Making it Happen

by Don Selle

Photo courtesy of Don Taylor

Because we are imaging dim objects, some barely brighter than the background sky, astro-imaging sets up a constant battle the imager must wage against noise.  In part-1 of this article, I explained in detail how longer exposures can improve the signal to noise ratio (SNR) of your target object when compared to the “noise” of the sky glow around it and the noise generated in your imaging electronics. 

What is Autoguiding?

Autoguiding uses images from your guide camera (separate from your imaging camera) takes rather rapidly. These images are analyzed by the computer controlling the guide camera to measure the small changes in the position of a guide star from image to image.

Using these measurements, the computer sends movement commands to your mount to restore the guide star to its original position in the next image. These rapid adjustments keep the mount tracking the sky more closely than an unguided mount would.

Why Autoguide?

The short answer to this question is that autoguiding is a technique that allows long exposures of the night sky, from a telescope which tracks the movement of the sky imperfectly.  When these imperfections lead to tracking errors, your images may develop “motion blurring” of your target and oblong background stars. 

Taking longer exposures means that your imaging mount must track the movement of the sky more accurately than shorter exposures require.  Keep in mind, with most mounts, longer exposures generally require your mount to be autoguided to prevent this.

Nice symmetrically round stars are the first indication of good autoguiding. Catching fine detail in your target, limited only by the image scale of your setup and how steady the atmosphere is (aka “seeing”) is the actual goal of autoguiding.

There are plenty of reasons why you might want to add autoguiding to your astro-imaging workflow. Perhaps you have been imaging using many short duration unguided sub-frames and you want to improve the quality of your results. Maybe you have gotten the bug to try your hand at narrow band imaging or want to blend some narrow band subframes into your final image to help bring out detail your current workflow seems to miss. Maybe you want to take longer sub-frames from your camera tracker, or you just want to take the next step on your astro-imaging journey. Whatever the reason, adding autoguiding to your workflow can be relatively straightforward, if you understand why autoguiding works, and what you need to do to successfully autoguide so you can take longer exposures.

This article assumes you are using either an equatorial mount or an alt azimuth mount on an equatorial wedge. When we speak of your mount “tracking” the sky, we are referring to how closely the average speed of the RA (Right Ascension) axis of your mount matches the speed of the rotation of the Earth. The instantaneous speed of the RA axis may vary (speed up or slow down) from the average, and this error (or variation from the mean) is usually expressed as +/- so many arcseconds.

 

In order to help get you started autoguiding, lets break things down a little bit into the following topics:

• A few preliminaries – what you need to know before getting started
• Equipment and software required
• Autoguiding Workflow

A Few Preliminaries

• How accurately does your mount need to track?

This question can be answered in two ways. The first way is that required tracking accuracy  depends on the “image scale” of your imaging scope and camera combination. Image scale is a function of the focal length of your scope or camera lens and the size of the pixels in the imaging chip of your camera and is expressed in arcseconds per pixel. 

In general, the larger the image scale, the less accurate your mount needs to track. This means that imaging wide field DSO’s with image scales ranging from 1.5 to 4 arcsecs per pixel will be much more forgiving of your mount tracking than high resolution imaging of galaxies or planetary nebula with 0.7 to 0.4 arcsecs per pixel.

Image scale is inversely proportional to the focal length (f) of the telescope or camera lens. This means that the longer the focal length is the smaller the image scale will be. Image scales can vary from greater than 40 arcsecs per pixel for a wide angle camera lens (f=35mm) and camera to less than 0.5 arcsecs per pixel for a large SCT (f=2400mm) with the same camera attached to it.

The second way to answer this question is based on the technical specifications of tracking accuracy for a specific mount design. Many manufacturers of equatorial mounts will specify their “tracking accuracy”. These are representative numbers and may vary a bit from mount to mount, but the idea here was to demonstrate that a given mount could be successfully autoguided. Keep in mind that this is only a relative measure of how well a given mount will track.

Back in the day, the rule of thumb was that a mount with +/- 15 arcsecs error and PEC (periodic error correction explained below) could be autoguided for widefield DSOs, but more demanding imaging would require a mount with less error. A good mount would have +/- 7 arcsecs and a very good mount (as in expensive)  +/- 5 arcsecs or less, and if they were really good, did not include PEC as it was unnecessary.

• What causes mount tracking errors?

The causes of mount tracking error can be grouped into three categories, imperfect polar alignment, mount mechanical issues, and environmental causes.

