The Age of Additive Manufacturing
Ian Wright posted on July 13, 2017 | 9050 views
From left to right: A paleolithic hand axe from Norfolk, UK, dated 800,000-600,000 BP. Sputnik 1, launched by the Soviet Union on October 4, 1957.  A gas turbine manufactured by Mitsubishi-Hitachi Power Systems Americas; the yellow blades are produced using additive technology.

Technology is one of the defining elements of human history.

The Stone Age was marked by the development of the first tools: hammer stones and hand axes. The Space Age began with the launch of Sputnik 1. Most recently, the rise of the digital computer ushered in the Information Age or—following the archaeological convention—the Silicon Age.

Forecasting the next technological era is difficult, but from a manufacturing perspective it looks like the fourth industrial revolution—a.k.a. Industry 4.0—will involve the synthesis of automation, big data, deep learning, the Industrial Internet of Things (IIoT) and 3D printing.

Each of these technologies has the potential to revolutionize the way we make things on an industrial scale, but different industries are adopting them at different rates.

Industrial robots, for example, are a favorite of the automotive industry, which accounts for half of all industrial robot sales. In contrast, automakers tend to be wary of additive manufacturing, due in large part to extended cycle times and higher costs per part.

That’s why the biggest adopters of additive technology to date have been the aerospace, medical and power generation industries. All three can reap enormous benefits from the capabilities additive technologies provide, and will gladly bear the higher costs and longer lead times to get them.

The ceramic coating process for MHPS-AMER's turbine blades. (Image courtesy of MHPS-AMER.)
The plasma spray ceramic coating process for MHPS-AMER's turbine blades. (Image courtesy of MHPS-AMER.)
“When we make a small improvement in efficiency, it has significant financial benefits for our customers,” explained Paul Browning, president and CEO of Mitsubishi-Hitachi Power Systems Americas (MHPS-AMER), which was founded as the result of business integration of power systems by Mitsubishi Heavy Industries, Ltd. and Hitachi Ltd.

“To give you some numbers: if we improve the fuel efficiency of a product by one percentage point, over the lifetime of the product it saves the customer more money than the product costs, he continued. “So, for example, the largest gas turbines we sell are about $50 million [USD] each, and a one percent improvement in efficiency saves our customers about $100 million. That’s part of the reason why we’re early adopters of additive manufacturing—it creates so much value for our customers.”

ENGINEERING.com had the opportunity to sit down with Mr. Browning and discuss MHPS-AMER’s investments in additive manufacturing and the benefits it provides in further detail.


You’ve talked about a transition from the Silicon Age to the Additive Age. What’s been the catalyst for this transition?

I’m a metallurgist by training, so I’d been thinking of this time as the Silicon Age, given the impact the transistor has had, but I came to realize that the silicon age was enabled by additive manufacturing.

To make a silicon chip, you grow a single crystal one layer at a time through a controlled solidification process. If you’re a metallurgist, you call it “epitaxial crystallization,” but it’s basically an additive process. We’re now getting into 3D printing, but that’s just the newest form of additive manufacturing.


What can you tell us about the additive technologies MHPS-AMER is using to make its gas turbine blades?

A gas turbine has three major sections: a compressor where the air is sucked into the front end and compressed; a combustion chamber where the compressed air is combined with natural gas and ignited; and the turbine section where that ball of hot gas passes a turbine and causes it to spin.
(Image courtesy of MHPS-AMER.)
(Image courtesy of MHPS-AMER.)
The turbine blade is either grown into a single crystal, just like the silicon wafer, or it’s directionally solidified, which means there’s more than one crystal but they’re all oriented in the same direction. That makes the blade stronger, since the weak points are the boundaries between those crystals.

So, we start with a seed crystal that has the right orientation, then solidify metal on top of that. As the metal goes from liquid to solid in a very controlled fashion—one layer at a time—we solidify the next layer of atoms and they adopt the same crystal orientation as that seed crystal.

Eventually, you end up with a cast blade about a foot tall where every atom is oriented the way you want it to be.

This technology was originally developed for aircraft engines—for much smaller components. But our power plants use very large turbines, and obviously it’s harder to create a single crystal blade in a large format compared to a smaller one.

Previously, the same part would have been cast in a way that’s called “equiaxed,” which means it has a large number of grains that all have a random orientation. With our method, all the grains have the same orientation, which improves the strength of the blade at very high temperatures.

The reason we do this in such a specialized way is that the gasses flowing past the turbine blade are 1,600C, which is hotter than the melting point of the blade. It’s internally cooled, but the temperature of the metal still gets up to 1,200-1,300C. At those temperatures, you need a very specialized alloy made with a sophisticated solidification process to be sure all the atoms are arranged with the right orientation.

The casting process that gives you that directional solidification is a form of additive manufacturing, because you’re growing the blade one layer of atoms at a time.

We also put a ceramic coating over top of the metallic blade, deposited one layer at a time using robots and a process called “plasma spraying.” That ceramic coating acts as a thermal barrier, insulating the metal from those hot gasses.


How much post-processing is required once you’ve cast the blades?

The blade is cast pretty much as a net shape; the only thing we have to do is some machining at the root. 

F-series gas turbines. (Image courtesy of MHPS-AMER.)

F-series gas turbines. (Image courtesy of MHPS-AMER.)

