Printed assemblies fail at the joint, not at the part. Two components can each measure correctly and still not assemble, because the clearance needed is the sum of the error on both surfaces plus whatever the process does to the features that meet.
How printed fits differ from machined fits
A machined bore and shaft are produced by removing material from a solid, and their error is small and predictable. A printed pair is built layer by layer in a machine that softens or cures material, so both members carry error from the same process and often in the same direction. The result is that a nominal clearance can shrink or grow depending on whether the features are built in the plane of the layer or across it.
| Fit type | Behaviour of printed parts | Design response |
|---|---|---|
| Snug fit that holds by friction | Depends heavily on process and orientation | Test with a ladder; expect to tune per material |
| Sliding fit | Needs clearance for roughness, not just size | Increase clearance until motion is free without play |
| Snap fit | Sensitive to layer direction and material toughness | Orient the flexing feature in the build plane |
| Press fit | Risky: printed surfaces deform locally | Use a boss and fastener, or plan for reaming |
| Threaded joint | Printed threads work only at coarse pitches | Use an insert or a machine-and-tap operation |
How much tolerance should a snug fit use?
Enough to assemble without force, then tuned by test.
A snug fit has to hold by friction, which makes it the most process-sensitive of the common fits: too little clearance and the parts will not go together, too much and the joint rattles. Because both members are printed, the practical approach is to design a small ladder of clearances into the first build and choose the pair that behaves correctly, then use that value for the rest of the assembly.
Two design habits make a snug fit more likely to work. Keep the mating surfaces cylindrical or planar rather than introducing complex curves, because simple geometry repeats more reliably, and keep the insertion length short, since friction accumulates with engagement length. Where a joint must hold firmly and repeatedly, a screw or a clip is more reliable than friction between printed surfaces.
What clearance works for a printed sliding fit
A sliding joint needs more than a snug one, because roughness contributes as much as size. A powder-bed surface is textured, so a nominal clearance that would slide on a machined part can bind on a printed one. Where the motion is important, the design should either increase clearance until movement is free without visible play or include a machined or reamed surface on the sliding element.
Orientation changes this more than most designers expect. A cylindrical bore built vertically may print slightly oval, while the same bore built horizontally can sag on its underside. Where a sliding joint is critical, the fastest way to establish a working clearance is to print the joint in two orientations on the same coupon and compare them.
Snap fits, and why orientation decides whether they work
A snap fit relies on a feature flexing and then recovering, which makes it sensitive to how the layers run through the flexing element. A cantilever built flat in the plane of the layers bends along the material, while the same feature built vertically is a stack of layers that delaminate when loaded. The design implication is to orient the flexing feature in the build plane and to give the latch a generous root radius so stress is not concentrated at a sharp transition.
Material choice matters as much as geometry. Resin is relatively brittle, so a snap feature in standard resin may break on first use, while a powder-bed nylon or a tougher engineering resin tolerates repeated flexing. Where a snap fit will be cycled many times, the honest answer may be to prototype the feature in the production material rather than in the quickest available one. Process terminology follows ASTM Committee F42, and application guidance is published by ASME.
Does build orientation change the fit?
Yes, and it is the variable most often left out.
Yes, and it is the variable most often left out of a clearance decision. Features measured in the plane of the build behave differently from those built across it, so the same nominal clearance can produce a joint that works in one orientation and binds in another. The coupon approach resolves this by printing the same joint both ways and measuring the result rather than assuming it.
There is a practical limit to how much can be fixed through orientation alone. Once a joint is at the edge of the process capability, the reliable options are to add clearance and accept some play, to use a fastener instead of friction, or to machine the critical surface after printing. Choosing between those is a design decision about what the joint has to do, not a tolerance value. Measurement practice is described by the NIST Manufacturing Extension Partnership, material references are published by ASM International, qualification guidance by UL Solutions, and process options on the 3D printing and materials pages.
Send the assembly with the joints that must work, and request a quote that includes a clearance test on the mating features.
FAQ
How much tolerance should a 3D printed snug fit use?
Enough clearance that the parts assemble without force, then tuned from a printed ladder rather than assumed. Because both members carry process error, the working value is usually specific to the material and the orientation used.
What clearance works for a printed sliding fit?
More than a snug fit, since surface texture contributes as much as size, and a powder-bed surface binds where a machined one would slide. Increase clearance until the motion is free without visible play.
Do printed snap fits need a different orientation?
Yes. A flexing cantilever should be built in the plane of the layers, because a vertically built feature is a stack of layers that can delaminate when it bends. A generous root radius also spreads the stress.