Working on my flat....

[Peter]

Tire(d) of trying to work out how to produce a good flat soout of desperation I worked on a range of exposures and though reaching the very right edge of the histogram shown on my Astrophotography Tool setup I found the following details when using Pixinsight's Histogram Transformation to see how the flat looked . Please look at the accompanying image and though I have increased the Histogram to 60 on the magnification scale the histogram looks identical to the normal magnification in the image. The usual double peak has disappeared and I am hoping that maybe finally I have an adequate Flat? Because of the 1 second exposure then I would expect to have to start taking Dark Flats to reduce the noise?An  example of the double peak flat is available for download at:

                  https://www.dropbox.com/s/mygm322jz7teeom/Flat_0001__27C.CR2?dl=0

And a download link for the shown jpeg is:

                  https://www.dropbox.com/s/kpfvgnrpf2yyzhy/Flat_0019__28C.CR2?dl=0

[pfile]

i think this flat is too hot in blue. the values are close to saturation. these cameras are 14-bit and so a saturated pixel shows up as 0.25 (or 16383) in a 16-bit space.

are you using a CLS filter? if so, and you are doing "t-shirt" flats, it helps to use a slightly pinkish tshirt to get a nice grey flat.

rob

[Peter]

Hello Rob, I am using a Gerd Neumann flat panel which seems to dominate in the blue. The reason I uploaded this flat frame was to try to balance the red green and blue and as you have observed this method leads to saturation but also the double peak has disappeared as well. Not sure why at more normal exposures there is always a double peak graph shown on the histogram graph. The first peak is close to the second peak and is usually 40-50% peak to the larger peak. Does this suggest a double exposure on my camera as I don't understand why I should have a gap between the two suggesting no pixels generated at this point? Thanks for having a look at my flats and will back down on the exposures, this was at 1.25 seconds were normally I shoot at 1/3 second. Sometimes experimentation can lead you to a dead end but still fun.....

[msmythers]

Peter

I don't know if this might be part of what you are seeing in your histogram. Here is a flat that has a very exaggerated double peak in the histogram. We are looking at the histogram of a image that hasn't been debayered. If we look at the debayered image we now see why the double peak is showing. The red and green channels are illuminated more then the blue. Also since the red and green are so close together in the gray image the histogram shows the red and green as a combined graph.


Mike

[Peter]

Thank you Mike, yes I debayered the flat frame and saw the same result that you mention. As the flat panel of Gerd Neumann gives a strong blue cast then is it because of the high combination of red and green channels that is causing this strong blue hue? I used to be a photographic printer so used to the CYM system, well before digital. The craftmanship has disappeared but the brain still gets frazzled with all these settings.

[msmythers]

Peter

I know. I'd rather deal with a mouse and settings and know that I can 'undo'. The old days with paper and pen, calculating proper temps for chemicals, which filter combinations, what exposure times and so on. Digital is so much more forgiving and you don't smell of chemicals at the end of the day.


Mike