Editor's note: A previous version of this story incorrectly stated that there are no standards for additively manufactured metal parts. The intended meaning was that there are no standards specifically for fatigue testing additively manufactured metal parts. This post has been updated to reflect the change.

In the near future, critical aerospace and automotive production parts may indeed come from a 3D printer. The technology’s potential is huge, agrees ASTM Fellow Steven Daniewicz. “But if we go too fast and we don’t know exactly what we’re doing, and there are failures associated with additive parts, we could deflate this balloon pretty quickly.”

Daniewicz says he and other fatigue testing experts remember the cautionary tale of powder metallurgy in the 1980s.

“We got excited about powder metallurgy like we’re excited now about additive,” he says. “And the Air Force—like a lot of people—rushed to get a part in an aircraft using powder metallurgy. There was a failure and a jet crashed. That industry took a nose dive, no pun intended. I’m not sure they ever really bounced back.”

NASA provided plausible explanation for the failure in a 1984 report, writing that the jet’s turbine disk “contained a large undetected flaw which propagated due to low cycle fatigue until it became critical and fracture occurred.” NASA reported that shortly after the crash, “The production of as-HIP powder metallurgy superalloys decreased dramatically.”

Fully understanding the material properties of additively manufactured metals, their resistance to fatigue and failure, and developing standards to match will be critical to prevent another costly mistake. In November, ASTM hosted its second Symposium on Fatigue and Fracture of Additive Manufactured Materials and Components. More sessions are planned in the future.

“There’s an enormous amount of work that needs to be done,” Daniewicz says. “The deeper we dig, the more we realize there’s a long way to go to ensure structural integrity. I think we’ve got a long way to go with some very exciting technology.”

To learn more about the future of fatigue testing additively manufactured parts, Quality spoke at length with Daniewicz, a professor in the University of Alabama’s department of mechanical engineering, along with John Tartaglia, engineering manager and senior metallurgical engineer at Element Materials Technology.

Quality Magazine: There are currently no international standards for fatigue testing additively manufactured (AM) metal parts, correct?

Steven Daniewicz: Yes, that’s correct. So what everyone is trying to do, including ASTM, is see how we have to modify what we currently have and take into account all these things that are different with an additive part from a conventional part.

QM: Which parts will need to be fatigue tested?

John Tartaglia: Everything that is 3D printed that sees any sort of cyclic load is going to require fatigue testing. Machine parts or castings that are going to be used in any kind of moving piece of machinery, whether it’s a car, airplane, rail, or whatever will all require fatigue testing, just like non 3D-printed parts require fatigue testing. If the thing is going to be static—in other words, if it’s just going to sit and support its own load, or just support some other static load, especially in compression, and especially not being subjected to the environmental fatigue forces like wind—then it’s not going to require fatigue testing.

QM: How much will the standards need to change, or how much will the AM parts need to change to meet the standards?

Daniewicz: Well, I think that’s part of the problem. The 3D printed parts are a lot different than conventional metal parts in terms of what’s going on with their metallurgy, which is what you see when you look at a piece of metal under a microscope. What may appear to you macroscopically—with your naked eye—to be the same, looks a whole lot different under a microscope. When I look at a part that was printed additively under some magnification, look inside that metal, I see a lot of differences between it and the conventionally machined part, which has a big impact on the fatigue behaviors. So the new standards, or the modified standards, have to somehow incorporate these vast differences that aren’t necessarily apparent on a macroscopic scale.

QM: How well do current AM parts stand up to traditional fatigue testing methods?

Daniewicz: As a rule, the additive parts are weaker. The additive part is a bunch of layers, and each of those layers is a tiny little weld, and it takes hundreds or thousands of layers over and over and over again to make a part. And so if you were to lay all those little welds end-to-end, you’ve got miles of weld in every part that’s made. So what you basically have is a very high probability of some defects. Therefore you’re going to have worse fatigue strength. There are too many opportunities to leave defects in the part because there are so many layers being put down to make that part.

Will different fatigue testing methods be needed for AM metal parts?

Tartaglia: No, the same methods will be used, in general. But how to specify parts is going to be an important thing, as you can see from the Symposium, how to set process and design parameters on fatigue and fracture behavior. So they’re especially interested in how you improve the fatigue performance of these materials which everybody knows are going to have some problems.

QM:  Now, high-speed, low-cost nondestructive evaluation techniques for additive manufacturing, why is that important?

Tartaglia: Well, the whole idea with additive manufacturing is you don’t do a lot of development. You’ve got a computer, and you’ve got a printer, and you want to be able to say, “OK, print this part, just make it.” And you don’t want to spend a lot of money on development, but you do want to have something just as a quick and high-speed, low-cost. Meaning, give me an X-ray of this thing that’s going to say that it’s going to last through the product life cycle. And people are going to be looking specifically at how you specify standards for additive manufactured parts. Because this has opened up the world, you don’t have to have a large investment in technology other than having the computer and the printer. John Smith in his garage somewhere could potentially have the ability to produce these parts. And using standards means that when the part goes into critical components like aerospace or cars, it’s going to last.

QM: What's your answer for fast and cheap testing for these parts?

Tartaglia: Well, it’s going to have to be something that’s visual. That’s another problem, X-ray radiography machines are expensive, but they they’re high speed. Having an X-ray machine that is relatively low-cost and able to produce an X-ray that gives a reasonable facsimile of what the porosity content of the part is, that’s going to be critical. And having standards for those X-rays, to say, “OK, this passes and this fails,” that’s going to be important. And those have not been developed yet.

CT scanning is also important. It’s fairly high-speed, but it’s not low cost. X-ray radiography is high-speed and it is lower cost, though it’s still in the tens of thousands of dollars. It can see certain types of pores. Low-quality additively manufactured parts can have pores that standard X-ray radiography will be able to detect, but pores that cause fatigue cracks can be much smaller than what X-ray radiography can find. So developing those techniques is going to be important.

QM: If the industry takes its time, do AM metal parts have the potential that people often describe?

Daniewicz: Yes, absolutely, huge potential. You can just print things that you can’t potentially make any other way, the complex geometries that can’t be made with conventional machining operations. So there’s a huge potential if you’ve got all that flexibility and can make complicated things that you can’t normally make. And so you free the engineer to think outside the box, and all of that has been great stuff. We just have to make sure we have more than just a part that’s complicatedly shaped and looks good. It's got to have the structural integrity inside as well.