new CCD

[Oddsocks]

8 hours ago, brrttpaul said:

 I was out last night left the rig set up and took roughly 10 30 minute exposures plus 4 darks bias and flats, ran it through maxim and the result was terrible. I dont know what more i could do,  it is so frustrating when you think you getting somewhere

Without being able to see your images it is impossible to say what is going wrong....

Looking at your set up HEQ 5, even guided, 30 minute exposures seems rather long for this mount. I used an SX H9C on a HEQ 5 PRO for a couple of years and guided was never able to get more than ten minutes max without guiding errors, I even invested in a SX active optics unit for it and local light pollution limited me to fifteen minutes max so I'm suprised you were going for thirty minute expoures unless that is a typo!

With a good guided mount And OSC in light polluted skies I would have thought twenty to thirty ten minute shots with at least fifteen darks and bias would be a good starting place.

After a while I settled for ten minutes max and as many frames as possible on any one target and only moved to mono when I eventuallly upgraded to a better mount and narrow band filters to help beat the local light pollution.

The thing I found with imaging was to only concentrate on one thing at a time until mastered otherwise you just get frustrated and confused and never get control of the various processes needed for the complex field that is astro imaging.

Maxim DL has been around for years, I have been using it since version 4, around fifteen years, and it is still used extensively by professional observatories around the world, the post processing side is very out of date now but the camera, mount and observatory control plus ability to work with other software is still unsurpassed, I recently installed a new observatory and run it using speciallised software (ACP) and this absolutely requires Maxim to work!

If Maxim has a problem it is the lack of any coherent instructions for a new user, the help files describe various options and processes but not how or when to use them, the same goes for Starlight Xpress, the documents they provide don't really help a beginner.

If I can make a couple of suggestions,

Open a free Dropbox or Google drive account, put your all your raw images from last night into light, dark and bias folders, upload them to one of the accounts and post the download link back here, there will be several Maxim users on SGL who can look at your images and suggest what might be causing you problems and the exact processing steps they used with your images.

The other suggestion is probably a bit more controversial here but invest in a couple of good books on astro imaging, we all have different ideas on how to work with any set of images and using forums like SGL often give you conflicting advice, don't do this, do that, I do this, I do that, etc etc and for a beginner which advice do you follow?

I have always thought there should be a separate section here on SGL for books so that reviews could be easily accessed but perhaps I am too much of a Luddite in this and books really are history?

It is really important to understand the fundamentals of astro imaging and the requirements of your own particular equipment before you will make much progress, from your comments above it seems you are not using Maxim correctly, if you have paid for this very pricey software I would be practising using this with other sources of raw data to get a grip on it before consigning it to the trash can.

 

Yes OSC cameras are limiting and yes mono cameras will give you greater flexibility and outperform OSC but even so some really great images are posted here and elsewhere from cooled OSC or DSLR cameras and you should be able to match them. The greatest limitation with the H9C is the small CCD size and resulting limited field of view, the best H9C images I ever saw were taken with short second hand camera lenses used on an eq mount ~ 200mm ~400mm focal length, instead of a telescope,  as these allowed a good field of view on the small chip and put less exacting requirements on the mount and guiding.

If you can not get passable images from OSC with the fewer processing steps required then you will find mono harder to manage and I can't help feeling that if you get a grip of Maxim and the OSC now then you will learn an awful lot and find it much easier to move on to mono.

William.

[Skipper Billy]

Hello Paul

It was me that you bought the camera from and I have been reading this thread with both interest and concern and felt I should chip in.

Much of what has been said about mono being more productive etc is all technically true but to be honest if you are struggling to take good OSC images and process them then going to mono is a world of pain waiting to happen.

To start with I would forget about calibration frames - just concentrate on getting good quality lights and stacking them and processing them.

The instructions I gave you with the camera about how to stack and process the lights in Maxim if followed to the letter will yield a good result in under 10 mins. If you have lost them I will have a look to see if I have another copy to send you.

This is an image taken with your camera and NO calibration frames of any kind - just stacked and stretched as per the above instructions.

Its not perfect by any means but I am happy to have it hanging on my office wall.

Just shout out if I can help you. The skills you will learn along the way will stand you in good stead if/when you go to mono.

 

 

 

[Gina]

My experience with Atik cameras is that they don't need bias or dark calibration frames - the noise is extremely low.

