Making a plastic injection mold involves designing two mating halves of a metal (or sometimes polymer) tool, machining precise cavities into them, and integrating systems for feeding plastic, cooling it, and ejecting the finished part. The process ranges from simple single-cavity prototypes you can produce in a small shop to complex multi-cavity production tools that require specialized equipment and months of lead time. The approach you choose depends on how many parts you need and how much you’re willing to invest.
Choosing Your Mold Material
The material you build the mold from determines how long it lasts, how fast it cycles, and how much it costs. The two most common choices are aluminum and tool steel, and they serve very different purposes.
Aluminum alloys (particularly 7075-grade) are the go-to for prototype and low-volume molds. Aluminum has a thermal conductivity of about 130 W/m·K, nearly three times that of P20 tool steel at 45 W/m·K. That means aluminum molds pull heat out of the plastic faster, shortening cycle times. They’re also much easier to machine, so you can have a finished mold in days rather than weeks. The trade-off is durability: aluminum molds typically handle thousands to tens of thousands of shots before wearing out, while hardened steel molds can run into the hundreds of thousands or even millions.
P20 steel is the workhorse of production mold making. It’s a pre-hardened, low-carbon steel that machines reasonably well but holds up far longer than aluminum under repeated injection pressure and heat. Raw aluminum costs roughly 3.5 times more per pound than P20 steel, but because aluminum machines faster, the total cost of a simple aluminum mold is often lower. For high-volume production, steel pays for itself through longevity.
If you only need a handful of parts for testing fit and function, 3D-printed mold inserts are an option worth considering. Inserts made from high-temperature resins using processes like PolyJet or FDM can survive around 15 injection cycles before the cumulative pressure permanently deforms them. Research comparing printed epoxy-acrylate inserts to aluminum found that the printed versions experienced strain exceeding 2% during molding, while aluminum inserts stayed below 0.03% under the same conditions. Printed inserts are strictly for validating a design, not for production.
Designing the Part for Moldability
A mold is only as good as the part design it’s built around. Several features need to be locked in before you start cutting metal.
Draft Angles
Every vertical wall in your part needs a slight taper so the finished piece can slide out of the mold without sticking or scraping. A minimum of 0.5 degrees on all vertical faces is strongly recommended, though 1 to 2 degrees works well in most situations. If your mold has shutoff surfaces where metal slides against metal, plan for at least 3 degrees. Textured surfaces need even more: 3 degrees minimum for a light bead-blast finish and 5 degrees or more for a heavier texture. Without adequate draft, you’ll damage parts on ejection and wear out the mold prematurely.
Wall Thickness
Uniform wall thickness prevents warping, sink marks, and incomplete fills. The minimum depends on the resin you’re molding. ABS needs walls of at least 0.045 inches (about 1.14 mm), while nylon can go thinner at 0.030 inches (0.76 mm). Thicker isn’t always better, either. Overly thick sections cool slowly and create internal voids. Keep walls as consistent as possible throughout the part, and use ribs or gussets instead of adding bulk where you need strength.
Shrinkage Compensation
Every plastic shrinks as it cools, and you need to oversize the mold cavity to account for it. The amount varies dramatically by material. ABS shrinks a modest 0.4 to 0.9%, so a 100 mm feature in the mold produces a part roughly 99.1 to 99.6 mm long. Polypropylene shrinks far more, at 1.0 to 2.5%. Polyethylene is the most variable: HDPE shrinks 1.5 to 4.0% and LDPE can shrink as much as 5.0%. Glass-fiber reinforcement cuts shrinkage significantly. ABS with 30% glass fiber, for example, shrinks only 0.2 to 0.3%. Your CAD model of the mold cavity must scale up every dimension by the expected shrinkage rate for the specific resin you plan to run.
Anatomy of the Mold
A standard injection mold has two main halves: a stationary side (the “A” side) that faces the injection machine’s nozzle, and a moving side (the “B” side) that pulls away to release the part. These halves are mounted in a mold base, sometimes called the mold frame, which provides the structural support and houses all the working components.
