Choosing a Manufacturing Method — A Decision Guide
Turning, milling, CNC, sheet cutting, tooling or 3D printing? Quantity, accuracy, material and geometry decide which route makes sense — and sometimes the answer is a combination of two.
The question that comes first almost every time is: "how should we make this part?"
The answer depends on six things, and none of them is "which machine is free."
The six factors that decide
- Quantity. One or ten thousand? This is the only factor that can change the answer on its own.
- Geometry. Rotational or prismatic? Flat or solid?
- Required accuracy. Actually required — not what you would like.
- Material. Metal, plastic, or both?
- Load and temperature. What does the part have to survive?
- Time. Needed tomorrow, or in three months?
A quick guide by geometry
If the part is rotational — shaft, bushing, pulley, flange, bearing seat — the answer is nearly always turning. High diametral accuracy, fast, and the cheapest option in low volume. Even for a single piece.
If the part is prismatic — bracket, housing, fixture, anything with pockets and slots — milling. One to a few pieces on a manual mill; from around five upward, CNC.
If the part is flat and comes from sheet — panel, frame, support, cover — sheet cutting (laser, plasma or waterjet) followed by bending. Far faster and cheaper than carving a block.
If the geometry cannot be reached by a cutting tool — internal channels, lattices, organic shapes — 3D printing, or casting if the volume is high.
Break-even: where each method pays
| Quantity | Usually the best option |
|---|---|
| 1–5 | Manual turning/milling, or printing for a prototype |
| 5–100 | CNC |
| 100–1000 | CNC with a dedicated fixture |
| Over 1000 | Injection tooling (plastic) or die casting / forging (metal) |
This table is approximate and sensitive to part complexity, but the order is nearly always right.
The factor usually overlooked is setup cost. CNC carries programming and fixturing that lands entirely on the first part. Tooling carries the mould cost, spread across the whole run. For one piece a manual machine almost always wins — even at a higher hourly rate.
Where 3D printing works, and where it does not
It works for:
- Prototypes and fit checks before committing to production
- Polymer parts under light to moderate load
- Geometry that is impossible or very expensive to machine
- Jigs and fixtures (many shops print their own)
- Runs under ten pieces of a complex part
It does not work for:
- Structural metal parts — machining or casting is the right route
- Service temperatures above 100 °C with common filaments
- Parts with tight-tolerance mating faces (unless machined afterwards)
- High volume — past a point each piece costs more, not less
Hybrid routes: the answer people miss
The best route is often not one process:
- Printed body + machined seat. Print the complex volume, turn only the face the bearing sits on.
- Cut sheet + bend + tap. Instead of carving a block, cut a sheet and fold it.
- Casting + machined mating faces. Cast the body, machine only where accuracy is needed.
- Standard profile + machining. Start from stock tube or profile rather than solid bar.
That last one saves the most and gets the least attention. If your part resembles a tube with some holes, buying tube and drilling it is far cheaper than boring out solid bar.
The question to ask yourself first
Before deciding on a process, answer this: which faces of this part actually matter?
Of the twenty surfaces on a typical part, two or three do work: where a bearing sits, where it bolts to something else, where it seals. The rest only have to exist.
Once you know that, the route reveals itself: make the surfaces that matter with the most accurate process, and everything else with the cheapest.
If you are not sure which category your part falls into, send a photo and rough dimensions. It is usually clear at a glance — and if the answer is "this should not be made that way," I will say so.