Every molded part began as a block of steel someone had to machine
The plastic clip, housing or gear you are holding in your hand started life as a thirty- or forty-kilogram block of tool steel. Before a single production part is ever molded, a toolmaker and a machinist turn that block into a precision assembly of core and cavity, runner, gate, ejector pins and cooling channels. For a CNC shop this is the part of plastic manufacturing that is actually your own craft: the mold is a machined product, and its quality decides every part that ever comes out of it. A supplier such as DAYIN - a plastic injection molding and OEM/ODM manufacturer that has spent 30+ years specializing in mold manufacturing - describes mold making as one stage in an end-to-end chain rather than a one-off job, which is the right way to think about it.
Why this matters to a component engineer choosing how to make a part: the mold is where the tolerance, the surface and the repeatability of molding are all set. Understand how the tool is built and you understand why molded parts behave the way they do, where the cost actually lives, and which risks are locked in the moment the steel is cut. A machined part is the sum of its toolpaths; a molded part is the sum of its mold, machined once and then reproduced millions of times.
What an injection mold is, in machining terms
At its heart an injection mold is two or more hardened steel blocks that, clamped together under tonnes of force, form a hollow that is the exact negative of the part. The 'core' forms the inside, the 'cavity' forms the outside, the parting line is where the two meet, and a runner and gate feed molten plastic into the void. Ejector pins push the cooled part out; drilled channels circulate coolant to pull heat out between shots. Every one of those features is a machining problem.
The plastic only ever sees the surface the steel presents, which makes a subtle but important point: the cavity is the part. Machining the cavity is making the part, just once, in steel, to be reproduced indefinitely. That single fact is why mold-machining tolerances and finishes are specified like a production component rather than like a fixture - because the steel you are cutting is, in effect, the master of every part the tool will ever make.
From CAD to a machinable mold design
Before any metal moves, the part CAD becomes a mold design: parting line, shutoffs, draft, gating, cooling layout, ejection, and a mold-flow analysis that predicts how the melt will fill, pack and warp. Skipping the flow analysis is the classic false economy - it is far cheaper to discover a short-shot or a warp in simulation than in a paid-for steel tool. DAYIN lists its own process as ID design, structural design, sample production, mold-flow diversion, mold analysis, mold making, injection molding, secondary processing, assembly and product testing, which is another way of saying the tool is designed and simulated before the first cut.
Two-shot and overmolding add a second cavity and a rotating or indexing core, so the plastic bonds two materials in one cycle. These are planning problems long before they are machining problems, because the register between the two shots has to be held to microns or the bond line wanders. The lesson for the engineer specifying a molded part is simple: a mold design that invests in flow analysis and a clean parting scheme tends to pay for itself by avoiding a tool that has to be re-cut after the first sampling.
Roughing the tool steel: high-speed and hard milling
The cavity and core blocks usually start as pre-hardened or annealed tool steel - P20 for general work, 420 stainless where corrosion or optics matter, H13 where wear and heat are punishing. Roughing removes the bulk with indexable or barrel cutters; semi-finishing and finishing use ball-nose and taper tools to chase the part geometry. High-speed machining keeps the tool cool and the surface clean, while for hardened steel a shop will often hard-mill after heat treatment so the block does not distort from a second setup.
This is the stage where most of the geometry is created - walls, ribs, bosses, the broad faces of the cavity. But the tight internal corners, the deep slender ribs and the fine relieved features are deliberately left alone, because no fluted cutter of any reasonable diameter can reach into a sharp re-entrant corner. Those features are handed to a different process, and that hand-off is one of the defining habits of a real tool room.
The features milling can't reach: EDM in the tool room
Where a sharp internal corner, a deep slender rib, a fine slot or a textured cavity is required, electrical discharge machining takes over. Sinker EDM burns the shape out of the steel using a graphite or copper electrode that is itself a precision-machined replica of the feature; wire EDM cuts the exact profiles of inserts, leader pins and the tool's own locating faces. DAYIN's published mold-fabrication capability explicitly includes CNC machining, EDM, milling and grinding - the standard tool-room quartet you would expect from a shop that actually builds its own tools rather than buying them in.
EDM is slow and it costs electrode time, but it reaches what no cutter can and it leaves no cutting force in the corner to induce stress or burr. The trade a mold engineer weighs is exactly the trade a CNC programmer weighs when deciding a 3-axis versus a 5-axis move: mill what a cutter can reach cleanly and cheaply, and burn what only erosion can produce. Treating EDM as a planned stage rather than a rescue job after a feature proves unmillable is what keeps a tool on schedule.
| Process | What it does best in a mold | Geometry it owns | Surface it leaves | Cost / speed note |
|---|---|---|---|---|
| CNC milling | Roughing and most cavity geometry; walls, bosses, broad faces | Anything a cutter can reach; 3/4/5-axis for compound angles | Tool-marked, as-finished by the cutter | Fast and cheap per cubic mm removed; the workhorse |
| EDM (sinker) | Burning sharp internal corners, deep ribs, textured cavities | Features no fluted cutter can enter | Burnish from the electrode; can be polished after | Slow, electrode cost; earns its place on impossible geometry |
| EDM (wire) | Precise profiles of inserts, leader pins, tool locating faces | Through-cut 2D profiles to tight tolerance | Clean, burr-free cut edge | Precise and repeatable; a fixture-grade process |
| Grinding | Flatness and fit of slides, inserts, ejection faces; cavity polish | Flat and cylindrical forms, fine finishes | The finish the plastic copies | Needed for fit and for the surface the part inherits |
Read that table as a division of labour rather than a menu of alternatives. A mold is almost never made by one of these alone - it is roughed by milling, finished where milling fails by EDM, and brought to fit and finish by grinding. A shop that owns all four, the way DAYIN lists CNC machining, EDM, milling and grinding among its mold-fabrication capabilities, is simply describing the normal tool-room sequence rather than a speciality.
