Jump to content

740427863_Terminatorchallenge.jpg.2f4cb93182b2ce715fac5aa75b0503c8.jpg

First DSO attempts (sugestions please!)


Recommended Posts

Hi there - Having got a Bahtinov mask and 2" extender, I was ready to move on moon / webcam imaging and have a crack at attaching my unmodified 300D to my C100ED f9 / CG5 mount (with Synscan EQ5 upgrade). Sodium streetlights, hedge/trees and house severely limit viewing at home (I'm trying to find an accessible dark site), and of course this probably isn't exactly a great time of the year to be trying this, but I did have a crack at M39 and M57 last week...

As I'm unguided, the subs are only 30s each @ ISO 1600 (with a 30s "rest" in between) and 4 x 30s darks. I did take 5 but somehow lost 1 somewhere(!) - I was also trying out DSLRFocus for the 1st time which somehow caused me to "lose" some frames. I know the camera took them, but I'm damned if I could find where they got stored, because they weren't on the camera and I couldn't find them on the laptop either... Still...

Post exposure, I stacked in DSS and then "default" processed using Astronomy Tools / Elements. There's no flats or bias frames (not sure how to do these yet), but these are my first crack at DSO's.

Firstly M39:

M39Stacked20100531edit.jpg

Although I set the remote to take 10x30 (and it did take them) I only ended up with 2 subs(?), so I know this isn't ideal... But I'm not really sure how many I should be taking anyway (the shopping list does include a guide camera, but that's a little way off). I did subtract the 4 darks anyway, but there seems to be a lot of noise in here which obviously I don't want. Is this a case of not enough subs / darks, too high ISO and/or lack of "elements" knowledge in removing noise?

Secondly M57:

M57Stacked20100531edit.jpg

My first thought on this is how small it is... and lacking in any colour, but at least the noise is better. Again, would I be right in thinking I need a shedload more subs/darks / lower ISO and also perhaps a good barlow? The one I currently have is only a standard Skywatcher 2x and I feel as though I need either a good 2x (ultima?) or possibly a 3x... or 4x... or 5x?

Any criticism would be very gratefully received - I'm hoping to use the summer months to try and learn as much as I can before Orion appears and in particular I really want to have a go at some galaxies and other DSO's when the skies start to turn dark again. I also have an alterior motive here - If I can get some reasonable images now (hence the cliched diffraction spikes!), my finance manager (aka wife) may allow me a bigger budget to indulge myself...!

Link to comment
Share on other sites

  • Replies 52
  • Created
  • Last Reply

Top Posters In This Topic

Not really qualified to pass criticism on the pics (but I like them).

Darks: there is a thread somewhere on the forum that crunches the numbers, but the result was that 20 is the optimum number. If you are obsessive like me, as they have to be at the same temperature as the lights and the temp varies through a session, you take ten before and ten after. I suspect that others will (quite rightly) say this is unnecessary and a set of 20 are fine.

Bias: Easiest of the lot. With the lens cap on, in a dark room (I put it in an old 'changing bag' from my film processing days) take a series of exposures (20 again) at the same ISO at minimum exposure. The resultant 'Bias masterframe' can be used on all sets of exposures for that ISO.

Lights: lots more. DSS only uses 80% on default settings. One advantage of short exposures is that you can get a lot in a short time. At 1 min per frame (30s + 30s gap) you can get 60 in an hour.

HTH

Edited by Demonperformer
Link to comment
Share on other sites

Hi bulldog - I tried to rework the saturation, but without it looking posterized, there's not really too much else I can do to it... but thanks very much for the suggestion as I'd never played with that before. However, there's simply not enough data in there to pull anything out...

DP - You're obviously totally right - I need a shedload more subs. I'll have a crack as you've suggested. I guess I could even try for 120 over 2 hours with 30s exposures...

I had read somewhere that 20 darks was acknowledged as the standard - I think I only took 10 as I had so few subs (which thinking about it now, is totally illogical!). It looks like being a fair night tonight though... on top of which it's 1/2 term and wife/daughter are both away, so I can stay up until dawn if I wish!

Also thanks for info on taking Bias frames - I had no idea it was so simple. I can obviously do that anytime, save them somewhere safe and then apply them in DSS each time. Thanks!

Link to comment
Share on other sites

I think at ISO1600 you'll be getting a lot of noise in the images compared to say ISO 400. Of course, if you drop back down in ISO, you'll also get less gain and so your 30 sec subs will not be as bright. Much less noise though.

Have you done a quick visual test to see what the noise is like on a 30 sec sub at the different ISO settings? I know on my 450D that ISO1600 is very noisy and ISO800 is about the most I'll use. I prefer ISO400 though.

