Cases for Adopting Additive Manufacturing Using your CNC Machine

by Jason Jones
Mar 30, 2021

The Peter Parker proverb, “With great power comes great responsibility” can be applied to the emerging class of hybrid CNC machines. Within the last decade, augmented CNCs with additive, subtractive, and/or inspection capabilities have emerged commercially under the term “hybrid machines.” While the multifaceted capabilities of hybrid machines have intriguing technical potential, what justifies the integration of such dissimilar capabilities? Furthermore, when do they actually stack up to be the most effective tool for the job? This presentation will address these crucial questions by reviewing multiple use cases for hybrid machines. Examples for both metal parts and polymer composite parts will be examined. Each will consider the manufacturing process chain and individual process steps with enough detail to evaluate time and cost implications. From this vantage point, observations and learning outcomes will be shared about when using hybrid machines do and do not add up. Attendees will leave with clear concepts for evaluating when hybrid machine adoption will deliver great value.



Transcript:

Stephen LaMarca:

And up next, we've got Jason Jones, co-founder and CEO of Hybrid Manufacturing Technologies. Jason, ready to take it away?

Jason Jones:

This is going to be an exciting half hour. I'm really looking forward to it. In conclusion, just in case, I'm based in Texas, so in case the power goes out, here are my conclusions. Number one, AM or additive manufacturing is more than just printing. Number two, what's in your utility belt really impacts your wallet. And number three we'll look more closely at, but as we look to do in-process finishing, inspection and optimization during additive, we then drive additional value. And that is evident as flexibility, agility and quality.

Jason Jones:

If you have any questions, please type them in and we'll get to them as we go throughout the presentation and at the end. And a couple of strange things that I've observed recently, I'd like to start with these. These came into my inbox, I subscribe to a variety of industry-standard newsfeeds. And I got this, and I've anonymized it. I don't want to ruin anybody's day. But I got this from a trusted channel. And as I looked carefully at this email I thought, "That's kind of strange." They're advertising 3D printing and yet, if you look carefully at the photo, that is not 3D printing.

Jason Jones:

And I kind of smiled and I thought, "Okay, well maybe the advisory board didn't catch it or something like that." And I went on with my day. And then, but clearly this is a 3D printing event, and it just doesn't show it. This is laser cutting. You can see, the sparks go below the piece of metal there that it's cutting.

Jason Jones:

So I thought, "Okay, fine." Within a few weeks I received another email about a different event, completely different event, and with a different sponsor. And once again, I looked at the photos, and I thought, "I don't see additive front and center. I see laser cutting."

Jason Jones:

Now, I get it that laser cutting photos are cool, and laser cutting is awesome. I love it. But to be advertising about additive and not to have photos that go with it, feels like a discontinuity to me. There's a little bit of just friction, and I can't quite get over it.

Jason Jones:

And as I've continued forward, then I'm like, "Okay." With these two in mind, this is kind of frame of reference and my backdrop, I think, "What's next?"

Jason Jones:

And recently, I got another email, and it said, "Band saws increase efficiency in 3D printing workflows." And it's like, wait a minute, we've been talking about laser cutting, and more laser cutting, and now band saws. What in the world do any of these things have to do with additive manufacturing or 3D printing?

Jason Jones:

And this really brings up an old debate if you will. The term 3D printing versus the term additive manufacturing. In 2009, when we founded the F42 Standards Committee at ASTM, we had a lengthy discussion, it went on for quite some time, trying to decide what are we going to call this industry? And we clearly saw the larger opportunity commercially to have an impact on the world, and to do good, to transition what was at the time, rapid prototyping and rapid manufacturing, into true additive manufacturing. Bearing in mind that we would still serve some of the prototyping and other needs as well.

Jason Jones:

But these terms are used very loosely and often by many people interchangeably, including myself. I have no issues with that. But I do want everyone that's watching to leave today with a very clear understanding that doing manufacturing with any kind of 3D printing technique is a much larger undertaking than simply doing prototyping on another lower threshold of applications.

Jason Jones:

Let's look for a minute at what are the steps that you have to do in order to use 3D printing in a manufacturing context. Almost all 3D-printing devices concern themselves with either forming and/or fusing, usually both. This is what the mainstream 3D printing providers are looking at. And I'm going to give you a little detail by what I mean from this. And this is true across all seven of the standardized additive techniques, and all additive techniques as far as I'm aware.

