Variable Stars

After imaging for a while, I discovered some difficulties with my work, particularly after I'd discovered my ultimate solutions described below.

• Not all variables need the same exposure.  Some are brighter, some are dimmer (duh).  The brighter ones can saturate or at least get out the linear response part of the camera's response curve.  The dimmer ones may not show up at all.  Experts frequently mentioned that I should cut my exposure because I was saturating stars.  It may have been true for mag 5-7 stars, but the mag 10-13 stars I seemed to have picked above were fine.  Cutting my exposure would mean that I'dhave trouble getting the dimmer ones I'd already started.

• If you enter the data for RA/Dec wrong in Sequence Generator Pro you unsurprisingly end up imaging the wrong field.  The variable you wish probably isn't there, but others may be.

• Imaging the correct RA/Dec with the correct exposure will give workable data for a particular variable, but may also reveal others.  Should they be submitted to the AAVSO (I did) and to this Program(I did)?

• Imaging from my driveway with an unattended setup sometimes gets excellent images of trees and my neighbor's house.  Unsurprisingly plate solving doesn't work well with trees or houses.  It also makes me uncomfortable leaving that much hardware in a place visible from the street.   It is a good neighborhood, but still...

• Weather conditions can dramatically change over an eight-hour imaging run, particularly when you are asleep.  I sometimes had to dump data because of clouds or dew.

The bottom line is that I didn't end up monitoring all 32 of the planned variable stars, but did end up monitoring others.  Were the new stars the required long period or semi-regular?  No clue, but I'm sure that at least fifteen on the monitored stars will work and with over 200 observations uploaded to the AAVSO my results should be fine.

Observing Locations:       Bortle     Latitude    Longitude    Elevation

• • Brommelsiek Park         orange     38.723N    90.815W       644 ft

  • Home                            red           38.632N   90.562W        567 ft

Hardware:

One of the things I wanted to do was to see how simple I could make my setup.    I knew that apparently some DSLR variable star imagers actually used a camera mounted on a tripod but I did want the convenience of GoTos and tracking.  I didn't want to use a mount that cost a fortune so I opted to use an Alt/Az Celestron SLT NexStar mount my son obtained on eBay for about $60.  He gave me the mount for Father's Day 2015.   I used my trusty Canon T2i with a manual 300mm Aetna Tele Rokunar lens coupled with an M42 to EOS adapter.  The lens was a $40 binge eBay buy.  Both the mount and camera were run off 12v as much as possible.  FWIW the poor 8 y.o. T2i died on 7/15/2017 prompting me to submit my results before starting calibrations with a new T7i.  Part of the current delay is that the software I use has yet to catch up the the T7i's release.

Data Collection:

When I started, I used Stellarium to find the targets and ByEOS to capture the images.  I found that the combination required constant attention...slew and center a new target, switch to ByEOS, and start the sequence.  I quickly found and switched to Sequence Generator Pro as a one-piece solution.  I made a sequence from my target variables, sorted in order of RA.  Each target contains the RA and Dec, commands to center the target with plate solving (PlateSolve2), and exposure information (number of frames-9, ISO-800, and exposure length-30 seconds).  Since I was using a DSLR to capture images no filters were used on the 300mm telephoto lens.  For the most part, I opted to use a cheap NexStar ALT/AZ mount, knowing that I would get less-than-perfect GoTos and some field rotation when objects were high in the sky and near the meridian.  I originally started with 15 targets, but as time went on I kept adding more until 31 were part of a nightly run.    I did find out that SGP was somewhat limited to about 35 targets in a sequence, crashing if more were attempted.  In the beginning, I used my Lenovo laptop, but it does require a good deal of power when I’m off the grid.  This past spring (2017) I switched to an Intel PC Stick computer that runs on +5v.  I control the PC Stick for focusing and snooping with an Android tablet.  Power consumption went down quickly, but it does run much slower particularly during plate solves.  By the time of this submission, the hardware and exposure parts of the puzzle ran flawlessly assuming no clouds.  Flats were taken anytime I'd change/remove the lens from the camera.  Bias frames were taken once.  Darks were taken before and after an imaging run and were combined.  I used MaxIm DL to create the Master frames.

Data Analysis:

While the data collection took up the greatest amount of time working on the program, it was for the most part hands off.  I’d start an imaging run out at Brommelsiek Park or in the driveway and assuming no problems, it would merrily run all night.  At the Park, I'd sleep next to the camera on a zero-gravity recliner.   The data analysis part of the program was hands-on pretty much all the way.  The first steps required calibrating the images for the particular variable star with the Masters, extracting and combining the two Bayer green channels for each image, stacking the nine images, and then doing a plate solve.  I chose to use MaxIm DL v.5 for the tasks.  I pretty much religiously followed the instructions for MaxIm DL in the AAVSO DSLR Photometry software tutorials.  My only point of deviation was to use a VB script from George (see below) to extract and combine the two green channels before stacking.   Customizing MaxIm DL by adding buttons for the various tasks made things go a bit easier, but still, it was hands-on, and time-intensive.

