Research
New techniques, materials, and innovations pushing additive manufacturing forward.
How Printers Learned to Calibrate Themselves
Automatic bed leveling and accelerometer-driven input shaping have quietly replaced two of the most tedious rituals in desktop 3D printing, turning manual tuning into something the machine now does for itself.
Squeezing More Life Out of SLS Powder
Selective laser sintering leaves most of a build's powder unsintered, and a growing body of research on how that leftover material ages is what lets shops recycle it instead of throwing it away.
What Makes Generative Design Actually Generative?
Generative design software doesn't sculpt parts the way a person would - it grows them from a set of rules, and understanding that difference explains both the organic shapes and the odd surprises engineers run into.
The Densification Curve for Printed Ceramics
Binder jetting and vat photopolymerization both start a ceramic part well short of full density, and a maturing toolkit of powder packing, isostatic pressing, and sintering control is closing that gap layer by layer.
How Print Simulation Sees Warping Coming
Thermal simulation software now models an FDM print layer by layer as it cools, catching the warping a part will develop before it's ever extruded. Recent research shows both classic finite-element models and newer machine-learning hybrids are closing the gap between predicted and measured distortion.
How Bioprinted Tissue Gets Its Blood Supply
Bioprinting solved the geometry of living tissue years before it solved the plumbing, and the last decade of vascularization research is largely the story of closing that gap.
Why 3D-Printed Parts Miss Their Tolerance
Dimensional drift in FDM and SLA parts traces back to a specific chain of thermal and cure kinetics - not just printer calibration - and understanding it changes how tolerances should be specified.
Where Two Materials Meet
Multi-material additive manufacturing lets a single build combine rigid and compliant, or dissimilar, materials without assembly, but the interface between them remains the part's weakest plane.
Continuous-Fiber Printing's Weak Interlayer
Continuous carbon-fiber 3D printing can match the tensile strength of conventional composites layer by layer, but interlaminar bonding and trapped voids between passes remain the process's defining weak point.
What Melt Pool Monitoring Can (and Can't) Catch in Metal 3D Printing
In-situ sensors watch the melt pool for the fraction of a second it exists, catching the keyhole instabilities that leave pores behind — but turning that detection into real-time process control remains the field's open problem.
Why 3D-Printed Ceramics Still Crack
Vat photopolymerization can print ceramic slurries to near-optical resolution, but debinding and sintering remain the steps that decide whether the part survives — and why shrinkage and cracking are still the field's defining problem.
Three Ways to Build in Metal
Laser powder bed fusion, directed energy deposition, and binder jetting reach full density through fundamentally different physics, each trading resolution, build rate, and cost differently.
Lattices That Design Themselves
Topology optimization and lattice infill are converging with additive manufacturing's own physical constraints, so that overhang angles and minimum feature size are now part of the mathematics that generates a part's shape, not an afterthought.
Why Two CT Scans of the Same AM Part Can Disagree: Porosity Metrology and the Partial-Volume Effect
X-ray computed tomography has replaced destructive sectioning as the primary way to inspect internal porosity in metal additive manufacturing, but its porosity numbers are only meaningful alongside the resolution and threshold used to produce them.
Melt, Deposit, or Bind: The Process Physics Behind Metal AM Selection
Laser powder bed fusion, directed energy deposition, and binder jetting each rely on a distinct thermal or chemical mechanism, and that mechanism, not the equipment, governs achievable density, microstructure, and part geometry.