Failing at flats!

[SiD the Turtle]

Hi all,

I've been trying to get decent flats and I think I'm failing. Basically my master flat doesn't really clean up the vignetting and dust motes. I've tried both a flat light source (laptop screen showing white, balanced over the lens) and the t-shirt method. The histogram is often off to the right, which doesn't seem to match most of the tutorials online. Setup:

• Canon 600D unmodified.
• William Optics ZS73.
• Controlled through EKOS/Astroberry.
• Taking FITS files.
• DSS for stacking, PixInsight for post-processing.

Here's the raw, stacked pictures, 23 5 minute subs at ISO800, with just bias frames applied and auto-stretched:

You can see pretty clear bright vignettes top and bottom, and a large dust mote above Andromeda here.

Here's the same again with the flats applied, just auto-stretched.:

So the vignetting is applied, but there's still a large gradient from left to right and the dust mote is still there.

And here's the flat, stretched, and then using Pixinsight's contour map script:

This contour map seems to pick up the vignetting, but not the dust motes.

Any ideas where to start?

[Budgie1]

How are you taking the flats (camera settings)?

Normally I put two layers of a t-shirt over the end of the scope and point it at a blank MS Word document on a PC screen. All I change on the camera (EOS 1300D) is the from "M" or Bulb to "Av" and let the camera choose the exposure. I can normally see any dust bunnies on my flats without stretching, so maybe the exposure is too short.

The last three sets of flats I took at ISO 800 had exposures of 1/8, 1/10 & 1/13.

If you want to try dark flats as well, then note the exposure time for the flats, put the cover over the scope, turn the camera setting button to "Tv", set the same exposure as the flats and fire off the dark flats.