Ngc3324

[peter_4059]

I got out to the AAQ dark site near Boonah last night for the first time in over a year. I'd forgotten how much work it was to pack everything in the car and then set up at a new location. I finally got imaging two hours after arriving and managed to capture almost three hours on this target before it was time to head home. I adjusted the colimation this time and think I've got the spacing for the paracorr about right.

Comments welcome...

[desler]

Got to be happy with that one Peter, well worth the trip.
Hard to tell with the jpeg, but are there some lighter artifacts in the left top and bottom corners? I'm not overly familiar with the target, but just something that attracted my eye!

Keep having fun.

Darren

[peter_4059]

Thanks Darren. These could be artifacts - still trying to work out flats with this camera however it's hard to say as there is so much going on around this object.

[peter_4059]

I got very ambitious today and took a full size copy of this to the printer. I was surprised at how different it comes out of the printer to how it looks on the screen. Where can I find some tips on getting an image ready for the printer to give it the best chance of looking good?

[desler]

I'll be watching the reply to the printer question with interest, haven't had the best of luck printing mine either!


Good luck

Darren

[allan gould]

As Dave Bowman should have said, "My God, its full of round perfectly focused stars from corner to corner".
Well worth the visit to the dark side., Luke
Really excellent shot and worthy of your new camera.

[troypiggo]

You're right. Fantastic mate!

[gregbradley]

Great shot. I think a bit of a gradient in the corners where there is a light patch possibly from overexposed flats.

Greg.

[multiweb]

Nice. Very crisp shot. Crystal clear. Good work. :thumbsup:

[peter_4059]

Thanks Allan, Troy and Greg.

Greg, can you provide any pointers on how to get the flats right? The flats I used for this were 20x1.5 seconds. I used Nebulosity to capture and based on the pixel stats over the entire image the pixel levels range from 6780 to 14432 (mean 10892). I thought the flat should be exposed at about 30% and 60% of the maximum?

[David Fitz-Henr]

Great image Peter !! Stars are well focused right into the corners - as a fellow Paracorr aficionado I salute you !!

[peter_4059]

Cheers David. I think I've finally got the spacing right and the colimation was having an impact. I've been playing some more in PS - here's my latest repro

[Hagar]

This is one very nice image Peter.... A shame about the vignetting/gradient/flats artifact in the corners but everything else in the image is really well done. How did you manage to highlight the reflection nebula and get the beautiful 3D effect. It almost looks like a bubble about to burst.
Even with the corners it is a great image. I would think it is a problem with your flats as I have had similar and found flats captured with an average of 23000 produced the best results for me. They have to be the hardest thing to get correct and all you can do is experiment. I actually sat one night and captured 10 different sets of flats after an imaging run one night and just processed the image 10 times then used the set which came out the best. Took a while but it did give me somewhere to start next time and I found they ended up being my flats setting for all future images with my QHY8 and Pro.
This is the post and comments made at the time: http://www.iceinspace.com.au/forum/showthread.php?t=56267

[allan gould]

Love the second redo. Really 3d and it stands out brilliantly.

[peter_4059]

Doug,

Thanks for the comments. To be honest I don't think I could replicate what I've done (lots of tweaking!). Thanks for the tips re flats - they are quick to gather so I'll try some longer exposures next time. It is good to have a target level to shoot for so I appreciate you sharing your experience.

Alan - thanks for getting me onto the QHY8.

[David Fitz-Henr]

That is sublimely beautiful Peter! Long live the Newtonian !!
You've probably done this but just in case - if you have changed the spacing of the Paracorr since taking the flats you will need to redo them. Also, it depends on your camera, but I believe the recommendation is to take longer exposures for flats to avoid shutter artefacts (of course you would need to filter the light or otherwise).

[peter_4059]

Yes - these flats were from last night straight after taking the lights so the spacing was the same. I have the dimmer on my light box at minimum so it looks light I need some extra resistance to allow longer exposures for my flats from what people are saying. Off to Jaycar tomorrow.....

