How Important Is Math In Electrical Engineering Technology Degree?/job

[ROBOT]

I just started work on my EE degree. It has been around 6-7 years since I had high school math. (I took all 3 required classes freshman year.) So I took intermediate algebra first as a refresher. (I am glad I did!)

Now how important is all this advanced graphing, matrices, factoring, simplifying, and systems of equations going to be on the job? I have never seen any of the EE's at my current job do any complex math. Just basic stuff calculated of datsheets.

I basically want to know do I really need to fully understand these concepts (like Calc and differential equations) or will a basic knowledge just enough to pass the class with some good notes be enough?

Let's assume the job does a few things:

Makes simple normally 2-4 layer, low frequency, 5v or 3v3 pcb's using off the shelf microcontrollers and other parts.

Works with customers on creating custom designs.


From my understanding now a days even advanced stuff like ASIC design is almost all done with off the shelf libraries and simulation software. A lot of stuff is being replaced by computer simulation.

[jaxbird]

In my experience you'll need very little math in your day to day job, but occasionally you will need to know how things work, as in knowing the relationship between various things and knowing how specific values are calculated (not as in knowing specif complex formulas off the top of your head, but at least knowing the relations) Also knowing whether specific values are reasonable or not and being able to tell by estimation that something is not right.

[ebclr]

The big difference between a technician and the engineer is the math background, Have two kinds of engineers the datasheet application cut and paste, and the math modeling guy . the seconds one normally have triple salary 

[CatalinaWOW]

I have worked with many engineers who had little grasp of or use for mathematics.  But the more you want to do, the better you want to understand what you are doing, and the better you want to do it, you will benefit from all of the math you can stand.  You may use some of it only once a decade, others will get annual touches, more will be touched more frequently. 

It is kind of like asking if you can do electronics without test equipment.  You can do quite a bit with your fingers, eyes and ears.  Add a DMM and many have all they need.  But add oscilloscopes and logic analyzers and you can find and do a lot more.  Even more sophisticated instruments can ferret out problems or allow design decisions that are not always necessary, but often beneficial.

[photon]

Mathematics is an ancient science. The civilizations that have developed it have been successful and strong. Ignore it at your own peril.

[joseph nicholas]

In many professions, including mine, after you start working there will be others in your profession that will try to determine if you are a charlatan and  try to get you disbarred or what have you.  My advise would be to find out what stuff they may use to determine if you actually went to the school you say you attended. It might be good to learn simple stuff like how to read the resistance color code.  I doubt you would be challenged about doing anything mathematical one the fly.

[tom66]

Mathematics is very important in engineering. Having an advanced understanding of algebraic principles, statistics and at least a reasonable understanding of calculus (logs, differentials, integrals and so on) is pretty much essential. Beyond that? Set theory, triple integrals, number theory and advanced statistically modelling (Matlab etc)...not that useful for the everyday engineer.

I would also say that the ability to make quick approximations and guesses within 20% of the actual answer is very useful, you can pretty much say whether something will work or not in five seconds. Learn your powers of two, as well.

[skipjackrc4]

I would say that the ability to manually perform integration and solve differential equations isn't really that useful for most engineers.  However, the understanding of the physical meaning behind vector operations, Fourier transforms (and general basis vector decomposition), convolution, etc... is very necessary, and "engineers" who don't understand these things wind up looking like complete morons.  In my experience, you need to intuitively understand what the math says, even if you can't actually calculate it by hand.  You should also understand at least basic statistics, which is something that I'm working on improving myself.  I've found that it's often necessary to write your own simulation code when commercial packages won't do what you want--this obviously requires math.

[ajb]

It's worth noting that it depends a lot on what subset of engineering you go into.  If you were to take all of the work of electrical engineering done across all subspecialties and industries and tally it up, probably 95% requires, at the most, basic algebra.  That last 5%, however, covers a wide range of math from algebra on up, and while a lot of the 5% is concentrated in the work of specialists in RF or semiconductor design or whatever, the rest of it is sprinkled all through the work of all the other EEs in the world.  So you might wind up in an area where 99% of your work only requires arithmetic, but there will be that 1% of situations where you'll need to be able to crack more advanced problems. 