Jason Jones:

I've chosen puzzle pieces to represent forming. This is the taking of feedstock, usually powder, wire, pellets, sometimes sheet material, and either stacking it up in layers, or collecting it in such a way that you actually form the geometry of a future part that you are printing. Now, often it's in lawyers, there are a few exceptions, but just go with me on this okay?

Jason Jones:

Next is fusing. Sometimes parts get fused as they go, and sometimes they get fused as a second step. But either way, what were discrete and individual particles or pieces have now been joined. And I use fusing in a very general sense. Fusing in a sense of consolidating material, often with heat, but not always. And if you get into the fine print, clearly there's all types of fusion. Right? Whether it's solid state, or liquid, or some other variations. There's chemical bonding, and crosslinking, and curing, and these types of things. And this fusing I'm referring to is this consolidation step.

Jason Jones:

This is the domain of most 3D-printing devices. Rarely do they go any further. My problem with this is that this is not a ... What is produced is rarely a finished article coming out of these machines. And for the sake of everybody, I would like to have a finished part come out of a machine.

Jason Jones:

Forming and fusing often happen in the same setup, but not always. And if you want more detail on that, we'll have a separate discussion later. But certainly, all the binder base processes have separate fusing steps after the initial forming with a little bit of fusing.

Jason Jones:

Next, to really take this into manufacturing, you have to address the finishing needs. Very rarely are surface finishes and precision, dimensional precision, certainly for mating parts, met by as-printed parts coming straight out of the 3D-printing device. Most often, and especially here I'm speaking of metal, knowing that AMT is largely a metal-oriented audience, parts get finished on a CNC machine. In fact, many estimate that it's a large percentage that go to a CNC machine, maybe as much as 95% or more of parts that are made in metal, 3D printed metal parts, go on for finishing in the machine.

Jason Jones:

Okay, but we're not quite done there. We've gotten a lot further. Parts are starting to look finished, but there's always a few things at the end, really for finalizing, and they really concern themselves with quality assurance and inspection, and the related things that go around it.

Jason Jones:

From my perspective, you have to deal with all four of these to truly get manufacturing-level tight parts. And some people, when they say additive manufacturing, they refer to this entire process chain, forming, fusing, finishing and finalizing. Others refer to doing forming and fusing at such a quality, that with the appropriate finishing and finalizing, you will have a good part. And it is always worth clarifying that. My mission today is to show you what is possible along this entire line in a single machine. It's an ambitious task. It doesn't work for all applications, but there are more and more applications that are starting to emerge, where it is possible to join more and more of these steps in an automated fashion.

Jason Jones:

So I come back to the last image that I showed you. Are bandsaws relevant for 3D printing workflows? And if you're going toward manufacturing, or if you're in materials that require bandsaw cutting, the answer is yes, this is actually a relevant email and I was perfectly happy to receive it. Whoever created this article, you can rest easy. But where do they stem into the workflow? Well, from my perspective there's a couple of places.

Jason Jones:

Certainly, if you look at finishing, absolutely. You need to remove metal printed parts from a build plate, and often it requires cutting, either EDM, or bandsaw, or something else. Also, if you're using build plates that are metal, you might even use them ahead of forming. You might be preparing those build plates for their position by cutting them initially, and then usually finishing those. So yes, bandsaws fit in. Lasers? Lasers fit into all types of additive, laser cutting occasionally, but not as much.

Jason Jones:

Okay, now, we've spent a little while building this foundation of these four Fs, what do we do with this understanding? The idea is, how can we automate? How can we get to an industry 4.0 future factory, where literally finished parts come out of machines without people having to do intervening steps?

Jason Jones:

If we take generic automation thinking, just very generalized, "Hey, what are we going to do?" These happen on separate devices. I know, what we'll do is we'll take a robot and put it in between the separate stations, and so then at least it'll get transferred along. And using a robot as a transfer mechanism is perfectly appropriate. And for some applications it's the best choice.

Jason Jones:

For other applications though, as I start to look at it, and looking across the seven families, about six of them can produce parts in metal, either directly or indirectly. And if 95% of those parts end up on a machining tool at some point, then the question is, is there a case for doing the forming and the fusing on a machine tool as well as the finishing? And of course, there are conditions where that can occur.