Inaccurate Polar Alignment. When you polar align your mount, you are adjusting the orientation of the RA axis so that it is absolutely parallel to the Earth’s axis of rotation. This orientation, and the correct speed of rotation of the RA axis will ensure that your OTA will accurately track any point in the sky as the Earth rotates.

It is not actually feasible to polar align without having some amount of error in how parallel your RA axis is to the rotation axis of Earth. These small differences over time will cause a star that is centered in your camera to drift from the center position. This drift will be mostly in the north-south direction and its speed will vary depending where in the sky you are pointed 

Mechanical Issues in your mount and imaging train. Your mount is a mechanism and as such, its performance is subject to its design and how precisely it has been manufactured. In addition, as your mount tracks, the center of gravity of your imaging system moves too. Many issues such as axes that are not at precise right angles, imperfections in bearings drive gears and drive motors may not be easily fixed, but some sources of tracking error can be. These include:

• Any loose fitting pieces in the mounting of your OTA or guidescope that will allow them to move as the mount tracks the sky.
• Looseness in the fittings that attach your cameras to OTA and guidescope which allow the camera to flex or rotate as the mount tracks
• Instability of the tripod and mount which might allow movement as the mount tracks
• Periodic error in the drive of your RA axis (dealt with in more detail below)
• And perhaps the most important thing is the weight of your imaging system. If it is too much for the mount to handle you must loose some weight or get a bigger mount.

PEC – Periodic Error Correction. Most mounts have worm gear drives on each axis. One of the characteristics of these drives is that due to mechanical imperfections, in the gears and support bearings, the speed at which the axis rotates will vary over time. Since the worm rotates many times for each rotation of the axis, any imperfections in the worm that cause this speed change will happen over and over at the rate the worm rotates, and are called periodic errors. 

Periodic errors (PE) can be reduced by mechanical adjustments or by upgrading the precision of the manufacturing tolerances of the worm and its bearings (as in a mount tune-up) and in truth higher end mounts cost more due in part to this increase in precision components. While PE can be reduced mechanically, it never disappears completely.

Most modern mounts include periodic error correction routines in their mount controllers. Once they are initialized with mount specific data, they will speed up or slow down the RA axis drive motor to counteract the mechanically induced PE.

Initialization typically involves autoguiding your mount at a rapid rate for several rotations of the worm, while the mount control system records the guide corrections. The timing of these corrections are stored in memory by the mount control system and used to make the correcting speed changes automatically. These speed changes are separate from those given to the mount by the computer controlling the guide camera.

In order that the guide corrections to be recorded will be as accurate as possible, there are a few things you need to do before collecting the PEC data. These will typically suggested in the instructions for your mount, but if not, here are steps you can take to ensure your PEC data is as accurate as possible. 

• use your imaging camera configured in software as your guide camera
• have it aligned as close to north/south as possible. You do this by turning mount tracking off while taking a long image. The star will drift making a streak on the image. Rotate the camera so that this aligns in the E/W direction.
• calibrate your camera as if you were guiding with it once it is aligned N/S.
• turn off any guide corrections in the N/S or Dec directions when you are collecting the PEC data

Mirror Flop. This is an error mechanism that occurs primarily on SCTs where the primary mirror moves on an internal light baffle in order to focus the OTA. The fit of the primary mirror over the baffle needs to be fairly loose so the mirror can move back and forth.

Sometimes the unconstrained primary will move oddly shifting the image on the camera. This movement is called mirror flop. 

Modern SCT’s (especially those with improved optics) designed for astro-imaging will have a means to lock the primary mirror in place once rough focus is achieved. Typically, this means that you will need to acquire a fine focuser that mounts on the visual back of your OTA since you will need to adjust your focus as you image throughout the night.

Environmental Conditions. Unless you are working in an observatory dome, astro-imaging is an outdoor activity. That means that your imaging system will be subject to the outside environment.

While your scope appears to be steady, wind gusts can play havoc on the tracking accuracy of your mount. The worst part is that this really cannot be corrected by autoguiding.

Some imagers, especially they will be set up for several days (such as when they are at star parties) will take the time to assemble a wind break, or even a pipe and tarp enclosure to break the wind.

Then there is that cow in the pasture that startles you when it comes to see what you are doing, causing you to bump your mount (not too violently I hope) knocking your mount out of whack. Yep this has happened!

• What new equipment do I need for autoguiding?

