Durable Design Through Fusion 360

PAUL SOHI: Afternoon, everybody. How are you all doing?

AUDIENCE: Good.

PAUL SOHI: Yeah, good? Fed? Feeling good? Energized? You ready for this?

We're going to throw a lot of information at you. So be prepared. So introductions first, we've got Michael Floyd, Sustainable Industry Technology Manager, Michael Aubry, Fusion 360 Evangelist Iconic Projects Lead and all-round amazing mechanical engineer and just a great guy, and myself.

I'm Paul Sohi. I'm also a Fusion 360 Evangelist. And I'm on the iconic projects initiative. I am a product designer. So I'm going to be coming at it from that perspective.

We've got a sustainability angle. And we also have an engineering angle on everything we're talking about today. So we're going to just give you a quick brief overview of our learning outcomes.

We want you to understand what we mean by durability and design. Obviously, that's going to be the critical component here. We're going to give you the wider environmental and sustainability impact of making and designing more durable products.

And then lastly, you're going to learn how to use Fusion 360 simulation environments as part of a design feedback loop to make your products more durable. I'm going to slow down because I'm feeling the effects of the caffeine right now. So I'm a little excitable.

So of course, we're talking about durability. But durability is only one component of durable design and good design. There are two key categories to making good products.

It's durability and repairability. So we might be kind of teaching you how to suck eggs here. But we want to make sure that we understand the difference between the two.

So when we're talking about durable design, we're talking about long-lasting products, where functionality is not compromised. They've got to withstand everyday use. And typically if we're using durable design, we're talking about products that are overengineered to increase the usable lifespan.

From a repairable point of view, we're looking at products that are using accessible components with user access to components, meaning that the user of the product can repair it themselves. We want to use commonly available parts so just to make things as kind of affordable as possible to individuals. And we want to use standard tools in manufacturing. Again, that really helps with the repairability.

It's always really important to remember that durability and repairability are not mutually exclusive. But you need to use both of them as part of feedback loops to create the best products possible.

So with that context in mind, we're going to use this summary to help keep in mind what it is that we're talking about when we say durability today. So just to read this out for you, "the ability of a product to perform its required function over a lengthy period under normal conditions of use without excessive expenditure on maintenance or repair." I hope that makes sense. Yeah? No arguments with that? All right, good.

So when we talk about designing for durability-- so looking at this from a design point of view first-- durability does not equate to tough materials only. That is probably the most common mistake we see. Intelligent design decisions factoring in how a product is used are really critical.

So that's not just the functionality of the product itself, but actually in the wider context of how that thing is being used. So if you think of a smartphone, it's not just the stuff that you're doing with the phone. But I'm throwing it in my pocket. I might drop the thing. All of these things kind of become part of that design process and making the product more durable.

You want to make choices. That ultimately means you want to make choices on how your product is made to make it durable, so things like the manufacturing processes, the material choices. All of these things also help in that.

And then lastly, the most important element of durable design is simulation, prototyping, and then redesign. You can't get there without those feedback loops. So when we're thinking about the materials we use in design, the assumption is often that the hardest and toughest materials are best.

However, in reality, materials that can be sacrificial, softer, or absorbing tend to work better. So if you think about things like drop tests or tough material, anything with low plasticity is not going to do it. You need something that's going to be able to take that brunt.

So balances need to be found in material rigidity and flexibility and to ensure protection and security, as well as maintenance. How you guys doing?

AUDIENCE: Good.

PAUL SOHI: You haven't missed much. So then when it comes to design choices-- protection of mechanical elements-- if you think about a watch, a G-Shock is a fantastic example of a durable product. It's designed specifically to ensure that the components of the watch, the mechanical elements, are well protected.

But ideally, you want to minimize the number of moving parts. So things like hard drives moving from the HDDs now to the SDDs, so your solid state drives, it makes a huge difference to the durability. I cannot tell you the number of times I've lost terabytes of data from dropping external hard drives that were mechanical. And now with solid-state, I just chuck them around.

So ultimately what we're talking about here is durability really does equate a lot to designing for everyday life. And then those feedback loops, so this is going to be the meat of the things we talk about with Fusion 360. There's a bunch of seats in the front right here. Welcome.

