More testing of the Flat-Man XL with artificial star reference

We had a professional astronomer find some fault with our last photometric test procedure, where we used a reference star relatively close to the target star. He suggested that we use an artificially generated reference star as our photometric reference.

Maxim D/L has a plugin that was written by Ajai Seghal that will embed a properly formed artificial star in the top left corner of the image and we used this as the photometric reference. We then exposed a series of 8 sets of 9 images of an unsaturated star near the zenith so that the star appeared in a different place along a diagonal line across the image from top left to bottom right corner in each of the sets.

We then did a photometric analysis of the star in each frame of each set after the frames were calibrated with a master dark and a bias corrected flat made with a Flat-Man XL. The telescope was our customer Mark Manner’s 16″ RCOS with an STL-11000 and a V photometric filter. Mark also did all the data acquisition.

Here’s the results of the test:

Artificial star photometry results of Flat-Man XL test

Note that the (mean+median)/2 overall variation in flux is around 3%, which for photometers, means a photometric accuracy of 0.03 magnitudes.

In practical terms, this means that our Flat-Man XL does a great job of correcting your images for dust donuts, vignetting, and other imperfections. For those of you not interested in photometry, but who want pretty pictures to turn out good after brutal non-linear stretching, I stretched the heck out of one of the XL-flatted images and could see no sign of vignetting in the XL calibrated image.

We did another test with a CDK 24″ and an ST16803 with a clear filter and a Flat-Man XL-30 to compare the electroluminescent panel flats with twilight flats.
Here are the results:

Comparison of twilight flats to Flat-Man XL flats

Here you can see that the Flat-Man XL outperforms the twilight flats at each position of the star on the diagonal. I’m not sure why neither flat corrected the corner position well. It may be because of some passing high clouds.

I think this is a conclusive test that shows that EL panel generated flats are just as good or better than twilight flats and provide a reliable, repeatable calibration source.

Leave a Reply




You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>