Magnetic dipole in scuff-scatter

[hdierckx]

Hi Homer,

I'm a new user to scuff-em. We would like to study the scattered field of a magnetic dipole source on a metal object of arbitrary shape. I found in the documentation that only electrical dipole sources can be used with scuff-scatter, and for magnetic dipoles one needs to use the API version in order to change the type parameter LIF_ELECTRIC_DIPOLE to LIF_MAGNETIC_DIPOLE. Is there perhaps a simple workaround to get magnetic dipoles working in scuff-scatter?

Thank you

[HomerReid]

Your timing is good---just recently I have added an "advanced mode" to SCUFF-SCATTER which allows the specification of magnetic point sources. Depending on when you last updated your code repository, you may need to pull from GitHub and rebuild to get this feature.

In this mode of operation, instead of using the --psLocation and --psStrength command-line arguments to SCUFF-SCATTER, you will want to write a little text file describing the source(s) of your your incident field(s). Then pass this file as a command-line option using the --IFFile argument.

Here is the documentation for how to write the file listing incident fields:

http://homerreid.github.io/scuff-em-documentation/reference/IncidentFields/#specifying-an-entire-list-of-incident-fields

Note that magnetic point sources are specified by the keyword MPS.

A bonus feature of doing things this way is that you can specify

• multiple point sources at various locations that are all active simultaneously (i.e. a single scattering problem with incident field consisting of a superposition of multiple point sources), and/or
• multiple point sources at various locations that are active individually (i.e. multiple scattering problems in which the incident field for each problem is a point source whose location moves through space).

Please let me know how it works. Feel free to ask further questions, or close the issue if satisfied.

[hdierckx]

I've implemented a toy problem on geophysical sensing now: given a PEC sphere of radius a=0.1m and a magnetic dipole at height h above its surface, what is the secondary magnetic field ($H^s$) in the same position and direction as the source dipole? The exact solution is found in Grant&West, Interpretation Theory in applied geophysics (1965), Fig 17-20. It is independent of the frequency, since skin depth goes to zero for a PEC.

I computed the same curve numerically with scuff-scatter, see graph below. Different colors indicate different levels of mesh refinement (red: clscale=0.5, green: clscale=0.4, blue: clscale=0.3)

I have 2 questions here:

• As I work with macroscopic objects and low frequency (9kHz), I put in the frequency omega in units of $c/1m$ instead of $c/1\mu m$. It seems that the space unit for geometry then also becomes 1m, and E and H scale by $10^18$. Is it correct to work this way, even in the presence of materials with $\epsilon, \mu$ different from their vacuum values?
• The plot above still shows a discrepancy with theory, which does not diminish by taking a finer mesh, or locally refining the mesh near the north pole only. Do you have any suggestions on how to improve the numerical result?

Thank you.