Every mechanical component starts as a material-and-process question. A gear can be machined from bar, sintered from powder, molded by MIM, or forged; a bushing can be turned from brass or bronze or molded from an engineering plastic; a housing can be machined from plate, die-cast, or fabricated from sheet. Each combination is a route with its own cost, tolerance, and lead-time story, and the route that wins depends on the component's function and quantity, not on habit.
What is a material-process decision really choosing for a mechanical component?
A mechanical component does its job through its material and its geometry: the alloy or polymer carries the load, the surface resists wear, the features locate it against the parts around it. The process choice determines which materials are even available, because a material in bar form can be machined, a material in powder form can be sintered or molded, and a material that only exists as a forging or a casting arrives near-net and needs machining for its fits. The decision is therefore not "machining versus casting" — it is a choice of complete routes, and each route has a tolerance, a surface, and a certificate story attached.
The starting point is always the function. Write down what the component must do: the load it carries, the speed and lubrication at any bearing surface, the environment it lives in, the features that mate with other parts, and the quantity you will actually buy. Function filters the materials to an eligible set; quantity and cost then pick the route within that set. A component whose drawing demands tight tolerances on its mating features is going to pass through machining at the end of any route, which means the real question is how much of the component machining must produce from solid versus finish from a blank.
Machined, sintered, MIM, or forged: how do you choose a gear route?
Gears are the classic near-net-versus-machining battleground because their tooth geometry is expensive to produce either way. A machined gear from bar or blank gives you the full control of the process: the tooth profile is cut to the drawing, the bore and keyway sit in true position to the pitch line, and the material is wrought steel in a controlled heat-treated condition. That combination makes the machined gear the benchmark and the default at low and medium volumes, and it is the only route when the tooth tolerances demand the machining band.
Powder-metal gears and MIM gears compete at high volume of a settled design, forming teeth near-net in quantity and saving the material waste of cutting them. They lose on density, on the tolerances of the as-formed teeth, and on the requirement for secondary sizing or machining of critical features. Forged gears earn their place where the load demands the grain flow and strength of a forging, and they too need machining on the bore, faces, and often the teeth. The comparison that decides is the finished cost per part at your real volume with tooling on its own line, and the revision cost, because a gear drawing that changes after the powder-metal tool is cut is money and weeks, while a machined gear absorbs it as a program change.
The quality conversation on gears is about measurement as much as machining. A gear drawing carries tooth-profile, spacing, and runout callouts that are meaningless without the instruments to verify them, and a supplier that cannot measure a gear cannot make one to a drawing — the CMM and gear measurement capability are part of the qualification, not a footnote. Ask how tooth spacing, profile, and runout are verified, and whether the inspection report carries measured values rather than a conformity line. On a near-net route the same question applies to the as-formed teeth: a powder-metal or MIM gear is only as good as the measurement that proves its form, and the secondary machining of the bore and faces has to be checked in the same datum frame as the teeth.
Brass, bronze, or plastic: which route wins for a plain bushing?
Plain bushings are where material choice and process choice are nearly the same decision, because the bearing material and the way it is formed travel together. A machined bushing from brass or bronze bar carries higher loads, runs against a steel shaft with lubrication, and holds the bore tolerance that the fit demands; bronze, in particular, handles wear and marginal lubrication better than brass and is the standard for heavier service. When the load is light and the speed moderate, a plastic bushing — machined from POM, nylon, or PTFE-bearing stock, or molded at volume — gives self-lubricating service with lower weight and cost, and its bore is machined or reamed to the fit it needs.
The engineering input that decides is the load-speed-lubrication triad, and the drawing should name all three. A bushing specified without the shaft material, the load, and the speed leaves the supplier to guess at a material that may wear out or seize in service. When the application is borderline, the honest approach is to test the candidate materials on the actual duty cycle before committing volume, because bearing materials fail by wear and temperature in ways that no drawing tolerance captures. Machined metal is the flexible default while the application is being proven; the molded or sintered route earns its place once the duty cycle is settled and the volume justifies it.
When should a housing be machined instead of cast or fabricated?
Housings carry the same route logic as every other component, scaled by size. A small, dense housing full of bores and mounting faces is a machining part — it starts as plate or billet, and the features that carry the bearings and the mating components are produced to the tolerance the assembly needs. A large, thin housing that would machine mostly into chips is a candidate for die casting at volume or sheet metal fabrication at larger sizes, and in each case the datum faces, bores, and threads still get machined afterward, which is why the finished cost is the number to compare.
The other variable is design maturity. A die-cast housing commits you to tooling the moment the design freezes; a machined housing stays flexible until it ships. For most programs the sensible sequence is to machine the early housings, prove the design through testing and field use, and then evaluate the die-cast or fabricated route once the geometry is truly settled and the volume justifies the tooling. The supplier that tells you this honestly — that machining is the right place to be while the design is moving — is the one to keep when the design does settle and the volume arrives.
When does a plastic component beat machined metal, and what changes in the design?
Plastics win a mechanical component when the loads, temperatures, and environment fit the polymer and the volume justifies the route. A molded plastic lever, bracket, or housing at high volume of a settled design is often dramatically cheaper than the machined metal equivalent, because the mold produces the shape in seconds and the material cost per part is low. A machined plastic component — PEEK, POM, nylon — wins at low volume and while the design moves, holding tighter tolerances than molding with no tooling commitment. The crossover sits wherever the mold amortizes below the machining premium.
