Injection Molding vs Extrusion — When Do You Choose Each Process, What Parts Does Each Make, and Can the Two Processes Be Combined?

Last updated: 2026-09-03 · Reading time: about 6 min

Injection molding makes discrete 3D parts in a closed cavity; extrusion pushes melt through a die to make a continuous, constant-cross-section profile. Choose by geometry — features that leave the cross-section (ribs, bosses, threads) need injection; a shape that is the same all the way along needs extrusion. Where one part needs both, three hybrid processes combine them.
See also related article.

Key facts

<>

  • Injection: discrete 3D parts — housings, gears, connectors — in a closed mold, ±0.05–0.2 mm, walls to 0.5 mm, draft required. Extrusion: continuous profiles — tube, pipe, seal, trim — through a die, ±0.1–0.5 mm, walls ≥0.8 mm, no draft.
  • The design rules conflict: injection needs uniform walls, draft and side-actions for undercuts; extrusion allows varying walls, no draft, and undercuts in flexible profiles. Design to the process you picked — not to both.
  • Three hybrids combine them: co-extrusion (a multi-layer profile), extrusion plus injection overmolding (a tube with a molded connector), and thermoforming extruded sheet (large, shallow parts).

    • Cycle time — The total time for one injection molding cycle: injection + packing + cooling + mold open/close + ejection. Typical range: 15–60 seconds. Cycle time determines production rate and cost per part — reducing cycle by 5 seconds saves 10–15% annual production cost. See also related guide.

    Where do the design rules conflict — injection against extrusion?

    The two processes share a barrel, a screw and a melt, but the tool imposes opposite rules. Knowing where they diverge tells you which process a part belongs to.

    Design rule Injection molding Extrusion The conflict
    Wall thickness Uniform, within 25%, to 0.5 mm Can vary; ≥0.8 mm to survive haul-off Injection needs uniform walls; extrusion balances varying walls in the die
    Draft angle 0.5–2° on vertical walls to eject None — the profile pulls free continuously Draft is essential in a mold, irrelevant to a die
    Undercuts Need a side-action, which adds cost Allowed in flexible profiles (TPE, TPU) that spring back Rigid undercuts must be molded; flexible ones can be extruded
    Tolerance ±0.05 mm precision to ±0.2 mm ±0.1–0.5 mm — the profile shrinks and warps free Hold tight dimensions in the closed cavity, not the open die
    Part form Discrete 3D, closed cavity Continuous 2D section, cut to length Features that leave the cross-section only injection can form

    Can the two processes be combined — and how?

    When one part needs a continuous profile and a discrete feature, three hybrids make it in fewer steps than molding two parts and assembling them.

    1. Co-extrusion — two or more melts through one die. A multi-layer profile leaves the die as one piece: a rigid layer for strength bonded to a soft layer for a seal, or a virgin skin over a regrind core. The layers combine inside the die, so there is no secondary bonding step.
    2. Extrusion plus injection overmolding — form the profile, then mold onto it. Extrude the tube or strip, cut it to length, place it in an injection mold, and mold a connector, end cap or mounting feature onto it. The profile and the feature come out as one part, with no assembly.
    3. Thermoforming extruded sheet — extrude a sheet, then form it. Extrude a flat sheet, reheat it, and vacuum- or pressure-form it into a large, shallow part — a tray, a cover, an enclosure — too big for an economical injection mold.

    Pick the process the geometry forces, then combine only where it pays. A constant cross-section is extrusion; a 3D part with features that leave the section is injection. When one part needs both, a hybrid beats making two parts and assembling them — but only when the volume justifies the second tool.

    “Buyers send us a profile and ask why we quote a mold instead of a die, or the other way round. The geometry decides before cost ever enters: if the section is constant, it extrudes; if a feature jumps sideways out of that section, it has to be molded. When a part truly needs both, we would rather co-extrude or overmold the profile in one step than hand the customer two parts and an assembly bill.”

    — Project Engineer, aCatMould

    Frequently asked questions

  • Key parameter identified — critical variable confirmed against spec.
  • Tolerance within achievable range — commercial ±0.15mm or tighter as needed.
  • Material matched to application — resin grade selected for end-use environment.
  • Process window validated at T1 — parameters locked after first trial.
  • Inspection method defined — CMM, visual, or functional test specified.

    • Which parts does each process actually make?

    Injection makes discrete, closed 3D parts — housings, connectors, gears, caps, snap-fit enclosures — anything with ribs, bosses, threads or undercuts that a straight die cannot form.

    Extrusion makes continuous constant-section profiles — tubing, pipe, weatherstrip, gasket, cable channel, trim, sheet and film. The profile runs at 1–20 m/min and is cut to length, so anything the same shape all the way along is a candidate.

    For a profile-shaped part, do you extrude or injection mold — where is the break-even?

    Length and volume decide. Short parts (under 50 mm) favor injection — the cycle is fast and the part ejects automatically. Long parts (over 500 mm) favor extrusion — the profile runs continuously and is cut to length.

    For medium parts (50–500 mm), weigh volume: injection tooling costs more up front but the per-part cost stays flat, while extrusion tooling costs far less but adds secondary cutting and end-machining. Once the secondary work is counted, the break-even usually lands in the low thousands of parts.

    How do extrusion defects differ from injection defects?

    They come from different physics. Extrusion faults are continuous-process ones: melt fracture (a rough surface when the die-exit speed is too high), die lines (scratches in the die transferred to the profile) and profile warpage from uneven cooling.

    Injection faults are cyclic-process ones: sink over thick sections, weld lines where flow fronts meet, flash at the parting line, short shots. Fix extrusion by tuning speed, die finish and cooling balance; fix injection by tuning pressure, packing and the mold.

    Key takeaways

    1. Geometry picks the process: a constant cross-section extrudes; a closed 3D part with features that leave the section injection molds.
    2. Their design rules conflict — draft, undercuts, wall uniformity and tolerance — so design to the process you chose, not to both at once.
    3. Combine them only where it pays: co-extrusion for multi-layer profiles, extrusion-plus-overmolding for a profile with molded ends, thermoformed extruded sheet for large shallow parts.

    Injection and extrusion share a barrel, a screw and a melt, but they part ways at the tool: a closed cavity makes discrete 3D parts, an open die makes continuous profiles. Let the geometry choose, design to that process’s rules, and for a hybrid only when one part genuinely needs both — then the combination saves the assembly, not just the material.

    Sources
    Plastics Technology — extrusion and injection processing comparison
    Society of Plastics Engineers — polymer processing fundamentals
    MoldMaking Technology — injection mold design rules
    Plastics Industry Association — plastics processing practice
    MatWeb — thermoplastic resin and melt-flow data

    Sources: Plastics Technology · Plastics Technology Online · Society of Plastics Engineers (SPE) · ThomasNet