How to overcome design-for-additive-manufacturing (DfAM) constraints that can be deterrents to mass production.
Benny Buller, founder and CEO, VELO3D
The manufacturing world’s mindset is shifting with the news of 3D-printed, certified-for-flight components and FDA-approved human joint replacements: additive manufacturing has shown signs it can produce end-use parts, but making metal additive manufacturing (AM) dependable enough for serial manufacturing remains an accomplishment of few companies.
Why is AM still falling short? After all it offers the promise of less tooling, shorter lead times, and fewer supply chain issues. Not to mention the enticing vision of consolidating multiple parts, and even the magic of “innovative designs that can’t be manufactured any other way.” The short answer is, while the technology has dramatically advanced in recent years, there are still many less-visible technology issues to be overcome.
One of these is the complex reality of shifting production from conventional methods to AM. A manufacturer will encounter this as soon as they start redesigning an existing part to ready it for 3D printing. Beware—this introduces significant, additional risk because it adds a host of new variables.
For one thing, you’re not comparing apples to apples. When you test old versus new manufactured parts and you see a difference, is that because of the design changes you made or due to the manufacturing technology itself? Or maybe a little bit of both? And when you begin to see financial benefits from the switch, will you be able to separate the benefits of improved manufacturing technology versus design changes that reduced the cost? Maybe not all of those benefits are unique to AM—will you be able to fend off those people who are skeptical of your math?
Another point is that redesign might not even be a viable option for various reasons:
- Does your business have the regulatory authorization to redesign the part (Design Authority)?
- Does your part have enough volume to justify a redesign and have you accurately captured all the design considerations that will make the part equivalent in operation to the original one?
- How many candidate-parts do you have and how much time will it take to redesign them in a way that truly makes an impact?
- What about the resources needed to validate your redesigns?
Finally, there’s just the sheer level of complexity of AM technology, particularly when it involves metals. Rey Chu, Co-founder of Phoenix, Analysis, and Design Technologies (PADT) says: “There are no guarantees of successful outcomes when printing with metals. There are so many variables coming together from the alloy, the machine architecture, the process parameters—they all compound into widespread complexity. Because of the nature of machine dependency, try getting the same result by printing the same part on a different machine requires significant effort in process development, validation and control ….it’s all hard.”
Is it possible to avoid or reduce this complexity? Is full up redesign required for most parts? In fact, lots of parts can be converted to AM now. They haven’t been in the past because they didn’t fit the rules of DfAM so they were being fabricated another way. Parts like impellers, heat exchangers, blisks, and volutes are excellent examples that have traditionally been manufactured outside of AM because of their geometries—low overhangs, large inner diameters, untenable aspect ratios, and complicated internal channels that make post-processing difficult or even impossible.
However recent developments in AM technology—most notably support-free manufacturing—are challenging the traditional mindset, providing high levels of quality control, and overcoming previous barriers such as those in the part-geometries above.
A little more about the implications of “support free” here: In AM, a self-supporting angle describes the feature’s angle relative to the build plate. The lower the angle, the less likely it is to support itself. Each material will perform slightly differently, but the general rule of thumb has been to avoid designing a self-supporting feature that is less than 45 degrees. As you can see in the picture below, as the angle decreases, the feature’s downward-facing surface becomes rougher and eventually the part will fail if the angle is reduced too far.
Support-free AM overcomes what were previously thought of as written-in-stone laws of physics, allowing you to accelerate your move to 3D printing without having to adhere to all the old rules of DfAM.
As Robert Smith, COO of Optisys, a leading aerospace/defense antenna-system OEM, says, “I have been designing and printing with additive manufacturing for more than 15 years. All that time, I’ve had to think about whether or not the machine tool can even access the support structures; if it can’t, then the design isn’t compatible with AM and it gets set aside. The ability to manufacture without consideration of support structures is transformative.”
Such a transformation will free you from being bogged down by outdated DfAM issues, allow your knowledge to mature as you carefully explore all the possibilities of AM, and enable apples-to-apples comparisons of manufacturing methodologies. Now you can focus on those critical challenges mentioned before: qualification and quality control.
All of these questions need to be considered, particularly if you are setting up a manufacturing flow for mission-critical, metal-AM parts. Addressed properly they will enable faster part qualification and quality controlled production—and allow you to manage the move from conventional methods to AM in a way that will also fit your financial goals.