How to Make an Injection Mold for Plastic Parts

Making an injection mold means designing and machining a precision metal tool that splits into two halves, forms a sealed cavity in the shape of your part, and survives thousands (or millions) of cycles under high pressure. The process spans CAD design, steel selection, CNC machining, and validation trials. A simple single-cavity mold for a small part can cost as little as $2,000 to $6,000, while complex multi-cavity production tools run $50,000 to $100,000 or more. Understanding each stage helps you make better decisions whether you’re building a mold yourself, working with a toolmaker, or sourcing overseas.

How an Injection Mold Is Structured

Every standard injection mold has two halves. The A-side (also called the cavity side) mounts to the fixed side of the molding press. The B-side (the core side) mounts to the moving clamp side. When the press closes, these two halves meet along a flat surface called the parting line, forming a sealed pocket in the shape of your part.

The cavity is the hollow space that shapes the outside surface of your part. The core is the solid form that shapes the inside. Think of a drinking glass: the outside would be formed by the cavity in the A-side, and the inside would be formed by the core on the B-side. When the mold opens, the part naturally clings to the core because plastic shrinks onto solid features as it cools.

That’s where the ejector system comes in. The B-side contains an ejector plate with a set of pins that push the finished part off the core once the mold opens. The press’s clamp mechanism actuates these pins automatically. Beyond these major components, the mold also includes a sprue (the channel where molten plastic first enters), runners (channels that distribute plastic from the sprue to the part), and gates (the narrow openings where plastic enters the cavity itself).

Designing the Part for Moldability

Before you cut any steel, the part design needs to work within the constraints of the molding process. Three principles matter most: draft angles, uniform wall thickness, and gate placement.

Draft is the slight taper applied to vertical walls so the part can slide out of the mold without sticking. A minimum of 0.5 degrees on all vertical faces is strongly advised, but 1 to 2 degrees works well in most situations. Textured surfaces need more: 3 degrees minimum for a light bead-blast finish, and 5 degrees or more for a heavier texture. Without enough draft, the part will drag against the mold surface during ejection, leaving scuff marks or refusing to release at all.

Wall thickness should be as consistent as possible throughout the part. Thick sections cool more slowly than thin ones, creating uneven shrinkage that leads to warping, sink marks, and internal stress. Where you can’t avoid thickness changes, use gradual transitions rather than abrupt steps.

Choosing the Right Gate Type

The gate controls how plastic enters the cavity, and the type you choose affects fill quality, cycle time, and cosmetics. Three common options cover most situations.

  • Edge gates sit along the parting line and are the simplest to machine. They can have a larger cross-sectional area than other gate types, which makes them ideal for filling bigger parts or sections with thicker walls. The larger opening also allows longer hold times, meaning the press can continue packing plastic into the cavity as it cools. After molding, you trim the gate manually, but the high-shear zone stays in the gate area itself, so stress marks are removed along with it.
  • Tunnel (submarine) gates are machined below the parting line so the gate shears off automatically when the part ejects. This makes them popular for small parts, high-cavity molds, or any application where manual trimming isn’t practical. The tradeoff is size: too large a tunnel gate can crack or leave cosmetic blemishes from the automatic shearing.
  • Fan gates are a variation of the edge gate with a wider opening that spreads the flow across a broader front. This improves dimensional stability and can reduce flow marks on cosmetically sensitive parts.

Designing the Cooling System

Cooling accounts for the majority of each molding cycle, so the layout of cooling channels inside the mold directly affects both part quality and production speed. The goal is uniform, balanced cooling across every surface of the part. Uneven cooling causes differential shrinkage, which shows up as warping, sink marks, or residual stress that weakens the part over time.

Conventional cooling channels are straight-drilled holes running through the mold plates. They’re simple to manufacture but can’t always follow complex part geometry, leaving hot spots in areas they can’t reach. Conformal cooling channels, which follow the contours of the part surface, solve this problem but require more advanced manufacturing methods like metal 3D printing or specialized machining.

The key variables are the distance between each channel and the part surface, the spacing between adjacent channels, and the channel diameter. These dimensions vary by material and part size. For example, a small ABS part might use channels spaced 6 mm apart and placed about 3.75 mm from the cavity surface, while a larger polypropylene part might need 14.4 mm spacing and 8.35 mm of distance from the surface. Getting these numbers wrong means longer cycle times at best and warped parts at worst. Mold flow simulation software can model cooling performance before you commit to cutting steel.

Machining the Mold

Most injection molds are built using high-speed CNC milling. A block of tool steel is clamped into a machining center, and cutters progressively remove material to form the cavity and core geometries. Modern five-axis CNC machines can handle complex 3D surfaces with tight tolerances, but they have limits: the rotating cutter always leaves a radius in internal corners, and it can’t reach into very deep, narrow features like thin ribs.

