Isotropy in 3D Printing: Comparing FDM, SLA, and SLS Technologies
There are many aspects to consider when preparing an object for 3D printing. Among them is the orientation of the object on the build platform: Is there enough surface area to adhere sufficiently, are unsupported overhangs minimized, and is there enough clearance between objects?
On top of these considerations, there is another vital topic of concern if the part is intended to withstand any significant amounts of stress: anisotropy.
Anisotropy, in this context, is defined as having physical properties that vary with respect to direction. For example, a 3D printed object may have different elongations at break or stiffness in the X, Y, and Z directions. Isotropy vs. anisotropy is a concept discussed often in the 3D printing world, but in some cases the underlying assumptions are based on generalizations.
In this article, we’ll discuss whether parts created with different 3D printing technologies are isotropic, and why isotropy matters in real-life applications. After comparing the mechanical strengths of parts created by the three most common plastic 3D printing technologies, fused deposition modeling (FDM), stereolithography (SLA) and selective laser sintering (SLS), we’ll determine their levels of anisotropy and discuss the advantages of leveraging each technology against each other and traditional manufacturing methods.
What Does Isotropy Mean in 3D Printing?
Plastics, or the range of synthetic and semi-synthetic materials that use polymers as the main ingredient, are used in every industry and in almost every facet of day to day life. Within this material group however, there is enormous variation. Mechanical properties vary widely between different types of plastics, and, depending on manufacturing method, can even vary within the plastic part itself.
That variation in mechanical properties within a part itself is defined as anisotropy: having material properties that are variable in respect to the X, Y, or Z direction. Isotropic parts, in contrast, have the same mechanical properties regardless of direction of load or strain. Different manufacturing techniques produce parts with varying levels of anisotropy. A common misconception is that 3D printers produce only anisotropic parts. However, this is not true—3D printed parts can be isotropic or close to isotropic, depending on the 3D printing process, as we’ll see from the test results in this article.
Traditionally, plastic parts are most commonly molded, for example using injection molding, where molten plastic pellets fill and then harden inside a mold cavity, or thermoforming, where a plastic sheet is heated then pressed over a mold and formed into a new shape. In both of these scenarios, the end result is an isotropic plastic part, meaning that there is no material property variability with regards to direction (the parts are just as strong in the X and Y direction as they are in the Z direction). In addition to cost-effectiveness and repeatability, the isotropic nature of traditionally manufactured parts is highly desirable for plastics manufacturers.
According to testing however, SLA printers and to some degree SLS printers, may be looked on as alternative methods of producing isotropic parts. Next, we’ll take a look at the different 3D printing technologies and the ways in which they create anisotropic or isotropic parts.
Why FDM Prints Are Anisotropic
FDM printers melt filaments and extrude them in a line many times, creating a ‘solid’ plane made of these lines, then adding additional planes on top to build the part vertically. The most widely used type of 3D printing at the consumer level, FDM printing creates mechanical adhesion (not chemical) both within the horizontal plane and between each layer.
As is demonstrated in this experiment, FDM layer surfaces are not fully adhered to each other. Even when the previous layer is partially melted, the surrounding layers are only partially adhered to their neighboring layers.
As a result, the emerging FDM-printed objects have different mechanical properties based on the direction mechanical stress is impacting them and are less dense than a similar object produced via alternative methods, such as injection molding. This is also the reason why it is difficult to produce watertight objects via FDM printing: FDM prints are pervaded by microscopic voids and holes.
When viewing this arrangement at a molecular level, there is a clear distinction between the forces inside each layer and the forces that hold the layers together: Each deposited line of PLA or ABS is composed of highly entangled polymer chains which hold together tightly and are therefore pretty strong, tough, and stiff.
As additional lines are deposited to the side or on top of each line, it is very difficult, even impossible, to get the same amount of entanglement between the individual lines, thus producing junctions in the filament extrusion bonding regions that are significantly less strong and stiff.
This means that given a particular line deposition pattern, the part will be strongest in the direction of the deposited line and less strong along the axes that are primarily composed of these filament extrusion bonding regions, namely the two spatial axes orthogonal to the line axis.
In short, FDM parts are not equally strong in all directions, and cannot be isotropic, meaning that orientation matters when designing and printing load-bearing parts.
Why SLA Prints are Isotropic: Theory
In resin 3D printing, there is no difference between the chemical bonds that make up the individual layers and the forces that hold the layers together.
As each layer is formed, the resin monomers react and form covalent bonds providing high degrees of lateral strength, but the polymerization reaction is not driven to completion; rather, the print process is modulated in a way that keeps the layer in a semi-reacted state called the “green state.”
This green state differs from the completely cured state in one very important way: there are still polymerizable groups on the surface that subsequent layers can covalently bond to.
As the next layer is cured, the polymerization reaction will also include the groups on the previous layer, thus forming covalent bonds not just laterally, but also with the previous layer. This crosslinking is typical of all resin 3D printing processes.
That means that on a molecular level, there is little to no difference between the Z-axis and the XY plane in terms of chemical bonds; each continuous part printed on an SLA machine is a single molecule. Since the SLA lines are fully bonded to their neighbors, there are also no voids or microscopic cracks typical in FDM prints; these prints are watertight and fully dense.
For all intents and purposes, SLA parts prepared in this way are isotropic.