Jason Jones:

And because that's the case, and because machine tools are such a dynamically accurate platform, the question then is, can it be the common denominator? Can you do everything that you need to do on a machining tool? And let's explore that a little bit deeper.

Jason Jones:

In order to get a machining tool to function in all of the different variety of different types of things it needs to do, we've got to empower it with quite a wide range of utility. And how do you get utility into something? How do you broaden its capability? And leaning on one of my favorite characters, I would have to say, what good is this suit without the utility belt? That has to be the most desirable thing about the entire suit, even though black is a very slimming color and I'm told that it's a great color to wear.

Jason Jones:

Transferring that same logic to a machine tool, essentially the utility belt is the tool magazine. And so let's load up that tool magazine with as many types of tools as possible. And in some small degree, then the tool-change arm, or tool-change mechanism becomes a little bit like, it takes the place of that transfer. Instead of moving parts back and forth, you're actually moving tools. Instead of a part flow, you might have a tool flow through a certain set of activities.

Jason Jones:

That then brings to mind, if you can tell a lot about a person by what they carry in their wallet, then the slightly less well known corollary is obviously that then you can tell a lot about a machinist by what they keep in their tool magazine. Let's explore some of the unusual things that you can put in a tool magazine, and what you can do with them.

Jason Jones:

And I'm going to address this within the framework of those Fs. And so first we're going to talk about finishing, and then we'll go to inspection, and a little bit of optimization across the whole board.

Jason Jones:

Finishing, obviously those top two steps, forming and fusing, they're already taken care of by mainstream 3D-printing equipment. And so now, the next thing to focus on is finishing. Now, let's jump right in to this part right here.

Jason Jones:

Well, first off, why in the world do you want to finish, or why would you want a finishing process? Obviously, it saves you the time and energy of moving parts around, there's something else it brings though. Much of the effort with machine tools is spent on fixturing, figuring out how you're going to hold parts rigidly, and you've got to design that in if you're able to do it all in a single step. And so it essentially consolidates some of the activities that you need to do.

Jason Jones:

By doing that, you also have the ability to maintain accuracy. In fact, your machine doesn't even have to recalibrate or re-register. It's all co-located not with a virtual coordinate system, but literally with the same coordinate system.

Jason Jones:

It also encourages additive to operate quickly. There are times where 3D-printing devices are slowed down in order to achieve a better surface finish. So the volumetric deposition rate is held hostage to surface finish. And if you combine a printing device with a machining device, you've already got a fully optimized way to address those surfaces, so long as you have access to them. Let the 3D-printing process operate as quickly as volumetrically possible, and then deal with the surface finish later. And so it decouples those and allows you a higher level of efficiency for many types of parts, especially as you scale up and get to larger layers, and larger surface textures into the stair-step effects.

Jason Jones:

Okay, so many of you I hope have heard of hybrid manufacturing for metal parts. I realize many of you are metalworking people. I would also like to draw your attention to the fact that there is a hybrid manufacturing opportunity for polymer and polymer composite parts. And I'd like to show you some of those in the spirit of it being less well known and newer.

Jason Jones:

This is enabled by having a deposition head that will actually fit in a CNC spindle. I happen to have one right here on hand. Literally this end is a tool holder, and it will go straight into your spindle. Here it is inside of a spindle of a normal machine tool. And it is building parts, and it does so very, very rapidly. It is designed to operate at the scale that a machining tool work envelope has. And so it operates, if you compare it to a typical desktop machine, you could be anywhere from multiple-hundreds to even a couple-thousand times faster volumetric deposition rates with this head. Because it's got a tool holder on it, it can also be stored with the other tools. And so it just gives you convenient access to be able to on demand, use this device that normally is for subtractive, but as needed, go ahead and print with this as well.

Jason Jones:

Here's a quick photo of a part that's being built. And as it gets built a little bit higher, and a little bit higher, this particular part has media that flows through it. It's important that the opening at the top, at the very top portion of it, is concentric with what is printed. If you've ever tried to fixture a part after printing, often, especially for starts, and stops and other things, there are some non-uniformities that you want to machine off. And if you're trying to set it up, either manually, or even with a touch probe, it can pose challenging. And so being able to, in the same setup, literally switch tools and be able to use your cutting tool at the same time, gives you this opportunity to get this registration perfect.

