Number of subs vs longer exposure

[R26 oldtimer]

Last week I was in a fairly dark location with my ed80 and unmodified 600d. Having an intervalometer now, and tried to up my game from the 30sec subs to 90 or 120.

Shooting some galaxies I found that the mount (azeq5) coped just fine unguided with no apparent trails. The histogram was now at about 1/4 instead of the " just cleared the left side" 30sec exposures.

All seemed great, but I didn't take into account the strong wind. When I checked the subs half of them showed wind trails.

Since I was imaging for a good three hours and expecting  at least 2,5 hours of decent data, I ended up with just 50mins.

Which makes me think.... Since my total exposure time was 3 hours should I had better gone playing it safe with three times as many subs at 30secs with lower drop rate, say 10-15%?

 

 

[DaveS]

Since you're not guiding them I'd be inclined to keep the subs short. The solution of course is to guide, but even there strong wind gusts can ruin subs.

In general though, my thinking is fewer but longer subs give deeper images.

[alan potts]

Even though I have a walled observatory I was amazed what a little wind did a couple of weeks back with only the top part of the scope showing above the walls, not serious but you could see it in the subs.

Alan

[Knight of Clear Skies]

Shame about the wind, but hope you got some decent data from the dark site.

11 hours ago, R26 oldtimer said:

Which makes me think.... Since my total exposure time was 3 hours should I had better gone playing it safe with three times as many subs at 30secs with lower drop rate, say 10-15%?

Probably, yes, in this particular case. Longer subs reduce read noise in the stack, but I'd be surprised if that could make up for losing 2/3rd of your data.

[david_taurus83]

As counter intuitive as it sounds, you do need to take longer subs for darker skies. This is because you need to saturate the sensor with more light to overcome the cameras read noise.

So if your imaging from light pollution at home there comes a point where the light pollution starts to overcome the faint detail your trying to pick up. Sure, your swamping the read noise but this is likely due to your LP.  Longer subs at home start to be detrimental.

So at your dark site with no LP, theres less photons hitting the sensor. The challenge then is to expose for long enough so the faint details overcome the sensor noise. But you will pick up much more detail and start to separate it from the background sky.

[Knight of Clear Skies]

49 minutes ago, david_taurus83 said:

As counter intuitive as it sounds, you do need to take longer subs for darker skies.

Your post is correct and well explained, but just to be clear increasing the sub length is likely to be counterproductive if it leads to a high proportion of lost subs.

It's difficult to know what the optimal sub length is, I believe it can only be calculated from the read noise of the camera, the local sky brightness and speed of the optics. But longer subs run into diminishing returns fairly quickly. If I'm remembering correctly from a dark site and a camera with moderate read noise such as a 600D the improvement in SNR becomes quite small beyond about the 2-4 minute mark. (Treat what I'm saying with a little caution here, I'm a bit sketchy on the subject). Long subs are most important when using a high read-noise CCD to image the faintest objects.

From my own experience, it's possible to pick up a fair bit of IFN using a DSLR and unguided 1 minute subs at f2.8. (This would be equivalent to a 4-minute sub at f5.6.)

[DaveS]

This was posted in the main Imaging forum

It's a bit detailed and wanders into *shock horror* Maths but should answer your questions regarding sub length, noise. and sky darkness (Or lack thereof).