M31 on SW 80 Evostar ED + 0,85 SW FF/Reducer + L-Pro + Canon 6D

[raf2020]

Hello Guys,

I am beginner and I look for advices what should I change/improve regarding my data collection and how to do it. I live in Bortle class 7 area so I need LPF. Below I will place links to my very first and 2nd sessions on Andromeda raw data:

M31 30-11-2020, 17 lights, 15 flats, 10 darks, 15 bias. ISO 400, 240 sec. Not to much of data and still few needs to be unchecked before registering in DSS as quality is poor because of clouds etc. I have taken flats on manual camera mode on exp. 6". Just white t-shirt and tablet on scope (as per few advices my goal was to keep all data in the middle of histogram). In DSS I have used default settings. I have tried to postprocess stacked material about 5 times. Very poor result. After this session I have cleaned well main scope glass and ff/reducer as dust was visible on photos. Stacked and postprocessed.

M31 05-12-2020, 41 lights, 34 flats on manual settings + 20 flats on AV mode, 20 darks, 30 bias. ISO 400, 150 sec exp. This time I have decided for the same ISO but shorter exposition as on previous I have seen strange "coma" around M31. This time I have asked neighbours to turn off security lights, on previous session it was on. Unfortunately dust is still visible so probably it is on CMOS of camera. LPF looks absolutely clean, same as main scope glass and ff/reducer. Tracking was great, usually error in PHD2 was around 50". This time I have took my flats in two different ways. First on camera manual mode, like the last time, I just tried to keep all data in the middle of histogram and I have reached it on 6" exp. Second way was same ISO on AV mode. Camera has picked 5sec. exp automatically but data on histogram is on its right side. No idea which one is better but tbh I see no difference between them after stacking. Another thing is that I see a lot of strange red, green, blue dots and lines after stacking on default DSS settings. I have tried recommended settings, even in various mixes, but general result was even worse. Stacked and postprocessed.

Next time I am going to try with ISO 800, 150sec exposure and I will try take around 60 lights. If someone experienced would like to take a deeper look on my raw data and give me some feedback/tips I will appreciate it a lot.

Cheers!

Edited December 7, 2020 by raf2020

[mackiedlm]

Hi

I'm surprised you have not had some help with this so let me see what I can do. But please understand I am not far off beginner myself and am certainly not the expert help you are looking for.

 

Anyway, I took your second set of data and ran it through Astro Pixel Processor for initial calibration and stacking. I highly recommend APP over DSS because I find it easier to use, it gives me much more consistent results and its easier to view your masters than in DSS. It will also give you a good gradient removal, colour calibration and take you 90% of the way to finished image.The following frames have had a moderate stretch (I found APPs stretching really useful too until I moved to Pixinsisght) applied so we can see whats going on.

1. Single uncalibrated light

2. Single flat

3 Single calibrated light (strong stretch to see whats there)

So we can see right away that there is something strange going on around the border of the image in both lights and flats and while it calibrates out a bit its far from good. I've never seen this before but maybe some of the other guys on here may have?? its not normal vignetting which is circular. I see you are using a Full frame camera and a full frame clip in CLS filter. My guess would be its something to do with that and I'd suggest you fire off a few flats without the clip in filter in place (save as JPG then you can view them easily). If that nasty border is not there on these new flats then thats whats causing it. If  its still there then hopefully someone else can help cause I have no more ideas!!

4. stacked image

Now that was after I rejected the worst 10% of lights - i did not have time to review them and the fist stack I did had nasty trails etc suggesting there were some bad frames. So what have we got? Well that nasty border looks even nastier now. And the black dust motes. The border I've addressed above. The dust motes are there in the lights (not easy to see but they are there), but not on the flats and thats why they have not calibrated out. I dont know how thats happened but it would be suggestive of the image train being changed/moved /reassembled between lights and flats (did you by any chance do the flats before the lights?)