So design, simulate, prototype, repeat-- all conveniently possible through Fusion 360. So let's take a case study. The Nintendo 3DS is a great example of exceptionable durable design.

This design has its roots in the Nintendo Game Boy Advance SP, which was released in February 2003. So this design hasn't changed really in over a decade. The design brief set by the then CEO, Satoru Iwata, was that the product has to be able to withstand a drop from a breast pocket. Things like the original Game Boy, even modern products, really can't do that.

So the designers and the engineers came together. And this is ultimately what we ended up with. This design has survived all the way until now. The latest product is still using this design. The clamshell actually came out of that brief.

So the clamshell allowed the most critical elements of the product to be protected. They're material choices. So the screens are not using glass but plastic.

The touchscreen on the bottom there is a resistive rather than capacitive, because they're going to last way longer. And ultimately they also kept their clients in mind, which are young children or adult boys like myself. Just to show you how durable this thing is, this is a drop test from 6 feet.

PRESENTER: It doesn't have the same effect without the sound.

PAUL SOHI: It's just so much better with the sound. Yeah [INAUDIBLE] [BWAAH] Let's try one more time.

PRESENTER: I'm ready. I'm ready.

AUDIENCE: The dog expression sums it up. [BA-DUNH]

PAUL SOHI: There we go. So you can actually see in that slow motion video right there, there's a spring mechanism in the hinge that's designed on the point of impact to close the thing. That also has a secondary function, where it locks the screen into different viewing positions.

The plastic materials used in the shell on the outside have also been considered here. There are elements of this that are designed specifically for drop tests, for things like children dropping them. There's loads of videos on YouTube of kids dropping them on purpose.

And my favorite one is a kid dropping it from five stories down a stairwell. And it survives, still playable. I mean, I wouldn't do that. But you can if you want.

But this is in high contrast to modern devices like smartphones. It's safe to say that a smartphone is one of the most ubiquitous devices of the 21st century. And we expect more and more from these products as time goes on.

But we're doing that at the sacrifice of their durability. So whether that's acceptingly or begrudgingly, these devices are becoming more and more delicate, way less durable, and we kind of treat them as sacred things. I'm sure at some point, everyone in this room may have dropped their phone and there's that freeze frame when you panic of like, is it going to survive the next five seconds? I wanted to see how many people's adrenaline started pumping if I showed them this.

But it wasn't always like that. We actually solved this problem in the mid-'90s. So looking back at old phones, it was, again, that clamshell design.

That's not to say that the clamshell is the only durable design out there. But it's certainly one of the best, using plastic materials, again, no glass in these products. They were great. I mean, who remembers this or my personal favorite one?

But that's not to say that we have to go back to these old designs. That's not to say that we need to just rewind and just do the same things we were doing in the mid-'90s. Samsung actually have just released this prototype product-- there's a couple of videos online you can check out-- where they're taking these modern technologies like flexible OLEDs and combining them with different mechanical components to come up with new concept phones.

They've actually stated that this will be a product that will be coming out soon. And the way that they're looking at durability, not only just in terms of lifespan but also ensuring that this product's going to survive everyday use, the screen is inside of a canister. So when it's not in use, you can protect the most fragile elements of the product. So with all of that context set, I'm going to hand over to Michael Floyd. He's going to talk to you about the sustainable impact of durable design.

MICHAEL FLOYD: Thank you, Paul, everyone. So Paul just did a great job walking us through a working definition of durability, explaining how durability and repairability fit hand in glove as complementary design strategies, talking about how durability supports enhanced user experience, better product performance, and gave us some simple key strategies for designing more durable products. I'm going to explore the environmental benefits of durable design. And then I'm going to pass things on to Mike Aubry, who's going to dive into Fusion, which is I know what you all came for.

So I'm going to be looking at ways to make your design more sustainable by boosting resource productivity. What do we mean by that? Resource productivity, a simple definition, is the value that you get out of something.

It can be of financial value. It can be productivity that you get out of a laptop. It can be pleasure you get out of a toy, anything, divided by the resources that go into it.

