"Viewpoint" is a
column authored by Terry Wohlers for Time-Compression
This column was published in the September/October 2008 issue.
Methods of additive fabrication will have a
profound impact on the future of rapid product development and
Many of us know that additive fabrication (AF) will have an effect on the future of product development. The difficulty can be in identifying the methods that will have the greatest impact. The following are my top picks. They are listed at random.
Often, people ask which AF products and companies will survive and thrive in the future. I usually respond by naming categories of technologies, such as inkjet printing, that I believe will do well. I feel strongly that inkjet technology will drive much of the business for AF systems and materials in the future, especially at the low end and mid-range of the pricing spectrum.
I’m referring to both the jetting of binder onto a powder, as well as the deposition of photopolymer. The leading companies in this space are Z Corp. for binder-based color printing and Objet Geometrics for the printing of photopolymers. Other methods of 3D inkjet printing are likely to develop in the future.
FDM is not the fastest AF technology and it does not produce the most beautiful parts. However, the technology is fundamentally simple and it builds parts using strong thermoplastics. The FDM products from Stratasys offer a good price/performance ratio and the company’s sales figures prove it.
The large installation base and momentum of FDM, coupled with the strengths of the technology, will carry it over the next 10 years. The speed of FDM will need to improve to keep pace with new generation methods of AF.
LS for thermoplastics is a mature technology that is preferred by many companies. When surveying organizations that are applying AF to the direct manufacture of products, you will find that the majority are using LS. The process is relatively fast and you can nest many parts in the powder of a single build. Part strength and surface finish are sufficient for rigorous prototyping and manufacturing applications.
The demand for LS will likely grow over the next several years as the use of AF for manufacturing grows. Whole layer sintering processes, as they develop and mature, will have an adverse impact on laser sintering. However, this is not likely to occur for some time.
LS is fast, but sintering an entire layer at once is much faster. Also, the machine price and maintenance costs of whole layer sintering promise to be lower. Sintermask Technologies of Sweden (formerly Speed Part) was first to introduce a machine of this type. It uses a photocopying process to produce individual masks that represent the inverse of the layers being produced. Infrared radiation is projected through the masks to sinter a layer in about 10 seconds. The material is polyamide powder, which is similar to what is used in LS systems.
Loughborough University has developed a whole layer sintering approach called High Speed Sintering, although it is not commercially available. Using inkjet printing, the system deposits black ink onto the surface of polyamide powder. Infrared radiation sinters the areas that are dark, leaving the light powder unsintered.
In the second half of the next 10 years, these approaches could have a strong commercial impact. Why not over the next five years? History shows that it takes several years for new technology platforms and start-up companies to hit their stride. The more money and resources that are available, the faster it can happen, but it’s still measured in years.
This is also a whole layer approach, but photopolymers are used instead of powders. Consequently, the layer solidification is initiated by light, coupled with a method of controlling the light. Digital Light Processing (DLP) technology from Texas Instruments is the preferred method. DLP uses tiny mirrors arranged in a matrix on a semiconductor chip. Each mirror represents one or more pixels from the image that is being projected. Data projectors and some large televisions use DLP.
Envisiontec, with its Perfactory systems, has commercialized AF systems based on the DLP chip. The systems are less expensive to purchase and maintain than stereolithography (SL) and are faster because of the whole layer DLP approach. Fewer materials are available, but the list is expanding. The V-Flash system from 3D Systems also solidifies an entire layer at once using DLP.
First generation laser melting systems made parts, but they were not impressive. Today’s systems are capable of producing usable components that would be expensive, difficult or impossible to machine cast and/or weld. The parts are being used for prototyping, custom and short-run production, and series manufacturing. The machines from EOS are arguably producing the highest quality parts, as measured by surface finish, edge definition and feature detail.
Systems from MTT (formerly MCP Tooling Technologies) and Concept Laser are not far behind. Also, Electron Beam Melting from Arcam has improved considerably over the years and is said to operate more efficiently than a laser-based system, although the surface finish is not nearly as good. Materials of choice for these systems are titanium alloys for aerospace and medical applications and cobalt-chrome for dental crowns and bridges.
