Nordic Optical Telescope Text Size: Smaller Text Normal Text Larger Text  
 

Horizontal versus vertical
Mounting slits
Mounting grisms
Echelle remarks
VPH grism #17 remarks
VPH grisms #18, #19, #20
Balancing the grism wheel

 

About horizontal and vertical grisms/slits in ALFOSC

This is an attempt to clarify what is a vertical grism, horizontal slit, etc.
  • A horizontal slit is imaged on the detector horizontally. Horizontal slits are the default for ALFOSC, and have to be used together with a vertical grism.

    If you take an image with a horizontal slit, and no grism, you get a horizontal bright stripe. An arc-line spectrum consists of a series of horizontal arc lines.

  • A vertical slit is imaged on the detector vertically. Vertical slits have to be used together with a horizontal grism. The Echelle-modes and grism#17 can only be used with vertical slit(let)s.

    If you take an image with a vertical slit, and no grism, you get a vertical bright stripe. An arc-line spectrum consists of a series of vertical arc lines.

  • A horizontal grism gives a horizontal spectrum on the detector. You need a vertical slit for this.

    If you look through a horizontally oriented grism towards a round lamp, the spectrum runs horizontally.

  • A vertical grism gives a vertical spectrum on the detector. You need a horizontal slit for this. This is the default for ALFOSC.

    If you look through a vertically oriented grism towards a round lamp, the spectrum runs vertically.

 

 

Mounting/exchanging ALFOSC slits

  • Vertical offset slits

    An offset slit has the slit cut off-center in the slit plate.

    These slits can be mounted in 2 different ways, normally and upside-down, and will cause either a redshift or a blueshift of the sampled wavelength range. The size of the shift is approximately 1/4 of the full wavelength range that a particular grism usually gives.

    Note: whether the slit gives rise to a blue- or a redshift depends on the orientation of the grism. In case the grism wheel contains a grism that images blue-to-red on the detector, and another grism that images red-to-blue on the detector, the same offset slit will shift the sampled regions to different directions for these 2 grisms.
    This is an undesirable feature, and I (JHT) will have to do something about it!

    1. Start the ALFOSC UIF
    2. Move the aperture wheel to the position opposite to the one you want to exchange
    3. Go to the dome floor and open the slit door
    4. Loosen the 4 screws in the holder
    5. Carefully remove the slit and put it in its box
    6. Check the new slit for dust on the slit jaws; blow any dust away with clean air
    7. Carefully put in the new slit (dont know whether normally or upside-down at this point)
    8. Tighten the 4 screws in the holder with only a tiny amount of force.
    9. Update the aperture definition file using the GUI editor EditSetup. Select the slitname with desired extension 'r' for a red-shifted wavelength range, or with extension 'b' for a blue-shifted wavelength range: e.g. VertOff_1.0r or VertOff_1.0b .
    10. Press 'Read Setup' on the UIF.
    11. Take an arc-line spectrum (e.g. with the Helium lamp) using the offset slit and the requested horizontal grism, for example using
        wheels   -s VertOff_1.0r   -g "Gr_#6 Hori"
        calibexp   -o "Test Helium"   He   3
      then use QuickExam to make a cross-cut, and compare the arc-line spectrum with that of the arc-line atlasses on this table of arc maps.
      Carefully check whether the lines are shifted in the correct direction: if not, then go back to step 2) above and turn the slit upside down.
    12. Align the slit and set the align-flag in the aperture definition file using the GUI editor EditSetup.
    13. Determine the position of the slit from a full-frame image, without a filter in the ALFOSC beam. Use the ThAr lamp as light source, with 0.2 second exposure time. Use alfosc.edit-slitpos to update the ~obs/alfosc/slitpos.def file accordingly. For CCD#8 we usually put the vertical spectrum at column x=1050.
      Note: the align tool (TIASGAT) will report the slit position when aligning the slit.

  • All other slits

    • Start the ALFOSC UIF.
    • Move the aperture wheel to the position opposite to the one you want to exchange.
    • Go to the dome floor and open the slit door.
    • Loosen the 4 screws in the holder.
    • Carefully remove the slit and put it in its box.
    • Check the new slit for dust on the slit jaws; blow any dust away with clean air.
    • Carefully put in the new slit with the defining side to the top.
    • Tighten the 4 screws in the holder with only a tiny amount of force.
    • Update the aperture definition file using the GUI editor EditSetup.
    • Press 'Read Setup' on the UIF.
    • Align the slit and set the align-flag in the aperture definition file using the GUI editor EditSetup.
    • Determine the position of the slit from a full-frame image, without a filter in the ALFOSC beam. Use the ThAr lamp as light source, with 0.2 second exposure time. Use alfosc.edit-slitpos to update the ~obs/alfosc/slitpos.def file accordingly. For CCD#8 we usually put the vertical spectrum at column x=1050.
      Note: the align tool (TIASGAT) will report the slit position when aligning the slit.

      slit 0.4 0.5 0.751.0 1.2 1.3 1.8 2.5 5.0 10.0
      Y-pos (pixel)999 1000100010211007 998102810031007 997
      Approximate Y-positions of the default horizontal slits on CCD#8 at X-pos=1050 .
      The error on the above values, for the telescope at zenith, should be +/- 3 pixels.
      This table is only indicative: allways remeasure !

    Note: Always check the setup with an halogen and an arc exposure to see if you have mounted the optical items correctly!