BUT, on the positive side you have some nice data in there, Star shapes are not too bad in terms of tracking but you've got a fair bit of coma at the corners. That could suggest you need to check the backfocus distance or it could just be because of your big sensor (I dont know as I only use APSC)

Heres a very very quick hack job at processing - cropped in hard to get rid of the border issue.

This is just to show you that you have data worth working with. With more time than I had to do this tonight it could become a pretty reasonable first image (starting with raising the black point a bit!)

 

On your own processed image the red and blue dots are hot pixels which have not been removed by dss in calibration. Someone else needs to tell you how to handle that in DSS. (Get APP!!!)

If you want the masters and the full files of the above PM me and I'll get them to you.

Stick at it, I tell people that AP is the most difficult and complex thing I've ever done but it is worth it when you get somewhere. Don't be disheartened by your first effort , there is good data there - you just need to learn how to pull it out and resolve the problems we can see above.

 

Good luck

 

David

Edited December 8, 2020 by mackiedlm

[Phillyo]

I don't think your flats are working correctly. They should be removing the dust from the image and it looks like they aren't. Can you explain how you're collecting the flats and the bias frames? Also, I wouldn't bother taking dark frames with a DSLR. They're only really useful for a set point controlled cooled astro camera; they can add more noise than remove it from none cooled cameras.

Phil

[raf2020]

Sorry Guys for late reply, busy week at work then Xmass tidies on weekend 0o

@mackiedlm thank you very much for you effort looking into my RAWs. This aberration close to the borders is caused by LPF. Without it photos look ok but because of my Bortle class 7 area I am not able to take any longer exposures without it. So I need to choose between 90 sec exposures without LPF or longer with it, but then I need to crop a lot my photos. In general I think I will stick with DSS to get maximum from it (now still I work only on default settings) then maybe I will try another solution so I will have full background picture to compare. Regarding flats I have took it straight away after lights, darks and bias. I have used manual mode to take it. I just looked on histogram to make sure data is on the middle of the graph. On manual mode because I have spotted that on AV and ISO lower than 800 data is not presented in the middle oh histogram what was recommended by a lot of users.

edit: another experienced user has convinced me to try with APP. I will go with trial and see what I can get from it.

Andromeda Galaxy M31 08-12-2020 ISO 800 90sek without LPF 60 lights, 20 darks, 25 bias, 25 flats (AV mode). If you would like to take a look into it as well to see the difference - would be great, thank you.  

@Phillyo in session pasted above dust is visible also od flats. For ISO 800 I have taken it on AV camera mode. Regarding darks - most of the users recommending to take it if you have a time. Afaik darks adding noise so it is recommended to take minimum 30. Please note I am very beginner. 

Thank you Guys again for replies!

Edited December 14, 2020 by raf2020

[mackiedlm]

On 12/12/2020 at 22:05, Phillyo said:

Also, I wouldn't bother taking dark frames with a DSLR. They're only really useful for a set point controlled cooled astro camera; they can add more noise than remove it from none cooled cameras.

Phil

This is one of those pieces of "recieved wisdom" which is handed down as undisputed fact to all new DSLR imagers - myself included. I dont think it takes into account several things;

1. Many DSLRs (most canons which is what the OP has) provide a temperature reading for each frame
2. With this, a bit of ingenuity, and a wife who turns a blind eye at a camera clicking away in the fridge/freezer it is fairly easy to collect reasonably temperature matched dark frames.
3. Some astro processing software will now scale darks. APP will scale for time, temperature and ISO (I confirmed that with Mabula some time ago) although I usually match time and ISO.

Below are two tight crops taken from one of my own M31 images. Each is a SINGLE frame (the same frame) with exactly the same stretch applied in APP. The first is uncalibrated. The second is calibrated with a master dark and a master bias (which allows for the scaling)

To my eye the calibration has given a positive benefit and certainly has not added any noise..