In this sense, designing sustainably means getting the most use out of the materials and energy that go into your product. So products go through a basic life cycle, from raw material extraction to material refining, manufacturing, logistics, use, and eventual disposal. And it's important to think of them in this larger context.

There are environmental impacts at every stage of the life cycle. We can't just look at the energy that a product consumes or the greenhouse gases associated with that energy consumption during its use phase. We have to think about the upstream supply chain impacts that contribute to a product's embodied environmental footprint, as well as its use phase, as well as what happens to a product at the end of its useful life.

Energy and waste are generated at each-- or consumed at each phase of the product life cycle. And this is not just about being eco-friendly. Energy and waste cost money. They cost money to businesses, to customers, and at the end of the day, it costs money to dispose of things.

And I'm going to argue that-- I think-- how many of you were in the keynote this morning and saw Andrew, CEO, speak? So he kind of already laid down these messages pretty firmly. So I'm just going to reinforce and build on top of what he said.

Today commodities are cheap. Polymers are cheap. Copper's cheap. These prices fluctuate. But we're not feeling the pain yet.

As we move towards 2050, we're going to be climbing some pretty steep curves that Andrew already identified-- population boom, rapid urbanization, middle class growing by around 56%-- all of that leading to about twice the energy demand globally. The companies that are going to survive and thrive in that context as we climb that curve are going to be the ones who are thinking about better resource productivity along the way. And they're future-proofed and prepared to confront these curves and come out on top because they've developed competitive advantage [INAUDIBLE]

So these changes are going to put additional pressure on the planet. They're going to require us to make better design decisions. Products have an impact on climate change-- not really. I mean it depends on the product-- if you're looking at a combustion engine vehicle, lots of impact in the use phase. A lot of products' impacts are upstream or downstream.

This is a NASA animation of global CO2 emissions over the course of the year. You'll see they're largely concentrated in the northern hemisphere. So the embodied energy, the embodied carbon that goes into a product, has to do, again, with its raw material extraction, processing that material, manufacturing, logistics. For example-- sorry?

AUDIENCE: [INAUDIBLE]

MICHAEL FLOYD: Oh, yeah, yeah. [INAUDIBLE] They're burning that coal. So out of the hundreds of materials that we come into contact with every day, I'll give you one example.

Production of raw aluminum accounts for 1% of all global greenhouse gas emissions. That's massive from one material. So it kind of highlights the importance of some of what Andrew was talking about earlier about material reuse, for example.

From a waste standpoint, virgin aluminum-- again, just to stick with that example-- to produce one kilo of virgin aluminum, you produce 85 kilos of waste. That's a bad ratio, 85 to 1. And just to give you a sense in the larger economy, plastics, metal, and glass, and paper are about half of all municipal waste. The rest is, I think, mostly construction and demo waste.

PRESENTER: Yes.

MICHAEL FLOYD: So let's get back to the example of the mobile phone. The mobile phone doesn't weigh much more than 100 grams. But it has an even worse waste-to-product ratio than virgin aluminum.

A mobile phone-- production of one mobile phone is associated with over 80 kilograms of waste in the supply chain. That's upstream here. We also need to think about downstream implications. At the end of a product's useful life, maybe it gets recycled. Global recycling rates are maybe around 20% if we're being optimistic.

This is a dramatic representation of e-waste by artist Chris Jordan. Fun fact, about 150 million mobile phones are discarded each year in the US. That's over 5 million pounds of copper and 116,000 pounds of silver if we say 20% recycling, right, 80% of which goes to landfill and is buried. That's material value down the drain in the ground again and hard to recover.

So again, I mean, what I'm trying to hearken back to and reinforce is the simple environmental benefit of taking this picture and just simply making it look more like this. If you double the lifespan of your product, half your manufacturing impact effectively goes away. That's akin to an engineering miracle.

So as Paul mentioned, a mobile phone's casing and components of course need to stand up to impacts from the occasional drop, rough treatment. But products also need to stand the test of time in other ways. So there's another dimension of durability that has to do with remaining not just functional but relevant, delightful over time, because my phone may work after eight years.

But I'm not going to be using the clamshell that I had in the mid-'90s just because it's still working. It needs to be relevant. So and again, the companies that thrive as we climb those big curves to 2050 are going to be the ones that take this into account.