Photometry:

My original photometry efforts involved using MaxIm DL and I’ll admit now that I almost gave up on the program.  It isn’t so much that MaxIm DL made it hard, it’s just that there was so much to do.  Hours were spent on the AAVSO site looking for targets, making finder charts, and consulting photometry tables.  Even more time was taken trying to identify the particular variable star targets, comp and check stars in my stacked images.  Everybody said it would get easier with practice.  It didn't.  MaxIm’s documentation is particularly sparse in regard to the information needed to fill out the form which eventually leads to the text file needed to submit the results to the AAVSO.  The AAVSO’s site isn’t particularly friendly about providing the information needed to be submitted when using a DSLR.  People who offered to help often suggested a CCD purchase along with a list of expensive filters.  As my level of frustration increased, I put off photometry on all but the first few targets, but continued to collect data (the easy part).  I did upload data to the AAVSO and it fit right in with the rest of the points in the existing light curves.

About five years ago I visited a local AAVSO member’s place out in the country and he showed me what he used to do the seventy variable stars that he was tracking regularly.  AAVSO members have access to VPHOT, a photometry tool that helps automate the process.  After almost a year collecting data I thought I’d give things one last shot and thought I’d look into joining the AAVSO so I could have access to VPHOT.  Original indications looked that it might help, but one post on the AAVSO forum looked to be a killer.  A DSLR using member flat-out said that VPHOT couldn’t be used to do DSLR variable work.  I emailed him about it and his reply pointed me at another member named George.  I emailed George with the same questions about VPHOT, DSLRs, and AAVSO membership.  He responded that my previously submitted data looked good and that VPHOT wasn’t written for DSLRs because of their frequent wide fields.  He said that he’d be glad to take a look at my data and would check to see if it worked.  I sent him a stacked, green plane image that I’d already submitted to the AAVSO.  Within a day, he responded to me that he’d solved it successfully, my headers were fine in uploads, and that my DSLR images would work.  At that point, I joined the AAVSO and started accessing VPHOT.  While VPHOT does have the ability to stack images, I opted to upload images that were already stacked to save bandwidth.  I deleted all AAVSO data that I’d submitted using only MaxIm DL and now all my AAVSO results were done with VPHOT.  Once my images are solved in VPHOT I download a file that is uploaded to the AAVSO's database (sounds redundant doesn't it).  Before the upload, I edit it in Notepad changing the camera from CCD to DSLR and the filter from V to TG.  Not exactly a plug-and-play solution, but it does work.  An added benefit is that VPHOT shows additional variables in my images and I've submitted my results for them too.

Sky Conditions:

Date            Observing Location     Seeing    Transparency

                                      (X/5)     (X/10)        

 April 2-3, 2016 Broemmelsiek Park      4          7             
    While I'd practiced what I was going to do at least four nights before this date, I decided that tonight would officially start collecting data for real tonight.

 April 8-9, 2016 Broemmelsiek Park      4          6             
    Collected data while at a Friday Night Open House at Broemmelsiek Park.

 May 3-4, 2016 Home                   3          5             

     The background sky is much lighter at the house.  Set up in the front driveway and worried about my equipment all night long.

 May 8-9, 2016 Home                   2          4               

     Set up behind the garage with a much smaller section of the sky available.  Got some nice pictures of trees and the neighbor's house.

 May 22-23, 2016  Home                    4          5             

     Set up behind the garage with the tripod extended as high as it would go.  Fewer problems with trees and the house.

 May 29-30, 2016   Home                   3          4             

 June 7-8, 2016 Home                   3          4             

 May 7-8, 2017 Home                   4          5               

     Started again after a wet spring with very poor skies.  First use of the Stick PC...plate solves really slow (30sec) but it did a great job of controlling the mount and camera.  Much easier to set up!

 May 28-29, 2017   Home                   2          3             

 June 10-11, 2017  Home                   3          3              

My Results:

The actual results of my work can be seen by clicking here to display the AAVSO's page with my 211 entries.

Final Comments:

It has been said that variable star observing can become addictive.  I can believe that.  I've joined the AAVSO and will continue imaging variable stars.  My only holdup is the "death" of the Canon T2i DSLR I used to collect data for my observations and Sequence Generator Pro updating to recognize my new Canon T7i.  The camera died about four weeks ago and I'm starting to miss gathering data. I've learned a good deal but know there is a good deal more to learn.  To that end, I've signed up for two AAVSO classes in the fall, one on DSLR Imaging Variable Stars and the second on VPHOT.  I'm currently thinking about what it would take to permanently place a camera and mount on my garage roof for a permanent setup to make data collection easier.