Author Topic: Astrometric distance between two stars  (Read 2328 times)

Offline CraigNZ

  • PixInsight Enthusiast
  • **
  • Posts: 94
Astrometric distance between two stars
« on: 2015 September 04 01:58:04 »
I use DynamicPSF to precisely determine the pixel location of stars on an image.  Is there a script for calculating the distance and angle between two stars?  I did not see it in aperture photometry but I guess it could be done using the spreadsheet tables and a macro.

Would be nicer if DynamicPSF could show the distance from the previous star selected to the current star.  If Image Solve has been performed on the image the distance can include any field curvature and displayed in arc seconds.

Offline CharlesW

  • PixInsight Enthusiast
  • **
  • Posts: 87
Re: Astrometric distance between two stars
« Reply #1 on: 2015 September 04 11:07:12 »
If you use TheSkyX there is an Angle and Seperation tool under Tools. Works like a charm.

Offline CraigNZ

  • PixInsight Enthusiast
  • **
  • Posts: 94
Re: Astrometric distance between two stars
« Reply #2 on: 2015 September 04 20:00:08 »
I do use SkyX for camera and drive control, but didn't see any way to apply the distance and angle tool to an image or series of images.  The tool seems to work only on the sky displayed or an Image Link image.

What I have done so far is use DynamicPSF and selected the two stars on a series of 10 images.  I then exported this to a csv file and loaded into Excel.  In Excel I then calculated the distance between the two stars on each image and graphed the results.

The precision is very good, down somewhere to 0.01 pixel or so .. haven't quantified this yet.  But if you run a trend line through the data you see a very distinct slope showing the effect of differential refraction.  I will verify this by correcting the position of each star for refraction effects and see if the slope is removed.

The equation for the trend line is: y  = 0.0019x + 68.537 where:
    y - separation in pixels
    x - image #

The image # represents the time because the images were taken as a software controlled series, each image was 45 seconds exposure + 10 seconds delay + a couple of seconds download time.  Or about 1 minute for each image.

You can see the slope is very small resulting in a drift of approx 0.03 pixel over the 10 images (about 10 minutes).

A lot more work is needed in confirming all of this but it does show the level of precision I am trying to achieve.  PI is a far better tool for doing this than SkyX.