Insights from NASA’s Ryan Watkins on how additive manufacturing is shaping the Mars Sample Return mission.
If you’re a nerd, as I must admit I am, there’s a particular feeling that comes with encountering your newest fixation. It’s a bit like being Alice on the edge of the rabbit hole: the irresistible pull of the unknown that demands to be grokked.
I felt it when I discovered dinosaurs as a child, science fiction as a teenager, and philosophy as a young adult. These days, as an elder millennial (emphasis on the qualifier) I don’t often get that same invigorating thirst for knowledge, except vicariously through my kids.
One notable exception was a presentation at this year’s AMUG Conference by Ryan Watkins, a research engineer at NASA’s Jet Propulsion Laboratory (JPL). His keynote, titled Built to Break, Designed to Protect, outlined his efforts to develop and implement crushable lattices for the Mars Sample Return mission.
The combination of the mission’s ambitious aims, the huge engineering challenges involved, and Watkins’ proposed solution to one particular problem made the keynote utterly captivating and undoubtedly the highlight of my first AMUG.
Additive manufacturing for Mars
Not surprisingly, 3D printing technology has already found its way to Mars in the form of nearly half a dozen titanium and Inconel parts on the Perseverance rover. (It didn’t come up in Watkins’ presentation, but this also means that the ratio of 3D printed parts to conventionally manufactured parts is orders of magnitude higher on Mars than it is on Earth.) Indeed, JPL isn’t just a user of additive manufacturing (AM); it’s a trend-setter.
“We had a lead contributing author for NASA Standard 6030 and also the part qualification handbook that came out in 2020,” Watkins said in his presentation. “In 2021, we took that NASA standard and fully implemented it to qualify our two main 3D printers. There’s a momentum that I’m starting to see where there is a high probability that a lot of our future missions are going to include more than a few metal 3D printed parts.”
As a research engineer, Watkins is charged with using advanced design and manufacturing technology to solve NASA’s most challenging problems. The Mars Sample Return has many, but the one he discussed involves returning samples taken from Mars to Earth in a capsule without any parachutes or rocket thrusters to slow its descent. In other words: How do you ensure the Mars sample container will survive a terminal-velocity impact?
The answer, of course, is to use lattices
Making lattices
As Watkins pointed out, lattices aren’t only common in 3D printing, they’re common in nature as well, in the form of animal and plant tissues. “What makes lattices unique,” he explained, “is the fact that they’re built upon very small building blocks called unit cells, which get repeated hundreds of thousands of times to build much larger structures.”
Beyond 3D printing, engineers use lattices for lightweighting, thermal management, and energy absorption – the last application being the focus of the research Watkins was presenting. Despite this broad range of applications, artificial lattices typically appear in two forms: foams and honeycombs. Not only that, but they’re also generally only made from aluminum and tend to be geometrically limited to either sheets of honeycombs or blocks of foam. However, with 3D printing, the possibility space for manufacturing lattices has grown significantly for both unit cells and overall geometry.
“There are significant developments that go into manufacturing these lattices,” said Watkins. “I think sometimes what’s forgotten in this context is how hard it is to design these structures.”
Designing lattices
After walking through how lattices behave under pressure, Watkins highlighted the two most important factors for designing crushable lattices to absorb energy: relative density and length-scale separation. The former is pretty much exactly what it sounds like, i.e., the solid volume divided by the total volume. “It’s essentially the inverse of porosity,” Watkins said. The second factor refers to the ratio between the size of a lattice’s unit cells and the overall size of the structure.
According to Watkins, current metal AM technologies still struggle to achieve good length-scale separation at low relative densities due to the minimum feature size of most metal 3D printers. While the JPL engineers tried to approach these issues by tuning their 3D printing parameters, “A lot of that felt like chasing your tail,” Watkins said. Their efforts to improve surface finish simultaneously resulted in more internal defects, and vice versa.
In the end, the best results the team achieved involved using chemical etching for post-processing to optimize material properties and relative densities.
“We’ve generalized this technique to a wide range of unit cells and materials: 6061 aluminum, Ti-64 and we’ve even been printing some more exotic materials like nickel-titanium with other partner centers at NASA,” Watkins said. “Today, we can comfortably get down to a relative density on the order of 2%, which is really what we want on the manufacturing side.”
A new tool for lattice design
As Watkins explained it, the challenges of designing and manufacturing lattices come down to the same issue: the length-scale separation. To illustrate why, he cited an unwritten rule in finite element analysis (FEA) – that your finite element mesh ought to fit three discrete elements through the minimum feature size.
With a diamond unit cell, that amounts to roughly 800 elements for single unit cell or 800,000 for a 10x10x10 lattice. For a gyroid unit cell, the number jumps to 100,000 elements; 100,000,000 elements for a 10x10x10 lattice.
“I don’t know how many finite element people we have in the audience,” Watkins joked, “but you kind of need a PhD in doing finite element analysis to solve a problem that big, and that’s only a thousand unit cells in the lattice. Our Mars Sample Return lattice has twenty thousand unit cells in it, so we’re now at sixteen million elements for the diamond and two billion elements for the gyroid.”
The problem becomes even more daunting when you consider that these are just two potential unit cell configurations. There are many possible unit cell designs and variations on them in terms of relative density and aspect ratio. This is what led Watkins to create a new, open-source design tool called UnitcellHub, which consists of three modules:
- UnitCellEngine: The “workhorse” that handles geometric modeling, meshing and FEA simulation.
- UnitCellDB: A database of pre-computed lattice properties including over 15,000 unit cell point designs.
- UnitCellApp: A GUI that takes inputs from the other two modules to make it easier for engineers to select the right lattice structure for their application.
Watkins gave a short demo of the tool during his presentation, but you can try it for yourself without downloading anything by visiting UnitcellApp.org.
As part of the demonstration, he explained how he used this tool to select the best lattice design for the Mars Sample Return mission.
“The key metric here is that the tool predicts when a lattice first yields,” he said. “Most lattices crush due to yielding, so we used this as a surrogate to downselect the design space. I plotted yield strength versus relative density for a titanium lattice – and what we found was that only about three lattices in our database met our crush strength requirement, which was around 2.5 MPa. Instead of randomly searching the literature and printing hundreds of coupons to find one that worked, I used this tool to select two lattices, did two tests, and was about 90% of the way to the final design we needed.”
To Mars and back again
There’s still a long way to go in terms of the design and manufacturing work necessary to get samples from the Red Planet back here intact. Watkins’ presentation slides included some fine print noting that, “NASA is currently considering alternative architectures for Mars Sample Return. This content reflects the planned program as of December 2023.”
Nevertheless, this kind of research demonstrates the value and versatility that 3D printing can contribute to some of the toughest engineering challenges ever conceived. Whether lattices designed with UnitCellHub find their way to Mars or not is still an open question but, after watching Watkins’ presentation, I sincerely hope that the tool he built at least finds its way into the hands of additive manufacturing engineers here on Earth.
