By Terry Wohlers, President, Wohlers Associates
The "Wohlers" column is authored by Terry Wohlers for Time
This column was published in the September/October 2010 issue.
In June 2010, I spoke to about 600 engineering and manufacturing professionals in Melbourne, Adelaide, Sydney, and Brisbane about additive manufacturing (AM) for part production. The presentations were a part of the “Factory of the Future” conferences and workshops organized by Formero (formero.com.au) and supported by the federal and state governments of Australia. Over the course of these events, I answered many questions, and here’s a recap of the most frequently asked questions.
They vary widely, depending on the AM process. Generally, they range from about $65 to $450 for 1 kg (2.2 lbs). Most AM plastics—both thermoplastics and photopolymers—are priced in the $175 to $250/kg range. For fused deposition modeling (FDM) from Stratasys, this translates to about $4.44/16.4 cm3 (1 in.3) for both model and support material. For machines from Z Corp., the material cost is $2.00 to $3.00/16.4 cm3 (1 in.3), which includes the powder, binder, infiltrant, and print head. While these prices are widely accepted for modeling and prototyping applications, for manufacturing quantities, however, they may be high. By comparison, injection molding plastics are priced from about $2.40 to $3.30/kg, meaning that AM plastics are 53 to 104 times more expensive. For injection molding, you need to also consider mold costs and lead times. Even so, I believe that prices for AM materials will decline as system manufacturers and material suppliers position their products as alternatives to molding parts. Also, as volume grows, competition is expected to stiffen.
Material prices for some of the most-established processes (stereolithography, laser sintering, and FDM) were set in the early 1990s and they have not changed much since then. A much wider range of materials are available today and some of them are more expensive, but prices of the most popular materials have changed little. To some extent, I believe the companies in the materials business are getting whatever prices the market will bear. Of course, the cost to produce and sell a unit of AM material is much higher than a unit of injection molding material simply due to the economy of scale. As Tom Mueller of Express Pattern stated, “The most popular AM materials are selling in the range of thousands of kilograms annually, while injection molding plastics are selling in the millions of kilograms per year.” As demand and volumes increase, prices for AM materials are likely to decline. Lower prices would help stimulate their use for production applications and that would, in turn, help push prices downward further.
Calculating the cost of AM parts can be quite involved. You need to consider all costs, including machine depreciation, maintenance, labor, material, and overhead. Including machine depreciation means that every part built includes a small fraction of the original cost of the machine. AM machines are typically depreciated over a period of 5 to 7 years. This cost can be significant if the machine purchase price was high. That’s why the cost of a part built on an inexpensive machine can be significantly lower than the same part build on an expensive machine, even if other costs are the same. At some companies, the cost of materials is about 15 to 20% of the cost of a part, although this can vary depending on the size of the part and the level of finish required. A high level of finish requires skilled labor and this portion of the overall cost can be much higher than the machine depreciation component. Material cost is also impacted by the shape of the part, which affects the amount of support material needed. Some machines use more support material than others. Also, waste comes from processes, such as laser sintering, which cannot recycle all unused material. In the case of laser sintering, about 30 to 40% of all loose powder in the build chamber after the parts are built becomes scrap.
Production numbers can range from one to thousands. The size of the part, especially the build height, influences the build speed and cost more than anything else. Loughborough University conducted a study to determine the break-even point of producing plastic parts by AM vs injection molding. It found that the breakeven point of a part measuring 77 x 48 x 32 mm (3.03 x 1.89 x 1.26 in.) is at around 7,700 pieces when using laser sintering. In other words, below 7,700 units of this part, use laser sintering; above it, use injection molding—if cost is the only criteria. Of course, many other factors besides cost enter into the decision process, including the physical properties of the part, surface finish, dimensional accuracy, design flexibility, and delivery. When manufacturing a part measuring 140 x 190 x 155 mm (5.5 x 7.5 x 6.1 in.), the breakeven point drops to about 180 pieces. These numbers can change dramatically from process to process. AM systems are becoming faster at building parts, so the numbers will improve in favor of additive manufacturing in the future.
Mueller of Express Pattern said that part complexity—the presence of features that increase the difficulty of manufacture and includes under-cuts, thin walls, disparate wall thickness, a very smooth surface finish, and tight tolerances—have a significant effect on the breakeven point. If there are part features that require a complex mold (e.g., slides or pulls), the breakeven point improves in favor of AM because of the additional tooling cost. However, he pointed out that there may not be a choice in some cases because accuracy or surface finish requirements rule out additive manufacturing as an option.
This varies widely. Machines that are among the most accurate are typically capable of holding a tolerance of about ± 0.125 mm/25 mm (0.005 in./in.). This does not mean that they cannot do better, but this is often what is quoted by service providers. Express Pattern found a standard deviation of just over ± 0.125 mm for any dimension that fits within a build volume of 500 x 500 x 585 mm (19.7 x 19.7 x 23 in.), regardless of length.
Plastics-based AM systems are not necessarily lagging behind. In fact, they have had more than a decade head start. The difference is in the way they process materials. Most AM machines that build metal parts melt the materials and try to achieve 100% density. While not all are fully dense, the result is usually a part that is relatively close to a machined or cast part. The thermoplastic-based AM systems, such as laser sintering and FDM, usually do not build fully dense parts because the material is not completely melted as it is in injection molding. Consequently, their mechanical properties differ from injection molded parts. The bond between layers can also impact the strength of an AM part. However, this is not to suggest that plastic parts from AM systems are not strong enough for many part production applications; it only says that they do not match the properties of the same material that is injection molded. TC
Thanks to Simon Marriott of Formero for this initiative. Also, thanks to Tom Mueller of Express Pattern and Michael Siemer of Mydea Technologies Corp. for their helpful input to this column.