Tilt of filters causes halos

gofraro

After the latest update (which ran flawlessly) , I did a functional test yesterday evening.
To start with the inialization, I went to Arcturus. Everything went fine, until I saw the first image evolving.
A huge halo, looking like a view through a Newtonian.
In tried another test on the same object with the LP-Filter on, and the reflections got worse.
As it can be seen in the comparison there is obviously a tilt with the filters, I cannot explain.
Compared to an image from last year of the same object, a huge difference.
Any thoughts and ideas about it ?
(I'm not a newbie and using different kind of telescopes since years).

Latest Update of App/Firmware/SkyAtlas; No fancy gadgets; Exp 10s; unprocessed; tilt checked with Siril

satch984

Its not software, this is hardware
Like something in the optical train was not aligned

gofraro

satch984
But, one thing has changed in the software. The application of flat calibration frames.
As described in the App:
"If the built-in flat-field file cannot meet the calibration requirements, you can control Seestar lens at the uniform light source and click the Shoot button, and the system will automatically take and save the flat-field file."

Where does the built-in flat-field file come from initially ? If a default file is used after the software update, this could explain it, because every optical train is very different.
And as long as there is no detailed description available, I don't want to use it. The only thing I can find and see are enthusiastic videos about flats and 60s exposures.

Support@Seestar

gofraro
The halo you’re seeing is caused by internal reflections, which are common when imaging very bright stars. We suggest avoiding overly bright targets to reduce this effect.

As for the tilt you mentioned, it may be due to a slight unevenness of the sensor plane. To better assess this, could you please provide a single raw frame for analysis?

Please note that software updates do not affect the physical optical path or alignment. This issue is not caused by the update. It's possible that over time, slight impacts or loosening of components during use may have caused some shift in the internal alignment.

gofraro

Support@Seestar
"This issue is not caused by the update. It's possible that over time, slight impacts or loosening of components during use may have caused some shift in the internal alignment."

What about the introduction of flat frames and the use of factory provided frames with this update ?
Dont't tell me, that you ever have used flat frames. If factory provided flatframes don't match the actual configuration, they can make things worse. I could live without flat frames since I have the S50. Why should I use it now ?
I not make it optional. But stop, then you have to consider it in your stacking code, which obviously didn't happen. Ask your software guys, why ?
Unfortunately, I deleted the files and cannot provide the single frame. I will try to shoot another one, when the weather conditions allows it.

Support@Seestar

gofraro No, the update did not force the use of flat frames, and they are not applied unless needed. The stacking algorithm does consider whether to use them based on actual conditions.

gofraro

Support@Seestar
it seems it is an endless discussion or im talking in circles.
Fact is (and please prove me wrong)

before this update, no factory flats existed and where used
the update introduced factory provided flats with the possibility to "Flat Shoot" customer flats
once a user use the "Flat Shoot" function, there's currently and actually no way back (no option to use no-frames or factory provided frames).
the S50 (I can't speak for the S30) has no vignetting due to the optics and sensor size and I have never experienced it
I don't know where dust on sensor or filter should come from, since it is a closed system. Dust on the front optics (as long as it isn't bird poop) won't produce dust mots.
the only reason for flats is therefore to deal with Pixel Response Non-Uniformity (PRNU)
the user has no control, how customer flats are produced. The app gives him only a go or nogo in the Flat-Shoot function. Whether they work, he can only see it later in the captured images.
there is currently no possibility to use the DeepSkyStack with or without flats to restack a session with different flats.
flats can contribute to noisy images and can make worse things worse, if they are not properly created. Using the T-Shirt method is oldschool, using a smartphone display can introduced interference patterns. Only a properly dimmed light panel can produce the flats that are needed to produce Lights with a certain quality.
using flats or no-flats requires a different stacking algorithm, currently not implemented in the firmware.
as Darks are sensor temperature related, Flats are influenced highly by changes in the optical path (but also temperature, therefore you have to use darkflats or bias frames when creating flats, do you ?). A slight tilt of the sensor or LP-filter when changing the focus (don't tell me that there are no tolerances), requires new flats. Otherwise you will see what I posted previously (images with halos and reflections).

I'm not a newbie. I have different scopes and cameras (one of them is a IMX294C and probably you know, what it means regarding flats) and I'm used to create calibration frames (primarily with Sharpcap Pro).