How do I work out what exposure I'm getting?

[Stewartfrom the Cotswolds]

I am just getting started with image processing and need a little advice. I have an ASI183mc pro (one shot colour, cooled). Capturing with SGPro and processing with Pix in Sight. What I'm not sure about is how to work out either in Pix or SGPro what level of exposure I am getting and there for how long my subs need to be. How do I check if I'm over exposing stars or galaxy centers, and how much exposure is a good guide to work from?
Regards
Stewart

[symmetal]

Hi Stewart,

You can check for over exposure in SGP by looking at the unstretched preview image after it's been downloaded. Any over exposed (clipped) stars or galaxy centres will show up as white dots or blobs. You can also hover the mouse over the preview image in any unstretched/stretched preview image to get a 16 bit ADU readout of the pixel under the mouse. Any ADU reading over 65500 or so is clipped. It should be 65520 for a 12 bit camera though my ASI1600 clips at 65504. Bright stars will always get clipped if you're aiming to expose for galaxies or other DSOs so don't worry too much about them.

You might assume it's optimum to expose long enough so your target is bright enough but not clipped. But it depends on how much light pollution you have which affects your sky background level. Once you have exposed long enough so that the ADU of the sky background level has sufficiently swamped the read noise from the camera then any further exposure gains you no real benefit. It's best to take many short exposures which are exposed to the 'optimum' sky background rather than a less number of longer exposures. This applies to CMOS cameras mainly which have a low read noise. For CCD longer exposures are recommended as the read noise is higher and being 16 bit have a higher dynamic range so clipping is less of an issue.

Accepted values by many for the 'swamping' ADU is either 10 * RN squared or a more simple 20 * RN where RN is the read noise (converted from -e to ADU). The 'swamping' ADU value is affected by the gain used along with the offset. At higher offsets this is the dominant factor in determining the swamping value as it adds a constant to the sky background value. For my ASI1600 which has similar data graphs to the ASI183 at unity gain 139 and recommended offset of 21 this works out at a background ADU of 826 for 10 * RN squared and 896 for 20 * RN. I use unity gain and an offset of 56 to avoid clipping of the noise so the background ADU for that is 1326 for 10 * RN squared.

You can get the background ADU from your image by hovering the mouse over it or just looking at the median value from the image statistics panel.

I have relatively dark sky and reach the background ADU value at about 45 seconds for luminance,  around 120 seconds for RGB and 600 seconds for Ha, so these are the exposure times I generally use. For bright objects these exposures may need to be reduced to avoid clipping.

As you have a colour camera which will have a bit different response than separate LRGB you can only expose for a RGB sky background level. I'm not sure what SGP reports in image statistics for a colour camera, whether it's three sets for RG and B or a single 'average' value. Hovering the mouse over the sky background in the preview image will give you a good idea of the RGB values and whether you need to expose for longer or shorter to reach the 'optimum' swamping values.

The ASI183 has unity gain at 120 (use unity when starting out and possibly all the time ). If you give me the offset value used at that gain I can give you an optimum sky background ADU for you. The ASI183 manual doesn't mention offset values but hopefully the Ascom driver used in SGP shows you what it is and whether it can be changed.

Hope this gives you some insight and others may have different views. 

Alan

[Stewartfrom the Cotswolds]

Thanks Alan, (I think haha) a few more things for me to learn, need to get my head around noise swamping.  I will get back to you with the offset next time I have the gear out.

[symmetal]

The noise swamping I mentioned above just refers to read noise. The sky background level is the faintest detail level you can achieve so as long as the background level is significantly greater than your read noise then you're limited by your sky background and not by the camera read electronics. There are several other types of noise present too which can be reduced by stacking many images, using darks and flats etc. Here's a quick intro. Exposing longer than the read noise swamping duration will improve your shot noise SNR but you will reduce your dynamic range as more stars will be clipped and become bloated. Stacking shorter exposures is a better way to improve the shot noise while maintaining dynamic range.

The 12 bit ADC is a limiting factor with CMOS cameras. The noise intro I linked to didn't mention quantizing noise which is caused by the limited bit depth of the ADC and causes posterizing effects. Using different gain settings on CMOS cameras is a swings and roundabouts method of trying to improve noise and dynamic range, some noise types may be improved by increasing gain while other noise types may be worsened. Likewise with reduced gain. Increased gain reduces your bit depth below 12 bits, reducing gain, it stays at 12 bits but you throw away photons. Keeping the gain at unity is a good mid point at balancing the noise and dynamic range.

@Vlaiv convinced me to stay at unity gain.  A big bonus is you don't need different dark frames for different gain settings. 

Alan

[ngwillym]

You might find the tables in this:-

https://www.cloudynights.com/topic/616524-sub-exposure-tables-for-the-zwo-asi183mm-and-qhy183m-and-colour-versions/?hl=+asi +183

a useful starting point

Regards

Neil

[symmetal]

3 hours ago, ngwillym said:

You might find the tables in this:-

https://www.cloudynights.com/topic/616524-sub-exposure-tables-for-the-zwo-asi183mm-and-qhy183m-and-colour-versions/?hl=+asi +183

a useful starting point

Regards

Neil

Thanks Neil, hadn't seen these before. I looked at the similar tables produced for the ASI1600 and the given exposure lengths for broadband are almost identical to the ones I found by experiment which is encouraging.  The narrowband exposures are longer than mine but I extrapolated the ADU from the shorter exposures I took. Looks like I possibly didn't deduct the offset from the extrapolation properly. 

Alan