Discussion Forum

I'm confused by your question and I think everyone else is too, since no one has replied to you yet. When you say that you want to "apply an RF field" to the plate, how do you intend to "apply" it? You refer to "the place" that you apply the field. So, are you saying that you don't intend to "apply" this RF field to the whole plate? How do you intend to limit where you do or do not "apply" this field? You are calling this a capacitor, with two plates, so I am guessing that, maybe, you actually intend to apply this "RF field" as a current, perhaps using wires? OK, let's assume that. But you also say " If I don't apply this perfectly at the center of the plate (which will be the case in practice), I will induce some current in the plate, which in turn will produce a magnetic field around the plate." Please explain why you think applying an RF current (if that's what you meant) at the center of one of the capacitor plates would/should not "induce some current in the plate." After all, if any current whatsoever flows from your wire (if you have a wire) to the plate, then surely there will be currents on that plate, regardless of whether the geometry is symmetric or not, and a magnetic field will also be generated. I really don't know how you could apply an RF current to a capacitor and not generate some magnetic fields, regardless of where/how you connected to the capacitor. I suggest you draw detailed pictures and label where you wish to apply currents (or is it some other way of applying RF?) and post that here.

I apologize for the confusion (this is my first post here) and please allow me to clarify. My setup is 2 metallic parallel plates (let's assume they are disk shaped, but the actual shape is not that relevant for my question and I am actually planning on trying different geometries), separated by 1 mm. I want to produce at the center of the plates (the actual volume I am interested in is roughly a sphere of radius 10 microns at the center in between the 2 plates) a sinusoidal electric field, of frequency ~50 MHz and amplitude ~ 100 V/cm. So these plates will be part of a resonant circuit and will act as a capacitor. If the plates are perfectly parallel to each other and the (ideal) electrical connections to the plates are made at the center of the plates, the magnetic field at very the center in between the 2 plates, in a direction perpendicular to the plates (which is the direction I am interested in) will be zero. In practice I will have magnetic fields from 2 main contributions:

1. If the plates are note perfectly parallel to each other (i.e. the time varying electric field will have a component parallel to the plates and it will thus induce a time varying magnetic field perpendicular to the plates).
2. Even if the plates are perfectly parallel, if the electrical connections are not made at the center of each of the 2 plates, the current induced in the plates won't be symmetric with respect to the point at the center in between the 2 plates and thus I will have a magnetic field induced there. If on top of this the 2 plates are also not perfectly parallel, this magnetic field due to induced currents will be along the direction perpendicular to the 2 plates (and it will be in addition to the magnetic field in 1.).

I want to compute the magnitude of the magnetic field due to 2. at the center in between the 2 plates, for different tilt angles between the 2 plates and for different electrical connections applied to the plates. I am also attaching a simplified drawing of the setup.

Please let me know if this is still not clear, I would be happy to provide more details. Thank you!

The attached quick example should be more than sufficient to get you started. Note that you'll need the AC/DC module and at least version 6.2 of Comsol Multiphysics. Enjoy.

Thank you so much! I will give this a try.

Is there a way to open this in 6.1? This is the version my university has.

Sorry. I did it in v6.2 because I thought you might not have v6.3 yet. Maybe you can find someone with a copy of v6.2 or v6.3?

Added note: Multiphysics v6.3 was released just last month, but v6.2 was released more than a year ago. See https://www.comsol.com/release-history

Would it be possible to export it as a java file? It seems like there is a way to open it in 6.1 using a java file.

OK. I haven't tried importing java files into Comsol Multiphysics, so I don't know how well that will work, but I've attached one that I exported as a java file. Good luck.