How do you choose length of subs?

[John C]

I don't have an auto guider yet. I'm using an 8" Edge HD with 0.7 x focal reducer and I've got a Canon 60d.

What would happen if I took lots of 30s subs instead of the much longer ones people normally do? I could reject any with star trailing.

I'm thinking of trying some DSOs.

Sent from my iPad using Tapatalk

[Snaxmuppet]

The way I understand it is that the longer subs will gather more light per pixel and so it will bring out the fainter detail.

Short subs properly expose the brighter areas but may not expose at all the fainter bits. Long subs may blowout the brighter bits but then properly expose the fainter areas. The stacking software basically takes each sub and uses the part that is best exposed.

If you only take short subs then you will not capture the fainter detail.

[Tinker1947]

You will need to get a balance between what you mount will allow when tracking, what sort of LP you have, global clusters might be a good starting point, not to dim and loads of stars....

[Freddie]

Not so sure that is correct. The stacking software will just stack what it is given. You have to layer different stacks if you have different levels of brightness to cope with and therefore take different exposures.

[Snaxmuppet]

Does that mean you have a 2 stage process if you take subs of different lengths? First stack together all subs of the same length into several different stacked images and then layer the resulting images? I thought stacking software like DSS took care of that?

[carastro]

Does that mean you have a 2 stage process if you take subs of different lengths? First stack together all subs of the same length into several different stacked images and then layer the resulting images? I thought stacking software like DSS took care of that?

I can't remember if DSS will accept subs of different lengths, it will certainly remind you in any case.  But if it won;t then stack the stacks separately and save as tiffs then stack the stacked tiffs.  

However mostly the purpose of taking different length subs is to be able to layer in the brighter areas separately so they don;t get blown out as you stretch the fainter areas.

Carole 

[Tinker1947]

I do believe that how it would work, different Darks as well due to change in temperature...

[ollypenrice]

There is very, very rarely any need for multiple sub lengths and in your setup over-exposing is going to be the least of your worries. Multi sub lengths are needed for M42 and... not much else. If you have them I would stack in two stages, probably.

You are imaging at about 0.66 arcseconds per pixel which is a tall order. It's what Yves and I were doing with his 14 inch ODK and we used a premum autoguided Mesu mount to deliver the tracking accuracy needed. You may well find that the seeing (atmospheric turbulence) won't allow you to acheive that kind of resolution anyway, but in order to produce pictures at the best possible resolution from your setup you really will need to autoguide and autoguide well.

In the mean time the answer to your question is simple. Make the subs as long as the mount will allow. They will be nowhere near as long as would be ideal.

Once you are guiding the subs should be as long as your skyglow will allow, which means when the histogram peak is about a third or a quarter of the way to the right. Multiple short subs are a reasonable way to start but long subs do pull in more faint data.

Olly

[Uranium235]

Depends on what youre looking for really, if its a bright target (M42, M57 etc) then you can get away with short(ish) subs. Darker, deeper stuff takes much longer exposures, for example CTB1 (a very dim supernova remnant) requires 1800s (30min) subs with a CCD camera in order to get anything useable.

Planetary nebulas are usually quite bright, and would proabably look alright in your Edge HD. You wont know until you try I guess, so just pick a target, take a few subs, then examine them to see if its worth chasing.

[Snaxmuppet]

So, just to confirm... for most targets we should just take subs of a single length being as long as our guiding/tracking will allow?

[jnb]

I don't have an auto guider yet. I'm using an 8" Edge HD with 0.7 x focal reducer and I've got a Canon 60d. What would happen if I took lots of 30s subs instead of the much longer ones people normally do? I could reject any with star trailing. I'm thinking of trying some DSOs. Sent from my iPad using Tapatalk

Your mount is more important than the optics. I am not using guiding yet and have to keep my subs short enough to avoid major problems from periodic error in the mount. In practice that means I have to keep exposures below 2 minutes and that is limiting the detail I can get from an image. It also means that to have usable data from those two minutes I need to go to higher ISO settings which in turn means more noise. I can probably improve that with a bit of tuning on the mount but ultimately I know I am going to have to add guiding to overcome those problems

[Tinker1947]

so guiding option are a Finder Scope Guider the cheapest option, a ST-80 many more stars with a set of guide rings it can be aligned perfectly with the main scope, OAG more expensive finicky to set-up if its not permanate you could spend a bit of time get a guide star. guide CCD starting from the top a Lodestar is good, there are other cheaper CCD'd but i don't have names or info.....

