Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Trouble making magnetic flux lines appear

Please login with a confirmed email address before reporting spam

Hey everyone,
I am trying to test the flux leakage from a solenoid (in DC) given a variety of core shapes. I am new to Comsol and, after a bit of playing, decided to simplify the problem to the field generated by a cylinder, and then to build the solenoid back up to make the analysis more thorough.
I am using the AC/DC module, and I have a cylinder with potentials at both ends. When running Comsol, the potential shows up on the cylinder, but I get no flux lines.
I've tried grounding one end, adding surface currents ( under boundary settings and subdomain settings and edge settings).
I feel I am missing something elementary here; as I mentioned, I am new to Comsol so any help would be appreciated!
Further (and finally), is there any way to create a helix or solenoid in 3.5? I found tutorials for 4.2 but the interface is completely different so I have no idea how to start it.
Thanks in advance!
Jonathan


3 Replies Last Post 29 gen 2013, 05:26 GMT-5

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 28 gen 2013, 03:30 GMT-5
Hey,

my guess is:
- you don't see any flux lines because there is no magnetic flux
- as you can find in postprocessing, magnetic flux density and magnetic vector potential is zero in your whole model
- the reason for this might be that there is no source term in your magnetic and electric fields interface
- try to bring an electric current density source into that interface with the current density from the electric currents interface (ec.Jx, ec.Jy, ec.Jz) as the source expression terms

By the way I don't understand your model. You have applied a voltage of 300 V to the edges of the cylinder along its sides. This goes straight to the grounded boundary. I am not sure if this is intended, but it doesn't seem to be physically correct.

On the other hand there is maybe some conversion error, since I opened your version 3.5 model with my 4.3 installation...

AFAIK the helix primitive was introduced in a later version than 3.5, maybe 4.1 or 4.1a. In an earlier version you can create a helix with some more effort, like approximating it with other primitives, using CAD import or a suitable MATLAB script...
Hey, my guess is: - you don't see any flux lines because there is no magnetic flux - as you can find in postprocessing, magnetic flux density and magnetic vector potential is zero in your whole model - the reason for this might be that there is no source term in your magnetic and electric fields interface - try to bring an electric current density source into that interface with the current density from the electric currents interface (ec.Jx, ec.Jy, ec.Jz) as the source expression terms By the way I don't understand your model. You have applied a voltage of 300 V to the edges of the cylinder along its sides. This goes straight to the grounded boundary. I am not sure if this is intended, but it doesn't seem to be physically correct. On the other hand there is maybe some conversion error, since I opened your version 3.5 model with my 4.3 installation... AFAIK the helix primitive was introduced in a later version than 3.5, maybe 4.1 or 4.1a. In an earlier version you can create a helix with some more effort, like approximating it with other primitives, using CAD import or a suitable MATLAB script...

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 28 gen 2013, 20:27 GMT-5
Hey,
thanks for the reply! I figured there was no flux on the model at all, but then now my question is how to I make it? I tried adding currents in a) subdomain settings, Qj and Je b) boundary settings by defining the 4 cylinder boundaries of 'current flow' type and defining currents there two.
I would have thought that one of these methods would have allowed current to flow along the cylinder (due to the potential across it) and thereby created a field. Am I missing something?
Would you be able to explain what you mean by this:
an electric current density source into that interface with the current density from the electric currents interface (ec.Jx, ec.Jy, ec.Jz) as the source expression terms

note: I did take the global equations and constants from the 'rail gun' comsol model; would this be a problem? Again, I am very unfamiliar with Comsol so I am not sure what I need to define or not in order to get this to work.
Thanks!
Jonathan
Hey, thanks for the reply! I figured there was no flux on the model at all, but then now my question is how to I make it? I tried adding currents in a) subdomain settings, Qj and Je b) boundary settings by defining the 4 cylinder boundaries of 'current flow' type and defining currents there two. I would have thought that one of these methods would have allowed current to flow along the cylinder (due to the potential across it) and thereby created a field. Am I missing something? Would you be able to explain what you mean by this: an electric current density source into that interface with the current density from the electric currents interface (ec.Jx, ec.Jy, ec.Jz) as the source expression terms note: I did take the global equations and constants from the 'rail gun' comsol model; would this be a problem? Again, I am very unfamiliar with Comsol so I am not sure what I need to define or not in order to get this to work. Thanks! Jonathan

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 29 gen 2013, 05:26 GMT-5
Okay, I think I cannot open and modify your model properly, since it was created with version 3.5 and I do not know what the intention of your simulation is.

But what I meant is that the magnetic interface, which solves for the magnetic vector potential and consequently yields your desired quantity magnetic flux density, has no source term, no excitation. So it won't find a quantity different from zero. I guess a possible excitation would be to use the current density from the electric currents interface. You define some electric potential BCs there that yield a current density, you can visualize that in postprocessing. Using this in the magnetic interface as a source will result in a magnetic field.

Then I saw that you didn't assign a material, but without a material parameter permeability you cannot get a magnetic flux density, because this is B = µ*H.

So try the following:

1) add a source to your magnetic interface
2) add a material with a permeability
Okay, I think I cannot open and modify your model properly, since it was created with version 3.5 and I do not know what the intention of your simulation is. But what I meant is that the magnetic interface, which solves for the magnetic vector potential and consequently yields your desired quantity magnetic flux density, has no source term, no excitation. So it won't find a quantity different from zero. I guess a possible excitation would be to use the current density from the electric currents interface. You define some electric potential BCs there that yield a current density, you can visualize that in postprocessing. Using this in the magnetic interface as a source will result in a magnetic field. Then I saw that you didn't assign a material, but without a material parameter permeability you cannot get a magnetic flux density, because this is B = µ*H. So try the following: 1) add a source to your magnetic interface 2) add a material with a permeability

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.