Injection Molding 101
The workhorse of mass production. Plastic pellets in, finished parts out — at scale, nothing matches it on cost per unit.
Injection molding is a manufacturing process where molten plastic is forced into a precision steel mold under high pressure. The plastic cools and solidifies in seconds, the mold opens, and a finished part is ejected. The same mold can produce hundreds of thousands of identical parts — or even millions — with near-zero variation.
It is the most common way to make plastic consumer products at scale. If you have ever held a phone case, a power tool housing, a toy, or a kitchen gadget, you have held an injection-molded part. For hardware founders, understanding molding is not optional — it directly determines your product’s cost, quality, and speed to market.
The process starts with granular plastic resin fed into a heated barrel. A rotating screw melts the plastic and pushes it forward. When enough material has accumulated in front of the screw, it rams forward at high pressure — typically 10,000 to 30,000 psi — injecting the molten plastic into a closed steel mold.
The mold consists of two halves: the cavity side (the outer shape) and the core side (the inner shape). Cooling channels inside the mold circulate water to bring the temperature down quickly. Within seconds to a minute — depending on wall thickness — the part solidifies. The mold opens, ejector pins push the part out, and the cycle repeats.
Mold types vary by complexity. A two-plate mold is the simplest and cheapest: the part and runner system come out together. A three-plate mold separates the runner from the part automatically. Hot-runner molds keep the plastic molten in the feed system, eliminating runner waste entirely — worth it for high volumes but more expensive to build.
Material choice is critical. ABS is the most common general-purpose plastic — tough, impact-resistant, and easy to mold. Polycarbonate (PC) is clear and strong but flows less easily. Polypropylene (PP) is cheap, flexible, and great for living hinges. Nylon (PA) offers excellent wear resistance but absorbs moisture. TPU and TPE give you soft-touch overmolded grips.
What goes wrong
Sink marks
Depressions on the surface where thick wall sections cool slower than surrounding areas. Fix: keep wall thickness uniform. If you need a thick section, core it out from the back.
Warping
The part twists or bends after ejection because different areas cooled at different rates. Fix: keep walls consistent, add ribs for structure instead of thickening walls.
Flash
A thin layer of plastic escapes between mold halves at the parting line. Fix: ensure clamping force is sufficient, and the mold fit is tight. Could also mean the mold is worn.
Short shots
Plastic does not fill the entire cavity. Fix: increase injection pressure or temperature, add more gates, or reduce flow distance from the gate.
Knit lines
Visible seams where two flow fronts meet around a hole or core pin. Cannot be eliminated, but can be minimized with gate placement and material selection.
What founders should remember
Mold cost is front-loaded, unit cost is nearly zero
A mold costs $2K–$50K+ depending on size, complexity, and cavitation. But once it is running, per-part cost can be pennies. This is why injection molding only makes sense above ~1,000 units.
Uniform wall thickness is the golden rule
Every DFM (Design for Manufacturing) conversation starts here. Aim for 1.5–3 mm for most consumer products. Varying thickness causes warping, sink, and longer cycle times.
Draft angles are non-negotiable
Vertical walls need at least 0.5–1 degree of taper so the part can release from the mold. Without draft, the part sticks and the mold is useless.
Undercuts add cost dramatically
An undercut is any feature that prevents the part from being pulled straight out of a two-part mold. Solutions exist — side actions, lifters, collapsible cores — but each one adds thousands to mold cost.