The trap is substitution without redesign. A plastic component is not a metal component in a different material: the wall thicknesses change, ribs and bosses carry the load, the fastening details differ, and the thermal expansion and creep behavior are different engineering. When a design was engineered for metal, switching it to plastic to save money usually fails in testing and costs more than the saving. The material-process decision belongs at the design stage, with the function requirements on the table, and a machining supplier's DFM review is a cheap place to test the assumption before the mold is cut.
How do tolerance and finish requirements on mating features choose the route?
Every mechanical component has features that mate, and those features carry the tolerances that decide the process. Machining holds fit and alignment features to ±0.003 mm, with capability of Cpk 1.67 or higher on key characteristics and Cpk ≥ 1.33 on general characteristics. Capability is reported per characteristic: every report names the feature, its tolerance band, the sample size, the sigma method, and the study window and precision grinding reaches ±0.002 mm where a fit demands roundness and finish that turning or milling cannot hold. Casting, forging, MIM, and molding deliver near-net shapes in looser bands, which means the bore that carries a bearing, the face that locates an assembly, and the shaft journal that runs in a bushing are machined features regardless of where the component's shape starts.
The drawing should make that division explicit. Put the tight callouts on the mating features and run the general envelope at a block tolerance, so the quote prices the real work rather than the worst case. When you compare a machined component against a near-net route, ask each supplier which features it will machine and to what tolerance, because the route comparison only means something when the finished features match the drawing. A near-net blank that saves material but cannot hold the bore tolerance is not a cheaper route — it is a blank that still needs the machine time you thought you removed.
The finish on mating features belongs in the same callout as the tolerance. A press-fit diameter machined to the tight band is still not interchangeable if its surface finish lets it gall in assembly or its edge carries a burr, so the drawing should state the finish and the edge break on the same features it tolerances. Heat treatment moves those same features: a gear that is case-hardened after machining changes its dimensions, which is why the finish grinding or honing of the bore and the teeth happens after the hardening step and carries its own certificate in the lot record. When you compare routes, compare the finished, treated, and measured component, because that is what actually assembles into your product.
How do you compare total cost across routes when sourcing from China?
Compare finished components at your real quantity with every element visible. The quote should separate material in its form, machining, tooling where it exists, secondary operations such as heat treatment and finishing, inspection, and freight — a supplier that hides tooling in the unit price is hiding the crossover where a near-net route would win. For a gear, put the material in its form: a machined gear buys bar or blank and cuts teeth into it, while a powder-metal blank starts near the tooth profile and trades material for tooling. The form decides how much material you buy, and the finished cost is the only number that compares.
Then add the two costs that never appear on a quote: qualification and revision. A new die, mold, or forging tool has a first-article cycle and a change cost that a machined component does not, and switching routes later means new tooling and re-approval. When the numbers are on the table, ask the supplier for the honest recommendation — a mechanical-components supplier worth keeping will tell you when a near-net route is the cheaper answer, because that honesty is what keeps the rest of your work coming back. The services page on this site explains how machining, finishing, and heat treatment are combined into a route, and the material-to-process FAQ covers which alloy pairs with which process for common components.
Frequently Asked Questions
When should a gear be machined rather than sintered, MIM, or forged?
Machine it at low and medium volume, when tooth tolerances demand the machining band, or when you need wrought or heat-treated steel properties. Powder-metal and MIM gears win at high volume of a settled design; forged gears win where grain flow carries the load. The machined gear is the benchmark every near-net route must beat on finished cost.
Brass, bronze, or plastic: which is right for a plain bushing?
The choice follows load, speed, and lubrication. Brass and bronze carry higher loads against steel shafts with lubrication; bronze handles wear better than brass. Plastic bushings suit lighter loads, lower speeds, and self-lubricating service at lower cost and weight. Name the shaft material, load, and speed on the drawing and the material choice follows.
When should a housing be machined instead of cast or fabricated?
Machine it at low volume, when it is small and dense with bores and mounting faces, or while the design is still changing. Die casting wins at high volume; sheet metal wins for large thin enclosures. Cast and fabricated routes still need machining for datums and fits, so compare the finished cost, not the blank price.
When does a molded plastic part beat machined metal for a structural component?
When the loads, temperatures, and environment fit the polymer and the volume justifies a mold. Molded plastic wins at high volume of a settled design; machined plastic or metal wins at low volume and while the design moves. Substituting plastic into a metal design without redesign usually fails — the decision belongs at the design stage.
What tolerance can each route deliver on a mechanical component?
Machining holds fit and alignment features to ±0.003 mm, with Cpk of 1.67 or higher on key characteristics and Cpk ≥ 1.33 on general characteristics. Capability is reported per characteristic: every report names the feature, its tolerance band, the sample size, the sigma method, and the study window., and precision grinding reaches ±0.002 mm. Casting, forging, MIM, and molding deliver near-net shapes in looser bands, so the features that mate and seal are machined no matter where the shape starts. Where the tight band rules, machining closes the route.
How do I compare a machined component against a near-net route fairly?
Compare finished parts at your real quantity with tooling on its own line. A near-net blank needs secondary machining for datums, bores, and threads, so the comparison is finished cost, not blank price. Add qualification and revision risk — a new tool costs money to change, while a machined part absorbs a revision as a program edit.
What is the cheapest way to test a material-process choice before committing volume?
Run a prototype in the production material through the closest practical process and measure it against the drawing. If the production path is a casting or molding, a machined prototype validates function but not the production process — budget for a first article from the actual route. The sample is cheaper than the discovery after a full order.
Not sure which material-process route wins for your component?
Email [email protected] or call +86 13126536955 for an honest route comparison — including the cases where machining is not the cheapest answer.
