Additive Manufacturing 101: Part IV

By Terry Wohlers, President, Wohlers Associates

The "Wohlers" column is authored by Terry Wohlers for Time Compression.
This column was published in the July/August 2010 issue.

[Editor’s note: This is the final article in the series on additive manufacturing processes that Mr. Wohlers started in the January/February 2010 issue of Time Compression.]

Additive manufacturing (AM) systems have been capable of building metal parts for more than a decade, but only recently have these systems begun to gain traction in the market place. Even so, relatively few systems have been sold compared to systems that produce parts in plastics or other materials. These systems have been available commercially for more than 20 years, so much less is understood about the metal systems in the market place.

The initial focus of these systems was on the production of tooling. As these systems developed, so did the capabilities of high-speed CNC machining. Companies got better at producing molds and dies quickly and at a lower cost using CNC technology. This made it difficult for the new AM systems to gain a foothold and compete with CNC machine tools and methods that were established and well understood. 

Surface finish is a consideration for tooling produced by AM. Startup costs are another major consideration. Consequently, the manufacturers and users of metal-based AM systems shifted much of their emphasis from molds and dies to the production of parts. As it turns out, that’s probably where they should have been in the first place. Complex parts are difficult and expensive to make using conventional methods of manufacturing, so this is an area where AM can be competitive and provide a reasonable return on investment.

Many names are used to refer to the commercially available systems that build parts in metal. The most common are “laser sintering,” “laser melting,” “selective laser melting,” and “direct metal deposition.” Other names, such as “direct metal laser sintering,” “electron beam melting,” and “laser engineered net shaping,” are specific to the company that offers the machine. “Laser sintering” (LS) is the most common term used to describe the process of using a laser to sinter a powder material, but it is most often associated with plastics. The ASTM International Committee F42 on Additive Manufacturing Technologies has defined and published the following definition for LS:

Production of objects from powdered materials using one or more lasers to selectively fuse or melt the particles at the surface, layer by layer, in an enclosed chamber. Discussion—Most LS machines partially or fully melt the materials they process. The word “sintering” is a historical term and a misnomer, as the process typically involves full or partial melting, as opposed to traditional powdered metal sintering using a mold and heat and/or pressure.

Metal-based AM systems typically melt the metal particles with the goal of achieving 100% dense parts. Some systems achieve this goal, while others reach 99+% density. Small, random voids are the reason they are not considered 100% dense. Some of these small voids are inherent in the powdered raw materials and are “bubbles” that do not burst in the additive manufacturing process but remain as inclusionary voids. 

The powder-bed systems usually work in an enclosed, heated, and vacuum-sealed chamber filled with a shielding gas such as argon. The argon gas reduces the oxygen concentration to a level that is safe to process reactive materials, such as titanium and aluminum. Also, it contributes to the integrity of the material and the performance of the finished parts. One of the processes, electron beam melting, uses a true vacuum instead of a shielding gas because of the need for a vacuum to have a properly functioning electron beam energy source.

The build process begins after the build chamber reaches operating temperature. For most laser-based processes, this is around 100ºC (212ºF), but for electron beam melting, it starts at around 700ºC (1,292ºF). After a very thin layer of powder is spread, a source of energy, usually a laser, strikes the surface, melting the powder in its path. The laser is driven by cross section data from the CAD or STL model. After a layer is complete, a new layer is spread and the process repeats based on the design data. Like most other AM processes, the parts are built from bottom to top. When finished, the parts are fully embedded in the powder.

The parts are usually built on a relatively thick metal plate that is later removed with the parts attached to them. Similar to the plastic LS processes, the powder underneath the parts being built serve as support. However, due to residual stresses and the need for heat sinks, solid anchors and support structures are often produced to secure the parts, reducing the likelihood of warping as they cool. The supports must be cut away from the build plate and removed from the parts. Removal may involve the use of a band saw, wire EDM, or manual tools, depending on the types of parts being made. Small parts, such as dental copings, can be removed with pliers, although remains from the support structures are usually ground away.

