Printing Cars from: IPMediaonline.com
Automakers around the world turn to additive manufacturing to reduce costs, cut development time, and increase flexibility
Don’t expect to 3D print the frame for a BMW X5 anytime soon. The engine block for an Audi TTS? Not a chance. No one in either the automotive or additive manufacturing industries would suggest that 3D printing, despite its amazing capabilities, can compete against the well-known manufacturing processes used to produce the nearly 100 million passenger and commercial vehicles sold each year1.
Feeling Bleu
At least, not yet. Mark Kirby, additive manufacturing business manager for Renishaw (Canada) Ltd., points to a number of automotive projects that he and other 3D printer companies are currently working on, activities that will undoubtedly lead to increased use of the technology in future, higher volume applications.
These include the Bleu car hinge, a “light-weighted, topologically-optimized, gullwing door bracket” that Renishaw produced in collaboration with 3D design software provider Dassault Systèmes2. Volkswagen is exploring potential weight reduction with 3D-printed window frame supports and water connectors3, and Ford is “proving the technical feasibility for high-volume 3D printing.”4
“It’s clear that we’re starting to see a large number of niche applications for additive manufacturing in the automotive sector,” says Kirby. “Aside from its obvious lightweighting potential, there are a number of opportunities for producing novel, high-performance product designs—by reducing part counts in turbocharger assemblies, for example, or improving the thermal characteristics of various engine components. Additive is definitely an enabler in this area.”
Kirby lives in the world of metal powder bed fusion. Whether you call it direct metal laser sintering (DMLS), selective laser melting (SLM), laser metal fusion (LMF), or any of the other provider-specific acronyms, it uses a laser to melt metal powder roughly the consistency of flour one paper-thin layer at a time. As each pass is completed, a fresh layer of metal powder is spread across the previous layer and the process begins again until the workpiece is complete.
Compared to machining, forming, casting, and other traditional metalworking processes, metal printing is a seemingly glacial way to produce something. Worse, the high amount of heat necessary to melt the metal powder, followed by rapid cooling, can generate extreme tension within the workpiece. Support structures are often needed to hold the part in place during the build process (which must be machined away afterwards) and heat treating is typically required to relieve stress. Both of these add cost and increase lead times.
A Printing Epiphany
Yet designers are learning ways to avoid these problems. Joshua Mook, engineering leader at GE Additive, points to the much-touted fuel nozzle used in the company’s LEAP jet engine as one instance where support structures were either eliminated or integrated into the finished part, greatly reducing post-production processing.
“We changed our design philosophy in a way that is amenable to the powder bed process,” he says. “Supports can be removed robotically or by hand, or sometimes they don’t need to be removed at all. These and similar techniques are making powder bed technology far more productive than it once was, and I would say the number one thing for me is teaching my fellow engineers to design with the printing process in mind, and to maximize the efficiencies it offers.”
What do parts that go into the Airbus A320neo and Boeing 777X passenger aircraft have to do with automobiles? Plenty. GE Additive is currently printing 60,000 fuel nozzles per year, and has its sights set on far more. It’s invested more than $1B US in 3D printing machine builders Concept Laser and Arcam, and is working hard to not only make 3D printing more productive, but less expensive as well5.
“Right now, everyone thinks of 3D printing in terms of high value and low quantity,” Mook says. “But even at our current production levels, we’re edging into automotive volumes, and as the technology continues to mature and printing becomes faster—which it already is—it will find its way into a number of higher volume applications. In the automotive space, for instance, there are terrific opportunities for complex mechanical components, drive train and engine parts, fuel cells, and more. We are very interested in pursuing these and other markets that can benefit from the radical new designs made possible with 3D printing.”
One way to move this along is by reducing material costs. Market pressures and increased use will help in this regard, but GE Additive is one of the many manufacturers either developing their own metal powders or working with suppliers such as Carpenter, Sandvik, and TLS in an effort to make gas atomized alloys more affordable. These include tool steel and aluminum, Inconel and stainless steel, and are just a few of the dozens of metal alloys that can be 3D printed into fully dense shapes, producing robust component parts as well as the tooling such as molds and fixtures needed to manufacture these parts via traditional means.
Keeping Cool
One notable example of this is the development of vertical machining centers with additive manufacturing capabilities. These hybrid machine tools can lay down and fuse a dozen or so layers of metal powder and then mill the resulting features to tolerances much tighter than the fractions of a millimeter currently possible with 3D printing. This allows engineers to create complex tooling previously impractical or downright impossible to produce, and is especially relevant for plastic injection molding. By printing conformal cooling channels directly within the molds and inserts used to produce millions of plastic parts annually, significant reductions in cycle time can be realized.
