Complementing Not Competing — An Automaker’s Mindset About AM and Castings
Is additive manufacturing (AM) poised to replace castings? At General Motors, that’s not the mindset and won’t be any time soon, according to Ante Lausic, Ph.D, lead process engineer for metal AM. For that to happen, a new leap in AM machine technology will have to occur, he says, and indeed the automaker is investing in multiple avenues from binder-jetting to area printing, which could eventually improve the economics of large-volume AM. Today, however, GM’s view is to use AM to complement castings, not compete with them.
Taking its knowledge of AM’s greatest benefits (such as enabling great complexities) and accepting its limitations (such as maximum build size and total machine throughout) the company has zeroed in on two highly advantageous applications for additive: steel tooling inserts and tool refurbishment.
AM, of course, is not one single method, but rather, like metalcasting, is a broad production category with different processes used to achieve a finished part. Lausic, who is a member of multiple standards developing organizations, has been developing and deploying metal additive technologies across the GM network for three years and has been part of AM as a whole for over 13 years. He says the definition of AM varies depending on who you ask, but whether it be ISO/ASTM or Wikipedia, three common elements are present when describing AM: three-dimensional, digital data, and layer-by-layer. GM, says, Lausic, defines AM as “a field of manufacturing processes that create three dimensional objects directly from digital data through successive addition of layers of material.”
The separate processes for metal AM include directed energy deposition (DED), binder jetting, material extrusion, sheet lamination, and powder bed fusion (PBF). The latter has the most industry acceptance and is the most commonly used today, said Lausic.
“You’re spreading powder layer by layer using a laser of high power, typically 400–1,000 watts,” he said, “tracing a 2D picture and melting all those powder particles together to form your 2D slice. Then your build plate goes down and you spread a new layer of powder and you melt that layer to the previous one. And you repeat that process over and over again.”
A major advantage of powder bed fusion is that it makes true conformal cooling possible in their tooling inserts with internal surfaces much smoother than other metal AM approaches, said Lausic. PBF can typically achieve 8–30 µm surface finish and fully dense parts (up to 99.995% dense), depending on the exact machine, material, and part size.
The down side is, it is an extremely slow process, typically printing 150 grams per hour per laser for a steel tool, Lausic said.
“It’s also very difficult to make very big parts,” he added. “So, if you have a part that’s 400 or 500 millimeters or larger in size, machines of that size are rare; and the parts are very difficult to manufacture because spreading that powder layer each time becomes exceptionally difficult on those large areas.
A Perfect Complement
GM has capitalized on the conformal cooling in tooling inserts via PBF to solve the challenge of thermal imbalance in die casting and has realized improved cycle time as well as reduced scrap and solder. One example is sub inserts in valve bodies, which have extensive small ribs that make them impossible to individually cool conventionally—but with the freedom and accuracy of PBF, water can be brought right up inside those ribs.
“We’re working to try and improve what we can do with tool steel inserts—this is one of our largest areas of metal additive work,” Lausic said. “What we’re working on is how do we specifically improve the limitations on the maximum size of a waterline you can put under the tool ... How do we get a bigger line to distribute more water without creating these stress concentrators in the line, which could be crack-initiation sites that might lead to a tool failing prematurely in production. We recently added two new metal powder bed machine machines in our building in the last couple of months that we’re hoping will allow us to make bigger lines with smoother surfaces all around. This will help increase the life of those inserts by preventing some of that early cracking in the insert, as well as allowing more cooling in some of those large areas.
“We’ve retrofitted an insert to be manufactured additively,” said Lausic. “To feed the waterline in a traditional piston tube, water travels through the insert where we need it based on the simulations that we run, and then it exits out the same hole as if it was a jet-cooled piston tube—we’re able, with minimal plumbing, to provide water to a very large area of the tool. And we have lots of examples of putting those water lines into extremely thin features all the way down to a 5mm rib.”
Additive can achieve those small features, Lausic said, but great care must be taken. Distances to surfaces are so small that any mistake in printing, machining, or heat treating can lead to non-symmetric loading and premature failure.
