0:04 All right, so I’m going to talk today a bit about process animation for design and analysis of injection welded parts. My name is Andrew Stelli, I’m the product management lead at Sena, I also handle all of our software partnerships. I spent almost all of my actual professional career working for engineering software companies, even though by schooling I’m a mechanical engineer. Maybe just quickly, from a a show, fans who’s heard about S before? Actually, okay, a de decent amount of folks here, that’s a, that’s a positive thing. So for those of you that haven’t heard about S before, we’re a company based out of the north of Germany and Breman, founded in 2018 as a spin-off of Alfred Vaga Institute, a research institute studying climate science.

0:50 But I think the most interesting thing about us is is not where we come from or who our investors are, but who our customers are, and if you look here at some of the the companies we have, whether it’s BMW, Volkswagen, Man Aron Group, also NASA, I think will also be talking a bit about us tomorrow. These companies are doing incredible things, and they’re starting to use Sara to elevate their processes. And so what are we actually doing with Sena? Well, we’re trying to solve the problem that a lot of engineers encounter today.

1:20 So for those of you that are engineers, you probably work in very sequential manner, you’re passing design data from one group to a simulation group to to a manufacturing group, and this transfer of information and knowledge is really difficult, and it slows down the product development process. Of course inherently there’s silos between all of these different groups that stagnate this process, and you know, if we think about something like injection molding, if you want to design a part for injection molding and use mold flow, for instance, to get some insights there, you really have to be an expert in that specific software to then bring those insights back into the entire development process, where you may be a design enger that doesn’t have that expertise.

2:00 So, you know, we took a step back and said, well, how are people trying to accelerate these processes today, and generally it’s in two different camps, either manual work or automation. So manual work, it’s pretty easy, you have a company that’s trying to develop faster and you just hire more people, but we all know that’s one very difficult to find the right quality of candidates, if we look at the room today, you know, it’s a small group here, and it’s a small group of folks that really can accelerate a company, and it’s it’s expensive to hire that. If we look at the automotive industry in Germany, a lot of times it’s outsourcing, so people will go through and work with engineering service providers to bring in engineers for a project basis to accelerate their teams, but there’s a lot of challenges with overhead, IP, and essentially the quality that comes out of some of these engineers.

2:52 The other aspect that we see is automation, right, a lot of us are using Python to automate different tasks that we do every day, or we’re using the scripting environments from products that we already use today to try to automate our work, but the problem we have with those is, whether it’s automating hyper mesh or Ansis or Excel, it ends up being some single purpose automations that don’t string together, or if you’re using Python, it’s great, but well, it lacks integrations into all the engineering tools that you’re using. And so that’s where we took a step back and said, okay, what is this perfect world that engineers could work in where they could really automate any aspect of their development process in a simple way?

3:35 We don’t want to have engineers writing complex codes, it’s hard enough to get engineers to document their processes, let’s not have them write complex Python scripts, we want that that work that’s automated to be sharable, reusable, transferable to other folks, and understandable by other people in the the engineering organization, not just the engineer, but the manager, the other people related to your projects. And so to do that, we ended up creating this visual language for engineers, so it looks pretty similar to those of you that may be not familiar with Scenar yet, but with Grasshopper, kind of themed after that, where you can create these automations in a visual way using a process canvas to construct your workflow, and then have the 3D visualizations you’re familiar with from your CAD tools, from your CAE tools, to actually understand what’s happening inside your automation.

4:24 Now, what types of things do people automate inside Scenario? This is always a complicated task for us, because we’re an open platform, you can automate pretty much anything, and we see that really in the the scheme of things that people end up automating. Ses an era, it can be as simple as pulling two parts out of your PLM system and comparing some attributes from them, all the way up to full vehicle concepts or spaceship design happening inside the same product. And so today I’m really going to focus a bit on what we’re doing in injection molded part design, because that’s an interesting area for us, especially in the automotive space, Bas, and consumer goods, where we see this as a mass manufacturing method for a lot of components.

5:05 So let’s take a look at a a simple part here, it looks pretty basic, but what are all the questions that go into actually trying to design a component like this? You know, first off, we have to think about what should my wall thickness be, now I could go pretty thin, but then maybe I don’t meet my structural criteria, I could go thicker, but then I’m going to actually have issues with my injection process. Of course I have to think, is my part actually moldable, can I demold it, can I get it out of my machine, at the end of the day, where should this parting line go, do I need multiple tools, will my part actually hold up to the the loading conditions that I defined, what about fatigue if I’m thinking about an automotive component.

