Nonlinear Simulation in Autodesk Fusion 360

MIKE FIEDLER: I think we'll begin.

First of all, we're on our last day here in the afternoon, and you guys showed up for a nonlinear simulation class. That's commendable right there, so you guys are our people, right? That's awesome.

So this class-- hopefully everybody saw the slide by now, and hopefully you're in the right place. This is Nonlinear Simulation in Autodesk Fusion 360. My name is Mike Fiedler. I'm one of the support specialists with Autodesk. So outside of this, I'm back in the office helping out people with simulation. And that could be Fusion, it could be InCAD. My background is simulation mechanical originally, so with that product for quite a few years.

But that's a little bit about me, and with me is Lee Taylor.

LEE TAYLOR: I'm Lee Taylor. I'm the developer of the explicit technology that is event sim in the Fusion product, and I'll be talking about some sneak preview of some of our technology towards the end of this talk.

MIKE FIEDLER: All right. Good thing you mentioned it. That leads us to our safe harbor. If you haven't seen one of these slides, we are most definitely going to show you-- talk to you about some things that are planned to come into the software-- not everything, obviously. Why would we give a class on something that nothing's going to be there, right. And we hope it's all there, but we will probably talk about some things, and we are definitely going to show some things, that are not yet part of the production level of the software.

So all that says in short is, you know, don't make a purchasing decision based on something that we might say here, anticipating that it's going to come to you next week or the week after, right. It's kind of on the roadmap, but we don't necessarily know the timeline for everything. A lot of this is going to be relevant, though.

When I get to a point-- there's about two different points in the presentation where I can think of that this comes into play, and when I get to that point of the presentation I'll mention it so you know specifically. You don't have to go the whole way through the presentation going, oh, is this in there already, or is this not there.

Next thing is just the welcome. And if anybody was in the class-- there was a class yesterday on an EAGLE and Fusion. Who attended that one? I think one of the things that James Youmatz did there that I like, and I also wanted to do, was the why-- like, why are we here in a nonlinear class, on Thursday afternoon after being at AU for so long. And I kind of highlighted in blue what I think-- this of course is the entire description that was here for the class, but I highlighted it in blue what I think is important. We're going to show nonlinear static stress and event simulation, and how we're going to do that is with Autodesk Fusion 360.

And I think that's one of the really key parts of the presentation. For instance, on Monday I was out, and I was talking to a salesman. I won't say whether he was with our company or not, but he asked what the class was. And I said, we're doing nonlinear analysis, and we're doing event simulation from the Fusion perspective. And he said, oh, I didn't know that was in there. And I went, exactly. That's the point.

So people need to know that Fusion 360 isn't necessarily I'm going to bring in a part, do this really simple linear static stress analysis, and that's the extent of it. We've got some really neat tools in it. Event simulation-- preparing for this class, going through a lot of event simulation work, it does some really neat things. So if you haven't used it yet, hopefully after this class you're inclined to test it out, experiment with it, let us know where the limitations are. I'm in product support. [? Mike Harris, ?] the simulation manager for Fusion. So give us feedback. Get my card at the end, and let us know how the experience is for you.

And then lastly, of course, we're going to examine the advanced material models and options that we can and should utilize in the interface. So the presentation is I would say largely four parts. We're going to look at when do we go to a non-linear analysis. And then because the materials are such a key part, or can be a key part of non-linear analysis, we're going to take a look at the material models that exist, how do we access them, things like that. And then we'll get into examples of non-linear static and event simulation, and then we're going to go into Lee, with some other things with event simulation and quasi-static. So you know where we're headed.

These were the four objectives that we outlined for the class-- limitations of linear static stress, three primary types of nonlinearity. Become-- this is basically what I just said. Become familiar with the advanced material models and how to set up the analyses. I was ahead in my head.

All right, so starting with objective one, limitations of linear static stress, and when to employ non-linear analysis. And I should preface this with I'm not trying to bash linear static stress analysis by any means. As I was typing up these slides I thought, wow, that really sounds really negative. You know, I keep going back through this is when you use nonlinear, and this is when you use event simulation. It's really more to give you some key points about do I have one of these, or more-- one or more of these components within my analysis. And presuming so, then I should probably consider having a non-linear analysis, or performing an event simulation.

