Analysis of 7 Common 3D Printing Technologies

VAT photopolymerization covers SLA, DLP, and LCD printing. A build platform lowers into a vat of liquid photopolymer and a light source cures one layer at a time. Layer heights run 0.025–0.1 mm, so surfaces come out smooth and fine features hold well. The catch is the resin. Most formulations are brittle and degrade under UV, so parts are best used for form-and-fit checks, mold masters, and visual models rather than loaded parts.

Material extrusion is FDM. A thermoplastic filament is heated and pushed through a nozzle that lays down a bead along each toolpath. ABS, PC, PETG, and PA-CF all print this way, which makes it the cheapest route to a large bracket or enclosure. Walls show layer lines, Z-direction strength sits well below the XY plane, and anything finer than ±0.3 mm needs post-work.

Where the two diverge is anisotropy. Cured resin is near-isotropic in the build plane; extruded plastic is not. If a part sees load across the layers, print it in another process or plan on a machined insert.

Laser powder bed fusion spreads a thin layer of polymer or metal powder and melts it with a laser or electron beam. SLS and MJF use nylon, usually PA12 or PA11, and produce strong, slightly porous parts with no support structures to remove. Metal versions (DMLS, SLM, EBM) reach full density in Ti-6Al-4V, 316L, and Inconel, which is why they turn up in aerospace and medical work.

The trade-off with bed fusion is surface finish and cost. As-built surfaces land around Ra 8–15 μm, and internal channels keep that texture. Metal powder is expensive, and every build needs stress relief and support removal. For a part with a critical bore or a sealing face, budget a secondary machining step.

Binder jetting takes a different route. A printhead deposits binder onto powder, and the green part is cured and then sintered or infiltrated. Throughput is high and the process handles large arrays of small parts at once. Dimensional shrink during sintering is significant, so holes and datums usually need machining after the furnace run.

Support removal is the dividing line here. SLS and MJF parts come out of the cake ready to use; metal powder bed and binder jet parts almost always need cutting, heat treat, and finishing before they fit anything.

Material jetting works like an inkjet printer for photopolymer. A printhead deposits droplets of resin and a UV lamp cures each pass. Multi-material and multi-color builds are routine, and dimensional accuracy is among the best of any polymer process. It is also the most expensive per part, and the resins age under sunlight. Good for full-color models, overmold-like prototypes, and dental or hearing-aid shells.

Directed energy deposition feeds metal powder or wire into a melt pool under a laser or arc. It is not a small-part process. Build rates are high and the working envelope can be large, so it suits repair of worn shafts, adding flanges to existing forgings, and near-net preforms that will be finished on a mill. As-deposited surfaces are rough, and tolerance is loose compared with powder bed.

Sheet lamination stacks and bonds sheets of paper, plastic, or metal, then cuts the outline with a blade or laser. Cost is low and the parts are large, but the material set is thin and the layers can delaminate. It survives in sand casting patterns and architectural models, not in functional hardware.

Put together, these last three fill gaps the first four cannot: full-color, large metal, and cheap big geometry. None of them replaces a machined surface where tolerance matters.

Printing wins when the geometry is organic, the batch is one or two, or the deadline is days. Machining wins when dimension, finish, or material properties are specified tightly. Our own shop holds ±0.005 mm on metals and Ra 0.2–0.8 μm on a fine finish, which no as-built print matches.

The common path is hybrid. Print a near-net shape, then machine the bores, sealing faces, threads, and datums. This keeps the weight-saving internal lattice from printing and puts a true metal surface where the part contacts something else. Metal powder bed plus 5-axis finishing is now a normal route for small aerospace and medical brackets.

Check the material callout before assuming a print will do. If the drawing names 7075, 17-4PH, or 316L with a hardness spec, printing will not deliver it in the same condition. Printed 316L is fine for a flow part; it is not fine for a wear surface.

Threads deserve their own note. Printed threads below M6 are unreliable and usually get drilled and tapped after printing. Same for any fit that needs a slip or press class. If the print is the finished part, design in a clearance and a printed-in thread relief so a tap has somewhere to go.