M31 first meager attempt!

[ianpwilliams]

So as some of you will know I finally managed to successfully polar align last night, and I attempted to photograph M31, using my HEQ5 Pro SynScan, 130PDS, and Canon D1100, with settings of 3200ISO, 30 x 30-second subs, and round about 10 dark bias (I ran out of memory card space). So I've stacked everything in DSS, and I'm wondering if I will be able to get anything half-decent with it. I've attached a JPEG of one of the subs, and a JPEG screengrab of DSS after stacking, and adjusting the RGB/K levels based on basic guides online.

I'm well aware that I may get nothing at all from this, but just polar aligning was enough (and getting a decent shot of the full moon). But any tips would be welcome, as well as tips for my next session. I was advised to go for 3200ISO, which no doubt explains the brightness. I wondered if I should go 800ISO instead, but maybe it's just all swings and roundabouts because I would end up with the same result anyway, unless I could get away with a lower ISO and longer subs unguided?

[Zuben Elgenubi]

Congratulations! 

I've been following your setting up thread on and off, great to see you are now achieving real results. I was told that the 1100D works well at 1600 ISO, although mine is modified. 

Keep at it, your images will only improve with practice and more experience 

[ianpwilliams]

Thanks! I'm just not sure if I can do any more with that image in DSS (my knowledge being very limited), and if I do stick with that, then I wonder if I could do anything more with it in an image manipulation program. Or is it just a list cause do you think!

I could also do with tips for my next session, so that I can hopefully get better results next time.

[lensman57]

Thanks! I'm just not sure if I can do any more with that image in DSS (my knowledge being very limited), and if I do stick with that, then I wonder if I could do anything more with it in an image manipulation program. Or is it just a list cause do you think!

I could also do with tips for my next session, so that I can hopefully get better results next time.

You seem to have decent data there and it is a good first attempt. Save the stack without any of DSS adjustments and try to stretch it in either Photoshop or Gimp. If feel like it you can drop the un molested raw stack in drop box and I could have a go at it for you.

Regards,

A.G

[almcl]

I think you can definitely do more with it. (My knowledge is also limited.)  I have stretched your jpeg a little and can see dust lanes and the start of the outer regions.  More data will help  as would some offset for the vignetting.  Perhaps a more knowledgeable member will point out other things?

[Knight of Clear Skies]

Well done, you are certainly up and running, that's real progress and something you can build on. I can see dust lanes in the DSS screen capture.

The sub is indeed too bright, this is either from light pollution or the Moon if you were imaging while it was out. ISO and exposure time need tuning to your local sky conditions. Looking at the DSS grab though I don't think you've lost any data from overexposure, the bright core does not seem to be blown out.

Flats would help greatly with the vignetting (dark corners). You'd need longer subs to bring out the fainter spiral arms and a darker sky would provide more contrast and help them stand out. Since you talk about taking a picture of the Moon I assume you were shooting M31 while it was out? Clusters and bright emission nebula are better targets under moonlight as they have higher surface brightness (brightness per unit area) than faint galaxies.

Would be interested to see your Lunar image too.

[mcinek82]

Great work! As for postprocessing I can give You only a bit of advice. My workflow includes saving the basic DSS output, then flattening the gamma curve (under Luminance tab) and levelling the color intensity (Your picture shows it is already in place) and then saving again. Afterwards I use Gimp (too bad it allows only 8bit per channel), usually with the second picture, and tweak the histogram to achieve darker background. Then I adjust the colors, brightness and contrast - it's usually trial-and-error for me.

However, I'm sure You can bring a lot of Your picture - just look at my effort of M42 (subs looked similar).

[ianpwilliams]

Well done, you are certainly up and running, that's real progress and something you can build on. I can see dust lanes in the DSS screen capture.

The sub is indeed too bright, this is either from light pollution or the Moon if you were imaging while it was out. ISO and exposure time need tuning to your local sky conditions. Looking at the DSS grab though I don't think you've lost any data from overexposure, the bright core does not seem to be blown out.

Flats would help greatly with the vignetting (dark corners). You'd need longer subs to bring out the fainter spiral arms and a darker sky would provide more contrast and help them stand out. Since you talk about taking a picture of the Moon I assume you were shooting M31 while it was out? Clusters and bright emission nebula are better targets under moonlight as they have higher surface brightness (brightness per unit area) than faint galaxies.