Jason Jones:

And so in this sense, this combination of adding and subtracting in the same machine, this is a hybrid approach. And it really does wonders for both surface finish and for registration. And I'm holding a part here, and it's beautiful, and nice, and smooth, and wonderful where it needs to be, where it interfaces with other parts.

Jason Jones:

That particular part that you're watching operate at about 200-times faster than a desktop fused-filament fabrication or material extrusion platform. And it really gives you this flexibility. Literally, this head could go in a drawer. You could forget about it for a few weeks, and then when you need it, you pull it out and use it.

Jason Jones:

If you have a CNC machine, and you want this capability but don't have the floor space, or don't want to justify an entirely separate system, maybe for cost or other reasons, there's the ability now to switch back and forth.

Jason Jones:

And of course, where you're trying to deal with surface finish, and quality, and concentricity, and all of the sorts of alignment and registration issues that come with the final steps in taking 3D printing and applying it to manufacturing applications, those get resolved as long as your work-holding is rigid enough.

Jason Jones:

Okay, so let's take a look at another part that is along the same theme. Again, this is angled duct, and it uses the same hardware that you just saw. From our perspective, we innovate first with hardware and electrical. And then we innovate with the software second, because you have to have the hardware to drive, to make all this stuff come out the other end. Here you can see though, we're building up this angled duct, and once it's built, then literally we switch straight over. We don't wait more than just the tool-change time, about a minute. And we're able then to start cutting. And you can see the important feature here that we're demonstrating, is the interface where it will slide over, mating duct needs to have a good surface finish. We want to be able to put sealant on that and get vacuum seal. And so that's why you see that machined surface.

Jason Jones:

And I'll just hold this up to the camera, you might see even a better view of it as I hold it right here. But you can see this nice smooth edge, all the way to the full depth that you need it. And that is what hybrid manufacturing of polymer and polymer composites brings you. This particular part was built at about 450-times what a typical desktop 3D printer would be, would have made it at. Literally, we're talking about minutes and not days. Again, you flexibly use your resource, and you get that quality that you need around the edge.

Jason Jones:

Yet a third example of, this is a checking fixture. Now, when we go for checking fixtures, we're really trying to tune in on accuracy. And so this particular part has an aspect ratio such that all of it can't sustain machining force while in its initial printing setup. That's okay, sometimes that happens. The desire is to reorient it. The catch of course, is that to put it, to hold it easily, you need to locate it on some surfaces. So it's highly advantageous in this case to be able to actually machine that top surface perfectly parallel with the bottom, so that when you reorient it and put it between two jaws, you're not guessing at where this part is. In this case, on your X-axis, it is exactly aligned. And that allows you then to machine with high precision and get to a beautiful surface finish. In this case, it was a checking fixture, only one surface needed to be finished, and you can see that on the far right.

Jason Jones:

And this is a slide just to show, hey, this stuff starts as pellets. It then gets transitioned into layers, a layer-based object, and then finally the surface gets finished. And it is a really efficient workflow. It's done on a single machine. There's no waiting, there's no line balancing. It doesn't sit off to the side for a long time. And so the priority on this part was getting the precision right, and we were able to do that, as well as printing about 250-times faster than you might see on a desktop machine.

Jason Jones:

Where time matters, hybrid machines can really help. And that applies to both turnaround time, as well as just duration of how long your operation's taking.

Jason Jones:

Okay, another example here. This is an automotive CAM. This particular CAM is used inside of older fuel pumps, and they rely on this cam. And this is actually a real manufacturing job. This was scanned by Hexagon, company Edgecam and Hexagon and Co. And later they gave us the data, which then we were able to print, so this is repairing this surface. This surface gets worn away over time, this customer is [Yankum 00:21:09] Diesel. They have been searching for a replacement for these parts for 25 years, and they have boxes of them.

