M104 Galaxy

[Swanssm]

The spectacular edge on Galaxy M104. Not sure what is more special. The Galaxy itself, the surrounding sky, or the fact that Vesto Slipher calculated in 1912 that it is moving away from us at 1000 Klms per second. This 81' x 54' field of view shows it against the background of a magnificent jet black dark australian sky.

60mins of HA (3nm)
ASA 16" Astrograph FLI PL11002 CCD

http://www.aartscope.com/index_files/g11-m104_sml_ha.jpg

Enjoy!!!!
www.aartscope.com

[Liz]

THAT is a beautiful image, very sharp .... and yes, it is a spectacular galaxy!! :thumbsup::thumbsup:

[dannat]

nice and sharp/pinpoint , the centre is very bright -takes away a little of the detail

[lesbehrens]

very very nice:eyepop:

[strongmanmike]

Looks quite promising :thumbsup:

Is that the 100% frame?...it's hard to critically analyse the image quality at that image size though. Was it really taken through a 3nm Halpha filter or was that a typo? Skies are always black and stars tighter through a Halpha filter of course but if that was indeed taken through a 3nm Ha filter it speaks volumes for the fast F3.6 and 16" of aperture!

I understand the 16" ASA had some flexure problems..? This was on U-Tube:

http://www.youtube.com/watch?v=iI0EyWufKPM

How did this flexure affect performance? Has this issue been sorted now and if so how was it acomplished? That FLI ProLine and filterwheel do look awesome on that big sexy red and black OTA!:eyepop:

Unfortunately I (and several others) got a lemon from ASA so I hope you guys at AART have had more luck? Would be nice to hear that ASA have sorted some of the problems they had in their early shipments as on paper they are the ultimate imaging scope in my opinion. Not sure on the new double vein spider though, I think the standard spider induced more natural looking diffraction spikes..?

I actually sent you guys an email a while back about these things but I never heard from you..?

Mike

[Swanssm]

Thanks all. Yes Mike we are fine tuning and nearly there. The hig res version is here.

http://www.aartscope.com/index_files/g11-m104astroswanny_ha.jpg

This is still a pretty raw image just a quick darks/flat calibration.

[gregbradley]

An unusual choice to image a galaxy with a 3nm Ha filter.
Be interesting to see a luminance image.

A fast astrograph is not really the right tool for galaxy imaging and your
16 inch aperture and F3.6 is better spent on something more widefield where it will show off its capabilities better. Something like Rho Ophiuchi or NGC6188.

The right tool for the right job. Also your PL11002 is a wide field imager at that focal length and a small chipped camera would get you some more magnification going.

To use your ASA 16 for a galaxy you probably need a Powermate 2 inch 4X or a Baader FFC or something to get extra focal length.

Stars look squared. Is that exaggerated by the 3nm Ha and less of a problem with usual luminance?

If not perhaps a different spider vein is needed one of those s shaped ones may be the go.

Greg.

[strongmanmike]

Can't say I agree with you on this Greg :shrug:

A 16" F3.6 astrograph provides plenty of focal length actually at nearly 1500mm and provides arguably a perfect image scale of about 1.2"/pixel with the PL11002. The notion of feeling the neccesity for a narrower FOV simply to frame galaxies or other smaller deep sky objects better at the expense of FOV is a bit strange IMO, this is what cropping (and upscaling if you want to) is for. It 's much better to have the FOV first for all those larger objects and then crop out a bit when you have a smaller object in the field than to be forced to mozaic to cover larger objects.

My deep Cen A image data set at 1140mm FL is a good example of this wonderful versatility IMO.

From this: http://www.pbase.com/strongmanmike2002/image/97061387/original

To this:
http://www.pbase.com/strongmanmike2002/image/108788806/original

All with the same data set and that was at 1.6"/pixel with just a 6" scope too so 1.3"/pixel with 16" of aperture and you have the potential to produce amazing results and FAST (F3.6!) on a very wide range of objects.