And sustainable design has lots of angles. And resource productivity also can be pursued in other ways through component reuse, remanufacture, recycling, and worst-case waste to energy. But durable design is really-- because you can achieve this like, factor-two, factor-three, factor-four benefits from multiplying your product's useful life-- is the first stop along the way to sustainable design, step 1. And I'm going to hand it off to Michael now, who's going to dive into Fusion a little bit and talk about some best practices for mechanical design for durability.

MICHAEL AUBRY: Absolutely.

MICHAEL FLOYD: Take it away, Mike.

MICHAEL AUBRY: Hi, guys. Now it would not be a good design class for engineers in the room if I didn't put up a sweet graph. So are any mechanical engineers in the audience today? What is that?

AUDIENCE: S-N curve.

MICHAEL AUBRY: That's an S-N curve. Yeah, so the beauty of this is one of my favorite classes in college because what it gave me was a set of basic tools to apply for all designs from it. When we talk about durability, you can't talk about durability without just the nature of the-- man, there's a lot of, how many times do you need to drop it before it breaks. You need to start talking about fatigue.

So with that, it's like fatigue is a very complex, nuanced-- people get PhDs in it. There are whole classes on it. It's a fun class.

But again, just to kind of-- it's kind of going down the rabbit hole, per se, on some of these challenges of what is the best theoretical way to approach it. Inevitably people end up creating physical prototypes from it. For those of you who are not mechanical engineers in here, why I love this particular S-N curve is it applies a couple of basic rules we can apply when we're going to create our upfront geometry.

It's stuff just to keep-- just some good designing best practices for all designs regardless of the material and the type of loading it goes for. So what this says-- and I won't make you guys read into this stuff although it is a pretty cool looking graph-- it just says that materials have a published ultimate tensile strength.

So oh my god. I'm pulling on it. And I'm Thor. And I just ripped the thing in half. That would be a [INAUDIBLE] of one.

Every time Thor get-- or I guess Hulk get mad and just starts hitting on that thing, that's a cycle of stuff. And there's different natures of loading. But it just says that, hey, the more times you whack on something, over time you're going to start to see a degradation.

And a general trend with most materials we see is that if it's going to be you're designing for 1,000 cycles of loading-- you're designing for 10,000 all the way up to 10 to the sixth, a million-- this one don't go higher than maybe 0.9 a year, depending on the type of loading, 0.9 of that tensile strength.

And then you're moving on eventually towards the tail end depending on the type of that-- you're really not trying-- you really want it to be super durable over an infinite life? You say, try and stay under a third of this. This is a general tool.

It's very much dependent on specific stresses and stuff, like I gotta Dude, at least you're not like-- screw it. So I think as we start talking about what is durability for us as designers, it's important that we go back and periodically revisit these sort of general trends, because you guys as you design really will dictate what the next generation of products are. And as you can see from the examples that Michael and Paul provided, we have good examples in the past.

But we have a real need for it in the future based on the needs on a global perspective-- being good global citizens-- but then also on a business level, making sure we stay relevant with business models that are more reliant on keeping current hardware current, as opposed to less of a consumptive throwaway. And just wait until next year. Wow, so I wanted to start with this first example here.

From that, I'm going to be showing you some stuff in Fusion around the theme of going back to basics with some of, what does it mean to do durable hardware design. So hopefully I'm preaching to the choir on this stuff. But it's really interesting to see how you can see it manifested in how we design.

The key thing is that hey, first off don't overload your material. I mean, if you want this thing to rock, you probably should understand how it performs and stay well underneath it or about to the level that's appropriate for the amount of cycles you're going to have. The next three-- I got some examples of showing you what they would look like in Fusion-- is those design practices around you making sure if you got a hard edge on something, if that's a load-bearing hard edge, that would be what it-- it's a stress raiser.

So you really should be adding some fillets. The next one is if your-- the cycles are based on the amount of pressure you apply to things. So if you have narrow cross-sections of stuff, you need to really be careful about making sure that you are paying attention to those areas, because those are the areas where you would see a failure.

Third one is stress concentrations in general. If you're pulling stuff, just a single point on something, that's a point load. Distributed loads tend to be more durable.

