search

Chandra-MARX / marx

Chandra X-ray Observatory ray-trace simulator
http://space.mit.edu/cxc/marx/
5 stars 4 forks source link

MARX PSF wrong for off-axis sources #21

Closed hamogu closed 8 years ago

hamogu commented 8 years ago

Comparing an observation of a bright, off-axis source (ObsID 1068), I discovered a major problem in the simulated PSF in MARX.

marxbug

Only MARX has that obvious big hole in the lower part op the PSF. The image shows the Chandra observation, a MARX only simulation and a SAOTrace ray-trace with MARX as the instrument model. The fact that this shows up only in MARX indicates that something is wrong with the MARX mirror.

Further investigation shows that this affects only off-axis sources (the source in the image is 25 arcmin off-axis) and only using the MARX HRMA mirror module.

hamogu commented 8 years ago

It turns out that this is due to a bug introduced in Marx 5.0 (i.e. it affects Marx 5.0, 5.1 and 5.2).

When a photon generated, it is placed randomly on the entrance aperture. In the MARX HRMA model, that entrance aperture is a ring with the inner and outer radius given by the maximum and minimum radius of the parabolic part of the Chandra mirror shells. This works well for rays parallel to the pointing direction, but for rays from an off-axis source, which come in at an angle, this laves some parts of the mirror unexposed (and a certain fraction of photons misses the mirror entirely), hence there is a hole in the simulated PSF.

hamogu commented 8 years ago

This fix increases the effective area of the Chandra mirrors, making the aperture big enough to to fill the aperture for all input angles. However, that also means that not all the photons will hit a mirror. The number of photons missed reduces the efficiency almost by a factor of 2 for an on-axis source.