Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-8589

Create HSC instrument model

    Details

    • Type: Story
    • Status: Done
    • Resolution: Done
    • Fix Version/s: None
    • Component/s: None
    • Labels:
      None

      Description

      In order to use CWFS or a forward modeling package (e.g., donutlib or galsim) for HSC images, we'll need a model for the instrument, including things like the focal length, pupil diameter, field-dependent obscuration (i.e., vignetting), and probably more. This ticket is to figure out exactly what's needed and how to model it.

        Attachments

          Issue Links

            Activity

            Hide
            jmeyers314 Joshua Meyers added a comment -

            I've been looking at pinhole HSC images to figure out how the pupil function varies over the HSC field of view. As Robert Lupton mentions here, this looks to be well modeled by a two-circle overlap. I used ds9 to overlay a series of circle regions on the pinhole image to more precisely specify this model. I also added a third circle for the obscuration by the camera itself. See attached screenshot.

            Note that very far from the center of the field of view, even this three circle model is insufficient, as some other optical stop apparently enters the pupil (see upper left pinhole in attached screenshot).

            Show
            jmeyers314 Joshua Meyers added a comment - I've been looking at pinhole HSC images to figure out how the pupil function varies over the HSC field of view. As Robert Lupton mentions here , this looks to be well modeled by a two-circle overlap. I used ds9 to overlay a series of circle regions on the pinhole image to more precisely specify this model. I also added a third circle for the obscuration by the camera itself. See attached screenshot. Note that very far from the center of the field of view, even this three circle model is insufficient, as some other optical stop apparently enters the pupil (see upper left pinhole in attached screenshot).
            Hide
            jmeyers314 Joshua Meyers added a comment -

            I then imported in python the region files I generated in ds9 for analysis. The attached quiver plots show how the camera shadow and lens obstruction (green circles and red circles, respectively) are displaced with respect to the image of the primary mirror (blue circles), as a function of location on the focal plane (defined by the image of the primary mirror center).

            Show
            jmeyers314 Joshua Meyers added a comment - I then imported in python the region files I generated in ds9 for analysis. The attached quiver plots show how the camera shadow and lens obstruction (green circles and red circles, respectively) are displaced with respect to the image of the primary mirror (blue circles), as a function of location on the focal plane (defined by the image of the primary mirror center).
            Hide
            jmeyers314 Joshua Meyers added a comment - - edited

            The attached regression plots show that the offsets are quite linear in field angle.

            Note that I made the y-axis units degrees in these plots, as obtained from the ds9 region files. I think the displacement will actually scale with how far in or out of focus the camera is though. What really matters for modeling the pupil is how large these displacements are with respect to the primary mirror. I measured the radius of the primary mirror image as about 128.9 arcseconds for this particular image (visit=904686).

            The regression slopes are 0.00558 degrees per degree for the camera shadow, and 0.0276 degrees per degree for the lens obstruction. I'm assuming the the directions of the displacements are exactly radial with respect to the focal plane center.

            In primary mirror diameters of displacement per degree of focal plane offset, these slopes are about 0.0779 and 0.385.

            Show
            jmeyers314 Joshua Meyers added a comment - - edited The attached regression plots show that the offsets are quite linear in field angle. Note that I made the y-axis units degrees in these plots, as obtained from the ds9 region files. I think the displacement will actually scale with how far in or out of focus the camera is though. What really matters for modeling the pupil is how large these displacements are with respect to the primary mirror. I measured the radius of the primary mirror image as about 128.9 arcseconds for this particular image (visit=904686). The regression slopes are 0.00558 degrees per degree for the camera shadow, and 0.0276 degrees per degree for the lens obstruction. I'm assuming the the directions of the displacements are exactly radial with respect to the focal plane center. In primary mirror diameters of displacement per degree of focal plane offset, these slopes are about 0.0779 and 0.385.
            Hide
            jmeyers314 Joshua Meyers added a comment -

            I happy with this now that fitting full focal planes of donuts seems to work a la DM-8565. Closing.

            Show
            jmeyers314 Joshua Meyers added a comment - I happy with this now that fitting full focal planes of donuts seems to work a la DM-8565 . Closing.

              People

              • Assignee:
                jmeyers314 Joshua Meyers
                Reporter:
                jmeyers314 Joshua Meyers
                Watchers:
                Joshua Meyers
              • Votes:
                0 Vote for this issue
                Watchers:
                1 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved:

                  Summary Panel