Self-Study in Electromagnetism

[icesalmon]

Recently ( due to me moving forward with my undergraduate studies in electrical engineering ) I purchased a plethora of books, aimed at self study, in an attempt to try and understand the concepts behind electrostatics, electrodynamics, and magnetostatics etc. I read reviews on books that would help solidify my understanding of these concepts and Griffith's Electrodynamics seemed like a good choice to start with, after working through it the past few weeks I feel more lost and confused than when i started on the subject. I really want to be able to understand the concepts behind electromagnetism and have the ability to do the problems in this and other books but I'm not sure this is the right way to go about it. I'm curious how an electrical engineer would recommend going about this process and any recommendations on books or resources to understand this topic more thoroughly would be massively appreciated.

 

[jtbell]

Have you studied a first-year calculus-based introductory physics course that uses a textbook such as Halliday/Resnick/Walker, and calculus including multivariable (vector) calculus? If not, then that is where you should begin. In the US at least, students do not take a course that uses Griffiths until after they have completed the lower-level intro course and studied enough calculus.

I'm speaking from the point of view of a physicist. I don't know the details of EE curricula, but I do know that engineering curricula generally start with introductory physics and calculus courses that are the same as or similar to the ones that physics students take.

 

[ZapperZ]

Recently ( due to me moving forward with my undergraduate studies in electrical engineering ) I purchased a plethora of books, aimed at self study, in an attempt to try and understand the concepts behind electrostatics, electrodynamics, and magnetostatics etc. I read reviews on books that would help solidify my understanding of these concepts and Griffith's Electrodynamics seemed like a good choice to start with, after working through it the past few weeks I feel more lost and confused than when i started on the subject. I really want to be able to understand the concepts behind electromagnetism and have the ability to do the problems in this and other books but I'm not sure this is the right way to go about it. I'm curious how an electrical engineer would recommend going about this process and any recommendations on books or resources to understand this topic more thoroughly would be massively appreciated.

The problem here is that you simply stated that you "... feel more lost and confused..." without describing exactly the nature of the problem. You also neglected to describe your background, i.e. what prerequisites do you already possess? As jtbell has asked, do you have knowledge of, say, vector calculus? Have you taken the standard General Physics sequence of courses that include basic E&M?

Note that you simply can't walk into a doctor's office and simply say "I don't feel well" and expect to have an accurate diagnose. And asking for resources recommendation right now is similar to asking for that doctor to give you medication without knowing what your medical problem is. There is no point in providing you even more resources without diagnosing the source of the problem.

Zz.

 

[icesalmon]

Have you studied a first-year calculus-based introductory physics course that uses a textbook such as Halliday/Resnick/Walker, and calculus including multivariable (vector) calculus? If not, then that is where you should begin. In the US at least, students do not take a course that uses Griffiths until after they have completed the lower-level intro course and studied enough calculus.

I'm speaking from the point of view of a physicist. I don't know the details of EE curricula, but I do know that engineering curricula generally start with introductory physics and calculus courses that are the same as or similar to the ones that physics students take.

I have taken calculus 1,2,3 including vector calculus, I have studied differential equations and linear algebra. I have studied calculus based mechanics, and electricity and magnetism. We have not used halliday/resnick/walker but I did use tipler and knight as supplemental texts to the course, we used an online system called "SmartPhysics" or "Flipitphysics" which was used heavily at the university of illinois urbana champaign for their engineering department.

 

[icesalmon]

The problem here is that you simply stated that you "... feel more lost and confused..." without describing exactly the nature of the problem. You also neglected to describe your background, i.e. what prerequisites do you already possess? As jtbell has asked, do you have knowledge of, say, vector calculus? Have you taken the standard General Physics sequence of courses that include basic E&M?

Note that you simply can't walk into a doctor's office and simply say "I don't feel well" and expect to have an accurate diagnose. And asking for resources recommendation right now is similar to asking for that doctor to give you medication without knowing what your medical problem is. There is no point in providing you even more resources without diagnosing the source of the problem.

Zz.

