assouptro

5 hours ago, windjammer said:

Over processed ? I don't think so - you've brought everything out and no artifacts AFIK.  It is very interesting.

Thank you windjammer I appreciate the comment 

I love the really natural look more experienced processors are capable of, I imagine it’s helped by using better optics and equipment than I currently own but it’s difficult to be completely satisfied with the final result when there is always another tweak! 

I think this is slightly better, but I’ve been looking at it for too long now!! 😵‍💫

Thanks again 

Bryan 

On 20/02/2023 at 19:48, newbie alert said:

That's just crazy...another old but still no slouch when it comes to using it

 

Hi newbie alert! 

I have to admit the 383 seems less sensitive and noisy in individual subs compared to some of the other cameras I’ve used and needs more time per sub leaving it vulnerable to satellite, plane, cloud or wind disruption but give it time and good conditions and it can produce the goods! 
I’m sure a more experienced processor could do if more justice, the stacks were really easy to work with! 
Thanks for the comment 

Bryan 

6 minutes ago, sharkmelley said:

The star blips are almost certainly caused by some subs that have "doubled stars" because of wind gusts or periodic error.  Blink through your raw subs and you are likely to find the offending ones.  Remove those faulty subs from your workflow.

Mark

My thoughts too

I have a heavy rig on a medium mount and when it’s gusty I get moments when there are double stars 

I always blink my subs and often remove 20-30% depending on the wind, if it’s particularly gusty I may only keep 20-30% 

I hope you get to the bottom of it 

Bryan 

2 hours ago, SiriusDoggy said:

Really amazing close-up details. Congrats.

Sorry to hear about your camera problems though.

Thanks for the comment and the sympathy SiriusDoggy it is appreciated 😊

Cheers 

Bryan 

6 minutes ago, Neil_104 said:

That's fascinating, never seen that before. Really gives a sense of movement

Thanks Neil 

I believe the Helical twists are caused by strong stellar winds sculpting the Bok Globule but I may be wrong? 
 

Thanks for the comment 😊

Bryan 

Hi stargazers 

I haven’t had a great deal of luck recently after buying a second hand qhy268m that was faulty, I was offered a refund but chose to attempt a service and repair which was luckily still within the 2 year warranty period 

I returned the camera to the vendor at the beginning of November and am still awaiting an update, as far as I know, the camera is still in the uk awaiting an rma number and return instructions from qhy 😟

Whilst waiting my Atik490 also developed a fault! 
That camera has been assessed and I have given instructions to repair, however the last communication was 19th of Dec 

I have since been playing with my noisy Atik383, a device camera that has a lot of lovers and loathers 

This is a closeup of the Helical Bok Globule in the rosette nebula, I have always found these potential star forming structures interesting and love imaging them 

This is an over processed version of the Globules comprising around 18 hours SHO, I would have liked to add some rgb stars but the weather hasn’t allowed 

Taken with my trusty Meade lx200 12” sct with a Starizona lf field flattener/reducer taking the F10  down to around F7 but more importantly helping control the star shapes 

It’s a jpeg, uploaded to my phone and tweaked in ps express 

Thanks for looking 

Cheers 

Bryan 
 

Epic! 
wonderful heart image 

I love it! 
 

Thanks for sharing 

Bryan

22 minutes ago, newbie alert said:

Nice image, what makes you say it's less than perfect?

Are you shooting wide open, ie f2?

Hi Newbie 

I’m not sure if you are referring to ..

22 hours ago, assouptro said:

I love my (less than perfect) Samyang 135 lens 

 

Or nmontec’s comment 

On 06/11/2022 at 18:55, nmontec said:

Objective: Bad quality Samyang 135mm f/2

Which my quote above was referring to? 

personally I find the Samyang 135mm lens a wonderfully versatile tool for Astrophotography, it’s fast optics and wide fov are perfect for many DSO but it does require some tweaking especially if using with a bigger sensor.

