3D printing, or additive manufacturing, of polymer parts for aircraft interiors is decades old. But metal is not naturally a convenient liquid, and metal aircraft parts tend to be safety-critical, so printing metal parts has taken longer to advance.

Nevertheless, the pace of metal printing is picking up. It started with an OEM of engines, where weight and performance are critical. It is now spreading well beyond to a variety of components and structures.

OEMs are also diversifying their use of 3D technologies. Metal powders as well as wires are being used, and different techniques for printing metal materials are being exploited.

The two simplest appeals of printing have been cutting waste and producing parts quickly in small volumes. The first can cut costs, while the second speeds parts to market, eliminating the need for massive inventories.

But, 3D printing’s real potential is in redesigning whole systems and components to the exact configuration and content required for their functions, not designing a part or collection of parts just to ease manufacture by conventional methods. This potential will only truly be tapped as new engines and aircraft are developed.

So there is much more coming in additive manufacturing. Smart OEMs are taking early steps, partly for reasons specific to each printed piece, but also to gain experience for the future additive aerospace industry.

Honeywell Aerospace has already begun printing metal parts for its auxiliary power units (APU). Donald Godfrey, an engineering fellow in additive manufacturing at Honeywell, explains the change: If a component already is in use, then printing it replaces sheet-metal forming, welding, casting or forging. “Honeywell is going through this exercise on many parts and applying additive manufacturing technology where it makes sense,” he says.

Godfrey notes that the basic differences between propulsion engines—the first to use printed parts—and APUs are simply size and horsepower. “Both are turbines and have the same sections: compressor, combustor, high-pressure and low-pressure turbines and exhaust,” he says. Printing APU components has thus been a high priority at Honeywell.

The OEM’s first printed part, a splash guard for an APU, was shipped to a customer Dec. 18, 2015. Godfrey says it was a very simple part and printed mostly so that Honeywell could understand what it took to get printed parts through its own quality-control system, engineering approvals and acceptance by the FAA and the customer.

Honeywell is now printing ducts and oil tubes for APUs. These parts were selected for one of four reasons: the OEM was having difficulty with vendors, printing could reduce lead times, printing could improve quality or printing could reduce costs.

To print these parts, Honeywell is using two 3D technologies. Most printing is done with laser powder bed fusion, also called direct metal laser sintering (DMLS). A smaller portion uses electron beam (EB) melting, also known as EB powder bed fusion. The OEM is also looking at binder jet printing and direct energy deposition (DED).

Honeywell is evaluating all nonrotating components of APUs as candidates for additive manufacturing. These include compressor and turbine cases, brackets, mounts, tubes, vanes and diffusers.

A much smaller, younger company, Norsk Titanium, has chosen a very different additive approach. The 140-person manufacturer uses its version of DED with wire to print footlong titanium fittings to hold aft galleys in place for 10 Boeing 787s each month.

Chief Commercial Officer Chet Fuller says small, intricate parts such as GE Aviation’s fuel nozzle are best printed by powder bed fusion techniques. For the big, structural titanium parts that Norsk wants to print, wire-fed technology is better. Norsk uses argon gas torches in a patented process called Rapid Plasma Deposition. “[Our parts] start at a pound or larger. Powder takes too long for that,” says Fuller. Indeed, the Norsk machines print kilograms per hour, not ounces per hour, as powder bed fusion does.

The wire-fed approach only produces parts to near net shape, after which excess metal must be machined away. But it is fast and economical for any volumes. And its great advantage is that it cuts the buy-to-fly (BTF) ratio for pricey titanium substantially, compared with cutting titanium blocks. Conventionally cutting a titanium structural part results in a 15:1 BTF, Fuller estimates. Norsk’s wire-fed printing has a BTF of about 3:1. That saves 24 kg (53 lb.) of titanium on a 2-kg part.

And because it is using wire, Norsk is not paying the exorbitant costs of metal powders. These can reduce or offset BTF gains. Titanium wire costs more per kilogram than titanium block, Fuller acknowledges, “but not dramatically more.” The OEM uses standard rolls of Ti64 titanium, made to its own specifications for twist and cleanliness.

The current Norsk machine has a work envelope of 900 X 600 X 300 mm (35 X 24 X 12 in.). The company is now deciding whether its next machine will aim for larger parts or just more complex ones. Fuller says that the best targets for his equipment are the latest aircraft—787s, Airbus A350s and the Bombardier C Series—that make heavy use of titanium.

Norsk is producing hundreds of test parts in various shapes and sizes to demonstrate to regulators and various OEMs the suitability of its methods for many more parts. Once it has proved the capabilities and consistency of its deposition process, certifying individual parts is much easier and takes much less time. The OEM can now use simulation software for some tasks leading to certification.

Fuller says the best candidates for printing are parts with a significant vertical dimension. In any case, the Norwegian OEM should have plenty of work ahead. It has partnered with Spirit AeroSystems, which builds thousands of titanium parts. Spirit executives expect that at least 30% of these parts could be candidates for Norsk’s rapid-deposition process.

Major aerospace companies that are not yet printing parts are eager to begin doing so. Eaton’s Aerospace Group is investing aggressively in additive manufacturing, with Mike York, director of aerospace additive manufacturing, leading a team of Eaton business lines dedicated to specific additive applications.

Eaton maintains a center of excellence in Southfield, Michigan, for developing both polymer and metal additives. The center is piloting initial development and will provide low-rate production parts.

In addition, the center is developing its own high-temperature polymer materials for use in additive manufacturing. Eaton will go to market to source metal material for 3D printing. Additionally, a larger Eaton team in Pune, India, is working on simulation of additive processes and products.

“We see this as a good opportunity in aerospace—to gain a competitive advantage in speed, cost and performance,” York says.

Eaton is chiefly interested in DMLS, EB, DED and cold spray techniques for metal additives. For polymers, it is concentrating on direct laser sintering and fused deposition modeling while also exploring other processes.

Eaton has identified manifolds, valves, filtration components and other products as candidates for metal additives. For polymers, its investigations include replacing some aluminum parts with additive polymers to save costs and reduce weight.

One critical factor in deciding the suitability of an additive for making a part is that part’s complexity. Traditional methods meant complexity increased cost, in either manufacturing or welding. With additives, the mantra is “complexity is free.” York says that as additive processes improve, the degree of complexity required to make it the less expensive option decreases. “Applications that did not look attractive two to three years ago now do,” he says.

And it is not just cost that can be saved. York says the new designs that additive manufacturing enables can improve performance by reducing weight or size, or improving fluid flows or pressures.

Eaton plans to launch its first additive-manufactured aerospace parts by the end of 2019, but most of the additive portfolio will come later. York predicts that in 5-10 years, a significant portion of the Eaton portfolio will be made through the additive process.