ResearchNews

MatterWave claims up to 10x fatigue life gain with vibration module for LPBF

US startup MatterWave reports early results showing up to a 10x improvement in fatigue life compared with conventional builds, using a retrofit module called Daedalus that sends tuned mechanical vibrations into the melt pool as it solidifies during laser powder bed fusion (LPBF). The figure comes from the company’s SME award submission, not a published peer-reviewed study.

Daedalus is a stack of transducers that sits at the build plate of an existing metal printer. It introduces vibrations, described as phonons, at a frequency and amplitude tuned to each material’s thermo-mechanical properties, shaping how crystals nucleate and align as the metal passes from liquid to solid. CEO Alex Tiliță says the mechanism lets alloying elements diffuse out of the crystal lattice before it fully forms, reducing the internal stress that makes high-carbon steels and other crack-prone alloys brittle when printed conventionally.

The company has installed Daedalus on an EOS M 290 and is running a qualification build for Inconel 718 under an aerospace contract. The goal, Tiliță says, is to show the technology can reduce the burden of heat treatment and qualification, which he says depends on narrowing the variability between machines, geometries and build-plate positions. He estimates qualifying a metal AM process typically takes one to two years and can cost around a million dollars or more. At the International Conference on Advanced Manufacturing in Orlando, he said customers asked about applying it to 7075 aluminum, Hastelloy, Inconel 738, 939 and 247, and tool steels.

MatterWave, co-founded by Tiliță, Ben Ferrar and Caitlin Oswald, is a subcontractor on a US government Small Business Innovation Research grant and expects a commercial launch in roughly two years. Tiliță says research groups have validated the approach he first published in 2017 across different machines, materials and frequency ranges, though he names none of them, and he cites no sample sizes, test standards or third-party data.

Sources