Learning how to do Finite Element Analysis took me several years. I admit I mostly learned with trial and error (there were no FEA online courses back then, or at least I haven’t found any!). So if you are at the beginning of this learning process, it’s natural to have doubts. It’s only fair to ask: does it make sense to put all that effort into learning FEA? To help you decide, let’s discuss what are the advantages of Finite Element Method!
Finite Element Analysis allows you to solve any engineering problem. This means you can verify a product/structure without any prototypes! Also, FEA allows for a great presentation of outcomes, making your reports look super professional. Finally, mastering FEA have an incredibly positive impact on your career!
All right then, let’s look into details!
Solving the unsolvable!
There is one huge advantage of Finite Element Analysis. You can reliably solve problems with it that are impossible to calculate otherwise. But before we get into that, let’s deal with something I hear all the time:
Engineering was done well before FEA was even invented!
Folks who dislike FEA often use this argument, and… it’s true! There are so many cool ways to calculate stuff by hand (not to mention prototypes)! I’m always fascinated by hand calculations. Knowing those methods is super useful even when you use FEA. They allow you to verify if the FEA outcomes are “reasonable”!
But let’s be honest: in most cases, the hand calculations are accurate for very simple problems! With anything more complex, you will more or less have to estimate the outcome.
I think it’s reasonable to divide engineering problems into several categories:
- Simple problems within the linear domain. The classical “simply supported beam” goes here. If you can reliably solve your problem with hand calculations, there is no advantage of using FEA. Well, maybe the way outcomes are presented can be an advantage… but I wouldn’t learn FEA for this alone!
- Complex problems with closed mathematical solutions. This is a funny group. There are problems you can accurately solve with hand calculations, but the equations and procedures are so complex that they make you cringe! In such cases, FEA offers a quicker and more pleasant way of dealing with such designs, which is a huge advantage in my book!
- Problems you can only estimate. I will make a big claim here: almost everything can be estimated. More or less accurately but it can be. I admit, however, that some estimations will be so inaccurate… that they are useless in design, making the problem “unsolvable”. With FEA, you can solve the same problems almost perfectly! Not only you will get “better outcomes”, but also you will be more certain about them. Since estimations are risky, codes and design procedures place a lot of safety factors on such designs. This is clearly seen in many industries leading to uneconomical design. You can read about this problem in silos design here. In those cases, FEA is pure gold! It allows you to accurately design stuff and perform optimization you couldn’t do otherwise! This is when it shines, and allows you to develop your career as an expert!
Where FEA shines?
If you are mostly solving the simplest problems FEA won’t give you much. My experience is, however, that people rarely do the simple stuff anymore. We tend to optimize everything so much, that we are encountering new problems all the time. Meaning that even the simple tasks become more and more complicated, and require advanced design approach, to obtain the desired level of optimization.
But I think that gold is somewhere else! If you are working with projects that can’t be calculated accurately, or the calculations take way too much effort… you will love FEA! It’s a tool that gives you the edge. Instead of guessing or wasting time on long mathematical procedures, you can do your FEA analysis. It will show you how the stress will be distributed in your model, and you won’t have to guess anymore. Not to mention other analysis that can inform you about vibrations, buckling and so much more!
FEA will also allow you to see where stresses will concentrate on your model and how badly! Try to learn the same with hand calculations!
FEA at a higher level
When you learn enough about FEA there is another level you can unlock. That is obtaining the actual failure mode of your model!
Before we get all excited here, let me tell you that this actually requires quite some learning.
You won’t do it overnight… but it’s doable for sure! I would estimate that with proper guidance and some practice you can get there within several months, maybe up to 1-2 years depending on what you want to do. Even here, there are more and less difficult things! Let me just tell you, that I learned this with trial and error (and sadly I mean it literally!). It took me 5 years to “get somewhere”. But this also shows that it’s doable even without ay external resources! Just don’t dive too deep into mathematics of this all… this won’t help you on the way!
This level is where the fun happens for sure! First of all, you are treated as an expert, and you earn like one I guess. I admit that when I was starting this was what motivated me. Funny enough it was more about the “expert” rather than “money” for me… I guess I’m weird like that!
Now I see things differently, sure being considered “an expert” and making a buck is great… but I also get to help a lot of folks! I also get to choose with whom I work (I admit that I resigned from some work as I didn’t like the people…). It’s such a great feeling to meet with people you like and help them out in their business! And there are plenty of ways you can help them! The most important are:
- Level up their products… cheaply! I just wrote that you can model an actual failure of something in FEA. Just as if you were testing a prototype… but in a fraction of costs and time required! Furthermore, with some programming, you can generate lots of various prototypes and verify all of them! If this is not a super-power I don’t know what is! It’s so cool to do a better product for companies, and see their growth and experience their gratitude. Definitely, something I love about the job!