5:47 When I start getting into the manufacturing process, where will there be defects, because there will be defects, where are my well lines going to be, where are my sink lines, is there going to be freezing, all of these going to be identified with molding simulation software, but getting that back into the design process is difficult. We have to think about how my mold should be designed, does my mold need conformal cooling channels, how should I manufacture my mold, where should I manufacture my parts, how much in the end is my part going to cost, the most important thing for a lot of companies, and then what material should I use. All of these things play a central role in the design for injection molded parts.

6:23 And now let’s think about how this actually translates into the work of engineers and designers, so typically when we think about this we think about maybe software tools, so you’re going to start out with your design process, jumping into a product like Fusion or NX or Creo, and start creating some sketches and some designs that generate some 3D models. Then maybe you’re going to look into Excel to figure out what materials your company has access to that you can use in the manufacturing process. Hopefully you’re using a a simulation tool like mold flow or CAD mold to actually look at what’s going to happen in the manufacturing process, so you’ll understand what the filling time is, get an understanding of the warpage of the component, and maybe some insights actually into what those defects are going to be in your part.

7:06 Then you need to use an FAA solver, maybe using Mark or Nastran, to understand the performance of the component, you’re going to use a tool like Fon to understand the cost and manufacturability of my part, and then maybe you’re going to use a CFD tool like Sim Scale to actually analyze the the cooling channels inside your component. Now, when we think about that, typically that ends up being multiple people in the process, so you have a designer, then maybe you have a business analyst or material specialist that tells you what materials you can use, you have an injection molding engineer, you have an FAA engineer, you have a cost engineer, you have a tooling expert.

7:47 All of these people are involved in the design for a single component, now it’s shown here as kind of being this this harmonous little circle of people, the reality I think many of you know is it’s not this harmonous circle, it’s chaos, it’s a lot of people trying to transfer data back and forth between one another, data is lost in the process, it kind of sucks. And so we take a step and say, okay, if we have an automation platform, how can we actually start to automate some of these tasks? And so, you know, the first example I have here is is a consumer products, e example, so it looks pretty basic, it looks pretty simple, it’s kind of a a door on a washing machine where you might put in the detergent, not a very complex complex example, but this is the reality of what engineers have to design for.

8:41 So this part is manufactured tens or hundreds of thousands of times going into washing machines, and the challenge is because it’s manufactured so much, you know, we have to make sure that it’s going to meet a lot of different criteria and of course be cost competitive. So we need to make sure that it’s going to have low warpage, because we’re going to put this inside an assembly, we need to be aware of where our weld lines and sync lines are going to be, because that’s going to influence the consumer perception of your products. And so in the end we’re going to end up with a part that looks like this, with some ribs on the back side of it to enable us to meet these design criteria.

9:20 And so now I’m going to step through what that process looks like inside Sara, so first off you see Sara here, it looks again pretty similar to Grasshopper for those of you that are familiar with it, and the process starts out pretty basic, every time we’re going to start here by importing some geometry. We also have connections to all the major CAD tools, so if you want to remote control Killa, NX, Creo, Solid Works, and so on, you can easily do that, but we’ll start, we’ll just import some basic geometry. And now after importing the geometry, we’re going to go ahead and start to create a workflow that generates ribs on our our project product here.

9:59 Now the nice thing is we have power solid baked into Sena, so we have robust solid modeling operations that allow us to construct this process, and you can see we have some sliders to start to play around with different settings of our ribs. Next up, we’re going to quickly set up a mold flow analysis, and I’ll get, I’ll speak a little bit later about how that integration works, but initially we’ll just define a few process parameters, like gating location for our analysis, and then go off and run the the analysis. Now the nice thing is, because we’ve set up an automation, we can now generate a lot of different data to understand this design process and start to do a DOE or an optimization to really select the right ribbing structure, the right thickness, to minimize warpage inside our part.

10:45 All of this again is it’s visually documented, so it’s very nice, it’s easy for folks to understand, and then of course, like a good engineer, I’m going to go off and I’m going to export some data to Excel and create some graphs for reports that I want to share with some of my colleagues. Again, all of this workflow is captured inside Scenario, so if you’re going to go and change the initial geometry, the whole process can replay, the whole automation can repeat itself. So in the end, we’re able to, with this automation, construct a design that meets our warpage criteria, that adds the ribbings that meet the performance criteria, and we’ve documented and simplified our process.

11:26 So how do we actually achieve this with Sara? So I like to say we’re not in this alone, we’re not trying to to reinvent the wheel, and so we do this through partnerships. So today we’re partners with over 20 different software companies, we have 56 add-ins on our marketplace, these partners range from Autodesk, Seamans, Hexagon, and big companies like that, to smaller companies, Intact Solutions here today, for those of you that were in Berlin, we’re partners also with Cognitive Design Systems, Rinex, and even we’re starting new partnership with Simcon, makers of CAD mold, to bring their injection molding simulation software inside S as well.