So let's see what those look like. So what I did is I created just a couple of basic examples here up in Fusion just to kind of get us started here. Let's be Is that shown here?

PRESENTER: No [INAUDIBLE]

MICHAEL AUBRY: Excellent. So as you can see in the one side, we have just an unpleasant red spot. That looks not necessarily menacing.

But what happens there is that's where cracks start to develop. Over time those cracks get bigger. And then you see failure. Yes, sir?

AUDIENCE: Plastic deformation.

MICHAEL AUBRY: The plastic-- yeah, that move turns it inelastic. And it creeps over time. And you're in trouble.

So it's amazing just by putting on just that one fillet, we see rather than a menacing red, you see a much more adaptable blue from it. Compromises have to be made along the way. But this is why we don't just have everything in infinite [INAUDIBLE].

Next one is around does cross-section matter. You can say, basically, hey, the same amount of surface area. One just has a little bit more thickness around where the holes are. Well that's because those cross-sectional areas across-- that's what determines--

PRESENTER: What will determine if it's going to fail or not?

MICHAEL AUBRY: That's an exaggerated thing. It's not made out of Laffy Taffy pulling from it. Third example, showing on what does a stress concentration look like.

And actually there is an interesting-- it's in the session that came up of the nature of how we apply loads. I think when we go back and describe this stuff, I think we can dig into what this thing is. If you have a load that's distributed across a larger area, you can see it pretty-- actually you guys can't see my cursor, can you? Oh, you can. Excellent.

Yeah, so if we're distributing across larger areas, obviously less stress from it, that's important because obviously we're trying to stay under certain thresh stress hold-- stress thresholds. Now let's do some Fusion. So how many in here use Fusion?

OK. How many have used Fusion simulation? How many have not? Excellent. So we have a mixed bag.

So I fear this every time I show sims. So this is one of those things I am trying to show basics of stuff but then also make sure I provide credible tips and tricks for those of you I'm sure have run simulations. So there is going to be I think a candid walk-it-out type of thing. So if you guys have questions, if you just let me finish the thought, let's get those questions asked. And we'll move on from it.

I want to show you three examples. The one is we've been harping on the iPhone, so why not break it? The next thing is there are some really interesting advanced simulation modules within Fusion that I think came out-- they got a little bit of love.

But I imagine that these are the types of things like why you come to Constance and say, hey, what if someone clicked on that? What would it do? So I have a couple of those things I clicked on. And I'm going to tell you what it does.

And then aside from there, I've figured there's some just sort of interesting things within the interface that people have told me throughout. And I just wanted to share with them like, hey, did you know that it can do this.

PRESENTER: Do you want to maximize?

MICHAEL AUBRY: Let's-- all right. So this guy-- we got the iPhone from this. And what I'm doing in this scenario is, oh my god. I have dropped it. And it's terrible.

And I've set up an event simulation to go through this. So this is one of those-- this is currently in preview on it. Has anyone tried event simulation so far? Yeah, OK.

So what this is, you go, no! And it drops. And boom, it shoots up. This is pretty sweet stuff, because one, it gets called a [INAUDIBLE] Yeah.

AUDIENCE: Yeah.

MICHAEL AUBRY: But you can get interesting data from this. So for the one guy in the back who's tried this, did you know that in addition to whatever you got out of that simulation, it only gives you however many times you told it to give it outputs?

But you can always go into this nice transient plot place, and then have it actually go and actually derive along the way what were those points. So you actually-- you'll see this thing move as I go across. And it'll calculate it.

For those of you who haven't seen that before, isn't this neat? We can go across, and throughout the entire simulation from this, go and figure out where are the points of peak stress in this thing. And by the way, isn't that interesting that in this particular simulation-- what is it at?

AUDIENCE: [INAUDIBLE]

MICHAEL AUBRY: Wow, we have some basics in design that is manifesting itself in an advanced simulation. Wow--

[BUMP]

--applied learning and a bump. So this is an interesting one. I want to give you some of the basics of how to set these things up.

I guess first off, do you guys care about event simulation? Who doesn't? Nobody's flipping me [INAUDIBLE] Let me give you just a couple of-- if you want to try this at home and I hope you do, I want to show you just some of the things I did to kind of set this up for success. And I had some things that have worked. And I had some things that didn't work.

