Why don't my flat frames work? - Forums

Topic starter

Dec 14, 2023

Hi everyone.

So i was imaging last night and i took 30 flat frames, which look good to me but when i stack the image in DSS its overcorrecting. My settings where ISO 200 180second exposure for the light frames, and same ISO WITH 1/8 exposure for the Flats?. I will attach 1 light frame, 1 flat, and the stacked image auto-streched so you guys can see what they look like. Do you have any ideas why this is happening?

Thank You.

RobDSJ

Dec 14, 2023

Hi Menelaos, I had this exact same issue. Are you taking any Dark Flats? These are needed so that the maths works in terms of calibration. I just took the flats and then repeated the process with the lens cap on to create the Dark flats (same number and exposure time). When I loaded these into DSS as Lights, Flats and Dark flats, it didn't work. I then added the Dark Flats as Bias frames. It worked perfectly! I am no expert in this area, but this is how I got round it by 'trial and error'. I hope this helps.
Cheers
Rob

Helpful Supportive

Menelaos

Topic starter

Dec 14, 2023

Hi Menelaos, I had this exact same issue. Are you taking any Dark Flats? These are needed so that the maths works in terms of calibration. I just took the flats and then repeated the process with the lens cap on to create the Dark flats (same number and exposure time). When I loaded these into DSS as Lights, Flats and Dark flats, it didn't work. I then added the Dark Flats as Bias frames. It worked perfectly! I am no expert in this area, but this is how I got round it by 'trial and error'. I hope this helps.
Cheers
Rob

I did take dark flats as well! Same number as flats same exposure and iso. I also took bias frames with the shortest exposure possible on my camera (1/4000).

Well Written

dkamen

Dec 14, 2023 · Edited: Dec 14, 2023

Try using the bias frames only. Or, as Rob says, the dark flats only but "declared" as bias.

dkamen

Dec 14, 2023

Also, you have a circle in your flat where the signal is peaking. This indicates some problem on the optical level, maybe distancing or reflection/stray light issues.

Menelaos

Topic starter

Dec 14, 2023

Also, you have a circle in your flat where the signal is peaking. This indicates some problem on the optical level, maybe distancing or reflection/stray light issues.

Can you explain a bit more on this please? I just bought a new coma corrector I assumed it was because of that. But either way shouldn’t the flats fix that?

Leonardo da Silva Garcia Leite

Dec 14, 2023

I had issues like this before. Try using longer flats frames, around 3-6s. You can use nina flat wizard.

Remember that flat is only influenced by the path of light, so you can lower your ISO to have longer exposures for your frames (besides the t-shirt method). Of course, your are going to need different darks and bias, for your differente ISO flat….

Brian Puhl

IOTD Staff

Dec 15, 2023

Take bias frames around 0.3 seconds. Try to get your flat exposure time up to around 1 second at least. No need for dark flats if you have bias. Worst case add a pedestal to your WBPP.

If that doesn't work you probably have light leak issues.

Concise

shenmesaodongxia

Dec 15, 2023

try use graxpert

Cuong Vu

Dec 15, 2023

I think your flats are too bright and it overcorrected your lights

try to reduce the brightness to make the signal sit at 2/5 of the histogram (from left to right)

Which camera are you using?

Menelaos

Topic starter

Dec 15, 2023

try use graxpert

Thats what i ended up doing actually i posted the image in my profile you can see the result there.

Menelaos

Topic starter

Dec 15, 2023

Cuong Vu:
I think your flats are too bright and it overcorrected your lights

try to reduce the brightness to make the signal sit at 2/5 of the histogram (from left to right)

Which camera are you using?

I was reading that the histogram peak should be in the centre or slightly to the right. So I’m guessing this is wrong then? It would make sense that the histogram matches the light frames to be honest is this what you’re suggesting?

Cuong Vu

Dec 15, 2023 · Edited: Dec 15, 2023

Cuong Vu:
I think your flats are too bright and it overcorrected your lights

try to reduce the brightness to make the signal sit at 2/5 of the histogram (from left to right)

Which camera are you using?

I was reading that the histogram peak should be in the centre or slightly to the right. So I’m guessing this is wrong then? It would make sense that the histogram matches the light frames to be honest is this what you’re suggesting?

No, the histogram is slightly left from the middle not always exact middle, also it depends on the camera and sky.
By the look of your images, I could say it is overcorrected, try reducing the flat's brightness and see

I had same issue with my ASI294MM PRO, reducing flat's brightness helped (Average ADU around 23-25k )

dkamen

Dec 15, 2023

Also, you have a circle in your flat where the signal is peaking. This indicates some problem on the optical level, maybe distancing or reflection/stray light issues.

Can you explain a bit more on this please? I just bought a new coma corrector I assumed it was because of that. But either way shouldn’t the flats fix that?

When you illuminate the sensor uniformly and with light that falls at exactly right angles, the expected image is a perfectly uniform and constant value. A flat gray for example. Flats correct any deviation from this. For example, you may have a dust mote meaning a part of the image will be darker than expected. Or the sensor may be a bit more sensitive in some parts which will be brighter than expected. Or your optics may have vignetting so things get darker as you move away from center.

When you have a reflection, the part of the image where the reflection falls will be brighter, of course. And this is a deviation from the perfectly uniform and constant value all across the sensor. But tjhat deviation is not caused by a different response to light that falls at exactly right angle. It is caused by the fact that some light falls at a non-right angle. In other words, it is not a non-uniform response to uniform field illumination. It is non-uniform response to non-uniform field illumination.

Essentially the reason a pixel is brighter when you have a reflection is the same reason a pixel gets brighter if you have a star at that part of the image (or gets red if you have an apple at that part of the image). And this is something flats obviously cannot correct.

Cheers,
Dimitris

Helpful Insightful Respectful Engaging