Jason Jones:

And so now, in a single setup, we're able to add metal back on them, finish it back to just precision as it was matched to the new ones. You can see on zoomed in picture there, let's take a little closer look at it. Look at the interesting texture on this part. You can see all of these little divots and valleys. To me it looks almost like a little footprint. That is all traced while you're driving, to govern the balance between your fuel and air ratio, your mixture as you drive. And it featured heavily in cars in the '70s and '80's, '60s, '70s and '80s, including the Le Mans winner, it took Porsche to win Le Mans for the very first time, this particular technology. A fun, for those who are enthusiasts of cars, really fun application. You'll hear more about this part as more details unfold and more video can come out, but I'm pleased to share it with you here first.

Jason Jones:

This is incredibly quick, and literally, you can do this in a few minutes. Really, the wait time on this has been 25 years, with some scanning and other things. But now it's several minutes to be able to do these. Another great example of advantages that come that way.

Jason Jones:

We've talked enough about finishing, at least for now. I'd like to jump back out, let's go to the next wrung down, which is inspection. Once again, forming and fusing, normally covered with 3D-printing devices. Finishing we've talked about. And the next step to automate is the finalizing, the quality-assurance step.

Jason Jones:

And so we'll take a good look at this part. As we do, I want to stop for a minute. This is about tooling insert, we also use a similar process for mold repair or mold edge enhancement. And the principles translate straight across, so let's look at that process for a second.

Jason Jones:

You can see that we've got, here on the left, is an original mold or a tooling insert where you've got edges that are the same material as the base material. And what we want to do is replace that material with a high-performance material. Often it's a very hard alloy, or it's a metal-matrix composite. It just depends on the application and the customer need. But to do that we will ... Again, this is a hybrid machine, this is now in metal. And so we will cut the edges, we'll prep those edges by removing some of the metal. We'll then add a new material to form the edge. We build that oversized slightly. And then we will come back and machine finish it and check that it's right while it's still in situ.

Jason Jones:

Now, there are a couple of different ways to check things. One, is visual inspection. Everybody loves that. That's why there's windows on CNC machines. Right? Essential, can't live without it. But when you need higher degrees of quality assurance, especially when you're trying to look at the bond when you've added a dissimilar material to an underlying parent material, how do you know that bond is good?

Jason Jones:

Another way that you can look and inspect these is with touch probes. Now, that'll give you an idea if the surface is dimensionally accurate, but it still doesn't quite give you the answer about what the condition of the bond is inside.

Jason Jones:

Yet another technique that's often used, if you want to look a little deeper, then people will often use dye penetrant. Dye penetrant is really convenient to use. I saw that, it's low cost, it's easy to get access to. You spray these chemicals on, often in 10 or 15 minutes it kind of develops. This is a really good deposit. You can see the crescent-shaped silvery portion in the middle of the screen. That is a deposition with no cracking whatsoever, it's perfect. If you look over here on the right, this is one where wrong parameters have been used, and there are some cracks, and they show up really well with this dye penetrant.

Jason Jones:

Now, the catch with dye penetrant, is because it takes as much time as it does, and it's a little bit messy, often this becomes an offline process, so you take it out of the machine to do it. And then you have the time to unload and reload, and the wait time in between, and managing the work-in-process in the meantime. And it's hard to document.

Jason Jones:

We have had great success with some initial applications of substituting a dye penetrant instead with eddy current. And we've packaged eddy-current probes to fit in machine tools, again, on a tool holder, it'll go in your spindle. They get powered externally. And here's an example of the workflow of using one of those heads.

Jason Jones:

This is prep. Again, this is a tooling insert. You can see the outer shape is being created on the left. And then the edge prep on the right. And then deposition. On the right-hand side you see the completed deposition of the hard-wearing material, some people call this hard facing. And now this final machining pass, you can actually see the two different metals. You notice that top cap around the edge is a different color and a different texture, that is the hard-wearing high-performance material, and that's what we want for this type of a tooling insert, to double, or triple, or sometimes quadruple its lifetime. And when you can deposit this sort of thing in a matter of a few minutes, maybe six, or seven, or eight, it really is a game changer.

Jason Jones:

Once the deposition and milling is complete, we then turn on the eddy current. And the eddy current then will go drive around these edges, and it will inspect to check for discontinuities that are subsurface. It will check for surface breaking cracks, just like dye penetrant does, but it will also detect into the depth of the material. We call it shallow subsurface checking.

