max sub length

[scitmon]

Assuming field rotation is not an issue, I was wondering how people determine what their maximum sub length is?  I read on a thread on cloudynights that one should aim for the histogram peak to be around the one quarter mark... is that correct?  When I was applying this rule last night imaging M31, I found I was hitting the 1qtr mark at ISO800 in around 90 seconds which isn't very long!  I was wondering why I would ever need to guide if my skies are that bright?  Or maybe this was down M31 being very bright and large for DSO's?

[carastro]

DSLR?

I was always told that 300secs and 800 iso should be the maximum because of noise.  I have tried 400iso and 600secs and it seemed to be a nice clean image.  

Images like Andromeda galaxy and M42 have very bright cores, and need long subs for the outer regions and shorter exposures = 30secs for the core and then combined in post-processing.

Carole  

 

 

[vlaiv]

Only two things really impact maximum length of exposure: well depth and guide precision. Well depth determines how much signal you can gather without clipping bright regions - like star cores and in some cases DSO cores / bright regions. If you guide, and guide good then you don't really care about maximum length - you can guide accurately for many hours, but if your guiding is poor or have trouble with wind or cable snag or something, you want shorter exposures so that in the end you end up with less wasted frames.

Need for long exposures is another matter - at a certain point there is no real benefit in going longer - in terms of SNR, but that depends on many factors (aperture, resolution that you are imaging at, broadband vs narrowband, sensor QE, ...). So you may choose to go with shorter exposures at some minor expense to SNR in order to end up with more data in the end - 1 minute wasted sub out of 100 subs is 1% data lost, 10 minute sub out of 10 subs is 10% waste - for same total 100 minutes imaging time.

[kens]

For DSOs at least, minimum sub length is more important. The faintest signal needs to be well above the read noise of the camera. Thats the rough basis for the histogram recommendation. Generally speaking, faintest signal corresponds to sky background. Even more important is integration time.

[StuartJPP]

Good answers from above... here's mine which pretty much says the same thing....

There is a sweet spot somewhere in the middle of exposure length and number of exposures.

Total exposure time is what matters the most. 

Expose for too long and you end up with saturated stars and the chances of throwing a lot of data away in the case of something going wrong for a single sub. Additionally you have to give the stacking algorithms something to work with, only having a couple of subs in the pot doesn't help.

Expose for too short and everything is stuck in the noise...

[scitmon]

So to bring it back to my example of M31... To get a good final image, I would take short subs like I was capturing last night, in order to get the detail of the core and not to overexpose the stars (i.e. 1/4 histogram peak) and blend that in photoshop with some longer subs to get the finer details of the galaxy's nebulosity.  These longer sub images as a whole would have blown out stars and DSO core, but those areas of the image would be thrown away.

That sound correct?

[vlaiv]

7 minutes ago, scitmon said:

So to bring it back to my example of M31... To get a good final image, I would take short subs like I was capturing last night, in order to get the detail of the core and not to overexpose the stars (i.e. 1/4 histogram peak) and blend that in photoshop with some longer subs to get the finer details of the galaxy's nebulosity.  These longer sub images as a whole would have blown out stars and DSO core, but those areas of the image would be thrown away.

That sound correct?

That would be one way of doing it, and it should work good. Also note that for star color to be good you done want to over expose the stars - similar to the core. So you can have two sets of subs, and then blend with luminosity mask - taking short subs stack only in places where there is "enough" light in long subs stack.

[Thalestris24]

I'd say it also depends on your local conditions i.e. skyglow and specific LP. If, like me, you have the misfortune to live in an urban area then skyglow especially limits the length of broadband exposures and even narrowband ones are affected.

Louise

[scitmon]

to the point where it's not worth buying a guide camera/scope?  That's what I am wondering at the moment.

[Thalestris24]

3 minutes ago, scitmon said:

to the point where it's not worth buying a guide camera/scope?  That's what I am wondering at the moment.

Oh I still guide  But, using a dslr, broadband exposures are limited to maybe 180-360s (depends on settings and it varies according to conditions). You have to find out what works best for you. I live in a 'red zone' and it never gets dark here. Things have improved slightly since a nearby streetlamp bulb went out  I expect they'll fix it sooner or later

Louise