Let’s pretend you know nothing about additive manufacturing (AM), more commonly referred to as 3D printing. Given that this industry-changing technology has been with us for more than three decades, that assumption is highly unlikely but is nonetheless the premise behind this “dummies” style article. If you’re already an expert on the topic, feel free to go do something more interesting, like 3D printing some cool parts or binge-watching the latest Netflix series.
For everyone else, let’s start with a very simplistic overview of additive manufacturing. Each of the seven AM technologies recognized by the American Society for Testing and Materials (ASTM) begins with a 3D CAD model of the desired workpiece. This file is digitally rendered like a loaf of bread into thousands or perhaps hundreds of thousands of paper-thin slices before being fed to the 3D printer.
Several of the most common technologies use a laser or LED light source to successively trace each layer’s profile and interior sections onto the surface of a resin vat or bed of metal or polymer powder, solidifying those areas. Once each layer is complete, additional material is pulled across the burgeoning workpiece and the process continues, over and over again from the bottom up until the part is finished.
There are also systems that use an extrusion head like a hot glue gun to build parts. Some spray metal powder or extrude thin wire into the path of a focused energy source (a laser or electron beam), thus depositing molten metal onto the work surface, while others selectively spray polymer binder onto a powder bed, creating a “green” part that must later be sintered in an oven. Other methods exist and additional details will follow, but that’s AM in a nutshell. Simple, right?
Birth of an Industry
As mentioned, AM has been with us a long time. Once limited to polymer printing, it has since expanded into engineering-grade ceramics, composite materials containing carbon fiber or aramid (Kevlar), and perhaps most notably, metals and their various alloys. We’ll discuss the 3D printing of polymers and other non-metallic materials in a future Additive Manufacturing Industry Report—the balance of this one, though, will focus on metal AM, the much younger (and for the time being, smaller yet fastest-growing) segment of what has become a multibillion-dollar market.
Hans Langer, founder of Electro Optical Systems (EOS) in Krailling, Germany, might argue the “much younger” point. In 1994—just eight years after stereolithography inventor Charles Hull founded 3D Systems—Langer’s company leveraged their expertise with polymer powder printing (also known as selective laser sintering, or SLS) to introduce the EOSINT M 160, a machine that he and many others consider to be the first metal 3D printer.
This machine used a blend of powdered metals such as nickel and bronze to print parts boasting mechanical properties similar to those made via metal-injection molding (MIM) technology. It was clearly a huge leap forward, but it would take another decade before EOS began selling 3D printers capable of creating “fully dense” metal parts, opening the door to increasingly widespread adoption throughout the aerospace, medical, transportation, and energy industries.
Langer and his team dubbed this early technology “direct metal laser sintering,” or DMLS, an acronym that is no longer wholly accurate. As noted, those early powder bed machines required bronze or a similar low-melting-temperature metal to act as a binder; in contrast, modern DMLS printers have sufficient laser power to melt or “fuse” even the most heat-resistant of materials, including titanium, Inconel, Hastelloy, and refractory metals like tungsten and niobium. This is why EOS has since swapped the DMLS verb “sintering” for “schmelzen” (German for “to melt”), a more accurate term and one that also allows them to retain their long-standing and trademarked acronym.
Perusing Powder Bed Fusion
Brief history lesson aside, EOS has plenty of competition from other 3D printer manufacturers, many with their own special acronyms. For instance, SLM Solutions Group AG has trademarked its namesake metal powder bed technology SLM, short for selective laser melting. Concept Laser, now part of GE Additive, has its LaserCUSING technology, 3D Systems offers DMP (direct metal printing), TRUMPF has developed laser metal fusion (LMF), and Velo3D offers its Sapphire system with its underlying Intelligent Fusion process.
All are examples of laser powder bed fusion (LPBF), a metal subset of the technology described by ASTM International as powder bed fusion (PBF), which also includes polymer printing. As its name implies, LPBF printers use a bed of metal powder, over which sits a laser or series of lasers that perform the task described at the outset—trace the outline and raster or otherwise fill the interior of each digital part layer. This creates a small “pool” of melted metal that fuses to the layer beneath, cools, and solidifies immediately once the laser has passed.
Closely related to LPBF is electron beam melting (EBM), which as you might guess, uses an electron beam in lieu of a laser to do the melting. As of this writing, only one major EBM provider exists—Arcam, now owned by GE Additive. Energy delivery method notwithstanding, the two processes are quite similar. Both are also much more complex than what’s been described so far, and this is where each 3D printer manufacturer begins to differentiate itself from the pack.
