Calibration data

In addition to my observations on pier 3, I also now booked some time on pier 1 and had a look at those files. Sorry if I'm too crtitical btw;

- The sky flats actually have stars in them, so that I would say is an issue.

The darks and bias for this one look good.

Hi Vincent, 

Bit late replying to this, but from those calibration files which I've just been using, I agree with you. Both the 300 and 600 second dark frames suffer from., as you say, some sort of light leak. They give the appearance of being flat frames, and with exposure times of 600 seconds it would have to be a very minor leak. Strange that the 900 seconds and others are fine. I used the 900 secs. The flat frames were giving real problems in the integrated images and with my limited knowledge of pixinsight I could not find a satisfactory way of dealing with the vignetting in post processing. Resorted to adjusting the master flats and recalibrating, which while far from perfect, gave a better result. 

Thanks for the confirmation, often I think it's what I'm doing or failing to do is the cause.  
Cheers, 
Ray 

With regards to Pier 3 the flats do need attention and we will be  getting a fresh set in the coming days. 

Over exposure On Pier 3 flats
(This will also apply to Piers 3, 10 and 13)
You are quite correct, the ideal signal on a regular flat frame should be between 30-50%  the total saturation of the sensor, this is true and works on most scopes.  However it is a very different story where the 3inch  Wynne Corrector is concerned as it suffers quite heavily of poor illumination in the corners.  The light fall-off  is very noticeable on large 50mm sensors and so to keep the signal strength up in the corners we do have to push exposure way over these typical values. That said  the histogram must not show any clipping what so ever so if you do have clipping do use these frames. You can get away with a very high  ADU value and generally you get a better correction from a flat with a more generous amount of signal where poor signal can do more harm when calibrating data sets.

Stars in Sky flats
As much as astronomers like to see stars in an image they do tend to get in the way when it comes down to flat frames.  Unfortunately when doing Sky Flats there is always a danger of seeing them in the flat fields and in most cases it is unavoidable. 
We we do dither the exposures, so you should not see stars in the same place on any of the flat frames.  When calibrating your data sets you should set your flat combine method to 'median' stacking this will reduce the the effect of stars in your sky flats.
We are staring to roll out flat panels on each pier so hopefully stars in flats will not be an issue in the future.    

Calibration frames are always a great topic of discussion, I hope this helps answer these queries.

Regards

Peter

Thank you Peter for this nice summary!

Regarding the sky-flats, dithering can help there but then I would say you need a bit more frames to get proper rejection for the rejection algorithms, usually this amounts to 15-20 frames. I'll try again with some better rejection parameters.

Regarding the light fall-off; good point indeed, however I would then say to stay at max 70% or so, I'd rather have some small issues in the corners which I can crop, then in the overall image. In this case I got a bit over over-correction in the corners, no change in the overall image (as it's clean already) and an added white stripe.

The light on the sensor for the darks is also an issue as when used, it gives massive over correction in the integrated image. I really couldn't use it.


Thanks a lot for the trouble! Vincent

edit: just found out the new data is almost ready, yay!

I couldn't find any sign of them just now Vincent , which probably means that they're being downloaded as I type this. Usually works that way. Phone up somebody about a payment, and yes it's in the post! 
Thanks for chasing this up. 
Cheers Ray 

The latest claibration files for Pier 3 flats are now up and available guys