Why do my 3D printed gears bind or jam?

Almost always because printed parts come out slightly oversized — extrusion width and elephant foot both add material to the tooth flank, which removes the backlash the pair needs. Fix it by opening the centre distance by 0.2 to 0.4 mm rather than reprinting, or by printing the pinion at 99 percent scale.

Which orientation is strongest for a printed gear?

Print flat, with the gear face on the bed. Layer lines then run around the tooth rather than across its root, so a tooth under load is not being peeled apart along a layer boundary — the standard failure mode for gears printed on edge.

Should I export STL or 3MF for printing a gear?

Either works in any slicer. 3MF carries units and metadata so it cannot be misread as inches, which makes it the safer default. Use STL when a tool or print service only accepts STL. Export STEP instead if the gear is going to be machined rather than printed, since STEP keeps the exact involute curve.

Which filament is best for 3D printed gears?

PETG for most loaded gears — it is tougher than PLA and does not creep as badly under sustained torque. PLA is fine for light or intermittent loads and prints the crispest teeth. Nylon is the durable choice for continuous running, at the cost of a harder print and moisture sensitivity.

3D Printed Gears — Generate a Printable Gear STL

Generate a printable gear and download the STL — describe what you need in plain English, set the module, tooth count, face width, and bore, and get real involute geometry rather than a rounded approximation. Then use the clearance and orientation rules below so the pair actually meshes off the bed.

Print one now: Generate a printable 20-tooth spur gear — opens the studio with that description already filled in. Change any number before you send it, then export STL or 3MF straight into your slicer.

Printable gear presets

Every link below opens the studio pre-filled with a description that prints reliably on a standard 0.4 mm nozzle. Edit the numbers before sending — tooth count and bore are the two you will most often want to change.

Gear	Why this one	Generate
Spur, module 2, 20T	The safe default. Crisp teeth, no overhangs, prints on anything.	Generate
Spur, module 1.5, 30T	Finer teeth, smoother running. About the practical floor for FDM.	Generate
Herringbone, module 2, 24T	Self-centring and quiet, with no axial thrust to react. A printing favourite.	Generate
Internal ring, module 2, 60T	The outer ring of a printed planetary set.	Generate
Rack, module 2	Linear motion. Pair it with any module 2 pinion.	Generate
Bevel, module 2, 20T	Turns drive through 90°. Needs support under the cone.	Generate

Pick a module your printer can actually resolve

Module is tooth size in millimetres. It is the single parameter that decides whether a printed gear works, because it sets how much material each tooth flank gets.

Module	On a 0.4 mm nozzle
< 1.0	Avoid. The flank is thinner than a couple of extrusion widths, so the involute profile rounds off into a bump and the pair binds or skips.
1.0 – 1.5	Workable if you slow down and tune flow. Expect to tune clearance by trial.
1.5 – 2.0	The reliable range. Teeth resolve cleanly, meshing is predictable.
> 2.0	Very robust, good for high torque. The gear simply gets physically larger.

Need a smaller gear? Drop the tooth count, not the module. Pitch diameter is module × teeth, so a module 2 gear with 14 teeth is only 28 mm across and still prints cleanly. Going to module 0.8 to hit the same diameter gives you teeth your nozzle cannot draw. Do not go below about 12 teeth on a 20° pressure angle or the roots undercut.

Clearance: why printed gears bind, and the fix

This is the single most common failure. The geometry is correct and the print is still tight, because a 3D printer adds material the model does not have: extrusion width lands slightly proud of the wall, and elephant foot spreads the first layers. Both land on the tooth flank, which is exactly where the backlash was supposed to be.

Three fixes, cheapest first:

Open the centre distance by 0.2–0.4 mm. No reprint needed — move the shaft holes in whatever the gears mount to. This is almost always the right answer.
Print the pinion at 99% scale. One slicer setting, one reprint of the smaller part.
Enable elephant-foot compensation (0.1–0.2 mm) so the bottom two layers stop flaring into the mating flank.

Nominal centre distance for a pair is module × (teeth₁ + teeth₂) ÷ 2. A module 2 pair of 20 and 40 teeth sits 60 mm apart in CAD — so design the mount for about 60.3 mm and it will run. More in 3D printed part tolerances.

Orientation, walls, and material

Print flat

Face on the bed. Layer lines then run around the tooth instead of across its root, so a loaded tooth is not being peeled along a layer boundary — the classic way printed gears shear off.

Walls beat infill

A tooth is nearly all perimeter. Three or four walls does far more for tooth strength than raising infill, and prints faster.

PETG for load

Tougher than PLA and less prone to creep under sustained torque. PLA is fine for light or intermittent duty and gives the crispest teeth. Nylon for continuous running.

Mind the overhangs

Spur and herringbone print unsupported. Helical teeth overhang along the helix; bevel needs support under the cone. Spur is the safe printed default.

Printing a gearbox: generate each gear in turn

PartWork.ai builds one part per request, so a gearbox is a short sequence rather than one assembly — which is usually what you want anyway, since each gear gets printed separately. A 3:1 reduction on 5 mm shafts, keeping the module identical so the teeth mesh:

	Pinion (driver)	Gear (driven)
Module	2	2 (must match)
Teeth	16	48
Pitch diameter	32 mm	96 mm
Bore	5 mm	5 mm

Ratio 48 ÷ 16 = 3:1. Centre distance = 2 × (16 + 48) ÷ 2 = 64 mm nominal, so build the plate at about 64.3 mm to leave printed backlash.

Generate the 16-tooth pinion
Generate the 48-tooth gear

For a planetary set, generate the sun, the planets, and the internal ring the same way — the ring tooth count must equal sun + 2 × planet teeth for the set to close.

Common questions

What module should I use for a 3D printed gear?

1.5 to 2 on a standard 0.4 mm nozzle. Below module 1 the nozzle cannot resolve the tooth flank and the profile degrades into a rounded bump. Need a smaller gear? Cut the tooth count instead.

Why do my printed gears bind?

Printed parts come out slightly oversized, which eats the backlash. Open the centre distance by 0.2–0.4 mm, or print the pinion at 99% scale.

STL or 3MF?

Either slices fine. 3MF carries units so it cannot be misread as inches, which makes it the safer default; STL when a service only takes STL. Export STEP instead if the gear is going to be machined — see STL vs STEP and STL vs 3MF.

Which filament?

PETG for most loaded gears, PLA for light duty and the crispest teeth, nylon for continuous running.

Can I get a gear that matches one I already have?

Yes, if you can measure it. Count the teeth and measure the outside diameter, then module ≈ OD ÷ (teeth + 2). Say those numbers in the studio — see measuring a part for replacement.

Generate your printable gear

Describe the gear — type, module, teeth, face width, bore — and PartWork.ai builds real involute geometry you can inspect, adjust in follow-up messages, and export as STL or 3MF. If you are unsure of a number, say what the gear is for and start from there.

Start from a ready-made description and edit the numbers: spur · herringbone · helical · bevel · rack · internal ring — or open a blank studio.