Nov 17, 2019 Want to write a piece for 3DHEALS Expert Corner? Email us: info@3dheals.com
Source An example of topology optimization: A bracket geometry is topology optimized, leaving only 50% of the material, which contributes the most to the stiffness.
Earlier we posted about Generative Design, the cutting edge assisted design process whereby a computer is given a set of criteria and creates a number of potential solutions in an evolutionary manner. There is often confusion between this process and Topology Optimization (for brevity I’ll be referring to it as TO) due to the visually similar nature of their results. This article will be discussing the latter; what it is, how it’s useful and how it can be used in medical applications.
What it is:
TO is a technique primarily used for lightweighting a design, whereby a designer or engineer defines a design space and TO software removes any part of that design space that is not contributing some defined percentage to the structural integrity of the product. TO does this by using FEA (Finite Element Analysis) to break a design into discrete pieces (polygons/vertices/edges – a mesh) and then analyze the forces applied to that mesh. Then it removes what is not needed creating organic and often surprising shapes.
For example; if I was designing a bracket, my design space would be the maximum dimensions of that bracket. I could then set a threshold by which to remove excess material; say 60%. Topology Optimization software will remove any material that does not contribute at least 60% of the overall strength to the object while keeping any hardpoints that I’ve defined as necessary in the design of the bracket.
Compared to Generative Design:
The biggest difference between this method and Generative Design is that the latter creates lots of designs in an evolutionary way whereas the former creates only one design that’s been optimized for structural integrity based on existing criteria. Topology Optimization is great when you have a set space and overall idea and just need the computer to make it as lightweight as possible.
By way of an example, if I wanted to design a drone that was optimized for weight, a specific type of material and a defined manufacturing process I would use Generative Design to come up with many different shapes for the end result. Then I would “evolve” the design by selecting some results to continue with, tweaking the algorithm and running another simulation.
If I then wanted to take that shape and create a bracket to hold a motor to the drone, then I would create a defined set of rules such as the position of the mounting screws and then use Topology Optimization to make that bracket as lightweight as possible.
In summary, all Generative Design uses Topology Optimization, but not the other way around – because TO only has one output. Topology Optimization is used when you have a shape that you want to make lighter – and don’t really care about much else. Generative Design is when you don’t know the shape you want and ask the computer to give you a lot of options, taking into consideration things like the desired material and manufacturing method.
Topology Optimization Design Considerations:
One thing to note is that when a model is produced using either of these methods, you cannot simply take the result and swap it out with a lattice structure. It is a common misconception that once you have optimized the topology of a model then you can just swap it out with a generated mesh from something like nTopology to take advantage of 3D printing’s strengths. However, a lattice structure will have a different structural capacity and a new simulation will have to be run on any latticework-based-design. Whether this simulation is TO, Generative or just straight FEA it is a requirement to ensure that the lattice structure can hold up to the same loads that a solid design can handle, especially after it’s been “topology optimized”.
Another consideration is that TO does not factor in construction methodology, unlike generative design. So in many cases, a topologically optimized model must be further massaged if it is to use a manufacturing method other than 3D printing, taking into account limitations of whatever process is chosen. Even if the design is meant to be 3D printed, there are always efficiencies to be gained in optimizing a design for additive manufacturing.
Finally, the raw mesh from topology optimization is often not smooth and is quite triangulated. Models can also benefit from additional care being given to the form of the mesh once the optimization process is complete.
Medical Applications:
Topology Optimization can be applied to the medical space quite effectively for the same reasons it’s used in aerospace and other high-performance industries: lightweighting. Should you be printing a replacement for a bone, group of bones or even a cast; topology optimization can be used to make those results as lightweight and comfortable for the patient as possible.
For example; in a cast design you could set the required “hardpoints” to immobilize the arm and then run a TO algorithm to remove any excess material that is not required for the mission. This will create a more lightweight, breathable cast that can then be tweaked to optimize it for manufacturing and application to the patient.
In another example, a complex bone replacement can be optimized to use as little material as possible. While something long and simple like a femur, one could use a more traditional metal, something like the femoral head or a hip replacement could be optimized using TO.
Examples:
Conclusion:
Topology optimization is a useful tool for lightweighting and differs from generative design in that it produces only a single output based on defined conditions; shaving away what does not contribute to the overall strength of the structure to give you the most lightweight product possible. It is easy to take advantage of and is available in many different CAD software packages including Onshape, Fusion 360, SolidWorks, Creo, etc. It is a valuable tool in a designer’s toolbox for creating advanced prosthetics and lightweight 3D printed devices.
About the Author:
Jordan Pelovitz
Jordan is a trained Industrial Designer and computer graphics artist. His skillset spans computer graphics, CAD & CAM and of course, 3D printing. In his professional career, he has designed a composite car chassis, consumer products, and even photorealistic architecture visualization. He is a “3D Engineer”.
Jordan is also very passionate about education and has taught a public class in CAD & 3D printing for the last 5 years. He has put on webinars and educational talks for 3DPrint.com, Formlabs and the Gnomon School. He was born on St. Croix in the USVI and currently resides in Massachusetts, USA. In his free time, he is into aviation, R/C planes, gaming, Virtual Reality and spending time with his wife and three cats. He is also 3DHEALS Boston Community Manager.
Related Articles:
Generative Design and 3D Printing
3D Bioprinting Bone – One Defect At A Time
3D Bioprinting: Chiasm of Art, Design, Science, Technology, Evolution
Does both the technique use same equations and/or algorithm ???