Defining Additive Manufacturing Standards
ASTM and ISO cover new ground.
At any given moment, there are dozens of new standards for additive manufacturing nearing completion, under advisement in working groups, or being considered for future committee sessions. It’s a race to both define the basics of the materials, processes and equipment, but also to organize a standards system that will stand the test of time in a fast-changing environment.
The work is being spearheaded by ASTM F42 Committee on Additive Manufacturing, working in cooperation with ISO TC 261. appliance DESIGN caught up with Carl Dekker, chairman of the F42 committee and president of Met-L-Flo Inc., to lend insight on how standards work hand in hand with a burgeoning field.
appliance DESIGN: Can standards drive a new industry and increase adoption of new technologies? Do they provide firmer footing for companies to move forward?
Carl Dekker: Standards can do an amazing amount for facilitating that growth. You look at organizations like Boeing or Lockheed Martin, they’ve got enough size to send people out to do the validation, qualification and confirmation that people are building things correctly. You start to get to some of your smaller manufacturers, and those that are building pieces for those large OEMs, and you now get to the point where people can say, “Manufacture it to that standard.” And there’s a greater confidence that they’re going to get what they’re expecting. You also have the ability to then look at the records and papers and documents and say, “You didn’t actually make your parts to this, because you’re missing these criteria.”
Now you have the potential to be able to disqualify some of those potential vendors who may not be meeting the quality standards. Just like an ISO certification or an AS9100, any of those. Now the amount of work needed to validate that somebody is in conformance with what you require for manufacturing becomes a lot easier because everybody’s already speaking the same language.
AD: How difficult is it to develop standards when the ground is always shifting? Is it tough to write standards that don’t become irrelevant if a technology quickly changes?
Dekker: That is a great question. It depends on how deep you’re going with the standard. For example, a standard defining the part’s orientation in the machine, with that one you could change the process and you’re probably going to be pretty much the same. The deeper you get into it though, it’s much more critical. If you’re building parts for the SpaceX rocket that’s going up with the Tesla, and then come back down, they want to be able to use these parts again and again. You’re going to have to have data that validates exactly what you’re doing—what machine did you build it on, what powder did you build it with, how was that powder manufactured? Was the operator trained to use it? Was the software the same version as the one that was qualified before? So you get to all these variables that come into the manufacturing process, and software upgrades in this are practically faster than Windows updates. You get to the point of, “How do you validate that the software update didn’t change some of your settings or variables within a qualified process?”
So there are still a lot of variables we don’t know about with the rate of technology advancement and change. We haven’t even gotten to the depth of what’s happening with the system before we’ve moved to the next level.
AD: The writing of standards is never done, but is there a short-term finish line for establishing a usable base of AM standards?
Dekker: The first answer is that there isn’t one. But I think the objective—we did it with one standard for mechanical parts using an electron beam into a powder bed. That was an example of at least one area where you’re able to manufacture a product with those standards. Now how do we get to the point where more products can be manufactured? So do we need to define what the nondestructive inspection criteria are? Do we need to define the evaluation and finished fabrication? Do we need to get the post-processing or the secondary treatments defined? Once we start to get a couple of these examples in different areas, I think we’ll have kind of that base line of, “OK, these guys have done it. Here’re their applications, what are the things that still apply over here?” You start opening the doorway to make it a lot easier to create or modify standards.
AD: How much awareness is there from the makers of 3D printing technology of the standards process?
Dekker: In 2008 we lobbied ASTM to start F42. In 2009 they started it, so it’s been around for almost a decade now. And yeah, we’ve got over 500 members; we took probably about four years before the term additive manufacturing was acknowledged as the industrial approach. And now another five years after that we’re starting to see some awareness from the major manufacturers on it. But there are still quite a lot of people out there who don’t have an extreme depth of knowledge of what’s going on, where it’s happening, and what it’s doing. There’s still a lot of misperception out there.
It’s amazing how many people are jumping into these smaller organizations that don’t have much information on where the history of the industry has been. But it’s going to take time. It’s unfortunate, you bring on a new technology and people only get so much time to learn new things.
AD: What’s the major push going forward with standards?
Dekker: NADCAP (National Aerospace and Defense Contractors Accreditation Program) has created a checklist for additive manufacturing, so that checklist getting evolved, or people being able to get certified to that checklist and having standards to be able to call out and manufacture to that checklist will be a big thing. Having standards that OEMs can reference for how to manufacture parts and define the testing criteria for them and know that you’re going to get the right part when you’re done is huge. So, as we see standards for testing and evaluation of these pieces—documentation of what’s being done and recording of it—I think we’ll see this steady growth of additional participants getting involved, as well as additional information on what we need to watch.
The reason I state that is, a lot of people talk about it, but when you analyze a part that’s been additively manufactured, it’s hard to say what the cause of a defect is and what’s an acceptable level of defect that you can have in the piece. How do you quantify those? Is it X-ray, is it destructive? And that doesn’t even begin to open the doorway for what can happen when we start to do gradient materials, or variable materials, which is just another whole compounding factor on it.
Trying to get enough clarification in the standard, or enough ability to clarify the requirements within the standard so that parts can be reliably manufactured by a variety of sources, and get an output that is within acceptable parameters, would be awesome. When we get to that point I think we’ll start to see the proliferation of the technologies go in all sorts of different directions. Then we’ll just have to watch as the configurations of materials, machines and software impact and how that gets managed and controlled.