- Get them out of trouble! I admit that I always take such jobs with a heavy heart, and I make sure this is not a “witch hunt” before I start. Let’s face it, things are being done, some aren’t great… some are straight-up bad. When something goes wrong the ability to model reality (or it’s a reasonable approximation) is super useful! You can tell what went wrong and why. This gives you a tremendous edge, and an ability to propose a good remedy for the problem. Just be aware that there are different kinds of such jobs:
- I like when companies approach me and say “our product has issues, we don’t know why – find how to fix it” – this is an awesome job. You get to talk to people, see how the stuff intended to work and brainstorm ideas on possible fixes. I think I’m the most energetic at such meetings!
- But there is also a darker part – especially in civil engineering where there is usually one “main” designer. And that is “find where *he* made mistakes”! While I understand the need for making fixes, I always make sure that my work won’t be used against someone directly (and luckily usually this is the case!). I learned from my professor at university while I was working on my first ever failure investigation (as a student back then). He told me: “Łukasz, never assume you would do this better than that guy. He had 2 weeks to complete the design, and we had several months and a staff of highly educated specialists to analyze it. In 2 weeks we would most likely make the same mistakes”. I guess what I’m trying to say is: don’t be an asshole. Sure, you may know more than someone else, but using this to “punish” him isn’t the best approach. This is why I want to make sure we are working toward a solution, not punishment before I start… and I would quit if it weren’t so. I don’t like to have a guilty conscious… after all as a famous XX century philosopher once said: “with great power comes great responsibility”!
- Do projects no one else dares to do! This is a fun one, connected a bit with “being an expert”. My experience is, that the more challenging the project is, the lower the competition is. Let’s face it, everyone want to do the simple work… which means they will do that work cheaper to get it. This is why the “simple” projects are so dirt cheap! If you can do the difficult things, then you are so much better off! Not to mention you get to be a hero for your Customer, as the only specialist who wanted to do the job! Again, this is a great advantage, and it comes directly from using FEA in many fields!
So as you can see, a lot of benefits comes from being able to use higher-level FEA. For me, the above list would be enough to learn, but it’s not all yet!
Learn as much as you want!
This is something I haven’t realized at the beginning. So I can’t tell you that this motivated me to learn! It’s one of the FEA perks I discovered by accident, but it’s such a great one!
You see, FEA is “just” a calculator, that is difficult to use. This is why you have to learn it, and this learning takes some considerable time. But when you reach a certain height, you realize that there aren’t all that many people running things the way you do. And since you already know how to use the calculator… you can calculate whatever you want!
This happened to me already, and I’m super excited about it! If you know how to solve complex FEA problems you can easily change industries! After all, the industry is more or less a set of “assumptions and regulations” about something, and the rest is “just calculations”. There are plenty of people knowing the assumptions and regulations, but relatively fewer that can pull off a proper analysis and interpret outcomes. Interpreting outcomes is much more difficult than doing the analysis itself I think!
I literally had a Customer already, who told me that they will teach me how to design stuff in their industry! So they explained what the codes and rules are, what are the best practices and all that… and thanks to this I could do FEA design for them. I think this is the best way to work, not only you get paid for doing what you are good at, but you also learn additional things making you even better!
A few years back, I just wanted to be “good”. Now I can be good at whatever I want within a certain radius from what I’m competent at. And the choices make my head spin!
And I know (being slightly paranoid) that when my industry will be in a bad shape globally, I can switch to a different one that is in ok shape. This elasticity makes me feel so much more relaxed. I think that being able to model reality in FEA is an ultimately flexible skill!
In a nutshell
I tried so far to write about advantages of FEA for you and your career. But of course, the method itself has few nice blows and whistles. It won’t take us long, so I figured a list would be nice:
- Description of what you want to do is easy. I really envy people who can simply read differential equations… well I can’t! If I would like to solve a complex problem using them… it would be my doom! But with FEA I don’t have to play this way. All I need is to “draw” whatever I want to calculate (like in CAD environment) and assign loads, supports, etc. to this. A much simpler process than formulating equations I think (+ you can actually show something at “show and tell” at your kids’ school!).
- Seeing outcomes makes it easier. Definitely, several times in my career seeing an outcome made me question my assumption and lead to finding mistakes. Visualization of outcomes definitely is a very strong suit for FEA! Not only you can see the outcome, but you can also animate it. This way, it is so much easier to understand the behavior of your model – definitely a nifty thing!
