Using Auto Focus During Astrophotography

Bahtinov Mask

Bahtinov Mask

 

Auto focus software can significantly improve image focus; especially when images are captured over several hours during an imaging session.  One of the biggest challenges of Astrophotography is making sure images are in focus.

Manual Focus Method

Most astrophotographers do their initial focus using a Bahtinov mask to perform the initial focus.  The Bahtimov mask causes a de-fraction pattern to appear in the image of the stars.

The Bahtinov mask is placed over the front of the telescope while the telescope is pointed at a bright star.  Short exposure images are then taken to assess the accuracy of the focus.

The following images show the defraction patterns for a star that is out of focus and the same star in focus.

The difference is subtle.  If you carefully examine the images you will notice that in the left image the vertical diffraction line passes to the right of the center of the X formed by the other two lines.  In the right image it passes to the left of the center of the X.  If you examine the center image you can see that it passes directly through the middle of the X.  A Bahtinov mask is considered the Gold standard for establishing focus when doing astrophotography.  Without the mask is is very difficult to tell if an image is at the best focus point.

Why Auto Focus?

Unfortunately, the correct focus point changes throughout the night because of the changing outdoor temperature.  The only way to keep the images in focus is redo the focus several times throughout the night.  This can be very time consuming because it usually requires moving the scope away from your target, pointing it to a bright star for focusing and then moving the scope back to the object you are imaging.  Therefore a lot of time can be saved by using auto focus software.

Implementing Auto Focus

In order to use auto focus software the telescope focuser has to be controllable by your computer.  Therefore the telescope need to have a software controllable motorized focuser.  There are many ways to add this capability to a telescope.  One way is to add a motor drive to your existing telescope focuser.  Several manufactures sell kits to motorize various types of manual focusers.  Another Approach is to replace the existing focuser with a new motorized focuser.

My Auto Focus Setup

Moonlight Motorized Focuser

My telescope uses a Moonlight focuser like the one shown here.  It has a precision stepper motor which enables the software to adjust the position of the focuser in extremely fine increments during the autofocus process.

I use a program called Sequence Generator Pro to collect my astro images. It can be used to fully automate a night of astrophotography, which can include multiple targets.  One of the tools it includes is an autofocus routine.

In order to use the autofocus routine, the telescope must be fairly close to focus before the autofocus routine is run.  This may seem counter-intuitive, but the primary purpose of the autofocus routine is to keep the telescope in focus throughout the night.  It is not designed to focus an unfocused telescope.

How Auto Focus Works

The autofocus process requires some initial setup before it will work with a specific telescope and focuser.  The user must specify the number of steps the stepper motor will move for each move for each focus adjustment.  The user also specifies the number of points that they would like included in the focus curve (typically 9).  The routine will then move the focuser out 5x the specified step change to collect the first point.  It will then take a short exposure with the focuser at this position and make a calculation to determine the accuracy of the focus.  The value calculated is called the Half-Flux-Radius, HFR.  The larger the value of HFR the poorer the focus.  The software will then move the focuser in by the specified step size, take another image and calculate the value of HFR.  This process continues until the number of specified points have been collected.

Plot of HFR Focus Curve

After the data has been analyzed two reqression lines are are plotted forming what is called a “V” curve.  The point where these two lines intersect is the best focus point.  The software will then move the focuser to this position and take another exposure to verify its result.  If the calculated HFR is less than or equal to the lowest value of HFR measured during the autofocus run the focuser will be left at this position and the software will begin collecting images.  If the value is greater than the minimum measured the routine will either repeat or utilize its sophisticated algorithms to select the best focus point. This can happen occasionally occur when the sky transparency is less than ideal.

The autofocus curve shown is from a recent imaging session of NGC 869 (Double Cluster).

The following image is the confirmation image from the Auto Focus routine.  Click on the image to see the full size image.

NGC 869 Raw Image (Click for full size image)

 

 

ARRL SSB Sweepstakes 2017

It has been a long time since I have spent much time operating a contest.  I decided to do a part time effort in this years ARRL SSB Sweepstakes.  My primary goal was to work all 83 ARRL Sections, which is called a Clean Sweep.  Unfortunately I fell one short of a Clean Sweep.  The section I missed was Puerto Rico.  That was not surprising since Puerto Rico is still struggling to recover from Hurricane Maria.  They have much more important things to deal with than ham radio.  Not hearing Puerto Rico made me realize how fortunate and blessed I am.