First with out cal

Second with Cal

So, perhaps, at least with Canons and APP, there may be value in using darks with uncoolled DSLR's.

Just my own beginners assessment/opinion.

Edited December 15, 2020 by mackiedlm

[Phillyo]

Interesting comparison and I agree that the second one looks better than the first, there are always exceptions to the rule. That being said, if someone just snaps 50 darks at the end of the nights imaging and uses them as darks I don't think it'll make much of a positive effect during calibration. Throwing it in the fridge is a good idea, or a freezer to get even colder but how many amateurs go to that trouble and is it well documented that that's a good thing to do? 

There's also the option of using in camera noise reduction whilst imaging which can work well. The camera captures a dark frame and subtracts it from the light raw frame directly after each image is taken. That can work well if you're imaging when it is fairly cold outside, if it's warm then it won't make much difference. 

Hot pixel removal and bias removal is very important for DSLRs. It would be interesting to see your comparison above done with just dark frames and not bias to see what difference dark frames alone make to a DSLR image?

Thanks for the comparison. Very interesting to see.

[mackiedlm]

2 hours ago, Phillyo said:

Interesting comparison and I agree that the second one looks better than the first, there are always exceptions to the rule. That being said, if someone just snaps 50 darks at the end of the nights imaging and uses them as darks I don't think it'll make much of a positive effect during calibration. Throwing it in the fridge is a good idea, or a freezer to get even colder but how many amateurs go to that trouble and is it well documented that that's a good thing to do? 

There's also the option of using in camera noise reduction whilst imaging which can work well. The camera captures a dark frame and subtracts it from the light raw frame directly after each image is taken. That can work well if you're imaging when it is fairly cold outside, if it's warm then it won't make much difference. 

Hot pixel removal and bias removal is very important for DSLRs. It would be interesting to see your comparison above done with just dark frames and not bias to see what difference dark frames alone make to a DSLR image?

Thanks for the comparison. Very interesting to see.

if someone just snaps 50 darks at the end of the nights imaging and uses them as darks I don't think it'll make much of a positive effect during calibration ..... I think that if the ambient temperature remains close to what it was during the lights, (as shown by the camera sensor temperature in the data) and scaling is used, it will do just as good a job as above.

or a freezer to get even colder but how many amateurs go to that trouble I'd say that any amateur who wanted to get decent DSLR calibration and who was told this would help and what the limitations were (Rather than being told darks are no use with a DSLR) would give it a go- lets face it - we do a lot more crazy stuff than that for the sake of AP!! I spent one miserable rainy week-end at it and not had to do it again, yet! and is it well documented that that's a good thing to do - the point of the fridge/freezer is to replicate the ambient temperature to which the camera is exposed during image capture anyway. How could it be a bad thing? I do stick it in a ziplock to keep it dry.

I cant talk to in camera ND as I've never used it

It would be interesting to see your comparison above done with just dark frames and not bias to see what difference dark frames alone make to a DSLR image? Excellent question! Below is just that - same image, same stretch, almost same crop, - calibrated in APP with the same masterdark but no bias used so no scaling. No real difference in the noise (or hot pixels etc) but this masterdark was within a degree or two of the lights anyway. there is a difference in the star colour and I dont really understand why not having the bias would do that.

 

Edited December 15, 2020 by mackiedlm

[mackiedlm]

On 14/12/2020 at 21:10, raf2020 said:

Sorry Guys for late reply, busy week at work then Xmass tidies on weekend 0o

@mackiedlm thank you very much for you effort looking into my RAWs. This aberration close to the borders is caused by LPF. Without it photos look ok but because of my Bortle class 7 area I am not able to take any longer exposures without it. So I need to choose between 90 sec exposures without LPF or longer with it, but then I need to crop a lot my photos. In general I think I will stick with DSS to get maximum from it (now still I work only on default settings) then maybe I will try another solution so I will have full background picture to compare. Regarding flats I have took it straight away after lights, darks and bias. I have used manual mode to take it. I just looked on histogram to make sure data is on the middle of the graph. On manual mode because I have spotted that on AV and ISO lower than 800 data is not presented in the middle oh histogram what was recommended by a lot of users.

edit: another experienced user has convinced me to try with APP. I will go with trial and see what I can get from it.

