Pixinsight Master Flat Generation Problem

[ashworthacca]

Hi all, 

I'm trying to create a Master Flat in Pixinsight and have a problem at the ImageCalibration stage.

I am using old modifed Canon 1000d data, have created a Master dark and Master Bias but during the ImageCalibration routine for the Master Flat I get a "Warning: No correlation between master dark and target frames (channel 0)." message and therefore I believe that the Master dark is being ignored?

Any advice how to remedy this?

For the Master Bias I have the calibrate box ticked. For Master Dark I have the calibrate and optimize boxes ticked.

Thanks.

[IanL]

No the dark won't be ignored, it just won't be scaled as the dark scaling routine couldn't correlate the noise between the flat and the dark.  You have a number of things to consider:

- For a dark to be successfully scaled it must be bias subtracted (which I think you are doing, as you have loaded a master bias and ticked calibrate on the master dark).

- You could try making a master dark that is a better match for the flat (same ISO, same exposure length, same sensor temperature).  If there is a big difference in exposure length or temperature it might not be possible to correlate the dark and the flat.  Making good darks is pretty hard to do - the EXIF temperature does not correlate particularly well to actual sensor temperature, plus the on-camera processing generally makes higher temperature and/or longer exposure darks less bright than lower temp/shorter ones, which is the opposite of what should happen.

- You could turn off dark scaling, i.e. the optimize option (or just ignore the error, same difference) if you are confident that the dark and the flat are a good match, but the error is telling you they aren't.

- You could just subtract the master bias and not bother with the dark.  That would probably work reasonably well as the flats should be pretty short exposures and there won't be much dark current.

No easy answers on this one, Canon DSLR darks are a minefield of problems.  In theory the PI dark scaling should help overcome some of the issues assuming you can get a correlation, but if you can't you may well be doing more harm than good to the image.  If you have similar problems on your light frames, you may want to try fixing them with the CosmeticCorrection process in lieu of using a dark.

[ashworthacca]

Thanks IanL, I'll give your suggestions a try and report back.

I've calibrated and stacked in PI a number of times but this is the first time I've seen this message.