Is engineering really like this?

[dudecoolname]

Hey guys, my first post here at physics forums. I am not actually an engineering major myself...I just wanted to know, do you guys really have to solve equations like this?!

I just took Calculus I...wasn't terribly difficult but THIS stuff...O_O

 

[Pengwuino]

Yes, MOST things you see people doing after studying it for 4+ years should be a bit shocking to someone who has just dipped their foot in the water. Engineering is no exception.

Although that is TOTALLY what I thought engineering was too. You don't want to know what I thought physics was that made me get into it.

 

[viscousflow]

Yep, you're about right. But most of the time, you mainly see those super nasty equations when you need to understand something more fundamental-basically you usually only see those once. In practice however, it would be cumbersome to need to go through that stuff on a daily basis.

 

[xxChrisxx]

Engineers don't really solve equations like that (though we need to know how to to understand what's going on). We get out whipping boy mathematicians to do it for us and then tabulate the data.

Machinerys Handbook.
Roarks Formula for stress and strain.
Engineering data Handbook.

(used daily)

 

[Pengwuino]

Engineers don't really solve equations like that (though we need to know how to to understand what's going on). We get out whipping boy mathematicians to do it for us and then tabulate the data.

Machinerys Handbook.
Roarks Formula for stress and strain.
Engineering data Handbook.

(used daily)

Can't leave engineering to the engineers!

 

[xxChrisxx]

Can't leave engineering to the engineers!

NevAR!

To be honest I really miss the challenge of academic engineering. Working on real products (unless on a development project) is really boring when you have strict design procedures and prechecked spreadsheets.

 

[Phrak]

I work with plenty of "engineers", and not one is an engineer--most couldn't. Invoking conversation about their presumed trade brings about slack expressions of incomprehension or the look of a deer in the headlights. They all seem to be witless in the finer points of their long lost education; have not advanced their education one bit since graduation and have landed careers as gofers and paper pushers.

If you want to do real engineering you have to look for it.

This should be enough to upset a lot of people, but after little formal engineering training (three classes) and some thirty or so design projects as principle designer, or subcontractor/consultant, I think I have a claim to some valid opinion.

 

[Mech_Engineer]

Engineering is equal parts paperwork, design and innovation, and calculation. I've found my job to be better in some aspects than I expected out of school (for example I get top-of-the-line computing equipment and software to facilitate my job), and in some aspects worse than I thought (the documentation and paperwork can be pretty tedious, especially in procurement).

To answer your question though yes I do use equations like that; not on a daily basis, but at least once every other week. In the grand scheme of things that equation really isn't that bad- it's just a bunch of addition/subtraction/multiplication/division paired with a few square roots. When you get into fluid dynamics, thermodynamics, and heat transfer things can get a little hairy. Calculating the natural convetive and radiative heat transfer in an electronics enclosure to confirm FEA results will make your head spin...

One thing I do appreciate is being able to use math software like MathCAD to do the tedious math for me, whether it's symbolic manipulation or otherwise. It helps reduce errors, and makes it easier for other engineers to check my work. Here's a good MathCAD sheet for you- https://www.physicsforums.com/showthread.php?t=209319&page=1":

https://www.physicsforums.com/attachment.php?attachmentid=12535&d=1202399550

 

[xxChrisxx]

Here's a good MathCAD sheet for you- https://www.physicsforums.com/showthread.php?t=209319&page=1":

https://www.physicsforums.com/attachment.php?attachmentid=12535&d=1202399550

Urgh! As nice as that is I'm far too lazy and it's computationally cheap enough just to stick something like that on Ansys.
tis quite a sexy mathcad sheet though.

I've recently learned to use MathCAD to so On bottom stability analysis. Save, save, save as there is no undo button.

 

[Topher925]

When I was an entry level engineer I used equations and stuff like that on a daily basis. Like Mech_Engineer said, the equations and math isn't really that bad and is something that should be embraced as a valuable tool, not feared. This is coming from someone who is terrible at math and failed both calc 2 and 3. When you apply these complex mathematical equations to something that you're actually building or testing in the lab, the math becomes a lot easier to understand and a lot more valuable than the stuff done in academia.

The not so good part of engineering includes the paperwork and reports and all the nonsense the bean counters and six-sigma morons put you through. I'll take equations over mind numbing paperwork any day.

 

[Mech_Engineer]

I'll take equations over mind numbing paperwork any day.

That is the truth!

 

[AlephZero]

Remember the Zen Buddhist saying:
When you first encounter something, everything seems simple.
As you look more closely, nothing seems simple.
As you look even more closely, everything becomes simple again.

You got hooked on engineering while you were at "stage 1", and didn't really know anything about it. You are currently at "stage 2". It's your choice whether to quit or keep going till you get to "stage 3" - which is not just looking like the guy in your picture, but also knowing what to do if you were in his situation, and a wrong decision would cost $millions, and/or kill people.

There is a famous quote from Hamming (Google him if you never heard of him): "The purpose of calculation is insight, not numbers". But most people can't get to the "insight" stage without going through the pain of learning how to crunch the equations. That's what you are doing right now.