Andromeda Galaxy M31 08-12-2020 ISO 800 90sek without LPF 60 lights, 20 darks, 25 bias, 25 flats (AV mode). If you would like to take a look into it as well to see the difference - would be great, thank you.  

@Phillyo in session pasted above dust is visible also od flats. For ISO 800 I have taken it on AV camera mode. Regarding darks - most of the users recommending to take it if you have a time. Afaik darks adding noise so it is recommended to take minimum 30. Please note I am very beginner. 

Thank you Guys again for replies!

So I had a chance to throw this new set through APP.

First an uncalibrated single frame

You can see that you have lost that square border by removing the clip in. In exchange you have normal vignetting - especially bad cause you are a full frame camera. Still quite a few dust bunnies etc.

Then calibrated with your darks, flats and bias - this is a high stretch so we can see what we have.

And there you have a good calibration, dust bunnies and vignetting removed very well, hot pixels gone and some level of noise reduction. There is a gradient (probably your LP) remaining bottom right to top left but thats easily removed either in APP, with DBE in PI or with Gradient Exterminator in PS.

Unfortunately your stars have gone to ****. First off I think your focus is soft (do you use a Bahtinov mask?) Then look out from the centre and you see mishaped stars - elongated streaks in top right, classic coma in bottom left. Because they are not all showing the same orientation of mis-shape, I think what you may have is some kind of tilt. Knowing that your scope is the 80ED (same as mine) and your camera is quite heavy Id suggest that you double check that you dont have slump in the focus tube, make sure the screw under the focuser is tight once you are in focus. Also make sure that the camera is tightly connected attached and perpendicular to the scope. Is your FR connected directly to the focuser tube or are you using a nosepiece on the reducer held in the standard eyepiece holder - that would be a classic location for slump/tilt. If its the Skywatcher FF/FR then remove the eyepiece holder and screw the FR straight on to the focuser tube. Did you adjust the t-ring at all. You did not have this in your earlier subs with the clip in filter so its not your sensor. Anyway, these stars are a real pity because you were heading for a good image there. So get that sorted now and you'll be in good shape.

 

As indication of where you can easly get to, here is a link to my M31 using very similar gear to you, SW 80 ED/0.85FR?EOS700D - Im bortle 5 or so.

david

Edited December 16, 2020 by mackiedlm

[raf2020]

David thank you very much for your effort. Another colleague has told me the same - this material is pretty ok'ish but you have missed the focus. I think I really need to take a closer look to my focuser as I have some problems with it. I use Bathinov mask, then while I am locking focus it's moving a bit so just before lock it need to be a bit out of focus to take into account this slight move. I have FF screwed to nosepiece and it's placed in focuser. Then on other side Canon t-ring and DSLR. I think I can feel small loose between t-ring and dslr. What do you mean by adjusting of t-ring?

Problem with my focuser (Crayford) is that if scope is in vertical position and if I push a bit (rly, rly tiny bit) locker screw, focuser is not moving at all. But after I loose it just by another tiny bit, focuser tube going down under load of DSLR. Not sure I can adjust it with those small screws on the bottom of focuser but I will try.  

Another problem probably is that usually I set the focus on Polaris, just after PA and then I am not adjusting in after GoTo to the object. In case of this material I did even worse. If I remember well I have rotated DSLR in focuser after GoTo object to catch better FOV without adjusting of focus after. 

Your M31 is brilliant  And I have switched to APP as per your advise but have some problems with it. I can not adjust RAM about and I work only on default 2GB so processing takes forever. Support look for the solution for me.

Clear skies David!