So there's the list right there. Deformations, small; strain and rotations, small; changes in stiffness are small; changes in boundary conditions are small; changes in load direction are small; and the material remains in the linear elastic range. I don't know why just read them all for you. I am going to, in the next couple of slides, just give you a couple of really simple examples of kind of things that I've come across over the years. So this type of lift, just as a real world example, in the years that I've been doing product support-- about 18 years, several times I have had-- not this exact brand, but I've had models of lifts or equivalent geometry where people go, oh, great, there's contact in linear static stress, so let me go ahead and use linear static stress to look at whenever we extend this boom out, and I apply a load on the end, and things like that. You don't want to do that with linear static stress. You want to do that with either nonlinear static or event simulation.

So I did it in a nonlinear static, just a really simple little short example, just so you can see that. This of course half symmetry, some sort of box channel. And all it is is fixed at one end, it's got a prescribed displacement on the other and to pull it out, and there was something else I was going to say-- oh, gravity is on. So it's all contact.

And what's your limitation? So you say to yourself, OK, how do I know that I want to use a nonlinear static or event simulation? You could utilize either. How do I know that I want to do that with that type of analysis? Well I always like to-- kind of like the level one is, it displacing or is it deforming, is it bending? If this was just not moving, but I was just putting a load on the end of it, well then I would consider that just a deformation analysis. So sure, it certainly could be done in linear static stress until it gets to a large deformation. But in this case, of course, we have displacement. So I would say any time that you just have bodies sliding along one another, or rotating-- you know, think about-- please don't, but if you had-- you were trying to simulate a nut going onto a threaded bolt, for instance, that would be displacement as opposed to deformation.

The limitation for contact, if it wasn't presented to you, if you were in the non-linear course from the Nastran end yesterday, the limitation in linear static stress is they anticipate movement of about the width of one element. So if two bodies are going to slide relative to one another greater than the distance of an element, then certainly I would consider either a non-linear static or event simulation. OK, make sense?

All right. The second thing, to look at the second point, strains and rotation should be small. And where I had the real world example before, granted I gave you two finite element images here, I was trying to, whenever I was capturing images, I was trying to-- they gave us some way in, whenever you're browsing the web for images, to look at only images that are publicly released. You can use them without permission and all that kind of stuff. And it's surprisingly hard to find a hyperelastic image whenever you do that, you get down to that level of detail.

But anyway, let me go ahead and play that animation. So again, I've noted down here, this is another analysis that was completed with non-linear static stress, and it's just a hyperelastic material fixed on the one end, and then there is a prescribed displacement on the other end. And you can see that we get quite a lot of deformation, and the strains are quite large. And I think it's really hard to see the numbers there. I apologize for that.

But it's, if I recall correctly, maybe around 15% strain. So certain materials and analyses such as hyperelastics experience large strain. And I kind of break that out into two different things. I said, certain materials and certain analyses. Certainly almost any time that you get into hyperelastic material, you're going to get into large strain, right. That's kind of what those material models are geared for, where you can get into I don't know, 20% strain, 100% strain. I think that the Mooney-Rivlin model is good up around 100% to 200% strain. And then of course you have other analyses that might not be a hyperelastic but still experience large strain. You could get into an analysis where you exceed the yield of a metal, so not necessarily a hyperelastic, but still could be a large strain type of analysis.

And then this point-- finally I found an image here with something hyperelastic and something not, but changes in the boundary condition-- excuse me, changes in the stiffness throughout the model should be small. So if you were doing this type of analysis that combines a metal and a hyperelastic material, that's generally one that's going to cause the program some issues. So at the minimum I would say you're going to get some warnings about changes and stiffness. It's looking at the stiffness of the elements in the model, and when it sees a drastic difference between the steel and the rubber in a linear static stress analysis, at a minimum it's going to give you a warning, if not fail the analysis with an abrupt change in stiffness type of warning.