Since you mentioned microcontrollers, even if you aren't doing anything computationally intensive embedded software in particular requires a very thorough understanding of numeric representation and algebraic manipulation.  In order to get correct results it's imperative to be able to spot where you'll lose precision, overflow a value, etc, and you'll need to know how to transform a given expression into a form that an embedded system can evaluate efficiently and accurately. 

[IanB]

Makes simple normally 2-4 layer, low frequency, 5v or 3v3 pcb's using off the shelf microcontrollers and other parts.

That's not an engineer. That's a hacker. You don't need a degree for that.

One thing that comes up frequently in this forum is the stability of feedback loops and control systems. Things like characteristic functions, poles and zeros, phase shifts and eigenvalues will become important when you try to solve such problems. You may consider that to be analog and not digital electronics. But in the real world all electronics are analog. The perfect square wave with sharp corners does not exist outside of textbooks.

You should expect that to be a competent engineer you will need a firm grasp of mathematics. Even if you don't use it all the time, learning mathematics will help train your mind to analyze and solve complex problems.

[nctnico]

In engineering you need math. I think I at least solve (re-arrange) one equation for every project.

[LabSpokane]

All engineering requires math - algebra, linear algebra, calculus, differential equations ordinary and partial, so on and so forth.  If you want the degree, do the work.

[TimFox]

As I tell high-school students, not only is there algebra (etc.) in the "real world", but it's all story problems.

[sarepairman2]

depends where you work. mathematical analysis may be seen as a waste of time by superiors in certain industries

more valuable is often the accounting 
you can go far sometimes by the simple "yes, but how much does this solution cost?"

[tggzzz]

Now how important is all this advanced graphing, matrices, factoring, simplifying, and systems of equations going to be on the job? I have never seen any of the EE's at my current job do any complex math. Just basic stuff calculated of datsheets.

I basically want to know do I really need to fully understand these concepts (like Calc and differential equations) or will a basic knowledge just enough to pass the class with some good notes be enough?

An engineer approaching a job will first determine what they do and don't need to understand and do to successfully complete the job. Frequently that will involve doing quick back-of-the-envelope calculations - which requires understanding their validity and limits. If quick calculations indicate some aspect is near the boundary of validity, then more accurate detailed calculations are required.

To do all that requires a good familiarity and understand of mathmatics.

Let's assume the job does a few things:
Makes simple normally 2-4 layer, low frequency, 5v or 3v3 pcb's using off the shelf microcontrollers and other parts.
Works with customers on creating custom designs.

Those are technican-level jobs. Nothing whatsoever wrong with that, but they barely touch on the full range of engineering activities.

From my understanding now a days even advanced stuff like ASIC design is almost all done with off the shelf libraries and simulation software. A lot of stuff is being replaced by computer simulation.

Simulation is, by definition, a series of tests. Tests never prove a design works - at best they can indicate some ways in which a design doesn't fail -- since the next test could show a problem.

[retrolefty]

All engineering requires math - algebra, linear algebra, calculus, differential equations ordinary and partial, so on and so forth.  If you want the degree, do the work.

 And the above math requirements are probably the biggest wash-out cause for all engineering degrees. Get to work if you want the degree.

 

[Maxlor]

In the last decade, I've needed quite a bit more of the math I learned at the uni than I thought I would, but far from everything that was taught. By this point I've forgotten more than half of it for sure. But I also noticed, if I need some of those things I've forgotten, it comes back quickly, often just the terse descriptions on mathworld suffice.

So I'd say, learn as much as you can, even though you don't know at this point what you'll use (and if you think you do, you're probably wrong,) so you can leave the uni with as big a toolbox as possible. It'll give you a good competitive advantage.

[coppice]

There's a very wide range of engineering jobs. Some require hardly any maths at all. Others require serious maths almost every day. The percentage of engineering jobs requiring more than basic maths day to day seems quite small. In many larger organisations there are a couple of well known mathies, to whom most of the seriously mathematical questions go. If you are in a small company, or work alone, you will need to cope with anything the work throws at you.

[D3f1ant]

The complexity varies greatly, but I doubt a single day goes by when there isn't something that needs calculating,  even if it's only the current through a Led. A good broad spectrum book that I wish I had as an undergraduate is Engineering Mathematics by Stroud, its the only dedicated  maths book I still have on my bookshelf.