 

[Angry Citizen]

Hey guys, my first post here at physics forums. I am not actually an engineering major myself...I just wanted to know, do you guys really have to solve equations like this?!

I just took Calculus I...wasn't terribly difficult but THIS stuff...O_O

That equation looks like a physical chemistry equation when you take the 'ideal' out of the ideal gas law. So no, you're not likely to encounter that equation as an engineer. Honestly, it really depends on what job you land. If you're a propulsion design engineer, you'll probably see equations like that all the time when you're modeling fluid flow. If you're a sales engineer, you'll probably go "bwuah?"

 

[minger]

That equation looks like a physical chemistry equation when you take the 'ideal' out of the ideal gas law. So no, you're not likely to encounter that equation as an engineer. Honestly, it really depends on what job you land. If you're a propulsion design engineer, you'll probably see equations like that all the time when you're modeling fluid flow. If you're a sales engineer, you'll probably go "bwuah?"

The ideal gas law is a very common closure equation used in fluid dynamics. When temperatures and pressures get high, we often try to use real gas approximations which "can" look like what the OP posted.

 

[Phrak]

I'll take equations over mind numbing paperwork any day.

That is the truth!

I'd like to know what sort of company you work for. The mind numbing megalith I work for seems to contract-out all it's engineering skills. I need a change.

 

[JakeBrodskyPE]

Nearly 25 years ago, my engineering mentor told me that if I was using any more than a pencil, paper, and a plain scientific calculator for a routine design, that I was probably doing something wrong. "There are rules of thumb for nearly everything," he said. "The more complicated the equations, the more likely it is that you'll make a mistake; find the rules of thumb and use them."

Moreover, we need margins for error and safety. You could engineer stuff to the gnat's eyelash, but it is likely to break when you need it most. We call this engineering with an extremely sharp pencil.

I encountered some sharp pencil engineering in my clothes dryer recently. The microcontroller balked because it couldn't see current in the electric heater. The relay contacts in the control board were undersized, and got pitted and burned. Had the designers followed the rule of thumb for heater current inrush and contact sizing, this wouldn't have happened.

So, yes, the equations are a good thing to know so that you don't trip yourself up. You need to know where those rules of thumb came from --because sooner or later you'll encounter a situation where the rules of thumb no longer apply. However, in daily use, we rarely have to refer to full equations like that.

 

[D H]

The not so good part of engineering includes the paperwork and reports and all the nonsense the bean counters and six-sigma morons put you through. I'll take equations over mind numbing paperwork any day.

Even worse is when you yourself are turned into a bean counter or six-sigma moron.

I love debugging the blank sheet of paper. ("See this blank sheet of paper? It's supposed to contain a spacecraft design. Fix it please.") Most of the stuff that ensues after debugging the blank sheet of paper isn't fun. However, much of it, or at least some of it, is necessary.

 

[Astronuc]

Engineering can be like the example in the OP, especially if one does R&D and application.

This day and age, many engineers use simulation programs (computational physics), and such programs have become more complex and sophisticated with the increase in capability of computational systems.

Not only do we solve complicated ordinary and partial differential equations, but we solve systems of differential equations.

Thermomechanical simulators require detailed models of thermodynamic and thermophysical properties in addition to the constitutive models of the materials. The greater the range of temperatures, pressures and energy densities, the more complex the simulator.

The there is the matter of the optimal solution (computational) scheme.

I should add one more point, which is becoming more important these days, and that is the ability of simulators to capture the physics of the evolution of microstructure or chemical changes of material systems in their operating environments. This is particuarly the case in power systems, and more so as the operating temperature increases.

An example - http://mech.fsv.cvut.cz/~dr/papers/Poofem/poofem.html

 

[Mech_Engineer]

Nearly 25 years ago, my engineering mentor told me that if I was using any more than a pencil, paper, and a plain scientific calculator for a routine design, that I was probably doing something wrong. "There are rules of thumb for nearly everything," he said. "The more complicated the equations, the more likely it is that you'll make a mistake; find the rules of thumb and use them."

That was 25 years ago, and probably held over from his college years 25 before that...

It's true that you can get a good order of magnitude estimate with analytics, but these days with competition what it is engineering is held to a higher level. Computer-based engineering is the standard for the 21st-century, and can make the difference in a product being competitive.

Moreover, we need margins for error and safety. You could engineer stuff to the gnat's eyelash, but it is likely to break when you need it most. We call this engineering with an extremely sharp pencil.

True, but safety factors are easily added into computer-based calculations such as FEA. In an analytical calculation of a complex geometry, the safety factors have to be higher than would otherwise be necessary to make up for inaccuracies and simplifications...

I encountered some sharp pencil engineering in my clothes dryer recently. The microcontroller balked because it couldn't see current in the electric heater. The relay contacts in the control board were undersized, and got pitted and burned. Had the designers followed the rule of thumb for heater current inrush and contact sizing, this wouldn't have happened.

There are lots of dryers that work great and utilize microcontrollers... sounds like your specific model may have made a few too many cuts to try and keep the price down. That being said, buy a better engineered brand if you aren't happy with your current unit.

