weird vignetting - assuming it is user error
- Thread starter joshtb86
- Start date
fredvanner
Well-known member
Twilight flats should be taken before sunrise / after sunset (so sun is not in the sky), and typically roughly overhead (I don't know if NINA has any advice about exactly the right "null spot"). Ideally your exposures should not be shorter than about 0.5s. If you do them as it gets darker you may capture some stars; you can fix this by dithering your flats then integrating with "median" selected.
fredvanner
Well-known member
Real sky flats are taken with a dark sky, using the "dither - median integrate" technique to eliminate stars.
These can be difficult - and are mainly used for precise calibration (e.g. for photometric work).
These can be difficult - and are mainly used for precise calibration (e.g. for photometric work).
Thanks I will try to do that in the future instead of what I've currently been doing
Is it possible that light leak could influence the flat frames and then cause these gradients?
Also, I have done flat frames the same using the T-shirt method in the past without this issue which makes me feel like it's either potentially from the moon or some sort of other calibration issue that I haven't yet figured out
fredvanner
Well-known member
... Oh, and of course twilight flats must be taken with a clear, cloudless sky (one reason why I've never used them ...).
cloudbait
Well-known member
I take twilight flats about 30° off the zenith opposite the setting/rising Sun. I turn off tracking, which is a very simple way to "dither" so that any residual stars calibrate out. My RGB flats are on the order of a second, my narrowband on the order of 10+ seconds. The exposure time changes very quickly as the Sun's distance below the horizon changes.
Is a flat panel just as effective?
cloudbait
Well-known member
No, but generally easier, and for most people in most cases, plenty good enough. But they can introduce problems. They can be imperfectly uniform. They can create aliasing artifacts if exposure times are short. Their radiation pattern can produce bad flats when they are located too near the end of the telescope. If you use a light panel, you should test it against good twilight flats. If you can subtract the two and see no structure (just noise) you should be able to use a panel. Otherwise, you need to either use the sky or try a different technique with the panel. For my most critical photometry, I use sky flats. But for aesthetic imaging I use a LED tracing panel, but it only works if I have it a good meter from the end of my scope.
fredvanner
Well-known member
I use a Gerd Neumann Aurora 220mm electroluminescent flat pannel. Placed in contact with the rim of my OTA it works fine for all my scopes (it just fits my Edge8HD, which is my largest aperture). The advantage of this panel is its very high uniformity; I have tested by comparing the image with the pannel at different angles, and I can see no difference at all. One disadvantage is a very weak red output (barely 10% of the G and B outputs) - but it is good enough for flat correction.
LED screens (e.g. from a laptop or tablet) can have a few problems. They are made of coloured dots, so it really helps to have some sort of diffuser (if only a sheet of paper) in front of them. LED light boxes are better, but they are often illuminated by edge-mounted LEDs using plastic "light guides" to spread the light across the pannel. This looks uniform to the eye, but is often less than uniform if you test it. The use of PWM switching to control the brightness means that it is safer to use these pannels at longer exposures (not less than 1s), using sheets of white paper to attenuate the light if needed.
LED screens (e.g. from a laptop or tablet) can have a few problems. They are made of coloured dots, so it really helps to have some sort of diffuser (if only a sheet of paper) in front of them. LED light boxes are better, but they are often illuminated by edge-mounted LEDs using plastic "light guides" to spread the light across the pannel. This looks uniform to the eye, but is often less than uniform if you test it. The use of PWM switching to control the brightness means that it is safer to use these pannels at longer exposures (not less than 1s), using sheets of white paper to attenuate the light if needed.
Last edited: