My first imaging rig issues

[AstroDab3k]

Hi everyone.
I'm very new to the hobby and had only few images with DSLR + Lens and Star Tracker.

I managed recently to purchase a nice new scope on a fairly good price from company TS-Optics. I've seen many good reviews about their refractors and apparently this is a rebranded Sharpstar with some minor changes.
Anyway I wonder whether I set this up correctly.  

It came originally as in below picture

However to put flattener in place I had to remove the two elements above (which is a pity as I liked the red accent). There was no manual with the scope so I hope I fitted out out correctly. 

 

Then I struggled to find place where to put 2" filter, and I had to disassemble flattener and put it in front of the glass. It was the only place, but I heard that it's always recommended to put it closer to the sensor, which in my case was not possible.

 

Then I had some old T2 ring adapter to canon DSLR which I used with my dobsonian reflector. It didn't fit straight away, but I noticed it has some small screws which were holding the inner ring. I removed it and et voilá it fits now the end of flattener.

I tested it yesterday and I found number of issues.

1. Flattener needs 55mm back focus, camera has 44mm and I thought that modified ring has 11mm (if I measure the thread/ entry to camera), but it seems I measured wrong as you can see on below picture, the stars are elongated from borders to center. If what I suspect is the case, I need to buy proper adapter.

2. Star colors. I know that L-eNhance does some strange things with color, but I thought it's only gradient which I tried to remove with flats in APP. However stars are still green/blue and I had difficulties removing the color. Is it normal with this filter or something wrong with camera? Should I take also exposures without filter?

3. The horrible noise on the right side of the image or actually accros the whole image, it looks like a smudge from left to right (I've seen the walking noise previously but it wasn't that prominent). Not sure if this is something wrong with my rig, or I just need more integration time and this is just nebulosity with not enough data. Any suggestions?

4. If I would like to add the OAG, how could I achieve the focus then? ZWO OAG has 16.5mm thickness and has m48 adapter to fit camera, but then I'll have more than 55mm?

5. Is filter in front of flattener creating some issues with data or not? Or should it always be between flattener and sensor, if yes what are the ways to attach it?

I've been shooting with below set-up at ISO800, 180s subs. Bortle 4. Total integration time 2h48m. I used calbration frames too.

My Rig:

TS-Optics PhotoLine 60 mm f/6 FPL53, 1.0x Flattener, Optolong 2" L-eNhance, Canon 600D (modded), AsiAir Mini

It was a very quick APP and PS processing, so maybe there's more into this data that could improve the image, but so far I see that there are some fundamental issues with data itself.

Sorry for long intro and post and thanks for all the answers.

[Swillis]

Hi, 

Hi, looks like there is a good amount of data in your image, hopefully something can be rescued. 

Did you crop the image as a first action before starting to process the image?  This removes stacking artefacts which can effect the image processing. 

Canon DSLRs can suffer from banding, which might be part of the issue with the noise.  I'm not familiar with APP but Siril can do banding reduction which may help, but can sometimes introduce other artefacts. 

What calibration frames did you use?  It's recommended with Canon's not to use darks or bias. Subtract 2048 instead of the bias frames. 

If your in a bottle 4, do you need the filter? Maybe try without and see.   I suspect it may not be possible to attach an OAG to a DSLR since most of the back focus distance is taken up by the camera itself. Was there a reason you wanted to go for an OAG rather than regular guide scope. 

Hope you can figure out the issue

Simon

[rnobleeddy]

Can't help with everything, but:

 

- In terms of the noise/banding, if not already, you will definitely want to dither, ideally between every frame, maybe every 2-3 frames if you're shooting short exposures.

- Make sure the screen on the camera is not  on whilst you're shooting, as there can be light leakage (it was an issue on a 550D I had)

- Back spacing is never exact, so you'll need a way to adjust the spacing, and then change it until you get good stars

- For star colours, this was par for the course for me with dual band filters. You're effectively just collecting 2 wavelengths. The centre of the star is overexposed (to collect enough data on the nebula) so that's white, then the edges are one colour or another depending on the colour of the star.  Options are to shoot some shorted exposures without the filter then blend the images, or use the processing software you use to repair the stars.

