3D printing makes inroads in manufacturing
New York, September 27, 2011
Major US manufacturers are now adapting the 3D printing technology to work in applications including jet engines, which require highly engineered metal parts able to stand up to extreme heat and pressure.
'The potential is huge from the standpoint of design flexibility,' said Todd Rockstroh, a consulting engineer at GE's aviation arm, which is the world's largest maker of jet engines. 'You can literally design and manufacture things that are generally unmanufacturable today.
GE, the largest US conglomerate, is working on the technology at its main research lab in Niskayuna, New York, outside Albany. There, engineers are using it to produce nozzles that will be used to spray fuel into jet engines through a process called 'powder-bed manufacturing.'
It works by laying down a thin layer of powdered metal -- cobalt chrome in this case -- and running a beam of computer-controlled laser light over it, which heats the powder enough to weld it into the shape of the first layer of the nozzle. When that layer cools, more powder is laid down, the laser makes another pass and slowly builds the shape up to its full size -- the nozzle itself is about the size of a very narrow computer mouse.
Additive manufacturing's name highlights the contrast between the technique and traditional manufacturing, where many engineered parts start out as blocks of metal that machinists grind down into the shapes they need.
The new technique cuts costs by several thousand dollars on a roughly $4 million engine by Rockstroh's estimate. It works in two ways, first turning what had been a complicated process of welding together 15 to 20 tiny parts into a two- to three-step procedure. Second, it saves raw material by not grinding away chips of metal that are not longer suited for use in jet engines.
The nozzles will be used in a jet engine GE plans to begin making in 2014.
Prabhjot Singh, the GE lab manager who has been working on the technique, said the main breakthrough was being sure the additive process could consistently produce parts that were dense enough to hold up inside a jet engine.
'The process was never this precise,' Singh said on a recent tour of his lab. 'Sure you could make show-and-tells, but when you are talking about the aviation industry, these parts are safety-critical.'
Even advocates of the technology say that it will only be used in limited cases -- GE will likely never produce an entire gas turbine in one pass, Singh said.
Holding down soaring metal costs
While saving labor and allowing engineers more freedom in their designs is part of the appeal of the additive approach, its advocates said that saving on raw materials costs is also a big part of the appeal, and with good reason.
Rising demand in rapidly industrializing China and India has driven metal prices sharply higher in the past few years. Copper and steel have more than doubled in price since early 2009 as measured by trading in their futures.
'Material is getting more expensive by the day,' said Tahany El-Wardany, principal engineer at United Tech.
The Hartford, Connecticut-based company uses the technology for making prototypes but does not yet believe it is ready for use in actual engines, said Agnes Klucha, an engineering manager at United Tech's Pratt & Whitney division.
El-Wardany estimated the additive approach can cut a manufacturer's raw material use on a given piece by 50 to 80 percent, depending on the design. That sort of cost savings will eventual catch Wall Street's eye, one analyst said.
'Every little bit helps towards lowering the overall costs of the product,' said Daniel Holland, an equity analyst at Morningstar in Chicago, who follows GE. 'You'd expect the company to continue to evolve and figure out better ways of doing things such that they can save investors money and push the margin profile of the business a little higher.'
The hurdle is how effective it will prove in producing the huge volumes of parts that big manufacturers need.
'It's this question of mass manufacturing, that's the place where it will go,' said Michael Cima, a professor at the Massachusetts Institute of Technology who patented one of the first 3D printers in 1993. 'When you get up into the millions of pieces, nobody has built the machine that will do that.” – Reuters