Quick Start

  1. Run MAlign and analyze the images in the default (spectroscopic) mode. Send the moves.
  2. Press the Load button on coma.tcl.
  3. After setting up on a field (in particular, make sure you are at the right physical rotator angle, if you are doing imaging), click on the appropriate mode button (Spectr. or Imaging).
  4. Start guiding once the DCS has corrected pointing for the new secondary tilts.
  5. When imaging for a long time, you may want to update the secondary tilts by clicking Imaging again in between exposures. PAUSE GUIDING when you do this, and wait for DCS pointing to compensate before resuming.

Note: If you are in spectroscopic mode you do not need to update secondary tilts for different rotator angles. All rotator angles use the same tilts.

Description of the Tool

The Coma Control Tool is shown below:

In operation, the main title should show the relevant instrument (e.g. "ESI Coma Control Tool" as shown here).

The two top blue buttons, labeled "Spectr" and "Imaging", are used to select one of the two modes. In spectroscopic mode the secondary tilts are independent of physical rotator angle (ROTPPOSN), and are optimized to eliminate coma in the telescope focal plane. In imaging mode the secondary is tilted to the position which will eliminate coma at the instrument's detector. The tool will minimize coma for imaging only at the current physical rotator angle, ROTPPOSN. It will not actively "cone" the secondary as is done on the high-precision f/25 secondary use with LWS. Hence if you are in imaging mode you may want to tweak up the secondary tilt in between exposures to keep image quality good.

Note that thebutton for the currently selected mode has a border around it. The current mode is also shown in the following line of text.

Below the mode buttons are three informational lines. The first line shows the mode, and in the case of imaging mode it shows you the value of ROTPPOSN at which you last selected imaging mode. The second line contains the current rotator angle and tilts. The third line contains an estimate of the coma error in imaging mode. (In spectroscopic mode or after clicking "Reload" this value is not relevant.) This value is calculated from the difference between the current value of ROTPPOSN and the value when you last clicked "Reload". This can be used as an indicator of when you should update the tilts if you have changed rotator physical angle significantly since last updating them.

The "Reload" button is used after running MAlign, analyzing images in the default (spectroscopic) mode, and sending the tilts. The coma control tool will read the current tilts and assume that these are the tilts relevant to spectroscopic mode. WARNING: if you click "Reload" while you are in Imaging mode you will incorrectly reload the current, imaging tilts as the spectroscopic tilts. Unfortunately the only way you can recover is to either reset to the spectroscopic tilts by hand and clicking "Reload" again, or by rerunning MAlign.

At bottom left is a Help button which briefly describes the use of the tool, and on bottom right a Quit button. WARNING: it is generally not a bad idea to click the "Spectr" button before quitting, if you are in Imaging mode.

A Wish List

Some other modifications could be made to the way we handle coma control. A number of DCS keywords could be implemented:

Also, it would be better if the MAlign analysis software measured the tilt offsets (as it now does), and then applies them to the secondary tilts stored in the image headers. Currently it adds the tilt corrections to the current values of (secthetx, secthety). Now if you were to send the corrections again (e.g. if you wer unsure as to whether they took effect the first time), you would overcorrect. If, instead, you were correcting the values stored in the FITS headers of the images, you would always be sending the secondary to the same absolute tilt. Also, if a problem occurs and you lose track of what tilts should be set, reanalyzing the latest MAlign data set will set to the exact same tilts as it did before.

This would require that instrument headers start carrying the secthetx and secthety keywords. The MAlign analysis software would look for these, and if it does not find them it would read the current values as it does now.