Most wall thickness advice stops at "use at least your nozzle diameter." That's the physical floor, not a useful target. What actually survives depends on the filament: a 0.8 mm wall that's perfectly solid in PLA can warp off the bed in ABS, tear in TPU, or delaminate in nylon. Here's how each common material behaves, and the sizing rule that matters more than any single number.
There are two separate limits at play, and conflating them is where most thin-wall problems start.
The extrusion minimum is geometric. Your nozzle produces a bead of plastic roughly its own diameter wide — a 0.4 mm nozzle cannot lay a line narrower than about 0.4 mm. Any wall below that can't be represented in G-code at all, so the slicer drops it or fakes it. This limit is the same regardless of what you load.
The material minimum is mechanical. It's the thinnest wall that still holds its shape while cooling, bonds to the layer below, and survives being handled. This is entirely material-dependent, and it is almost always the higher of the two numbers. A wall can be perfectly printable and still be useless.
Rule of thumb: the extrusion minimum tells you what the printer can do. The material minimum tells you what you should do. Design to the second one.
The figures below assume a standard 0.4 mm nozzle. They're practical working minimums drawn from how these materials behave in typical desktop FDM conditions, not absolute physical limits.
| Material | Practical minimum | Comfortable target | What goes wrong when you push it |
|---|---|---|---|
| PLA | 0.8 mm | 1.2 mm | The most forgiving common filament. Stiff, low shrinkage, bonds well. Thin walls print cleanly but become brittle — they snap rather than flex. |
| PETG | 1.0 mm | 1.6 mm | More ductile than PLA but stringier and slower to solidify. Thin walls tend to look rough and can droop on tall unsupported sections. |
| ABS / ASA | 1.2 mm | 2.0 mm | High shrinkage on cooling. Thin walls are where warping and layer splitting start — the wall contracts faster than it can bond. An enclosure helps, but thickness helps more. |
| TPU (flexible) | 1.2 mm | 2.0 mm | A thin TPU wall doesn't break — it tears. Flexible parts concentrate stress at thin sections, so anything that bends in use needs real material behind it. |
| Nylon (PA) | 1.2 mm | 2.0 mm | Hygroscopic and prone to warping. Thin walls amplify moisture defects and delaminate under load, which defeats the point of choosing nylon. |
| Polycarbonate | 1.2 mm | 2.0 mm | Needs high temperatures and strong interlayer bonding. Thin walls cool too fast to bond properly, producing parts far weaker than PC should be. |
| Filled composites (carbon-fibre, glass-filled) |
1.2 mm | 2.0 mm | Stiff but notch-sensitive. The added fibres reduce ductility, so thin sections fracture rather than deform. |
Here's the part that trips people up. Slicers build walls out of whole extrusion lines. If your line width is 0.4 mm and you design a 1.0 mm wall, the slicer can fit two full lines (0.8 mm) and is left with 0.2 mm of nowhere-to-go. Depending on the slicer and its settings, that leftover becomes a gap, a thin under-extruded sliver, or a slightly over-extruded bulge.
Walls sized to an exact multiple of your line width avoid this entirely:
This is why 1.2 mm shows up so often as a recommendation for 0.4 mm nozzles. It isn't a magic strength threshold — it's three clean perimeters with nothing left over. Modern slicers have variable-width extrusion features that handle awkward in-between thicknesses far better than older ones did, but designing to a clean multiple still produces the most predictable result.
If you change nozzle size, these numbers all move. On a 0.6 mm nozzle the clean multiples become 1.2 mm, 1.8 mm and 2.4 mm — so a model designed around 0.4 mm walls no longer lines up.
Height matters as much as thickness. A 1 mm wall that's 5 mm tall is fine; the same wall at 80 mm tall will wobble as the nozzle passes, producing ringing artefacts and possibly detaching from the bed. Once a wall's height exceeds roughly 20 times its thickness, treat it as a structural risk and either thicken it or add a rib.
For anything bearing weight, taking impact, or being screwed into, the material minimums above are far too low. Start at 2.5–3 mm and remember that FDM parts are anisotropic — much weaker across layer lines than along them. Orientation often matters more than thickness for real strength.
Detail work is the one case where going below these numbers is reasonable, because the parts aren't structural. A 0.2 mm nozzle brings the extrusion minimum down to roughly 0.2 mm and makes 0.4–0.6 mm walls viable for decorative features. Accept that they'll be fragile.
This is the most common way people end up with unprintable walls without realising it. Scaling a model to 60% scales every wall to 60%. A comfortable 1.6 mm wall becomes 0.96 mm; a 1.2 mm wall becomes 0.72 mm and is now below the practical minimum for most materials. Always re-check thickness after scaling down.
Reading thickness off a model by eye is unreliable, especially on organic shapes or imported CAD where thin regions hide inside the geometry. The dependable approach is a thickness analysis pass that measures through the solid at every point on the surface and flags everything below your chosen threshold.
LuminaMesh's wall thickness analyzer does this in the browser: set the minimum for your material from the table above, and problem regions are highlighted directly on the 3D model. If the model is thin in places you can't easily redesign, the same tool can thicken those regions in place.
Upload an STL, set your minimum wall thickness, and see exactly which regions fall short — free, in your browser, no upload to a server.
Open the Wall Thickness Analyzer