I don't know if it does detect the difference between that being the material-related or shape-related, but that's where one of the problems might be. We do those checks for element quality. So if all the elements were all the exact same size, and shape, and material, they would all have basically the same stiffness. You get some really long skewed elements in there, and then it has a much drastically different stiffnesses. So it's looking for that kind of stuff at a linear static stress, one of the reasons that linear static stress doesn't like that sort of thing.

So like I said in the first bullet point there, in assemblies, it is anticipated that the materials have reasonably similar stiffness. So as soon as somebody sends me a model like this, I'm starting to wonder, if it's set up as a linear static stress, is it going to run, for one, or is it going to error out on them, or give them warning messages that they're going to worry about?

And two, I mean, we do have some sort of rubber material there. So is it even proper that he's running it linear static stress? Because if it's linear static stress, then that means he's using a linear elastic material model, which probably doesn't describe that rubber material appropriately, right. And then as I mentioned, in addition, the stiffness is not updated in a linear static stress analysis.

So what the stiffness is at the beginning is the stiffness throughout the analysis, whereas in a nonlinear static or an event simulation, you can see that this little exercise that I did here, this one was done in an event sim. It's an O-ring compress. So basically that's about a 1-inch diameter sphere, steel plate in the bottom, steel plate on the top. It is a quarter of a 3D model. I'm just looking at it from the side here. And if we play that animation, you can see that it compresses that O-ring. I think the displacement on there was a half-inch, so we're really pressing down on it. It runs just fine. That was a great, quick analysis to do.

So again, the event sim or non-linear gives us the material models that we can utilize for that, plus the fact that it's updating the stiffness throughout the analysis. I'm not again strictly stuck on it being a hyperelastic, but again, if you exceeded the yield of the material, since we're updating the stiffnesses as we go through the solution with event sim or non-linear static, that would be a more appropriate solution.

Are you now wondering why you came to this class in the afternoon of your third day? Is it tiring you out? Or are you getting excited about what you can do? Hopefully the latter.

OK, so the next one from the list-- I think there's maybe one or two more. So then we'll get into some good stuff here. Changes in boundary conditions are smaller, one of the other limitations. And by the way, I don't know if you saw on the first slide where I had them all listed out. That came right out of the mechanical event simulation training manuals, where this list that I came from to highlight, just so I can give you some references about where I'm getting some of the information. And you'll see a couple of other slides where I have the information here, so you know I'm just not making it up as I go along.

So we have changes in boundary condition. And this problem, we really have two different locations where that applies. First of all, the top of the model, right. We're going to take this model, we're going to displace it a large amount. So in the first slide where I was showing that lift-- similar thing, right? Large displacement.

And then the second thing that we have is the contact. So as we play the animation, you can see that the model has to deform a good bit in order to be inserted. Again, another really nice, quick analysis to do in event simulation. That was a good one. I like that a lot.

So I had mentioned it earlier that in linear static stress with contact, whenever you have a sliding contact, we expect maybe movement of about the distance of an element, right. And of course this one in that displacement is covering, or sliding by many, many, many elements. So I think that's a pretty good example in a couple of different ways. It probably would have been good if I would have displayed the mesh for you. It has displacement to it, plus the boundary conditions. So is that clear? Good slide?

All right, this is one of the things where we need to go back to the safe harbor statement. So one of the two or three things that I said that I would mention-- changes in the load direction with deformation are small. So in linear static stress, when we apply a force, or we apply a pressure, the direction that we apply it in is the direction that load stays in throughout the entirety of the analysis, right. So I have a very simple model here of a cantilever, and then there's something to bend it around.

So if you were to do this analysis in linear static stress-- there's a mini-axis there. It's really small. But the blue is my z direction. So if I were to say I want to bend this around that mandrel, and I apply a force in the z direction, it's never going to happen in linear static stress, right, because that force is never going to change direction. So it'll bend it down, but again, it's never going to update with the geometry. It doesn't have the steps to be able to update that direction.

