Am I over doing it?

[MadMaxwellSmart]

I wanted to ask if there is a possibility that I'm over doing my images for the equipment I have, Canon T2i, celestron c6a. I took an images of M31 with my camera piggybacked on my scope and the image was 5 lights at 120 sec each! took a few darks and that was all. Processed in Star Tools and it came out the best I've done, really was a nice image...

M31

M31,M32 andM110 - Copy.jpg (344.93 KiB) Viewed 4201 times

. I done the same with the Triangulum and got the same results. After purchasing my Kapaxen power adapter for my camera I went up on my exposure times, This is where I think my trouble began...I know that the longer I keep my shutter open the more noise I will get....would it help if I backed my time to around 120 to 180 seconds and done twice as many lights? Thanks for any help....The wide field image I took of the horse head and M42 was the same set up but I took 25 240 sec Lights and ran in to some post processing woes.

[ChrisLX200]

The longer you expose then yes, more background noise will accumulate - however, if the target is actually brighter than the background then the signal will always accumulate faster than background noise. You are only interested in signal that is above the background noise level, that is the only data you will be able to process into a useful visible image and anything below that is effectively lost. The longer you can expose then the more signal data you will get. So longer is better (with some caveats arround linearity, sensor saturation and heat build-up in DSLRs over time). I've gone out to 10mins with a DSLR (my old Canon 350D) during Winter nights and had good results.

Whether you get a good result will depend on how bright the target is, because for faint objects there is no substitute for longer exposures - simply adding lots of short exposures together won't add up to a few longer ones even though the total exposure time might be the same (the signal will always be down there just above the background noise making it very difficult to separate the two). However averaging lots of exposures does help things in another way by smoothing out the (random) noise component, and by doing that it allows you to see the faintest signals you captured which may only be marginally above the noise threashold. For non-random noise (i.e., hot pixels) you need dark frames in order to cancel them out (or use an algorithm that can remove them during post-processing, DSS has such an algorthim). The minimum number of light frames to average is around 10 (which allows sigma clipping - an effective method of reducing noise during stacking) but the benefits of stacking multiple frames starts to fall away (due to diminishing returns) above 30.

So 10 frames @ 2-5mins would certainly not be 'over-doing' it and will get you good results if you can focus and guide well for that period of time.

This image of the Veil Neb was taken a number of years ago (long before I had StarTools!) with the un-modded 350D and (IIRC) exposures around 5min long. I suppose I should go back to the raw data and reprocess it with StarTools to remove the gradients and stretch the data properly etc., but as an example of a faint nebula captured with a DSLR and long-ish exposures it will suffice.

ChrisH

[MadMaxwellSmart]

The longer you expose then yes, more background noise will accumulate - however, if the target is actually brighter than the background then the signal will always accumulate faster than background noise. You are only interested in signal that is above the background noise level, that is the only data you will be able to process into a useful visible image and anything below that is effectively lost. The longer you can expose then the more signal data you will get. So longer is better (with some caveats arround linearity, sensor saturation and heat build-up in DSLRs over time). I've gone out to 10mins with a DSLR (my old Canon 350D) during Winter nights and had good results.

Whether you get a good result will depend on how bright the target is, because for faint objects there is no substitute for longer exposures - simply adding lots of short exposures together won't add up to a few longer ones even though the total exposure time might be the same (the signal will always be down there just above the background noise making it very difficult to separate the two). However averaging lots of exposures does help things in another way by smoothing out the (random) noise component, and by doing that it allows you to see the faintest signals you captured which may only be marginally above the noise threashold. For non-random noise (i.e., hot pixels) you need dark frames in order to cancel them out (or use an algorithm that can remove them during post-processing, DSS has such an algorthim). The minimum number of light frames to average is around 10 (which allows sigma clipping - an effective method of reducing noise during stacking) but the benefits of stacking multiple frames starts to fall away (due to diminishing returns) above 30.

So 10 frames @ 2-5mins would certainly not be 'over-doing' it and will get you good results if you can focus and guide well for that period of time.

This image of the Veil Neb was taken a number of years ago (long before I had StarTools!) with the un-modded 350D and (IIRC) exposures around 5min long. I suppose I should go back to the raw data and reprocess it with StarTools to remove the gradients and stretch the data properly etc., but as an example of a faint nebula captured with a DSLR and long-ish exposures it will suffice.

ChrisH

This is a good explanation Chris. I'm going to try the same target again when ever it decides to clear up. Thanks ....Glenn

[admin]

Great answer! Thanks Chris.