3D printed fashion style: how geometry, fit and finish get built

Most fashion pieces start as a rigid or semi-rigid shell: a corset panel, a shoulder plate, a lattice cuff. For those, powder bed fusion on nylon gives the best strength-to-weight ratio and needs no support removal on internal channels. When the surface has to look like metal, laser powder bed fusion in stainless or titanium is the usual route, though part cost climbs fast above a 200 mm footprint.

Resin processes matter when the visible surface is the product. Stereolithography and digital light processing hold detail down to a few tenths of a millimeter, so a lace-like pattern or a fine rib reads cleanly without post-polishing. The trade-off is brittleness: thin resin ribs crack under repeated bending, which is why resin suits display and runway pieces more than daily wear.

Material extrusion is the cheap entry point and the one most studios already own. Fused deposition modeling handles large flat panels and quick fit checks, but layer lines are visible and the anisotropic strength means a panel printed in one orientation fails along the layer boundaries. Treat it as a prototyping tool, not a production one.

Metal binder jetting and investment casting sit between the two. Binder jetting builds a green part that is sintered afterward, which shrinks it by a known percentage and leaves a slightly porous surface. For clasps, buckles and hinge pins, that is often enough. For a load-bearing structural arm, machined metal is still the safer answer.

Standard sizing is a compromise. A scan gives you the actual surface, and printing lets you build the garment to that surface instead of cutting fabric to approximate it. A handheld or photogrammetry scan at 1 mm resolution is enough for a shell that sits over clothing; below 0.5 mm you are paying for detail the printer cannot hold anyway.

The scan is not the CAD model. Point clouds arrive noisy, with hair, folds and motion artifacts that have to be cleaned before anything is offset. We typically work from a decimated mesh, then offset 2–4 mm outward for clearance over skin and 6–10 mm over a base layer. The clearance number is what decides whether a piece feels tight or forgettable.

Where the body moves, the geometry has to move with it. Hips, shoulders and the rib cage expand and rotate, so a rigid single-piece shell will bind. Splitting the shell into overlapping segments joined by printed living hinges or captured pins keeps the silhouette while allowing travel. That segmentation is the part most first-time designs miss.

Iteration is cheap once the scan exists. A 10 mm change to a waistband or a 5 degree change to a shoulder slope is a CAD edit and a reprint, not a new pattern and a new toile. Two or three physical rounds usually settle the fit before tooling decisions are made.

An unfinished printed surface looks like a prototype. The layer texture, the support scars and the matte grey of raw nylon all say the same thing. Finishing is what converts it into a wearable object, and it also changes how the part behaves in contact with skin.

For nylon shells, bead blasting evens out the layer lines and gives a soft, uniform matte. Dyeing after blasting takes color into the surface rather than laying it on top, so abrasion does not expose white plastic. A clear or colored anodize is the equivalent move on aluminum components, and hardcoat anodize adds a wear layer where the part rubs against itself.

Metal pieces usually get tumbled or brushed first, then a plating or a PVD-style color layer. Electroless nickel gives a bright, even finish on complex geometry that electroplating struggles to reach. Gold and silver plating are available for small hardware, though they wear through on high-contact edges, so keep plated parts away from where fingers land repeatedly.

Sealing matters for anything worn against skin. A printed lattice traps dust and skin oils in its internal volume, and cleaning it is difficult. A thin lacquer or a UV-cured clear coat closes the pores, makes the surface wipeable, and stops dye from transferring to fabric underneath.

Flat panels with no internal geometry belong on a CNC or a waterjet. A 3 mm aluminum plate cut and bent costs less than the same plate printed, holds tighter tolerances, and takes a finish that will not delaminate. Printing a flat part is paying additive prices for a subtractive job.

High-cycle flexible joints are another weak spot. A printed living hinge survives hundreds of cycles, not tens of thousands. Where a hinge opens and closes daily for years, a machined pin in a printed housing or an entirely machined assembly lasts longer and is easier to service.

Large one-piece shells are usually a mistake too. The build envelope sets the ceiling, and a 600 mm tall single piece needs support everywhere and often warps in the lower layers. Splitting it into three or four bolted or pinned sections costs less, prints flatter, and lets you replace the part that actually broke.

Textiles still win on drape. Nothing printed folds and flows like woven cloth, and a hybrid piece, printed structure over a fabric base, usually looks better and wears cooler than an all-printed garment.