Guide Camera. It is obvious, but you will need to have a second camera if you want to autoguide.  Cameras for many years have been designed and produced specifically for this service.  These cameras have much smaller imaging chips, and are designed for easy mounting to a guide scope or off axis guider (see below).

Typically the imaging chips on these cameras have pixels which are smaller than your imaging camera. This is intentional, as we want to keep the image scale as small as possible, so that changes in the position of the guide star on the chip affect a larger number of pixels, thus the guide camera can measure smaller shifts from image to image.

Back in the day, the only way for the imaging computer to make guide corrections was by communicating with the guide camera. The guide corrections would be passed to the mount by “relays” built into the camera, through a guide cable connected to the mount controls. As a result, guide cameras have this feature built into them.

Guide scope or Off Axis Guider. Since you will be using a second camera to autoguide, that camera must be able to image a patch of sky very near to your target so that it can find a guidestar to monitor. This means that you will need one of two additions to your imaging system, either a second smaller OTA is mounted on to your imaging OTA, or a device called an off axis guider (OAG) is used. 

An OAG mounts in front of your imaging camera. It has a small “pick off” prism in the light path which redirects a small part of the light at right angles, to where a mounting is provided for your guide camera.

An OAG is much smaller and lighter than the typical guide scope. Since it is connected directly in front of your imaging camera and may be attached to it, an OAG does not require any mounting hardware.

A guide scope is just what it sounds like. It is typically a small refractor which is mounted on a rail connected to your OTA or to the mounting rings holding your OTA. The guide scope must also have a way for you to mount the guide camera and to focus it. This focus need not be automated unless you are imaging remotely.

So which is better an OAG or a guide scope? For imaging systems using a short to medium focal length OTA, a guide scope will be the best choice, and works well. It is in many ways more convenient to use once aligned with the main scope. It is generally easier to find suitable guide stars as the guide scope will generally have a shorter focal length than the main scope, giving the much smaller chip in the guide camera a larger field of view. 

The downside of a guide scope is increased cost and added weight. Mounting the guide scope on top of the main scope also shifts the center of gravity higher and may require additional counterweights. All of this can be an added burden on your mount which in extreme cases can degrade your mount’s tracking.

If you are imaging at long focal lengths though, you may not have a choice but to use an OAG. due to “differential flexure”. Differential flexure occurs when the weight of the guide scope causes very minute changes in the pointing of the guide scope as the OTAs tracks the sky. 

At small image scales this flexure need only be a few millimeters, and because the alignment of the two OTAs change ever so slightly it causes the image in the main camera to shift. This appears as drift or poor tracking in the main image which is actually caused by the guiding. 

I first encountered this when I upgraded to a 12 inch RC telescope with a focal length of 2400mm and it was quite frustrating to solve. Your experience may vary.

Mounting accessories. Your guide scope must be firmly attached to your imaging OTA. The mounting should also allow you to align the guide scope fairly closely with your imaging OTA. 

For smaller guide scopes which are the size of a finder scope, such a mounting is relatively lightweight and uses the same type of alignment guide rings as a finder. It might even mount in the same fitting you would use to mount your guider! 

For larger guide scopes, the requirements for mounting become greater. Rail clamps and larger guide rings are available, as are devices that allow you to align the guide scope with alt/az adjustments built in. All of this will add to the weight and cost of your system so keep that in mind when deciding which to choose.

Software for autoguiding. Most software for controlling your imaging camera will also contain a means for autoguiding. 

At a minimum your guiding software should have these features:

• The means to control your camera’s exposure including exposure delay. 
• The ability to dark subtract your guide images on the fly. This is critical for eliminating hot pixels which may look like stars or can otherwise interfere with your guiding
• A means to turn guide corrections on and off in each axis
• An automated means to calibrate your guide corrections (discussed below)
• The capability of adjusting the guide corrections based on the declination of your target. This feature allows you to reuse your guide calibration for multiple targets in an imaging session, and if you are permanently mounted in an observatory, also from session to session.
• A graphing feature so that you can visualize the guide corrections which are sent to your mount
• A means of selecting how guide corrections are sent to your mount. If you have an alternative such as direct guide or pulse guide use it and eliminate the guide cable. This sends the corrections directly to your mount’s controller from your computer.
• Dither – this changes the center of each of your image subframes by a few pixels, by moving position of the guide star in the guide image. This is a very effective tool for noise reduction in your final image.

