SharpCap or the camera? Weird histogram on QHY183C

[si-cho]

Hi everyone,
I am having recently quite a number of problems with my QHY183C, I am getting extreme halos, not possible to remove using flats, and even is not easy to get flats using SharpCap for the camera, and maybe is related with the anomalous histogram. This is the picture of what I get:



Trying to solve it, I went for new flats, but taking flats, there is no way to keep the white peak between 45% to 55%, there are several whites peaks, or if I try to get the ok, on the taking flats option, all the spectra moves to the right, the white moves above 80%, so something is a bit estrange, but my first question is if the spectra looks weird or not, with so many peaks (like a gamma ray spectra...), maybe there is no problem with SharpCap and is the camera...I have licence for SharpCap in one of my computers.
I would appreciate any guidance,
Just in case, you can see the pictures with the pathological vignetting here:
https://www.flickr.com/photos/80831869@ ... ed-public/

[admin]

Hi,

the image showing the problem with the histogram was missing unfortunately, but the final image that you shared on Flickr looks a lot to me like you are having problems with dew or frost forming on the sensor or window. Try reducing or turning off the cooling to see if the problem goes away – if it does then you can start thinking about a better solution that allows you to still use the cooling.

Cheers, Robin

[si-cho]

Thanks a lot for the answer, here is the histogram, is important your opinion about it,
https://www.flickr.com/photos/80831869@N00/50810299223/
I hope you can see it, because I did turn the cooling off and I got the same result, unfortunately...

[admin]

Hi,

I don't think there's anything particularly odd with histogram – remember that the vertical scale is logarithmic, so the height of the main peak is about 400,000 pixels, the heights of the smaller peaks are only a few hundred pixels each, which is not enough to affect the overall image significantly.

If you and tick the 'logarithmic' box at the top right of the histogram then you will see the pixel counts on a linear vertical scale, which might make the situation clearer in this case.

Cheers, Robin

[si-cho]

Thanks a lot Robin, that brings my breath back, I will continue doing some other test, experimenting with the cooling, hoping I can take a decent set of flats, but still I don't know if I go for the rule of 45% to 55% or to move until I get ok in the window for taking flats under histogram.
Thanks again,
Claudio

[admin]

Hi,

SharpCap's suggestions for flat frame exposure are based on the idea of keeping the 0-10% and 90-100% regions of the histogram and used. Anything over 90% and you are getting into the range where the sensor might no longer be responding linearly, so it's best to avoid that. If you're getting areas down below 10% then those areas are going to be experiencing an incredibly strong boost when the flat correction is applied. Any small errors in the measurement of the flat in those regions could well be magnified by the amount boost applied.

If you have to choose between the two, I'd suggest allowing pixels to be below 10% rather than allowing them to be over 90%. Any areas that are at 10% or lower brightness when the main part of the image is in the middle of the histogram are probably so badly vignetting as to be a lost cause anyway.

Cheers, Robin

[si-cho]

Thanks a lot Robin, I will do a few test trying your suggestion. It is now a bit more clear to me the topic.
Cheers
Claudio