RCC autoguiding

From CCDSH
Revision as of 16:19, 28 September 2016 by Apal (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

RCC autoguiding

Currently, two autoguider mechanisms are installed on the RCC telescope:

  • There is an off-axis guider, installed at the spectrograph frontend. This guider camera observers a field approximately 14' south from from the main optical axis of the main (1m) telescope.
  • There is a guider telescope on which an FLI guider camera is installed. This guider camera sees roughly

the same field what the main (1m) telescope is actually observing.

Both autoguider cameras can be controlled using the same type of script, however, the off-axis guider needs some tweaking and focusing before observations. The control scripts can be invoked from the m5 computer, where these are named as

  • rcc-autoguide.qsi for the off-axis guider; and
  • rcc-autoguide.fli for the guider telescope guider.

Important note

Please note that the autoguider algorithms are rather sensitive to the roll of the respective cameras. Since both cameras are now attached via simple eyepiece support mounts, this roll can easily be misaligned (either unintentionally or deliberately). One must certainly take responsibility for any intentional aligning and/or (un)mounting of these cameras.

General usage

Usually the default options are adequate, therefore only running it without arguments should suffice. Otherwise, the exposure time and the relaxing time can be set with the following command:

$rcc-autoguide --exptime T --relax m

where T and m are integers in seconds. Here, and in the following we refer the autoguide scripts as $rcc-autoguide which can either be rcc-autoguide.qsi if we are doing off-axis guiding or rcc-autoguide.fli if we use the guider telescope and the FLI camera mounted on the guider telescope.

Before starting the autoguiding process, check the temperature of the respective guider camera. In case of off-axis guiding, please also check the guider X-stage as well as set the telescope focus approriately (see below for further details).

Basically, the autoguide process performs the following.

  • First, an initial image is acquired (with an exposure time of T) which is used as a reference throughout the autoguiding process.
  • In an autoguiding cycle, an image is exposed with the exposure time of T, the necessary autoguiding offset is obtained by computing the offset between this image and the reference image and then the telescope position is adjusted accordingly.
  • After adjusting the telescope position, the autoguider process waits for m seconds. This part is called relaxing, i.e. when the whole autoguiding subsystem does nothing, just simply waiting (and relaxing).

There are two algorithms available in the autoguiding scripts which compute the offset between the reference and the current image.

  • The fine matching algorithm needs less number of stars, even 3-4 stars can be adequate for a nice and safe autoguiding. Use the argument --fine for both $rcc-autoguide scripts to use this mode.
  • The point matching algorithm needs more stars (usually at least 8 or 10, but the safe is to have 14-16 or more). Use the argument --point to use this mode.

The main difference of these two modes is the following. The fine matching algorithm expects that the difference between the current and the reference image is within a few arcseconds and the autoguiding process will completely fail if (due to some reason) the offset is too large. In the case of the point matching, this offset can be very large (even in the range of arcminutes), the algorithm is capable to recover and lock again the autoguiding. However, this robustness needs more reference stars as well...

The scripts print the number of stars found on the reference image and the current images as well as the match success ratio and the status of the autoguiding, so one can easily fine-tune the parameters.

Some notes:

  • The autoguider process can safely be breaked during relaxing.
  • In general, use the fine matching algorithm for the off-axis guider and use the point matching algorithm in the case when the guider telescope is employed.
  • There is a small but significant mechanical "backlash" between the main telescope and the guider telescope which means that the pointing of these two telescopes can have a long-term drift of 3-5 arcseconds. This drift can be significant for the fiber-fed spectrograph, so use the off-axis guider (and hence the fine matching) if observing with the spectrograph for longer periods.

Testing and configuration

In order to fine-tune on the autoguiding process, first, one can acquire a test image:

$rcc-autoguide --exptime T --output test.fits

Use the command line argument -x or --xpa to let the autoguider script display the acquired image automatically in a DS9 window.

Both the autoguiding process and this test image acquisition prints the CCD temperature. Use the command

$rcc-autoguide --temperature TEMP 

to set the temperature to TEMP degrees of Celsius. -30 degrees of Celsius are usually fine, but if the cooling power can be stabilized at a maximum of 80-90%, even colder CCD temperature setpoints can be used.

Off-axis guiding: centering the guider X-stage

In order to efficiently use the off-axis guider, one has to ensure that the guider X-stage mechanism is centered. If unsure, first reset the X-stage with the command

acectrl --guide-xstage-reset

Then, move it to the central position with the command

acectrl --guide-xstage-center

The first step (resetting) is neccessary if the spectrograph state is unknown or after a power loss. If the guider X-stage offset is reported to 58000 by the acectrl --status command, one can almost be sure that the X-stage is centered. However, unintentional tampering of the X-stage motor moves the X-stage mechanism while the controller do not sense this motion (i.e. the reported guider X-stage offset won't be altered).

If the test images acquired by rcc-autoguide.qsi is not in focus, use the acectrl --guide-focus +S or acectrl --guide-focus -S command in order to fine-tune in the focus of the off-axis mechanism. 1 step is close to 0.01mm motion in the main focal plane, so focusing in 50-100 step units might be efficient for an f/13.5 telescope. Note that the off-axis guider is quite sensitive to temperature variations.

Guider telescope guiding

The guider telescope does not need any fine-tuning. It is always (more or less) in focus. However, ensure with the CCDSH command

CCD> rcc guidercap open

that the cap of the guider telescope is opened.

Navigation
Personal tools