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 The information provided may be out of date.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.

electric field

Please login with a confirmed email address before reporting spam

The full physical definition of electric field is:

E = -grad(V) - dA/dt

This takes into account electric field from a point charge (electrostatics) and also from a time-varying magnetic field.

It seems strange to me why this formulation is not provided in a pre-defined physics module.

Has anyone else has had similar thoughts?

6 Replies Last Post 30 oct. 2011, 14:38 UTC−4
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 29 oct. 2011, 03:27 UTC−4
Hi

for me that is what COMSOL is using too ;) but you will not see the dA/dt term if you do not turn on temporal time dependent solving.

Are you checking the "Equations" nodes, you will find yourbabies therein, but the list is long ;)

--
Good luck
Ivar
Hi for me that is what COMSOL is using too ;) but you will not see the dA/dt term if you do not turn on temporal time dependent solving. Are you checking the "Equations" nodes, you will find yourbabies therein, but the list is long ;) -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 29 oct. 2011, 12:20 UTC−4
Not all physics modules provided in AC/DC solve for V and A.

For instance, Electric Currents only solves for the steady-state term, grad(V) and Magnetic Fields only solves for the time-varying term, dA/dt.

But there is not a pre-defined physics which includes both.

Is there a reason for this?
Not all physics modules provided in AC/DC solve for V and A. For instance, Electric Currents only solves for the steady-state term, grad(V) and Magnetic Fields only solves for the time-varying term, dA/dt. But there is not a pre-defined physics which includes both. Is there a reason for this?

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 30 oct. 2011, 07:15 UTC−4
Hi

solving for V (a scalar) and A (a vector) can be rather RAM heavy, so if you do not need all, why not reduce the number of dependent variables. ACDC is trickier than HT as you have vector components and cross items to consider.

With MEF you solve for V & A, hence time dependent MEF will solve for both. But with more dependnet variables, you must define many more BC's too, and get them correct w.r.t. each others

--
Good luck
Ivar
Hi solving for V (a scalar) and A (a vector) can be rather RAM heavy, so if you do not need all, why not reduce the number of dependent variables. ACDC is trickier than HT as you have vector components and cross items to consider. With MEF you solve for V & A, hence time dependent MEF will solve for both. But with more dependnet variables, you must define many more BC's too, and get them correct w.r.t. each others -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 30 oct. 2011, 11:40 UTC−4
Ivar,

You have a point about reducing degrees of freedom when possible. So, I suppose that it does make sense to have pre-defined physics with incomplete physical definitions and to couple them, only when, and where, necessary.

But let us assume that you need both V and A, in time:

You mention a time-dependent solution with the MEF formulation, which I thought was magnetostatic only.

Just to clarify, is it possible to do a transient analysis with MEF?

Perhaps I have been wrong all this time, and have been doing it the hard way!

Ivar, You have a point about reducing degrees of freedom when possible. So, I suppose that it does make sense to have pre-defined physics with incomplete physical definitions and to couple them, only when, and where, necessary. But let us assume that you need both V and A, in time: You mention a time-dependent solution with the MEF formulation, which I thought was magnetostatic only. Just to clarify, is it possible to do a transient analysis with MEF? Perhaps I have been wrong all this time, and have been doing it the hard way!

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 30 oct. 2011, 12:56 UTC−4
Hi

you are right, my fault, I tend to forget the EM distinction ACDC / RF. In ACDC the hypothesis is that the geoemtry is "small compared to a wavelength, and the magnetic effects are rather "immediat" w.r.t any time change.

But if you have dA/dt, how do you incorporate vxB terms ?
I suppose you need today to use MF + EC to get the full formulation, in V and A, in time stepping with COMSOL.

Or you split your work to run the structural part in time domain and MEF in stationary, assuming the magnetic effects to be instantenous w.r.t the rotation or motion of your mechancal part. In MEF as in MF you have the velocity Domain conditions that can be added

--
Good luck
Ivar
Hi you are right, my fault, I tend to forget the EM distinction ACDC / RF. In ACDC the hypothesis is that the geoemtry is "small compared to a wavelength, and the magnetic effects are rather "immediat" w.r.t any time change. But if you have dA/dt, how do you incorporate vxB terms ? I suppose you need today to use MF + EC to get the full formulation, in V and A, in time stepping with COMSOL. Or you split your work to run the structural part in time domain and MEF in stationary, assuming the magnetic effects to be instantenous w.r.t the rotation or motion of your mechancal part. In MEF as in MF you have the velocity Domain conditions that can be added -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 30 oct. 2011, 14:38 UTC−4
I was hoping that I was wrong, and there was a way to get a time-dependent formulation of the MEF governing equations, even though the documentation is very clear.

From page 235 of ACDC Guide:

"...Note that the Magnetic and Electric Currents interface only supports the stationary and frequency domain study types—that is, there is no transient formulation available..."

Thanks for the responses Ivar.
I was hoping that I was wrong, and there was a way to get a time-dependent formulation of the MEF governing equations, even though the documentation is very clear. From page 235 of ACDC Guide: "...Note that the Magnetic and Electric Currents interface only supports the stationary and frequency domain study types—that is, there is no transient formulation available..." Thanks for the responses Ivar.

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.