[vodka]

In the last decade, I've needed quite a bit more of the math I learned at the uni than I thought I would, but far from everything that was taught. By this point I've forgotten more than half of it for sure. But I also noticed, if I need some of those things I've forgotten, it comes back quickly, often just the terse descriptions on mathworld suffice.

So I'd say, learn as much as you can, even though you don't know at this point what you'll use (and if you think you do, you're probably wrong,) so you can leave the uni with as big a toolbox as possible. It'll give you a good competitive advantage.

I think that the maths bad focused because the teachers tend to teach it as when hadn't computer. So i believe that resolve  math mechanically haven't sense today.
The most logic would be teach the application of the maths at the labour or real world .

Maxlor . all the world have forgotten the stuff that have learnt at past, who say otherwise LIE. 

[tggzzz]

I think that the maths bad focused because the teachers tend to teach it as when hadn't computer.

Maths should be taught without using a (black box) computer, for several reasons:

• you will understand why the maths works (and when you it is safe to simply skip directly to the results)
• it will help you mentally visualise what's happening in your systems (so you can synthesise new concepts and debug existing systems)
• you will develop an understanding of what you can and can't predict, and why (crucial for any engineer with significant responsibility, and avoid having salesman blind you with lies)
• you will learn how to formally understand and analyse a system (ought to be necessary before any large investment is made)

So i believe that resolve  math mechanically haven't sense today.

Those of us that have been around a while have frequently seen people that have accepted ridiculous results simply because that's the answer that came out of a black box.

The most logic would be teach the application of the maths at the labour or real world .

Not at university for an engineer - the prime responsibility is to teach technical fundamentals and how to learn. Time spent learning which button to press on software program X is a waste of time.

For technicians, the answer may be different - knowing today's package may be what it necessary to get them a job tomorrow. But it won't be of much use in 5 years, whereas the maths certainly will.

[Maxlor]

I think that the maths bad focused because the teachers tend to teach it as when hadn't computer. So i believe that resolve  math mechanically haven't sense today.
The most logic would be teach the application of the maths at the labour or real world .

I don't think there's much of a difference at the uni level, really. Even when you use programs to help you, you still need to understand what they're doing to use them correctly.

[EEVblog]

Now how important is all this advanced graphing, matrices, factoring, simplifying, and systems of equations going to be on the job? I have never seen any of the EE's at my current job do any complex math. Just basic stuff calculated of datsheets.

Yep, that's the majority of electronics design jobs in my experience.

I basically want to know do I really need to fully understand these concepts (like Calc and differential equations) or will a basic knowledge just enough to pass the class with some good notes be enough?

Uni level math is not easy for most. You more likely than not won't pass it without understanding it. Although I have seen math classes were it was simply enough to study past exams to bum your way through it because the teachers were lazy and didn't change much between semesters.
No one hiring you will care what score you got in your math classes, unless it's for some more obscure job that requires a lot of math. So technically, yes, you can get away with not really understanding it, but that could come back to bite you.

[EEVblog]

An engineer approaching a job will first determine what they do and don't need to understand and do to successfully complete the job. Frequently that will involve doing quick back-of-the-envelope calculations - which requires understanding their validity and limits. If quick calculations indicate some aspect is near the boundary of validity, then more accurate detailed calculations are required.

Yes, but the issue is with the type of math required.
In the majority of the cases it's relatively simple stuff.
Most practical electronics engineers won't be solving triple integrals.

A practical electronics engineer is likely going to need statistics more than calculus in their career for example.
To do all that requires a good familiarity and understand of mathmatics.

[tggzzz]

No one hiring you will care what score you got in your math classes, unless it's for some more obscure job that requires a lot of math.

While they didn't care about a score, two of my employers did explicitly care about maths ability during interview - and I wouldn't have got a job without the maths.

With one I discussed the problems I had previously had (in my first job out of uni) with analytical solutions to PLL lock-in when noise is present, and how N-Path filters and Z-domain transforms circumvented the problem. (I had previously needed the maths when designing a test set that measured the attenuation in optical fibres)

In another we guided candidates through the maths behind IQ modulation and demodulation - which is used in most RF devices today.

So technically, yes, you can get away with not really understanding it, but that could come back to bite you.

Yes indeed. Too often the people that baldly claim "maths is unnecessary" don't understand it themselves and/or aren't involved in anything novel.