry types
Flat end mill is the default for pockets, slots, and facing. Square shoulders, good for vertical walls. The corner is sharp and chips easily — if finish matters, use a corner-radius tool on the finishing pass.
Ball nose is for 3D surfaces and contours. The curved tip creates smooth sculptured surfaces but leaves witness lines in flat cuts (unavoidable scallops between passes).
Corner radius (bull nose) is the best choice for most pocket finishing. A small radius (0.010"–0.030" / 0.25–0.75 mm) on the corner dramatically improves edge strength and leaves better finish on the pocket floor without the scalloping of a ball nose. If you're buying tools for general work, corner-radius variants of your common sizes are a better investment than flat-end mills.
Rougher (corncob, chip breaker) is for maximum material removal. The serrated cutting edge breaks chips aggressively. The finish is terrible; that's the tradeoff. Use it for roughing only, follow with a flat or corner-radius end mill.
Chamfer mill cuts bevels and deburrs edges. Not a general milling tool.
Carbide vs. HSS
Carbide runs faster, stays harder at temperature, and deflects less (higher modulus of elasticity). It's brittle — drops, crashes, and interrupted cuts chip carbide that would flex in HSS. On a rigid machine with good workholding, carbide is almost always the right choice.
HSS is tougher and cheaper. It dulls faster and can't run as fast, but it survives abuse that would fracture carbide. For large-diameter facing operations, interrupted cuts in difficult setups, and learning situations where tool crashes are likely, HSS makes economic sense. A $6 HSS end mill that breaks in a crash costs a lot less than a $25 carbide one.
For hobby machines with occasional crashes and marginal rigidity: carbide for cutting operations, HSS for learning setups on expensive materials.
Coatings
For most hobby work, geometry and chip load matter more than coating. That said, getting the wrong coating in aluminum will ruin tools fast.
Uncoated carbide is the safe default for aluminum. No chemical interaction, no adhesion issues.
ZrN (Zirconium Nitride) and DLC (Diamond-Like Carbon) both work well in aluminum. Low friction, chemically inert, and they resist chip welding. If your carbide tools have either coating, use them without concern.
TiAlN and AlTiN are excellent coatings for steel, stainless, and high-temperature alloys — they're designed to run hot. Avoid them in aluminum. The titanium content has chemical affinity to aluminum that promotes built-up edge, where aluminum welds to the flute and ruins the cut.
TiN (Titanium Nitride) is the gold-colored general-purpose coating on many cheaper tools. It performs adequately in wood and mild steel. Not a problem in aluminum, just not the first choice.
Diameter selection
Use the largest diameter that fits the feature you're machining. Larger diameter means:
- More flutes for the same chip room
- Higher surface footage at the same RPM (or lower RPM for the same surface footage)
- More rigidity and less deflection for the same reach
The exception is tight corners and inside radii — you can't cut a 0.1" (2.5 mm) inside corner with a 1/4" (6 mm) end mill. The tool diameter sets the minimum inside radius you can produce.
In hobby aluminum work specifically, there's a tendency to reach for 1/4" (6 mm) or larger tools because they feel more substantial. But a 1/4" (6 mm) tool at 18,000 RPM demands 36+ IPM to stay above minimum chip load — a feed rate many hobby machines struggle to hold cleanly. A 1/8" (3 mm) O-flute at the same RPM needs 18 IPM for the same chip load. The smaller tool stays inside the operating envelope; the larger one pushes against it. Match tool size to what your machine can actually drive, not to what feels sturdy.
For roughing, match the tool diameter to the pocket width when possible: a 3/8" (10 mm) end mill roughing a 3/4" (19 mm) pocket (2× diameter passes) is generally efficient. For finishing, step down to a smaller tool if the corner radius allows it.
Stick-out and the 3× rule
Keep stick-out at or below 3× tool diameter for general milling. At 3× reach, a 1/4" (6 mm) end mill will show measurable deflection in steel. Longer than 3× and you need to significantly reduce DOC and feed to compensate. The math is not linear — deflection scales with the cube of stick-out length.
If you genuinely need long reach (deep pockets, far-side features), look at long-reach or extended-reach end mills with reduced-diameter shank sections. These are engineered to minimize the diameter change along the reach length, which is structurally better than just hanging more tool out of a standard collet.