Jason Jones:

Here is a quick video, it'll play for you. Here is a tool change to a deposition head. It has a supply mechanism that comes down. And now we're depositing. This is after edge prep is complete. Switch back over to machine it. Here's the completion of the finishing. And now you're about to see the finalizing steps of quality assurance for the first time, all in a single setup. Come in and detect along that edge, know that it's good, and you're finished. And you literally have created a finished part that you know is good before it leaves your machine. You can send it off with confidence, you don't have to wonder or brace yourself that quality assurance is going to come back.

Jason Jones:

Here's a little video of the data capture. I enjoyed Mateo's remarks just before mine about how critical managing data securely is, what protocols we should be using. And here is an example. This is a data stream, you can collect in any way you'd like using [MC 00:28:18] Connect or other protocols. And you can store it locally. There's an edge-computing device that sits next to these deposition systems that will manage it for you, or you can send it out to your ERP, or put it in the cloud, or whatever you'd like to do. Again, a very critical piece of taking this technology and making it mainstream.

Jason Jones:

This particular part, less than 10-minute cycle time. We've seen several parts of this, right? It'll get down to just a few minutes depending on how big they are and how long they last, but it is everything complete. Print, mill, inspect, all done together in a single setup. This, to me, is the ultimate form of flexibility. You want operational flexibility in an environment like now, you have a machine that can do that. You push a button and it becomes whatever you need it to be.

Jason Jones:

And very critically, because the inspection is built in, it is a no-faults-forward. If there is anything that is wrong that is found, you have the option to stop and repair it on the spot, check if it was the wrong parameters, or the wrong feedstock material. But you know, because you've just measured it, and you've got data that documents that, that you can pedigree that part going forward all the way to your customer.

Jason Jones:

That is our discussion about finalizing, and now, I'd like to talk about optimization for just a few minutes as we round things out. There is no single processing head that can do it all. There just isn't. Even within, if we go even within ... We just talked about finishing and finalizing, but if we look up to forming and fusing, which is within an additive context, right? Normally in 3D-printing machine, it is not ... In a machine-tool world, we're used to having flexibility. We can switch back and forth between different cutters no problem. In 3D printing we usually get one head, or maybe two side by side if we're lucky. And that is in selective-deposition techniques. And so as we look at that, you say, "Okay, I'm going to select the most universal, well-rounded head that I can, straight deposition head. Here it is for metal." And something, it'll be about this size, something like this. It can have a tool holder on it, so it'll go right into your machine. Here you are, you're depositing. You're depositing on a rectangular block and everything's great.

Jason Jones:

Now, somebody comes along and says, "Hey, this is really good, but what I really need is to clad on the outside of a cylinder." You're like, "Okay, I can rotate that horizontally, hold it in an A-axis, no problem. I can deposit straight lines [C-licks 00:30:56] whatever it needs to be. We're good."

Jason Jones:

And then they come back and they say, "Ah, I forgot to tell you, I actually want to deposit on the inside of that cylinder or of that tube." So you get your best five-axis guy out there on his machine, and you get the head sort of partly into the cylinder and then you realize, you've compromised. It's okay, sort of, but the laser energy isn't normal to the surface, you're not using the head as efficiently as you could. You're overusing feedstock, and you can't go all the way through the inside of that tube. And that's the bottom line.

Jason Jones:

And so when you get to a compromise, most people in additive are not used to having other options, because there's only one head, or two. And they're almost always facing straight down, because they're building their whole thing from scratch in layers and they're not equipped for accessibility issues.

Jason Jones:

And yet, there are many times where there's a more optimal path. And having interchangeable heads and a hybrid solution brings you that. What you want is an internal cladding head, or an angled head, or some other variation, a high-access head, where you can activate your tool change arm, go right over to the tool magazine, have it index around and find the head that you really want to use. And in this case, we have an internal cladding head on this machine. It loads straight into the spindle, and there you have inside and out, optimal deposition, changeover between those two in a matter of seconds. This is not a long project. Literally, you push the button and it happens.

Jason Jones:

No single processing head can do it all. And for that reason, having this flexibility allows you to optimize in a way that we've never been able to before.

