Discussion Forum

The specified frequency does indeed get applied to the background field. Now, if you don't apply any fields on the boundaries (but only periodic conditions) then you had better apply a field (or some other sort of excitation, port, etc.) elsewhere, or your field will be zero everywhere. You can also specify a background field, if you choose the scattered-wave formulation. Regardless, if you are having a frustrating problem with a particular model in which something unexpected appears to be happening to the frequency, I suggest you post it to the forum so that others can take a closer look at it to help you find out what is going wrong.

Thank you for the response. I have attached my model below. I have a background field applied with all periodic boundary conditions, but it fails to converge. It's solved without the frequency applied, which is why i asked the question.

In the Model Builder, go to Electromagnetic Waves, Frequency Domain..." then look in the Settings, and Find "Analysis Methodolology." Expand that (down arrow) and replace your setting under "Methodology options:" (which you have as "Fast") with "Robust" instead. Your model will then execute to completion. {Note: I did this with version 5.5. I don't know if it will do the same in 5.4 (but it probably will).} There are other potential issues with your model, since your unit-cell scale is very, very small compared to the wavelength (which, I suspect, contributed to the failed execution). Physics-wise, I wouldn't be confident that the assumed properties of "soft iron" (with a mu_r = 1200 in your model) are valid at a frequency of 1 GHz (which is the frequency you currently have specified, under "Step 1: Frequency Domain"), but that is something for you to decide if you believe or not. At least the model executes, now. But I suspect that you have more work ahead of you to make this representative of something in the real world. Good luck.

Thank you! This did solve. Would you be able to explain the difference of the two options to me?

As for the material values and frequencies, those are things I willa djust, the 1GHz was a comsol default. I have experimental data I will use to tune my model to more meaningful results. The Iron is also a placeholder for which I am adjusting the values. That permeability was actually taken from the included B-H curve in comsol.

Glad to hear it is working for you. In regard to why "robust" converges and "fast" didn't, I don't know! But the robust one is evidently more robust in terms of convergence, based on your own example. :-) Now, in regard to your use of placeholder material properties, or leaving the frequency at its default of 1GHz if/when you know that you are actually interested in something possibly much different, may I offer a suggestion/word of caution: Bear in mind that the convergence (or not) of a Comsol Multiphysics model may depend on such parameters! So it is a good idea to set those (and other) parameters appropriately (even if only roughly), before concluding that your model won't work in cases of actual interest.