Dave

Link to comment
Share on other sites

Thanks both - Yes, I was a bit concerned about the ISO setting when I took them, but I was also conscious of the fact that the scope is an f9, my polar alignment might be a bit out (I haven't fully got my head around drift aligning yet, so try and set according to polarscope app) and therefore wanted to keep the time down as much as poss... and I thought I might get away with it... but I didn't!

I tried 800 last night (even though I'm still a little uncomfortable) and took 60x30s subs, 20x30s darks and 20 bias. My first "cheap and cheerful" processing seems to have produced some MUCH better results (M81, M3 and Double Cluster) although the light gave out on me a bit on the last one. However, I think what I've actually done is proven how little I also know of Elements, so I'll obviously need to learn that too...

I'll work on them a bit more (especially M81 - I can just about make out the spiral and I really hope I can bring it out a bit more) and post those to see if anyone thinks they're any better.

Tonight looks like being another cloudless night (fingers crossed), but I think I'll try and just pick one object and take c 120x30s subs at ISO400 and see how well that works (probably M81 again)

(If only I could afford a fairly sensitive autoguider I'd be MUCH happier!)

Link to comment
Share on other sites

Okay... Having taken on boards everyone's suggestions, I had an other go (and just got around to trying to process them... whilst imaging the next batch!). This time I was taking 60x30s subs, 20x30s darks and using 20 ISO800 Bias frames.

M81:

M81Stacked20100602editlowres.jpg

(I had a bit of a processing problem with this - I'm sure I could have teased out more if I had sufficient skills in Elements...)

M3:

M3Stacked20100603editlowres.jpg

(I was quite pleased with this one - I'd only ever seen it as a fuzzy before)

The Double Cluster (NGC869):

NGC869Stacked20100603editlowres.jpg

(I ran out of time with this one - It was approaching 3:30am and the sky was beginning to turn blue. I had to really fiddle with this one to get it like this as there was a nasty blue cast across the bottom...)

I learnt a lesson from trying to take these - 3 in one night is simply too much. In future I intend just to stick to one and take as many 30s subs as I can...

Again, I'd welcome any critical suggestions (please) - If I can get better at this (and my wife can see the fruits and is suitably impressed!), she just might let me buy a guide camera...!

Link to comment
Share on other sites

Okay, these will be the last one's for now (wife and daughter coming home).

I set out last night to only image M81 with shedloads of 30s subs... but unfortunately after 165 frames, some stray clouds moved in, so I had to stop. However, as it was only 2am, I didn't want to waste the sky so also had a quick go at M31 (which had just cleared next door's garage). However I could only squeeze in an hour before the sky started to brighten...

M81:

M81Stacked20100603editlowres.jpg

(I think the extra subs have provided better resolution / colour)

M31:

M31Stacked20100603lowres.jpg

Again, I still used ISO800 (I wanted to do a comparison with M81 from the previous night), but I still can't see how I can tease out any further info from M81. It's quite frustrating as I know it's there... and ditto with M31, as if I crank the midrange towards black I can begin to see the structure (God, M31 is HUGE - I think it might actually overflow the frame!).

However, it doesn't look anything like the other images I've seen posted... Maybe I need to try and up the exposure time(?) or simply take hundreds more subs? I think I've seen that others take mulitple exposures of 20 mins or more for the outer detail, but unguided, that's going to be out of my league at the moment.

Link to comment
Share on other sites

This really is the last one...

Looking on Stellarium, I thought I'd have a go at M13, with the intention of taking 2+ hours worth of 30s subs. Unfortunately things didn't go the way I'd planned as after 30 the camera was getting awfully close to the tripod, so I had to abort.

I therefore chose to go back to M57 and put all the advice above to practice (and also advice on taking darks, before, during and after imaging). I can now REALLY see the difference in resolution that taking more subs makes and how much more latitude there is for "pushing" the processing further (I can now even see some colour in here and at pixel level can make out the central star). This was 120x30s subs, 30x30 darks, 20x30s ISO800 bias:

M57Stacked20100605editcopylowres.jpg

Just a thought though - With these smaller objects, would I get much more benefit from using an ED 2x barlow (or maybe something more powerful?). I know it will push up the f ratio as well but it would enable me to see even more detail, then perhaps it might be worth the investment...

Link to comment
Share on other sites

Looks like you're getting a lot better data now. Some of the earlier ones though, particularly M81/82 looked overexposed so that the stars were ok, but the DSOs were essentially whitish blobs with minimal or no detail.

I'm not sure if that was a result of processing post-stacking, or if the stack came back like that.

