Getting to Zero Defects in Parts Manufacturing!

Recently, Avonix’s Ben Conners was a guest on Floodlight’s NDE 4.0 Podcast. Here’s how it turned out:

 

HOST: Hello, everyone, and welcome to today’s episode of Floodlight’s NDE 4.0 Podcast, where we posed five questions to a variety of NDE 4.0 experts and explore the benefits and challenges in this emerging field. 

Today we are joined by Ben Conners, an expert in technical X-ray and computed tomography. Ben Connors graduated from Bridgewater College with an Applied Associate Science AAS Degree in Nondestructive Testing in 1998.

He worked as an NDE technician in various industries, including nuclear power, bridge construction, aerospace castings, and structural steel. In this capacity, he helped ASMT Level Two Certifications in RT, UT, MT, PT, VT, as well as AWS Certified Welding Inspector. 

Since 2005, he has been in the digital X-ray and CT scanning arena as it has grown from a novel technology to being used in mainstream manufacturing, inspection, and failure analysis. Ben is currently the Director of Inspection Services at Avonix Imaging. His mission is to build up team and operations capabilities to meet the market’s future needs, specifically in scanning parts for aerospace, defense, electronics, additive, and medical devices with digital X-ray and CT. The high resolution and high power microfocus technology is fueling growth. So today’s discussion will focus on the manufacturing side of NDE and how we can bring together several of the latest technologies to become smarter and more effective at manufacturing good parts.

Welcome, Ben, to Floodlight’s NDE 4.0 Podcast.

 

BEN: Well, thank you. 

 

HOST: I will pose five questions designed to dig into some of the most meaningful and interesting aspects of NDE 4.0. 

 

First, can you tell our audience a little more about CT scanning? What makes the Avonix solution unique, and the benefits that your customers will achieve due to that?

 

BEN: For those that don’t know, CT scanning is an X-ray process. But rather than just one x-ray where we would see a two-dimensional image, and maybe you would see a defect but not necessarily have a depth of field, CT scanning rotates the part, and we take x-rays as the part spins 360 degrees.

So you might take 1,500 or 3,000 images in a rotation. As we collect that x-ray image, while the part is rotating, we’re able to see the location of different defects, both in depth and in size. CT is x-ray, but a far more advanced x-ray process. Think about when you go to the doctor for a CT scan —the x-ray tube and detector spin around the body. In industrial x-ray, what we do is the tube and detector are static and we spin the part. 

When we’re looking at a metal part or electronics part, we can give it quite a bit of radiation, and the part comes out fine.  CT really enables us to see defects that, depending on the situation, someone never saw before without cutting their part in half.

Avonix Imaging is a company that started in 2012 as a CT scanning service business, and we still do that today. Brian Ruether and Jeff Diehm started the company and partnered with Nikon, and the main reason we partnered with Nikon is because of their x-ray tube technology, which has certainly advanced over the last ten years.

That x-ray tube technology, that’s the base of the physics of x-ray.  And the way that these Nikon tubes are built and the way that we utilize them is we manage heat differently.

One of the things that we do is use a rotating target, but we do it in a micro-focus (very high-resolution) setting. And so, with that rotating target, we can see defects no one else can see.

The physics is fascinating with these types of x-ray tubes. We’ve done round-robin tests, and can pretty openly and confidently say we can see a defect no one else can see. And that’s a lot of fun because we get to hang out with incredibly intelligent people, and they’re blown away with what we can do. We get to help with a small portion of building a rocket or building an electric car or building a heart device—impactful things. We’re a small part of helping companies build them better. 

 

HOST: And how long has your technology been around? Has it been around since 2012?

 

BEN: Avonix has been in business for 10 years, but x-rays were discovered long before that.  The first x-ray was taken in 1895 and the first medical CT scan was in the 1970s. Using x-ray and CT in an industrial setting came some time later. 

Those industrial x-ray tubes are developed in the UK, and the unique thing is there are only a few places in the world where micro-focused x-ray tubes are built. Germany, most certainly. There’s the UK, specifically the Nikon or XTech product, and then there’s a little bit in Asia. But it’s such a niche market — we just manage the physics differently.

