AstroArt5 and Flats

[ollypenrice]

When our flats were not working they inverted everything, corners and dust bunnies becoming brighter rather than darker.

I think all this has nothing to do with darks-for-lights. As I think Harry is saying, this is all to do with your darks-for-flats (where everyone agrees that a master bias will do?) What I suspect is that a tiny alteration of the master bias used to calibrate the flats is capable of having a huge effect (but only on reflector images?) Gordon Bennett, put me out of this misery.

SInce my recent success followed the same process as my many previous failures I'm inclined to suspect that some small chance variation this time has done the trick. If Harry says that adding a miserly 50 ADU to the bias determines how the flats work then I can guess that my recent bias might have been just sufficiently 'different' to work.

Still guessing, though. I think Yves is very busy at the moment but he has our new data for analysis.

Olly

[harry page]

Hi

Or adding 50 adu to the bias that you are going to use as a master dark and even adding 10 made a differance

Harry

[ollypenrice]

Hi

Or adding 50 adu to the bias that you are going to use as a master dark and even adding 10 made a differance

Harry

In our case I don't think this matters. I have tried stacking using a classic dark for lights and under your evil* influence have also tried a bad pixel map and a master bias as a dark for lights. This had no effect on the flats not working nor did it make any difference to the new set of data in which the flats did work. Surely it has to be the bias for calibrating the flats that decides whether they work or not?

Olly

* but it works, I admit.

[harry page]

Hi

Yes it has a bigger effect on the flats , I spent ages trying this to save some images which I did in the end , but is nice when it just works

harry

[Jessun]

Or adding 50 adu to the bias that you are going to use as a master dark

Is it a right move to use the same bias a F.Dark as well as Dark? (I use the M25C so proper darks seem a waste of time)

[harry page]

Hi

Yes thats what I do most of the time , use a master bias for everthing

Harry

[Jessun]

Thanks Harry!

[samtheeagle]

Woo hoo! I finally got an image to process correctly

The vignetting seems to be correctly removed, and the dust bunnies are gone! I didn't supply any kind of dark frame for the flats, and I "borrowed" some old darks from a previous session. I suppose I should reprocess without those darks, to verify that it is supplying the master bias as the flat dark that is causing the issue for me. Oh yeah, this is all done in a single pass with the pre-processing tool.

Edit - Scratch that! I was trying out a number of reprocesses, including different things at a time, and I did what I thought gave me a good image before again just to be sure, and it failed miserably. So a few experiments later and I realize that I did indeed use the master bias as the flat dark. BUT that only works if I supply light darks too. I've tried all kinds of combinations and this is the only one that seems to work correctly, with this set of data at least.

[harry page]

Hi

This is the normal way of doing it , Some people like me use a masterbias in the place of a dark

Harry