Design for Additive Manufacturing (DfAM) is the process of adjusting your design to make it cheaper, faster, or more effective to manufacture. Executing proper design for additive manufacturing techniques can increase yield and save time and costs when building your manufacturing process by ensuring that your 3D printers are used effectively.
In the last few years, advancement in 3D printing technology has created new opportunities in the design for the additive manufacturing process. For the first time, geometrically complex parts can be as easy to create as simple parts. More importantly, these parts can be designed with functional requirements in mind.
There are several key areas where design for additive manufacturing can prove transformational to the modern manufacturing and development processes.
Design for Additive Manufacturing vs Design for Traditional Manufacturing
Traditionally, when designing a part for CNC milling, you need to consider your mill’s make and model, capabilities, and working volume. These considerations are all machine-driven, while high-tolerance features, tool changes, and setups are all part driven.
Beyond machine- versus part-driven considerations, there are details like spindle speed, materials, and tool type that further impact the manufacturing process with CNC milling. If you have a complex part, it's easier to make on a more complex machine, but those machines are more expensive.
Design for CNC milling requires upfront consideration of every operation required to produce a part, as well as the associated tool needed to perform that operation.
Designing for 3D printing shares some similarities to traditional manufacturing. Some aspects of 3D printing design are process dependent and some are printer dependent. Since 3D printing is an additive versus a subtractive process, cross-sections of parts are extruded on top of one another, layer by layer, to build up your model. The number of operations and required tools is simplified to just one or two, which eliminates operational considerations like tool changes, tool clearance, setups, and custom workholding that are typically needed before starting a CNC job.
For example, a part produced with CNC milling, consisting of five distinct operations, using three different cutting tools, two setups, and a set of custom soft jaws could be produced with a 3D printer in a single operation, using one tool, and no additional setup. The same simplification also applies to parts requiring 20 operations, eight tools, and four setups. This opens up a wide range of design opportunities but also comes with its limitations that should be accounted for in your design.
One of additive manufacturing’s greatest advantages is that a complex part is just as simple to set up as a basic one.
Let’s take two different part designs as an example — Design #1 is a simple part with a vertical hole and Design #2 features an angled hole, which is not as simple to manufacture.
Design #1 would require a simple machining setup if you were milling. Design #2 would need either a more complex machine or a more involved fixturing setup. Two completely different machining approaches for a slight difference in design.
With 3D printing, you don’t need two different approaches. You can send both parts to the 3D printing software and press print. The printer does all the setup for you, so a geometrically complex part takes the same amount of time and effort to set up as a simple one.
Isolating Geometrically Complex Features in DfAM
One of the drawbacks of a deposition-based plastic printing process is that the parts are anisotropic, so the material strength will be different along planes parallel to the print bed and along the axis that's normal to it. Think of it like a stack of post-its. It's hard to break across the surfaces, but it's easy to pull apart at the seams between the discreet slices of material.
So, with design for additive manufacturing, it's important to put thought not into just a part's printability, but its performance and how it meets your functional requirements.
We can sum this up using a simple tetrahedron shape as an example. The first iteration of this is a basic kind of blocky shape. It works, but it's basically like a block CAD model. It's easy to get to this stage, call it good, and hit print. This would be a nine-hour print and cost $12.63 USD.
If we expect to go through multiple copies or revisions of this model, we may want to make some improvements to save time and cost. We can reduce the print time by cutting out a lot of the material in the center but maintaining the structural integrity of the part with ribs instead of a solid block.
That version of the part would take six hours and cost $6.12 USD to print, but from a structural standpoint, this part is anisotropic. When we apply a load to the part it can shear along the layer lines. So, we need to rethink the requirements of this part.
We care about the strength since this may need to endure a large load and we also care about the angles required to make it a regular tetrahedron. Those angles are complex geometries that need to be precise, and the part can't break on these beams. If we want to rapidly iterate on this version of the part and modify aspects of the design, we have to do so in cycles of six hours or more. This is where we can really use 3D printing to our advantage.
With 3D printing, you can isolate the critical complex components, which in this case are the corners. We circumvented the anisotropic elements of the part with dowels while conserving the overall geometry. If anything needs to be changed about the part, each corner unit is a half-hour print job and costs $0.50 USD, so you can iterate much faster on each joint if you need to, and if you'd like to change the size of the part all you have to do is swap out the dowel pins without reprinting the corners.
Here, 3D printing is perfect for this design because we've isolated the geometrically complex features. This is the key to designing for additive manufacturing. Identify what aspects of the design can lend itself to the layer-by-layer process. This affects costs and print time, improves workflow and part functionality, and makes it easier to iterate and modify.