Simulation of the effect of the external calibrator bar on the image quality

SL 9/12/16. Update 10/12/16 (part 3)

For the NIKA2 run 7 one of the goal was to test the possibility to use an external calibrator to make a power modulation on the signal. There's several possible advantages for such a technique, the main one being a better measurement of the KID response variation than the tone modulation. If we validate the technique we would like to install this external calibrator into the secondary mirror. This is a delicate operation, that can't be done without a clear proof of the advantages of the technique. So to make this test we installed the calibrator at the middle of the beam in the telescope vertex cabin hole passing through the back structure of the primary mirror; the calibrator is fixed on a bar attached to the side walls of the cabin, as shown on the pictures below:

To know the effects of this installation on the observation I did Zemax simulations. The main outcome can be seen on the following series of images. The 1st series represent the system without the calibrator, and the second series with the calibrator.

• Image 1 = a 3D shaded model representing the ray tracing for 12 fields from the M1 vertex (hole at the centre of the surface of the primary mirror) down to the arrays inside the NIKA2 cryostat; the grey surfaces represent some cabin limits, cryostat, and optical elements.
• Images 2 and 3 = footprint of the central field rays on the primary mirror without the vignetted rays (by the aperture stop at the cold pupil), and PSF of this field on the array surface.
• Images 4 and 5 = footprint of the high elevation edge field rays on the primary mirror without the vignetted rays (by the aperture stop at the cold pupil), and PSF of this field on the array surface.

1. System without the calibrator

2. System with the calibrator

3. System with the calibrator showing light emitted from the calibrator itself, not coming from the sky

Green rays below = portion of cone of light emitted by the 1mm hole of the external calibrator which totally illuminate the M3 mirror (only portion that can reach the cryostat window). As you can see on the picture the 1mm hole is not centred on the calibrator box, which itself is slightly lower than the optical axis of the telescope.
Blue rays below = reference of what would have been the cone emitted by the calibrator if its hole would have been placed exactly on the optical axis of the telescope (in the images above the bar was centred, which is not quite correct, but changes nothing to the qualitative information given by the simulation).

Left image below = footprint of the calibrator cone of light on the cryostat window. Right image = footprint of the calibrator cone of light on the cryostat cold stop (""cold pupil")

==> Conclusion: using the calibrator in the vertex window does not allow aligning the cold optics, because it isn't vignetted by the cryostat cold optics but by the M3 mirror.
Note that it's purely by chance that at 60 degrees elevation at which we did our measurement with Alain the calibrator offset creates a footprint offset on the cryostat window identical to the cold optics offsets that we rechecked on blank sky without calibrator the following day!