Multiaxis Large-Scale AM with Netfabb and PowerMill High-Rate Technology

DOMINIQUE MUELLER: So good morning, everyone. My name is Dominique, and I'm a research engineer on the Netfabb team and based in Germany. And I'm mainly working on customer research projects, and, at the moment, it's mainly large-scale additive manufacturing.

So first of all, I get the question a lot, what means large-scale additive manufacturing? This question is really hard to answer because there is not just one answer. It always depends on which industry it is, which material it is, which system it is. So with metal, we are a lot smaller than we can go with concrete, for example. And there are no standards out there, right now. There are a lot of people working on that, but it is still not defined what large scale means.

So if we, for example, look into the construction industry, we can build complete walls already. We can build complete house structures already. And it is also used for design. So the whole building gets a complete new structure because we can now integrate the design. And we don't need to have corners anymore. We can have round houses. And we can also combine materials, now. So we can use complete new materials for construction. We can use frameworks and print these frameworks.

And in construction, large-scale additive manufacturing doesn't mean it's one piece. For them, it's just big in total. So it can be 100 pieces, and if they assemble everything for them, it's large scale. It just matters that it is printed.

And we can now combine materials. So we are just printing parts of the structure. And then, we use traditional methods we're already used to to get a better installation, to have more light-weighted constructions to have new shapes.

One example I want to look into a little bit deeper, is Dubai. Dubai is a very innovative city. And the government is really looking into new technologies. And this year, they announced that they want to have a new 3D printing strategy. And that means by 2025, they want to have 25% of the building structure of each new building printed.

That is not so far away from today. It's only seven years. And to reach that goal, they are investing in the new technologies now. They are teaching-- they are starting to teach additive manufacturing in schools. And they are not only looking into constructions, it's also about medical applications, for example.

But the main reason is Dubai made the first printed office structure. And they ran into some issues. So because there were no standards and it was really a research project, they had to completely rethink about the whole structure, about the whole design.

They were working with a Chinese company. And this Chinese company was printing the stuff in China because they have this safe environment. They can print everything in the factory. It is controlled. They can print more pieces faster, use less material, and ship it afterwards. But what came out in the end is that the shipping costs were higher than the whole building. It was more complex to ship that stuff instead of building it on site.

We are having some technical issues, so I need to jump out each time I'm showing a video. I'm sorry.

So they used a gantry system to print these structures. And what they did here is, if you look into the structure, itself, they used the shape to increase the insulation for the walls to use less energy.

And we're seeing that a lot in architecture. That it's not only about having 3D printed pieces, saving costs, saving materials, it's also about bringing functionality into your building. So we're talking about having your wires maybe already integrated. So there are a lot of future projects coming up where people are looking, how can we increase the insulation? How can we make our buildings more environmentally friendly? How can we integrate tools, like water pipes and all that kind of stuff, already into the building?

So if we look into architecture, 3D printing can be really large. But if we look into the manufacturing industry, we are scaling it a little bit down because we have a lot more complex materials. We have a lot of material combinations. And the applications are different.

So this is an example from our team in the UK. They work with the Port of Rotterdam and the RAMLAB, which is a research institute of the Port of Rotterdam. And they were trying to decrease costs, not so much about the additive manufacturing process, itself, but storing costs for all these parts, all the spare parts, are really high. You need a lot of space. And 30% of these parts are never used.

So they were trying to build this stuff on site when it's needed. And this propeller, for example, will be tested on Monday. And it's going out on the water the first time. And also, here, they are starting to define regulations and set standards.

But we can also go and integrate design into our printing. So we can use, for example, if you look at the chair, we can use support structure and integrate it into our design so that we don't need to remove the support structure.

And then if we look into polymers, we are having machines that are big as apartments. And here, it's about volume. They want to print big pieces. It's for molds and for tooling. The tooling industry can save a lot of money by using polymer printed and machine toolings instead of having metal toolings, especially when it's just for prototypes.

So these pieces are printed. And if you have a cubic meter, it's like half an hour to print that piece. And afterwards, you can machine it without any problems. It is open-source because it is a pretty simple extruder. You can use almost every material. You can use composites but you can also use every normal polymer you're using in your daily life. So you are not fixed to materials anymore-- what you often have when you use FDM process.

And this is an example we did together with the University of Milano in Italy. We used continuous fibers, and we printed these continuous fibers in one piece. Because of this, we could save around one kilo of weight for that BMX. We also used some other tools like Autodesk generative design to have customized brackets for that bike, to have less single pieces on that bike and integrate functionalities into just one part.

So what we did in that project is we redefined the shape of the bike. You're only having the forces you know and we're using a voxel-based technology to simulate our stress directions. In the end, we get the result of tension and compression and we could align our tool pass to these directions to make our bike more stiff, but also more light.