Would be interested to see your Lunar image too.

Thanks, it definitely does fell like progress.

I think that was a combination of light pollution and the full moon that we had that night (which I will upload). Is there a way I can work out the best ISO setting for my light pollution levels, maybe with and without the moon too?

Can you give some examples of clusters and bright emission nebula?

[wxsatuser]

Use iso 1600 for faint stuff like nebs and 800 for clusters/galaxies.

Use sub lengths that give you an in camera histogram between 25 and 40%.

Without guiding it will restrict you a bit, you may not get to the 25>40%, depends on how bright your sky is.

[Scott]

Definately up and running now Ian. I think iso3200 would just about be possible from a dark sight on a moonless night. Last night with a full moon, darn near impossible so take comfort in knowing you're making progress and pa is becoming easier. when capturing your subs, have a look at the histogram and make sure the peak is between 1/3 from the left and half way. good luck and keep going

[wxsatuser]

If you want to use DSS to see the result, this is not the finished result btw, get your histogram like mine.

Get the steepest portion going through the bottom third of the histogram.

Use the sliders to adjust the curve to look similar to mine and push the saturation upto 20%.

Proper processing in other software is best but this will give you an idea whats in your image.

Here is one of the California Nebula.

[Knight of Clear Skies]

I think that was a combination of light pollution and the full moon that we had that night (which I will upload). Is there a way I can work out the best ISO setting for my light pollution levels, maybe with and without the moon too?

Can you give some examples of clusters and bright emission nebula?

The easiest way is to use the histogram display on your camera. On the 1100D, take a test shot and press the play button to review it. Then press Disp once or twice until the histogram comes up. If the peak of the graph is too far left then up the exposure time or boost the ISO level. If it's too far to the right then do the opposite. Getting the peak somewhere in the 25-40% range makes it easier to check framing and review subs as they come in, while leaving headroom to prevent overexposure of any bright objects. You want it to look a bit like the histogram in your DSS screen-grab.

Imaging galaxies when the Moon is bright is very tricky, they are faint objects. A common mistake made by beginners and something that I didn't realise for a long time is that magnitude is a poor guide to visibility. For example, Andromeda is mag 3.4 which is quite bright, but its light is spread over a large area. Surface brightness (brightness per unit area) gives a far better indication of how easy DSOs are to image.

A couple of good open clusters to image at the moment are the Pleiades and the Double Cluster but there are many others. Not sure about the best nebulae to target with your setup. Possibly the Dumbbell (M27) early in the evening or the Orion Nebula (M42) later?

[ianpwilliams]

Thanks I'll give that a try. Is it possible to save the stacked and unmodified project, in case you want to go back and start again without having to stack again?

[glowingturnip]

congratulations !!  that's looking very good.  You're on your way !

Now starts the dark art of post-processing - you'll be amazed how much more detail there is in that picture than you think you have,

As lensman says, if you can post a link to the unmodified tiff, I'd be happy to have a fiddle with it too.  Save it from DSS with 'settings embedded but not applied'.

Re ISO, I'm getting a modded 1100d for Xmas and Juan from CheapAstrophotography advised using ISO1600 with it.

[Scott]

Am I right in thinking that you are unguided at the moment Ian? If this is the case, and you are not drift aligning, I'd suggest trying for aound 60-90 sec exposures (maybe a touch more if you can keep the stars round). this in a way will determine your iso setting. Me, with guiding, I use either 400 or 800 to try and keep the noise down, but I'm in no way restricted by exposure times anywhere near what you are. If it's like last night with a full moon, add in any light pollutionand I'd say 800 max, possibly 400.

There are so many variations between different cameras, light pollution varies significantly so trial and error is really the best bet. I'd also concider a light pollution filter if you've not got one already 

Best of luck

[ianpwilliams]

Yes I am unguided. I'll try a lower ISO next time, maybe 800ISO, and hopefully 60 second subs. And I'll check the histogram for the test shot.

If anyone wants to have a play around with my TIFF file then I'd be happy for them to do it. I've been doing it all day but I don't really know what I'm doing. Would you need to email me your email address so I can share the folder with you, or could I message the link or something?