It's so awesome to see Mike smile here. I can tell him after the class what didn't work. But I'm going to tell you how I was successful.

So the way this guy is set up-- it's a way you set up an event simulation like this-- is you go for a minute. And you guys will see that it's actually in preview. So I'll go [INAUDIBLE] And then the whole way simulation is set up, if you guys notice, just the UX I think is really sleekly laid out. It goes left to right. So when you start off there and say like, we need to make our geometry right, make material selection, go apply what's fixed, choose the loads-- then we're dealing with things-- context, mesh.

So it's like-- it's a left-to-right process is the best practice, how we deal with it. The first thing you want to do with this stuff is really kind of set it up so it's-- so you're removing all the extra materials in this. So this is-- can I just clone something?

Let me go back to my-- actually I want to clone this one just to show you some of the stuff I do. You want to minimize the amount of details that don't matter. And I had [INAUDIBLE].

I loaded my simplified version of this so I already did the great part. I'll do it on a different part. Bear with me a second, guys.

Let's load you. No. What? Yeah, this guy.

So this model that was provided originally when Paul approached me-- he said, hey, Mike, I want to break an iPhone. I was like, oh you mean like an idealized thing? He's like no, I got a model off the internet. Can we prepare it?

And so I was like, OK, great. I'm going to take somebody else's geometry. And I'm going to take this. And we're going to mash up all these little things.

We're just going to hit Go on it, because that's what we do. We're going to leave the iPhone. We're going to leave the Apple.

And I tried it just to pull it out there. But realize when event simulations go on, it's very computationally intensive. It's one of the awesome things about Fusion.

You can throw it in the cloud. And you can tell the Amazon server to do it. But there's a smart way to do it. I mean, there's the showman way to do it. And then there's the way to actually get it done.

So let me show you. If you guys have not seen in simulation how we prepare to get things ready in simulation, this simplified tool rocks. So showing you some of this stuff, it's basically all the stuff we have in our direct modeling area, all in kind of just a couple of different spaces.

So as I'm going to go and just kind of start and get rid of these things-- cool tip is like so I don't need these buttons, so I can take these things-- just hit the E button. And I can just kind of quickly start moving away. If you click on the face, it'll just go and pull it away. And we're moving around.

If you guys have played around [INAUDIBLE] back side, you can go and it'll heal things like faces. So if I go and just choose these edges right here-- if you just delete, you can start deleting them. So there's a lot of really good healing tools that are available with this stuff. Actually, I can even go in [INAUDIBLE]

Other stuff we can do is there are whole Remove Features sets some of this stuff. I can go in there and just say, hey, select stuff for this whole body. Let's just go and find different things.

Like can we get rid of certain things like fillets, holes, chamfers? And you give it basically a feature size from which you can choose from. So there's a lot of things you can just automate up front.

Oh, what the hell? Let's try it. See how it goes. We can go and fix the other stuff that it doesn't.

So it will eliminate things like the tiny chamfers and stuff. So by using these different tools from this space, other things replace repremise probably don't need it for this one, although we probably could just simulate it as a big block. Just want to make sure you guys knew that was there.

Some of the more anal-retentive people on the simulation sites-- especially people are coming from historical tools-- will want to actually have their own separate simulation model. In some cases, that may be the case. And what's so awesome about this simplification environment is it does create a different instance. So you don't have to worry about screwing up your actual model. You'll have this dedicated position of where the simplified model is present to do it.

So I just want to show you just a couple of simplifying tools. I'll get back to the one that's more idealized for simulation. How am I doing on pacing? Everyone tells me I talk too fast. We OK? Am I OK, guys?

OK. All right, so next up on how I set up the simulation is when you think-- you might notice that, oh my gosh, if you're going to hit the ground, why not create something that's super big? Well I pulled this in just from trying to only keep the mesh that you need, because you can improve your times on it. The next thing on the materials-- choosing it.

This one I kept things simple and just said let's make it as brittle as possible, because I just want to see where the peak stresses are. This is an important context for how you use simulation. There is the simulation to get the answer. And there's simulation to get the upfront understanding of how the stresses is distributed in its upfront design.

