When a warship breaks down, the Navy needs to get it running again right away. A new GE Global Research program is developing ways to scan and 3D-print replacement parts out of metal melted with lasers, to get them back to the ship as quickly as possible.

The Navy cares about this technology because its fleet is aging, with ships dating back decades. Many of the parts that make up these vessels are so old that substitutes are no longer manufactured, and replacement parts have to be custom-designed.

This isn't unique to the military. For other specialty-part orders, such as those for classic cars, drivers are already buying 3D-printed parts, made to spec when antique parts aren't available. In fact, BMW used 3D printing to rebuild a BMW 507 once owned by Elvis Presley. But the rapid-response aspect is crucial for military vessels and for the hardware they carry. 'For example, if there is a Navy aircraft that has an issue, they can print the parts online and send them to a ship at sea,' instead of waiting longer to search for an out-of-stock part or refabricate it using a slower method, says Ade Makinde, a principal engineer for additive technologies at GE Global Research, who is leading the project.

The technology, called laser powder-bed fusion, already exists. In fact, GE is already printing fuel nozzles for jet engines, gas turbines and other components. But the new research program is looking to embed new sensors in the process that can monitor the part as each metal layer takes shape, speeding up production.

This type of 3D printing, or additive manufacturing, uses a metal powder laid on a flat plate. With a powerful laser, the machine quickly melts the powder one small section at a time. As the powder melts into the correct shape, the machine deposits another layer of powder, corresponding to the geometry of the part. Software algorithms determine where exactly the powder should go and how the laser should fuse it. 'We are going to embed sensors that will actively inform the user about what is going on as the part is being built layer by layer, when certain defects occur,' Makinde says. The machine could self-correct if the sensors detect problems with the material. If the new project succeeds in giving the machines' more insights, they could start working faster.

That's important because the replacement part not only has to be precisely the same size as the original part, but also has to share the same exact physical properties. 'You have to validate the part in terms of its quality and attributes before you can safely say, 'I can put this in a car or on a ship,'' Makinde says.

A mechanical engineer who's been with GE for 21 years, Makinde is leading a team of a half-dozen materials scientists and other researchers based in Niskayuna, New York, near the upstate town of Schenectady. He sees parallels between this project and prior work he's done on casting. 'You have no idea what's going on until you break the mold,' he says.

The research team will be working on ways to see inside the 3D printer and determine if any correction is needed. They will also be considering the mechanics of the process, Makinde says: 'When the laser strikes the powder, how does that liquid form, and what shape does the liquid take?'

Makinde's team, GE Aviation and GE Additive are working with a consortium of companies - Honeywell, Penn State University, Lawrence Livermore National Lab, Nuclear National Lab and the National Center for Defense Manufacturing and Machining - on the $4 million Navy-funded project, which is expected to take four years.

Top image:In the Navy: Sailors and Marines with the 11th Marine Expeditionary Unit scrub down the flight deck of the amphibious assault ship USS Makin Island. The ship is the Navy's first vessel powered by hybrid propulsion. Image credit: U.S. Navy.

Attachments

  • Original document
  • Permalink

Disclaimer

GE - General Electric Company published this content on 02 May 2018 and is solely responsible for the information contained herein. Distributed by Public, unedited and unaltered, on 02 May 2018 14:42:10 UTC