Almost every printer ships with a 0.4 mm nozzle, almost every published profile assumes one, and almost every functional part would be better off on a 0.6. The 0.4 is not a considered choice; it is an inherited one, and it persists because changing a nozzle feels like a bigger commitment than it is.
Here is the case for reconsidering, with the arithmetic attached.
The same bracket on three nozzles
The test part is a bracket 80 mm tall enclosing 95 cm³, given three perimeters and 20% infill at 80 mm/s. Each nozzle then runs at the layer height it can bond reliably, which is roughly three-quarters of its diameter.
| Nozzle | Layer | Layers | Time | Extruded volume |
|---|---|---|---|---|
| 0.4 mm | 0.2 mm | 400 | 5.827 h | 113,974 mm³ |
| 0.6 mm | 0.3 mm | 267 | 2.883 h | 126,956 mm³ |
| 0.8 mm | 0.4 mm | 200 | 1.779 h | 137,483 mm³ |
Notice that the extruded volume rises as the time falls. The bigger nozzle lays down more plastic, not less, because wider walls and coarser layers both add material — you are buying hours with grams, and that is the trade the print time estimator makes visible where a single slicer estimate does not.
Stepping from 0.4 to 0.6 took roughly half the time off a job that was not remotely near any hardware limit. The step to 0.8 halved it again. If you print functional parts and your machine sits busy, this is the single largest time saving available to you that costs nothing but a nozzle.
The catch in the right-hand column
Look at the extruded volume. The same part uses progressively more plastic as the nozzle grows, and this surprises people who expect a bigger nozzle to be more efficient.
The reason is the walls. Wall count is specified as a number of perimeters, not as a thickness, so three walls at a 0.45 mm line width is a 1.35 mm shell while three walls at a 0.9 mm line width is a 2.7 mm shell. You did not ask for a thicker part; you asked for three walls and the slicer gave you three fatter ones.
That extra shell is not wasted — it is genuinely stronger — but it is filament you are paying for. If you move to a larger nozzle and want the same shell, drop the wall count as you go: three walls at 0.4 is roughly two at 0.6. The corollary is that a coarse nozzle with the same wall count as a fine one is a quiet, permanent increase in your material bill.
The ceiling that decides whether your speed setting means anything
A nozzle does not limit speed. The hotend's ability to melt plastic does, and that limit is a volume per second rather than a distance per second.
Take the 0.8 mm run above and ask for 150 mm/s instead of 80. The requested flow is 54 mm³/s; the hotend's ceiling in this model is 37.5 mm³/s, so the estimator reports the job flow limited and the head averages 104.2 mm/s regardless of what the profile says. The print takes 1.379 h instead of the 1.779 it took at the lower speed — a real saving, but nothing like the one the speed number implies.
This is why large-nozzle printing rewards a high-flow hotend more than a faster motion system, and why people who fit a 0.8 to a stock hotend and turn the speed up are disappointed. Print speed and what actually limits it works through the rest of that relationship.
What a bigger nozzle costs you
- Minimum feature size. A 0.8 mm nozzle cannot draw a 1 mm wall as two lines, so thin features either vanish or come out oversized. If your model has detail near the line width, it will be reinterpreted.
- Text and engraving. Anything relying on a fine groove disappears first.
- Small holes. The wider bead bulges further into a tight radius, so hole error grows with nozzle size.
- Overhangs and bridges. A heavier bead sags more before it sets.
- Visible layer lines. At 0.4 mm layers they are unmistakable and no amount of light sanding hides them.
None of that matters on a bracket. All of it matters on a display piece, which is why the honest answer is to own more than one nozzle rather than to pick a winner.
The abrasive question, which is not really about size
Carbon-filled, glass-filled, wood-filled, glow-in-the-dark and metal-filled filaments all wear a brass nozzle out, some of them within tens of hours. Two consequences, and only one of them is about diameter.
The first is hardness: those materials need a hardened steel, ruby or tungsten-carbide nozzle, at any diameter. The second is clearance: wood flour and metal powder are coarse, so 0.6 mm is a practical minimum and 0.4 clogs. Fibre-filled grades run fine through a hardened 0.4 mm, but they flow better wider.
If you are buying a nozzle in order to print something filled, buy the hardened one and buy it at 0.6.
Which nozzle for which work
- 0.2 and 0.25 mm — miniatures, jewellery masters, fine text. Slow enough that the time cost is the design constraint.
- 0.4 mm — keep it as the baseline every published profile assumes, and use it when a part mixes detail with structure.
- 0.6 mm — the default for functional parts, filled filaments and anything printed more than once.
- 0.8 mm — large parts, vase mode, moulds and formers, and anything where layer lines are part of the aesthetic.
- 1.0 mm — only with a hotend built for the flow, since the nozzle will not be the limit.
Changing a nozzle is a calibration event
Swap the brass part and the profile that worked yesterday is wrong today. Line width, flow, pressure advance and the first-layer squish all move together, and a nozzle change that skips them produces exactly the symptoms people then blame on the nozzle: inconsistent extrusion width, gaps between walls, and a first layer that will not settle.
Budget twenty minutes for a line-width test and a fresh first-layer check every time you change diameter, and keep a saved profile per nozzle rather than editing one in place. That habit is worth more than the nozzle.