M27 - The Dumbell Nebula

[PhotoGav]

Here is my first image with the QSI on the Celestron EdgeHD 800:

Specs:

All at 2x2 bin and -20C

L = 11 x 600s

RGB = 3 x 600s each

Calibrated with flats, bias and darks

I was interested to see what could be done 'quickly' with this set up, i.e. an image in one night. I'm not convinced that 2x2 was a good option for the L data and I think that I will shoot a set of L subs at 1x1 and go longer too and see if that helps with definition. It will certainly help with file size so that a larger print can be made. The trouble is that the set-up will be operating at 0.55"/pixel at 1x1, which is way below the recommended resolution...

Please let me know what you think of this one.

Thanks!

[adamphillips]

that is the first time ive heard of this recommended resolution thing. i also just tried the new field of view calculator v2 and they have that stat in there.

but ive tried some different telescope and camera combinations and .55"/pixel seems like a pretty small number, im finding it hard to get much better.

although my 5d mark ii on my edge 11 gets me 47"/pixel, and the atik 383L that i plan on buying would get me .40"/pixel

it just seems like you need some serious focal length to get those smaller resolutions.

where did you find the recommended numbers for resolution as you were talking about? i don't see how all these people with their ccd cameras on little refractors could be getting that high of resolution with such a low focal length, but some of the images i see definitely look printable.

[ChrisLX200]

Well the result looks good to me Gav, I guess over-sampling is better than under-sampling which would result in square stars. I usually image at either 1.1 or 1.4 arc-sec/pixl (@1x1 binning without/with the reducer) using the refractor, which probably has the finest definition optics of my scopes. I have not used my 10" SCT much but that produces around 0.3 @ f/10 so I tend to use binning. The optics of my classic LX200 are nowhere near as sharp as the refractor anyway, but your Edge will be better.

Chrish

[swag72]

That's not bad for one nights work Gav!! I'd take the resolution thing with a pinch of salt and actually try a real life comparison to decide what actually works for you. Theories on paper are all very well, but they do not mirror real life and all the stuff that goes with it. It may be a chore, but to see how your system will work best I definitely think it's worth spending some time to take 1x1 bin and give it a fair comparison.

[PhotoGav]

Thanks @adamphillips, Chris & Sara.

Yes, I think that's right, a scientific study needs all the data to draw a conclusion, so I will capture those 1x1 subs and see what it produces... I'll use the colour subs that I have already - they seem to be working fine.

As for the whole resolution thing @adamphillips, I can't profess to totally understand it properly, but you are aiming for a resolution of around 1"/pixel to acheive the (theoretical) optimum capture (the actual figure is around half your seeing resolution, which for most is around 2"/pixel, I think). More or less results in over sampling and under sampling, I haven't quite got my head round that yet and what each actually means... More reading required. Anyway, you certainly aren't trying to get the smallest figure. Sorry to be so vague, but this topic really strains my small brain!

I'll post my findings soon.

[Coco]

Awesome 1 night data set, binning never did anything to speed things up but it did loose resolution.. That's what I found with the 383 & 314 mono cams

[PhotoGav]

Thanks for that Guy. I will bin only one thing... The theory!!!

[lensman57]

Really nice Gav, I think the new gear is working very well here. I'd bin the colour if I were you but reduce the sub length and increase the numbers, like 15X450s but I'd use the Ha as Lum and then blend it in with a layer with blend mode set to Luminosity, this will also serve to control the size and the number of the stars too . Blending it into the Red channel does not always work as it throws the other colours out too much at times.

Regards,

A.G

[PhotoGav]

Thanks AG. I will hopefully get the 1x1 data tonight. I haven't shot any Ha on this target though, just LRGB. I'm thinking that the Elephants Trunk will be my foray into HaRGB imaging soon... Skies willing!

[adamphillips]

very interesting, like I said this is the first time ive ever heard of the recommended resolution thing. my natural instincts just told me that you would be looking for the smallest pixels possible.

[PhotoGav]

Here's the 1x1 bin pic:

I did put this in a new thread, but that has generated absolutely no response at all. That points towards the image being pretty uninteresting! Holds head in hands and ponders over future of edgeHD 800........

[theo]

Im sure that's not the reason Gav...its a very fine image and nicely processed.

Pete

[PhotoGav]

Thank you Pete! Virtual hug felt!

[swag72]

As a comparison Gav it may be easier to see a difference if you look at your 1x1 luminance stack and your 2x2 luminance stack then stretch them the same and crop into an interesting area with detail. Present them side by side and you will be able to easily see if there's a difference. I think that the difference could be very subtle and would perhaps get lost in a large image.

[PhotoGav]

You're right, that would be the most efficient way to compare - I will prepare that when I can. In the meantime, I think that 1x1 is winning pretty easily. The quality is certainly no worse and I think better than 2x2 and the file size is much bigger. I am now in that position of not sure about the scope and whether it is a really useful imaging choice. I will give it a few more chances and see how things progress.

[swag72]

You mention in your OP that you wanted to see what was doable in one night - A quickie so to speak () Well this certainly stands up as a fine attempt. Having recently been imaging at f10, don't underestimate just how much data this scope will need. There are no shortcuts at this speed, no one night master pieces - It needs data and lots of it in my experience.

Perhaps you are not prepared for the amount of data that this scope will actually need in the UK weather.

Lets put it this way - using my Tak at f3.9, in order to get the same amount of detail at f10, I will need approximately 8x the amount of data. If you think that I will produce a mono Ha image from 7 hours worth of data (rarely less) that means that to get an equivalent amount of signal etc I would be looking at 7x8 hours ................... That sort of data is not for the faint hearted and neither is it for one nighters.

Of course this can be negated a bit by processing - But it's not easy.

[PhotoGav]

Very good thoughts and advice Sara, thank you. Yes, perhaps it's more a question of matching scope to conditions / time available. Though would increased signal actually improve sharpness?

My plan is to give The Crescent a go with the reducer on and to collect a goodly amount of data. I'll see how it turns out and form an opinion then.