pspecbeam.PSpecBeamUV updates for efield beam #143
Conversation
ghost
assigned 
| @@ -429,7 +440,7 @@ def power_beam_sq_int(self, pol='pI'): | |||
| primary_beam_area: float, array-like | |||
| """ | |||
| if hasattr(self.primary_beam, 'get_beam_area'): | |||
| return self.primary_beam.get_beam_sq_area(pol) | |||
| return np.real(self.primary_beam.get_beam_sq_area(pol)) | |||
There was a problem hiding this comment.
Should this function be returning complex numbers to begin with? Or is there a bug in get_beam_sq_area?
There was a problem hiding this comment.
That's what I was asking @bhazelton, and I never got a response, but I think I came to the conclusion that in principle it can be complex (e.g. when asking for XY and YX areas), but in practice we will always be using XX or YY or pI or pQ beams, in which case it is real. But @bhazelton let me know if you have any opinions on that.
| # setup primary power beam | ||
| self.primary_beam = uvb | ||
| if uvb.beam_type == 'efield': | ||
| self.primary_beam.efield_to_power(inplace=True) |
There was a problem hiding this comment.
efield_to_power is called for linear polarizations, for pseudo-stokes you need to use the function efield_to_pstokes.
There was a problem hiding this comment.
So this is done assuming one only wants linear dipole polarization. This function is (purposely) not trying to be smart and will not convert the efield to pstokes. To pspecbeam for pstokes parameters, one needs to feed a pstokes power beam. (see the unit tests added where I test for exactly this case). I'll make a comment specifying this in the docstring!
Updates the beam files in the
hera_pspec/datadirectory, and allowspspecbeam.PSpecBeamUVto take an efield beam and convert to power beams.