[brrttpaul]

Hi Bill, I agreee with you (think I said earlier that is was user error not the camera), I also totally agree  with what the other guy said about instructions not being very good etc. Let me explain the biggest problem and its not intended as a pop at anyone ( im just useless writing what I mean). Lets say you wamted to build a wall but had no knowledge of it and you asked a bricklayer ( which is what I am) the bricklayer says you need sand so you go away and you get some sand, all excited you say " i got the sand" and i say  "you got cement?" so you go away again and come back with cement have you got bricks? mixing ratio gauging etc etc its endless. Thats how it feels sometimes with astrophotography which is a lot more complex than building a simple wall. I was taking a single image, then clicking digital developement ( as per instructions) then get told no dont use that use this,  take flats so take flats and its all over the shop no mention of a boxcar filter and unless someone tells you you not going to know are you lol. Im sorry if it sounds like whinging it honestly isnt frustrated yes as i know i have come a long way. I would say my biggest gripe is these darn specks which i presume are hot pixels (bright blue and bright red) it looks like a clear night tonight what I will do is take some images and put them here. the dropbox thing sounds a good idea so might set that up.

[Robp]

Could you upload one of your 30 minute exposures here so we can have a look at your starting point before you stack?

 

[D4N]

Bright dots are normally hot pixels, these are easy to remove with dithering during acquisition and pixel rejection during stacking.

If you didn't dither then they are much harder to remove.

[Skipper Billy]

Hi Paul

I think you are at where I was two or three years ago and it would drive many people insane !!!

What I have learned is to take baby steps and try to keep it simple.

First off get your polar alignment sorted, get good focus, if you are guiding get it sorted - if you aren't guiding then keep the length of lights down to whatever your mount can give you whilst keeping the stars round.

Then just take lights - don't worry about calibration frames - that's a very quiet camera and hot and cold pixels can easily be got rid of in processing.

Then just follow the instructions I agve you with the camera on using Maxim DL5 and if your data is good then you should end up with a good image.

Pop your lights in a dropbox file and post the link in this thread and I and many others will have a pop at it for you and hopefully be able to identify where any issues lie.

I feel your frustration and your building a wall analogy is a good one - just like anything else though - remove as many variables as possible and it will all come together.

I have never heard of nor used a 'boxcar filter' !!!

 

[brrttpaul]

I think this is one of them, my workflow is all over the shop at the mo as its on a laptop, im pretty sure this one is calibrated and colour converted as you can see

CCD Image 27MAX.tiff

[Robp]

Thanks I'll have a look when I'm at home.

Do you have one from before calibration as well?

 

[Oddsocks]

1 hour ago, Skipper Billy said:

 

I have never heard of nor used a 'boxcar filter' !!!

 

Hi Skipper,

The Boxcar filter has been in Maxim for a while, it is only used while producing the calibration flat master frames, I have posted a link below which explains what it does and shows how Maxim uses the Boxcar filter and also shows a manual calibration method which achieves the same, but in essence it looks at only the flat frame and averages each 4x4 pixel set - RGGB - into a single average luminance value and then pastes this back into the 4x4 pixel set (boxcar) so each red and blue pixel has the full average and each green pixel has half the average. The boxcar filter is applied to each of the flats before stacking and averaging into the master flat, it is not used on any other of the calibration frames.

Why?

The reason is that when you create a flat frame the likely hood is you won't have a perfect "white" light source and any flats from a OSC will then contain an imprint of the colour temperature of the flat light source, if you then apply these uncorrected flats to the light frames you end up introducing a colour cast into the final image which you would then have to remove in post processing (PixInsight would shrug away these colour casts using the DBE and Linear Fit tools but these powerful tools do not exist in Maxim which is where Maxim has fallen behind so much in post processing).

So the Boxcar filter is not a post processing filter and it will not harm your final image, Maxim applies the Boxcar implementation only on the master flat creation and it is only used for OSC or DSLR images, which is why it appears as a selectable tick-box in the image calibration dialogue, the result of the Boxcar filter is a perfect white balanced master flat that will not introduce a colour bias into the calibrated light frames.

Pauls use of the Boxcar is the correct method for flat frame calibration with a OSC / DSLR camera.

(Here endeth todays sermon)

HTH

William

http://www.astrosurf.com/comolli/flatfieldcolor2.htm

[Oddsocks]

1 hour ago, brrttpaul said:

I think this is one of them, my workflow is all over the shop at the mo as its on a laptop, im pretty sure this one is calibrated and colour converted as you can see

CCD Image 27MAX.tiff

Hi Paul,

Looking at your image there is something there, a bit of nebulosity and dark stuff in the top corner and again possibly lower left but the digital development tool has just over enhanced the stars.

So far as hot pixels are concerned then of course dithering will remove them but you can work on your existing images before stretching using the Maxim hot pixel / dead pixel filter.

To really see what is going on we do need to see the raw,  virgin,  fit files before you do anything to them.