Finder Scope Guiding

ST-80 Guiding (my choice)

Off Axis Guiding

[Freddie]

So, just to confirm... for most targets we should just take subs of a single length being as long as our guiding/tracking will allow?

Correct. If you are not guiding then your tracking will be the limiting factor not your sky conditions (LP etc)

[ollypenrice]

Correct. If you are not guiding then your tracking will be the limiting factor not your sky conditions (LP etc)

Another 'yes' to this.

A thing that beginners sometimes overlook is the need to understand the relationship between pixel size and focal length, determining your pixel scale. It is this pixel scale, measured in arcseconds per pixel, which really determines the guiding accuracy you need. 3.5 arcsecs per pixel is tolerant. That's what we have in our dual Tak rig with 530mm FL and huge 9 micron pixels. Resolution is low, though. 0.6 arcsecs per pixeel is highly intolerant. With mono cameras there is an easy solution, you bin the pixels 2x2 to make them 4x bigger. You can't do this with OSC cameras though. If you did you'd end up binning your reds and blues and greens together and that wouldn't be a pretty sight! 

As DSLR pixels become ever smaller their ability to work sensibly at long focal length diminshes. They are best mated to short focal lengths and fast F ratios.

Olly

http://ollypenrice.smugmug.com/Other/Best-of-Les-Granges/22435624_WLMPTM#!i=2266922474&k=Sc3kgzc

Edit; I've never found any need to use guide rings for guidescopes. The ones bundled with the finder guiders are fine but using ST80s I just bung them on in a 'bolted down hard' manner and that's all you need.

[Tinker1947]

Olly, my ST-80 worked fine without the guide rings, i added them as i wanted to use the ST-80 as a Finder scope, so what's in the centre of the PHD screen is also in the centre of the Camera sensor so i use a GoTo to slew then adjust on the PHD screen then just take a image, it also works when i do the PA watch the chosen star move away and come back then adjust to get it spot on......so its not necessary to have guide rings but i like them...

[John C]

Thanks everyone, I've got lots to learn. I've been reading up on the relationship between pixel size and focal length. I'll get a guide camera in the autumn probably.

Sent from my iPad using Tapatalk

[dph1nm]

Using shorter subs makes your observing less efficient -  but it does not stop you seeing faint detail. You just have to keep taking more subs!

The ideal sub is one where the read noise of the camera per pixel is substantially below the other sources of noise (which in most cases probably means the noise from the sky background) per pixel.  The larger your pixel size in arcseconds the more sky you get in it and hence the less likely it is to be affected by read noise. Paradoxically, light pollution will also help!

What "substantially" means is probably up to the individual to decide (e.g. a factor 10 might satisfy some people but not all), but the one thing I would note, though, is that some modern DSLRs  have VERY low read noise (e.g. 1-2 electrons), and you may be able to get away with quite short exposures on these cameras.

NigelM

[MattJenko]

Also, some targets absolutely don't want 20 minute+ exposure times. A star cluster for instance will have simply lots of white saturated stars at that exposure length, so I actually do much shorted subs for things like these objects.

 

Cheers

Matt

[John C]

Thanks everybody. I'll experiment and see what happens.

Sent from my iPad using Tapatalk

[lensman57]

Thanks everybody. I'll experiment and see what happens. Sent from my iPad using Tapatalk

Experiment is the order of the day. You'd probably find that your subs length are more determined by you local sky and LP since the camera and the scope are a constant factor.

A.G

[fielderda]

I've just bought a EDGE HD 8" SCT with a 0.7x reducer and using my Canon 60D.

PHD guiding with an Orion Autoguider allows excellent tracking and I've done 10 x 5 minutes subs with a few Nebulas with good results (as a newbie). I've haven't tried longer subs yet.  I've also done multi-30 sec exposures but it's too early to identify what works best. I guess it depends on the object that is being photographed.

I would concur with some of the other comments that without a auto guider, the length of subs will be shorter.

Clear Skies!

[Physicist13]

one of my "pet" topics... :/.... a good one hour talk... but bottom line is...try to get out of the "read noise" of whatever imaging device you are using..then only total imaging time matters....with read noise you will need a longer total imaging time to get a particular SNR. 

Pat