As 3D printing in metal and plastic becomes more mainstream there is an increasingly complicated alphabet soup of acronyms which describe the various processes. The differences between SLS, SLA, MJM and DLP can be daunting even to seasoned veterans, so we’d like to take this opportunity to sort them all out for you, which should help you to choose the best method for your next rapid prototyping or low volume production project.
As we previously did with the top 7 methods for forming plastics we’ll talk about the history of these processes, what they are best suited for and the relative advantages and disadvantages of each. Note: One term that you’re going to read a lot here is “additive manufacturing”. Simply put, this describes the making of a finished part by selectively adding layers of material in a controlled step-by-step fashion, usually via a 3D CAD/CAM drawing (Computer Aided Design / Computer Aided Manufacturing). There is one exception to this general ‘layering’ technique which we’ll discuss later in the article. This is compared to the more traditional machining work, which is a subtractive process (selectively removing material).
The history of all modern 3D prototyping technologies begins with Mr. Hideo Koyama, then of the Nagoya Municipal Industrial Research Institute in Japan. In 1981 Mr. Koyama developed a process for creating a solid object using a photoreactive polymer in combination with a UV light, where a cross-sectional model of the part is hardened, or developed, by exposure to the UV light . Famously Mr. Koyama, a scientific researcher and not a businessman, failed to correctly patent this process, leaving it open to other researchers to refine and commercialize the process.
One such was Mr. Charles Hull, who furthered this field by helping to create a standard stereolithography, or .stl, file format to be used for the necessary computer control of the automated process. Charles Hull did patent his innovations and later started the company 3D Systems, now the largest such company using this technology and one of the leaders in the field. It is also Mr. Hull who first used the terms 3D printing and rapid prototyping, thus ushering us all into a manufacturing revolution which is still undergoing remarkable growth and innovation.
Like many other 3D printing processes, SLA first requires that the part in question be modeled with 3D drawing software. Then the geometry of the part can be analyzed, divided into cross-sectional layers, and additional supports added as needed. Modern software is designed to add these supports automatically, but the product designer needs to monitor this carefully.
This idea about supports is an important one, for this technique and other 3D printing methods. Parts are designed to be developed on a build platform that moves up and down, along the vertical or “Z” axis. After the completion of each layer, the platform moves down a precise amount, equal to the thickness of a single layer. A recoater blade then deposits fresh material across the top surface and the process is repeated. This means that a certain mechanical stress is induced across the face of the part, in addition to the force of gravity which may be pulling down upon a newly-formed thin layer of material. The use of supporting struts is therefore critical to maintain the part geometry until it is finished and cured. The correct location of such supports may indeed require a modification to the initial design or the rotation of the part configuration on one of its axes in order to account for gravitational effects.
SLA printing has the advantage of being relatively fast. Parts can often be finished in one day or less, though part size is typically limited to 50 x 50 x 60 cm. This method is considered expensive, due to the cost of the photopolymer and the sophistication of the machines involved. The finished part is a solid plastic piece that can be machined or used as the master model for making a plastic injection molding die, for blow molding or other industrial processes. SLA is most often used for rapid prototyping and to test form, fit and function of new design ideas. The surface quality and precision of small features in SLA is considered excellent.
Selective Laser Sintering (SLS)
This method derives from the mid-1980s, under a university program sponsored by the US government. A private company called DTS was formed to make the necessary machines, which was eventually bought out by the aforementioned 3D Systems.
A word about “sintering”, because that’s going to come up from time to time. This is the joining of metals or plastics in powdered or granular form using heat and/or pressure. The powder is heated below the melting point. Therefore it is not liquified and thus the process is not welding. Welding and sintering produce differences in the microcrystalline structure of the material, but for our purposes both methods make a solid part that is mechanically strong, machinable, etc. Sintering, however, is not fully dense and this might be an important consideration depending on the intended application of the part.
SLS is closely related to Direct Metal Laser Sintering (DMLS), so they will be discussed together. Both are examples of so-called “powder bed” methods. As with SLA above, the part will be divided into thousands of cross-sectional slices via a computer program. This program is then fed into the machine that will control the application of a fiber laser to sinter the material. Powder is introduced into the machine chamber on top of a build platform. The laser then draws the first layer of the design, sintering the powder into a solid. The build platform then descends the vertical distance of one layer’s thickness, typically in the range of 30~50μ. A wiper blade passes over the platform, depositing another layer of powder and the process is repeated until a finished part emerges.
Unlike with SLA above, support structures are not needed here, since the part being made is at all times surrounded and supported by unsintered powder material. This powder will later be brushed away upon completion, revealing the finished part. In most cases, unused powder can be recycled so there is little waste. One disadvantage of SLS is that the part is not fully dense which may make it inappropriate for some applications. The processing time is relatively slow and therefore not suited for normal volume production, although that is being improved upon. There are also a limited number of metals that lend themselves to this process. The main advantage here is that, because the part requires no secondary support struts, it is free to be constructed in any way that the designer envisions. And that includes the making of shapes that simply cannot be produced any other way, including with CNC machine tools. Although relatively expensive per piece, there is no need for the making of hard tools or dies – just a computer drawing file is all that’s required, and a part can be made in a single day or less, making this technique (and others closely related to it) ideal for working prototypes, one-offs and solid models requiring complex geometries.