So likewise with pressure-- granted, your pressure could be normal to the surface. Pressure doesn't have to be XYZ. You could take a curved surface and apply a pressure, but in linear static stress, that pressure direction always stays in the direction that it was initially oriented due to that surface. In other words, if it goes through some sort of large deformation-- think about a balloon. It starts out flat. So if you start to blow air in it, you'd get most of your pressure up and down. And then as that balloon inflates, now the pressure is kind of radially outward in a circle. That's a non-linear effect. So it's not going to happen in linear static stress. You would have to do that in a nonlinear static or an event simulation.

So with this one what I did was a following pressure on the tip of that cantilever, and you can see that as it goes through the simulation, that pressure changes direction that stayed normal with that cantilever beam, and wrapped it right around.

This is on the road map. So the reason that I mentioned the safe harbor is, right now the ability for it to follow is not something that is in Fusion, but we're aware that that is something that needs to be added in the future. What else was I going to say about that? I don't know if there's anything else I was going to say about that. Oh, I know what I was going to say about that. I am in the support world. I mentioned that. All right, so part of my job is to find solutions-- answer questions, but we also find solutions.

So if you run into a problem like that, and you say, yes, I understand I need to use a non-linear analysis, or I need to use an event simulation, but I also think that I need to have a following force or a following pressure, give us a call. Let's do a screenshare, let's look at what you got, because there's other ways to solve that problem potentially. We could maybe use a dye, or maybe a series of dyes that come in and help form it in the directions that you need. so there's always-- shouldn't say always. Most times there is an alternate solution. There's more than one way to set up the problem. That's what I'm saying. So we might be able to accomplish it in a different way.

All right, I think this is the last one, the material remains in the linear elastic range. In linear static stress, and that is to say that the slope is purely defined by the modulus of elasticity, you can certainly exceed the order of the material, but the values that you arrive that are going to be defined by what the stress strain curve looks like. And for linear static stress, of course, it's just a straight line. And I have a pretty good example of that later which will highlight that to you.

But this was in another event sim, just a little compression of a tube. This is something-- prior to getting into the world of FEA, I worked for a company that did sheet metal hydro-forming, we'd roll some tubes, flat stock-- roll it, weld it, and then put on some nuts onto the thing, and then we'd flare the ends basically so it would capture the components onto the tube. So that was just something that was in my head. But anyway, event simulation analysis, and of course using a non-linear material. So that way we see the permanent defamation. It comes in, exceeds the yield, tip of that tube, we retract our dye, and you can see the permanent deformation just fine. Cool?

All right. So in conclusion of all that, why do we care about non-linear analysis? Well, there are limitations to static stress, and they could be opportunities for us to leverage nonlinear static or event simulation. And I underlined, maybe, because I'm not saying that nonlinear static stress-- I'm a realist. Not in linear static and event simulation is it necessarily going to do every single analysis that you might want to do in the world.

You still have to worry about the size of the problem, for instance, the complexity of it, things like that. But if we know what the bounds of linear static stress are, then we also know when we can go into nonlinear static stress, when we can go into event simulation. It just makes us better analysts. If we know more about the tools, we know when to employ them, than that's just good learning.

And the awesome thing is that Fusion does it. Unlike the salesperson that doesn't know about it, Fusion does do nonlinear static and event simulation.

All right, so our second objective, the three primary types of nonlinearity. Just real quick, generally speaking, there are three types of nonlinearity. Some people break it into four, and I highlighted that. So if you were to look at the Autodesk Nastran InCAD, section 14, Nonlinear Static Analysis, they'll break it down into three, which is material, geometry, and then they lump forces and boundary conditions together. Other places, I've heard it referred to as four, and they just basically separate out forces and boundary conditions.

So these are kind of the coveralls of what makes a problem nonlinear. So it could be any one of these, we'll say four, things-- material, geometry, forces, and boundary conditions. And the slides that I just went through are basically specific examples of things that would fall in at least one of these categories, maybe more.