You can put the ceramic coating on the older blades that were made without additive processes, but the way the technologies were developed, the directionally solidified blade came first and the ceramic coating came later.

There’s nothing that prevents you from putting that coating on equiaxed blades—and we do that sometimes—but generally it’s used in combination with directionally solidified or single-crystal components.

This is much slower, since you have to control the way the material solidifies to get that single crystal orientation. But we think about this in terms of the temperature capability of the part when it’s done, and by using this advanced solidification technique we get about 200C of additional temperature capability, and the thermal coating adds another 150C on top of that.

In our business, 350C is worth hundreds of millions of dollars over the lifetime of the power plant. It gives you much better fuel efficiency, so your fuel costs are much lower, and it means lower emissions of carbon dioxide and other pollutants.


Did your investments in additive technology require additional investments in safety measures or training?

The plasma spray booths were a different animal for us, and anytime we do anything different there’s obviously safety training involved, so there is some specialized safety training for those folks. But the real training is for our design engineers.

Inside a plasma spray booth. (Image courtesy of MHPS-AMER.)
Inside a plasma spray booth. (Image courtesy of MHPS-AMER.)
Most people who become design engineers think about materials having properties that are the same in all directions, but when we use this process to create a part where we’ve purposely oriented all the atoms in the same direction, it gives the material very different properties in one direction than in other directions.

So, you have to train your design engineers on how to design with that material. That was a big part of implementing this, and it’s not just training your people—you have to develop design tools that allow your engineers to work with materials that have very different properties.


What can you tell us about qualifying parts made with additive manufacturing?

One of the things we do involves an x-ray process that ensures the orientation of the crystals in the direction we planned for it to be. In a normal alloy, you don’t have to worry about that; but for these, we have to check to make sure we got the crystallographic orientation that we wanted. That’s one big difference with additive.

These are what we think of as mature additive technologies in our industry. Even though this is pretty high-tech stuff, we’ve been doing it for decades. What we’re doing nowadays is actually 3D printing, not of these parts, but other parts that in the past we would have cast in a traditional casting process. Now we have a process where we can use metallic powders and laser sintering to print parts one layer at a time and build up complex 3D structures with internal cooling passageways.

G-series gas turbines. (Image courtesy of MHPS-AMER.)
G-series gas turbines. (Image courtesy of MHPS-AMER.)

There are some things that you just can’t manufacture using traditional methods, keeping in mind that the biggest challenge on these is the internal cooling passageways. It’s hard to manufacture those using traditional techniques. This gives our design engineers a whole new level of freedom, where they can design pretty much anything they want. Anything that can be printed, we can now produce. The big challenge is making sure that, in this laser sintered powder system, that we’re getting the same mechanical properties that we would have gotten from a traditional casting process.

During the laser sintering process, there’s a lot of in situ monitoring of temperatures and dimensions. Once we’ve manufactured the part, there’s also destructive testing for mechanical properties and dimensional criteria, among other things.


What do you take to be the biggest barrier to adopting additive manufacturing?

The biggest barrier is, very simply, cost. In the aerospace and power generation industries, there are some very specialized components that cost thousands of dollars per part to produce. These are also products that have a huge impact on the performance of the machines we sell. So, they cost thousands of dollars to produce and have millions of dollars of value to our customers.

For these kinds of applications, we can afford to spend a lot of money to get exactly what we want. A small improvement in performance is worth a lot to our customers. So, I think it makes sense that we’re the pioneers in implementing and commercializing these additive technologies.

But as the cost of these technologies, particularly 3D printing, is falling rapidly, I think you’ll start to see it become more of a mainstream technology that gets adopted by the more cost-conscious industries, like automotive.
The first M501J gas turbine manufactured in U.S. shipped from the MHPS-AMER Savanah Machinery Works last year. MHPS-AMER has grown from five employees to 2,000 over the last fifteen years. (Image courtesy of MHPS-AMER.)
The first M501J gas turbine manufactured in U.S. shipped from the MHPS-AMER Savanah Machinery Works last year. MHPS-AMER has grown from five employees to 2,000 over the last fifteen years. (Image courtesy of MHPS-AMER.)

There are some small start-up companies that can handle 3D printing in plastics, but 3D printing in metal is a much more technologically challenging application. It was certainly pioneered by companies with the kind of in-house research capabilities that smaller companies just don’t have. But I do think that metallic 3D printing is now to a point where we’re starting to see mid-size companies in our supply chain make investments in that equipment. I think, over time, you’re going to see it become more and more mainstream for small and large companies.

There are some components on our machines that we will probably never transition over to additive manufacturing, just because there’s no huge benefit and the cost position is probably never going to be significantly advantageous. But, over time, we expect more and more of the components we manufacture will be done with additive technologies.


Do you have any advice to new adopters of metal additive technologies?

My biggest piece of advice to new adopters is to hire a good metallurgist. The biggest challenge of metal additive is getting the metallurgy right. Printing something that looks like what you want it to look like is actually not that hard, but printing something that performs the way you want it to perform—that’s the real challenge. In the metallic space, all of the challenges are in materials engineering.

Qualifying the parts is important, but it’s getting the properties right and being able to verify that they’re right that’s the biggest challenge.


For more information, visit the MHPS-AMER website.

Recommended For You