Tom Houle, director of Matsuura Machinery’s LUMEX line of hybrid machines, says automotive OEMs and their suppliers alike are pushing hard on additive technology for this very reason. “The smart companies see the writing on the wall,” he says. “By using conformal cooling channels, molding times are easily 30-percent shorter, and are oftentimes cut in half. Additive manufacturing also allows us to create porous surfaces within the mold, so you can simplify venting in deep ribs and similarly complex features that are often problematic with conventional machining—things like door components, dashboard grills, that sort of thing. It makes a huge difference.”
Because the mold and mold insert layers are alternately printed and machined, there’s no need for post-build stress relief as is common with most additively manufactured metal parts; in many cases, users can remove the tool from the 3D printer’s build chamber and bring it directly to production. However, because the Lumex and other metal powder bed fusion machines employ a metal build plate as a platform on which to print parts, and because these parts are basically “stuck” to the plate when the job is done, it’s possible that some small part movement can occur when the two are separated
Predicting and avoiding such build events is something that virtually everyone in the 3D printing industry is working on, Matsuura included. This is why companies such as AutoDesk and 3DSystems have developed simulation software that offers a predictive view of what happens when laser beam meets metal, and continuously improving the outcome.
“A number of companies are fully invested in developing additive simulation solutions,” Houle says, “In our case, we’re interested in designing the most efficient conformal cooling lines possible based on the model geometry. But anyone designing 3D printed parts is looking for ways to leverage its capabilities. This usually means using organic, skeletal shapes that do a better job of strengthening part geometry where it’s needed, and produce lighter, more robust, and more cost-effective parts overall.”
Truing Up
Another way to cut costs is by increasing printer speed. Renishaw, Concept Laser, and various other machine builders are doing this by adding more lasers to their wares, and using them to build multiple parts at one time, or make a team effort approach and build a single part more quickly. The trick is to do so in the most efficient manner possible.
Metalworking machine builder Trumpf is one of these. Automotive industry manager for additive manufacturing Frank Nachtigall says that adding two or three lasers to a 3D printer does not double or triple the price of the machine, but it does offer the potential of significantly faster build speeds.
“Aside from supporting the known industries such as aerospace, medical, and tool and die that currently use powder bed manufacturing, we are also participating in a number of research and development projects in the automotive market,” he says. “These will require continuing collaboration between the machine builders, the OEMs and their Tier 1 and Tier 2 suppliers, but there’s definitely a good deal of interest in the mass production of automotive parts.”
In the meantime, Trumpf is one of the many additive machine builders focusing a big chunk of their efforts on motorsports such as Formula 1 racing, which as Nachtigall points out have used additive manufacturing for the past 15 years, and can easily justify the high cost of 3D printed vehicle components by their lower weight and improved performance.
“We’re also seeing a shift towards the luxury segment for certain sports cars applications, which are not produced in very large numbers,” he says. “That might mean anywhere between 5,000 to maybe 30,000 per year of parts for the power train, cooling systems, ventilation—anywhere you can gain efficiency by combining several pieces. The bottom line is that there are very many interesting parts with the potential to be additively manufactured.”
The problem remains the cost, a factor that everyone in the industry knows automakers are sensitive to; even if thousands of machines were purchased and production scaled up to the point that part runs in the millions became possible, the sad reality is that current powder bed fusion technology would remain too expensive for mass production of commuter car and F-150 pickup truck components.
Farming the Future
Fused deposition modeling (FDM) has been around for nearly three decades. It prints parts by squirting molten plastic from a filament-fed nozzle onto a build platform. Because different materials can be combined in a single build, it is a favorite for prototyping multipart assemblies, but it is nowhere near fast enough nor does it produce parts robust enough for automotive mass production.
Still, the concept is sound. By replacing the filament with a metal-infused thermoplastic and then sintering the finished product in a high temperature oven, the plastic binder melts away in a manner much like metal injection molding (MIM). This produces fully dense metal parts in large, cost-effective quantities, and does so much more quickly than other metal-based printing methods. It requires a new acronym, however: ADAM, short for atomic diffusion additive manufacturing
“Even though the technology is still in its infancy, we’re comfortable talking about production runs in excess of 100,000 pieces.” That’s according to Jon Reilly, product manager for Markforged, who envisions massive farms of such print devices. “In the early days of computing, you’d have a mainframe server that needed an entire room to house it and a skilled technician to keep it running,” he says. “ADAM is more like Amazon Web Services, which is a farm of blade servers whose processing power anyone can access at the click of a button. By organizing dozens or even hundreds of relatively inexpensive machines in parallel, we believe it will offer print speeds and production costs low enough for the automotive and consumer products industries.”
