Obs-Session Data

Here's some stuff from recent obs-sessions.

First, here's a light curve from the eclipsing binary star HW Viriginis:

The position of HW Vir in the sky seems to be a place where a lot of satellites move.  Here's an example of a single 3s image:

Here's a combined RGB image of open cluster Messier 46, with planetary nebula 2348 in the background.  The color information of the stars and nebula will hopefully tell me a little bit about their situation:

A long exposure (about 450, 3 second images) of the Eskimo Nebula (planetary nebula, NGC 2392):

Here's the same image as above, but stretched to show the faintest stars:

Here's a long exposure (about 450, 3 second images) of the Globular Cluster Messier 5:

Here's a video showing the motion of asteroid (6) Hebe:

... and asteroid (433) Eros:

Observing asteroid (6) Hebe

From about 03:30 - 05:30 UTC on 19 April 2019, I collected some reflected photons from the main-belt asteroid (6) Hebe.

At the time of the observation, (6) Hebe is in the constellation of Gemini.  The approximate RA and DEC was: 06h 55m 24.3s +19d 38m 20s.

I took 700, 4 second exposures.  Even during the observation, I noted that I probably should have done 3 second exposures.  When I decided on 4s exposures, the tracking was looking really nice.  Maximizing the exposure time improves the signal-to-noise, so being able to do 4s exposures would be better than 3s exposures.  The problem is that the tracking quality kept changing, so much so that many of the star images were short streaks.

So because of the pretty poor tracking, my first run at stacking and making some photometric measurements on the two reference stars in the field-of-view used 544 images.  So that means I've lost 156 images, or 22.3%.  That's a bit much, so I'll need to go back into the data to see if I can modify the parameters in the code to pull out a few more images.

Here's a single image:

Here is the stack of all 544 images (inverted greyscale):

Here's a movie showing the motion of (6) Hebe:

https://youtu.be/HA3sQQUsDLU

There are some strange things in this dataset that I need to look at carefully.  More on this later...

Eclipsing Binary Star HW Virginis

On 23 March 2019, I observed the eclipsing binary star HW Virginis.  Here is a preliminary light curve of the event:



The x-axis is the Julian date, and the y-axis is the relative magnitude compared to a reference star in the same field of view.

Here's a long exposure of the field of view:

Deepest Image Sofar

This image is a stack of 954, 3 second exposures of the XZ Canis Minoris region.  I'm pretty easily seeing 17th magnitude stars, and maybe some hints of 18th.

Obs-Session 190322

Had a pretty nice obs-session last night.  I collected data on R Leo, and the asteroids (2) Pallas and (433) Eros.  My only complaint is that I had trouble finding the targets and that task took too long for every target.  With my first target - R Leo - I even came back into my office and checked the coordinates because I couldn't seem to find the star patterns on the chart.  But I'm very stubborn so I finally found it.  Collected 100, 2 second, 2x2 binned images and 40 darks of R Leo itself, and then another 100, 2 second, 2x2 binned images of a nearby comparison (or reference) star.

I then proceeded to the asteroid Eros, which was still high in the sky but getting lower by the minute.  Had trouble finding the target, but again prevailed and finally figured it out.  The main trouble was that there are sooooooo many stars!

And here's Eros moving by showing you the first image followed by the last image in a loop:

I've also made a movie of the entire session with Eros:

https://youtu.be/31JWsSTi4Jk

I then moved on to (2) Pallas.  Same trouble finding the the target, but for the opposite reason.  In this part of the sky, the stars are SPARCE!  So simply finding a pattern of stars that fit into my small field of view (12 arcmin by 9 arcmin) was a challenge.  I was about to give up for the night when I finally saw a recognizable pattern which was surprisingly close to the target!

Here's an image of (2) Pallas:

... and here's the same kind of movie as above.  Notice how much less Pallas moves versus Eros.

my manual stacking here isn't perfect, so you'll notice that the star at the lower right appears to be moving.  It's not.  (2) Pallas is that bright one at upper left.

I collected 200, 2 second, 2x2 binned images of (2) Pallas.  Wow that's a lot of two's.

So now I need to reduce the data and then make the photometric and astrometric measurements.  That's all just tedious work which I'll get done in a blast of time in the next couple of days.

Photometric Transformations

On the evening of 01 March 2019 (UTC), I collected a new set of photometric data using my 'G' and 'R' filters.  The target was, once again, Messier 46.  I like this particular object because it's a rich open cluster with nice selection of stars and it has a bonus object NGC 2438.

I took 200, 4 second images with 2x2 binning with each filter.  Running it through my stacking program, about 180 individual images were actually used for each stack.

Here is the Green stacked image:

and the Red stacked image:

After running the stacked image through Astrometry.net to get the WCS transformation (which is used to convert x,y pixel locations to RA, DEC coordinates), I use the IRAF task 'starfind' to locate the stars in the field of view.

The next step is to perform aperture photometry on each of the detected stars.  The result of this will a measured flux for each star.  The flux can then be converted to astronomical magnitudes following the transformation algorithm described Chapter 6 of the CCD Photometry Guide published by the AAVSO.  To calculate the transformation coefficients, I used the Sloan g' and Sloan r' photometric standards from the The AAVSO Photometric All-Sky Survey Data Release 10.

Here is my plot for T_gr = 1.570281 (slope = 0.636828)

and for T_g_gr = 0.291380:

and finally T_r_gr = 0.022768:

So then I can compute the magnitude of each star relative to a bright standard star I've chosen within the field of view.  After computing these magnitudes, I can compare my measurements to the Sloan g' and Sloan r' magnitudes.  The results are shown in the following two plots:

Along the x-axis is the Sloan g' or r' magnitude of the star, and along the y-axis is the percent difference between my measured magnitudes and the catalog magnitudes. As you can see, the accuracy is less than 1.5% even for stars as faint as magnitude 14.5.  The accuracy drops to less than 0.5% for stars brighter than about 12th magnitude.

So overall I'm fairly pleased with these results.  I'm going to take one more set of data using my 'B' filter, which I can then compare to the Johnson B and Johnson V magnitudes to see if I get similar results.

Not that it will be scientifically viable, but at that point I'll have 'R', 'G', and 'B' images of this region, so I'll probably combine them to produce a color image as well.  It might be pretty to look at.