Adding a barlow only slows the scope down too, nup I'd say once (if) they have it sorted the 16" F3.6 is just about the perfect imaging scope :thumbsup:

Before I knew of the heartache and dissapointment the 12" ASA was going to be :scared:...I would have gone for a 16" ASA too... if I had something bigger than the NJP to mount it on of course (oh and the extra money :rolleyes:)...then again, it too may have been a lemon just like the 12" ASA was :mad2: so it's all irrelavent really...............:sadeyes:

Mike

[gregbradley]

Interesting point you bring up Mike.

Say for example a 16 inch RCOS with 3 metres of focal length or more
would give you a very zoomed in image without much of FOV perhaps
but from what I have seen in images the detail is far greater than say a crop of a fast scope to get the same image scale.

I suppose the proof would be in a direct comparison of a cropped ASA 16 inch image versus a 16 inch RCOS image of the same object and both using a 11002 chipped camera.

Cropped images to get a zoom in factor are not as detailed as long focal length in the first place in my experience but I may not be comparing them correctly. As you point out the arc seconds/pixel comes into it here. Though if you are oversampled as this shot probably is, you will simply lose some of your F3.6 gain and it will take longer to get the signal right?

Greg.

[strongmanmike]

Well, Martin Pugh has done what I consider to be the best narrow field Centaurus A image in the World using a scope under 20". Martin used a 12.5" RCOS at 2850mm FL with a KAI11002 chip and AOL (adaptive optics) riding on a PME. LRGB image - exposure time over 6 hours.

Here I have compared Martins superb image directly to my cropped RGB image taken at just 1140mm FL with a 6" scope and then croped out of the much wider field of view. RGB image only - exposure time 4.5 hours

Not sure what Martins processing included but I certainly used no deconvolution on my image. It is clear that Martins image is superior but the difference isn't "that" great really...?

http://www.pbase.com/strongmanmike2002/image/113745498/original

To have the luxury of also being able to get the following full fame image from the same data set with the 1140mm FL scope means a wide FOV with adequate focal length is the way to go to maximise imaging versatility without the need for multiple outfits/equipment

http://www.pbase.com/strongmanmike2002/image/97061387/original

Just my thoughts of course :P and Martin if you are reading this I hope you are ok with the comparison - your image is better in this particular comparison for sure, I am just trying to make a case to answer the age old quest for "what is the perfect imaging scope" :thumbsup:

Mike

[gregbradley]

Hi Mike,

You make a strong case there.

Although I think sampling has a lot to do with it. If your camera is well
matched to the scope to get your 1.6 arc sec/pixel type sampling then it
should work out similarly with the biggest aperture usually winning out
especially on dimmer objects.

The comparison may not be so true for dimmer objects do you think?

Greg.

[strongmanmike]

OK, getting back to your original idea that a 16" F3.6 scope wasn't well suited to imaging galaxies, well actually it is, becasue it has enough FL and thus image scale to produce impressively detailed images "and" a nice wide FOV when it is really needed. How you crop around and thus "frame" the galaxy is then totally up to the imager which is better than having to be limited to a narrow FOV (ala Fred V :P) and the restrictions that can place on your framing, think of NGC 253, M33, NGC 55 etc... not to mention all the great medium sized nebulae out there. Here is another case in point, take my recent NGC 3603 & NGC 3576 image set ie. several images were possible, from wide field to close in detailed crop, simply becasue I had the FOV "and" adequate (but not too much) FL...it's a pretty simple concept really :)

In a nut shell with a standard 9 micron 35mm sized sensor a FL anywhere between 1000mm and 1500mm is pretty well perfect...IMO of course

If you were imaging from a location with sub arc sec seeing (very rare places) well then, that's another ballgame altogether, that's when the <1"/pixel image scale would come into its own :thumbsup:

Mike

[gregbradley]

Hi Mike,

I posted this question on another group and got a few interesting replies.

Firstly unlike sound where it is oversampled 2.2X (CDs) and that is acceptable, Stan Moore contends that oversampling with light needs to be 3.3X. Now that is an educated theory but it does seem to work out.

So resolution is limited by your FWHM of stars and thus your seeing but if you assume an average of 2.0 (not sure how real that is) and divide by your 3.3 you get around .6 arc seconds/pixel as the theoretical limit of your resolution for that area.