By switching print materials, this same technology can be used to produce carbon fiber reinforced plastic components as well, offering additional appeal to automakers aiming to reduce product weight while preserving integrity. And since the metal powder and sintering process Markforged employs is virtually identical to that used by metal injection molders for the past fifty years, there should be little problem with acceptance within the engineering and quality communities.
Like many 3D printing methods, Markforged promises to dramatically shorten multi-year automotive development cycles, a paradigm that, as Reilly points out, has lasted more than a century. “Ford Motor Company was founded in 1903 and shipped the first Model T in 1908. It took them five years. Tesla was founded in 2003 and shipped the Roadster in 2008; one hundred years later and it still took them five years to ship their first car. Additive manufacturing is beginning to change all that, and will one day shorten development cycles from years to months. That’s our goal”
A Top Down Approach
Printing powertrain and engine components is well and good, but what about all the connectors and gaskets and knobs and ductwork? We discussed the potential for 3D printing of plastic injection molds earlier, but what if an automaker could just skip the molding process entirely? Not only would investment costs be substantially reduced, but so too would the length of the development cycle. So the question is this: if the 3D printing industry can produce bespoke tennis shoes, why not custom dashboard parts?
Someone addressing this is Phil DeSimone, vice-president of business development at Carbon. “Automotive is definitely becoming more and more of a niche market,” he says. “In the US alone, there are over ten million vehicles sold every year that are produced in sub-twenty thousand unit quantities. That’s an area we do very well in.”
To those familiar with the grandfather of commercial 3D printing, stereolithography (SL), Carbon might at first glance seem similar, in that a pool of photo-curable resin and a light source are used to build parts. The similarities end there however. There’s no bottom-up layer building with Carbon, nor lasers tracing part outlines and patiently filling in the spaces between. It relies on an entirely new technology called CLIP (continuous liquid interface production), which as its name implies builds parts continuously, from the top down, and is able to “digitally program” the material’s properties. And according DeSimone, it’s also up to 100 times faster than competing thermoplastic printing technologies, and far more accurate.
He adds that Carbon is working with BMW on customized external components, electrical connectors with Delphi, and a wide variety of plastic components such as automotive brackets, sensor housings, and grommets. The biggest challenge in each of these cases has been material validation. “To get a new material approved for a vehicle, you’re looking at about 12 months for the first one, after which it becomes much easier,” he says. “In the case of BMW, the parts have gone through multiple Bavarian winters and summers and we’ve proven that 3D printed resins are every bit as strong and weather resistant as traditional polycarbonate and ABS components.”
Next Steps
As an increasing number of 3D printed parts have gone into this “end-use” mode, additive manufacturing machine builders have begun placing great emphasis on understanding, visualizing, and documenting the entire process. Each of the providers mentioned here are actively monitoring each layer of each build—in Carbon’s case, every part is tagged with a code that identifies what machine, stage, window, and resin was used to build it, and can be traced back to the frame by frame manufacturing data.
The same can be said for the metal parts building processes, where the melt pool, laser power, optics performance, powder bed temperature and humidity, and related manufacturing factors are continuously recorded and analyzed. This provides complete traceability in the event of a product failure, and also promotes continuous improvement in this rapidly evolving industry.
To this end, machine tool builders and manufacturers are beginning to fully leverage Industry 4.0 and the IIoT (industrial internet of things), and 3D printing equipment suppliers are no exception. In many cases, machine metrics and build data are continuously fed to cloud-based servers for predictive maintenance purposes, and regular software updates pushed downstream to the end user.
As GE Additive’s Mook points out, it’s all about making intelligent decisions. “With 3D printing, you have to close the loop between the machine, the build parameters, the materials, and the part. Not only will this help users better manage their shop, but it will also increase throughput and improve quality. And if I can guarantee the quality of any given component at the completion of the build, that’s in turn a huge step for reducing costs, no matter what you’re making.”
“If you think about the motivation behind GE getting into the additive market, there were really two key things,” he says. “One, because we were early users of the technology and it disrupted some of our existing product manufacturing, we recognized its potential to disrupt all industries, not just aviation. And then, combined with what we perceived as a state of immaturity in the market, we ran the numbers and saw a 76 billion-dollar opportunity. So we’re going to build a business around it. We’re bringing our expertise and knowledge to the table, and as we expand beyond aerospace into automotive and other industries, we think 3D printing will ultimately make good sense for everyone.”