That’s where electrical discharge machining (EDM) comes in. EDM uses controlled electrical sparks to erode metal rather than cut it, which means it can produce sharp internal corners, V-shaped features, and deep narrow slots that milling tools can’t reach. A shaped electrode (typically copper or graphite) is sunk into the steel workpiece, and the spark gap removes material with extreme precision. Most mold shops use EDM selectively for features that standard milling can’t produce, not for the entire mold.

Additional operations include surface grinding for flat mating surfaces, polishing the cavity to the required surface finish, and fitting components like ejector pins, guide pins, and cooling line fittings. If the part has features that can’t be formed by the two main mold halves alone (undercuts, holes perpendicular to the mold opening direction), you’ll need side actions or lifters, which are moving components inside the mold that retract before ejection.

Selecting Mold Steel and Class

The plastics industry classifies molds into five tiers based on expected production volume. These SPI (Society of the Plastics Industry) classes determine what steel to use, how robust the mold needs to be, and ultimately what it will cost.

  • Class 105 (prototype): built for 500 cycles or fewer. Often made from aluminum or soft steel. Used to validate part design before investing in production tooling.
  • Class 104 (low volume): up to 100,000 cycles. Suitable for limited production runs or parts with short market lives.
  • Class 103 (medium volume): up to 500,000 cycles. A common choice for mid-range production.
  • Class 102 (high volume): up to one million cycles. Built with hardened steel, good for abrasive materials or parts requiring close tolerances.
  • Class 101 (extremely high volume): over one million cycles. The most robust and expensive tier, using premium hardened steels throughout.

Common mold steels include P20 (a pre-hardened steel for general-purpose molds), H13 (a hardened tool steel for higher volumes and abrasive plastics), and S136 stainless (used when corrosion resistance matters, such as molding PVC or when optical clarity requires a mirror-polished cavity).

3D Printed Mold Inserts for Prototyping

If you need a handful of parts to test fit and function, 3D printed mold inserts offer a fast, low-cost alternative to machined steel. These inserts fit into a standard mold base and form the cavity and core for short runs.

Polymer-based printed inserts (made from filled resins or extruded thermoplastics) are the cheapest option, but they deform quickly under injection pressure. In testing, optimized designs with 100% infill and reinforced shutoff surfaces lasted about 15 cycles before the cumulative pressure permanently deformed the insert, causing flash along the parting line. Parts molded from these inserts also tend to have lower mechanical properties than parts from steel tooling, because the insert can’t withstand the same packing pressures.

Metal inserts made through powder bed fusion (a metal 3D printing process using steel or bronze powder) perform much closer to traditionally machined tools in both longevity and dimensional accuracy. They’re significantly more expensive than polymer prints but can serve as bridge tooling while a production mold is being built.

Testing and Validating the Mold

Once the mold is assembled, it goes through a series of trial runs before it’s approved for production. The first meaningful milestone is the T1 trial, which gives you your first real look at how the mold performs under production-like conditions.

During T1, you’re examining the molded parts for appearance, dimensions, and functional fit. The surface should be free of short shots (incomplete fills), burn marks, whitening, flow lines, flash (thin excess plastic along the parting line), bubbles, weld lines, and sink marks. Dimensions are measured against the part drawing to check that shrinkage is within tolerance. If the part has snap fits, threads, or mating surfaces, those get tested for functional compatibility.

It’s rare for a T1 sample to be perfect. Most molds require at least one round of adjustments: tweaking gate size, modifying venting to let trapped air escape, adjusting cooling to address warping, or removing steel in areas where dimensions are too tight. A T2 trial follows the changes, and sometimes a T3 is needed for particularly demanding parts. Each iteration adds time and cost, which is why investing in thorough design and simulation upfront pays off.

Typical Costs and Timelines

Mold cost is driven by size, complexity, number of cavities, and steel grade. Here’s what to expect across the spectrum:

  • Simple prototype mold (single cavity, small part, aluminum or P20 steel): $2,000 to $6,000
  • Production single-cavity mold (medium part, P20 or similar): $6,000 to $18,000
  • Multi-cavity production mold (4 to 8 cavities, with side actions): $18,000 to $50,000
  • High-cavitation hot runner mold (16 to 32 cavities, precision components): $50,000 to $120,000
  • Large complex mold (parts over 300 mm, hardened steel): $30,000 to $80,000

Domestic mold shops in the U.S. or Europe typically deliver faster but at higher cost. Offshore tooling from China or Southeast Asia can cut the price significantly, but lead times stretch longer when you factor in shipping, communication cycles, and trial sample logistics. For a straightforward single-cavity mold, expect 4 to 8 weeks from design approval to T1 samples domestically, and 6 to 12 weeks offshore. Complex multi-cavity tools with hot runners and side actions can take 12 to 20 weeks regardless of where they’re made.