Search
Text Size:
Smaller Text Normal Text Larger Text

Facebook page

Twitter profile

YouTube channel

ALFOSC setup for polarimetry

Calcite

Mounting calcite plates in the aperture wheel

For imaging polarimetry only one calcite plate is needed, the one labelled "Cal-90" on the holder but named "Calcite" in the optical elements database. This is "Calcite 1" corresponding to optical element ID 604. It is already correctly aligned in its aperture wheel holder to take the orientation "Ver90", which splits the two components horizontally on the CCD. (The 90 degree orientation of the calcite crystal is referred to as Ver90, because angles are defined as polar coordinates, and the crystal orientation is vertical when the target is split horizontally.)

It is per default mounted in the aperture wheel, preferrably in slot 0, where it is aligned at stepper motor position of 319000, which fits with the window setting of polwin. (You have to unscrew all four screws and remove completely the slitholder before you can mount the Calcite in its holder.) Update with editsetup.

For spectro-polarimetry you need to mount another calcite plate in addition to the one above. This one has a polarimetry slitlet mounted on top of it, and its name in the optical elements database is either Pol_1.0" (ID 508), Pol_1.4" (ID 509), or Pol_1.8" (ID 510), depending on the slitlet. Per default the Pol_1.8" slitlet is used (actual width is 1.5"), unless the observer wants a narrower slit.

When not mounted, this combined "Calcite + slit" item is found it the Calcite plates drawer on the wall. It is already correctly aligned in its holder to have the calcite in Ver90 orientation, as for the imaging calcite. With the slit correctly mounted on top, the slit will be horizontal, and therefore the orientation of the combined optical element is referred to as "Horizontal" in editsetup and in the optical elements database. This splits the horizontal slitlet horizontally, giving two vertical spectra when used with the default vertical grisms.

Staff Instructions:

  • Handle the calcite plates with care. They are very soft material and should never be cleaned with any liquids nor rubbed in any way. Remove dust by a light blow of air only! Don't touch with fingers.

    When they are not mounted, the calcite plates are stored in their holders wrapped in optical paper inside a plastic envelope in the Calcite plates drawer in the filter carousel.

  • This step is normally not needed. Both calcites should be correctly and permanently mounted in their holders: But if you have to take the calcite plates out of their holders, then check their orientation as follows. Use the piece of paper in the calcite drawer of the filter carousel called: Calcite alignment tool, which contains lines rotated by angles 0, 45 and 90 degrees. Place the calcite plate (in its holder) such that the largest edge of the holder is parallel to the lower line of the frame of the plot. If it is mounted correctly, all lines will appear as double except for the horizontal one (because it is displaced horizontally). If this is not the case, then loosen the inner tube by unscrewing the screw holding the tubes together (need a small hex key), and correct the alignment. NB! It is crucial that the two calcite plates are both oriented in exactly the same way (i.e. not even 180 degrees offset), otherwise the slit aquisition for spectropolarimetry will not work. If you have to rotate the calcites, keep this in mind, because the access to the actual calcites in their holders is opposite for the two holders! One you operate from "above" and the other you operate from "below". NB!

  • Normally we use the Pol_1.8" slitlet and this one is installed by default and aligned, and you do not need to read further on this step. The polarimetry slitlet is mounted on top of the calcite plate (on the side where light enters from the telescope), and is fixed to the calcite plate holder by tape. Align it by looking through the calcite at the slit, which now will appear as two slits horizontally offset to each others. Look directly at the calcite such that you see your eye reflected in the center. The slit is aligned with respect to the calcite plate when the two slit images are on a line. See below for aligning the slit with respect to the CCD.

  • Mount the calcite with the slit in the aperture wheel. You have to unscrew all four screws and take out the the whole slitholder. (Take care that screws don't fall into the instrument!)

  • Update the ALFOSC aperture wheel content using the program editsetup as user obs on selena. Note that the 3 polarimetry slitlets when used on top of the Calcite are named: Pol_1.0", Pol_1.4", and Pol_1.8". In the database they can have orientation either vertical or horizontal. NB! As from April 26th 2012 they should always be installed horizontally! This also means that for spectro-polarimetry you can use grisms in their default vertical orientation.

Aligning the polarimetry slitlet with respect to the CCD:

  1. Put the polarimetry slitlet in the beam and set all other positions to open.

  2. Run alfosc.focus-offset. This sets the internalt focus to 870 units less than the default value for ALFOSC which is 1610.

  3. Turn on the ThAr lamp in the ALFOSC calib unit.

  4. Make a small window by typing alfosc.polwin.

  5. Take a test image with exp 1 and use imexam to find the y-position of the left and right parts of the two slitlets split by the calcite. Measure the two y-positions with the key 'k' and determine the difference dy.

  6. If the slit is tilted, change the encoder position of the wheel in the program editsetup and iterate until the slit is aligned. About -1400 units is needed to rotate the slitlet clockwise by an amount corresponding to dy = 3 pixels from left side of left slit to right end of left slit. (For reference: when mounted in aperture slot 6 the best alignment was found at encoder value 237200.)

  7. When the slit is aligned, type resetxy and take a full image to determine the slit position, i.e. the center of the left image of the slit, in order to update the file ~obs/alfosc/slitpos.def. Instruct the observer to use the left image of the target for slit acquisition.

  8. Make an appropriate window for spectroscopy for the user by setting xbeg, ybeg, xsize, ysize in a script, such as for instance (and always in this order), and store it in the script directory of the observer: 
    #!/bin/tcsh
alfosc.resetxy
alfosc.readspeed 200
alfosc.xsize 250
alfosc.xbeg 1150
alfosc.ysize 1350
alfosc.ybeg 1
alfosc.bin 2
    The above is an example for Grism#10 where the upper part (showing the zeroth order) has been cut. Depending on the grism, you might wish to keep the full size in the vertical direction.