 

HOST: Cool. So moving to question number two: 

 

Let’s talk about CT scanning in the 3D printing industry. How do you combine those technologies, and what are the benefits?

 

BEN: Sure. The 3D printing industry is a lot of fun. If you look at where it’s at today compared to two years ago, compared to ten years ago —there are 3D-printed parts that are flying today, on aircraft, rockets, defense things, whatever the case may be. 

One of the big deals about additive manufacturing is that engineers can iterate very quickly, so they can make a design or a CAD model today. Maybe it’ll take a couple of days or a week to print whatever or two weeks to get your design on the machine, but relatively quick. For an older process, if you were to make a design and send it to a casting house, that casting house may take weeks or months to make tooling. It could take a long time to get your first prototype. Now they’re able to iterate very quickly with their design.

So our x-ray and CT piece of it: we’ll scan a part after it’s created. We can see inside of it, and they can see if they’re making it with or without a defect. With this microfocus 450, we’re able to see very small defects. Our customers are able to make a smaller, lighter, more high-quality part because they absolutely know that we can see defects no one else can see. 

When you start adding up weight reduction and quality, your confidence in fairly cutting-edge manufacturing situations, it’s a big deal to see those kinds of defects. Our impact is that feedback loop. 

To be clear, currently, there’s not a big red easy button that goes from the additive printing machine to CT and then tells the additive machine to change the product and change the manufacturing.

We had a study with a pretty high-end university recently that does defense work, and we scanned about 100 samples. And on the additive manufacturing machine, they set different settings.

The gist of the study was they could take the settings from the additive machine and the ground truth of the CT data. And they actually did destructive testing as well to prove out CT. Then they could make statements inside their organization about how to manufacture the next part.

They could use CT and their settings to make a good decision on the next round of parts.

 

HOST: That’s interesting. I mean, what we’re talking about here is almost real-time manufacturing changes in the manufacturing process. In the old days, when it was expensive to create the process that created the actual part, you wouldn’t want to change it too frequently because that would be costly, right? So the tradeoff is that you sort of wait until the process is making bad parts, so much so that you really need to change the process. Whereas now, as you say, we can make the process better almost in real time so that we don’t have to tolerate any kind of defect. 

 

BEN: Absolutely. And there are a couple of components. One is let’s squeeze defects out of the process. Any good engineering process wants to meet that defect at its creation point and fix it. The worst-case scenario is you’re always finding it in the inspection. 

And you’ve got 20% or 40% drop off. You’re throwing parts away. That’s terrible. So we want to feed that back. But the other thing with CT, high resolution 450 seeing defects that weren’t seen before is parts can be lighter, right? Because any engineering factor has to consider what defect could be in there that we couldn’t find.

 

HOST: Right. Right.

 

BEN: And the higher the risk, the more that’s a concern. But if we can see small defects confidently now, we can say, let’s make this a little bit lighter. So for making a tractor, who cares? As long as it works, it’s not a big issue. But cars that are trying to be more efficient or planes. There’s a lot of activity on composite going into aircraft for a reason. It’s going to use less fuel, or you look at pound for pound in the rocket business, with all these satellites going up. That’s a big deal.

I don’t want to project that we’re designing these rockets — we’re not. But we do look at their components,

 

HOST: It takes the whole process to make a change. So it’s fascinating to think about these things. One thing I’d like to dig into a bit. You’ve touched on it already, but you’ve talked about this feedback loop between inspecting parts and the configuration or settings of the 3D printer that manufactures future parts.

 

So for question number three, can you explain a bit how first of all, how it works today? I know it’s not magic, but how do you anticipate it working as you improve that process?

 

BEN: Sure. I’ll refer back to our university friends because that’s an easy one that I can talk about somewhat openly. In that feedback loop, they designed that test of making good and bad parts and the CT to train future operators and engineers on that additive system. 

They would always refer to as the ground truth. They set the knobs a certain way. And then these people can make decisions on: if I’m making this area of this part this thick, I need to think about these things, etc. I can see that process becoming a little less manual.

