Master dark question

[vlaiv]

29 minutes ago, Anthonyexmouth said:

Ok, heres a new dark. 

D_2019-07-21_12-36-16_Bin1x1_240s__-10C.fit 22.31 MB · 1 download

Now that is proper dark

Odd warm pixel here and there and amp glow visible, nothing else in that image (well there is bias signal embedded there but it looks like noise in single sub).

Histogram confirms everything is proper:

[Singlin]

Do I need to do darks for every exposure legnths?

[vlaiv]

1 minute ago, Singlin said:

Do I need to do darks for every exposure legnths?

Depends on what sort of sensor do you have.

If your sensor has proper bias (and most CMOS sensors have some issues with bias signal) you can use only longest exposure length darks and do dark scaling.

Dark current has linear dependence on time (unlike temperature where there is exponential dependence), therefore you can take 10 minute dark, remove bias and divide it with 2 to calibrate 5 minute light with it. Stacking software should be able to do this automatically for you.

With CMOS sensors and their bias issues, you can't subtract bias from darks (it changes between exposure lengths for some reason, or even between power cycles of sensor if there is some sort of internal calibration) and there fore you can't scale darks because bias does not depend on exposure length and temperature in predictable way.

If you have CMOS sensor, then it is best to do a set of darks for every exposure length you use. On the other hand, depending on the scopes you use and filters - you should not have too many exposures that you use. In principle you can get away with one exposure length per scope, maybe two if you run a risk of over exposing star cores - one "proper" exposure length and one short "filler" exposure length.

Exposure length will largely depend on your conditions and capability of your mount - go with longest exposure length that you can manage to guide/track without any trailing issues.

How short with exposure length you can go depends on how high your read noise is - lower read noise means you can use shorter exposures as main exposure. To be precise you need to compare read noise to other noise sources - when it becomes dominant noise component - increase exposure length.

There are other considerations for exposure length - how much data you want to store per session (shorter subs - more of them, more data), and how likely is that you are going to have to discard a sub (wind gust, earthquake, cable snag, sudden flash of light in direction of scope - passing airplane or car headlights, etc ...). It is better to discard 1 minute sub (1 minute lost) than it is to discard 30 minute sub (that is 30 minutes of imaging time lost).

[Anthonyexmouth]

16 minutes ago, vlaiv said:

Now that is proper dark

Odd warm pixel here and there and amp glow visible, nothing else in that image (well there is bias signal embedded there but it looks like noise in single sub).

Histogram confirms everything is proper:

thanks, i'll set APT off to do 50 while i'm out this afternoon. see what the stack looks like. 

[Anthonyexmouth]

quick kinda related, are these asi cooled cameras ok on 13.8v? its whats feeding my pier from the workshop or should i run a dedicated 12v supply?

[Singlin]

I have an Atik 383L 8300 sensor.

Reading the above I think that I will play it safe and take new darks for my different exposure legnths.

[vlaiv]

4 minutes ago, Singlin said:

I have an Atik 383L 8300 sensor.

Reading the above I think that I will play it safe and take new darks for my different exposure legnths.

That is CCD, I'm fairly certain that dark scaling will work properly with that sensor, although it is better if someone using that same sensor could confirm that.

[vlaiv]

8 minutes ago, Anthonyexmouth said:

quick kinda related, are these asi cooled cameras ok on 13.8v? its whats feeding my pier from the workshop or should i run a dedicated 12v supply?

Honestly, I don't know, but I suspect it should work fine

[Anthonyexmouth]

Well here's my first image with a cooled camera. Seems re-taking the darks sorted the problem.

Just have to save up for Pixinsight now.

[mark skelton]

When you do the master darks, do you do master bias as well or will the darks do both?

[spillage]

@mark skelton Bias frames will be very very short exposures and the darks must match the same length of exposure as the light frames. If you use DSS then it will save a master in the same folders as the calibration frames.

[Anthonyexmouth]

32 minutes ago, mark skelton said:

When you do the master darks, do you do master bias as well or will the darks do both?

not done any bias frames, been told they aint needed so much with this camera. I'll start making the effort to do flats though. 

[mark skelton]

On 22/07/2019 at 20:41, Anthonyexmouth said:

not done any bias frames, been told they aint needed so much with this camera. I'll start making the effort to do flats though. 

Any recommendation  on what the exposure for flats should be

[spillage]

Flat will be dependant on the ADU level. I think I went for 18,000, but this may depend on your camera.  I see you have listed a 674 mono and I am sure that this camera may not need any darks taken so I would just try flats to take out dust bunnies and vignetting. I would maybe drop sx a quick email regarding their recommendations and flats and darks, I have always find them very helpful.

[mark skelton]

49 minutes ago, spillage said:

Flat will be dependant on the ADU level. I think I went for 18,000, but this may depend on your camera.  I see you have listed a 674 mono and I am sure that this camera may not need any darks taken so I would just try flats to take out dust bunnies and vignetting. I would maybe drop sx a quick email regarding their recommendations and flats and darks, I have always find them very helpful.

Ok I'll give them an email to see what they say. cheers

[ollypenrice]

On 21/07/2019 at 14:36, Singlin said:

I have an Atik 383L 8300 sensor.

Reading the above I think that I will play it safe and take new darks for my different exposure legnths.

Are you sure you need lots of exposure lengths? Look at the linear subs to see what is saturated.

Olly