So the example below was 1.93 arc seconds/pixel so there is potentially 3X worth of resolution left to obtain under good seeing conditions.

This oversampling factor probably is the difference between your image and Martins. As we know that last little bit of difference is what the thousands of dollars pays for. Also your scope is an AP with super high strehl and the ability to cut through seeing more easily than a bigger scope.

So in the end for optimum you would still need a long focal length scope of the same aperture to get best resolution but as you well point out excellent results can be gotten by cropping. Depends on how far you want to take it to get that last bit of detail.

I have seen Rolanf Christen use cropping to great effect from his refractor shots as well.

Greg.

[Martin Pugh]

Hi Mike and thanks for the kind words regarding this image.

Comparisons are always a good exercise and yours is a fine example. For me it illustrates a rather simple conclusion - resolution yields finer detail - of course it does! On the other hand, the versatility you refer to with a fine refractor and a a large chip camera, such as your set up, is best exploited by creating composite images.

I have to ask a question though. On your Cent A deep-field page, you link to a full-size image here:
http://www.pbase.com/strongmanmike2002/image/97573250/original

When posting this image, did you resize this in any way? I ask, because the actual galaxy is somewhat smaller than the comparision you posted. Was your comparison post just a crop or upscaled to match my image for comparison purposes?

Regardless, back to the main aim of the thread. I have to agree, that to use a 16" Astrograph at f3.6 to image galaxies would really only be useful to capture the Ha regions for blending into traditional RGB e.g. face on spirals. Once again, another way to create a stunning composite image.

good thread though...and its always nice to see the nyquist theorem popping up again when at the end of the day, there are a dozen other factors that come into play in any one single imaging session that will affect the resultant image.

When I was preparing my talk for Lostock, one of my opening slides was entitled 'The 10 Commandments of Astrophotography' and just about all of them would affect the final image in one way or another even before you start to think about matching optics to pixel size.

cheers
Martin

[Octane]

I thought the aim of the thread was to discuss/critique the original poster's image.

Perhaps the mods can crop the equipment discussion and post it to another thread in the relevant forum.

I feel sorry for the OP having had their thread hijacked.

Regards,
Humayun

[strongmanmike]

Oi Humi, chill man, I am sure Peter (Swannsm) would be quite happy about this discussion as I am trying to support the usefullness of his comercial scope that took, what was by his own admission, very much a test image mate :thumbsup:

The 16" will be available for use by the wider comunity soon and I recon it will be a purler once some initial teething problems are sorted.

Mike

[strongmanmike]

Yes that link is to the raw image size just converted to a jpeg, if you right click on the image you will note under properties that the pixel dimensions are almost that of the KAI11002, "almost" only becasue I did crop the non registered edges off the final combine (125 exposures over 3 nights with meridian crosses remember).

Yes I carefully coregistered my shot to match yours, so Astroart did the matching :thumbsup:. I think the comparison is good too, good because it shows just how fine the details are you have captured and good because it shows that, as far as fine resolution goes, a high quality medium FL 6" APO on a nice mount (but no PEC) can hang with a research quality 12.5" long focal length RC using adaptive optics on a fully tricked PME....without being (too) embarrased :P.

My point was that the nearly 1500mm FL of the 16" F3.6 is more than adequate to produce high resolution images of galaxies with a 9 micron pixel chip and if it is a big chip (11002 or 16803) then the focal length is still low enough to fit plenty of other things in too when needed. I would give my eye teeth for a (well made :rolleyes:) 16" F3.6 Astrograph - perhaps the perfect still amateur sized imaging scope in my opinion..? :).

Mike

[Peter Ward]

Bah! Humbug !

You are wussing out at less than 2500mm IMHO :P

[strongmanmike]

Ok Peter Vanderhaven :lol:

Yes would be ok if you have a 60mm X 60mm chip and correction to match...christ imagine that:eyepop:actually our Starfires could handle such a chip, not sure about your RC though :whistle:

Mike

[Bassnut]

One thing ive never understood. Bear with me, this gets to a puzzle to the need for megapixel sensors for long FL, other than oversampling, and convienient framing as Mike has picked on me for.

