Discussion Forum

If I am reading the documentation correctly then COMSOL always solves the Maxwell's equations in the temporal gauge. If this is true and there is no way to fix the gauge at the time of specifying the physics or the solver then the question becomes how does one compute gauge functions in COMSOL to transform the potentials obtained in the temporal gauge into a gauge of choice.

I don't think that I can help you directly with your question, but I'm commenting here because I'd sure like to learn more about why you are asking it! From my albeit naive theoretical perspective, the most essential feature of "gauge invariance" is that all physically-meaningful quantities are gauge-independent. So, if the numerical calculations are being done correctly (hmm, might that be your true concern?) then it shouldn't matter what gauge is being assumed, since all physically-meaningful results should be unchanged. So again, I wonder why you care about the choice of gauge. Might your concern have to do with properly setting some scalar or vector potential-related boundary condition, perhaps?

I am preparing to use COMSOL for teaching. The point for me is precisely to have COMSOL solve for potentials in a variety of gauges, then to visually explore the resulting solutions and then to verify that they all yield the same fields in the end. Calculating the fields is the least significant part of the exercise here.