3D Printed Shrek Toothpaste Extruder: What the Gimmick Teaches Engineers

• 1
  How a 3D Printed Shrek Toothpaste Extruder Actually Works
• 2
  Failure Modes You Can See on a Desk Toy
• 3
  From Novelty Cap to Production Part: What Changes
• 4
  Choosing a Process for a Squeeze-Fit Cap

The part is a cap with an internal bore that slips over the threaded neck of a toothpaste tube, plus some kind of plunger or roller behind it. The user squeezes the tube, the paste moves forward, and the character head at the front decides where it lands. That is the whole mechanism. There is no valve, no pump, no seal that has to hold pressure for weeks.

Because the load is so low, almost any desktop process can make one. Fused deposition modeling with PLA at a 0.4 mm nozzle and 0.2 mm layers is the usual choice. A typical print of this size takes 40 to 90 minutes and uses maybe 15 g of filament. No supports are needed if the bore is oriented vertically.

The interesting part is the fit. A toothpaste tube neck is roughly Ø28 mm to Ø32 mm with a coarse thread, and the cap has to grip it without cracking. That means the bore is a light interference fit, usually 0.2 mm to 0.4 mm undersize on the minor diameter so the print flexes over the thread instead of splitting along a layer line.

• 1
  Bore fit0.2–0.4 mm undersize so the cap flexes over the thread.
• 2
  Wall thickness2 mm minimum around the bore; thinner walls split at the layer seam.
• 3
  Layer directionPrint the bore vertically so hoop stress stays in-plane, not across layers.
• 4
  Nozzle0.4 mm is enough; go to 0.6 mm only if the head is large.

The usual failure is a split that runs along a layer line, not across it. FDM parts are weakest in the Z direction, often 30 to 50 percent weaker than the same material printed flat. A cap that grips a threaded neck loads the wall in hoop tension, so if the layers run horizontally the seam takes the whole load. Rotate the model 90 degrees and the same geometry lasts far longer.

The second failure is creep. PLA softens near 60 °C, and a bathroom shelf in summer can hit that. A cap left clamped on a tube will slowly lose its grip, then leak. PETG holds to about 80 °C and tolerates a bit more strain. Neither is a long-term answer for a part that stays under load.

The third failure is chemistry. Toothpaste is a mild abrasive paste with fluoride salts, surfactants and water. PLA and PETG are not certified food-contact materials, and printed surfaces have layer grooves that trap residue. If the part touches paste every day, that is a hygiene problem, not a strength problem.

None of this matters for a novelty item that sits on a shelf. It matters the moment the same squeeze-fit cap becomes a real product: a dosing cap, a pump housing, a medical applicator. Then the load, the temperature and the cleaning cycle all get harder, and the printed version stops being good enough.

• 1
  Layer splitHoop stress across Z layers is the number one crack path.
• 2
  CreepPLA near 60 °C, PETG near 80 °C; both lose clamp load over time.
• 3
  HygieneLayer grooves hold paste; no food-contact certification on typical FDM plastics.
• 4
  Thread wearRepeated on/off cycles round off printed threads within a few dozen uses.

A novelty cap is loaded once and forgotten. A production squeeze-fit cap is loaded every day, cleaned with hot water or alcohol, and expected to hold a seal for years. The geometry barely changes. The material, tolerance and surface finish change completely.

Take the bore. A printed bore might measure Ø27.6 mm with a 0.3 mm layer ridge and still work, because the plastic flexes. A machined POM cap at Ø28.00 mm ±0.05 mm gives a repeatable grip that does not depend on how the operator threads it on. That repeatability is what lets a manufacturer promise a seal.

Hoop stress is the other change. A printed wall carries load through layer bonds. A machined wall carries it through solid material in every direction. For a cap in 6061-T6 we can hold ±0.005 mm on the bore and a finish of Ra 0.8–1.6 μm, which is smooth enough for an O-ring to seat without tearing.

Cleaning is usually what pushes a project from plastic to stainless. Alcohol wipes, autoclave cycles and caustic wash-down attack most printed plastics. We machine 316L and 17-4PH for parts that see those conditions, and we can apply bead blasting or electropolishing afterward.

The decision is rarely about strength alone. It is about how many cycles the part has to survive, what it touches, and whether anyone has to prove it was made the same way twice. A toy fails that test. A machined cap passes it.

Cost only makes sense once you know the run size. One printed cap costs cents. One machined cap costs more, but the hundredth costs the same as the first because the setup is already done. Below about 50 pieces, printing usually wins. Above a few hundred, machining or molding wins on unit price and on consistency.

• 1
  Tolerance±0.005 mm on machined bores; printed bores vary by layer height.
• 2
  FinishRa 0.8–1.6 μm for sealing surfaces; printed ridges are far rougher.
• 3
  CleaningStainless grades survive alcohol and autoclave; most printed plastics do not.
• 4
  RepeatabilityMachined parts measure the same across a run; printed parts drift with the machine.