Also, with your calibration images, you might want to consider taking Flats instead of Bias images, because basically the Darks you are taking also contain the Bias signal. Therefore, when you stack the Darks, you end up with a master Dark that also happens to contain a Bias Signal. Subtracting the Master Dark from your light frames will therefore also subtract the Bias. So you don't need Bias frames or you might end up subtracting the Bias signal twice. Bias frames are only necessary if you have a pile of Dark frames that do not match the exposure times of your main Light frames. They allow the Darks to be "scaled" to match the exposure time of the Light frames.

What is more useful is to start taking some Flats (and Flat Darks to calibrate and produce a master Flat). This will help your stacking software eliminate any dust noise and other uneven-ness in the exposures.

What software are you using to stack with?

David

Link to comment
Share on other sites

Thanks Paul - You're very kind..! I'd really like to get closer in to M57, but I have a feeling that maybe I don't have the right scope for that. Even with a 2x barlow, I'd be working at f18 (and, for that particular object, I still don't think it would be enough) and with working with a 3x (4x?), the number of 30s frames to get a similar resolution would be huge... My wife and I find the open clusters quite fascinating - She's always wanted to see the "stars between the stars" and NGC869 is perfect (although I'm sure it'll be easier for me later in the year)

DAVE...

Yes - Comparing the 1st M57 to the latter, I can see proof positive that the data is MUCH better (thanks to all the advice I'm getting from this forum!) but I too am frustrated with the M81/82 (and M31) images - The frames actually all came back like that with virtually 0 definition of the surrounding area at all. You can JUST begin to make out the surrounding DSO on the latter image, but try as I might I couldn't get anything more out (with my limited Elements skills) without splitting the image.

I thought that the solution to getting definition in these areas may require me to take either shedloads more frames (and blend two exposures using layers - one for the centre and the other for the surrounds) and/or take longer exposures (which I don't think is really an option for me at the moment as I'm unguided...)

For stacking, I use DeepSkyStacker (using the default settings). I did try having a go at drizzling on M57 to see if that would make any difference but my laptop REALLY didn't like it at all!

Thanks also for info about the Bias signal being contained within the Darks - I hadn't actually thought about that until you mentioned it and I can now see the logic...

FLATS - I've read about these, and understand what they are and what they do, but I haven't figured out exactly how to take them yet (well, I took one in daylight once using my camera, but only to check the sensor by photographing a sheet of white paper...). Am I right in thinking that, like darks, these need to be taken before every session? Also, as I don't have a lightbox, what's the simplest way of taking them?

I've read about using the twilit sky pre-session (which sounds simple to me!), but I must admit the other idea I read of putting a white hanky / cloth over the end and then shining a white torch at it doesn't sound too good for night vision to me, and might be a bit tricky when I'm on my own... Or are these something I can take any time during the day?

DARK FLATS - The first I saw of these was when I first opened up DSS. Never having heard of them before, I'm afraid I ignored it(!). However, as a result of your post, I've literally just looked it up so now I know what they are (same as flat, but with scope cap on)

I'm sorry to be so naive about all this - There's just so much to learn (and then of course remember!)

Link to comment
Share on other sites

Ok, so basically I'm also just starting out in imaging. I am about to start doing some serious FLAT taking, but I have a few problems with my mount that I have to sort out first.

The idea with flats is to take them AFTER your imaging session but before you modify anything. So focus position and camera orientation will all be exactly the same. This ensures that dust motes and other debris will all be in the same position in the FLAT as they are in the LIGHT frames. I think I am going to build a simple light box out of foam core panels and white LEDs (there are numerous DIY projects on the net you can find). I've also considered point the scope at a white screen on my computer and doing a FLAT from that.

Once you have a suitable FLAT light source, you're supposed to adjust the exposure time (ideally using a live histogram) so that you end up with roughly 2/3s of the maximum exposure capacity if that makes sense. If you google a bit, there are examples of what good flats look like and also how to take them etc.

Hope this helps,

David

Link to comment
Share on other sites

Thanks Dave, doing this at the end makes sense - I wondered if this was going to be another one of those "10 frames before and 10 after" jobs, although it must be a bit awkward with a light box if you're not supposed to move the camera away from it's imaging position and you've been targetting something half way close to the celestial pole(?)

I'll have a look on the web and see what I can find... And thanks again!

Link to comment
Share on other sites

Hi again Dave - Your comments about there being no definition around the core of M81/M82... I think I may have found the answer after searching the web for other images and noting their exposure times.