For instance, builders heat their metal powder beds to reduce the thermal shock that occurs when laser light or high-energy electrons strikes otherwise cold metal powder—depending on the material, printer, and part geometry, this temperature might be several hundred degrees C (around 600 °F) or more.
And because oxygen and humidity wreak all kinds of havoc to electron beams and molten melt pools, and also tend to react with metals such as titanium and aluminum, the environment within a metal 3D printer’s build chamber must be tightly controlled. In most cases, a vacuum is generated to pull air and other impurities from the chamber, after which it is filled with a precise amount of argon or similarly inert gas. Here again, each builder has its own unique approach, even though each must follow the same laws of physics.
Of Supports and Surfaces
Another important consideration is the method of introducing fresh raw material once each layer is complete. Some 3D printers use a hard metal or plastic blade to scrape a thin layer of powder across the burgeoning workpiece. Others use a roller, compacting the material as it goes, and some employ a stiff rubber squeegee approach or alternative non-contact recoater system. The challenge in all cases is to deliver a consistent and predictable amount of powder across a surface that is not necessarily flat, and may actually have small protrusions and sharp edges that can catch the blade as it passes, an unfortunate event known as “crashing the build.”
This last point leads us to the Holy Grail of laser powder bed fusion—managing the intense thermal stresses that cause individual part layers to warp or curl upwards during the build, stresses that require heat-treatment afterward to alleviate. As with certain types of polymer printing, the solution here is to anchor the part to the build plate (the removable fixture on which most metal parts are constructed) and to other parts or part sections using strategically placed scaffold-like structures. However, precise control of the laser output and atmosphere within the build chamber can help to reduce or in some cases eliminate the need for such structures.
Regardless, once the 3D-printed part has been cut away from the build plate using a high-precision saw or wire EDM machine, these “supports” must then be removed via CNC machining, manual or robotic grinding, or vibratory deburring methods. Here is also an opportunity to smooth the characteristically rough surface finishes generated by most AM processes, and to finish machine close-tolerance or application-critical part features. All of these are an integral part of the 3D printing process. For this reason among others, metal 3D printing and traditional machining—a.k.a subtractive manufacturing technologies—will continue to complement one another well into the foreseeable future.
There are more ways to skin the metal AM cat than LPBF and EBM. One of these is directed energy deposition (DED), a technology that is commonly used to repair parts such as turbine blades but is also quite suitable for building metal components from scratch. As with laser powder bed—and indeed all 3D printing technologies—numerous brand-specific iterations exist, among them laser engineered net shaping (LENS) from Optomec, TRUMPF’s laser metal deposition (LMD), and direct metal deposition (DMD) by Precision Optical Manufacturing (POM).
Many DED systems inject a stream of metal powder into the path of a high-power laser that is directed at an existing workpiece or onto a build plate. As the two merge, a melt pool forms on the surface, resulting in a net deposition of material within a well-defined region. A sealed build chamber filled with inert gas is required for reactive metals like titanium, otherwise a shielding gas that surrounds the beam and eliminates oxygen from the build area will be used on less reactive alloys such as stainless steel.
Other DED machine builders consume wire feedstock rather than metal powder. One of these is Sciaky’s EBAM, short for electron beam additive manufacturing. But there’s also Norsk Titanium with its rapid plasma deposition (RPD), Gefertec’s 3DMP (3D metal print), Lincoln Electric and WAAM (wire arc additive manufacturing), and numerous others. Broadly speaking, all fall under the DED umbrella, whether it’s wire-DED, DED-arc, or WAAM.
Ironically, this last is the earliest known form of metal AM. In 1925, inventor Ralph Baker of Wilkinsburg, Pa., together with his employer Westinghouse Electric, filed for a patent on Baker’s “Method of Making Decorative Articles.” It describes using arc welding “to produce receptacles or containers of ornamental and useful shapes,” a process virtually indistinguishable from that of today’s wire cladding, hard facing, and other well-known methods of building up part surfaces.
What’s changed from Baker’s era is the use of a CNC gantry or robotic motion control to drive the deposition head, allowing it to trace complex part geometries in three dimensions. This gives DED great flexibility, never mind its productivity. Some DED systems boast deposition rates of a dozen or so kilograms per hour in a broad range of high-performance alloys, copper, titanium, stainless steel and refractory metals, among them.