 

All in, I'd have been very happy with that for my first image

[almcl]

There is a specific Canon OAG which attaches directly to the camera and on my 700d this gives exactly 55 mm, 44 mm in-camera and 11 mm of OAG.

Here's the one I used https://www.teleskop-express.de/shop/product_info.php/language/en/info/p2722_TS-Optics-Off-Axis-Guider-fuer-Canon-EOS-Kameras---ersetzt-den-T-Ring.html

As for filters you might want to look at something like the Astronomic clip filters (https://www.astronomik.com/en/clip-filter-system.html) these fit inside the camera body and sit very close to the sensor.

I found the OAG made a huge difference in guiding quality and was very worthwhile.

[Rallemikken]

9 hours ago, AstroDab3k said:

I tested it yesterday and I found number of issues.

Not bad for such a narrow scope. Keep the optical train simple in the beginning. Put away the flattener and the filter for the time being. Mount a 50mm finderscope with a ASI 120 mini in the finder shoe as a dedicated guide camera. OAG seems cool at first glance, but does not offer any advantages other than weight savings versus a dedicated guide scope. Learn to guide and dither. I point my guide scope slightly off-target. You dont want to guide on the center of a galaxy or star cluster. 70% of the image lies in the processing. Stack each series of subs at least twice, in two different programs. I use Siril, DSS and Sequator. I can never tell in advance witch stack I will keep. In some cases my finished image is a blend of two different stacks, stacked in two different applications. Don't tie yourself to one application, or one way to do things. Finish the image in a decent, regular Photo Editor, that being Photoshop, Gimp or any other.

180 seconds @ ISO800 is exactly my favourite setting on my 600D. I live under bortle 4 skies, shoots with f/4 and f/5 scopes, never uses filters and only uses flatteners to increase field-of-view. The 600D have small pixels and with my focal lenghts (600 and 1000mm) I rather crop out the edges with the elongated stars.

[AstroDab3k]

Thank all for very helpful insights!

Allow me to reply in one post, but before that I captured some more data yesterday and I noticed that pattern already on single subs. At the end of the session I removed the filter and did 2 test shots. I haven't stretched them in the PC software yet (just used the ASIAIR mobile app) but I already noticed that the strange horizontal pattern is not there anymore, so I suspect that the filter is cause of troubles.

 

On 25/09/2023 at 22:53, Swillis said:

Hi, 

Hi, looks like there is a good amount of data in your image, hopefully something can be rescued. 

Did you crop the image as a first action before starting to process the image?  This removes stacking artefacts which can effect the image processing. 

Canon DSLRs can suffer from banding, which might be part of the issue with the noise.  I'm not familiar with APP but Siril can do banding reduction which may help, but can sometimes introduce other artefacts. 

What calibration frames did you use?  It's recommended with Canon's not to use darks or bias. Subtract 2048 instead of the bias frames. 

If your in a bottle 4, do you need the filter? Maybe try without and see.   I suspect it may not be possible to attach an OAG to a DSLR since most of the back focus distance is taken up by the camera itself. Was there a reason you wanted to go for an OAG rather than regular guide scope. 

Hope you can figure out the issue

Simon

Thank you Simon, I read about DSLR pattern as well and found that there's a free software called ImagesPlus that apparently removes similar issue with good effects, before stacking.

I always crop after stacking but I saw that in APP there's an option to crop before, so I have to try this. Thanks!

I used Flats, Darks and Bias. So far I read that they are always needed (especially darks), since the Canon sensor can get quite hot. What do you mean by subtracting this value?. Is there such option in usual stacking software?