It seems intuative that a larger chip with the same pixel size as a smaller chip just allows a bigger FOV, and with croping to the same FOV, would render essentially the same image quality generally.

But no, except for a few fine examples, generally, croping, or zooming in it seems does not mean the same sort of image you would get with a small chip with the same FOV as the crop. I have seen many many examples from the best that just simply dont cut it, zoomed in.

So I wonder why?. When processing a wider view, colour balance and dynamic range especially must suit the image as a whole, so the core of an object is often overexposed (ie, individual parts of the subject are not processed individually, which I guess would be very difficult to blend in with an overall pleasing result). Or, maybe, the lower QE of megapixel cams gives a different result alone.

Im not convinced at all that a bigger chip always gives a subjectively "better" or even equal :eyepop:result (with the same image scale) at long focal lengths, despite the advantages of convienient framing. Croping on a small object is just a waste of money spent on the cam.

Mike, I like narrow field high QE and cheaper imaging for the same or arguably better results than a croped megacam. Bite me :D.

[Bassnut]

AND, Peter, am I right or wrong, OTA optical performance is generally measured by image quality over the whole illumination circle (I might have the terminology wrong, sorry), which is important with a large chip that is fully illuminated. If a small chip only captures a small proportion of a large aperture OTA illumination, then edge performance is far less critical?.

[multiweb]

:lol: I see your Schwartz is as big as mine..... thread's gone waayyy off topic from M104. :lol:

[Bassnut]

Opps, OK, yes, thats a nice M104 too.

Oh, and welcome to GRAS Peter :D

[strongmanmike]

....? :stupid::rofl:

I think some confusion has crept in here me hearties and I cana be bothered typing anymore (said in Scottish accent).. suffice to say.....I wish I could just image more often :( :sad::bashcomp:

Mike

[Peter Ward]

Que? You asking me?

I suppose by definition Strehl looks at the whole wavefront.

But this is a bit moot, as is field curvature, if you just use the say centre 1/3 of the available field. That said, I suspect visual observers would not be happy with an optic that was only sharp in the centre 30 degrees of a 100 degree Ethos field!

But with imaging provided your optic pumps most of the available flux into the airy disks at the centre third..or what ever...and if your CCD is too small and can't "see" the rest you will have a high resolution solution of sorts. But wait there's more.

If you happen to image from within the atmosphere you are sky limited.

Lets say at 500mm you can get 1000 counts per pixel from the sky, plus 100 counts from the star sampled at 4 arc sec per 1 pixel. Total flux 1100 but less than 1/10 is signal

Let's now increase the FL to 2000mm . Our sky contribution over the same pixel has reduced by 4x. The object flux remains the same, Total flux 350, but now over 1/3 is signal!

Having more signal vs noise is pretty handy....de-convolution works better
plus when noise is pushed down to trivial levels the signal clearly and cleanly resolves above it.

Yes, narrow field rules :)

[Bassnut]

Fair enough, but I think I do have a point anyway (albiet not expressed very well), despite the math and $, there is more than meets the eye (no, begum, its what meets the eye :whistle:). Horses for courses. When it comes to Narrowfield, megapixels and mega $ is not always the obvious answer.

[Bassnut]

Thanks Peter. high QE will do me over low QE big FOV anyday......... for narrow field anyway :D

[Peter Ward]

Well....You can sort of have your cake and eat it too...I bit the bullet (instead of the cake) and sent off the deposit for a 16803 based camera yesterday...... Good QE and a field that makes 2" accessories way too small :)

[Bassnut]

I guess you mean the $11,875 Sbig model (well, less for you, natch ;)). At 60% QE, no big shakes, but huge FOV (what happened to NF rules :shrug:), and fancy guiding. and oh those filters, there goes the one-to-be-obeyed funding permission (for a piece of coloured glass, you have to be KIDDING :lol:)

[Peter Ward]

Yup about $19K in Pacific Pesos with the fancy coloured glass wheel......:confuse3:
second thoughts I'm probably mad given the weather to date!