Also, the DSS output file is quite a bit darker now for some reason. I don't know if it's maybe because I chose the "Embed adjustment settings" option as you said? It's still very bright in the DSS window, but the TIFF is quite a bit darker.

Also, this is the 2nd time that I've stacked the images, and this time I modified the slider beforehand so that the number of stars detected was around 50 as per an online tutorial. But I don't know if I should have done that or not.

[glowingturnip]

it seems a good way of sharing files is to set up a Dropbox account, upload the file to there, and then post a link to here (although as for the details, I've never done it, so I'll leave that to you).

Don't worry about the tiff being dark, that will come out in stretching.  To be honest, although the histogram controls in DSS are quite user-friendly, and can make the image look good in the DSS screen, I don't think they're up to scratch for proper post-processing (I've found that it clips histograms amongst other things).  Hence the 'embed but don't apply' which means the saved tiff doesn't have any of the DSS post-processing done to it.

50 stars sounds fine

What programme are you using for post-processing by the way ?  Lots available, and many people on here can give you advice.  GIMP is free and lots of people use it, lots of people use Photoshop, and several geeks like me use Pixinsight (which is a tad pricey and a very steep learning curve !)

[ianpwilliams]

I've got GIMP at the moment. I'll not be investing in a paid one unless this takes off I can make some headway in the post processing!

Here's the link to the TIFF:

https://www.dropbox.com/s/if85opwa22d3mmp/M31%20DSS.TIF?dl=0

I am able to bring out M31 through the various methods mentioned. The tricky bit is differentiating it from the area above and to the right.

[ianpwilliams]

And here is a montage of three moon shots That I took recently, for those who aren't following the other thread:

https://flic.kr/p/pnJjMP

[lensman57]

Hi Ian,

I have conservatively stretched the stack in PI. very little processing has been done as it would be pointless.There are quite a few issues with this probably related to the mount as the stars exhibit a trail in two directions. Examine the frames as this may just be limited to just one frame and take it out of the stack.

There is severe banding issue with the camera though this can be removed by software, I only removed just enough to leave some. It would also be good if you could get the galaxy in the centre. I suggest that you return to this target when the weather and time allows.

Regards,

A.G

[ianpwilliams]

Wow thanks AG, fantastic job! How did you manage to get rid of all the stuff above M31? Whenever I managed to bring M31 out a bit, the rest of it came out too, causing it to look like a dome.

Also, could you give me an idea of what you did? Then maybe I could try to recreate something similar, which would no doubt teach me a lot. And what software did you use? If it was a paid product like Photoshop, would I be able to produce something similar in GIMP?

Also I'll check the frames today.

[ianpwilliams]

I've had a look through the JPEG versions of the subs for star trailing. Here's an example of a good sub

Then there are four subs which have bad star trailing, and this is one of them:

Then there are two (maybe one or two others as well) which have slight star trails, which I'm not sure about, like this one:

I'm guessing I should restack the raws minus the four bad subs. But should I also remove the ones that are on the cusp? There may only be two, but there may be more like five when I look again.

[wxsatuser]

Wow thanks AG, fantastic job! How did you manage to get rid of all the stuff above M31? Whenever I managed to bring M31 out a bit, the rest of it came out too, causing it to look like a dome.

Also, could you give me an idea of what you did? Then maybe I could try to recreate something similar, which would no doubt teach me a lot. And what software did you use? If it was a paid product like Photoshop, would I be able to produce something similar in GIMP?

Also I'll check the frames today.

Sign up for a trial of Pixinsight and then goto Harry Page's tutorials.

[NickK]

I'm guessing I should restack the raws minus the four bad subs. But should I also remove the ones that are on the cusp? There may only be two, but there may be more like five when I look again.

Yup, I found that it's better to have less high quality subs.

[glowingturnip]

do post up the new tiff without the badly trailed frames, lets have another look

Taking flat frames will help you a lot with the vignetting you have , and should enable you to differentiate the galaxy from the background better.  Should be plenty of tutorials on the web, so something to do for next time

edit:  your third frame above is a reject, I'm afraid.  Only use frames like the first one, which actually looks pretty good