In terms of filter, it seems after yesterday session that it may be an cause of the problem. I read that it's recommended regardless of sky quality on emission nebulas as it brings the details significantly...but if this will cause this strange pattern, I will get rid of it (assuming that it's not a faulty batch)

And for OAG, I'm still struggling with my decision. I read a lot about issues with differential flexure when using scope and a lot of recommendation for OAG even for lower focal lenghts. I found the OAG (also recomended in post below) from TS-Optics that would be part of the M48 ring that goes to camera and I could achieve the 55m backfocus. Pricewise the set-up would cost me similarly to the guide scope and camera (vs OAG + camera), at least where I live, but the downside would be that if I decide to upgrade to dedicated camera I will have not use for this OAG most probably.

On 25/09/2023 at 23:09, rnobleeddy said:

Can't help with everything, but:

 

- In terms of the noise/banding, if not already, you will definitely want to dither, ideally between every frame, maybe every 2-3 frames if you're shooting short exposures.

- Make sure the screen on the camera is not  on whilst you're shooting, as there can be light leakage (it was an issue on a 550D I had)

- Back spacing is never exact, so you'll need a way to adjust the spacing, and then change it until you get good stars

- For star colours, this was par for the course for me with dual band filters. You're effectively just collecting 2 wavelengths. The centre of the star is overexposed (to collect enough data on the nebula) so that's white, then the edges are one colour or another depending on the colour of the star.  Options are to shoot some shorted exposures without the filter then blend the images, or use the processing software you use to repair the stars.

 

All in, I'd have been very happy with that for my first image

Thank you. I agree I didn't pay attention to turn off the screen camera and I noticed that there's some leakage/amp glow on the darks. 

On 25/09/2023 at 23:25, almcl said:

There is a specific Canon OAG which attaches directly to the camera and on my 700d this gives exactly 55 mm, 44 mm in-camera and 11 mm of OAG.

Here's the one I used https://www.teleskop-express.de/shop/product_info.php/language/en/info/p2722_TS-Optics-Off-Axis-Guider-fuer-Canon-EOS-Kameras---ersetzt-den-T-Ring.html

As for filters you might want to look at something like the Astronomic clip filters (https://www.astronomik.com/en/clip-filter-system.html) these fit inside the camera body and sit very close to the sensor.

I found the OAG made a huge difference in guiding quality and was very worthwhile.

Thank you! Yes, this is the same OAG I'm thinking about. As for the filters, I'm struggling a bit now whether I even need them for my sky quality.

On 26/09/2023 at 00:51, Rallemikken said:

Not bad for such a narrow scope. Keep the optical train simple in the beginning. Put away the flattener and the filter for the time being. Mount a 50mm finderscope with a ASI 120 mini in the finder shoe as a dedicated guide camera. OAG seems cool at first glance, but does not offer any advantages other than weight savings versus a dedicated guide scope. Learn to guide and dither. I point my guide scope slightly off-target. You dont want to guide on the center of a galaxy or star cluster. 70% of the image lies in the processing. Stack each series of subs at least twice, in two different programs. I use Siril, DSS and Sequator. I can never tell in advance witch stack I will keep. In some cases my finished image is a blend of two different stacks, stacked in two different applications. Don't tie yourself to one application, or one way to do things. Finish the image in a decent, regular Photo Editor, that being Photoshop, Gimp or any other.

180 seconds @ ISO800 is exactly my favourite setting on my 600D. I live under bortle 4 skies, shoots with f/4 and f/5 scopes, never uses filters and only uses flatteners to increase field-of-view. The 600D have small pixels and with my focal lenghts (600 and 1000mm) I rather crop out the edges with the elongated stars.

Thanks, I didn't think about simply cropping the image to remove the elongated stars, but on the other hand looking at my imageI would loose a lot of frame? Are there any downsides to flatteners (i.e. less light coming through?).