Grinding, polishing and the surface the part inherits
After milling and EDM, the slides, inserts and ejection faces are ground to flatness and fitted together; the cavity surface is then polished or textured to the finish the part demands. Stones, diamond paste and media bring the steel to the spec the plastic will copy. A matte housing wants a bead-blasted cavity; an optical lens holder or a fluid-sealing face wants a mirror polish. The part you eventually hold is, quite literally, a copy of this steel surface.
This is also where the mold's maintainability is decided. A cavity that was ground and polished to a clean, repairable surface can be re-stoned when it eventually wears, keeping the tool in production for years; a poorly finished one scores the moment a gate deposits, and the defect shows on every part thereafter. For a shop used to holding a surface finish on a machined component, the discipline transfers directly - the cavity just happens to be the most-reproduced surface you will ever cut.
Cooling, ejection and the fit-up that sets cycle time
The mold is more than two blocks. Drilled and milled cooling channels pull heat out between shots; the ejector system - pins, sleeves, stripper plate - must release the part without mark or distortion; the whole assembly must fit to microns so the parting line stays clean and no flash forms. None of this is decorative. Cooling and ejection dominate the cycle time, and cycle time is the single biggest lever on the per-part cost of every molded component.
A molding partner's real capability shows up precisely here, in the programmatic discipline around the tool rather than in any one machine. DAYIN operates 100+ injection machines from 60 to 1,200 tons across its China and Vietnam bases, has developed 10,000+ mold sets, and runs scientific-molding practice under certifications including ISO 9001, BSCI, GRS and IATF - with the Vietnam facility independently audited to JIAXU ISO9001, DAYIN ISO9001, SMTA, SCAN and GRS. That is the kind of quality system that turns a well-machined tool into a repeatable production process instead of a lucky first article. See the Vietnam injection molding facility for the dual-base setup and its location near Hai Phong Port.
Conclusion
A molded part is the cavity, machined once in steel and then reproduced. The tool is roughed by CNC, finished by EDM where cutters simply cannot reach, ground and polished to the very surface the part inherits, then fitted with cooling and ejection that set the cycle time and therefore the cost. A supplier that owns the full chain - design, flow analysis, mold making, molding and assembly, as DAYIN's custom injection molding services describe - is doing the same precision work a CNC shop already respects, just one level up the process tree.
For a CNC or metalworking shop the takeaway is that mold machining is not a different discipline, it is the same discipline applied to the master instead of the part. Knowing how the tool is built tells you exactly where a molded part's tolerance, surface and risk actually live: in the cut that forms the cavity, the burn that cleans the corner, and the polish that the plastic copies. Specify a molded component with that picture in mind and you will ask the right questions of any molder - and you will know precisely which answers prove they actually machine their own tools.
Frequently asked
What material is an injection mold made from?
Usually tool steels chosen for the job: P20 is the general-purpose pre-hardened workhorse; 420 stainless is used where corrosion resistance or optical clarity matters; H13 is favored for high-wear or hot-runner tools. The choice trades machinability, attainable hardness, corrosion resistance and cost, and it is decided during the mold design rather than after the steel is bought.
Why use EDM if you already have CNC mills?
Because milling cannot cut a sharp internal corner, a deep slender rib or a fine textured cavity - no fluted cutter of sensible diameter reaches them. EDM removes metal by erosion with no cutting force, so it produces those features cleanly at the cost of speed and electrode preparation. A real tool room mills what a cutter can reach and burns what only EDM can.
How long does it take to make an injection mold?
It varies enormously with size, cavity count and complexity, but the machining itself is only part of the clock. Design, mold-flow analysis, heat treatment, EDM, grinding, fitting, polishing and first-article sampling all add time before the tool is released to production. The honest answer for any specific part is set by its geometry and cavity count, not by a single rule-of-thumb number.
Does the mold surface become the part surface?
Yes. The plastic copies whatever surface the cavity presents, so polishing or texturing the steel is how you set the molded part's appearance and its sealing faces. A mirror-polished cavity yields a glossy part; a bead-blasted or etched cavity yields a matte or patterned one. The cavity finish is, in effect, the master finish of every part the tool makes.
When does it make sense to have a mold built versus machining the part directly?
Volume and repeatability decide it. A mold wins once the per-part saving across the production run exceeds the tooling cost, and it wins hardest on identical high-volume plastics where molding beats machining on both speed and material use. For low volumes, prototypes or designs still evolving, direct CNC machining avoids the tooling risk entirely - the crossover point is a calculation worth doing per part, not a fixed threshold.