If you do opt for a Dropbox account, it is free up to the point you exceed your space allowance and then the account is locked until you pay the monthly subscription, and canny operators that they are when you delete a folder from the dropbox account it disappears from view but stays in the trash can thus adding to your total space used, crafty or what?  so just remember when using Dropbox once you have finished with a file or folder you place there remember to go into the trash can and empty that too or you will end up like many others with a locked account that is no longer free to use, the limit is 2Gb, you don't get a demand for cash if you exceed the allowance, what you use is free up to that point and you don't give credit card details when setting up the account but if you exceed the limit you can no longer access the Dropbox account to delete the excess files unless you sign up for the paid option. 

If you think you might put up lots of files then Google drive is the better option with 15Gb of free space available. In the end your broadband speed will probably dictate how much space you use as conventional broadband upload speeds are only a fraction of the download speed achieved and it can take hours to upload a really big folder.

William.

[Skipper Billy]

32 minutes ago, Oddsocks said:

 

2 hours ago, Skipper Billy said:

 

I have never heard of nor used a 'boxcar filter' !!!

 

Hi Skipper,

The Boxcar filter has been in Maxim for a while, it is only used while producing the calibration flat master frames, I have posted a link below which explains what it does and shows how Maxim uses the Boxcar filter and also shows a manual calibration method which achieves the same, but in essence it looks at only the flat frame and averages each 4x4 pixel set - RGGB - into a single average luminance value and then pastes this back into the 4x4 pixel set (boxcar) so each red and blue pixel has the full average and each green pixel has half the average. The boxcar filter is applied to each of the flats before stacking and averaging into the master flat, it is not used on any other of the calibration frames.

Why?

The reason is that when you create a flat frame the likely hood is you won't have a perfect "white" light source and any flats from a OSC will then contain an imprint of the colour temperature of the flat light source, if you then apply these uncorrected flats to the light frames you end up introducing a colour cast into the final image which you would then have to remove in post processing (PixInsight would shrug away these colour casts using the DBE and Linear Fit tools but these powerful tools do not exist in Maxim which is where Maxim has fallen behind so much in post processing).

So the Boxcar filter is not a post processing filter and it will not harm your final image, Maxim applies the Boxcar implementation only on the master flat creation and it is only used for OSC or DSLR images, which is why it appears as a selectable tick-box in the image calibration dialogue, the result of the Boxcar filter is a perfect white balanced master flat that will not introduce a colour bias into the calibrated light frames.

Pauls use of the Boxcar is the correct method for flat frame calibration with a OSC / DSLR camera.

(Here endeth todays sermon)

HTH

William

http://www.astrosurf.com/comolli/flatfieldcolor2.htm

 

Thanks William - you live and learn !!!

[Skipper Billy]

Paul - there is something in the top right and bottom left of your image - this is a very quick and dirty stretch of your file in PS.

As said above we need to see the virgin Fits files via DropBox etc

David

 

[brrttpaul]

Ok hopefully I will get some sorted tonight I have a habit of clicking convert to colour as soon as the photo pops up to see whats there etc. just looking at a few other pics ( I wernt bothered about where in the sky I was just experimenting with different exposures) plot 3 is 300sec exposure  and plot 4 is 900 secs, but i notice in the FITS header that dark subtraction for the 900 sec is a 300 dark? never noticed but cant say ive looked before.  its those pesky blue and red pixels that really gets to me. another thing i just noticed while uploading is they both the same size even though one is 900sec and one is 300 sec .

CCD plot3.fit

CCD plot4.fit

[John78]

The fit file will always be the same size irrespective of the exposure length as it contains the same matrix of pixels at the same colour depth with no compression (it will be the same size even if it was 1/4000sec and every pixel black).

 

There's data in there but its overexposed (light pollution?) so its hard to pull the signal out of the noise - I'd try shorter exposures maybe? (the 300sec is easier to work with than the 900sec)

[Skipper Billy]

At least your stars are nice and round !!!

[John78]

I thought that too they are nice!

The moons pesky at the minute that's really not going to make 900sec exposures viable, try pointing at a known target away from the moon.

[brrttpaul]

That was the experiment if im honest I  was just seeing how long I could go for before they start going egg shaped, before I got a guiding camera and PHD it would be 5 maybe 10 mins, then I got PHD and thought right im guiding, I done the polar alignment, drift alignment everything but could not improve on it, then I read somewhere ( I also wanted to dither with Maxim) I read that maxim and PHD dont communicate with each other so looked into the maxim guiding, it was so easy and this was the result basically ( I was going to take it to an hour but time run out). The dithering I am not quite sure about because it has 3 options  if I remember correctly which one i tick to set up dithering not sure yet I,ll have to read up on it

[brrttpaul]

ok just set up and took two exposures both 30 secs long. The first is untouched at all the 2nd is just colour converted nothing else, no darks/flats bias nothing on either of them. I tell you what I do notice is when im going through my star alignment I usually have maxim on continuos with the cross hairs on and use the hand controller to centre the stars, but if for example I slew to altair and its off screen slightly I get the chequered effect but as soon as altair comes into view the screen changes to a nice black background and a nice well lit star, I presume thats just because the star is so bright that its doing that yes?