Sure I can read through sections on electric flux / gauss's law in differential and integral form or laplaces / poisson's equations and remember some symbols and the process by which, say, electric flux is determined for a generic surface such as a hollow or solid sphere, planar conductor or infinitely long charged wire or even a cylinder or some system of conducting concentric spheres, I'm worried that this "understanding" is completely isolated to these types of problems which i had encountered in my course on E&M previously. I want to understand the ideas behind electromagnetism to the point where I can begin to think about things in the real world that are built from these concepts and attack applied problems in circuit analysis and power systems analysis.

icesalmon

 

[ZapperZ]

Sure I can read through sections on electric flux / gauss's law in differential and integral form or laplaces / poisson's equations and remember some symbols and the process by which, say, electric flux is determined for a generic surface such as a hollow or solid sphere, planar conductor or infinitely long charged wire or even a cylinder or some system of conducting concentric spheres, I'm worried that this "understanding" is completely isolated to these types of problems which i had encountered in my course on E&M previously. I want to understand the ideas behind electromagnetism to the point where I can begin to think about things in the real world that are built from these concepts and attack applied problems in circuit analysis and power systems analysis.

icesalmon

But this is different than being "... more lost and confused ..." about the subject.

It appears that you're trying to relate this to what you are doing in EE. Remember that what you are looking at is the study of EM FIELDS. This knowledge "sneaks" into "circuit analysis and power system" not in the obvious way. One may look at the various components of such circuits or systems and see how EM theory is in action (induction, Faraday's law, etc..), but you're not going to be using EM theory directly in those topics.

In my experience, many EE students, especially at the graduate level, tend to take advanced classical EM fields (i.e. at the graduate level using Jackson's text) when they are either concentrating on being RF engineers, working on waveguides, designing antennae, etc... These are areas which make direct use of EM fields in various geometries, and where direct use of Maxwell equations will be evident.

Zz.

 

[icesalmon]

Okay, so for the sake of simply trying to understand EM Fields might you recommend anything for me to continue building my knowledge alongside Griffith's text? I am not sure how much Griffith's helps to build the intuition of electrical engineers, but I would also like to be able to work on that aspect of my skill set as well.

Also, what is the difference between electrodynamics and electromagnetics?

 

[K Murty]

First understand circuits. Then at the end will be transformers and basics of electromagnetism, make sure to understand these really well. Then move onto higher concepts.

One book which bridges circuits and EM seemingly well is Fawaz Utaby fundamentals of applied electromagnetism.

 

[icesalmon]

First understand circuits. Then at the end will be transformers and basics of electromagnetism, make sure to understand these really well. Then move onto higher concepts.

One book which bridges circuits and EM seemingly well is Fawaz Utaby fundamentals of applied electromagnetism.

when you say "first understand circuits" do you mean ALL of circuit analysis? what parts if you don't? also, aren't the concepts behind circuits just the same as the concepts of electromagnetism (electric potential, current, induction, electric fields, magnetic fields etc.) so wouldn't i want to get the best possible understanding of all of those things before i really start trying to study circuit analysis?

icesalmon

 

[ZapperZ]

when you say "first understand circuits" do you mean ALL of circuit analysis? what parts if you don't? also, aren't the concepts behind circuits just the same as the concepts of electromagnetism (electric potential, current, induction, electric fields, magnetic fields etc.) so wouldn't i want to get the best possible understanding of all of those things before i really start trying to study circuit analysis?

icesalmon

Let's start with an example of Ohm's Law. If you study E&M or just know Maxwell Equations, you'll never be able to derive Ohm's law.

You get it from Statistical Mechanics, and maybe some application of semi-classical theory of Drude Model to derive Ohm's law and know where that came from.

So even if you did study E&M first, it doesn't mean that you get all the microscopic origins of why things do what they do in those circuits. In fact, the properties of semiconductors that make up your pn-junctions and transistors will require you know about the semiconductor band gaps, and that is already solid state physics, not E&M.

A lot of people learn about circuit analysis without needing E&M beyond General Physics courses. Like I said earlier, EE majors who deal with RF fields, etc. are usually those who do need to know quite a bit of the same level of E&M that physics undergraduates and graduates know.

Zz.

 

[K Murty]

when you say "first understand circuits" do you mean ALL of circuit analysis? what parts if you don't? also, aren't the concepts behind circuits just the same as the concepts of electromagnetism (electric potential, current, induction, electric fields, magnetic fields etc.) so wouldn't i want to get the best possible understanding of all of those things before i really start trying to study circuit analysis?

icesalmon

Hello IceSalmon,

My apologies for not making the post clearer and more polite.

As you know, it is not really practical for you to start learning such higher theories at first. I am an electrical engineering student and I would like for you to watch a brilliant video which explains what you pointed out, that the circuit concepts are the same as those of EM, but under certain constraints:

I recommend you to watch the whole video, as he derives KCL KVL and lumped elements from maxwells equations.
Mr Anant Agarwal also has published a brilliant book in which he derives KCL, KVL from maxwells equations and electromagnetism.
Here are the pictures of the first two pages of his chapter:

As to your question, in my opinion it is not possible to know those things in detail to the extent that you want, mainly because of time and for the mathematics involved.
Circuit analysis exams test knowledge about circuits, not primarily EM but introduces basic concepts of EM.

The video linked above by Mr Anant Agarwal explains the layers of abstractions.
It is not necessary to know it all by heart. Indeed not even possible.

I recommend the path I am following, first I got to know circuit analysis really well, follow a book like Boylestadt (traditional) and Anant Agarwals circuits and linear electronics book too.

After that, take if possible, some courses on electromagnetism and transmission lines. Circuit analysis is really nice and easy to understand, and a good start to understand electricity especially as you study capacitances, inductances, magnetic circuits, and trasnformers. This is the foundation for your higher knowledge in EM.

So what SHOULD ONE understand first? Or rather, what was REQUIRED in my program to understand?
Well the whole of boylestadt was very important.

However, some parts were not in boylstadt that I found in Anant Agarwals book.
These were the chapters on RL, RC circuits, RLC and higher order circuits, transients and the solution to such differential equations I learned in mathematics but found easier to understand in prof Agarwals book.

I know any year long circuits course would cover the entirety of boylestadt, however boylestad does not have RLC and higher order differential equation circuits, for which books like prof Agarwals are better. Hayts engineering circuit analysis is also good for transients and challenging questions. The second circuits course that was required to pass in our program taught transient analysis, filter design and system design, transformers and magnetic circuits, and signal measurement techniques and mathematics behind them.Despite having finished most of these books there are still some very challenging questions in Hayt that I am unable to solve.

Have you taken any circuits class yet? Circuit analysis is very important to understand well for EEs. So I advise you to focus on that, there will be basics of EM and more in an year long circuit analysis course. But first priority should be solving as many challennging circuits as possible.

 

[icesalmon]

Let's start with an example of Ohm's Law. If you study E&M or just know Maxwell Equations, you'll never be able to derive Ohm's law.

You get it from Statistical Mechanics, and maybe some application of semi-classical theory of Drude Model to derive Ohm's law and know where that came from.

So even if you did study E&M first, it doesn't mean that you get all the microscopic origins of why things do what they do in those circuits. In fact, the properties of semiconductors that make up your pn-junctions and transistors will require you know about the semiconductor band gaps, and that is already solid state physics, not E&M.

A lot of people learn about circuit analysis without needing E&M beyond General Physics courses. Like I said earlier, EE majors who deal with RF fields, etc. are usually those who do need to know quite a bit of the same level of E&M that physics undergraduates and graduates know.

Zz.

So even if i wanted to study this stuff in depth in hopes of getting a strong core, that understanding would have holes in it still. I've got to say, that's frustrating but i suppose even when doing math, hardly anyone understands why 1+1 = 2, yet almost everyone over the age of 3-4 can tell you what the solution is. Maybe someday i'll find the time to really get back to "basics" with some of these underlying principles. I suppose going forward i'll fine tune what I need to know based on where my path takes me. Regardless, i'll be back to this thread later on for more guidance / directives. For now i'll just hit the books.

 

[icesalmon]

Well the whole of boylestadt was very important.

However, some parts were not in boylstadt that I found in Anant Agarwals book.
These were the chapters on RL, RC circuits, RLC and higher order circuits, transients and the solution to such differential equations I learned in mathematics but found easier to understand in prof Agarwals book.

Despite having finished most of these books there are still some very challenging questions in Hayt that I am unable to solve.

Have you taken any circuits class yet? Circuit analysis is very important to understand well for EEs. So I advise you to focus on that, there will be basics of EM and more in an year long circuit analysis course. But first priority should be solving as many challennging circuits as possible.