In terms of scope, do you mean to point a star in the software or physically turn the scope slightly off? but then all this talk about differential flexure and suggestions to use the scope with rings instead of finder to keep it ideally with the scope. I'm confused now ;D, so much to learn 🙂

That's a great advise about using different stacks, I got fixated with APP after first use as it removed a lot of issues with removing vigneting, background calibration and initial stretching that I've had previosly using DSS and PS only, but I need to try other options, world belongs to the brave ones 😉

 

Thanks again to all of you for very helpful information, so many new ways to solve the problem and try which works best for my situation!

 

[alacant]

4 hours ago, AstroDab3k said:

, Darks and Bias. So far I read that they are always needed (especially darks),

EOS? Absolutely not needed; the opposite.

Simply subtract the offset. 

4 hours ago, AstroDab3k said:

Is there such option in usual stacking software?

Yes:

https://linuxcb.blogspot.com/2023/09/siril-dslr-processing.html

Cheers

Edited September 27, 2023 by alacant

[Elp]

Nothing wrong with flatteners, they're designed so you can get the best flat field possible across the whole illuminated sensor frame. It helps if you've got a range of mm sizes m42 or m48 (depending on your flattener and image train adaptor sizes) extension rings and spacers at hand so you can finely sort the backspacing from the flattener to your camera sensor. With a DSLR this issue is usually moot as the t ring combined with the dslr body gives the correct backspacing distance typically 55mm, note not all flatteners have a 55mm backspacing specification.

[Swillis]

@AstroDab3k glad you have made some improvement. It looks like you have also had some very good suggestions above. 

16 hours ago, AstroDab3k said:

always crop after stacking but I saw that in APP there's an option to crop before

It sounds like you are doing what I suggested already. Not sure what effect cropping before stacking would have, but could be worth a try if you're still observing issues. 

 

16 hours ago, AstroDab3k said:

So far I read that they are always needed (especially darks), since the Canon sensor can get quite hot. What do you mean by subtracting this value?. Is there such option in usual stacking software?

When I started I thought they were all important, as everything I read or watched was telling me this, but as several people on this forum have pointed out darks from a Canon DSLR introduces more noise rather than removes it. Apparently they do some magic behind the scenes...  I have not tried APP so cannot comment on this but SIRIL has the option to input a value.  @alacant has pointed you in the right direction.  There is a bit of a learning curve with SIRIL, but well worth it once you get going.  There are plenty on here willing to help, and lost of tutorials online. 

 

16 hours ago, AstroDab3k said:

try which works best for my situation!

Good luck finding the one which works for you. Please do post your progress. 

Thanks

Simon

 

[AstroDab3k]

15 hours ago, alacant said:

EOS? Absolutely not needed; the opposite.

Simply subtract the offset. 

Yes:

https://linuxcb.blogspot.com/2023/09/siril-dslr-processing.html

Cheers

Thank you, will give it a try!

[AstroDab3k]

5 minutes ago, Swillis said:

When I started I thought they were all important, as everything I read or watched was telling me this, but as several people on this forum have pointed out darks from a Canon DSLR introduces more noise rather than removes it. Apparently they do some magic behind the scenes...  I have not tried APP so cannot comment on this but SIRIL has the option to input a value.  @alacant has pointed you in the right direction.  There is a bit of a learning curve with SIRIL, but well worth it once you get going.  There are plenty on here willing to help, and lost of tutorials online. 

Thanks, that's a very good point. I see that there are a lot of conflicting opinions on many topics, but the best way is often to check by myself (as long as it's not related to spend money, then I prefer to do the thorough review ;-))

 

[david_taurus83]

If you are in Bortle 4 area do you have any local light pollution? I would say you could get away with no filter at all. The L-Enhance is a narrowband filter so you will need much longer integration with a DSLR. Below is 5 hours with my 6D at 135mm and a standard LP filter. Regards to noise, do you make sure to flip open the screen and switch the display off? As mentioned above, you don't really need darks. I use a master bias file and subtract that from the lights and flats.

