Last updated: 2026-09-05 · 5 min read

For molded gears, cams and sliding mechanisms, two resins dominate the short list: acetal (POM) and nylon (PA). Nylon brings toughness and abrasion behavior; POM brings something gears usually value more — a low, stable coefficient of friction, high stiffness, and near-total indifference to humidity, which means the tooth profile you molded is the tooth profile that keeps meshing months later. When the application is precision motion rather than impact survival, POM is usually the better gear material, and this article explains where that line sits and what molding POM demands from the tool.

Key facts

<>

  • POM absorbs very little moisture, so its dimensions and friction barely drift in humid service — the property that makes it the precision-gear default.
  • Nylon, conditioned by moisture, can run quieter and survive worse impacts, but its tooth size and clearances move with the seasons unless the design planned for it.
  • POM shrinks noticeably and crystallizes fast — the tool is cut with compensation, and gear profiles need shrink correction calculated per direction, not guessed.
  • Copolymer and homopolymer POM differ: homopolymer is a touch stronger and stiffer; copolymer processes with a wider window and handles hot water better — the grade decision is real.
  • Lubricated and filled POM variants extend the resin into quieter, longer-wearing gears — at the cost of brittleness or cost per kilo depending on the additive.

    • Flash — Thin excess material that escapes between the mold parting line or around ejector pins. Cause: insufficient clamp force, worn parting line, excessive injection pressure, or contaminated mold surface. Flash thickness: 0.02–0.1mm. Requires secondary trimming to remove. The related guide covers adjacent topics.

    Why does POM dominate gears and sliding parts?

    A gear is a fatigue part: every tooth loads and unloads millions of times, in sliding contact, often unlubricated, while holding a profile that was cut into steel at a fraction of a millimeter’s precision. POM’s property set reads like the requirements list for that job. Its friction against itself and against most partners is low and — critically — stable, because it does not soak up humidity the way nylon does. Its stiffness keeps tooth deflection down under load. Its fatigue resistance, especially in the copolymer grades, carries millions of cycles in well-designed teeth. And its low moisture uptake means the fit you set at assembly is the fit you keep, whether the gearbox ships to a desert or a dockside.

    None of this makes POM universally better. Nylon conditioned in service becomes tough and compliant, which quietens gear trains and survives abusive handling that would chip acetal teeth. But compliance is a moving target: as a nylon gear drinks and loses moisture with the seasons, its tooth thickness, its clearance and its noise signature all wander. Precision motion — metering mechanisms, printer drives, actuators, clockwork — punishes that wander, and it is precisely where POM wins the material meeting. The comparison that matters is not which resin is stronger on a datasheet; it is which resin changes least in the environment the gearbox will actually live in.

    How do POM and nylon compare on the properties gears care about?

    Gear property POM (acetal) Nylon (PA6/PA66)
    Friction and wear Low, stable friction; excellent for unlubricated meshes. Good wear behavior; friction rises and falls with moisture content.
    Dimensional stability Excellent — minimal moisture uptake, predictable shrinkage. Drifts with humidity; tooth thickness and clearance move in service.
    Stiffness High; tooth deflection stays modest. Comparable dry, softer when conditioned.
    Toughness Good but less forgiving; sharp teeth can chip under abuse. Excellent, especially conditioned; survives rough handling.
    Fatigue Strong millions-of-cycles record in copolymer grades. Also strong; moisture state changes the response.
    Noise Crisp and consistent; slightly harder mesh. Often quieter conditioned — the plasticized surface damps.
    Process window Hot, fast-crystallizing, needs disciplined tooling. Wider, forgiving; needs serious drying.

    The table’s quiet lesson is that the two resins fail differently, not just perform differently. POM’s failure mode is brittleness — an abused tooth chips — and precision loss from a badly compensated tool. Nylon’s failure mode is drift — a gearbox that measured perfect in January hums in July. Design reviews should name which failure the application cannot tolerate, because that choice picks the resin more reliably than any property midpoint.

    What does low moisture mean for tolerances?

    Everything about gear quality is dimensional: profile, tooth thickness, center distance, backlash. POM’s near-total indifference to humidity is what lets a designer hold those dimensions as fixed promises rather than ranges. The same stability shows up in molding itself: POM’s shrinkage, while larger than commodity resins, is predictable and — because the resin does not continue absorbing water after molding — the part you inspect at T1 is the part you ship for the life of the program. Nylon cannot make that promise; it equilibrates over weeks, and a rigorous nylon-gear program specifies the conditioning state at inspection and dimensions for the mid-range of service humidity. With POM, that entire layer of engineering disappears — one reason acetal gears dominate applications where the mesh is calculated to hundredths of a millimeter and the tolerance callouts are genuinely tight.

    How should you design molded gears?

  • Design in the molded state, not the cut state — involute profiles are moldable, but root fillets, tip relief and tooth thickness must be specified with the resin’s shrinkage compensation built into the tool geometry.
  • Radius the tooth roots — a sharp root notch is a fatigue initiation site in any material; a generous root fillet is the cheapest gear life you can buy.
  • Gate where the profile is honest — a center gate on a spur gear gives a symmetric, predictable shrink pattern; an edge gate on a precision gear invites out-of-round and profile distortion.
  • Control vestige — the gate mark sits on a working surface; submarine or pin-point gating, or a designed boss for the gate to live in, keeps the mesh from grinding on a gate vestige.
  • Vent the mesh — gear teeth trap air at the root; under-vented roots burn and leave weak seams exactly where the tooth bends.
  • Match the mating part — specify POM against POM, or POM against a softer partner deliberately; the friction pair is a system decision, not two independent part decisions.