I mean, that was purely in an R&D environment. And the people we were working with are writing additive standards. They’re ultimately trying to cascade a set of rules like ASTM or other procedures, other guidelines. They’re trying to cascade that to the market that builds things that fly.

So that’s kind of where we’re at at the moment. If you go to specific manufacturers based on their pressure, they’re doing that quicker. Yeah, but we’re certainly a ways away from a CT scan; just cut, paste, drag and drop on an additive machine, and then some knobs are turning.

 

HOST: Right, right, right.

 

BEN: I’m not aware of that happening today. But there’s potential. The quicker we can help someone make a good part, the better off they’re going to succeed. 

 

HOST: Definitely. One of the things that’s always the case with new technologies is there tends to be resistance. 

 

Do you think there will be any roadblocks for organizations interested in this technology in terms of adopting an automated, automated feedback loop or the general technology?

And if so, what do you think the roadblocks will be? What form do you think that resistance will come in?

 

BEN: A lot of the time, we can use history as a guide. It’s not perfect, but it’ll give you some clues. The easiest one for me in this business to refer to is a film to CT conversion. Where there’s an old product or methodology that everyone’s comfortable with. In the past, before ASTM matured, there were a lot of arguments against digital. But you always have to weigh the value.

I think back to a customer I worked with years ago that spent $1,000,000 a year on x-ray film. You come to them with a digital x-ray, and they’re like, “Oh, we can stop using film tomorrow.” Well, you need a couple of these machines type deal. And then we had to start proving it out: Can we see the defects or not? 

And so the social, not the technical part that’s probably the hardest hurdle. It’s that kind of commitment to the old thing that’s hard to overcome. But I think usually the proof is in the pudding. If you can just do a side-by-side study, again, filmed to digital, as an example, I can see a defect here. I can see it here. Yes. Go to the next one, etc. You have to do that over and over and over again. The next hurdle after you’ve kind of proven yes, I can see it today is: Can I see the defect over a period of time, which leads to a lot of the ASTM stuff.

There’s a friend of mine that worked pretty heavily on the ASTM Digital and Defense Working Group type stuff, and it took quite a while for standards to adopt exactly how we’re going to test this machine to prove if I saw the defect today, will I see it in a year? Now, that’s all very normal. If you go in, anyone doing some kind of digital x-ray work, they’ve got some sort of an ASTM E2737 or something to prove their machine and the panel works right.

Pushing far forward into using CT to be some kind of a feedback loop, you have to prove it. The social aspect of getting past someone that’s got their heels dug in, that’s hard. It’s more psychology than technical.

 

HOST: It’s funny, in all the conversations I have with folks about the challenges, hardly anybody talks about the technical challenges. We spend so much time on technical aspects, testing, figuring stuff out, by the time we can take it to market, we’ve ironed out the technical issues to the point where we know it’s going to work. I feel like we end up bringing people along as more of an afterthought because that’s where the resistance is. It’s people that don’t believe that it works, and they have to be convinced, to be shown. That adds another year or two to the adoption cycle.

I don’t know that anybody’s figured that out, but what I have noticed is that almost always resistance is, like you say, a social thing, versus a technological one. 

 

BEN: It is. And I’ll maybe take a little detour. We were at an ASMT conference about a year ago. As a group of ASNT Level III’s and high-end engineers, there was an open panel discussion we were involved in. It detoured into a great discussion about the core values of the people that you’re dealing with.

So our company, maybe about a year and a half, myself, the owners, and a couple of other people, and a coach got involved with setting core values for our organization. Like any company, we want to have success for our team, for our customers, our strategic partners, vendors, and anybody that comes in contact with us.

If my vendor doesn’t succeed, how am I going to succeed? We’ve got to have some level of mutual success. 

We were brainstorming on a whiteboard about what should our saying and core values be. 

Three things stuck out when looking back on the people we’ve worked with.

They were authenticity, competency, and empathy, which we coined ACE.

Competency is what most of us technical-type people talk about. That one’s easy. Either you can do the job, or you cannot. The C is the easy part.

Authenticity is our way of showing up, joyfully, humbly, imperfectly (actually, there’s a debate on that last one, but I’m glad it’s in there). Be vulnerable, compassionate, curious, and grateful to serve others. If someone shows up authentic, you can be confident that they’re not fighting you for the wrong reasons. 