So if we take a look at that last exercise, or last little video that I played, material nonlinearity. We impacted it until it exceeded the yield, went into permanent deformation, and we saw that. And then we had the one example where I pushed down on that tab and it kind of followed the contour of the other. So that would be something that has a boundary condition nonlinearity, the contact-- large contact. So all those things just fall into one or more of those four buckets.

And then I kind of just broke it down a little bit more for you. And what I would say is, you know, this is my quick indicator. So when somebody sends me an analysis, and I'm taking a look at it, you know, these are kind of the things that I scan through in that person's setup and their analysis. And I say, OK, well, let me see what his materials are. Is there going to be plasticity? If the guy tells me that I'm going to ram it, I want to see it break. I want to bend it all around. And then I see that he's using a linear static stress. I say, well, maybe we're going to have some plasticity, especially if we're going to permanently deform it. Do we need the nonlinear stress-strain curve?

And then I take a look at what he has going on in the analysis. Is it a prescribed displacement? Is he trying to rotate the part? Is it large displacement? And if it is, then again, that lets me know it should probably be a nonlinear static or event simulation. And then lastly we get into the forces and boundary conditions. So is it a following or dependent force, or is there a lot of contact interaction?

So honest to goodness, that's a simple list to kind of keep in mind as you're setting up an analysis, and say to yourself, do any of these conditions exist, and if so, then would I get a better solution if I use a nonlinear static or event sim? You've probably all seen that 100,000 times if you've been to 100,000 nonlinear classes. These are not specific to Fusion 360. These are specific to non-linear analysis, or I should say, general to nonlinear analysis, I suppose.

All right, so one of the big things with nonlinear analysis is the ability to leverage nonlinear materials. I don't know what the beep was. I thought my laptop was dying.

So we're going to spend a little bit of time looking at the advanced material models that you have currently available within Autodesk Fusion 360. And the first slide is just kind of general here. There are currently five different material models for you to utilize, the very first one being the linear isotropic material model, and then you have a hyperelastic Mooney-Rivlin for hyperelastic material models. You have a non-linear elastic isotropic. Anybody use nonlinear in Nastran InCAD or Nastran, by the way? OK, yeah, good.

All right, so you guys have seen these material models before-- a lot of sharing going on between the three programs here. Then there's a non-linear plastic isotopic and a nonlinear elasto-plastic or bi-linear, and we're going to expand upon these briefly. So first of all, there is a linear isotropic, and that is your default material model. The stress-strain relationship has to be linear, or it has a slope defined by the modulus of elasticity, and the analysis should remain in the linear portion of the curve. It doesn't have to, but it just means that the results that you get past the order of the material are going to be a little bit spurious. You know that you have results that have exceeded the yield of the material.

And again, I have an example that I will show you with that. Again, I would just emphasize a little bit that it is the default.

A number of times, especially if you're just getting started-- this might make sense if you've explored nonlinear in some of the other programs, but one of the pitfalls that I've come across over the years is as soon as people switch to a nonlinear analysis or an event simulation they go, great, I'm in the nonlinear world. So let me go ahead and take that cantilever beam, and I wrap it around the mandrel, and they expect to see permanent deformation, and it doesn't happen. They unload it, and next thing you know it's right back in its initial condition. And the reason for that is because the default material model is a linear isotropic, so it doesn't know that it has exceeded the yield.

And then we have a hyperelastic Mooney-Rivlin in here. It's a two-constant standard Mooney-Rivlin that you can utilize for hyperelastic materials-- not sure why I'm going over here. I thought I had to push my mouse, but I got this handy clicker.

The other three material models that you have-- isotropic nonlinear elastic. And you can see with that one that you have the curve that you can define in both tension and compression. So for a nonlinear elastic material they will spring back elastically. There is no plastic region to that particular material model, so no permanent deformation in it. And you get to enter-- I should say, you get to enter. You need to enter in the full stress-strain data in the form of a table in order to define or utilize that material model. And by the way, we are going to show you a little bit of this inside the software, so it's not all going to be words. We'll show you what some of these look like.