 

[Dountoun]

Interesting, I have the same telescope with the same field flattener but a Canon 700D (modded) and I use a Astronomik L-2 UV/IR cut clip in filter.

I do not have any problems with backfocus on my setup.

 

[Rallemikken]

On 27/09/2023 at 15:23, AstroDab3k said:

Are there any downsides to flatteners (i.e. less light coming through?).

Not so much that, rather it will be one more thing to cause possible trouble and complexity. I'm a reflector guy, and I try to image without any glass between the stars and camera sensor. I'm no expert on refractors, but I think a field flattener works about the same way as a coma corrector on a reflector; it fixes the misshaped stars along the edges and in the corners. With small targets at long focal lenghts the "area of interrest" often gets small, and there is room to crop. On bigger targets I use my CC to get nice finish all out, but mostly I leave it in the drawer.

I have two scopes, one 6" f/4 and one 8" f/5. Thats 600 and 1000mm focal lenght. I have 3 Canon's, one 5D MkII fullframe, one 600D cropsensor, and one 450D, also cropsensor. My coma corrector is a Skywatcher 0.9. Together, this relatively cheap setup offers a waste number of combinations.

As for the guiding, I turn my guidescope slighly off target. Not much, but enough that I don't get my main target in the view. I guide my 8" reflector with the smallest mono camera on a 50mm SW finderscope in the findershoe on the big scope. Actually the same finderscope that came with the main scope. Works great. Don't spend big bucks on guiding hardware before you have a sturdy mount, and you master the software. Guiding up to focal lenghts of 1000mm is relatively painless. Just needs to be learned. Keep everything tight, the guidescope in perfect focus, and a tripod or pier that's rock solid. The better you master your current setup, the better the chance is that your future upgrades will be spot-on.

[AstroDab3k]

19 hours ago, Dountoun said:

Interesting, I have the same telescope with the same field flattener but a Canon 700D (modded) and I use a Astronomik L-2 UV/IR cut clip in filter.

In my case, I'm 90% sure it's the wrong EOS M48 ring (I'm mean it's not even M48 ring, I had somethign at home that I modified, but I measured wrongly the thickness)

 

22 hours ago, david_taurus83 said:

If you are in Bortle 4 area do you have any local light pollution? I would say you could get away with no filter at all.

no not really, I'm planning to do next session(s) without it. I have accumulated 4+ hours with the filter and all frames suffer from horrible horizontal banding, or maybe it's not even that. I can see that already before stacking when I stretch single exposure. First session had better seeing and it wasn't that bad but with second session and worse seeing, after stacking all looks like crap. I will upload images later today when I get back home to show what I mean.

I tried two frames without the filter on the same object the same night, and after quick stretch in Asiair app I noticed that this banding issue is gone. I will need to dig into that a bit more. Maybe I have a "bad batch" filter and will have to return it. I'm a bit angry as I'm not sure if I will be able to recover anythign from this data, so 4+ hours of imaging gone

 

[david_taurus83]

The banding won't be the filter. It's narrowband so far less signal getting through from the background (space) of the target so when you stretch the image you are also stretching the noise. Really, 2 and a bit hours with a narrowband filter on a 600D is not enough signal. Without the filter you would have less emphasis on the target but I think you would still have a nice image with less noise.

[alacant]

On 29/09/2023 at 08:51, AstroDab3k said:

horrible horizontal banding,

The 18mp sensor is usually quite clean.

Anyway, 600d, so are you sure:

• you are powering the camera with either an in-camera battery or a stable DC supply?
• you are not using bias and/or dark frames taken In-camera during pre-processing?
• there is no memory card in the camera?
• that the screen is folded out?
• you are dithering at least 10 pixels between frames?
• sending the pre-processed sequence through banding reduction before registration?
• stacking using a clipping algorithm?

Then cables, hubs, usb sockets, software; anything you can check by substitution... then... etc. etc.