[desler]

The second repro is amazing Peter, definately one you should try in the printer!

Darren

[RobF]

Interesting object and one I'm not familiar with, but certainly well collected/processed it would seem Peter. Sounds like you had a good night out.

[peter_4059]

Doug,

Thanks for the link to the thread on Flats. I'm looking for a bit of clarification on the measurements referred to in the thread. Are people referring to the mean or maximum pixel value? Ie you suggest 22k ADU - is this the maximum or mean pixel value in the flat? What software are you using to measure this? Nebulosity has a pixel stats window that gives min, mean and maximum pixel values - will this work?

Did you end up plotting the exposure vs pixel value curve for the QHY8pro? Mine still looks pretty linear up to 26180 maximum pixel value - might have to try some longer exposures.

Peter

[gregbradley]

Here is what I have done for flats for years with good results, others may have different approaches:

1. I put a white T-Shirt over the end of my scope and smooth out any wrinkles so it is firm and flat (may need to be tied to the scope or a rubber band or something).

2. At dawn or dusk (usually dusk) I point the scope towards the west area and arund 15-20 degrees up into clear sky.

3. I take a flat using CCDsoft software. I take 3 flats for each filter. I aim for 1/3rd total exposure. For a 16 bit camera that is a figure around the 20,000 ADU mark. As dusk proceeds you will see that number drop from flat to flat so you have to move fast and you may need to increasr the exposure time as you go along to compensate for approaching darkness. I get them all done in about 4 or 5 minutes and this is where fast download speeds of the FLI cameras helps.

4. Because my FLI camera uses a physical shutter I find exposure times have to be about 3 seconds minimum otherwise you can see the shutter in the flats. SBIG cameras don't have this problem with their blade shaped shutters which don't have that effect. I believe you can also take flats with FLI cameras with some chips using video mode but that is not accessible to me in CCDsoft and I believe you need to use Maxim to get that feature.

5. I also take a bias, usually this is one from my library. Some say it is not needed, I take Stan Moores advice on this (the writer of CCDstack). I think it is a minor point and doesn't make a big difference but there was a reason which I don't recall clearly now. I think it was do with scaling images.

6. I combine the 3 flats I take, subtract the bias and use median combine or average to make a master flat.

That works fine. You do need to make sure the flat is taken with the same optical setup. If you change anything with your optical setup the flat is invalid. For example if you rotate the camera to frame a shot some dust donuts that were on the scope will no longer line up.

I take flats for each filter and they do vary in appearance from filter to filter so you can't just take one and use it on all of them. You also need to do your flats at the same binning you imaged at.

I hope the above helps. It is simple and easy to use but you have to remember to be setup and ready just before dusk. You can easily miss this short period of time when the flats have to be taken.

Greg.

[peter_4059]

Thanks for the info Greg. When you say you aim for 1/3 of the total exposure (20,000 ADU) how is this figure measured? Is it an average or peak value? Does CCDsoft report this figure or are you looking at the histogram?

Peter

[peter_4059]

After a lot of experimenting and asking questions I've had another go at this image - this time with some bias frames.

[RobF]

Nice repro Peter
Deeper colours, better balance (green tamed nicely) and soft colourful stars. Beauty! :thumbsup:

[peter_4059]

Thanks Rob. I realised after I posted that the JPEG compression really didn't do this justice. I'm really pleased to have the flats issue worked out!

[troypiggo]

Wonderful repro.

[ozstronomer]

Peter

Thats really well done. What flats did you eventually decide on ?

Do you think the bias made much of a difference in getting rid of the green or was that more to the photoshop repo?

Geoff

[peter_4059]

Thanks Troy. Geoff, the bias made a lot of difference to the flats processing as I was getting a light ring around the images without the bias. I've done the colour balancing manually.

Peter

[gregbradley]

A very striking image Peter. Nicely reprocessed.

Greg.

[peter_4059]

Thanks Greg.