And then we get into the last two, and the last two are what really gets me excited. You know when you spring out of bed in the morning, and you're like, yee-haw, I get to go to work and do some simulation. This is where the stuff gets really kind of neat. In the isotropic nonlinear plastic, you get to enter in the stress-strain data post-yield.

So if you look at that little graph or image in the middle there, you can see that the yield is defined, and then it has kind of a smooth curve to it, right. So you'll define with that yield is, and you'll find the stress-strain data post the yield. So that way as you're going through the analysis, it knows, OK, this region of the model, or these elements in the model have exceeded the yield. I need to start to do something different.

And then you have your isotropic nonlinear elasto-plastic bi-linear. And you can see there, instead of a nice smooth curve, you basically have two curves, a bi-linear. So you're defining the modulus of elasticity, and you're defining the tangent modulus, and then the yield of course, and the program knows what to do from there.

Now, should I go into that? Yes, I will go into something-- a little bit of an aside. In the world of simulation mechanical, I always taught, leverage this first, because this is the simpler of the two. You can find the tangent modulus. You can find the modulus elasticity everywhere. Random people walking down the street-- no, OK, maybe not that far, but-- so I would always say, start here. This is the simpler of the two material models. And that's going to give you some idea about how that model runs. And then from there, if you need to get more accurate data, then move on to a nonlinear plastic analysis with all that stress-strain data in there, because you're going to get more accurate results then. You could bring in that data from test results.

But not very long ago, I just heard that kind of-- and remember, this is based on what we're getting from Nastran, or a lot of the solution is based on Nastran software. The recommendation via Nastran is really, when you can, to utilize the isotropic nonlinear plastic, because this transition is a little bit more difficult, to go from just one slope directly to another slope. So they actually encourage use of the nonlinear plastic over the elasto-plastic bi-linear to help convergence.

So to be honest with you, I haven't gone through a million and one examples to tell you that that is fact, but it is what their documentation recommends, so I would just maybe follow the same.

OK, so what about materials? We have all these different material models. You know that you can now leverage a nonlinear static, or you can leverage an event simulation. You're ready to get going. What are you going to do when you get in there, you bring your part in, and you're ready to start setting it up.

There is a built-in library with non-linear materials in it. I don't know that everybody's aware. Currently it has 12 different materials in it. There's three aluminum, there's two copper, there's five steel, there's two titanium, but it is located in a different little section of your menu pull-down. So whenever you go to define the material, make sure you drop that down, and right where the red arrow is, you can see there is a Fusion 360 nonlinear material library. That is where you will access those different those 12 different materials, and then of course build your own. I don't know, I suppose there's a possibility that those 12 materials cover everything that you will ever need to utilize. Chances are that they won't. So in that case, you're going to select Manage Physical Materials, and then you can enter in your own material data.

I am going to expand upon this a little bit, because in my own experience, I think coming from something like Simulation Mechanical into Fusion, getting my own material properties--

[DISTANT APPLAUSE]

--they applauded my transition. Getting my own material properties into the program was something that I thought was maybe a little bit challenging. And then just recently I was working with a user who was doing the same. One of our users, he's been using Simulation Mechanical for quite a few years, and that was also one of his feedbacks about it, was it was maybe a little bit maybe not so clear.

So what I'm going to do is I have this demonstration right here, and I'm going to play through it, and I'll talk about it. Basically I already have my geometry inside the program, and we'll take a look at A, how do we access the other nonlinear materials that are in there, and then how do we define our own. So it'll go the whole way through the process. I'm going to talk about it as we go through.

So basically to assign a material, you go to Material, go to Study Materials. You can see that it's a default steel to begin with, and if you expand Properties, and highlight the steel, then you can see the linear material properties. That's what you'd normally see, the Young's modulus, Poisson's ratio. It does have a yield strength and UTS there, but those are only used whenever you're using a linear material model in the context of a safety factor. So there you can see, if you go to the pull-down, and select the Fusion nonlinear material, and then I go to Assign and the material, there is the aluminum, the copper, the five steels, two titaniums.