Jason Jones:

I'm Dr. Jason Jones. It's been a pleasure to spend some time with you. I'm looking forward to answering some questions. You can literally do an end-to-end automation solution all based on a machining tool, and it's a beautiful, elegant solution. Additive is more than just 3D printing. What's in your utility belt impacts your wallet. And as you can implement in-process finishing, inspection and optimization, you will derive much greater flexibility, agility and quality.

Jason Jones:

Thank you so much. I'm happy to take questions. And keep typing into that chat feature.

Stephen LaMarca:

All right, Jason thank you so much. That was awesome. Instead of taking the easy way out and handing it right off to Thomas right away, I just want to say that as questions came up in your presentation, it seemed like a couple seconds later you answered it right there, so that was really cool. But the first thing that I would like to talk about is, at the very beginning in your presentation, you were kind of calling out pictures that were used improperly in additive material going out in emails. And I've got to say, at first I was like, "Dude, I'm about to call this guy out, because look at the picture behind him on the wall. It's got a tool holder in it with some weird red egg in it." And I didn't know what it was. But then you explained that was a deposition tool. Am I right? That's a deposition right?

Jason Jones:

That's correct. Yup.

Stephen LaMarca:

In a CNC tool holder. So I guess my first question would be, how easy is it to employ one of your additive tools into your, I don't want to say standard, but an average CNC machine like a Haas VF-2, or a Mazak INTEGREX I-100?

Jason Jones:

Sure.

Stephen LaMarca:

Your video-

Jason Jones:

[crosstalk 00:35:00]

Stephen LaMarca:

Made it look really easy. It was just a tool swap like it was a normal tool. But clearly there had to be something else going on. There must've been some wireless connection going to the tool head and-

Jason Jones:

There's a couple other layers to it. But yes, if it's a vertical CNC machine, it's fully engineered. It drops pretty much straight in if it supports [Amnesia 00:35:17], 40-Taper, HSK63, all that kind of stuff. If it's a mill turn with Mazak, we work directly with them, but they engineer their own solution on the mill turns.

Stephen LaMarca:

All right. And how many different types of tool holders, types of spindles do you guys make deposition tools for? Can I get one on a Pocket NC?

Jason Jones:

So-

Stephen LaMarca:

Probably not yet.

Jason Jones:

Well, we've used every controller, almost, that I know of, in terms of controller integration from Heidenhain, to Mazatrol, to Mitsubishi, to Haas, to you name it, Hurco, Siemens, [Vannic 00:35:57]. I don't know if I've missed anybody Open CNC. Literally, this is a few extra M-codes from a machine-tool perspective, and an [e-stop 00:36:06] connection.

Stephen LaMarca:

Wow. Let me get up my other questions here. Thomas, do you have one for-

Thomas Feldhausen:

Yeah, I'll jump in. Great talk Jason. I always love talking about hybrid manufacturing. I'm a little biased towards it. Can we talk briefly about some of the CAM programming?

Jason Jones:

Yes.

Thomas Feldhausen:

We see a lot of people that do additive, they get scared of machining and vice-versa. Is there solutions out there that can do everything?

Jason Jones:

About the 10 largest CAD/CAM providers now offer some variation of additive and/or a hybrid module to go along with subtractive machine tool pads. Now, all of them are in a state of maturing. They're not nearly as mature, additive or deposition tool pads are not as mature as subtractive. But I could name eight, or nine, or 10 CAD/CAM companies right now. If we use a mainstream CAD/CAM package, probably there's a module already ready for you.

Thomas Feldhausen:

Yeah, that makes it the total package, right? One machine and one programming software?

Jason Jones:

It creates the ability to track problems, and responsibility, and accountability. It really is, route-cause analysis lies usually with one or two people at the most.

Stephen LaMarca:

... Oh right, oh geez. Wait-

Jason Jones:

Which is a great thing mostly, unless of course you're a Nazi.

Stephen LaMarca:

With additive-part metallurgy, the metallurgy of an additively produced part is typically compared to that of a cast part. Is there any risk in using this type of technology in a machine that might not necessarily be able to mill or turn material that has been cast?

Jason Jones:

The deposition steps are, depending on which technology you have, it's either non-contact, or it's very low force relative to cutting. And so a lot of these deposition heads are used in robots, and gantries, and far less rigid platforms. You can do those additive steps without concern, for the most part. When it comes to machining, that's really when you have to think carefully about which platform am I going to put it on. Or am I going to downstream it? Right?