30sec exp-unbayered.fit

30sec exp-bayered.fit

[Oddsocks]

22 minutes ago, brrttpaul said:

  I tell you what I do notice is when im going through my star alignment I usually have maxim on continuos with the cross hairs on and use the hand controller to centre the stars, but if for example I slew to altair and its off screen slightly I get the chequered effect but as soon as altair comes into view the screen changes to a nice black background and a nice well lit star, presume thats just because the star is so bright that its doing that yes?

 

Yep..That is just the auto screen stretch in Maxim adjusting the image to best display the incoming image pixel range, it has no effect on the actual images captured.

[Oddsocks]

Ok the main problem with your 30sec exp-unbayered.fit is that the focus is off.

This is just debayered in Maxim DL 6 using the SX M25 default setting and then one go through Filter > Kernel Filter > Hot Pixel with the threshold set to 0%

Some of your hot pixels are adjacent on the CCD so these won't be removed with the hot pixel filter because Maxim thinks two or more pixels adjacent are stars.

You can get rid of these in order of simplest to use by either of the following options:

1: Dithering and using the Sigma reject option in stacking, make sure you dither by at least 6 pixels otherwise the bad pixels will end up stacking on top of one another.

2: Using a dark frame in the calibration setup.

3: Create a "Bad Pixel Map" in Maxim and use this on each of the incoming lights before debayering or stacking.

Options two and three have to applied with the same exposure time, binning and CCD temperature set in Maxim that your lights are created with.

 

 

[brrttpaul]

thx for the reply, funny enough I just stacked three images that I just done and noticed the focus is off ( I can only presume I nudged the focus wheel as I used the bat mask, also i left the dither pixel at 1.5 so way to small by the looks of it, and my hot pixel remover is set at 80% so thats wrong for me also, I will read up on the bad pixel map . Thx I,ll have another go now and refocus

[Oddsocks]

Paul,

Ok looking at your earlier image, CCD image 27 Max, that caused you so much grief I think the the following will help you understand why the results were not what you expected.

There are no processing steps here for you to follow since I don't have the raw files but there is still much to be picked up.

First thing I did was to try and find out where the telescope was pointing when the image was taken so I took the image into photoshop and messed with the levels and curves then converted it to mono to use in a plate solving program on the net. You can find the service here http://nova.astrometry.net  upload any colour or mono image and after a few minutes wait you will (hopefully) get a "Success" message appear and then click on the results page to see what's what.

The results page will tell you exactly where the telescope is pointing, what the pixel scale is, and show you any archived images from the same location so you get an idea what your image should look like.

if you can use this live while imaging it will help you to know if you are on target or not.

Here is your modified image that I used at http://nova.astrometry.net

 

Here are the results of the plate solving routine, your pixel scale is 2.67 arcsec per pixel by the way:

 

 

Your image enhanced with true star colours and background suppression via the Astrometry.net service 

 

This is what your own processed image looks like:

 

And this is what archived images from the same location look like (not so different from your image once you know what to expect):

 

Finally I uploaded the modified image into my copy of the SkyX and asked the SkyX to plate solve and overlay your image at the same scale onto the sky chart.

The little grey square at the bottom of the North American Nebular is your image at the same scale and exact location.

 

Now I can see you were aiming for the North American Nebular but at your pixel scale you were never going to be able to capture it in a single frame, your image is mostly overlying one of the boring bits with not much going on just a lot of mainly red background, loads of tiny stars with a few dark clouds and slight colour variation and this partly explains the problems you were having.

So hope this helps, I think your doing quite well really and a bit more practice with Maxim and you'll be there!

 

William. 

[brrttpaul]

Thanks very much for all your help it has been much appreciated,  trying ATM to focus but clouds have rolled in so im on stop a minute. I am using CDC and  basically looking East, I seen that and it was high up so just slewed to it, not only that i knew the red would be faint and i was exposing for several minutes i thought why not. I do use the astronomy net just to see exactly where I am (not registered though). As a side note I have just started using astrotortilla and go through the motions it comes back solved in 15 secs or whatever but thats it as far as i can see, i thought a list or something would show but it just says solved ( yip i know im missing something but cant see it yet lol).

Seriously though thanks once again for your time been appreciated

[brrttpaul]

just an update, I took some 30 sec subs>colour converted each one and applied the hot pixel to each one> then I stacked them (sigma clipping) then I used the hot pixel filter on the stacked file no darks nothing but I can see a big improvement already, only stacked 3 as the other 7 clouds came in so just picked the best. Think that was the procedure i used anyway

Group2.fit