And then empathy, you’ve got to look through other people’s eyes, not your own.

 

HOST: That’s hard. 

 

BEN: If your customer has a line down and can’t ship parts today, you need to be falling all over yourself to help them. So back to the original idea, how do we get over resistance? A lot of that comes down to the core values of the person or organization.

Are you going to battle this progress, or will you help it along? I’m not saying do something foolishly — it’s got to work. Whatever the technology is, it’s got to work.

 

HOST: And a lot of the resistance, I feel comes from something you touched on, which is really risk-tolerance. That’s what this whole industry is about, is addressing risk. And so if the customer or organization is bringing in new technology, they have to be darn sure it’s going to work better than what they’re already doing. Otherwise, they’re bringing more risk to their organization and their customers. 

 

BEN: So and in some of that is technology, but some of that comes down to the P&L. The dollars have to get involved at some point. And if you can make a lighter part quicker and get to market quicker, you’re going to succeed. Or you’re really slow and heavy, you’re having a hard time flying; it’s going to be awkward.

 

HOST: It’s interesting to think about the dynamics between psychology and technology. And moving forward, change. All of those things are part of this.

 

BEN: Yup, there are still people. 

 

HOST: So, let’s move on to our final question of the day.

 

If you could jump five years into the future, what would you hope to see different, changed, or improved, and what impacts do you think that will have?

 

BEN: Oof, Well, I’d like to go to the stratosphere and say that there’s a red easy button for everything that…

 

HOST: Might be!

 

BEN: …just fixes the part. But just like anything, how do we eat an elephant? One bite at a time, right? Part of CT adoption is the standards maturing, right? The OEM and the top-end type customers are adopting CT and putting it on the print.

When I say on the print, I mean if you’ve got an approved engineering document of some sort and it says it’s got to measure this and be made of this material, and it’s got to be scanned, we’re starting to see that. But most of the time, if we get a print like that, it’s the first time this person is doing that.

It’s the first time they’re going and asking, the first time they’re receiving. Once that is rolling, then people look at it and say, okay, there’s a lot of manual work on taking this CT scan and talking back to the machine, right? So get it on the print. It becomes normalized as something that’s just as normal as a CMM or a laser scanner. It’s just part of the quality environment. 

Ten years ago, there was a 90% chance that somebody I talked to didn’t know that CT scanning existed in this form. Now, I feel with new customers, it’s more like 40, 50%.

We’re still in the education phase. Five years from now, I hope that we’re in more rooms and conversations where a customer comes in and they say, Oh, yeah, I’ve got a  450 Micro Focus, and I’ve got a 4343 panel, and I’ve got this, and I like cesium.

That’s when we push the data back around. We certainly have educated customers who have written standards and are certainly brilliant. We’ve got other customers where today might be the first time they’ve ever heard of CT. 

I think as the market matures education-wise in how CT works, that will drive NDE 4.0. Because you got to know the footings in the foundation before you start getting further along.

 

HOST: Yeah, I agree. There’s definitely a learning process, as you say. And getting used to technology from our world at Floodlight, it’s kind of similar. We’re providing a digital foundation for customers so they can get their data digitized, which is step one in this much longer digital transformation process.

And some customers are well on their way; some companies are struggling with these first steps. And you’re exactly right. So I think there’s a lot of correlation there.

In any case, this was a fantastic discussion. Thank you so much, Ben, for being here today and bringing us such an interesting perspective. 

 

BEN: Well, thank you for having me. That’s my first podcast!

 

HOST: Well, you did fantastic. And I look forward to seeing some new advanced advancements coming out of Avonix. And definitely, we’ll continue to follow you, Ben. And I encourage our listeners to follow Ben and Avonix. You’ll find links on our podcast web page as well. 

If any of our listeners have any feedback or would like to nominate an individual or organization to be a guest on a future episode, please send a message to one of us here through the contact us form on our website. Thanks again, Ben, for being with us today, and thanks to our listeners for joining. We’ll see you next time.

 

For more on what Avonix Imaging is up to, check us out here or contact us today!

Share with a Friend