Lausic says additive has revolutionized die builds. Conventional dies require extensive cross-drilling just to bring cooling water to a simple area. In addition, thermal imbalance leads to shortened die life and more frequent maintenance.
“What if you rethink how you design the main insert using additive as your way of achieving waterlines in areas you couldn’t get traditionally?” he noted.
“Maybe your main insert is no longer the cavity insert at all. Maybe your cavity inserts are the additive portions … And because we’re making it additively, we can bring water to the right places, we can balance the tool, it’s producing better castings, and it’s faster—not to mention the reduction of plumbing, the ease of assembly, and the ease of repair. There are a lot of benefits on top of conformal cooling.”
Looking ahead, GM has its sights on moving toward machines able to produce bigger, 600mm inserts, but, according to Lausic, what will further improve shot life of dies is better materials. Thus far, the company has employed maraging steel, the prevalent tool steel material in the market. While easy to print due to its low carbon content, it comprises many expensive alloying elements such as cobalt and nickel, and its high-temperature performance doesn’t compare to that of traditional H13, he said.
“What we really want from our powder suppliers is weldable H13-like alloys that give us the right hardness, the right toughness, and the right thermal stability—that will be a game-changer for our industry as a whole,” Lausic said. “There’s about six or seven different novel tool steel powders being manufactured today from metal powder suppliers, all geared toward die cast tooling, which is likely the largest volume application in automotive metal AM right now.”
Dollars and Sense
It’s not uncommon for the additive group at GM to receive requests from different areas of the company to convert a sheet metal part into an additively-made part, but Lausic says the cost justification is often lacking in these scenarios. Broadly speaking, he explained, there exists a point at which AM’s volume limitations make conventional processes more cost effective.
“I think about it in terms of the economics—additive is very, very expensive,” he said. “Can it create something that a casting can’t create? Definitely—it offers design freedom and makes complexity possible. But we consider ourselves an enabler of conventional manufacturing rather than a displacer of conventional manufacturing. The reason is all geared toward the dollars and cents. Breakeven volumes will be exceptionally low compared to casting, stamping, or molding. Our current on-vehicle success stories are in low-volume programs like luxury vehicles or early prototypes.”
An example is the shift knob emblem on the Cadillac Blackwing program, he added. They are manufactured with binder jetting technology using stainless steel. It’s a non-safety-critical component for a low-volume program, he said, which is why it made sense to go down the additive route for this individual component.
Making the business case for additive boils down to this: Even in an ideal scenario of technologies like area printing panning out (in which powder particles could be exposed in a large area all at once), the cost is estimated to still be 50 cents to a dollar per cubic centimeter (cc) compared to 20 cents or less per cc with die casting.
“Our sci-fi future needs to have another magnitude jump to get to a point where they can actually start competing with the casting world,” Lausic said. “And even then, we’re talking about small components versus making large structural components, for example. When the parts get big, it gets harder to make a business case for AM.
“So, is additive going to replace all of our castings? That’s not where we see additive going. What we see additive doing is being another tool in the toolbox, a way of complementing what we’re already doing in our traditional conventional manufacturing roles.”
The Next Exciting Application
Lausic and his team have also focused on another noteworthy application for additive: a hybrid strategy to give tooling a second, third, and fourth life through AM refurbishment. Ordinarily, when a tool cracks, diecasters will have to wait four to 16 weeks for a new tool to be made. But now, they can have the old tool sent over to the metal AM team to be repurposed and brought back into production rapidly.
“We go through cycles where we use our tools and throw them out because they’re out of tolerance or they’re no longer capable of being reused, he said. “Additive offers us a realm where we can take that old insert, remove a certain portion of it, and ‘grow’ back the geometry, make an engineering change, or anything else you want to put on top of the tooling piece and repurpose that old material.
“This strategy has enabled us to effectively halve the costs of our additive inserts. We’re able to actually repurpose that ‘business end’ of the tool or even start from the very beginning with a forged block of some other material and then print the end on top of it with all the features that we may want inside of it.
“Repurposing tools is a very exciting realm. And we can just keep adding automation to make this bigger and faster. So instead of using the powder bed approach, we can also use our robotic approaches with DED to find ways to repair tools for several more life cycles.”