Dithering preperation

[ianpwilliams]

I had a go at stacking my 10-min and 5-min Andromeda sessions together in DSS and it seems have worked well. It's really easy too, you just load all the frames in for a session as usual, and then do the same for each other session in a new tab each time.

 

So then I started looking back at my old unguided Andromeda sessions (I've done three unguided ones prior to my recent two guided ones). In one of them Andromeda isn't even centered so that's out. And another one might have been a possibility if it hadn't been for the fact that the galaxy is 90 degrees rotated, so I'm not sure what would happen there. There is another one though which is actually quite close (90-second subs), so I had a go at including that with the two new ones.

 

Looking at the three TIFs after applying screen transfer (just to get an idea) it's hard to see any improvement really. But then maybe I would see an improvement if I attempted to process all three first and then compare. I also noticed that the three-session combination (bottom-right) has more unwelcome colours (vignetting?), and so I wondered if including my old session might be a bad idea, particularly given that the subs from that session have light pollution, and neither of the two new sessions do. But then it is extra data.

[PhotoGav]

Looking good! The only thing I can say is that the quality of the final image rests mainly in the quality of the starting data. If you have a set that is sub-standard you will do better to not include it. That is true of subs within a set too. It's tempting to just add in a sub even though you know it's not really up to standard. I'm pretty certain that it is detrimental to the stack.

Once you have done all the gradient removal and colour correction I think you will make a pretty decent image with what you have. The dust lanes are already starting to emerge and the core seems nicely under control. It certainly is a spectacular object and to think that we are a small blue speck about two thirds of the way along one of those spiral arms in a similar looking Galaxy only two and a half million light years from there!!

[ianpwilliams]

Yes I think you're right. I'll probably have a go at processing all three to compare and contrast, but my suspicion is that the two recent guided sessions combined will provide the best results.

 

It's a dangerous thing combining multiple sessions though, because that means you could theoretically go on forever! I could get carried away with this, given enough clear nights!

 

And yes, the distances are mind-blowing. Did the light from Andromeda really take 2.5 million years to reach us? Given how fast light travels, how can that even be comprehended? And of course it makes it even more miraculous that we are able to grab any photons at all over such huge distances. Despite all the frustration and challenge, and the time, expense and effort that is required, it really is an incredible hobby.

[PhotoGav]

It is addictive!

[ianpwilliams]

9 hours ago, PhotoGav said:

Looking good! The only thing I can say is that the quality of the final image rests mainly in the quality of the starting data. If you have a set that is sub-standard you will do better to not include it. That is true of subs within a set too. It's tempting to just add in a sub even though you know it's not really up to standard. I'm pretty certain that it is detrimental to the stack.

It's tricky to know which sessions to include at the moment.

 

The first session (90-sec, 800 ISO) was unguided, no LP filter, no dithering (but lots of darks).

 

The second session (10-min, 100 ISO) was guided, no dithering (but I took a few darks), no LP filter.

 

And then my third session (5-min, 800 ISO), was guided, dithered, and with LP filter.

 

Maybe I should just process the three stack versions and see how they compare.

[ianpwilliams]

Am I right in saying that Flats aren't temperature-dependant?

 

I've got some clear nights forecast this week, and my original plan was to take some 30-second subs and some Flats in the first session, and then take 5-min subs with no calibration frames in the other sessions because I've got my Darks from the other week and so I won't be disconnecting my camera any time soon. 

 

But if the Flats aren't temperature-dependant then it would be a lot easier to do them indoors, and concentrate on nothing but subs when outside.

[PhotoGav]

I think that's right. I suppose the thing is that the flats are a short exposure time so noise won't be an issue.

[ianpwilliams]

Yes it seems that they aren't:

 

http://www.budgetastro.net/calibration-frames.html

 

Although he says that the ISO should be the same as the Lights, but I was told to go 100 ISO in AV mode, and aim for a one-third to half histogram peak. Anyway, looks like I could do them indoors.

 

 

[ianpwilliams]

Thinking about it, I'm wondering if his method might be better, doing the same ISO as the Lights, in manual mode, starting with a 1/80th shot, and adjusting the sub length until the peak it at 50%. At least that would be quick once you had the right sub length, unlike in AV mode where the camera decides the sub length, and where I've often found that the sub length can be really quite long (20+ seconds), no doubt because of the 100 ISO. 

 

Do you do your Flats in AV or Manual mode?

[ianpwilliams]

I think I'll try manual mode, as I doubt it'll matter as long as the peak is in the right place.

 

I'm also thinking that maybe thirty 30-subs should be enough for layering a processed stack on top of the other stacks, and that'll only take 30 mins total when including 30-second intervals, so hopefully that might leave enough time to also do an hour's worth of 5-min subs too.

[PhotoGav]

I almost always do flats indoors the next day. I use a CCD now, so it's a slightly different process for exposure. However, when I used a DSLR I used AV mode at the same ISO as the lights and darks. I used an iPad white screen with my 80ED & then bought a light panel to use with my 8" scope. The exposure times were always fast, 1/100s or thereabouts ish. Not sure why you are getting long exposure times on AV - is your light source very dim? It will certainly work with Manual mode, you just need to do some test shots to select the correct exposure time.

Sounds like a good plan for your next imaging session, get those short exposures in the can and then go back to the long subs. The problem is that while the forecast is clear for the next few nights, we have a full moon to put up with - definitely not ideal imaging conditions...

[ianpwilliams]

I'm not sure why the exposures where long in AV mode. Maybe it had something to do with my remote timer, although I always set the exposure length to zero on that as AV decides the exposure length anyway. Or maybe the laptop was too dim. I've always put a sheet of A4 paper over the laptop screen when I've done my flats, which I read somewhere. I don't know what the reason for that was, maybe just to soften the pixels on the screen or something. I'll definitely try manual mode because then I have full control over it.