Now if everything you design is one of those 12, great, but that's not always the case. But you notice, whenever you select a nonlinear there, it tells you it's using a nonlinear. Do you want to view the properties? And I said yes, so I had chosen a copper, and I'm going over, and I'm saying edit. And now I'm going to go over and look at the physical properties, and that right there is your clue that it's a nonlinear, the checkbox on Advanced Properties. And you can see that it's an isotropic nonlinear plastic, so it has the stress-strain data post the yield point of the material. So that's one of the standard library materials that's in there. So you can just open it up, review the properties are there, know what's there.

Let's define our own material. So I'm going to go to Material, and I'm going to access Manage Physical Materials. All I'm going to do is go to the Fusion library, the metals that exist there. I'll just take any old material that's in there. Right-click, add to my favorites. Once I add it to my favorites-- again, I'm just starting with a basic material that's in there. I'm going to rename it. I'm going to make it my own.

So I always just-- pretty much they're always aluminum. I Right-click the aluminum. I copy it up to my favorites. I rename it. Be careful. Once you rename it, it re-sequences itself in the list. And then I just went over to the right, to the edit, so I can start to edit the name or description. So I'm just giving myself a little bit of a reminder, when I look at this material later, what the heck is that material? Why do I have this in my library?

And now we can go over to Physical, and notice that the checkbox is not there under Advanced Properties. So right now it's just a linear elastic material. But I check the box for Advanced Properties, and now as soon as you do that, this is where you get into those five different material models. It's isotropic linear to begin with, but if you toggle on isotropic, you can change from isotropic to hyperelastic, and there is your two constant, Mooney Rivlin, A01, A10. D1's are your input. Changing it back to isotropic, looking at the behavior, so there's the three others. We have linear, and then temperature. There's your types right there-- elastic. I'm looking at it here. I'll change that in a moment.

So there's your stress-strain input for the elastic one. There is the curve input. If I change from a nonlinear elastic to nonlinear plastic, that's the one like we had just looked at, you enter in your stress-strain data points. Again, it's starting with 00, and then the yield of the material. Here I'm showing that you can import it. So if you have stress-strain data, just chop everything off below the yield, and then just import the portion of the curve post the yield. And then there's the elasto-plastic bi-linear tangent modulus and yield stress. So that's all the inputs that that one requires. And I'm going to go ahead and utilize that one. So I'll just put in some value so we can recognize it-- 3.984.

All right, OK, so I have my material there. The last step I'm going to do with it in defining my own is create a new library. So I'm going to create a new library somewhere on my machine. So that way I can access my own library, put my own materials in that library-- whoops, forgot about the mic. So just giving the library a name, and then I'll go back up to the Favorites. When I grab my material and favorites, there it as. As soon as I can find it, Right-click, and I'm just going to say add it to my other library. So that took all of, I believe that video is about five minutes. That's all the steps. And of course, I was going kind of slowly so that I could show that.

And then the only thing I'm going to do here now is go to the material pull-down, and see if we can access it. So from the material library dropdown, since I made my own library-- remember, I named it my AU library. I'll select that library, and there's my AU example material. Again, when you click on Properties, if it's a nonlinear material, you're going to see that, but you can click that button so that you can review the material properties.

All right, so now I'm just checking it because I'm kind of paranoid. There's the advanced properties, and there's the value that I had edited. All right. Make sense? You got a question?

AUDIENCE: I just want to add one point to there. There's a subtle thing about materials in Fusion that Mike hasn't pointed out. Well, he did point it out, but it's subtle, and [INAUDIBLE]. We are always requiring you to pick the existing material from the library and edit it. Aluminum from the library, put it into his favorites, edited the Advanced Properties. What we didn't do is said, well, what if this was some other material being used? Maybe we went out and picked maybe a plastic, some special grade plastic, and he was lazy and picked aluminum because he just wanted to make it first on the list, and then just go off properties.

