CCD colour

[jaygpoo]

I need help as I am loosing the will. I have a QHY8L one shot colour CCD. Everything I take is BW but the members have pointed out that in DSS you need to change the camera type from QHY8L to generic GBRG in settings. This I have done. Last night I set up and took 12  /  6min exposures of M51 (guided) due to time I could take no more but felt the amount of time would give me good data. The images where almost white as if a light was on but the moon was full and I am guessing that was the issue?. I then used DSS and saved the image as a fit which was then run through Fits Liberator and saved as a tif. I opened up SC6 and cropped the image as in photo 1 and then adjusted the levels and curves but although I do not have the experience to remove the light polution the image shows no sign of colour / What am I doing wrong?

[Earl]

I thought it was RGGB for the QHY8L, it has been a long time since i played with one mind you.

[mhard26339]

Hi

I am not sure that this will help too much but I think most people have serious problems when first trying to sort out OSC CCD cameras; I know I did with an SXVR-H9, just the same frustrations. I use Maxim DL and unless you are a member of an astronomical society with help readily at hand it really is a matter of trial and error with the settings; the guys on this site will be great in helping you and keeping you going, it is worth it when the magic moment happens. Keep going!!

Regards

Mike

[Astroboffin]

Hi,

Is it not something as simple as a debayering option while processing??

Just a thought

AB

[Zakalwe]

Generic GBRG is the correct one to use. It looks like you have this correct as there is no matrix visible in your shot.

You will need to bump up the saturation to about 17% in DSS to see any colour.

Shooting OSC data under an almost full moon will be darned difficult. Best waiting for darker nights.

[martin_h]

Why save as a FIT then convert to TIFF, when you could just save as TIFF in the first place?

[Thalestris24]

Hi Jay

My qhy8l has always given good colour but it can 'disappear' from the output image when stacking with dss. First of all, I can confirm the bayer matrix setting should be GBRG and, obviously, the box is ticked under the FITS tab. You should see colour in individual subs within dss. If you don't, then something has probably gone wrong with the capture. I use the 'per channel background calibration' option in dss. You may like to use dss to shift the histogram back to the left using the linked slide controls. You can increase the saturation in the final image by adjusting the saturation slide control to 25-30% (well, adjust to taste!). If you use dss thusly then you need to save the altered image (with a new file name). That way you will have the Autosave.tiff (which has not had the saturation adjusted) and the save file which has had it adjusted. Thereafter, subsequent adjustments will be done in whatever post-processing software you're using.

Hope that helps

Louise

[Thalestris24]

Hi again

Just did a quick and dirty stacking of an M37 I did last August on my 150pds/qhy8l with idas lp filter. Only 5 x 600s.

Here is a single sub (can see colour):

Here the 'raw' output after stacking (colour drained):

Adjust histogram (colour still drained):

Saturation set to 25% (colour improved):

I always suffer from gradients because of a nearby streetlamp but I've not done any other processing/gradient removal with this. Should give you and idea of what a qhy8l image should look like at each step.

Cheers

Louise

[steppenwolf]

I then used DSS and saved the image as a fit which was then run through Fits Liberator and saved as a tif.

Did you extract each of the three planes within FITS Liberator? FITS Liberator has to manually import each of the Red, Green and Blue planes individually which you then combine in something like PS or the Gimp to produce a colour (RGB) image.

[jaygpoo]

Thanks Louise and others for your time and thoughts. While I am asking can someone explain the bin 1 / 2 / 3 / 4 options . I always use bin 1 . Jay

[Zakalwe]

Thanks Louise and others for your time and thoughts. While I am asking can someone explain the bin 1 / 2 / 3 / 4 options . I always use bin 1 . Jay

Binning uses software to "combne" groups of pixels into one "super" pixel. it makes the camera more sensitive at the expense of resolution.

http://www.andor.com/learning-academy/ccd-binning-what-does-binning-mean

It can't be used on a colour camera as it won't work with a Bayer matrix.

[D4N]

There are a few sensors that can colour bin but they are the exception to the norm. It depends on the colour matrix, read pattern and sensor firmware.

Sent from my iPad using Tapatalk

[lensman57]

There is no colour information in the processed image. If possible put the subs and the calibration frames in drop box and post a link and we see what can be done.

A.G

PS: I am not sure if QH8L and QHY8 have the same Bayer Pattern but for my QHY8 I use RGGB in Pixinsight.

[Zakalwe]

PS: I am not sure if QH8L and QHY8 have the same Bayer Pattern but for my QHY8 I use RGGB in Pixinsight.

The Bayer matrices are different.