So it's awesome. Just by making the whole thing really brittle, we can be just really harsh and critical about where those stresses are and ultimately look at where they are. And we can make our upfront decisions without having to really worry about, oh, I don't know if I totally trust this, because you can make a couple other design situations.

And as you begin to better understand your specific simulation type, then you can, over time, start being more nuanced and looking for those specific cases. The simulation is extremely accurate. But you have to know within the context of how you set up your simulation.

So in this case, I'm looking for trends. I'm just doing this glass. Does anyone disagree with that? [INAUDIBLE] He hasn't even rolled his eyes once. So I think we're OK.

[LAUGHTER]

So I set this thing up with saying, we've got a mixture of aluminum, steel, and glass. The next thing we go with this stuff is how I constrained it for this, is I put my rigid constraint on the bottom of it. And then I turned on gravity in the top.

One of the things you'll notice is this is a drop test. But you're like, man, that really didn't drop a whole lot. Well why not?

Well it's because of the way event simulation operates is it is actually taking these little micro-moments. The entire simulation I ran here is-- oh, wait if I go to this guy-- is I'm running this for a hundredth of a second. So to drop something that I can just go and do my 1/2mv squared equals MGH calculations on this-- instead of having to take that entire simulation time, I was able to save a bunch of time just by doing the quick calculation of saying, how fast is this going to be when it hits.

And I applied it under my load cases as a initial linear velocity. So we got this thing. Basically, it's 2 meters per second when the thing whacks in. So as a way to speed up the simulation and ensure that it works.

Some of the other stuff we're doing to make this guy work-- so [INAUDIBLE] loads on the contacts to make this thing set is in these global con-- not global contacts-- if I go to my Manage Contacts, the way you read this particular table is there are three things in here.

I have it set up so that there is the glass screen, there is the body, and there's the phone itself. I'm assuming those things. The way you read it is, OK, what's the glass screen doing the glass screen?

You actually can make it so if it was some sort of highly deformed thing, you could make it so it does wash to see if it touches itself. In this case, it's a block. It's not going to touch itself.

So to tell Fusion to crack that doesn't make sense-- the same thing with all these ones. So I actually did this Suppress ALL Self-Contacts. So if-- in here, so [INAUDIBLE] That it makes it so I don't have to worry about these. And it speeds it up a little bit.

The next thing we had is how you read these other things. It's basically what gets to move and what doesn't. So we're saying that the phone and the glass screen are bonded together.

And then you're saying that the other ones are separated so that they will move freely. So that's how you read that thing. If you need to change it, you just go in there.

It's a right click on this stuff. So edit the contact. And then you can just choose what type.

You get these different types. That's-- wouldn't call it-- not easy but not hard, right? I mean once you get somebody to show it out there, this is totally doable.

So from this stuff you have-- you managing your contacts. And then from there with the way the solve worked, the way I got this-- the mesh if I were to show you what it looks like, so we can go ahead and turn this on. So enmeshing, it's-- meshing is super-- oh, OK, we'll recalculate it. Oh, it's because I'm doing a new one.

Meshing is super important because that determines the character shape of how the stress moves across it in terms of your computing time. This is really super important. Actually I don't even know what settings I had. Let's just see what this comes up with. I can show you the one I did that worked.

So what I've found is that if you're concerned about a specific particular stress raiser, a particular, specific hole, you want a lot of mesh around that. This is a good example of a terrible mesh for event simulation. Why? It's really, really dense.

This thing's going to take forever. And it's going to throw up-- it's going to give you that, hey, this is going to take more than 24 hours to-- 12 hours to solve. Yeah, go stick it where the sun don't shine.

The one that actually worked for me on this actually has a mesh that's much coarser. And it's-- so if you'll notice around how many nodes we have across this particular bend, you might say if we're actually trying to get the accurate result from this, you would want more mesh. But if you're looking for trends and a result to indicate where the stress is, this turns out to be the Goldilocks amount for this particular simulation.

And what that looks like under the settings is-- I have in my settings-- is actually I'm using the absolute size for this particular type of simulation. What that is different than our model-based size or some of our relatives' tools we have, this basically says, let's basically use the same distribution of nodes across different parts. That just makes for a more stable simulation for this stuff.