NASA, for example, is exploring the use of powder-based DED to construct exhaust nozzles many meters across. Similarly, Relativity Space has successfully 3D-printed equally massive fuel tanks and is said to have reduced the parts needed to construct a typical rocket from 100,000 to just 1000 components. Lincoln Electric and other DED providers are able to quickly produce large-scale tooling and machinery components that once required months to build. By outfitting five-axis machining centers with DED heads as machine builders DMG MORI, Okuma, and Mazak have done, so-called hybrid manufacturing of highly accurate and complex workpieces can be accomplished in a single operation.
Bound for Success
Powder can be applied in other ways as well. For instance, Markforged has developed the means to bind metal powder together with a wax-like material into long filament spools that look much like those found in weed whackers. Identical in operation to polymer-based fused filament fabrication (FFF), this metal-filled material is fed through a heated extrusion head and deposited onto the work surface below, building parts as it goes. When done, these are washed to remove most of the binding agent then sintered in a furnace until cured. They call their process metal FFF.
Desktop Metal has invented a similar technology. Its Studio System employs bound metal deposition (BMD), which as its name implies, relies on rods of pre-packaged bound metal powder that are extruded layer by layer to build the workpiece. Rather than a secondary wash operation, however, the parts go through a two-stage sintering process. Both systems are designed to simplify the printing process and eliminate the need for loose metal powders, allowing their use in even office environments.
Then there’s binder jetting. Here, a polymer-based binder is selectively sprayed over the surface of a metal powder bed, temporarily holding the particles together. Similar to LPBF systems, a layer of fresh powder is then pulled across the surface and the process repeats, ultimately producing a “green” part. As with other sintering-reliant AM systems, these relatively fragile constructions must be washed in solvent or cured with UV light before heading to the oven for the final fusing process, where they become fully dense.
There’s much more beyond sinter-based technologies. For instance, Fabrisonic’s ultrasonic additive manufacturing (UAM) uses a “horn” to project extreme high-frequency sound waves onto thin sheets of metal, bonding them together. Even dissimilar metals like titanium and aluminum can be joined to create metal sandwiches, and when combined with a CNC milling head, complex parts containing integrated electronics are possible. Then there’s Jason Jones of Hybrid Manufacturing Technologies, who has invented laser cladding and polymer extrusion heads that can be retrofitted to any CNC milling machine, multitasking lathe, or robotic arm. And machine tool builders Matsuura and Sodick have taken LPBF one step further by incorporating metal AM into certain models of vertical machining center, giving manufacturers the ability to print and machine plastic-injection molds with conformal cooling channels in a single operation.
The takeaway? Metal 3D printing is just getting started.
Design and Cost Considerations
No article on metal AM would be complete without a brief mention of design for additively manufactured parts, principles that many refer to as DfAM. In some ways, it’s the most important part of the conversation, especially where metal AM is concerned. That’s because all forms of 3D printing—metal, polymer, or otherwise—present designers with previously unavailable opportunities for stronger, lighter weight, and more effective products that have been optimized for the intended application.
Taking advantage of these capabilities, however, depends on two things—the designer’s skill, and the software she uses. Attaining the first of these requires education, experience, and no small amount of hard work. The good news is that 3D printer manufacturers and the AM community in general offer plenty of resources for each, never mind the fact that universities including Penn State have begun offering degrees in additive engineering. For a young person (or even one not so young) it’s a great time to be in manufacturing.
As for AM-related software, the CAD industry has done a good job of keeping up with and in some cases surpassing 3D printer development, and now offers products that go well beyond the slicing and build preparation tools alluded to earlier. Topology optimization, generative design, AM workflow management—these are just a few of the systems available to additive manufacturers large and small, all of which serve to make AM part designs more robust and the 3D printing process more efficient.
There are some very good reasons for all this. Additive manufacturing promises a host of benefits that extend well beyond the factory floor. As Velo3D CEO Benny Buller said recently, AM makes it possible to print locally and disrupt globally. This seemingly simple statement has profound implications. It means shorter supply chains, accelerated design and development cycles, reduced equipment downtime, and greater product efficiency. Perhaps most importantly, it means far less waste—in time, energy, and natural resources alike, all of which is good news for the planet and the people who depend on her.
Special thanks to additive manufacturing experts and solution providers 3D Systems, Nexa3D, Markforged, Optomec, Stratasys Direct Manufacturing, and Velo3D for their input and fact-checking during the writing of this article.