Gradient in dark frames

I have this same camera and have never seen anything like this… except when I had a light leak in my OAG.

Maybe try taking darks..in the dark and see if this still happens?

This second post was sent in error. On my phone, and I can’t delete it.

Lukas Bauer · Mar 5, 2026, 11:22 AM

But I still have a question: I understand that the darks can be taken at anytime because of the cooling of the sensor. But how about dark flats?

Anything with “dark” in the name can be taken whenever since no light is supposed to be getting to the sensor. You just want to make sure the gain, offset, temperature, and exposure length match. With that said, I use bias just fine with my IMX571 sensor so I don’t use dark flats.

Lukas Bauer · Mar 5, 2026, 11:22 AM

Thanks for all the replys!

I will try to take the dark frames in a completely dark room when I carry the telescope indoor again once the clouds come back (it should be clear for a few nights in a row now - hopefully). I’ll let you know if this solved my problem.

But I still have a question: I understand that the darks can be taken at anytime because of the cooling of the sensor. But how about dark flats? Shouldn't they be taken right after taking the flat frames? Or is it just important that the exposure time and sensor temperature is the same?

Furthermore, I’d like to ask if the flats for one night need to be exactly the same brightness in every flat. With sky flats the sky gets brighter while I'm taking the flat frames (over a period of a few minutes). This change is barely noticeable by eye but the value of the midtones in the histogram changes marginally.

While darks do provide some benefits (very negligible tho) , dark flats do not. At least not with the camera sensor you’re using - the adu difference between dark-calibrated vs bias-calibrated is at most a tiny fraction of a percentage, barely measurable, let alone noticable. Unless you’re heating your sensor to like 50C, neither temperature nor exposure time are too relevant for calibrating flats.

As to the flat exposure length changing, Pixinsight will normalize them multiplicatively iirc, so it’s not an issue whatsoever.

Lukas Bauer · Mar 5, 2026, 08:44 AM

Hi everyone!

I recently bought the Player One Poseidon C Pro camera and after a very long period of cloudy nights it finally cleared and I got to try it. When taking the lights everything was normal. When the sun rose again the pictures got very saturated (half the pixels were just violet and a the other half were just green). I stopped capturing and went on to take the flats and flat darks. I have always used the sky to take my flat frames and everything went smoothly (with my old DSLR). The flats are also great to shoot with the new Camera.
But when it got to taking the flat darks things got strange. I put on the telescope cover and covered the whole telescope with the camera and everything with a blanket. That way I should eliminate any possible light rays getting into the image train. But the darks don't look like the usual darks. Instead, there is a very noticable gradient, as you can see in the attached picture.

📷 Dark_Poseidon.jpg

Afterwards I also tried taking bias frames like usual (same setup with blanket). As you can see in the second picture the bias frame also doesn't look like a bias frame should look like.

📷 Bias_Poseidon.jpg

For both the dark and bias frames I reduced the gain and offset of the camera to 0 (when shooting lights I use gain: 125 and offset: 25).

Does anyone have a idea what could cause the very strang looking dark frames and bias frames? Has anyone else had this issue before and how did you fix it?

Thank you very much in advance!

Lukas

Lukas, I think your main problem was a light leak when taking your darks.

I know that when you were shooting with a DSLR you needed to take darks during the session because of the lack of temp. control. Now that you have a cooled camera, you don’t need to do that anymore. The best approach is to pull the camera from your rig, cap it to totally eliminate any possible light leaks and methodically build a library of dark frames. Pick a temp that you’ll always use, -10 is a common setting. Take twenty frames for every exposure time you are likely to use and stick em in a folder. Going forward there will be no need to take darks at the scope at all. Just pull as needed from the library.

On flats, these are going to be stacked in your processing program to make a master so as long as they don’t vary by much more than 15%, you’ll be fine. Sky flats are a good idea and super easy. If you use NINA you can use the flat wizard whish will compensate if the sky brightness changes to much. You can even automate the taking of sky flats as part of your imaging sequence. I always have the camera cooled to my standard shooting temp when taking flats but I’m not sure that’s necessary.

True bias frames are done with the shortest possible exposure with the camera capped. The camera doesn’t need to be cooled. You can do them when you make your dark library.

There’s a lot of confusing talk on this thread about dark, dark flats, flat darks with tarter sauce. A lot of it comes from the needs of different processing programs and sky conditions.

I would suggest starting with the basic calibration frames your program needs to start with and press on with imaging.

And some software like Siril really require them…

The OP should keep in mind that while using a flat library can work, it’s not recommended. If anything changes in your full optical train such as a dust mote on the sensor window you are pretty much SOL if you didn’t take flats during the imaging session. Flats are like bread, best enjoyed fresh….:)

Lukas Bauer · Mar 5, 2026, 06:51 PM

Thanks for the help!
I really like the idea of putting the camera even without a telescope in a dark room to build a flat library, I will try that. Although some might say that it isn’t necessary to take darks, I think that it isn’t that much work to build such a dark library and in my opinion it is better to have it even if not really needed. I haven’t read anywhere that the addition of dark frames made the picture look worse, so why not add them…

Exactly, as long as you take enough that you aren’t -adding- noise from the darks to each frame. I take about 50 for each exposure time. I do just use biases for my flats since the exposure time varies and, at 50% histogram and short exposures, any dark noise is absolutely 100% swamped.

I’ve tested just using biases vs using darks for light exposures, and the LRGB, there really wasn’t any difference. But for Narrowband, darks did help some since the signal is low and the exposures longer. But I just spend the little non-sky time to build the library and use them now going on over a year.

I use a flat panel for flats and they just always turn out great. Use flat wizard in Nina and spend 5 minutes in the morning (I use a rotator so need to take flats each session). 10 flats per filter seems to work great.

Tony Gondola · Mar 6, 2026, 07:43 PM

SonnyE · Mar 6, 2026, 04:46 PM

I’m not the only one who uses a library.

It works for me, and my friends who use this method.

So don't knock it till you've tried it. 😉

Sorry Sonny, As we are giving advice to a newbe I have to stick to my guns on this one. It might work for you but it’s not a good idea. Sure, people get away with it but it’s not good general practice. Like a lot of things, it’s good until it isn’t….:)

Whatever suits you, Tony.

We’ll have to agree to disagree.