calculus

Dismantling the #1 myth about being an engineer

01 June

It was a bit of a shock to me and some of my peers when we started our first jobs.  The real world of engineering was NOT like the stereotype or like school.  21st century engineering requires a wide range of skills and our managers were asking us to do tasks school hadn’t prepared us for.

Surprise surprise, another stereotype isn’t true.  (sarcasm)

I’m going to dismantle the engineer stereotype, bit by bit, in 6 posts.

PLEASE share these posts with any science teachers, teens or their or parents in your life.

I especially want teens who are considering engineering or any other science field to read my posts.  In my tutoring I see many brilliant students who think they can’t be this or that profession because they don’t fit the stereotype.  It breaks my heart.  The stereotype is increasingly false in 21st century engineering, and those who do not fit it are highly sought after by smart companies.

The first misconception is a big one.  Hang on, it may be a bumpy ride, but it’s worth it.

Myth #1:  You need straight A’s to be an engineer.

I am breaking a huge taboo by saying this, but you do NOT need straight A’s to be a good engineer.  And I would guess this is true of many other professions.  Sorry teachers and parents, but it is true.

After my first job no one cared what my GPA was.  All they cared about was that I completed my degree and went to a good school they had heard of.  That is it.

I myself did not get straight A’s in high school and earned solid B’s in college.   I was not valedictorian of my class, barely made it into the top 15%. Yet I have worked in the Aerospace industry on high tech carbon fiber composite parts using advanced Finite Element Analysis Software.  Cutting edge stuff.

Very few of my colleagues were straight A students either.  Most of them were solid B students like me, only a couple of “geniuses”.

What set us apart were not our grades but our work ethic.  Hard work, perseverance, persistence, sweating the details.  All of these habits paid off in the real world of engineering.

Some of you reading this may argue that to get straight A’s a student has to work hard.  That is true for some students, but others learn new concepts easily and coast through school. They frequently do not learn “grit”, strategies to handle difficult problems.

I remember struggling to solve difficult calculus problems in college and feeling like it was a personal failing that I didn’t get it the first time.  That self judgement doubled my stress level, and college is stressful enough all on its own.  I felt ashamed to go to office hours or ask my peers for help.  I did it anyway and it paid off.

Calculus is complicated, nearly no-one gets a differential equation the first time they see it.  DiffEQ is abstract and counter-intuitive.  I would never expect myself to make a hole in one my first time playing golf, so its unreasonable to expect the same of my brain.

It is well known that in sports, music, or dance, sweat will get you a lot further than simply coasting on raw talent.  Engineering is a brain sport and the brain works a lot like a muscle.  The more you use your brain, the stronger (i.e. smarter) it gets.  Science says so.

In fact, being a straight A student can be a problem once you graduate into the real world of engineering.  Why?

Click here to read about the second biggest myth.  

5 Responses to “Dismantling the #1 myth about being an engineer”

  1. Dante D June 4, 2015 at 6:13 pm #

    I worked with engineers in aerospace doing composites with finite element analysis software. Cutting edge? Meh, a lot I could train a high school student to do.

    Many of these aerospace engineers were indeed B and C students from some good schools, mostly 2nd and 3rd tier. They were schools known locally.

    These engineers were absolutely terrible, and if you ever wonder why many aerospace projects are 2 or 3 or 10x over budget, it is because of numbskulls like these.

    So while there are indeed a wide range of engineers, from your MIT PhDs to your Long Beach State “BS”, there is a definite difference in the actual engineering.

    • NastranLover June 5, 2015 at 1:00 am #

      Dante, I’m sure that a high school kid could do FEA with a little training. They could also be taught to build a bridge with a little training. That doesn’t mean I’m going to drive my car over it.

      Besides, why does it matter which college these numbskulls went to if you could train a high school kid to do it? Are you saying that knowing some fundamental concepts about force and stress makes the numbskulls HARDER to train than your high school students who don’t even have a vocabulary to use, or know about Hooke’s law and couldn’t construct a free body diagram to save their life?

      Studies show that the more we know about a subject, the less simple we think it is. The converse is true as well. Your comments make me think that you think you’re in the camp which thinks it’s simple.

      While anybody could, with limited training, be taught to use a GUI and build an FEA given a few rules of thumb, that doesn’t mean that the model will be useful. FEA is often under-rated, easpecially by project engineers and designers who have seen one too many such crappy analyses. To get a useful model, you have to know how to model the physics which, spoiler alert, frequently looks nothing like the geometry in the world if FEA. You cannot simply hit mesh and run– not if you want anything useful.

      Even a simple beam model can be screwed up by someone who doesn’t understand the difference between a Timishenko beam and a classic Euler-Bernoulli beam. Shear relief, warping, etc. Shells, plates, and membranes have a whole gammut of much more complicated issues. And anybody who tells you that solid elements are more accurate is either lying to you or is grossly incompetent and should be fired immediately.

      I have been doing both stress and dynamic FEA for over a decade, under the mentorship of a guy who wrote one of the codes we use today. I have never met a real-world problem that a monkey high school kid running a GUI and hitting solve could get right after just a little training. It takes two years of half time training to get an engineerto the point where they are even marginally competent to run basic problems without supervision. If you have seen it, then the problems must have been essentially textbook and probably should have been done with a hand calculation anyway (the first question I ask myself is could good enough answer be obtained without an FEA).