If you do that, there is-- there is not only your properties, but remember, there is also the basic mechanical properties. So if you're creating something, that is slightly different from the base materials that you've selected, you should also do the basic property. And if you are going to use that, like for your struts for analysis, if you're not expecting to put steel, just make sure that you've got the right properties on both the basic animal and the advanced animal. Because if you're at the linear studies, it will use the basic properties, and if you're in the nonlinear, you'll be using the advanced panel.

So you want to make sure that that's clear. Because as Mike said, we've gotten customers who are trying to work with the Fusion materials, and the input process is maybe not as ideal as we would like it today, and maybe not as familiar with what modes you were using, be it mechanical or [INAUDIBLE]. So try to work things better, but I want to make sure that what we show is super-clear as to what their limitations are.

MIKE FIEDLER: Yeah, two good points. You reminded me, one of the models I did, I had just changed than the nonlinear or the plastic properties of it, and then I ran the analysis, and I'm looking at the results of the analysis. And when you get into the results of the analysis it shows you the safety factor by default. And I'm looking at this analysis, and I'm thinking, how did it fail? There's no way that this analysis is failing. And it's basically that it was reading the yield from the linear properties, and that's what it's using to compare as far as the safety factor is concerned. So I had forgotten to adjust those properties. A good point. Thank you.

The second thing that I would mention, and I just walked through that-- we went through that kind of quickly, five minutes. And again, I think it was a little bit-- it's maybe a little bit daunting of a material definition workflow to begin with-- maybe a little. But anyway, if you haven't downloaded the class handout, in the class handout, if you haven't seen it already, it's somewhere between 30 and 40 pages, I think. And all the first steps that I just did there are outlined in the document. So it would be a good reference, the first time you go through it, maybe, just to use that as a guide so that way you have the sequence down.

And then very shortly here we're going to get into setting up a nonlinear static analysis, and setting up an event simulation. And in the handout I also have a download link so you can download the file that you will see in these videos, and then the first step on how to set those up as well. So if you haven't downloaded it, certainly something to look at.

Yes.

AUDIENCE: Is it nonlinear [INAUDIBLE]? Is this available also for the Nastran and then for InCAD? Using an instrument database?

MIKE FIEDLER: Those 12 different materials.

AUDIENCE: Yeah, InCAD does not use the same material libraries.

AUDIENCE: And there's a different structure?

AUDIENCE: Yes, different structure.

AUDIENCE: So we need to define different materials [INAUDIBLE]?

MIKE FIEDLER: Fusion's is-- well.

AUDIENCE: So the one thing that you need-- we should talk about this offline, because I do have an idea about how it might work. So if you invent your library-- not the-- but I think InCAD allows you to key things from the Inventor library. And the Inventor library and the Fusion library are the same file format. So we can try it offline.

AUDIENCE: [INAUDIBLE]

LEE TAYLOR: Those material-- those 12 metals that are in the nonlinear material library that he showed you, I built those and put them into Fusion. They're not in InCAD right now. Yeah.

AUDIENCE: Would it be possible to put a simple warning if you're using [INAUDIBLE] or the wrong base material for the nonlinear, so there's a disconnect [INAUDIBLE] to say the rest of your material doesn't match your advanced material?

AUDIENCE: So we went as far to say if you are doing a linear study-- any of the linear study types, and you're using a material with the advanced panel checks, we're just saying, hey, we noticed that you were using a material with advanced properties in a linear study. Just mentally check that everything is right.

Vice versa, if you're in a nonlinear study, nonlinear static or [INAUDIBLE], if you did what Mike said, you say, oh, I'm in the nonlinear world. I'm just going to pick something. If it doesn't have properties, it will also flag it in the pre-check and say, hey, Mr. User, you're in a nonlinear study. You don't have any advanced properties. Check your material inputs.

AUDIENCE: Well, you were saying where it's like [INAUDIBLE] a different material, turning that into an advanced material. Were you making a point that he chose the wrong basic material, and you got the wrong [INAUDIBLE]. I was just wondering if you [INAUDIBLE]?

AUDIENCE: Yeah, we probably should take a close look at that stuff. Yeah, we'll take a look at that. That's good feedback.