There are limitations in some surface features, so additional “subtractive” machine work is often necessary, for example in threading tapped holes. Also, the surface finish is rough and usually needs extra polishing, sandblasting, etc.
Fused Deposition Modeling (FDM)
Developed by S. Scott Crump in the late 1980s, this process was bought by Stratasys who had the exclusive right to this patented technology and the name by which it was described. That patent has since expired and many other players have now created a vibrant DIY community to develop new uses for 3D printers of this type, which has also consequently greatly reduced the cost.
This is the type of 3D printer that most laymen are now becoming familiar with. A spool of thermoforming plastic [hyperlink] or wire extrudes a supply of raw material to a dispensing nozzle. This nozzle will lay down successive layers of material onto a base platform, but unlike the other techniques mentioned above it is free to move in both the vertical and horizontal directions.
The advantages here are that the machines are getting smaller and cheaper, and many different kinds of thermoforming plastics can be used which reduces cost. Even more than one kind of material can be printed during a single build, which increases versatility. The number and type of materials that can be printed this way is increasing constantly and now includes biodegradable, starch-based plastics for environmentally sensitive applications and even cement-like substrate for large-scale construction projects.
This technique is still too slow for large production runs but is ideal for rapid prototyping and rapid production while remaining cost effective. The presence of the extruding nozzle is a physical obstruction that limits the resulting fine detail, but it can be ideal when simple geometries of medium precision are sufficient. Recently this type of printer was used to manufacture parts on the International Space Station [hyperlink] to test the effects of zero-gravity on this process and the resulting parts.
Selective Laser Melting (SLM)
Again a CAD drawing file is used to translate a three-dimensional object into a series of 2D layers that can be successively printed. Unlike with SLS, the granulated metal in this process is welded and not merely sintered, producing a part that is fully dense.
This process was invented in Germany in 1995. A high-powered laser is used, usually an ytterbium fiber laser. The metals that can be melted include cobalt chrome, tool steel, aluminum, stainless steel and titanium. All of the material must be finely and uniformly atomized, and the build takes place inside a sealed chamber filled with an inert gas such as argon. This is because powderized metals can be highly explosive if melted with a laser in normal atmospheric conditions.
This process is becoming more popular not only for rapid prototyping but also low volume production. It also lends itself very well to complex engineering designs which have hidden pockets, conformal cooling channels and other internal features which can decrease the weight while increasing strength for demanding applications in aerospace, automotive, medical and other fields.
Laminated Object Manufacturing (LOM)
Here a series of thin laminates are laid out on a build platform. The laminates can be paper, plastic sheet or metal foil. With each layer, a computer controlled laser or other cutting device traces out the pattern for that layer, cutting away that unwanted excess material. The platform then drops by the thickness of one layer, a new laminate is glued on top and the process continues. This stacking process makes a finished part which is less sophisticated than a SLS or SLM equivalent, but it is cheaper and does not require especially controlled working conditions. Also, if paper is used as the laminate the finished part will be similar to solid wood and can be worked accordingly.
Multi Jet Modeling (MJM)
Another process developed by 3D Systems, this is something of a hybrid 3D printing device. As with SLA, a thermosetting polymer is used. But instead of having the finished part emerge from a bath of liquid material, an array of inkjet nozzles moves horizontally across the platform depositing a thin film for the 2D cross-sectional layer. Polymerization quickly solidifies the plastic in that layer, then the platform descends by one thickness and the process repeated.
Digital Light Processing (DLP)
Another variation on the polymerization of a curable resin, this process is very similar to SLA printing. It cures the resin with a more conventional light source, but it also requires support structures and post-build curing. The process is generally faster and a more shallow reservoir of photoresin can be used which also saves on cost. Like with SLA, the finished part has excellent dimensional tolerances and surface finish.
An interesting derivation of this process is called CLIP (Continuous Liquid Interface Production). Here the part is pulled from the vat in a continuous motion – there are no layers, it is an uninterrupted process. As the part is withdrawn it crosses a light barrier that is programmed to alter its configuration to produce the requisite cross-sectional pattern on the plastic. More on that process here.
These processes continued to evolve and to be refined. There is still some industry confusion about the correct nomenclature for many of them, since some names or acronyms are proprietary and therefore competitors in the marketplace use a polyglot of names to describe essentially the same process.
Each of them has advantages in the right circumstances, so which one you choose will depend on the intended application, financial constraints, part geometry, volume and speed of manufacture. Remember to always consult with your project management specialist to choose the right application for you!