      I’ve reviewed and rejected analyses by PhD’s for mis-applying a few rules of thumb, not thinking about how their results compare the the physics, and missing the mark entirely. I’ve taught people FEA, both people who have never done any and those who had been trained but who had not been taught anything beyond basic mechanics. It’s never a fast process and I am constantly developing them, and pointing out a nuance of the physics that they missed. Either they were overly detailed for the level of confidence expected, making them take much longer than necessary, or they don’t understand why something has such an important effect and then they have to re-work. Both appeoaches waste time and money, which could have been avoided if they knew the physics instead of having just a little GUI training in high school.

      Throw in composite materials and the nuance/difficulty of getting an answer quadruples. It’s more than just assigning properties. Will the layup have important cross-coupling? Gotta do something tricky. And if you don’t know about an ABD matrix then you’ve got a no hope. All of a sudden with composites, edge stresses matter. Does a high school kid even know about edge stresses? There is even less good data to verify your model. Do I have enough nodes? Can I allow my software to lump it as a equivalent isotopic material? What if it’s not a laminate? How does the manufacturing variation affect results? What’s conservative? These questions only get more difficult for FEA when composites are in the mix.

      The only FEA analysts who get it right are the ones with extensive theoretical backgrounds in graduate level mechanics, dynamics, materials, numerical methods, and FE Theory, or who have been taught an equivalent level of theory by their mentors. They’re sometimes at the top of the class, and sometimes earning C grades, and sometimes in the middle. The ones who are used to straight A’s often can’t cope with real world problems, and the ones worth C averages who were working and raising families in college often out perform the A students. Some schools do produce students who are much better than others. But their success is much more profoundly influenced by their experience on the job, and the quality of the mentorship they get.

      If you can’t get the fresh-outs who start with the disadvantage of a lesser school working to the same level as the better educated ones by the time they’ve both been working for two years, then I’d say the problem is your mentorship ability and not the school. Once you’re out of school, it’s the experience that matters, not the grades you got. School gives you a vocabulary, everything else is on the job education and determination.

      Maybe you could find some genius high school kid with the time and interest to compress 5 years of full time training (theory plus rules of thumb plus on the job) before they graduate, but anything short of that will result in work that wastes all the time and money you speak about. The waste is most painful when I reject the entire assessment for the sheer ineptitude of the FEA they provided to substantiate their claims. It’s especially frustrating when a few basic checks or test cases could have revealed the error to them if only they understood more than the GUI training they got in high school.

      People who think that a high school kid could be trained to profuce believable and useful results probably also think that the elements are actually a real thing that exists mathematically. They don’t. Nodes exist in theFEA world, but the elements do not. See if you can get one of your high school trainees to explain why, if they’re so competent and capable.

  2. Dave June 4, 2015 at 6:52 pm #

    Dante’s right; there is a wide range of engineers. Some of the very best I’ve met, the ones who solve problems the fastest, who come up with elegant and effective solutions, were farm kids. They got used to thinking about how things work and how they might be fixed, and just kept getting better at it. Then there was the fellow with a Ph.D. who didn’t know which end of a hex wrench to use…

    But there’s room in industry for all of them. Engineering runs the gamut from lab-coat science projects to overalls and wrenches, from late-night coding sessions to dressed-up sales pitches. An engineer may need to be well-versed in sales, marketing, customer service, project management (which is where budgets are made and broken), or any of dozens of ancillary subjects. And the ones who excelled in differential equations will have work building tools for the rest of us to use.

    Collaboration and specialization make all of us more effective, and make good use of the skills we have. Some good students go on to become good engineers, but not all good engineers were once good students.

  3. NastranLover June 5, 2015 at 1:08 am #

    Dante may be right in claiming that FEA has been around so long that the tool itself isn’t cutting edge. But what we do with it is. We are constantly pushing it to its limits, trying to figure out how to solve problems that were once unsolvable. What do you think all the user conferences are about? You think people pay thousands just to go socialize? No. They’re cutting the edge.

  4. Oscar Gordon June 5, 2015 at 4:16 pm #

    I’m going to second everything NastranLover said, especially as it applies to CFD (my field).

    As for the variation of engineers, if a person makes it through an engineering program, they have value. No one actually coasts through any reputable engineering program, some may sweat less than others, but everybody sweats. In the real world, if you have an engineer who is not hacking it, then either A) their mentor sucks (I had one of those, it was horrible, I quit that job over it), or B) they are in the wrong job. If it’s B, and you are their mentor, perhaps you could be a decent person, take a look at their strengths and weaknesses as you see them, and guide/advise them to a different aspect of engineering where their skills & talents can shine.

    I’ve studied CFD for almost 20 years, I work for a company that produces CFD software, I occasionally do CFD analysis, but I spend the bulk of time writing software tools so other engineers can do CFD effectively. This was not the job I imagined for myself as I graduated college, but it is the job that fits my skillset exceptionally well. Of course I had to have a few jobs that were bad fits before I finally accepted that the stereotypical/traditional engineer job was not going to work for me.

Leave a Reply to Oscar Gordon Click here to cancel reply.

This site uses Akismet to reduce spam. Learn how your comment data is processed.