The sprue bushing sits on the stationary side and acts as the interface between the machine’s nozzle and the mold’s internal channel system. Molten plastic flows through the sprue, into runners, and through gates into the cavity. Guide pins and bushings on both halves ensure the two sides align precisely every time the mold closes. Even a fraction of a millimeter of misalignment creates visible parting lines or flash on the finished part.
On the moving side, ejector pins push the cooled part out of the cavity when the mold opens. These pins retract flush with the cavity surface during injection and only extend when the B side pulls back. Their placement matters: put them under flat surfaces or structural ribs where the small circular witness marks they leave won’t affect appearance or function.
Machining the Mold Cavities
Two processes do the heavy lifting in mold making: CNC milling and electrical discharge machining (EDM). Most molds require both.
CNC milling handles the bulk material removal. A series of progressively finer cutting tools carve out the rough cavity shape, then refine it to near-final dimensions. Modern CNC machines achieve tight tolerances and good surface finishes on most geometry, but they have a fundamental limitation: because the cutting tools are round, they leave radiused internal corners. If your part design requires sharp internal edges, CNC alone won’t get you there.
That’s where EDM comes in. Instead of a spinning cutter, EDM uses electrical sparks to erode metal away. A shaped electrode (typically copper or graphite) is plunged into the workpiece, and controlled discharges vaporize material in the exact shape of the electrode. This process excels at creating sharp internal corners, deep narrow ribs, and fine details that would break a cutting tool. It also works equally well on hardened steel, since the material’s hardness doesn’t limit the process the way it does with conventional machining.
EDM naturally produces a surface covered in tiny craters, but by running at low power and slow speed, the process can achieve a mirror-like surface finish of around 4 micro-inches Ra, smooth enough for optical-quality mold surfaces without additional polishing. Unlike CNC milling, EDM leaves no directional tool marks, which matters for parts where surface appearance is critical.
Designing the Gate System
The gate is the point where molten plastic enters the cavity, and its type and placement affect fill quality, cycle time, and the appearance of the finished part.
Edge gates are the simplest option. They sit along the parting line at the edge of the part and work well for larger components or parts with thicker wall sections that need generous flow. They leave a visible mark at the gate location, but that mark is on the edge where it’s easy to trim.
Tunnel gates (also called submarine gates) are cut below the parting line at an angle. When the mold opens, the runner system shears automatically at the gate, so you don’t need a secondary trimming step. They’re ideal for small components and parts where the gate mark needs to be hidden, but they can struggle to fill large cavities.
Fan gates spread the flow across a wide opening, making them a good choice for parts with large flat surfaces or complex geometries where you need even filling to avoid weld lines and warping. They require more runner material, which increases waste unless you’re using a hot runner system.
Adding Cooling Channels
Cooling accounts for the majority of the injection molding cycle. How you route coolant through the mold directly impacts part quality and production speed.
Conventional cooling uses straight-drilled channels bored through the mold plates. These are simple and cheap to produce, but their straight-line geometry can’t follow the contours of complex parts. Areas far from a cooling channel cool slowly, creating hot spots that cause warping or longer cycle times.
Conformal cooling channels follow the shape of the cavity at a consistent distance from every surface. They deliver more uniform cooling, which reduces warping and improves dimensional accuracy. The challenge is manufacturing: conformal channels are too complex to drill conventionally, so they’re typically produced by metal 3D printing (direct metal laser sintering) or by building up mold sections in layers. The investment pays off for complex parts where cycle time and quality are priorities.
Assembly and Testing
Once all components are machined and finished, the mold is assembled into its base, the cooling lines are connected, and the ejector system is set up. Before running production parts, the mold goes through a series of test shots, often called “T1 sampling.” These initial shots reveal problems like short fills, flash at the parting line, sink marks, or poor surface finish.
Expect to make adjustments. Gate sizes may need to be opened up or relocated, vents may need to be deepened to let trapped air escape, and cooling channels may need flow adjustments to balance temperatures across the cavity. Most molds go through two or three rounds of sampling and correction before they’re production-ready. Each round involves running parts, measuring them against the print, and making targeted steel modifications. This iterative process is normal and should be factored into your timeline and budget from the start.