12:02 But really the key here is the marketplace enables easy access and connections to all these different tools, so if I think about what engineers do today when they want to automate things, they’re crafting that by themselves, they’re writing their own scripts, and what we’re trying to do is provide those easy off-the-shelf connections to those tools, so you can focus on the engineering work and not the integration work. And so with moldflow we worked with Code Product Solution, actually a engineering company based in the Netherlands that does a lot of consulting work for automotive car seats, and they created this integration inside Scara and released it themselves, and allows you essentially to remote control mold flow, generating your studies and bringing all that data back inside your Scenario process.

12:51 Now the second example I want to talk about here is getting now more complicated, so we started with a designer and an ection molding engineer, and now we’re going to add in a finite element analysist, and in this case we have a structural component that we want to again add some ribs to, we want to minimize warpage, and we want to make sure that we’re producing a part that has the lowest possible mass that still meets all of these different criteria. So we’re going to combine not only mold Clow, but we’re going to do FEA, we’re going to combine our CAD tool here as well.

13:30 And so again, I really love the the visual environment inside Sara, it makes it very easy to understand the process using document everything as you go along, and like before, we’re going to start off by importing some geometry, in this case bringing in a step file here, already colorcoded with different information embedded there. I’ve created here a nice little embedded workflow that’s going to actually go off and generate the FEA mesh, and so again the the nice thing is we’re not just providing nodes that generate FAA meshes inside Scara, we’re also connecting to your favorite tools like Ana hyper mesh, so that you can use the tools that you’re already using inside your workflow, inside your process, rather than just providing what’s provided by one vendor or another.

14:19 Next, we’re going to set up a basic FAA analysis on this component, again nicely embedded inside a little note here, but we’re using our internal solver here, but you can also make a connection to Nastran, Anis mechanical, Abacus, Optistruct, even now MC Mark, so whatever you’re comfortable with, you can start using that inside your process. In this case, we’re going to extract a displacement response to understand the stiffness of our part, and then we’re going to move on to the next step, we’re going to run a mold flow analysis again, so we’re going to pull out some warpage values here.

14:56 And so we’re going to have an evaluation now of a few different components, stiffness, warpage, mass, and so you probably notice these little purple boxes around some of my nodes on my canvas here, this is so we can do a design exploration and really explore the solution space of our model here. So we went through and we generated 100 different possible design variations here, running them all through our FAA solver, through a mold flow solver, to really understand all of these different aspects that go into design of an inject molded part. And of course I can select any one of these, or I can just show my design response space and evaluate that directly inside Sar and visualizing in the canvas.

15:49 And so again, we took another example here where we started to combine multiple user personas inside a single workflow, and the nice thing with Scar is, you know, those multiple individuals can work on the same document, construct a very complex workflow, and all of that is executed without any manual intervention there once it’s constructed. And I’ll just touch briefly on one last example here, which is something we put together with our partner Simscale and my colleague Max did for his his master’s thesis, which is now actually design and optimization of cooling channels.

16:24 So Max had this great idea that, you know, you see on the left kind of the more traditional cooling channels that you might see, he had the idea that there’s a better way to to generate those, and he had came up with his own algorithm to design this kind of more vein like cooling channel structure. And so he was a able to actually create this optimization inside Sena, evaluate the results, and actually came to the conclusion the traditional design was better than his new design, as can be the case sometimes, but the nice thing here, besides the the pretty graphic that we created, is actually that pretty much anything is possible inside Scar, if you put your mind to it, it’s an open platform, an open environment for you to start to play around with and explore different possibilities.

17:13 And so if we think back now to the original part we had, you know, how thick should my walls be, is my part modable, all of these initial questions that we have are things that you can start to evaluate inside Sara, not because Senaro provides the off the box or offthe self solution for all of these things, but because we’re providing you connections to all the tools that enable you to already evaluate these, but you’re not able to connect them together in one single automated process. So whether it’s tying in mold flow to evaluate the manufacturing defects of the part, or Naster and a Mark to evaluate the stiffness of the component, maybe you want to do remanufacturing with Cognitive Design Systems to really redesign the part for manufacturability, all of these things can happen inside Senero as long as you’re an engineer that is open to constructing their own automated processes.

18:06 And so I’d say, if you have challenges with manual work, your team’s struggling to get through the tall stack of problems that you have at the end of the day, take a look at Sara and think about what we can automate together.