Help requested with taking and applying flat frames
Hi!
I'm relatively new to telescopic astrophotography, having purchased a 200mm f4 telescope and kit secondhand a couple of months ago. I'm having quite a bit of success and am really enjoying myself, but the learning curve is very steep and I'm having a few problems which I haven't been able to work out myself yet. First of all I'd like to ask about flat frames.
The attached file shows the problem I'm currently having with my flat frames obviously not doing what they're supposed to. In the past I've done without, or I've shot subjects bright enough that the signal overwhelmed the faint variations in my field. The Pencil Nebula however doesn't give me that luxury and I have to work this out.
The image shows the results of applying bias, darks and flats to 164 minutes of integrated light frames, then applying a background neutralisation and screen stretch in Pixinsight. Obviously the flats aren't working as intended, and seem to be misaligned.
To capture my flats I've been pointing the scope at the ceiling, laying a sheet of white paper over the aperture, positioning and focusing the camera as it was during the image session, and taking shots with a camera flash bouncing off the white ceiling. I've attached a jpeg of my master flat, which was made from 128 subs.
Can anyone tell from looking, where I might be going wrong? I can fill in any missing details if you ask me questions.
To my knowledge, it is best not to move the camera and take flats exactly at the same orientation and focus as it was for imaging session. Also, if signal in flats is too weak, software won't correct properly for uneven illuminations due to optical imperfections. I have found through experimentations that I need to reach about 2/3 of the maximum values for pixel wells in the flats for them to work properly. I hope it helps a bit
That flat you have shown looks very dark it should be a lot brighter.
Because the lights are divided by the flats, if the flat is too low in value, it will cause over correcting and you end up with bright corners and dark centres.
As Suavi said get the histogram of the flats about 2/3rds of maximum value.
I'd second the advice here.. try getting a brighter exposure with your current flat. Then try a twilight sky flat (pointing overhead just before stars start becoming visible in the flat).
I have an EL panel at home you can try also, but Peter's flat boxes are even better - I got one with my FS-102 and use it regularly with camera lenses etc.
Thanks for all the suggestions everyone. I'll re-take my flats with a higher exposure. I have a few questions about shooting flats:
- Do I need to calibrate the flats with bias and/or darks?
- Do the flat calibration frames have to match the flats in exposure settings and temperature?
- Do the flats have to match the lights with the temperature and exposure settings?
- If I make sure the focus and camera/tube orientation match the lights, can I take the flats at a different time? And if I can always use the same focus and orientation can I always use the same flats? (I'm aware I have to take new flats if I clean the sensor/mirror and if I change collimation)
- I'm thinking of using a flash in a soft box (I have studio lighting equipment) and white reflex paper over the tube as my light source. Is that sufficient?
- If this won't do, and I absolutely must shoot flats at the end of a session, can I order one of Peter's light boxes without the box? It's pretty big and fragile and I'd much rather something flat and more portable I could put on the end of the tube instead.
Unfortunately, shooting all night and staying up till dawn to shoot sky flats isn't an option for me.
In theory you should subtract a bias or a flat dark to get them calibrated correctly. However you can subtract an offset using pixel maths to get the same result.
Flat darks must be the same exposure length, in my case I always use 5sec flats and flat darks. Bias subs are at 0 sec exposure time.
Flats do not need to be taken at the same temperature as the lights and definitely not the same exposure time.
Yes you can take the flats at any time as long as nothing has changed ( focus, orientation etc). You can use these flats over and over as long as nothing has changed.
The master flat I am using is now a month old and as long as I keep the optics clean and orientation the same it will always work.
I wouldn't use a flash light to create flats, better off pointing the scope at a white wall or white cardboard with even illumination.
I use Astroart V5 software and I can create a Pseudo Flat from the stacked lights and then divide the pseudo flat into the stacked lights, this works OK as well.
So this is another option to consider, in fact after stretching the stack, if a gradient appears I can create another Pseudo flat to remove the gradient.
My way:
Lay the telescope horizontal.
Put a white T shirt over the front of it- taped tightly & flat.
Use even thickness layers of cloth.
Block the rear of a Newt. with black cloth.
Turn the lights out.
Put your large computer monitor on the floor so that the middle of the monitor is at the center line of the scope.
Place it about 1 meter away
Make sure you make the monitor screen all white using a photoshop
pic made all white full screen.
Take flats with the camera at the same temperature as your light frames.
Make sure you go at least to 1/3 rd full well depth in the corners of the frame.
Adjust exposure time accordingly.
5 flats should be enough for each filter.