We dont know much about E.Fs do we?

[dE_logics]

At least compared to E.F.

After knowing the charge density of an object, and knowledge of its geometry, you can predict at which places how/how much will the E.F be.



Since there're no monopoles of M.Fs...doing such a thing is not possible (well, actually I think it is, I modeled something to do so, but I don't know if its a common technique or not); moreover we have no alternatives to 'charge density' when it comes to M.F...all we can know about is the pole strength, AND it will vary at different places in the magnet (for instance in a bar magnet the centre section has no pole strength).

Actually I think that is too not possible, cause the definition of pole strength is only applicable to point monopoles AND coulomb's law is too applicable for point monopoles only, so anyway to figuring out the intensity too gets ruled out.

So we can state that there's no perfect way to model a M.F around an objects of various different geometries...we got to do it practically.

 

[jtbell]

In classical electrodynamics, magnetic fields are caused by currents and by time-varying electric fields.

In a magnetostatic situation (current does not vary with time), if we know the distribution of current density, we can calculate the magnetic field that it produces, either via the Biot-Savart law or Ampére's law.

Unfortunately, we can't really "explain" the magnetism of a permanent magnet in classical electrodynamics, because it ultimately comes from the interactions between intrinsic magnetic dipole moments of electrons in the iron or whatever material the magnet is made of, which is a quantum-mechanical effect.

Nevertheless, we can model the field produced by a permanent magnet in terms of a fictitious bound current density which we can relate to the alignment of those intrinsic dipole moments (magnetic polarization = magnetization). At some point we have to feed into the analysis, the magnetic susceptibility of the material on question, which we measure emprically. I don't know how well we can predict these magnetic properties from a quantum-theoretical analysis, although I'm sure it's one of the major topics of condensed-matter physics.

This is rather similar to the way we model the electric field produced by a electrically polarized insulating object (dielectric) by using bound charge densities, which are related to the electric susceptibility of the material, which we measure empirically and (try to) derive from quantum theory of materials.

 

[Born2bwire]

There isn't much in classical electrodynamics that we can't model for want of a fast enough computer with large enough memory. Maxwell's equations, and the Lorentz force pretty much describe the entire behavior of classical electrodynamics. We can solve these using a myriad of methods. Even magnetic fields are fairly easy to model because they still follow specific boundary conditions are are dictated by differential equations. Depending on the frequencies involved, we can simplify Maxwell's equations into the quasi-static, electrostatic, magnetostatic, or circuit equations. The simplifications allow us to use different and more appropriate techniques.

Take a look at Harrington's book on Method of Moments or better yet, Walter Gibson's The Method of Moments in Electromagnetics to see techniques in solving electrostatic and full wave problems. If you want to model the microscopic level, then that can be done using Van Der Waals forces and such like that. Atomic Scale Simulations would be the topic I think and they use different techniques like Monte Carlo simulations and such. I don't know a good book on that, I have a text, probably back in the States, that I used but it wasn't one I would recommend.

Magnetic fields may not have monopoles, but the simplest unit is a dipole. And classical electrodynamics does model things like ferromagnetics as collections of dipoles and such.

 

[dE_logics]

First of all sorry for the wrong subject, it was "We don't know much about M.Fs...do we?"

Unfortunately, we can't really "explain" the magnetism of a permanent magnet in classical electrodynamics, because it ultimately comes from the interactions between intrinsic magnetic dipole moments of electrons in the iron or whatever material the magnet is made of, which is a quantum-mechanical effect.

Yeah, that's what I was thinking about...its primarily cause of the electrons spin, and if the electrons spin, it forms 2 poles, i.e 2 poles are forming within that electron, so this thing creates the confusion.

Nevertheless, we can model the field produced by a permanent magnet in terms of a fictitious bound current density which we can relate to the alignment of those intrinsic dipole moments (magnetic polarization = magnetization).

I'm absolutely not aware of that..."intrinsic dipole moments".

Point is, we don't use it actually (or I won't find it in my course books...its not common) to predict the behaviour of the magnet right?...it's still in R&D.

There isn't much in classical electrodynamics that we can't model for want of a fast enough computer with large enough memory.

Well...we have Nvidia Tesla cards with CUDA enabled to do the job.

1 cuda GPU core requires 4 GB memory (minimum)...and we can get 4 cores in a system at most, i.e ATLEAST 16 GB memory

Magnetic fields may not have monopoles, but the simplest unit is a dipole. And classical electrodynamics does model things like ferromagnetics as collections of dipoles and such.

You mean modeling using small barmagents?

 

[jtbell]

I'm absolutely not aware of that..."intrinsic dipole moments".

Do you know about electron "spin" (intrinsic angular momentum)? It causes the electron to have a fixed intrinsic magnetic dipole moment.

To get an idea of what is happening inside a permanent magnet like a bar magnet, I suggest you look in your textbooks or on line for descriptions of ferromagnetism. It would also probably help to look up paramagnetism and diamagnetism which are the two basic types of "temporary" magnetism that exist only while an external magnetic field is applied to the material. They're simpler to describe and analyze than ferromagnetism, so it might be a good idea to look at them first.

 

[dE_logics]

Ok...thanks for answering, I'll see to it later.