    • What does POM demand from the molding process?

    1. Respect the melt window. POM processes hot and fast-crystallizing, and it degrades with an acrid signature when overheated or held — the datasheet window is a working rule, not a suggestion.
    2. Compensate shrinkage in the tool, per direction. Gear profiles shrink toward the center; the tool is cut oversize by calculated compensation, and the profile is verified on measured parts, not assumed.
    3. Vent aggressively at thin roots. Fast crystallization means fast fronts; gear roots trap air and burn unless the parting line carries deliberate vents.
    4. Hold the residence window. POM tolerates less dead time in the barrel than many resins; a stopped machine purges before resuming, or the first shots carry degradation.
    5. Inspect with the crystal structure in mind. POM reaches final dimensions as it crystallizes — parts measured hot read differently than parts measured cooled, so the inspection timing belongs on the process sheet.

    When would nylon or another resin beat POM?

    The substitution list is honest and short. Under repeated impact or abusive assembly, conditioned nylon outlasts acetal — its compliance absorbs what would chip a POM tooth. At sustained temperatures above POM’s comfort zone, the conversation moves to PA66 or to high-temperature resins. Where the gear train must be as quiet as possible in a stable indoor environment, a conditioned nylon mesh damps noise in a way stiff acetal does not. And where cost dominates a low-load, low-precision application, acetal’s processing discipline may buy precision nobody needs. The discipline that keeps these decisions honest is writing down the failure mode first — chip, drift, heat, noise or budget — because each failure points to a different resin, and the datasheet midpoints will happily support whichever bias arrived first.

    Parting line — The line where the two mold halves (A-side and B-side) meet. Parting line location affects: gate placement, ejector pin marks, witness lines, and cosmetic appearance. A well-designed parting line minimizes visible witness marks. The related guide covers adjacent topics.

    Gear programs are where shrinkage compensation stops being a number on a drawing and becomes the whole job. I have cut POM gears where the tooth profile was corrected, the out-of-round was corrected, and the root fillet was verified on molded parts under a profile projector — and that level of care is exactly why the gearbox ran silent for years. The acetal gears I have seen fail all failed the same way: someone treated the tool like a metal gear blank, cut no compensation, and discovered at T1 that the mesh had no backlash. The resin was never the problem. The tooling math was.

    — Arthur, Tooling Engineer at aCatMould

    Frequently asked questions

    Is POM stronger than nylon?

    They trade. POM is typically stiffer and holds its properties more steadily across humidity; nylon is tougher, especially conditioned, and survives impact that would chip acetal. For gears, the steadiness usually matters more than the peak — which is why POM dominates precision motion.

    Why do POM gears hold their size while nylon gears drift?

    Moisture. POM absorbs very little water, so its dimensions and friction barely change in service. Nylon equilibrates with ambient humidity over weeks, thickening and softening as it does — tooth thickness, backlash and noise all shift unless the design dimensioned for the mid-range.

    Do POM gears need lubrication?

    Often not. Acetal runs well dry against itself and many partners, which is a design feature in sealed gearboxes and food-adjacent mechanisms where grease is unwelcome. Lubricated POM grades exist for harder service; the pairing decision should treat the mesh as a system.

    Why is shrinkage such a big deal on molded gears?

    A gear is a precision profile, and POM shrinks several percent as it crystallizes. The tool is cut with directional compensation so the molded teeth land on the intended involute — skip the math and the mesh measures out-of-round or binds at the roots. It is tooling geometry, not process tuning.

    Copolymer or homopolymer POM for gears?

    Both mold gears well. Homopolymer carries a slight edge in stiffness and strength; copolymer offers a wider processing window and better hot-water behavior, and it is the more common gear stock. The grade datasheet and the service environment decide; the family name alone should not.

    What should a POM gear T1 report include?

    Profile and out-of-round measurements on actual molded parts, the shrinkage compensation used in the tool, vent condition at the roots, gate vestige against the agreed limit, and parts measured at a defined cooled state — so the mesh conversation starts from evidence instead of opinions.

    Key takeaways

    1. POM wins gear programs on stability: low, moisture-indifferent friction and dimensions that stay where they were molded.
    2. Nylon fights back on toughness and conditioned noise damping — and loses ground every time humidity moves a tooth.
    3. POM tooling is precision tooling: directional shrinkage compensation, aggressive venting at roots, and honest profile verification at T1.
    4. The resin decision should start from the failure mode — chip, drift, heat, noise or cost — not from a property midpoint.
    5. Grade choice inside POM (copolymer vs homopolymer, lubricated vs neat) is a real decision with real consequences for wear and processing.

    If your program involves gears, cams or sliding mechanisms where the mesh has to stay true, send us the gear data, the mating-part description and the environment before the deposit. You will get the resin and grade recommendation in writing, the shrinkage compensation and gate plan for the tool, and the T1 measurement list — from a named tooling engineer, before any steel is ordered.

    And if the honest answer is that nylon fits your application better — because the part must survive impact, or the mesh must damp — you will get that answer too, with the conditioning plan that makes nylon work: the inspection state, the mid-range dimensions and the drying discipline. A resin recommendation that survives contact with the service environment is worth more than one that merely survives the quotation meeting, and that is the standard we hold our own material decisions to.

    Sources
    MatWeb — material property database for POM homopolymer and copolymer grades
    ISO 20457:2018 — Plastics moulded parts: Tolerances and acceptance conditions, International Organization for Standardization
    ASTM D638 — Standard Test Method for Tensile Properties of Plastics, ASTM International
    Autodesk Moldflow — injection molding simulation for shrinkage and filling balance, Autodesk

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