Guiding and Tracking Techniques

Guided Imaging

Long exposures generally consist of a number of shorter sub-exposures and because of that, the telescope must be kept accurately pointing to the target during the course of the sub-exposures.  A very precise mount may be able to track accurately enough for the sub-exposure duration but most do not.  That is where guided imaging comes in.  A guide star is selected and its centroid calculated with every exposure.  As the centroid moves from the starting position, commands are sent to the telescope to bring the centroid back to its original position.  However, the guiding routine is always acting after the error has occurred so there is an inherent delay  in this correction.  By the time the correction has been sent to the telescope has been sent, there might be a different and even opposite correction required.  A number of guide parameters can be adjusted to minimize this effect.

However, recall the discussion about pattern noise.  That can be minimized by dithering (intentionally moving the guide star position between sub-exposures).  When the resultant sub-exposures are aligned, any pattern noise will not be reinforced by alignment but sky details will.  If a subsequent statistical combining method is applied to the aligned, dithered images, the pattern noise will be greatly reduced.  Of course dithering must be such that the same location is not duplicated and CCDAutoPilot provides such a capability with its Enhanced Dithering option.  

There are a number of parameters that must be properly set to achieve successful guiding.
  • Minimum Move Time: This is the time (or movement) that must be exceeded by the guiding algorithm to cause the telescope to move.  This must be set to avoid "chasing the seeing" and causing frequent but unnecessary movement of the telescope.  In other words, set this value so that only those corrections that will impact the image will be sent to the telescope.
  • Maximum Move Time: This is the maximum time (or movement) that will be sent by the guiding algorithm to the telescope.  It must be set high enough to have a reasonable correction response time but not so long that an occasional cosmic ray or other effect can cause the telescope to move too far so that it has to come back on the next correction.
  • Aggressiveness: This is a measure of how much of the calculated correction is actually sent to the telescope.  At first blush, one might be tempted to send the complete correction but recall there is a lag from the time the error is calculated until the telescope is moved.  So my sending less than 100% of the correction, the tendency to overshoot is minimized.  On the other hand, setting it too low may mean the telescope never catches up to the starting guide star position.
  • Guide Exposure: During the guide star exposure, the guide star position is essentially being averaged for the duration of the guide exposure.  If the guide exposure is set too short, the telescope won't have time to respond, there will always be a lag and we will be chasing the seeing again.  Set it too long and the corrections will be delayed too much to properly correct for mount tracking and elongated stars will result.  On the other hand, the guide exposure has to be long enough to get an adequate SNR for the guide routine to be able to accurately calculate the star's centroid.  (Note: there is a type of guiding called AO for Adaptive Optics, in which a mirror or piece of glass is moved.  As such it can be moved more quickly due to its lower mass, compared to a telescope.  In this case, shorter exposures are desirable since the lag is greatly reduced because of this lower mass.)
  • Dithering: How much dithering is enough?  Too little and the pattern noise will not be shifted enough from frame to frame so that it won't be removed in stacking; too much and the guider will spend a lot of time recovering from the dither, reducing data gathering efficiency.
  • Guide Star Selection Algorithm: Normally for guided imaging, you want the brightest guide star you can find for best guiding performance.  With narrow field imaging and guiding, getting a sufficiently bright guide star is generally a challenge.  However, when using a wide-field guide scope, you may accidentally find a guide star in the FOV that is too bright (saturated).  As you might expect, guiding accuracy is impacted if this saturated star, being the brightest in the FOV, were chosen to guide on.  CCDAutoPilot will reject any guide star whose peak value exceeds 55,000 ADU to avoid this problem.  It will choose a star whose peak value is less than 55,000 ADU automatically.