 

And yes, I too noticed that the moon is full. It is far away from Andromeda at least, but still not ideal. But I have to get out there while it's clear!

 

[ianpwilliams]

Ok I've got 50 Flats, all with a nice 50% peak. Can't believe how much easier it is to do that indoors, and on manual mode.

 

Would you still do some sessions this week, given that it is a full moon? It would seem such a waste not to when there are currently three clear nights ahead.

[PhotoGav]

Well done with the flats - yes much easier when warm and fully awake!

It's always a tricky one, the Full Moon. It's very tempting to make use of a clear night despite the Moon, but it does compromise the quality of the subs. The temptation is stronger when the weather is rubbish most of the time and a clear night is rare. I do try and avoid the temptation though. I sometimes use the clear but moonlit nights for framing of next targets and have even been known to image the Moon itself. The other thing is to do some planetary imaging when things are well placed, unfortunately Jupiter is too close to the Full Moon for it to be any good at the moment. I am, however, very tempted to try and gather some subs for my current target over the next few nights. I know I shouldn't, but my target, The Jellyfish Nebula at the feet of Gemini, is rapidly sinking into the western horizon and I don't have much opportunity remaining!

[ianpwilliams]

Yes, to not go out over the next few nights when I've been waiting for weeks would be frustrating to say the least! In this thread here:

 

http://www.cloudynights.com/topic/380612-astrophotography-during-a-full-moon/

 

It does seem like trying something as faint as Andromeda would be a waste of time based on the information in that thread. However, one person in that thread says "Another option is to do objects that don't require the darkest skies, open and globular clusters are good examples.", which got me thinking that maybe I could have another crack at Pleiades, as that will be high in the south moving west while the moon is low in the east heading south. I could even do three sessions on it, all 5-min subs seeing as there is no galaxy core to saturate, although I'm not sure if there would be a benefit of doing three sessions of long subs with Pleiades being so bright anyway.

 

What do you think? Might it be worth doing? Or if not that then maybe another cluster (although I'm not sure what)?

 

[ianpwilliams]

Also on this page:

http://astrodoc.ca/deep-sky-imaging-in-moonlight/

"I have found that taking many relatively short exposures gives better results than fewer long exposures. I have found this to be generally true, but especially when the skies are bright from moonlight or light pollution. One reason is that shorter images have fewer (or no) saturated pixels in the bright areas. Another reason is that stacking more images does a better job of removing artifacts in individual frames, like hot or cold pixels. Finally, a lower background brightness in the image gives more dynamic range between the darkest and brightest pixels in the image."

I'm not sure if I would want to take fewer shorter exposures rather than less longer exposures as a general rule (otherwise what would be the point of guided imaging instead of unguided imaging?), but maybe it might be worth considering when there's a bright moon around. Any opinion on shorter/longer exposures, particularly under a bright moon, but also just generally?

 

[PhotoGav]

Moon light, especially at Full Moon, will make processing more difficult as it introduces gradients into the subs and will wash out contrast and detail. So, ideally there shouldn't be too much moonlight around to get the best quality subs. Having said that, it's hard not to, so give it a try and see how you get on. You will soon make your own mind up about the issue!

Few long subs vs. many short subs? I find in practice that I get best results with the longest subs possible. With narrowband filters I generally use 1800s subs and try to get as many as possible, at least ten hours of total integration time and much more if I'm able. With LRGB filters it really depends on the target, a faint galaxy will need longer subs when compared to a bright cluster. As with much of this, it's a case of experimenting and seeing what your sky conditions let you get away with.

[ianpwilliams]

There are a few clouds dotted about tonight, and with local training ground floodlights and the full moon too, it's not happening. So I suppose it's a case of deciding how much smaller the moon should be before going for Andromeda again, while hoping for clear skies. One day at a time!

[PhotoGav]

I tend to limit serious imaging with any moonlight to the two and a bit weeks around the New Moon. Then there are all sorts of cheeky opportunities for bits of the night before Moon rise in the days after Full Moon. I always like those stolen hours before a bright Moon rises, it feels like you have cheated the little sucker!

[ianpwilliams]

I think I'll leave it until close to a new moon and hope for clear skies for Andromeda.

[ianpwilliams]

Clear skies tonight finally! I'm hoping to head back to those M31 coordinates that you found me. I've got them saved in APT under Custom, so hopefully it should just be a case of selecting them and clicking GOTO++ in PlateSolving and letting APT do all the work...

[PhotoGav]

Good luck!

Have you seen the Aurora up where you are?

[ianpwilliams]

No, apparently it was visible as far south as Liverpool. I must have slept through it! I hope to see it one day though. Did you?

[PhotoGav]

No I'm way down south... There are some great pics on here though. One day I will definitely go in search of it.

[ianpwilliams]

Well that was a complete disaster. Whenever I tried to use GoTo++ in PointCraft I kept getting errors, including runtime errors, timeouts, and requests to check the ASPS settings. And when GoTo++ did "work" it was taking far too long  eg 10+ minutes just to do 3 of 5 steps, so I stopped it. And I got myself confused again in PointCraft, trying all the various buttons like Auto, Solve, Blind etc.

 

Also I could clearly see that the scope was pointing too far to the right whenever I slewed to a target using GoTo, despite the fact that I believe I did a reasonable polar alignment. Focusing was fine, as was guiding, and dithering seemed to be fine too, APT automatically going into dither mode. It was just the slewing that was the issue. 

 

Anyone got any tips on how I can avoid this problem next time? Maybe I should have just polar aligned again or something.