There is also an awesome open software guiding program called PHD2 which many imagers use in lieu of the built in guiding software. You can’t go wrong using it, and there is lots of information and support for using it.

https://openphdguiding.org/about-open-phd/

• Autoguiding workflow. You will benefit greatly by using a specific workflow for your own autoguiding. Here is a suggested workflow you may wish to start with.

Prior to an imaging session

Here are a few tasks you may want to complete, some on a night not so good for imaging, some when its cloudy outside.

• Know the parameters of your imaging system such as image scale.
• Do the PEC training for your mount
• Understand your guiding software – what algorithm is used for calibration, how do you set up dither etc. 
• Select the method of sending guide corrections to your mount and test it  if your mount can 
• Other tasks you may want to complete are addressed in the AP Corner article from May 2021 

https://www.astronomyhouston.org/newsletters/guidestar/ap-corner-%E2%80%93-may-2021

During your imaging session. Having a specific workflow for your imaging session is very important if you want. Here are a few steps you might consider to 

Accurate polar alignment.  The more accurate your polar alignment is, the less declination drift you will have to correct. The October 2021 Guidestar has an article that describes how to do this. You can find it here: https://www.astronomyhouston.org/newsletters/guidestar/ap-corner-polar-alignment

Calibrate your guide corrections. In order for your guiding to be accurate, your guiding software must be calibrated in order to make guide corrections of the right amount to recenter your imaging camera. This is typically done with an automated process in software which takes an image with your guide camera, drives your mount at guide speed for a specific time period or a specific distance amount and measures the offset of a guide star in a subsequent image.  This is done in all four directions your mount moves and a guide rate is calculated for each direction. In addition, the angle your camera is oriented from N/S is also measured and used to adjust the amount of correction that is sent to the mount after each guide image is measured.

• Set up and take down. For an imaging system you set up and take down you will need to do a guide calibration each time you set up and polar align your system. This is due to the fact that each time you set up and take down your imaging system, the alignment of your system changes. Your polar alignment changes as does your weight distribution as well as other things. It is good practice to recalibrate each time you set up.

If you plan to image over several nights from the same position, you could reuse your first calibration. I would recommend however that you recalibrate, as it does not take too much time and could save you more time if your calibration is thrown off by things you cannot anticipate.

 

• Permanently mounted. If you are permanently mounted, it is safe to re-use your calibration from session to session. It is good practice however, to periodically check your polar alignment and if any changes are made, you may also want to recalibrate your guiding.

 

• Camera Rotation. If you have a camera rotator or rotate your camera by hand for different targets, re-calibration may be necessary. With a camera rotator, if your guiding software takes into account the angle of your rotator when the calibration is done and can adjust the guide corrections for subsequent rotation angles, you probably will not have to re-calibrate. If you adjust your camera angle by hand, you must re-calibrate.

 

• Do I need to calibrate on the celestial equator (CE)? This is a practice that is suggested by many, as the tracking errors on the CE will actually be larger in terms of pixels than if the calibration is done closer to the pole. If you choose to do this, you must be sure that your guiding software compensates for changes in declination.

In my opinion, this is a good practice if your imaging system is permanently set up as you can reuse a previous calibration provided you use a camera rotator, and your software compensates for the rotator angle. If you are using a set up take down system though, it will work but you might want to use a different approach.

Typically, I will find a fairly bright – say 5^th magnitude star nearby may target. Calibrating on such a star works very well. The brightness of the star ensures that your calibration routine will pick it up in each guide image. When your calibration is completed, it is a good target to get your initial focus. If you work more than a single target each night, especially if you change your camera angle by hand for each target, this approach works will and is pretty time efficient too.

Use Dither 

If your software supports dithering between sub-images, you should use it. It takes up a little bit of imaging time but the results are always worth it.

Dithering is a big help in reducing random noise in your image. It is good practice to set a fairly large dither value, especially if you use a one-shot color camera. These cameras tend to have a bit of pattern in the color noise which dithering knocks out.

Check your sub-images when you first start up to make sure that the dither routine is doing what you think it should. Catching a problem early in an imaging sequence can save a lot of lost time later.

 

Finally, the authors of PHD2 have put together a best practices document for using PHD2. It is a good source of information and while specific to PHD2, it contains many tips and suggestions that apply generally to all autoguiding.

https://openphdguiding.org/PHD2_BestPractices_2019-12.pdf

Check it out and good imagining!