SLS 3D Printing: Slightly Isotropic, Material Variability
SLS 3D printing uses a high powered laser to build parts by sintering nanoparticles of polymer powder. The printer heats up a build chamber of powder to a point just below melting point, then a high powered laser traces the shape of a cross-section of the part file, heating the particles to just below or right at the melting point of the material and fusing them into a solid shape. The printer then deposits another thin layer of powder on top, and the platform drops down, exposing the fresh powder to be cured by the laser.
Within the same layer, particles are chemically bonded and mechanically bonded to each other, creating strength when stress is applied in the XY plane direction. With regards to the Z-axis, layers cannot fully cool down before the next layer is smoothed on top, and are thus partially fused to the powder layer above them. The bond is strengthened as that layer in turn reaches melting point. Though the parts are still strong, especially due to the materials’ inherent mechanical properties, SLS parts are not fully isotropic, and will have a variation in strength when stress is applied in different directions.
The most common materials, such as Nylon 12 Powder and Nylon 11 Powder are single-component powders, but SLS 3D printing also offers advanced composite materials, in which this variation is more prevalent. For example, Nylon 11 Carbon Fiber from Formlabs is a composite material with both nylon 11 particles as well as short carbon fibers. Because the roller in the Fuse 1+ 30W printer deposits the powder in one direction, the carbon fibers are laid out in a specific orientation in each layer. Users should be aware of this directional specificity, and orient their parts to align these carbon fibers with the direction that stress will be applied to the final part.
To summarize, SLS parts are not fully isotropic, and will have a variation in strength when stress is applied in different directions. However, due to the advanced mechanical properties of SLS materials, SLS is often still the best choice for load-bearing parts.
Why Does Isotropy Matter?
Improvements to hardware, software, and materials have made more applications possible with 3D printing, including functional testing of prototypes and manufacturing end-use parts. While 3D printing may have been used mostly for looks-like prototypes, fit tests, or display models a decade ago, strength, durability, and reliability are now integral.
Product designers, engineers, technicians, and manufacturers need to be able to rely on a 3D printed part’s performance in the real world, and therefore must know ahead of time if, and how, a part might break under a certain stress. In a manufacturing scenario, this means that development teams should consider material properties alongside isotropy and build in some extra safety to account for anisotropy depending on the printing process.
We decided to put isotropy to the test in a real life example: a 3D printed eye-hook, mimicking end-use parts that might be used to secure a boat line, secure a repelling rock-climber, or fasten a package to a pallet for shipping.
We printed the same eye-hook design on an FDM, SLA (Form 3+), and SLS (Fuse 1) printer, then tested it by attaching a large weight to the hook and recording the break point for each material. The FDM printed part broke at 50kg, at the specific point where two layers weren’t chemically bonded to one another, further solidifying our theory that FDM parts are anisotropic.
This knowledge can be applied in every 3D printing application when deciding how to choose which technology serves your needs best. When designing a part that requires load-bearing capacity from several directions, or will be under stress in several different planes of motion, SLA and SLS are best. When deciding a looks-like or fit-test prototype, FDM parts are perfectly sufficient.
Testing the Theory
Though printing a part and seeing how (or if) it fails under stress is a potential way of determining its isotropy, there are internationally agreed-upon testing procedures that can produce the same data, without any loss of property. We printed five ASTM D638 defined type IV tensile bars for each position (at 0°, 15°, 30°, 45°, 60°, 75° and 90° in relation to the build platform) on each of the three technologies—FDM, SLA, and SLS. Each bar was post-processed according to the guidelines of the specific process, and then clamped into an industry-standard tensile tester.
Each tensile specimen was tested in accordance with the ASTM D638 test method, and we took the average value from the five samples at each angle. The tensile tester stretches the bars at a constant speed until they break while recording how much force the bars exert in response. Based on this data, it’s possible to determine a host of material properties, most relevant among them the maximum strength measurement and elongation at break.
Maximum Stress (MPa)
Maximum stress is the highest amount of sustained pull that the bar can withstand before it breaks; this is usually what people mean when referring to a material being “strong.”
It’s easy to see how maximum stress could be used as a measure of anisotropy: a part with weak chemical bonds between layers would show significantly diminished strength when the tensile forces are exerted perpendicular to the XY plane.
In fact, that’s exactly what happened in our experiment on FDM prints where the Z-axis yield strength of a given part was approximately 55 percent of that in the X-axis.
With SLA 3D printing, the difference is less than 10%, supporting our theory that SLA parts are isotropic.
SLS 3D printing shows a slightly higher degree of variability than SLA, around 20%, but this means that the parts are still close to isotropic.
Elongation is the amount that a material will deform up until a breaking point, given as a percentage of the original measurement of the sample. The final length of the sample is compared to the original length to determine the percent elongation. This measurement is closely related to ductility, or the amount that a material can have its shape changed without fracturing. For instance, a rubber band is highly ductile, it can be stretched to multiple times its original length without breaking.
While the differences in elongation are bigger compared to the maximum stress, once again SLA and SLS showed the least variation. Both SLA and SLS parts showed variation generally within the expected random error range, while FDM parts showed a much more significant variation in elongation when printed parallel and perpendicular to the build platform. FDM parts printed at 90° exhibited an elongation of 1.8%, which means that the test bars broke almost immediately when force was applied, there was no yield before failure.
Part testing presented results aligned with the theory: FDM parts are anisotropic, with the greatest difference between parts printed at 0° and 90° in relation to the build platform, SLA parts are, for all intents and purposes, isotropic, and SLS parts are close to isotropic, with substantially more consistent strength across different orientations than with FDM.