     

     

    Mounting/exchanging ALFOSC grisms

    • Start the ALFOSC UIF.
    • Move the grism wheel to the position opposite to the one you want to exchange (see notes on balancing below).
    • Go to the dome floor and open the access box.
    • Loosen the 3 screws in the holder.
    • Carefully remove the grism and put it in its box.
    • Take the new grism out of its box and check its general appearance for scratches, bad coatings, dust, etc. Update the grism status table if necessary.
    • Check the direction of dispersion of the grism: it should be at right angles with the slit orientation.
    • Carefully put in the new grism with the grooves towards the camera.
      • For high-res grisms the grooved side is the steep one.
      • For low-res grisms look at the grism faces under a shallow angle: the grooved side is the least reflective one.
      • All grisms are marked with an arrow that should point towards the camera.
    • Tighten the 3 screws in the holder with only a tiny amount of force.
    • Update the grism definition file using the GUI editor EditSetup.
    • Press 'Read Setup' on the UIF.
    • Align the grism and set the align-flag in the grism definition file using the GUI editor EditSetup.

    Note: Always check the setup with an halogen and an arc exposure to see if you have mounted the optical items correctly!

     

     

    Echelle-mode remarks

    For Echelle mode the cross-dispersing grisms (#10, #11, #12) are to be mounted in the filter wheel, in a similar fashion as described above. The cross-dispersing grisms should be mounted with the dispersion direction vertical. The cross-dispersing grisms occasionally have to be rotated in their holder in order to allow for correct mounting in the filter or grism wheel.

    The Echelle grism (#9, #13) has to be mounted with horizontal dispersion, if used together with a cross disperser.
    Note that grism #13 can be used in the standard long-slit mode, together with a dedicated order-sorting filter (filter #124) in the ALFOSC filter wheel. Grism #13 has its mounting holder reversed wrt the other grisms.

    The Echelle slitlets are to be mounted vertically in the aperture wheel, such that the ThAr lines appear vertically (see example image below).
    The Echelle slits are called Ech_0.7" up to Ech_2.2" in the setup GUI EditSetup.
    Although the Echelle slits do not have to be aligned, it is important to edit the center of the slitlet in the ~obs/alfosc/slitpos.def file (use alfosc.edit-slitpos). The position of slit center (in CCD pixel coordinates) must be measured from a unbinned full-frame slit image, without a filter in the ALFOSC beam. Use the ThAr lamp with 0.2 sec exposure.

    Example calibration images, Halogen (left) and ThAr, for Echelle grism #9 with grism #11 as cross disperser. At the top of the image, the zeroth order of the cross-disperser is visible: this order is discarded in data reduction (i.e. is useless). The useful Echelle orders run from the middle of the chip (blue) to the bottom (red).

    Note: Always check the setup with an halogen and an arc exposure to see if you have mounted the optical items correctly!  

     

    VPH grism #17

    The grism has a different holder than the other grisms. On the holder, there is an adjustable rotational definer that should not be loosened, as it currently defines the correct rotational alignment of the grism.

    When mounting grism #17, put in the grism and rotate it anticlockwise until the rotational definer hits the mounting pin on the wheel. Then secure the grism with the 3 clamps.

    The grism can only be mounted in one of the wheel positions that does not have an extra baffle on the shutter-side of the wheel: this leaves positiones 1, 2, 6, 7 (Open Lyot). Sofar we have always mounted grism #17 in position 2.

    The grism can only be mounted horizontally, and the holder fits quite tightly in the wheel (i only tried wheel position 2 so far). When mounting the grism, the stepper units in EditSetup should first be set to -4800 wrt the nominal value of that wheel position. Nominal values are 0, 40000, 80000, etc. Then the stepper units need minor adjustment in order to align the grism (see below). This procedure ensures that the grism, which is mounted offset in its holder, will be in the center of the ALFOSC beam.

    Note that the grism is big and heavy, so please keep wheel balance in mind.

    The associated vertical slits are kept on the same shelf as the other ALFOSC slits. See also the carrousel.

    Aligning grism and slits: the grism/slit can be aligned with the detector rows/columns, by manually changing the stepper-motor units by amounts of 50 or so (possibly much more for the slits). Note that the slit must be aligned with the grism in the beam, using arc lines.  

     

    VPH grisms #18, #19, #20

    These grisms are to be mounted as any other grism, but because the barrels are a bit wider than the normal grisms these VPH grisms only fit in slots that do not have an extra baffle mounted on the shutter side of the grism wheel. Dedicated slots are slots 2 and 6, and slot 1 should be available as well. Slot 5 is also dedicated from August 17th 2015 (the baffle on the shutter side was removed for this slot).

    Note that these grisms are heavy, so please keep wheel balance in mind.

    Note that these grisms consist of optical components that are stuck together with oil, rather than optical cement/glue.
    Be extra careful when handling them and when placing and storing them.  

     

    Balancing the grism wheel

    Everyone tells us that we have to balance the grism wheel, in order to avoid grism wheel timeouts.

    This requires that we know the weights of all items, including the focus pyramid and the Lyot stop.
    See the table below, which list the weights of the items including that of their holders.
    grism weight (gram)
    Gr#3 100
    Gr#4 105
    Gr#5 110
    Gr#6 110
    Gr#7 120
    Gr#8 165
    Gr#9 160
    Gr#10 105
    Gr#11 110
    Gr#12 105
    Gr#13 150
    Gr#14 110
    Gr#15 100
    Gr#16 120
    Gr#17 230
    Gr#18 ~160
    Gr#19 ~160
    Gr#20 ~160
    Focus Pyramide   90
    Open Lyot 100
    WeDoWo 110

    Furthermore, wheel positions 0 (Focus-Pyr), 3, 4, all have an extra baffle mounted on the shutter-side of the wheel, which weigh 35 grams each.


Back to top Last modified: 18-Aug-2015