Improved materials for stereolithography will help to extend the life of the machine technology. SL is mature and used by many service providers, but it is relatively expensive. The cost is often justified when specific materials are needed.
A relatively new resin is the DMX-SL 100 from DSM Somos. The material offers up to 4x the impact resistance of other SL resins and up to twice the impact strength of FDM (ABS and polycarbonate) and some laser-sintered polyamides, according to DSM. Overall performance of the material is similar to laser-sintered polyamides.
Compared to thermoplastics, such as ABS and polyamides, the properties of photopolymers are unstable over time. This means that strength and other properties usually decline as the part ages. It is too early to know whether the DMX-SL chemistry will suffer this same problem. It is not a problem for most prototyping applications because the life of a prototype is relatively short. For part manufacturing, the rules are dramatically different and stability is critical.
New chemistries will help to propel other photopolymer-based systems into the future. Customers of Objet Geometries and Envisiontec will benefit from the new formulations.
Most AF systems perform well for making models, prototypes and patterns. They also perform sufficiently for the manufacture of parts in some industries such as custom furniture, home and personal accessories, and collectables such as action figures. For aerospace, automotive and medical applications, they often fall short. Much of the problem is with controlling the build process. Most AF machines lack feedback systems that provide a permanent record of the build process as it occurs. This information can be used to identify problems should a part fail to meet quality standards. Data derived from closed-loop systems can be used to monitor and adjust build parameters (such as laser power) as the machine is producing each voxel of a part. Without process controls, AF will be met with resistance in industries where failed products can cause personal injury or worse.
Software is being developed to dramatically speed the process of preparing data for manufacturing by additive fabrication. Examples are hearing aids and dental crowns and bridges. Without this software, it is impractical to manufacture in this way. 3Shape, Geomagic, and Materialise have created special software for hearing aid and dental applications.
Software is being developed for other rapid manufacturing applications. For example, FigurePrints is spending considerable resources on software that can handle World of Warcraft characters that the company accepts from game players for manufacturing. The problem is significant because it must handle the submission of thousands of models that lead to consumer products that are sold at a low cost. The software improves the resolution and identifies thin areas and makes them thicker so that the features of the character are of sufficient strength. What’s more, it allows the customer to change the pose of the character.
To make AF work for manufacturing, a great deal of effort, trial ‘n error, fine tuning and rewrites will occur to develop the front end of the process. The way in which it is handled could make the difference between success and failure.
Similar to the front end of the process, the back end requires considerable effort. Cleaning and finishing prototypes can be a challenge, but it’s nothing like the problem that companies face as they attempt to manufacture hundreds or thousands of parts with AF. With prototypes, a considerable margin is included in the price for post processing. A similar margin for part manufacturing with AF would make it cost prohibitive.
To make AF work for manufacturing, organizations must develop streamlined methods of removing excess material and finishing the parts. Removing support material from holes, slots and trapped areas normally require expensive labor, so these material removal problems must be resolved. Hand sanding is also time-consuming and expensive and could become a “show stopper” if the task cannot be automated. In some cases, the parts may require strengthening using one method or another, such as infiltration. Parts may also require paint or one or more clear coats. If the post-processing requires little human intervention, the use of AF for manufacturing has a reasonable chance of success.
Several methods of additive fabrication will have an effect on design and manufacturing. Determining the top ones can help organizations anticipate and plan for the future. The machines and materials are important to consider, as well as the three basic steps of additive fabrication— preprocessing, the build process and post-processing—to ensure efficiency and quality, especially for manufacturing applications. TCT
Industry consultant and analyst Terry Wohlers is principal consultant and president of Wohlers Associates, Inc. (Fort Collins, CO). Wohlers has provided consulting assistance to more than 150 organizations in 20 countries For more information, visit http://wohlersassociates.com.