So if you guys-- what are my best practices to say if you want to try this yourself? Try that. Other experts?

AUDIENCE: [INAUDIBLE]

MICHAEL AUBRY: Absolutely.

AUDIENCE: [INAUDIBLE]

MICHAEL AUBRY: Yeah, the question is around mesh customization in general. Let's do that on a simpler example. I can dig into that because there's some other stuff I wanted to share.

So you guys feeling good about event simulation? I wanted to-- yeah? Good? Anybody bad? I think it's pretty cool.

I mean, there's a whole bunch of other things. If you guys look online, it's a great one. Is that an Alex Lobos?

AUDIENCE: [INAUDIBLE]

PRESENTER: It is an Alex Lobos.

MICHAEL AUBRY: We have a legend in the crowd here. Are you taking this class? When's your class?

ALEX LOBOS: Tomorrow at 2:15.

MICHAEL AUBRY: OK, tomorrow at 2:15-- Alex Lobos is a professor at RIT and incredible industrial designer and-- hey, we'll talk after class.

ALEX LOBOS: I'm on vacation.

[LAUGHTER]

MICHAEL AUBRY: So let me pull up-- I'd like to pull up just a basic one. So we're going from like crazy, complex, cool iPhone to like, a sweet-looking Frisbee disc. The reason why I'm pulling this up is actually the gentlemen in the front, Bill, this morning was asking-- said there was a part of Fusion in how we apply force loads that he was looking at. And he said that he didn't know it was in there. And actually I knew it was in there.

When I was preparing this presentation, I was seeing some interesting behavior. And I figured I would share this as a tip for you guys. Like know thine force distributions to this.

So for this simulation what I set up was that I had these different loads-- hey, how's it going-- is I set up these loads as-- actually let me go back into the model itself. The way I set it up is I wanted to have a specific cross section that I was applying a force to. And I actually did it-- that cross section that I was applying the load to-- in CAD.

So I had a model-- or a sketch that I created it. So I don't know if I want to edit this guy. Right here, so one cross section, another cross section, and then I use the Split Body tool to actually split the face, saying I would like to apply force here-- kind of a slick way to do it, using CAD to set yourself up to go and do it.

So I go here, split it, go set up my simulation. Actually I can just show you guys this from scratch. So if I set up a study, static stress-- this is awesome-- well you go, no need to simplify that. That's pretty basic.

My material and steel-- I like that. If I said constraints, I'll just say, I'm going to just [INAUDIBLE] them this thing [INAUDIBLE] So [INAUDIBLE] change it up. I'll do this-- so constrain the sides, and the point that I'm making around there is that the way we do a load is-- what this is when you pull in your loads, that force will be applied uniformly over an area.

That's the default. So I was able to go-- then take and do this. So we'll just-- give me a load. What do you guys want to do? 10? [INAUDIBLE]

PRESENTER: Taking bids now.

MICHAEL AUBRY: 10 kilonewtons of force have been applied. And it's going to be uniform. Let's do it a different way for this guy actually.

So I got this face. I got this face. Actually let me go back. I'm going to suppress this.

If I go into simulation and take my load and say force-- so that's going across everything-- did you know that if you hit this guy, the angled data, and there is a little super secret power up limit target button you can apply across the face? Did you know that? All right. I knew it. I was surprised when I saw results on this, listening into what it's doing.

So let me-- so I would assume that if we have something basically the same size-- we go with this stuff-- oh, it can be a little different just for funsies-- I've got this thing where we did like 10 kilonewtons or something. Was it 1 or 10?

PRESENTER: It was 10.

MICHAEL AUBRY: OK, thanks. So I go and pull these things off there. Let's keep a-- I'm going to keep a coarse mesh on this thing, so just to kind of prove the point.

So if I go and-- you're asking around ways to manipulate the mesh-- so if I just generate it off the default-- it's just got some basic defaults and you can look at it-- general rule for how you set up a mesh is you want it to basically mimic the contours of what it is you're simulating and have mesh where the critical details are and then where they aren't, obviously cut some corners, and get rid of it.