So, yes, the equations are a good thing to know so that you don't trip yourself up. You need to know where those rules of thumb came from --because sooner or later you'll encounter a situation where the rules of thumb no longer apply. However, in daily use, we rarely have to refer to full equations like that.

Rules of thumb can be very useful, but in these days of competitive high-performance product and precision optimized design they're for order-of magnitude estimates to get into the ballpark, not for the final solution. I use rules of thumb and simplified analytical estimates to confirm the FEA solution is in the right order of magnitude, not visa-versa. In the end, the FEA drives the final design.

 

[PhanthomJay]

Nearly 25 years ago, my engineering mentor told me that if I was using any more than a pencil, paper, and a plain scientific calculator for a routine design, that I was probably doing something wrong.

Yeah, and the one without the hyperbolic functions works quite well, thank you. Of course, the computer comes in handy...as long as you're able to interpret and verify its input and results using your backup calculator and engineering expertise.

 

[JakeBrodskyPE]

Rules of thumb can be very useful, but in these days of competitive high-performance product and precision optimized design they're for order-of magnitude estimates to get into the ballpark, not for the final solution. I use rules of thumb and simplified analytical estimates to confirm the FEA solution is in the right order of magnitude, not visa-versa. In the end, the FEA drives the final design.

Yeah, it's great until you make a minor typographical error and don't realize it. What I'm talking about is an intuitive feel for the process or device that one is engineering. These computer programs are really really cool and they can result in some interesting and highly optimized designs. However, I've also seen too many stupid mistakes made by people who had no idea what their software was doing.

I use computer models to do all sorts of things. I use automation to get things done with far fewer employees and far better quality. However, those who still work here need to know what is being automated. You can not merely push a button and watch a process happen without knowing exactly what it is supposed to do for you and what the results should look like.

Before you even start walking toward the computer, you need to have some idea of what you're expecting from it. If the results do not look right to you, YOU MUST INVESTIGATE!

I don't mean to sound like a Luddite. However, if you've been in the field like me for most of your career, you'll know that putting things in autopilot does not absolve you of the responsibility for knowing where you're going --even if you're going somewhere you didn't intend to.

Thus: "If you're engineering anything with more than a pencil, paper, and a calculator, you're probably doing something wrong." Go ahead and optimize with whatever software you have. I have those tools too, and I use them. But you damned well better know what the result should look like in advance, or you will eventually make a terrifying mistake.

My designs are refined with software, not driven by them.

Jacob Brodsky, PE

 

[viscousflow]

My designs are refined with software, not driven by them.

the FEA drives the final design

I think both of you are talking about the opposite side of the same coin. Both of you have mentioned that you have to check whether or not FEA is in the correct ballpark or not, or if the final solution even seems correct or not.

However, the key thing is, when both of you have concluded that the FEA software has given the correct answer you have to run with that answer. Unless you ever feel like solving millions of simultaneous differential equations recursively to proof check the code.

 

[coucher630]

This is a pretty humorous picture...and yes, I'd say accurate too. But don't let that discourage you.

Often I find myself nose deep in equations that look like this when you sprawl them all out to look complicated, and that's something I would have found horrifying in the early years of my education. Now, I do this kind of stuff and then take a step back and try to take in everything that just went into those equations-- it's really awe-inspiring to see how everything seems to work out when all the number crunching is done. To me, that can be more rewarding than walking around with a hard hat and a toolbelt (I keep those for weekend projects at home) .

The key is to find a job that gives you a mix. Enough hands on stuff to keep your finger nails from getting too clean and enough math magic to keep your mind sharp and inspired.

 

[Mech_Engineer]

Yeah, it's great until you make a minor typographical error and don't realize it.

This is just as easy in an analytical calculation... At least if you do FEA and analytical analysis together (if possible) you can compare and hope they're in same ballpark. If not, time to delve deeper and decide which is right.

What I'm talking about is an intuitive feel for the process or device that one is engineering. These computer programs are really really cool and they can result in some interesting and highly optimized designs. However, I've also seen too many stupid mistakes made by people who had no idea what their software was doing.

The software is a tool, and only as effective as the person using it. I too know lots of people that think the software can analyze a design better than they can- obviously a recipe for failure. Intuitive feel only gets you half way there IMO, the rest is critically applying engineering skills (including software analysis) to get the final solution.

Before you even start walking toward the computer, you need to have some idea of what you're expecting from it. If the results do not look right to you, YOU MUST INVESTIGATE!

That's the point of comparing to analytical calcs. Otherwise it's hard to know if what the program is telling you is correct (unless you have an "intuitive feel" for it, not good enough in my organization).

My designs are refined with software, not driven by them.

You're pretty much right, my point was just that the final result of stress or safety factor from a complex design's analysis comes from the FEA analysis, not the analytical calc that was used to see if the FEA was in the ballpark.

However, the key thing is, when both of you have concluded that the FEA software has given the correct answer you have to run with that answer. Unless you ever feel like solving millions of simultaneous differential equations recursively to proof check the code.

Exactly.