A wide range of materials are available for use with the powder-bed systems. Among them are tool steel, stainless steel, cobalt-chrome alloys, pure titanium and titanium alloys, aluminum, nickel-based super alloys (e.g., Inconel 718), and even gold. The specific set of materials can vary from one system to the next. 

Companies that offer powder-bed systems include Arcam (arcam.com), Concept Laser (concept-laser.de), EOS (eos.info), Ex One (exone.com), MTT (mtt-group.com.), and Phenix Systems (phenix-systems.com). All of them use a laser for the energy source, except for Arcam, which uses an electron beam. Arcam’s electron beam melting (EBM) is said to be several times more efficient than a laser because the energy is better absorbed by the material. Much of a laser beam’s energy is reflected by the powder particles, making it less efficient. The commercial systems available today show that lasers are capable of producing finer detail and a better surface finish than an electron beam. However, recent advances in improved software and multiple electron beams functioning simultaneously promise to reduce the difference in surface detail between laser and EBM parts

ProMetal (prometal-rct.com) also produces powder-bed systems, but they do not use a laser or electron beam to heat and melt the metal powder. Instead, these systems deposit a liquid binder through an inkjet print head onto the surface of the powder. After the last layer is produced, the parts are carefully removed from the powder bed and cleaned. The parts are then placed in a furnace where the binder is burned out and a second material, usually bronze, is infiltrated to bring the parts to near full density. The result is a part that consists of 60% base material, such as steel, and 40% bronze.

Another broad class of systems uses a powder deposition process. They typically operate by depositing metal powder in the path of a laser beam, layer by layer. Unlike powder-bed systems, they do not spread a layer of material across the build area. Consequently, the parts are not fully embedded in powder at the end of the build cycle. Support structures are built to support the parts, including overhanging features, as they are produced from bottom to top. The supports are later removed using methods similar to those described for parts built on powder-bed systems.

Accufusion (accufusion.com), Irepa Laser (irepa-laser.com), Optomec (optomec.com), and POM (pomgroup.com) are companies that offer powder deposition systems. These systems are suited not only for building parts, but also for the repair of damaged metal parts and tooling—a capability that sets them apart from the powder-bed systems. The systems often employ a robot arm to position the powder deposition and laser beam, so the build volume can be quite large.

Metal-based additive manufacturing systems are not inexpensive. The powder-bed systems range from about $149,000 for a small (70-mm diameter x 40-mm height) build volume to $780,000 for a machine that builds parts that fit inside a 250 x 250 x 350-mm build volume. Powder deposition systems are more expensive. They start at $350,000 for a 1500 x 1200 x 1300-mm system to more than $1-million for a 3962 x 2743 x 3657-mm system. Note that a system’s build volume is one of several factors that establishes the price.

Solidica (solidica.com) offers an AM system that uses ultrasonic consolidation (UC), which does not fit in either of the previous categories. The UC process uses aluminum alloy “tape” and a hybrid approach that combines ultrasonic welding and CNC milling to produce parts. The machine’s design is based on a CNC machine and combines part buildup, layer by layer, and machining in a single system.

The powder-bed systems are, by far, the most popular. Customers are using them to manufacture orthopedic implants for hip replacements and custom cranial implants. Over the past two years, they have been used increasingly to manufacture copings for dental crowns and bridges. In fact, an estimated 6,000 dental copings are manufactured daily, with as many as 1 million produced annually. 

The systems are being used increasingly for aerospace applications. At some point, they are expected to be used for production flight hardware, such low-pressure vanes, stator rings, and combustion components, which can be difficult and expensive to produce using conventional manufacturing methods. For now, the parts are used as prototypes for testing and other fit and functional applications. TC

Author’s Note: Thanks to Andy Christensen of Medical Modeling Inc. and Greg Morris of Morris Technologies for their input to this article.