DSS Fail. Help please!

[Iainp]

Hi all. I'm just beginning to experiment with deep sky imaging, using Nikon D3100 (unmodded), on 25 year old 10" SCT. The drive is quite unreliable and unsteady and I seem to be limited to around 20-30 seconds exposures at the moment. I tried some 10 second trials last week, and was excited to get some 'decent' results (first 2 images below; Ring nebula and M15) so last night I started to take things a little more seriously by tethering the Nikon to a laptop, and using a focusing mask.  For the first time I also took bias frames and flats too. (I read somewhere here that darks with a DSLR often add noise, so can be left out, please correct me if that's wrong ;-) 

I too 20+ light frames of the dumbbell nebula, ISO 800, 20 seconds each. As you can see below, the individual frames are really dark, and stars are pretty much invisible. This is odd, it seems to me. Anyway, I put everything into DSS and it says "no stars visible". I understand the need to play around with % slider to pick out enough stars to stack images, but whatever I did, DSS couldn't see any stars. These are RAW (nef) files, so can be stretched in UFRaw to show lots of stars and detail in the nebula. (An "extreme" stretched and denoised one shown below the original, just to show what information is in there), so I used the "Exposure Compensation in EV" slider to reveal enough stars, then saved these as uncompressed Tiffs. (example below). Putting everything back into DSS produces the message that "The checked pictures are not compatibles (width,height, number of colors, number of channels").  The same thing happened when saving the stretched images as PPM files.  Next I tried converting the bias  and flat frames to tiffs too. This produces the message that there are no stars visible, even though there are clearly dozens of stars!

I've reached the end of my tether, so to speak. After finished the imaging session at 1.00 this morning, I made the mistake of trying to do the processing straightaway rather than leave it till the morning. As a result, i've had virtually no sleep, and still nothing to show for it! Can anyone help please? Is there a way to stretch the images within DSS to reveal the stars? I tried playing with the levels slider in the top right hand corner of DSS screen,  which makes the stars visible on screen, but doesn't seem to help DSS actually register the stars.  What's extra frustrating is that I managed to get a result with less effort before.   The only change I made to the camera this time, was that I noticed the EV compensation was set to 2 in the settings, and I put that back to zero. Perhaps I should have left it at a high level? As I say, it seems odd that virtually no stars are visible in the original RAW file, even with 20 second exposures on a 10" scope. Thanks, Iain   

[Ant]

I'm at work right now, and I haven't imaged for quite some time! But I do remember that in one of the options area in DSS, there is a star detection threashold setting.

I'm pretty certain that you need to increase or decrease that - I seem to remember that it worked the opposite way to how my brain said it should do - so try increasing, and then if that doesn't work decrease it and try again.

 

Good luck

Ant 

[Iainp]

8 minutes ago, Ant said:

a star detection threashold setting.

Thanks Ant, yes, that's the % thing (Sorry, I couldn't remember the proper name for it!).  That doesn't seem to be the issue though, as I've tried it on all settings, right up to the point where it "recognises" noise as stars and the whole thing falls apart ;-).  The image itself is too dim to reveal the stars at all, it seems, unless image is stretched and saved as something else, but then I can't get DSS to accept the modified files...

Cheers, Iain 

[Owmuchonomy]

Can you post your DSS settings?  Also, 20 seconds ISO 800 is a bit short to be honest.

[Iainp]

Thanks Chris, will do. And yes, I'll try 30 seconds tonight at ISO 1600. The wind was awful last night, so not sure how much of the poor image quality was due to tracking issues or just the scope wobbling. 

Iain 

[Iainp]

Hi Chris, I just had a good look at the setting, and it's the standard default setting throughout, eg. "average" and "automatic" for all the Stacking parameters ,and for Star Detection I've experimented with the full range right down to 2% but still No Stars detected . I'm at a loss what to do now, except to exposure for longer times and to increase ISO, which I'll try tonight, if the forecasters aren't lying about the clear skies...

Iain 

[Iainp]

Result! I increased ISO to 1600 last night, and set the EV compensation to 3 in the camera, and that brightened things up a lot, and I've just produced a fairly decent Dumbbell  . I'll post it up now. Thanks again, Iain 

[Tiny Small]

If you load TIF files, the stars won't be detected and if you try to mix RAW and TIF, they won't stack. All the photos need to be RAW, the same colour space, bit depth and size otherwise DSS throws a fit. I had similar problems this week.

[Iainp]

Thanks.  I'm struggling on to get DSS to work for me, despite the enthusiasm of my previous comment.  I started taking longer exposures, assuming the problem was lack of bright stars but this didn't work, and all my frames were consistently rejected, despite trying every setting I could think of. In the end I decided the slight streaking of the stars due to poor alignment and tracking was upsetting DSS. I reduced exposure back to 20 seconds, and raised the EV compensation in the camera to 4. Even though stars were dimmer and less were visible, they were more point-like and this seems to have solved the problem for now, though I'm surprised it's so sensitive to even the slightest imperfections in star shape.  I'm still having fun with these experiments but suspect the frustrations of my set up will test patience and stamina to the limits, and I may end up giving up and having to wait until I've got some cash to do this thing properly! 

Iain