    CCDAutoPilot has tools to enable efficient starting point settings for these variables.  The Guide Calculator on the Tools page suggests minimum and maximum moves.  The Suggest button on the Tracking & Guiding page recommends a Maximum Dither that is appropriate for your system, based on your entries in the Settings page.  For non-AO operation, I suggest a minimum guide exposure of 3 sec.  The maximum guide exposure depends on how well your mount tracks.  Longer gives more averaging of the guide star's position.  There is an Auto Guide Exposure facility that allows you to set the minimum and maximum guide exposure so that your Target Guide ADU can be achieved by automatically setting the guide exposure between those two limits.

    Finally, CCDAutoPilot has a number of recovery options in case the guide star is temporarily lost.  See the Guiding topic for more details.

    Guiding with Adaptive Optics (AO)

    AO guiding consists of using an additional optical element, either a mirror or a piece of glass, in the optical path.  This element is driven to correct for positional errors in the guide star due to seeing (to a first order), mount and other sources of error.  Instead of moving the entire telescope, as is the case with conventional guiding, only the optical element, which is a much lower mass than the mount is moved.  This has the benefit of being able to move much faster than is possible with a mount and can correct for some seeing issues.  

    The optical element has limited travel and at some point, it will run out of range.  This limitation is resolved by moving the mount when the AO gets near its maximum movement.  This is causes "bumping" the mount.  The mount is moved while the AO is guiding so the AO effectively corrects for any disturbance induced by the mount bumping.

    There are therefore two calibrations that are required - mount bumping and AO.  AO calibration is done once (manually) and mount bumping normally has to be done at every sky location.  CCDAutoPilot will properly provide the calibration for mount bumping once initialized.  Thus, all you need to do is calibrate the AO using your camera control program, initialize CCDAutoPilot and you won't ever have to worry about any calibration for your AO system unless you change something in your physical camera arrangement.

    Unguided Imaging

    If a mount tracks well enough for a sub-exposure duration, then unguided imaging is a much simpler operation.  Of course, one should dither for the same reasons one dithers with guided mitigate pattern noise.  However, even the best mounts will slowly drift off target over a time if a substantial number of sub-exposures are taken.  CCDAutoPilot has the ability to correct pointing to the target periodically.  Use the Realign To Target Frequency feature and set the period to whatever value you find you need.  30 minutes might be a good starting point.

    Non-sidereal Objects

    When imaging objects that move at a non-sidereal rate, such as asteroids and comets, unguided imaging is the best way to go. CCDAutoPilot, in concert with TheSky, can make this job a lot easier. When an object is entered into CcdautoPilot via the Get function on the Session page, the relevant differential tracking rate information is entered into the target database. Additionally, when the time comes to image that object, the object's coordinates will be updated from TheSky so that the mount can slew to the correct coordinates at the time of target imaging, and not when the session plan was set-up or indeed when the session was started. For faster moving non-sidereal objects, the mount can be optionally commanded to slew to the updated coordinates before each exposure by checking Move to updated coordinates on the Options page. Assuming your mount is up to the task, this can make subsequent alignment for stacking of comets much easier. With TheSkyX, the native precision can approach 1 arc-sec., depending on how heavily the mount is loaded. Thus, not only are the tracking rates updated but also the target coordinates are updated and optionally the mount is moved to those updated coordinates before each exposure.

    For comets, a good choice is to use the Shuffle template in combination with a suitable Loop_series setting, to insure adequate stellar movement between exposures so that the background stars can be easily removed in processing. For example, if you are doing LRGB imaging with 60 sec. exposures, there will be more than 3 minutes of RGB between one L and the next. The same holds true for R, G and B. Loop_series determines the number of LRGB sets you take.

    Meridian Crossing with a German Equatorial Mount

    For best quality imaging, it is always desirable to collect data through the least amount of atmosphere and this means around the meridian.  Unfortunately most German equatorial mounts (GEM) can't track indefinitely through the meridian.  CCDAutoPilot provides automatic meridian crossing detection and supports a number of setup options to maximize imaging near and through the meridian so that as little time is lost as possible.  There is also an option to terminate imaging at the meridian if that is your desire.