Dust doughnuts even when using flats, why?

[bendiddley]

Took this shot of M101 a few nights ago, which I am pleased with. Was the first time I have taken flats so glad to see the back of the vignetting. However I am still getting dust doughnuts. I checked my flats files and you can see the doughnuts on them so not sure why deep sky stacker hasn't taken them out. Is there a setting I have missed or any other trick. It's got rid of the vignetting nicely which I am glad to see the back of, but not sure why I've still got the doughnuts, hope someone can help.

[carastro]

It looks like the flats have generally worked, but it is possible you have a very bad bit of muck on the sensor too much for the flats to cope with.  I had a similar situation once and once that bad bit was removed the flats dealt with the other specks of dust OK.

Carole 

[Davey-T]

Worth stretching a light frame and a flat to check there is a corresponding spot on both.

Dave

[bobro]

Looking at the two doughnuts adjacent to the galaxy, one is dark and one is light. Both are the same size, making me think the light doughnut should correct the dark doughnut. As the doughnuts are quite large the dust is more likely to be some distance away from the sensor, possibly on a coma corrector? So something, such as a coma corrector or the camera position possibly has moved between taking the image and flats. This movement wouldn't necessarily affect vignetting correction.

Just a theory......hope it helps.

 

 

[Thalestris24]

Another possibility is that your flats are too dark for your lights. I get that sometimes as my lights are often exposed to the right and correctly exposed flats don't seem to work properly.

Louise

[Juicy6]

If calculator at http://www.ccdware.com/resources/dust.cfm is correct the dust is 3.8 mm from sensor surface (pixel size 5.7 µm - f 5).

Christer, Sweden

[bendiddley]

Thanks Guys. I use an astronomik clip filter in the camera and also a coma corrector. I've just had a look at the clip filter and there are some dust specs on it, the coma corrector could probably also do with a clean too so hopefully that will sort the problem. Nice easy fix if it is! Cheers.

[bobro]

Following the post from @Juicy6 above, I did a similar calculation based on a spec of dust at a distance of 1/2 the focal length from the sensor would cast a shadow of the same size as the scope's aperture and the shadow size varying linearly with distance from the sensor. The result was the same - about 3.8mm. Looks like the dust will be on the sensor glass.

It would be interesting to see jpegs of the master flat and a single RAW sub converted to jpeg to see how the dust looks.