And we used the robot to print these pieces because we needed a lot of freedom. We needed an intelligent way of printing it so that we don't have to change direction too much and to cut the fibers very often. And to print that whole frame took us about one week, but only because it was the first ever printed bike in this way. But if you know how to do it and if you want to repeat it, you can print that bike in one day.

And we used endless fibers, so continuous fibers. We coated them with a liquid resin and then we used a laser to harden the resins. And because of this, you can do some big bridgings. You can do almost 3D tool passes with this system. In this case, we cleaned it up afterwards, but normally you would put a coating on it. But we wanted to have the original shape so that people can see what we did.

But you can put a foam on it. You can color it. You can customize your bike for every need. You can customize it for your weight, how tall you are, what you want to do with the bike. So if you look further, for example, if you have a mountain bike, you could use these fiber systems and use it as your suspension. Or you can make a racing bike more aerodynamic.

I brought a simple piece of that bike. Please don't touch it inside, but on the outside, it's clean so you won't get any fibers on your fingers. So that is a piece. It was one of the first test parts we printed. And you can park a car on it without any problems. Yeah, so we assembled that bike. And it was more a showcase to tell people what kind of technology we have out there and what unique shapes we can get out of it.

And because of this, I want to look a little bit more into the system. So if any one of you want to start and print large-scale additive manufactured pieces, you should ask yourself and understand your application. Is it for architecture? Is it for aerospace? Is it for the automotive industry? Is it maybe something for biology or medicine?

And you should also know your requirements. So do you focus on costs? Is maybe weight reduction your main goal? What temperatures are you dealing with? Or do you need water resistance for your system? So these are questions you should ask yourself every time you're doing one of these things because we don't have an out of the box solution. There are many solutions for every application.

And if you're sure about that, you should ask yourself, OK, which system should I use? Do I want to stay with a robot? Do I want to have the freedom of a robot? It is always a good starting point because a robot is relatively cheap, and you have a lot of freedom to play with it. Or do I really want to use a gantry system, because I already have a gantry system and I can customize it and put an extruder on it. Or do I want to have a hybrid machine, where I have additive and substractive in the same machine?

And you should, of course, ask yourself, what kind of material am I working with? Is it metals? Am I working with polymers? Do I have, maybe, multi-material systems? And if you know all these facts, you can start looking into the software. And I would recommend to have a long and deep look into what kind of tools you want to use. Because it can be different from polymers to metals but if you want, you can always stay with the same system. It is on you what you want to use and what is working best for you.

So I'm always using somehow Netfabb because I'm from the Netfabb team. I like that tool. And it gives us a lot of opportunities. We can simulate printing processes in metal. We can customize tool passes. We can use generative design to have unique shapes. And we can bring in lattice structures. We can optimize the topology. So we have a lot of small pieces in Netfabb we can use. And-- I need to get out again, I'm sorry.

What we also have is we're having simulation in our system. So now we can simulate large-scale metal printings before we print. That means we can save time and we can save costs because instead of doing it 100 times, we may only can do it three times untested because we can simulate it before. And if we see an issue in our simulation, we can change the tool pass, for example, to reduce the heat in one area. Or we can change the material if we see that it's maybe not the right material to use. We can also understand, where are the stress points in my material?

But we can combine tools, so we don't have to stay in the Netfabb software just because it's called additive software. We can use Fusion for the design. We can use PowerShape to prepare our part before we print it. We always need an offset if we print it because you have to machine it afterwards. And then we can have Polymer to do the tool pathing, to do the machine control, to have the process simulation, to avoid collisions when I'm using a robot, for example.

And the Netfabb team and the Polymer team were working together the last months and created a channel between the systems. So now, we have pieces of Netfabb in Polymer. And we can do additive in Polymer. That means if you have-- use Polymer for substractive, you can now stay in your system and use it also for additive applications. Or if you print a part, you can prepare it in Polymer, you can machine it afterwards because you have your part already in there. And here, it doesn't matter if it's a robot, or if it's a kind of a gantry system, you can have more than one machine in your system and load them all in the same project.

The Polymer team was working on many applications. And they figured out that these are the most, or the often asked applications in the industry. So it's about coating. It's about surfaces, or having features on cylindric shapes. And this is what Polymer is awesome to use for. It helps you. It is easy. And it is good to understand. So if you use Polymer, you can do all these applications. Of course, you need all the training for it, and it takes some time to learn it, but it is possible for everyone.

But if we look into the bike, for example, because we customized the tool paths. We put a kind of intelligence into the tool path so that we don't destroy our fibers or that we don't have to cut them as often as we normally do.

