Focusing on the tools and processes, experienced energy and aerospace AM experts discuss some of the strategies you can use to efficiently and confidently test your AM parts.
– Hello everyone and welcome to ASME AM Industry Summit Panel discussion about non-destructive evaluation strategies. Abbreviated as NDE. I am Dr. Jason Jones, and I will be your moderator for today’s session. We have some fantastic panelists that I can’t wait to introduce to you, but before I do, I wanna emphasize the chat feature in your browser will allow you to ask questions, which we will be answering as part of this discussion. So please use the chat, post the questions throughout the whole time that we’re speaking. And now we’ll introduce our panelists. First up, we have Shana Telesz of the larger Baker Hughes Group. Shana, give us a little intro about yourself.
– Hi, I’m Shana Telesz, I’m with the Waygate Technologies Group in Baker Hughes. We do non-destructive testing, we’re equipment manufacturers. I’ve spent the better part of the last 17 years specifically sort of working on the radiography technology and a variety all the way from film all the way up to digital, and now working a lot on CT in the last five years seeing what radiography and inspection technologies can do for additive.
– Oh, that’s great. Shana, thanks for joining us. Next, we’ve got Andy Kwas of Northrop Grumman. Andy, tell us a little bit about yourself.
– Hi everybody, my name is Andy Kwass. I’m a Northrop Grumman fellow responsible for space activities and additive manufacturing for the space system sector. I’ve been with the company 42 years, my 42nd year. So obviously I much like it here, and we’re doing some really exciting things at Northrop with regard to AM and NDE fits in a very important part of this I’m glad to tell you all about it.
– Fantastic. Well thank you for joining us and let’s launch right into questions. The first question we have, is tell us what factors helped you select the right NDE techniques for any given part? And Shana, why don’t you start off?
– Yeah, I think for any given part, it depends on what what the part entails? How complex the giving component is? What features it has? What are you trying to see? What’s its mission? What’s its purpose? And obviously I tend to spend most of my time on the CT side of things. So we typically see parts that fit well for CT as being things with complex internal geometries, things with very critical missions, the consequences of failure is very high. Things that go on aviation components or space components, but again, kind of depends on what you’re looking at. You can kind of go all the way from ultrasound or visual penetrant, but really depends on again, that complexity of the part, what you have access to versus what you wanna see and what you wanna validate.
– Oh, that’s great. And Andy, what can you add?
– Well, obviously the requirements are key. It’s hard to justify spending a lot of money on an NDE technique like a CT scan when the part cost a hundred bucks, right? So on the other hand, if it’s a mission critical part as Shana talks about then maybe it’s valid, maybe it’s the way you have to do it. So it really boils down to the requirements. Decide exactly where that part is going and deciding how important it is to the overall mission and how important it is to make sure it works. If it’s a ground test article, it’s probably not that critical, but if it’s something in space that’s mission critical or has human involvement life risk, it’s obviously very important and then you spend whatever you have to, to make it work.
– Got it, and in doing a little research ahead of time and talking with you all NDE the acronym for non-destructive evaluation also is used very popularly for near death experience. So if we tie those two together, if it’s likely a failure is likely to result in a near death or death experience, then you should validate with very sophisticated NDE. And let’s transition to our next question, which is. Both of you have talked about the importance of CT or more sophisticated methods. Where do you start when you have those lower value, perhaps less safety critical applications, what lower penetration or less invasive NDE methods do you rely on.
– To me?
– Please Andy. Okay. Yeah, it’s important to, again, base your NDE work on the value of the item itself. And we usually step our way through the cheaper ways to do things more cost affordable ways to test those things out. But again, we have to expect the reliability to be there. For AM parts that are so complex now, it’s very easy to have voids or unfused particles or something in the mix that are very difficult to see. And so we’ll try first to use the low-level affordable techniques, but if we’re not confident that we quickly move up the line to CT scans or other things.
– Okay thanks, Shana.