[Thalestris24]

There is no colour information in the processed image. If possible put the subs and the calibration frames in drop box and post a link and we see what can be done.

A.G

PS: I am not sure if QH8L and QHY8 have the same Bayer Pattern but for my QHY8 I use RGGB in Pixinsight.

I suspect the op lost the colour info as a consequence of saving via fits liberator...

[jaygpoo]

I set up again last night and using my 9.25 with a focal reducer and guided I took 20 180sec subs of M13. I ran them through DSS and as told set the saturation at about 25%. saved as a tiff and opened it in sc6. 1st image (screen shot). copied and used levels and then curves 2nd image and did that again 3rd image. Not much to show for an evening and all I can see is a shade of yellow. I know m13 is not colourful as I have imaged it many times. I did try at M101 but clouds came.  I do not feel as if I am getting anywhere at present but will read as much as I can and try looking for answers. enclose 2 previous images of m13

[Thalestris24]

Hi

At least you seem to have some colour now. Did you stack with calibration frames? Also, meant to ask before if you have set the gain and offset of the camera?

Perhaps you could post a link to the dss autosave.tiff file?

Louise

[jaygpoo]

I am showing my thick side as I can not work out how to post a link to a file on my computer? and its too big to download on this site.

[Thalestris24]

Hi

If you have, say, Dropbox, you can just copy paste the url of the file into a post or paste the url into the link facility in the post reply box (bottom row, 9th from left).

Hope that's clear.

Louise

[jaygpoo]

https://www.dropbox.com/s/0u6fk0jl4a29hfp/Autosave011.fit?dl=0

https://www.dropbox.com/s/n6ie8wtk2lr1bn9/Autosave012.fit?dl=0

https://www.dropbox.com/s/neuhdeu0krrm1zq/Autosave013.fit?dl=0

Well thats been an interesting exercise in linking files. I learnt something there. 3 files of recent images. to be fair I am not sure which is the right one but they all show something that I have taken.  Jay

[Thalestris24]

Hi

What are the 3 versions?

I'll have a look tomorrow (Sunday)

Louise

[Zakalwe]

There's really bad trailing in the 2nd and 3rd file. I'd concentrate on getting the foundations correct first, tracking and focus. There really isn't much point in trying to wrestle with less-than-optimal data (I know...I've tried it often enough). You'll only become really frustrated trying to get something out of it in post processing.

The first one is better (M13), though there appears to be some dodgy subs in the mix. Inspect each sub before stacking and delete any with trails. It's hard. I know, to delete hard-won data, but poor subs just lower the overall quality of the result. The focus also looks a bit soft..how are you judging focus? A Bhatinov mask is worth it's weight in gold here.

Are you using a focal reducer on your C9.25? Even with a focal reducer, you are imaging at a fairly long focal length. That's going to put demands on your tracking (are you auto guiding?).

Finally, there appears to be a lot of vertical lines in the images. Carol Pope (Carastro on here) had a similar issue. It might be worth speaking to her to see if she got it sorted. It can be down to interference on the USB line.

[Thalestris24]

Hi Jay

I have to concur with Zakalwe's comments. Also, you are using an old driver which overscans the sensor producing unwanted margins. So do download the latest one before doing any more imaging. I'm not sure if the driver version is related to the vertical banding issue.

Anyway, I had a quick play with the first tiff. I did some adjustments in dss. It's not the recommended way of doing it but still. I also loaded the saved adjusted image into Startools to remove the vignetting. Stacking the lights with flats (and bias) calibration frames is the best way... With short exposures, you can probably get away without darks.  You didn't say whether you set the camera gain and offset in your capture program?

DSS adjustments:

Post Startools:

There is clearly colour present

As Zakalwe says, do concentrate on sorting out the fundamentals first.

Good luck!

Louise

[nucdoc]

Here is a very quick run thru PixInsight, definitely some color in the image, you lose a bit with conversion to Jpeg

[lensman57]

https://www.dropbox.com/s/0u6fk0jl4a29hfp/Autosave011.fit?dl=0

https://www.dropbox.com/s/n6ie8wtk2lr1bn9/Autosave012.fit?dl=0

https://www.dropbox.com/s/neuhdeu0krrm1zq/Autosave013.fit?dl=0

Well thats been an interesting exercise in linking files. I learnt something there. 3 files of recent images. to be fair I am not sure which is the right one but they all show something that I have taken.  Jay

Thank you for posting the files. The good news is that you have M13 with a hint of the propeller showing too. You also have colour and it is correct. The bad news is really evident for itself. The vertical bands could be either a driver or USB issue or both. You also need to calibrate with good flats .

Regards,

A.G