    Retarder plates

    lambda/2

    The idea is that FAPOL will always be mounted in FASU and the default retarder plate is the 1/2 wave plate (lambda/2), which is used to measure linear polarimetry. When the lambda/2 is correctly mounted in FAPOL, as described below, its optical axis is at 0 degrees with respect to the principle plane of the calcite crystal. The orientation of the lambda/2 is not very critical, however, as the zero-point correction of the angle can be found with high-polarisation standard stars.

    lambda/4

    In order to measure circular polarimetry, the 1/4 wave plate (lambda/4) has to be installed. (Note that it is possible to measure simultaneously circular and linear polarisation with the 1/4 wave plate, but at a cost of 50% efficiency for the linear mode.)

    When the lambda/4 is correctly mounted in FAPOL, as described below, its optical axis is at 45 degrees with respect to the principle plane of the calcite crystal, that is the plane defined by the o- and e- rays.

    Mount lambda/2 or lambda/4 in FAPOL

    If you have to change plates, make sure you handle them with care. These plates are expensive. They are located in the their marked boxes in the cupboard above the filter table in the electronics room.
    1. Make sure the FAPOL carriage is out of the beam.

    2. Power off the FASU controller.



    3. Remove the round connector from FAPOL.

    4. Disconnect the DB25 connector.



    5. Unscrew the 4 screws that mounts FAPOL in FASU. The unit is heavy and will fall down a few mm if not held.

    6. Slide the FAPOL unit carefully out of its hole and take it down to the control room.



    7. Sit at a table and carefully unscrew the 4 screws with rubber washers that hold the retarder plate in place. Align the 1/2 wave plate such that the red pen mark aligns with the white zero on the holder. The arrow points to which side is up. Both the 1/2 and the 1/4 wave plate should be mounted such that the circular mark on the plate holder points to the circular mark on the carriage (which is at the outer end of the carriage).



    8. For the 1/4 wave plate fix one screw per side of the plate.

      For the 1/2 wave plate two screws on two opposite sides is the best way.

    9. Take the unit back, slide it in and fix it, connect the cables again and power on the FASU controller. Don't forget to open the FASU shutter.
    10. NB! Remember to use 'editsetup' so the UIF knows which plate is in and the FITS headers become right! See below.

      Changing the setup information for the retarder plate

      • Use editsetup to update the name of the mounted retarder plate.
      • Do 'shutdownobssys alfosc'.
      • Do 'startobssys alfosc'.

      Physically mounting the FAPOL unit on FASU

      • There is a 13 mm gap between the top side of FASU B wheel and the underneath of the FAPOL unit. FASU filter holders stick out about 2mm above the wheel. To be conservative, don't put in thick filters in FASU B which will protrude more than 9mm out of the filter holders.
      • Power off the FASU controller.
      • Remove the black plate that is facing towards the staircase.
      • Mount FAPOL in the hole and fasten it with the screws.
      • Remove the blind round connector to the right of the hole.
      • Connect the round connector from FAPOL to the connector to the right of the hole.
      • Connect the DB25 connector to the FASU controller.
      • Power on the FASU controller (Don't forget to open the shutter).

      Connecting FAPOL on a normal instrument mount

      • Connect the network cable to the network switch mounted on the adapter. You can use whatever port you like.
      • Connect the 220V cable to a 220V outlet.

      11. Misc

      Start of the control programme

      The control programme for FAPOL is started from any of the lisa X-terminal windows by giving the command (logged in on sandra):

      • /usr/local/bin/fapo-control


      Since February-2007 FAPOL starts up together with the rest of ALFOSC. If for some reason it doesn't, then check the processes and kill all old FAPOL processes that might be running. Start up FAPOL manually with the above command

      Note that after having moved the calibration mirror, it is normal that FAPOL initializes when you command carriage to go in. The initialization should take about 10s, and the total time to get FAPOL in is about 40s. If it hangs, then interrupt the FAPOL s/w by CTRL-C and power cycle the FASU controller (remember to open the shutter afterwards). Restart FAPOL s/w.

      16 retader plate positions

      • For testing purposes there are 16 retader plate positions available (the step size is 22.5 degrees). Non-UIF positions are not updated in the UIF, i.e. the retader-plate might have different position than the UIF indicates, however the FITS-headers are updated in a correct way. The use of the "hidden" positions is the normal FAPO-script way e.g. "!fapo-112.5" etc.

      Comments on previous orientations of the Calcites

      Current status: Since April 26th 2012 the two calcites are again mounted at 90 degrees orientation in their holders. This means that the beam is split in two orthogonally polarized components horizontally on the CCD. (NB! Note that in the optical elements database the 90 degree orientation is referred to as Ver90, because angles were defined as polar coordinates.) For spectropolarimetry the horizontal slit is displaced horizontally giving two vertical spectra.

      Old setup info: From April 2003 to November 2008 the two calcite plates were mounted at 90 degrees orientation in their holders. This way the slit was horizontal and spectro-polarimetry done with grisms in the vertical (default) orientation. From November 2008 to April 2012 the calcites were mounted at a 0 degree orientation of the calcites, splitting the beam in two orthogonally polarized components vertically on the CCD, needing the slit to be vertical and the grisms horizontal! This was done to reduce the readout time for fast spectro-polarimetry. With the arrival of the new ALFOSC controller in June 2012 there was no longer a speed issue, and it was decided to go back to the old setup, where the (default) vertically mounted grisms could be used.

      Comments to Anlaug Amanda Djupvik