Normally I do search and pounce contesting, but this time I spent a couple of hours running stations on 15 meters in order to improve my confidence running stations.  I ran stations on 15 meters until the band started to fade late in the afternoon.  When 15 started to fade I decided that was a good for a dinner break–soup, fresh baked dinner rolls and hot homemade apple pie with ice cream!  My wife really takes good care of me!

After dinner I spent a little time on 40 meters doing search and pounce.  The pickings were pretty slim, so I decided to throw in the towel and relax the rest of the evening.

Here is a summary of my score.

Start Date : 2017-11-18

CallSign Used : KC9QQ
Operator(s) : KC9QQ

Operator Category : SINGLE-OP
Assisted Category : ASSISTED
Band : ALL
Power : HIGH
Mode : SSB
ARRL Section : IN
Club/Team : SOCIETY OF MIDWEST CONTESTERS
Software : N1MM Logger+ 1.0.6903.0

Band QSOs Pts Sec Pt/Q
3.5    64    128    22    2.0
7       81    162    24    2.0
14     75    150    17    2.0
21    113   226    19    2.0
Total 333   666    82    2.0
Score : 54,612

73,

Fred, KC9QQ

First Astrophoto Processed with PixInsight

I am relatively new to astrophotography.  I have found it very challenging,  turning the nearly invisible smudges my camera captures during my imaging sessions into full color photos.  A typical astronomical object will look like the following image before it has been processed.

M1 “Crab Nebula” RAW

Weak Signal

As you can see, it doesn’t look anything like the beautiful images one is used to seeing. This occurs because signal is very weak.  Therefore most of the pixel data are congregated into the lower part of the photos histogram.

The object in the image can be made visible by a process called “stretching”.  Stretching can be done in Photoshop, but astrophotographers typically use software designed to specifically process astronomical images.  There are many special purpose astrophotography programs available.  Some are free, others are nominally priced while the more sophisticated programs are  priced similar to Photoshop.

After trying some of the free and low priced programs with mixed results I decided to download a trial version of the very popular PixInsight program.  PixInsight is like the Swiss Army Knife of Astronomical processing software.  It has hundreds of special functions, each designed for a specific processing task.

Trying out PixInsight

After I downloaded the PixInsight trial  version,  I watched a number of video tutorials trying to learn the basics of the software.  While the online videos were very good at explaining the use of specific tools in the PixInsight toolbox, none of them did a good job of showing the basic steps needed to process an image. After watching several videos, I still did not have a clear understanding of how to use the program.  However, I should point out that now that I have learned the basics of PixInsight, the videos have been very helpful in improving my skills.

An Excellent Book on PixInsight

Warren Keller’s Book on PixInsight

Just as I was about to give up on PixInsight, Warren Keller (no relation) brought out an outstanding book on PixInsight.  Warren’s book turned out to be just what I needed to learn the basics of PixInsight.    The first six chapters of his book walk the reader through the essential steps needed to process an astronomical image.  After the completing those steps I was able to produce  presentable images.  The rest of the book walks the reader through many of the advanced features of the program.  These advanced topics can be used to make your images really pop (assuming you have enough good data).

I won’t attempt to explain each of the steps that I used to process the “data” I collected on M1.  However, I only used the techniques taught in the first 6 chapters of the book.  I am still experimenting with some of the more advanced features found later in the book.

Astrophotographers refer to the group of images (sub-exposures) they collect on a specific astronomical object as data rather than photos.  I think the reason they use this terminology is that until all of the sub-exposures are combined and processed they really don’t resemble a  normal photograph.

My Processed M1 Image

For M1, I collected twenty 3 minute sub-exposures;   one hour of total exposure time. Most of the premier astrophotographers will collect 6 to 30 hours of data on a given target.  Therefore, I didn’t expect my image to be as awe inspiring as the images of these highly skilled amateurs.

You can check out the work of some of the worlds best astrophotographers on NASA’s Astronomical Photo of the Day site.

M1 Crab Nebula

Here is my finished photo. It is a significant improvement over photos I have processed by simpler means.  I highly recommend PixInsight software along with Warren Keller’s excellent book “Inside PixInsight”.  His book gave me the set of training wheels I needed to get started with this incredible program.

Fred