next previous
Up: Instrumental polarization caused

2. Geometry of rays incident on primary mirror

The primary mirror of the telescope is in general a concave paraboloid. We consider a paraboloid, having its axis coinciding with the Z-axis and the vertex coinciding with the origin of the co-ordinate system. We consider one incident ray in the ZX plane, making an angle tex2html_wrap_inline1318 with Z axis (and tex2html_wrap_inline1322 with X axis) and passing through the origin. The Y axis will be perpendicular to this ray. The direction cosine (henceforth d.c.) of the incident ray can be written as:


equation1046
Now we consider a beam of rays which are parallel to this above ray and coming from a celestial object having a field angle tex2html_wrap_inline1318 with respect to the telescope axis. The rays which lie in the outer periphery of this beam, will be incident on the parabolloid mirror at points defining a circle, with radius h1 and the centre of the circle lying on the Z axis. (where tex2html_wrap_inline1334 is the diameter of the paraboloid mirror). The plane defined by this circle will have Z axis perpendicular to it. The (x,y) co-ordinates of any point on this circle can be expressed as tex2html_wrap_inline1340. In this case tex2html_wrap_inline1342 is the azimuthal angle of the ray. For example the rays which are contained in the ZX plane will have tex2html_wrap_inline1346. We further assume that f1 is the f-number of the primary mirror, which is nothing but the ratio of focal length to the diameter.

Now the equation of the above paraboloid can be expressed as:


equation1049

Substituting the values tex2html_wrap_inline1352, we get the Z coordinate of the point of incidence as tex2html_wrap_inline1356. Now the d.c. of the normal at this point tex2html_wrap_inline1358 can be expressed as:


eqnarray1051

The angle of incidence (pi) between the normal and incident ray can be expressed as:


equation1053

The d.c of the reflected ray tex2html_wrap_inline1362, which makes an angle pi with the normal and tex2html_wrap_inline1366 with the incident ray can be expressed as:


eqnarray1055

Now we shall find the d.c. of the other vectors connected to the incident plane. The electric field vectors of the incident and reflected rays can be resolved in two directions, one perpendicular to the plane of incidence (s-direction) and the other orthogonal to the s-direction (we call it p-direction). Actually there are two p-directions, one corresponds to the incident ray (pi-direction) and the other corresponds to the reflected ray (pr-direction). While discussing reflection on primary mirror, we refer to these directions as ps, ppi and ppr.

While considering the reflection on the secondary mirror, we shall denote the corresponding quantities as ss, spi and spr vectors.

The ps-vector is actually perpendicular to the plane containing vectors tex2html_wrap_inline1392 and tex2html_wrap_inline1394. Therefore the d.c. of the ps-vector tex2html_wrap_inline1398 will be proportional to


displaymath1312

After proper substitution and normalisation, these values can be determined as:


equation1059

Similarly the ppi-vector with directions (plpi, pmpi, pnpi), is perpendicular to the directions tex2html_wrap_inline1398 and tex2html_wrap_inline1410 and therefore, one can derive


equation1061

The ppr-vector with direction (plpr, pmpr, pnpr), is perpendicular to the directions tex2html_wrap_inline1398 and tex2html_wrap_inline1362 and one can similarly derive


equation1063
where we have substituted


equation1065


next previous
Up: Instrumental polarization caused

Copyright by the European Southern Observatory (ESO)
web@ed-phys.fr