I found an image be someone using a similar f9 scope who'd taken 2.5 hrs of Luminance and 1.25 hrs each of RGB. I think this may go quite some way to explaining why, with my 80 mins, I haven't got any detail in the outer areas... Perhaps I'd better wait until Autumn / winter :D

Link to comment
Share on other sites

You can move the camera, or rather point the telescope wherever you like after you take your light frames. What you CAN'T do is change the focus setting or rotate the camera in the focuser.

Some people, for example, have a light box thing on their observatory wall. After taking a series of LIGHTs, they slew back to a position which points directly at the light box and snap off a few FLATs. Then they cover up the scope and snap off the DARKs and FLAT DARKs.

On the M81/M82 images, I don't really know enough about that to comment I'm afraid. But I've read up a lot recently about the theory of image calibration with Darks, Flats, and so forth and I'm making my own software to help me do that. Or at least to experiment with various techniques. It does wonders for one's understanding of how this stuff works...quite enjoying it actually.

Anyway, hope this helps.

Dave

Link to comment
Share on other sites

Oh... My naivety showing again... I imagined everyone having to stand on chairs at the end of star parties all with their lightboxes above their scopes for a few minutes whilst they shot off their flats... which would have looked ridiculous! Thanks for clearing that up...

Programming... Now that's a SERIOUS dark art! I project manage a team of 12 and I can hardly understand a word they say (a lot of stuff about objects, classes, cache server pages and lost focus jargon) - All I have to do is provide them with detailed functional specs but they have my deepest respect as I wouldn't have the faintest idea where to start. May your work be bug-free...

Link to comment
Share on other sites

  • 2 weeks later...

Hi Mike - That's very generous of you (thank you very much!)

I think I've really got the bug for this - It's been DESPERATELY frustrating this week as I've spent the last 2 nights in a hotel in Newcastle and the skies have been totally clear (no scope, no camera, nothing)

On the back of these, I'm really itching for a guide camera now, although I've now having to replace my DSLR as a higher priority. I don't want to to but I'm having to get a new laptop this week and it's coming with Windows 7... and I discovered at the weekend that Canon have decided not to produce drivers for a 300D :icon_scratch:. Hopefully I can bring the weather back down south with me and tonight will be clear at home so I can make the most of the last few days of my 300D :)

Link to comment
Share on other sites

Hi Mike - That's very generous of you (thank you very much!)

I think I've really got the bug for this - It's been DESPERATELY frustrating this week as I've spent the last 2 nights in a hotel in Newcastle and the skies have been totally clear (no scope, no camera, nothing)

On the back of these, I'm really itching for a guide camera now, although I've now having to replace my DSLR as a higher priority. I don't want to to but I'm having to get a new laptop this week and it's coming with Windows 7... and I discovered at the weekend that Canon have decided not to produce drivers for a 300D :icon_scratch:. Hopefully I can bring the weather back down south with me and tonight will be clear at home so I can make the most of the last few days of my 300D :)

I think it is possible to install a vista driver, if you are careful and run the installation program as administrator

Link to comment
Share on other sites

The chance of clear weather is inversely proportional to the aperture of the nearest available scope and directly proportional to the distance to that scope
Oh how true...!

I'm back home now, and it looks as though I DID bring the weather back... I'm trying my first widefield as I type - Haven't a clue what it's going to come out like, but that's the exciting bit!

Thanks for info on seeing if a Vista driver on W7 for the 300D will work - I'll give it a whirl... I've got nothing to lose by trying, and if it does work, I can revert back to plan A and hopefully get a guide camera in the next couple of months (and you'll have my permanent thanks!)

If not, well, I've set my sights on a 40D... I've dreamt of a 5D (full frame sensor) for many years, but I know it's not going to happen!

Link to comment
Share on other sites

Be aware that a lot of the advice on darks/biases is given by people who take long exposures. With lots of 30sec exposures life is somewhat different. If you take, say, 100 30sec light frames, then using only 20 darks will almost certainly *increase* the random noise in your final image.

NigelM

Link to comment
Share on other sites

Hi Nigel - Thanks for mentioning that... I'm aware that 30s in this world is pretty miniscule and that many talk in terms of 15-20 mins. Is this a bit like in "old" 35mm days when film reticulated(?)on long time exposures...?

I'm assuming you're saying that if/when using 30s exposures, I should be taking significantly more Darks / Bias frames... Is there perhaps some kind of guideline for this? (ie n subs at 30s = n/2 darks ot something). I've read that a guideline is 20, but as you say, that may well apply to n minutes exposures. I've also seen some guides that advise it should be a 1-1 correlation...

Also, since OzDave noted in his response (above) that taking Darks also covers the Bias signal as well, I've been neglecting these since (not wishing to remove the bias twice)... I hope this isn't one of those contentious debating points - Darks+Bias vs Darks only?

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.