Can I simulate the movement of the iron dust in a static or dynamic magnetic field in the ELMER?

General discussion about Elmer
Post Reply
Jenwel
Posts: 27
Joined: 22 Jun 2017, 07:07
Antispam: Yes
Contact:

Can I simulate the movement of the iron dust in a static or dynamic magnetic field in the ELMER?

Post by Jenwel »

Hi, Dr. Raback,
I wonder can I simulate the movement of the iron dust in a static or dynamic magnetic field in the ELMER? Which solver should I use?
As we know the iron dust is much smaller, and the size is approximate 4-20 microns. In my case, the magnetic field is produced by a coil with direct current, with much greater size, of course. Actually, I want to optimize the dimension of the coil and the value of the current.
Thank you in advance.
Hope you and your family are healthy and safe during these uncertain and unprecedented times.

Best Regards,
Dr. Jenwel
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Can I simulate the movement of the iron dust in a static or dynamic magnetic field in the ELMER?

Post by kevinarden »

I have simulated the movement of an iron plate using a magnet. The sif turns the magnet on and off in order to vibrate the plate. Moving iron dust would require imagining how to model the dust. It may be worth considering applying magnetic properties to a fluid and treating the dust as a fluid, perhaps a magnetic field would make the magnetic fluid flow.
displace.sif
(5.24 KiB) Downloaded 218 times
magnet.png
magnet.png (44.9 KiB) Viewed 2929 times
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: Can I simulate the movement of the iron dust in a static or dynamic magnetic field in the ELMER?

Post by raback »

Hi

You can certainly model magnetic fields in Elmer.

An essential question for the iron dust is whether they act as tracer particles or also affect the magnetic field?

Assuming that these are tracer particles you could try to couple the ParticleDynamics solver with the magnetic fields. Unfortunately there is currently no force of such type implemented but still I feel this would be the simplest way to proceed.

What has been implemented already is electrostatic force acting on particle. So effectively adding f=-q*\grad(phi). What you could do is to formulate similar force in terms of the magnetic field and dipole moment of the particle. Adding that would then be pretty smooth operation. If you would have a good test case with analytical solution that would be even better.

If you're looking for bidirectional coupling then that would be a little more difficult... You would then have to estimate the effect of the particles to the electrical properties of the medium.

-Peter
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Can I simulate the movement of the iron dust in a static or dynamic magnetic field in the ELMER?

Post by kevinarden »

The magnetic nodal forces are calculated for a magnetic fluid but they are not active/used in the fluid solver. It seems magnetic forces are only implemented in the elasticity solvers.
Jenwel
Posts: 27
Joined: 22 Jun 2017, 07:07
Antispam: Yes
Contact:

Re: Can I simulate the movement of the iron dust in a static or dynamic magnetic field in the ELMER?

Post by Jenwel »

raback wrote: 26 Dec 2020, 23:47 Hi

You can certainly model magnetic fields in Elmer.

An essential question for the iron dust is whether they act as tracer particles or also affect the magnetic field?

Assuming that these are tracer particles you could try to couple the ParticleDynamics solver with the magnetic fields. Unfortunately there is currently no force of such type implemented but still I feel this would be the simplest way to proceed.

What has been implemented already is electrostatic force acting on particle. So effectively adding f=-q*\grad(phi). What you could do is to formulate similar force in terms of the magnetic field and dipole moment of the particle. Adding that would then be pretty smooth operation. If you would have a good test case with analytical solution that would be even better.

If you're looking for bidirectional coupling then that would be a little more difficult... You would then have to estimate the effect of the particles to the electrical properties of the medium.

-Peter
Dear Peter,

Sorry for the late response, and thank you again for the helpful advices!
Unfortunately, that project is not launched. Indeed, coupling the ParticleDynamics solver with the magnetic fields is a simplest choice.

Best Regards,
Jenwel
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: Can I simulate the movement of the iron dust in a static or dynamic magnetic field in the ELMER?

Post by raback »

Hi Jenwel,

Luckily there were other things in the meantime...

As it happens there was also a case involving magnetic fields and particle dynamics. Hence it is nowadays possible to 1st compute the magnetic fields and then follow particles taking the Lorentz force into consideration just giving name for the "Magnetic Field Variable" in ParticleDynamics. This is without any backcoupling though...

-Peter
Post Reply