Stephen LaMarca:

Got you.

Jason Jones:

I've given you examples today of trying to pull it all together, because it simplifies things, but there are times where it makes more sense to keep them separate.

Stephen LaMarca:

Got you. Our first question that came in is from Dr. Cooper of the NSF of course. And he says, "Great talk. Any ideas how you would finish internal geometries where there's no line of sight for the tool?" And actually, before I hand it over to you, the actual expert who knows this, I'd like to take a guess. Is the answer to that, to finish an internal geometry that there's no line of sight to, is the answer to finish, as you're building up a part layer by layer, you finish it before you get on to the next layer?

Jason Jones:

Absolutely, while you still have access to it.

Stephen LaMarca:

Awesome.

Jason Jones:

You're able to very easily access what I would consider to be up-skins and verticals. The down-skin is tricky. There is no, if you're building from above, then you don't get access to it, but everything ...

Stephen LaMarca:

Oh, did we just lose Jason?

Thomas Feldhausen:

Looks like it. Good thing he started with his conclusions.

Stephen LaMarca:

Good thing he started with the conclusions, and the other good thing is we were in the question section anyway.

Thomas Feldhausen:

Yeah.

Stephen LaMarca:

But-

Thomas Feldhausen:

I can actually take a stab. Our other question from Lawrence, he asks, "I don't see a feeder mechanism that supplies the FDM material attached to the CAT 40 tool holder." Funny enough, we have one of Jason's polymer processing heads at the national lab here. And so I can answer that, where it actually comes in from a feeder tube on the side. It uses a vacuum to pull the polymer pellets and then it feeds it directly into a holder, so you don't have to make any changes to your actual tool holder.

Stephen LaMarca:

Oh wow, so you-

Thomas Feldhausen:

It's a pretty quick set up.

Stephen LaMarca:

You actually have one of these on site to [ORNO 00:40:15]?

Thomas Feldhausen:

Yeah.

Stephen LaMarca:

Oh wow. So that was kind of getting at my first question, there's actually a feeder tube going into the tool?

Thomas Feldhausen:

Yeah, I'll let Jason tag on here. We were just talking about Lawrence's question about where you don't see the pellets being fed through CAT 40 tool holders.

Jason Jones:

Yeah, they don't go directly through the CAT 40 tool holders. There is a supply mechanism. I showed it in one of the videos. And then the different types of heads have slightly different ways of handling the feedstock, but there's something for pellets, and powder, and wire, and anything else you might need.

Thomas Feldhausen:

I've got a question Jason. How important is data flow through hybrid manufacturing? Kind of building off our previous speaker, is there a need for data flow? And what's out there, and where are the current gaps?

Jason Jones:

Yeah, from our perspective, there's a huge opportunity to make that data flow really rich. And so much of what we've done in research for decades, if not centuries, is make something and go cut it up, go destroy it. That's our de facto standard. And getting the data back from that loop, it's a long loop. And so our approach has been, what can we do in situ on the machine to provide a level of confidence? It won't be the same as putting it in a CT scanner, but at least you get some feedback right away.

Jason Jones:

And so from an integration standpoint, you have the option to do very little data, but it's very beneficial to do more.

Thomas Feldhausen:

Great. Do you have any questions Steve? If not, I'll keep them coming.

Stephen LaMarca:

Actually, like I said, he answered all my questions as they were coming up during the presentation. I'm still taken aback by how awesome it was. But yeah, no, you go ahead if you've still got more Tom.

Thomas Feldhausen:

Yeah, I'll just ask one more Jason. I love how you've got multiple feedstocks in one single machine. How do people who want to take a dive into additive manufacturing, how do you know, do I want metal powder, or metal wire, or polymer, or do I want all of them? How do you determine that?

Jason Jones:

The application that you have is always going to drive you down ... There is an optimized solution for your application. And if you're not sure what to do, we always recommend to people to start with polymer because it's the easiest to implement. It's a few hours on a machine and you'll be using it. It's a familiar environment.

Thomas Feldhausen:

Okay. I think we're at the end of this time, right?

Stephen LaMarca:

We're just about. Jason, thank you so much ...

PicturePicture