I'm working out of boylestadt at the moment and hayt will be arriving to me in the next week so i plan on working through that soon. I have taken circuit analysis and digital systems so I've dealth with both discrete and continuous circuits at this point. My goal in the next two months or so is to really solidify as much of E&M and Circuit Analysis as I can. I also am planning on working on understanding the theory behind Fourier analysis, laplace transforms and z-transforms as well as differential equations to start for self study in signals and systems by oppenheim, as well as my QM / Solid State Physics for a course called "electronic circuits" which studies the ideas behind p/n junctions, BJT etc. I also will start working out of Power Systems Analysis and Design by Arthur Gross, Microelectronics by sedra/smith and control systems by Ogata. I'm really just trying to get a feel for what the next few years will hold and to fine tune my interests so i can concentrate in something (i'm leaning towards power engineering at this point). Your post was never impolite, so don't mention it!

 

[Joshy]

I really liked Ulaby's book recommended above, which we used for my upper division electromagnetism module; I had previous exposure in the lower division physics using Young and Freedman.

I personally feel the story sticks while bookkeeping the units, and thinking about waves used in optics, sound, and fluid waves because I can picture it a lot more easily compared to an electric field, characteristic impedance, etc. Some of the RF engineers I spoke to struggle with KCL and KVL after a while, and I think it's because they don't use it frequently. Problems become very geometrical playing around with a lot of dimensions and understanding electrical characteristics of materials. I do like that it was mentioned because it's a lot like learning how to solve for the area of a rectangle using base times height, but later finding out about integrals (thinking of circuitry as waves instead and realizing - like integrals - that the method works on simple problems as well)... I still think it's too broad of a learning plan although it's good to know.

People continuing in RF will probably mention Pozar's book in Microwave Engineering. I hear about it all the time and it's really a great book, but it also assumes some background. It's not an easy topic. I think people even with the guidance from their education and in their careers... there are just times where it seems like they've never done a transmission line problem before.

Materials sciences and semiconductors physics could be another approach if the RF engineering isn't sinking in. Learning how to solve circuits using simplifications and then trying to relearn it as waves might be a bigger barrier than thinking about how the electric field is used to create a channel in transistors, what happens when you forward and reverse bias a diode, or more about what these units such as surface charge and boundary conditions mean when looking at a device. If one story doesn't work out for you, then try another ;)

 

[K Murty]

I'm working out of boylestadt at the moment and hayt will be arriving to me in the next week so i plan on working through that soon. I have taken circuit analysis and digital systems so I've dealth with both discrete and continuous circuits at this point. My goal in the next two months or so is to really solidify as much of E&M and Circuit Analysis as I can. I also am planning on working on understanding the theory behind Fourier analysis, laplace transforms and z-transforms as well as differential equations to start for self study in signals and systems by oppenheim, as well as my QM / Solid State Physics for a course called "electronic circuits" which studies the ideas behind p/n junctions, BJT etc. I also will start working out of Power Systems Analysis and Design by Arthur Gross, Microelectronics by sedra/smith and control systems by Ogata. I'm really just trying to get a feel for what the next few years will hold and to fine tune my interests so i can concentrate in something (i'm leaning towards power engineering at this point). Your post was never impolite, so don't mention it!

For power systems prof DP Kotharis lectures are the best that I have found online:

His published books are also excellent.

 

[ZapperZ]

So even if i wanted to study this stuff in depth in hopes of getting a strong core, that understanding would have holes in it still. I've got to say, that's frustrating but i suppose even when doing math, hardly anyone understands why 1+1 = 2, yet almost everyone over the age of 3-4 can tell you what the solution is. Maybe someday i'll find the time to really get back to "basics" with some of these underlying principles. I suppose going forward i'll fine tune what I need to know based on where my path takes me. Regardless, i'll be back to this thread later on for more guidance / directives. For now i'll just hit the books.

You have an unrealistic expectation that studying one subject will answer everything. As you get older and as you progress in your studies, you'll discover how naive of an outlook that is.

The field of solid state/condensed matter physics is the LARGEST division and topic area in physics. This is the area of physics that is the foundation for many of the application, devices, and components that you use in circuit theory and analysis. This is the area of physics that "explains" why these things work the way they do, right down to the basic physics of transport of charges in a conductor! And yet, this area of study is a separate topic from E&M. It is a subject area that makes use of E&M and quantum mechanics.

This is why I explained why you will not get what you think you will with just studying E&M alone, because there a whole group of knowledge that goes beyond that to bridge E&M and circuit theory.

Zz.