Units of amplitude in UniformPlaneSource and units of power in PoyntingFluxDetector

[galcerte]

So I thought that amplitude in UniformPlaneSource was in SI units (so V/m), like time and space in the rest of the framework, but when I plop a PoyntingFluxDetector in front of a source, I don't get a maximum of 2.7 mW/m², and when I do reduce_volume=True, instead of getting a maximum of 1.08 x 10⁻⁷ W (which should be what's expected given the area of the detector), I get a maximum of 1, which got me a bit suspicious. Is amplitude normalized using power somehow? And the detector is in these units as well? Detector plots don't even use units in the Y axis, so I don't think I was seeing Watts anyways...

[ymahlau]

Currently, the unit conventions in fdtdx are a bit all over the place, which is why I am currently working on an actual unit system (see #114). When this is implemented, it should be very clear which values fdtdx has as input and output. I am pretty sure that in the next few weeks I should finish this feature.

Currently, sources are supposed to be normalized to a poynting flux of 1 (arbitrary unit, fdtd is currently scale invariant), but there is a good chance that there is also a bug present.

[galcerte]

I am currently working on an actual unit system (see #114)

Oh you posted that shortly before this discussion, I didn't see it, hah.

sources are supposed to be normalized to a poynting flux of 1

Just to confirm, do you mean a maximum power per unit area of 1, or a maximum power over the source's area (that is, integrating/summing up the Poynting vector's normal component to the plane that is the source, over the area of the source) of 1? I think it is the latter, given the plots from the detectors I saw...

[ymahlau]

Yes, the source is normalized over the total area of the source :)

[galcerte]

Thank you, sorry for the dumb question but I really wanted to make sure, hah.

[ymahlau]

There are no dumb questions here :)