– Yeah, I think obviously we’ve seen a lot of different kinds of parts. Again people can do things from, if they’re smaller parts, putting multiple ones on the build plate, just choosing one of them to sacrifice and sort of destructively test. I told you an example yesterday that we’ve got some customers that I’ve seen kind of trying to throw the kitchen sink kind of NDE options kind of at a given part look at, “hey, can ultrasound see some of the single wall areas?” Can penetrant look out for some of those surface defects. We’ve seen people throw Eddy current at it. We’ve seen people put borescopes inside, kind of look at those features. Like, does everything look okay inside? Before they sort of get to, is radiography required? Is CT required? So I think like Andy said that it really depends on the part, what the requirements are, and what the features are. I mean, some of them can just be a visual inspection too.
– Okay, and for people who have, the opportunity either during a design stage or the manufacturing stage, what can be done to make evaluations easier for these parts?
– I think from my perspective, I think that this is one of probably our biggest challenges in the industry right now. As we selfishly look at it from the NDE side, additive technology gives the design engineers much more degrees of freedom and much more creativity available to them to make much more complex parts. So they’re having a field day kind of being super creative, making these really crazy things. And then we look at them from an NDE perspective and we’re like, oh no. You know what does that do to us? And I think personally I had a moment, a couple of years ago, I started working in the additive space a little bit more and kept seeing these larger and larger build plates going to meet our build plates with high density metals going, what are we gonna do? What does this look like? So I think it’s really important the design engineers and the NDE teams kind of stay tied off in some of these initial phases over the last two years. I’ve seen design requirements or drawing requirements that are just not practical. They have one meter parts and they’re asking for 5,000, 10,000 sort of defects, which just isn’t really possible to sort of see them validate. So there’s been a lot of negotiation going back and forth with design engineers, letting them know what’s possible for us to validate from an NDE perspective with our various methods, and then having them go back and kind of understand, is that tolerance level acceptable? Can we change the drive to that or do we need to rethink the design of the whole part? So I think getting ahead of this is super important and the people that are working together with additive teams, they’re having the time of their lives and making it super challenging for us at the same time.
– At the expense of you huh?
– Andy, what can you add?
– Yeah, I agree. It all comes down to complexity, at Northrop we’ll take a system or a component that’s 15, 20, 30 different parts all bolted together, and now I’m making it as a single part. And so I’m turning corners and I’m going around edges and I’m doing things that are so complex. And I agree with Shana that the designers are having a field day, it’s endless what they can do, right? But the poor guy that has to guarantee that it’s gonna work is looking back at it saying, Hey, the best CT scanner with the highest power. I can’t see everything. And that opens up some interesting things. There’s concerns in open literature and other places where if I can make a part that I can introduce voids or some kind of discontinuity in a corner or some particular place that you can CT scan all day and not see that part, failure method, it’s a good chance it’s gonna fail on you. And so at an unknown load, that’s not predetermined, but not where it’s supposed to be. And so that’s got me real scared. I’m real concerned that the technology of AM is moving so fast, that the ability to verify its performance of a part is actually lagging considerably right now. An answer for that. By the way, I don’t know what else to do.
– I was going to say maybe just to comment on that a little bit. I think that that’s a really good point that Andy brings up, one of the things that we’ve been working with a lot of customers on, is also to look at creating an RQI or some sort of reference quality indicator. So some sort of representative part, if the part is affordable to make an extra one and put known defects especially in those high risk areas like Andy mentioned, around some corner or whatever, so that we can inspect that one periodically before, after every scan or once a week or something to make sure that we’re reliably seeing those defects in those high risk risk areas, whether it be trap powder or some sort of porosity, distributed porosity, some kind of metrology feature, et cetera. So I think it’s super important that we continue to run with the practices, especially in AM, creating some sort of RQI that proves that if that defect were there, we would see it in that specific location.