We switched to Dynamo because in Dynamo we could customize all of that. And we can also have a robot control in Dynamo. So it is a tool where you need to understand logics, but you don't really need to program your system. We translated all the machine language in a way that, for you, it's just I want my robot to go left. So then you can tell the robot in Dynamo that it should go left. And then you can load it in any robot application. Often they come with KUKA or ABB where you can simulate what your robot is doing afterwards.

So we have a similar application we can do in Polymers and Dynamo. And what we also did here is we can have a real-time control. That means if you see, for example, you print metal and your layer is not as high as you calculated, you can adjust your system. You can tell your robot please go one millimeter down and adjust new.

Anyway, using Dynamo for also big data sets. Because of having all this information in your tool paths, especially with complex systems-- so where do I need my laser on? Where do I need my laser off? Where do I need to cut my fibers? Where do I need to go slower around the corner? Where can I go faster? The data system gets really, really big. And Dynamo is a powerful tool to handle all of this. We can also do 3D tool pathing in there.

So I ask myself, very often, which kind of tools I should use for additive applications. And I put a slide together with all the tools I ever use for additive manufacturing. So it is a lot. I know that Autodesk is having so many tools. And it's really hard to understand all these tools. But I would recommend if you start a new system, and you are not forced to use any software, then check it out what is out there. Because you can combine tools. We created a lot of channels between tools and there are a lot more coming, that you can load files from one platform to another one. And you can customize your complete system and your complete software workflow with it.

So it was only an overall overview because this topic is complex. It's big. It has so many applications. So if you start, ask yourself, always, the question, what is my application? What requirements do I have? Which system do I want to use and which materials I'm using? And then get started with the software. Thank you.

[APPLAUSE]

So I'm hoping for questions, if anyone has a question. Yeah?

AUDIENCE: Could you-- going back to your last point, could you summarize the differences between Netfabb, generative design, and Dynamo, and how you might use those tools.

DOMINIQUE MUELLER: Yes. So Netfabb is the overall tool I use for additive because you can do the tool pathing in there. You can optimize your tool.

And Autodesk generative design is really for these unique shapes you see. So if you want to have a complete new shape and you don't know exactly what you want to get out. So you don't want to design your part yourself, then it's a great tool to go for because it gives you ideas. You can say, I want to have the highest stiffness or you can focus on the cost and that kind of stuff, and the system gives you more opportunities you can choose from. So in the end, you just choose the design you like the most.

And with Dynamo, we are more doing the whole control of the system. So controlling the robot.

But the whole preparation we did in Netfabb. So we created the tool paths in Netfabb. And then we load all this information we got from the tool paths in Dynamo. And then we adjusted that and translated it in the machine code we need so that we can communicate with the robot.

But theoretically, you can do your tool paths also in Dynamo. But for me, I'm used to Netfabb. I like it. I'm using Netfabb.

Any other questions? OK, then-- yep?

AUDIENCE: Could you just go into a little bit-- so you mentioned that in Polymer there's now some Netfabb in Polymer?

DOMINIQUE MUELLER: Yes.

AUDIENCE: Could you just address that a little bit more?

DOMINIQUE MUELLER: So we have some of the tool path strategies we had in Netfabb. And we now have them integrated in Polymer. So you don't need the Netfabb software for it. It is pieces taken from Netfabb and integrated in Polymer. And you also have the opportunity to use the simulation tool of Netfabb in Polymer now.

AUDIENCE: Inside Polymer. So one of your slides, you have all those different software tools you're using, but if you were starting somewhere from scratch, would Polymer be a good place to start? Or what additional skills or software do you recommend?

DOMINIQUE MUELLER: I would say it is a good place to start because it is a really sustainable software. It takes a lot of training to get into it. But if you know how to use it, I would say it is the easiest way to start. Yep?

AUDIENCE: So some of the tool path control that's in Polymer sounds like it's the same type of control that's in Dynamo. Because you can do tool path editing in Polymer that changes-- it makes those small changes, the speed up [INAUDIBLE].

DOMINIQUE MUELLER: Yes.

AUDIENCE: So would you say that Dynamo's not needed if you have Polymer, then?

DOMINIQUE MUELLER: It's just different approaches. You can do the same stuff sometimes, but in Dynamo you have a lot more freedom because you can program the stuff yourself. And Polymer is more for the end user who doesn't want to program at all.

And I also realize that, for example, designers they want to go with Dynamo because they're used to that stuff. But someone from manufacturing want to go with Polymer because they know, I load my system, I load my project, and it works. So yes, you have sometimes the same applications in the same system, but this is where we give our customer the freedom to choose.

AUDIENCE: Maybe it'd be kind of nice if you had a summary sheet for every one of those software packages that you showed there in the video. [INAUDIBLE] process going. Materials here, here, here. Just you know, a paragraph under each one of those things [INAUDIBLE].

DOMINIQUE MUELLER: Yeah, I hope I can address that to the right people. OK, then, thank you very much.

[APPLAUSE]