What's gone wrong? I have some ideas...M42 in DSS

[Fordos Moon]

Canon 1100D, ED80 on unguided HEQ5. 21 x 60 secs ISO 800, darks, flats and bias. Never done flats and bias before.

Image seems too bright but only on left hand side.

My flats may have had a bit of blue in them from sky.

My flats may have not been focussed same as lights.

Have I done something obviously wrong?

Image appears grainy?

[Fordos Moon]

Update. I am now reprocessing in DSS without flats to see what effect they had.

[Fordos Moon]

Without flats:

[IanL]

Yes it looks like a flats problem:

- Don't remove the camera from the OTA, re-orient it or re-focus it between lights and flats. It is hard to avoid changes to the imaging train if you wait until the next day to get your flats. Even dust can migrate around in that time (as I have found to my cost).

- Assume these are T-Shirt type flats? Ideally you should take flats against an evenly overcast sky. This will reduce colour balance problems (but even so you can usually re-balance in post-processing). The main issue though is that if the sun is falling on the scope, you may well find that one side of the flat is brighter than the other because the light on a sunny day is directional. Even if you put the scope in shade, reflected light from buildings, etc. may cause an issue.

- The simplest solution is to make a light box or buy an EL panel to do your flats, since you can get them in a couple of minutes straight after the lights, avoiding all the problems above.

- The grainy quality of the image simply means that you need more exposure. Turn the ISO down to 400 or 200, since all that 800 is doing is reducing your dynamic range and causing the core to burn out sooner. Shoot as many subs as you can. The more you have the more you will be able to stretch and bring out details without bringing out the grainy noise. Make sure you have plenty of flats to stack and also darks. I'd go for at least 30 of each if you have the time available to make them.

- Shoot a bunch of subs at 30s, 60s, 120s and either do a HDR stack or stack each set separately and layer them together in a composite with masks (lots of tutorials around on how to do that). Unlike most objects, you will need multiple exposure lengths for M42 since the is such a big range of brightness between the core and the outer parts. There is no way you will be able to get the outer regions fully exposed without burning out the core using a DSLR or OSC, so this is the only way to do it.

[Fordos Moon]

Ian thanks so much for taking the time to help me with this problem and explaining all! It's great to know how I can reduce the grain too!

[DaveS]

Wasn't there a thread on here a while ago regarding iso noise in Canon DSLRs? The conclusion was that iso 800 was probably the worst iso to use, either iso 400 or 1600 giving substantially better noise, though iso 1600 will have reduced dynamic range and colour depth.

I've been seeing a troublesome background gradient in my latest attempt at M42 which I've put down to moonlight in the lights, but I'll try stacking without flats, see if it goes away.

Looking again at your images, although you've lost the gradient in your second version, you've also lost The Running Man nebula as well.

[Fordos Moon]

Thanks for the info Dave about ISO800. I think the running man ran off into the flats!

[IanL]

The noise issue is a bit debatable in terms of which ISO is optimal. It certainly varies between different Canon models, and probably varies significantly between individual cameras. Sky backrgound is also a factor; simplistically and the more LP you have, the less noise really matters and you should favour lower ISOs so you can expose for longer. There are means by which you can identify the best ISO for your camera in terms of noise contribution, sky background, etc. but it's somewhat technical and didn't want to get in to all that.

Accepted wisdom for a basic approach is that anything over about ISO400 is a waste of dynamic range compared to any improvement in noise. Even the difference between ISO200 and 400 is probably not that significant and I'd favour more dynamic range (lower ISO number) in this case since the exposures are going to be pretty short come what may, so you can take plenty of them vs. a target that needs a 10 or 20 minute exposure per sub.

[DaveS]

I'm coming round to the idea of not going above ISO 400 as well. The only thing is, until I start guiding I'm stuck at a max of 2 min before trailing.

In any case I think we've lost M42 for this year, since by the time the moon wanes and gets out of the way Orion will be getting low in the east by the time it's dark enough for imaging, even if we get clear skies.

Time to look for other targets to practice on.

[dph1nm]

Accepted wisdom for a basic approach is that anything over about ISO400 is a waste of dynamic range compared to any improvement in noise. Even the difference between ISO200 and 400 is probably not that significant and I'd favour more dynamic range (lower ISO number) in this case since the exposures are going to be pretty short come what may, so you can take plenty of them

As you say it depends on your camera, but for most Canons if you are doing short exposures you really want to avoid low ISO. If you look at the 1100D, the read noise is 4x higher at ISO200 than at ISO1600. Even in a light polluted site you are going to struggle to avoid being read-noise limited at ISO200.

NigelM

[IanL]

As you say it depends on your camera, but for most Canons if you are doing short exposures you really want to avoid low ISO. If you look at the 1100D, the read noise is 4x higher at ISO200 than at ISO1600. Even in a light polluted site you are going to struggle to avoid being read-noise limited at ISO200.

Would be interested in some references for that conclusion, there certainly does seem to be a fair bit of contradictory advice on this issue. I haven't got an 1100D so can't test, but with a 500D, the results at lower ISOs are less noisy than at higher ones, but that hardly counts as an exhaustive set of tests. Craig Stark's conclusion when testing a 450D (http://www.cloudynights.com/item.php?item_id=2786) was:

"The camera’s internal gain (e-/ADU) for each ISO value points toward limiting the ISO to 400 and not using higher values. Both in theory and in practice, using higher values limits the dynamic range and does not let you pull out fainter details from the noise (even if they look brighter). The exact optimal ISO value will likely vary from model to model, but it's unlikely to be the high ISO settings."