Plastic gears run the drivetrains of printers, kitchen appliances, power window mechanisms, and medical pumps. They are also the component that fails first in cheap products, which leaves engineers with a split intuition: plastic gearing is either obviously fine or obviously a corner cut, depending on which product they last took apart. Both intuitions are right. A molded acetal train correctly sized for its torque outlives the product; the same gear 30 percent over its rating, or run 20 degrees C hotter than the designer assumed, wears its teeth into rounded stubs within months. The difference is not the material but whether anyone computed the load.

The Materials and What Each Is For

Acetal (POM) is the default and the right starting point for most designs: dimensionally stable, almost no moisture uptake, low friction against itself and against steel, clean to mold and machine. Its weaknesses are notch sensitivity and a service ceiling around 90 degrees C continuous.

Nylon (PA6, PA66) is tougher on shock loads and quieter. Its problem is water: nylon absorbs 2 to 3 percent moisture at equilibrium and grows 0.5 to 0.8 percent doing it, so a gear cut to nominal and installed at a tight center distance binds after two weeks in a humid room. Design backlash for the wet condition.

Glass-filled grades roughly double stiffness and strength and raise the temperature ceiling. They also make the gear abrasive, so a filled gear meshing an unfilled one grinds its mate away — use filled material on both or neither. PEEK and PPS handle 150 to 250 degrees C at $80 to $200 per pound.

Internally lubricated grades — PTFE-, silicone-, or molybdenum-filled — cut friction and heat at some cost in strength, and are usually the right specification for an unlubricated pair that must last. They sit alongside the comparison in nylon versus acetal for moving parts.

The PV Limit Is the Real Constraint

Plastic gears rarely fail by tooth breakage. They fail by wear, governed by PV: contact pressure times sliding velocity, in psi-fpm. Every gear plastic has a published PV limit, generally 2,500 to 6,000 psi-fpm dry and several times that lubricated.

PV is a heat statement in disguise. Plastic conducts heat about two hundred times worse than steel, so frictional energy at the tooth flank stays in the tooth. Local flank temperature runs well above bulk gear temperature, and once the surface passes the softening range the teeth deform, mesh geometry degrades, and failure accelerates. This is why plastic gears survive a long bench test and fail in the field: the bench had airflow and the product does not. Practical consequences:

  • Derate the published PV for actual ambient temperature. A gear rated at 23 degrees C may have half the capacity inside a 60 degree C enclosure.
  • Intermittent duty gives real headroom: a mechanism running 5 seconds per minute dissipates heat between cycles.
  • Pitch line velocity above roughly 1,500 feet per minute is difficult for unlubricated plastic regardless of load.

Lubrication

Plastic gears will run dry, which is their appeal where oil is unacceptable — at a cost of roughly half the load capacity. A single application of a compatible PAO or silicone grease at assembly raises capacity, cuts noise, and lasts the life of a sealed mechanism. Compatibility matters more than performance: some greases stress-crack acetal or plasticize nylon, so verify against the resin datasheet before production.

Molded or Cut

Injection-molded gears are the volume answer. Tooling runs $8,000 to $35,000 for a single-cavity mold, per-part cost drops to cents, and molded teeth carry a smooth skin that machined teeth cannot match. Acetal shrinks 1.8 to 2.2 percent and not uniformly, so the first article rarely hits the tooth profile and the tool gets corrected. Gate location drives the result: a single side gate produces asymmetric shrink and an out-of-round gear, so a diaphragm or multi-point hub gate is standard. Wall discipline applies to the web and hub as to any molded part, per wall thickness for injection molded parts and injection molding defects.

Hobbed gears cut from extruded stock cost $30 to $250 each with no tooling, hold tighter tolerances, and are the only option for prototypes, with slightly different wear because the machined surface has no molded skin. Volume decides: below about 2,500 units a year machine them, above 10,000 mold them, and in between run the numbers with tooling amortization included, as injection molding costs lays out.

Sizing Rules That Keep Plastic Gears Alive

  1. Give the small gear more tooth. The pinion sees more cycles per revolution of the pair and wears first. Use at least 18 teeth where geometry allows.
  2. Widen the face. Face width of 6 to 10 times the module spreads load and drops contact stress, and plastic's compliance distributes it well despite misalignment.
  3. Design in extra backlash — two to three times a steel pair's — to absorb thermal expansion, moisture growth, and molding variation.
  4. Mesh plastic against steel where you can. A plastic pinion against a steel gear conducts heat out through the steel and lasts far longer than plastic against plastic.
  5. Apply a real factor of safety: two to three on torque against the published rating, more if the mechanism can jam, reasoned through as in factor of safety in mechanical design.
  6. Support the bore properly. Most plastic gear failures traced back far enough are alignment failures — excess bore clearance tips the gear and concentrates load on one end of the tooth.

The Noise Advantage Is Genuine

Replacing a steel gear pair with plastic typically drops noise 5 to 10 dBA, and plastic's damping removes the high-frequency whine people find most objectionable. For an appliance or an office machine this is often the reason plastic is chosen, independent of cost. A belt drive achieves something similar by another route, compared in belt drive or gear drive.

Test to Failure, Not to Spec

A plastic gear train needs life testing at the top of its temperature range with the real lubricant and duty cycle, run past target life until something breaks. What breaks and when is the only trustworthy input to derating. Wear is progressive rather than sudden, unlike the fractures in material fatigue in product design, so a partly worn train gives warning.

Sizing a Gear Train That Lasts

Projects House designs plastic and hybrid gear trains against real torque, duty cycle, and ambient temperature: material selection, tooth sizing, molding strategy, and a life test plan that finds the limit before customers do. Send your requirements through the contact form.