Discussion Forum

Posted: 4 years ago 9 nov 2020, 10:21 GMT-5

Updated: 4 years ago 9 nov 2020, 10:36 GMT-5

First, in regard to your remark that "...our lab was told that..." you need to make that decision *for yourself*, based on the details of the particular problem that you want to solve! If the *only* basis for being "told" that you need to use the RF module was that you want to do models at MHz and "higher" frequencies, then that simply isn't good enough (at least, all by itself). You really need to look at other considerations. For example, does your electrode EEG model need to include (specifically, within the computational space) the propagation of electromagnetic waves? I.e., does it *require* (even if only in the frequency domain) that the physics of a time-varying electric *field* (not merely a time-varying electric *current*) will induce a time-varying magnetic field? Note that the AC/DC module handles Faraday's law (i.e., that a time varying magnetic field will induce an electric field) just fine, as well as Ampere's law for computing magnetic fields due to electric *currents*. But it leaves out Maxwell's "displacement current" term (i.e., that a time varying electric field will induce a magnetic field). But Maxwell's term *typically* doesn't come into play in geometries that are very small compared to a wavelength. At 1 MHz, a free-space wavelength is 300 meters. At 10 MHz, it's 30 meters. These lengths get shorter in media (such as dielectrics, water, etc.). But if your regime is very small compared to the characteristic wavelengths, you just *might not* need to use the RF module after all. Also, although the RF module *can* be used in such small volumes if it is applied carefully, it is more usually (and more easily) used in computational volumes at least half a wavelength across. So, if you are going to use the RF module in a much smaller volume than that, you will need to be extra careful about specifying your boundary conditions (although, in practice, you should *always* be careful about how you specify your boundary conditions, in any Comsol Multiphysics problem). (By the way, you will find that properly handling "circuit-like" situations using the RF module is trickier than with the AC/DC module.) Finally, I expect that the official Comsol technical support folks can (and will be happy to) offer you better guidance. So... ask them! And I expect (at least, I hope) that they will base their guidance to you on the details of the actual physics needed to be modeled, not just some arbitrary frequency range. Good luck.