Hi Hui,

Thanks for your question!

To set up an output from an angle polished fiber in Sequential Mode, you can use either a Coordinate Break surface or a Tilted surface after the Object plane to define the angle tilt you’re looking to achieve. You’d need to define your Object surface to be the same index as your fiber core, I think, since it would be leaving the fiber and emitting into the rest of your system. So you’d want to take some care in defining your beam parameters to ensure that the output represents the mode traveling out of the fiber into air (as in, checking your ray angles at 1/e^2 if you’re using geometric rays or checking beam properties after the angle polish if you’re using Skew Gaussian Beam or POP). And, as always, if you’re using geometric rays to model your fiber output, you’ll want to be careful in how you define those rays based on the position of the next element with respect to your Rayleigh range:

As for Non-Sequential Mode, I think a similar approach would apply -- you’d want to define the beam inside some medium such that it would refract out of the core index with the properties in air that you’d expect. You could use an object like Cylinder 2 Volume to directly define a tilt on one of the cylinder faces. The caveat in NSC mode is that it can only approximate beam propagation with geometric rays, so this will be something you need to keep in mind and verify that it remains a good estimate of your performance. We also have a Source DLL on the Code Exchange that defines rays to model the change in waist size of a diffracting beam, which may be of help to you: DLL (Source): Non-sequential Astigmatic Gaussian | Zemax Community

Let us know if you have any further questions!