– Right, sometimes in AM, we’ve summarized by saying, “oh, complexity is free.” But obviously when it comes to NDE, if it’s not free, is it? So let’s talk for a minute about innovations. What are you doing that’s innovative maybe that people don’t know or that you’ve seen, that are really trying to address some of these bigger issues and make them practical. Andy, let’s start with you.
– Well, again, there’s limitations on what you can do and an area that I think has some real promise is the ability to identify a part with an electronic signature. That’s not an ultrasound, everybody thinks that’s ultrasound. It’s really not. It’s the ability to literally run a charge through your park and look at what the response is to that electronic signature of that park. And it’s probably not gonna tell you where the problem is, but it’s going to help you identify quickly for a cheap part. That it’s a good part or a bad part based on some standard signature you see on what you know to be a good part. So I see that as a method of screening, quickly screening parts that decide that one goes in the waste basket or this one actually gets flown. So I think innovation wise, I see things along the lines of what Chris Williams is doing at the dreams lab in Virginia tech, as an example in that area as a very proactively moving towards a new way to do business.
– No, that’s great. Shana what else can you share?
– I think from my side as an equipment manufacturer, we’re constantly understanding what are the new challenges, what is AM bringing, I kind of mentioned earlier that couple years ago, I started to get scared about some of these build envelopes, looking at meter cube to kind of parts and now even bigger with some of the new metal printers. And that immediately brought us to the fact that we didn’t have enough energy, to penetrate some of these parts. Most people use 450 KV maybe 600 KV kind of technology, which penetrates maybe two inches of kind of cumulative wall thickness of a dense metal. And some of these parts that just wasn’t gonna cut it. They were gonna become an inspectable, especially the complex ones that are more mission critical. So one of the things we just did is we just installed a 9 MeV CT Cell at our facility and CT Linac to kind of help with that part of the industry, help people kind of to validate those larger components and introduce that kind of technology. The other thing that I sort of talk about a lot is making things production ready, right? It has to be affordable, NDE has to be affordable or no one’s gonna use it, or it’s gonna make additive not worth going to. So a lot of things, a lot of innovations that we’ve done over the last couple of years ago have really focused on production speed. So how do we make scans faster? And either getting more energy to the part quicker, making the detectors read out faster, we introduced scatter crack technology, which we’re also using in our high energy cell, to basically correct that and be able to scan parts kinda faster, use flat panel detectors where only LDAs were possible before. So we focused a lot of efforts on how do we do things faster, how do we make it more affordable? How do we, can’t be a 30 hour scan? It needs to be, really ideally it needs to be minutes, but we’ll start by going from 30 hours to five hours and start going from there. But I think those are, I think two key areas where our business is really focused on continuing to develop technology and make additive more realistic for people.
– Oh, that’s great. Thanks for trying to transitioning into what’s next, and Andy can you add a little bit about what’s next for you all?
– Yeah, what’s really next. And people say it’s something that gets you excited to get up in the morning. And I tell him it’s more why I didn’t sleep last night, but is the whole idea of how to do NDE and space. Northrop’s very actively involved with in-space manufacturing assembly and operations there working with organizations and other companies to actually print things in space. And the last time I looked, I don’t believe I saw a CT scanner floating around in space, or around the moon or Mars for allowing me to go on and take my printed part and see how good it is. So I’ve got a real challenge right now to be able to print something. And I can’t rely on necessarily, although it’s part of it in situ monitoring, watching the print, but I’ve got to come up with a way to be able to test that part after I’ve made it up in orbit, around a planet and know that it functions properly, not a trivial task. So I don’t have an answer for that yet.
– Yeah we need something that is inspirational is what happens far, far away. And again one time ago in a galaxy far away and transition that to in the very near term orbiting in our very own solar system.
– Some amazing innovations. Well, I can’t thank you both enough. This has been a fantastic discussion. And of course, this is just the beginning. We will now transition and let start to take live questions from the audience. Again, if you have not yet submitted your question via the chat feature, please do. Thanks for joining us for this first part. And now we transition to your question.