- Boundary conditions. If you read my blog for some time, you most likely noticed that I’m “big” on boundary conditions. The fact that you can see your model (and how it deforms!) can greatly help you assign proper boundary conditions. This is so important! A huge FEA advantage right here!
- Do whatever you want. I already wrote about it in the previous section. You can really apply FEM in a lot of various applications. Making it a really flexible tool at your disposal!
- Save on prototypes. This is a good one of course! If you can model a good approximation of reality with FEA that’s it! You can save a lot on creating early-stage prototypes, and be a hero!
- You don’t do the math! Yup, your computer does it for you! And you know what? It doesn’t get bored or tired, and it’s not distracted by YouTube! All of this reduces a lot of possibilities for errors in analysis… but you still need to be careful while creating the model!
- So many interesting things to calculate. I would guess that usually FEA is associated with linear static. You know the nice stress plots you see everywhere online. But you can do a lot more! Some more complex problems (dynamics, nonlinear stuff) are basically unsolvable outside of FEA. This is another super cool advantage of this approach to solving engineering problems!
So as you can see, the method itself has a lot of advantages over its “competitors” (mainly hand calculations and prototyping). But since I’m a fan of FEA, I would say that this is not an “honest” competition. At a certain level, FEA is a dominating approach in my opinion (but I know there are folks who I respect, and who would disagree).
It’s not only “sunshine and rainbows”!
I don’t want to be onesided here. This is why I think it’s only fair to talk about FEA disadvantages as well!
I think that at least some of those may surprise you!
- It looks easy! Yea, that is actually an issue. You see, a lot of people *think* they know how to do FEA. Unfortunately, it’s rather easy to get the wrong answers. Some will be “just” wrong, but sometimes you may get “terribly wrong” answers as well. If you don’t possess enough knowledge about what you do, you may not notice that the outcomes are bad. And this is where the potential problems start…
- It’s good to have a proper gear! FEA is a beast when it comes to calculation possibilities. But those take time… sometimes even hours if not days! We tend to think that computer calculates stuff in an instant, but in FEA there are a LOT of things to calculate. The longest analysis I did so far commercially took almost 30h to compute! Good gear is definitely an important thing to have here!
- It’s rather expensive… I’ve spent the first 3-4 years in my first company to save up money so we could buy our first robust FEA license. Sure, there are countries where people earn more and all that, but still, license costs are something to consider. There are now online and open-source FEA codes, so if this is something that terrifies you, check them out!
- Searching for mistakes may be scary! For big problems/models you will have to prepare a lot of input data. If you make a mistake somewhere it will be hard to find it. Especially when you don’t verify your work during model creation. Sometimes it’s even easier to start a model from scratch than to search for a mistake…
- It’s actually easier to convince someone to a stupid idea with FEA… This is not something you should try. But when you get a stupid idea, you can “prove it” in FEA with some wrong assumptions. It will look way more professional and “expert-like” so you have a higher chance to convince others to your stupid idea. It works so well, that I even convinced myself once or twice. Luckily Asia who worked with me never hesitated to ask those critical questions!
- FEA is complex, and there is a lot to learn. I’m not saying you need all of it at the beginning (God knows I did my first FEA designs knowing next to nothing about this!). But even as you don’t have to know all, you definitely need to know enough. And it’s actually quite a few things! You know, things like basics of meshing, validation of outcomes, supporting your models, etc. This ties to the first point on this list… since FEA “looks simple” to use, you may think that you don’t need that knowledge. Well… I would be careful there! I’m far from saying that you need a Ph.D. in all of those topics, you really don’t. But ignoring them completely isn’t a great solution either!
A short summary
Without a doubt, FEA is a powerful and flexible thing.
Since we are discussing the advantages of FEA here, first of I would say that FEA gives you a tremendous career advantage! If you master this art, you will be able to relatively easily switch industries to hunt new cool things to do. You will be also able to save your Customers time and money with your work. This is definitely a good place to be!
FEA itself is an interesting tool. While it looks deceptively simple (which sometimes is a problem for sure!) it really goes deep. Thanks to this fact, it can allow you to calculate problems that are otherwise unsolvable. And with a proper understanding of mechanics, there is little to no things you can’t actually do in FEA!
Solid learning at the end!
Thank you for reading! I really hope you liked the article, and to be honest I hope I managed to at least interest you in FEA a bit : )
If you want to learn the most important things about FEA that I learned during my 10 years in the industry, definitely sign up to the free online course I just finished. Just sing up below – I’m pretty sure you will enjoy it! See you around!