Cheers 

 

 

Edited September 30, 2023 by alacant

[AstroDab3k]

14 hours ago, david_taurus83 said:

The banding won't be the filter. It's narrowband so far less signal getting through from the background (space) of the target so when you stretch the image you are also stretching the noise. Really, 2 and a bit hours with a narrowband filter on a 600D is not enough signal.

I've done 4+ hours in total with the filter and results was even worse (i.e. the horizontal banding was more prominent). I tried stacking in different software and also even without using any calibration frames (just lights). Same results.

However I have done a comparison with shooting with and without filter (although I managed to shoot only 2 frames w/o filter). I took single frames and stretched them only to show the banding problem (see below).

The no filter frames have normal noise patern which would be removed with more integration, but the fitler ones have regular horizontal lines pattern

Session 1 Filter (single frame)

Session 2 Filter (single frame) - dont mind the glow from the left, I forgot to turn of the LCD screen

Session 2 NoFilter (single frame)

[AstroDab3k]

12 hours ago, alacant said:

• you are powering the camera with either an in-camera battery or a stable DC supply?
• you are not using bias and/or dark frames taken In-camera during pre-processing?
• there is no memory card in the camera?
• that the screen is folded out?
• you are dithering at least 10 pixels between frames?
• sending the pre-processed sequence through banding reduction before registration?
• stacking using a clipping algorithm?

1. Stable DC supply

2. Well I tried both and results are the same

3. There is memory card in the camera (should I remove it if I shoot with Asiair?)

3. Yes and turned off (however with second session I forgot about it 🤪)

4. No dithering. I'm using Asiair App to shoot and I thought it's only possible with guiding (which I'm yet to upgrade)

5. Well, I'm using APP or DSS, no idea how to do this. I tried software ImagesPlus (found someone recommending this for horizontal banding in DSLRs) but couldn't improve anything, just introducing even more banding...

6. In DSS I'm using Kappa-Sigma clipping, but in APP I follow the standard ones

[david_taurus83]

So what is the average ADU value of the single frames with filter and without filter? My point is the frame without filter has more signal so will be masking the horizontal banding. The 600D is well known to produce banding, there is even a specific script in Pixinsight to remove Canon horizontal banding!

[AstroDab3k]

10 minutes ago, david_taurus83 said:

My point is the frame without filter has more signal so will be masking the horizontal banding

Ok I see your point. Assuming that I'm checking in correct place, the avg ADU for filter frames is 8978 (Session 2) 8533 (Session 1) and for no filter 12540.

So basically filter is doing the good job, it's just bringing the 600d banding to the point that it kills the picture? I've done 4h+ of data and it's just increasing the problem rather removing :/. I didn't use the PixInsight yet, I'll see if this will help somehow

 

[alacant]

Recommendation: upgrade to siril. 

Here are your before and after images. With the banding routine applied to the whole sequence, it becomes negligible.

Yes, you must remove the sd card.

To dither, you do not need to guide.

HTH

 

[david_taurus83]

Agree with alacant, remove the SD card from the camera. I'm sure I read on one of the ZWO pages that this causes issues for some people.

[AstroDab3k]

16 minutes ago, alacant said:

Here are your before and after images. With the banding routine applied to the whole sequence, it becomes negligible.

😮I literally had to pick up my jaw from the floor...

I tried it even on my first version with 2h data and it already improved by a lot. I never used this software but I think I will start from now on...are you using it for stacking too? or something else?

Edited September 30, 2023 by AstroDab3k

[alacant]

1 hour ago, AstroDab3k said:

are you using it for stacking

Yes. It makes the alternatives seem decidedly aged;)

But I'm strill struggling to see why the banding is so pronounced.
JTOL: do you have the latest firmware?

Also as @david_taurus83suggests, the l-enhance may not be doing you any favours. For HOO, we much prefer the UHC. Same effect, differnt price!

Cheers

Edited September 30, 2023 by alacant