I can only speak from personal experience

My version is the Canon bayonet version and I find most adapters (m42/48 or m56 in my case) have a “springiness” in them that easily allows the lens or camera to sag, resulting in tilt if not completely aligned and  supported so spoiling the outer regions of the image 

Also it is better to “step down”the f ratio of the lens to prevent weird star shapes, but doing this with the in-built aperture can also lead to “weird star shapes” or diffraction spikes so it can be best to reduce the outer  lens to about 50mm instead or the standard 77mm (I think) with reductive filter ring adapters that may also add a reflective surface for light (from moon etc) to bounce off from the lens causing further issues 

So to sum up, It’s a brilliant lens, but if you are a pixel peeper you will have to fine tune it to get the best from it 

I hope that answers your question, unless you are referring to the original image and in which case I apologise for my hijack?! 😊

Cheers 

Bryan

Stunning image!

I love my (less than perfect) Samyang 135 lens 

There’s so much going on in there!! 
Have you tried star removal? I’d love to see all that nebulosity without the stars! 

Thanks for sharing 

Bryan 

On 07/11/2022 at 20:38, Rodd said:

Thanks--had a look.  Gees.  Why does everything have to be so complicated.  It will take me a long time to figure it out and I am sure something will not work as per usual.  Thanks though for the link

Amazing detail in the image Rodd, I can see the microlensing issue but it didn’t distract me

It’s not a problem I’ve had to deal with so I’m sorry I can’t advise but I have installed starnet2 with Pixinsight and if was a reasonably simple (and free) procedure
I do however prefer starXterminator though It just seems to do a better job?

I guess this is just one of the wonderfully frustrating things about the hobby/obsession that is astrophotography!

I have no doubt you will sort the software and make good use of it! 

 Thanks for sharing 

Bryan 

I find m45 a deceptively Tricky object to process 

I think you’ve done a Stirling job! 
 

Thanks for sharing 

Bryan 

Hi Stargazers!

I am waiting to hear back from QHY regarding the 268m I bought second hand which seems to have some problems with well depth 

I have sent it back to the original retailer where Rupert of Astrograph.net has agreed to help me with the diagnosis and repair 

In the meantime whilst waiting  I turned to my faithful Atik cameras! I was using my 460 initially but realised my 490 with its smaller pixels is better suited to the small focal length of the Samyang 135 lens

This was taken on Bonfire night where patchy cloud, smoke, and an 87% moon, I ended up with an hour Ha and about 45 min Oiii and Sii 

I had to crop due to tilt and possible distance issues but I’m fairly happy for a one night quick image 

I am itching to get an answer on the QHY268m but in the meantime I am lucky that when the clouds allow I can still “play”!

Thanks for looking 

Bryan

And starless

10 hours ago, Richard_ said:

Ah mate, that sucks to hear. I would be mortified if I pushed on a faulty item, so is there any chance you can get back in touch with the previous owner to explain? Pretty bad if they sold a faulty item and didn't say anything about it. 

But yeah, have a look at the firmware too and get in touch with QHY as you say. 

Quick update 

I have contacted the chap I bought the camera from and he has offered me a full refund! I have also contacted QHY as I would prefer to get the camera serviced and fixed if possible as I feel invested in the Qhy infrastructure and would love to get to grips with a working version of this camera

I am currently waiting for QHY to get back to me and hope the camera can be fixed without too much time and cost involved 

In fact the chap I bought it from has a spare 268c which he would be willing to loan me if the timescale of a repair is long!

I will report back here once I know what QHY plan to do 

Once again, I want to thank you for your help! It's been great to be able to reach out to the stargazers lounge community and get some really useful advice!

Kind Regards 

Bryan

8 minutes ago, Richard_ said:

Ah mate, that sucks to hear. I would be mortified if I pushed on a faulty item, so is there any chance you can get back in touch with the previous owner to explain? Pretty bad if they sold a faulty item and didn't say anything about it. 

But yeah, have a look at the firmware too and get in touch with QHY as you say. 

Thank you  for your sympathy

I will contact the chap I bought it from and see what we can do? 
He seemed reasonable, there was an issue with the filter wheel which he declared and I took a risk with that hoping I could fix it which I did! Then I had to research and buy the filters, and arrange the right adapters for me to get it up and running 

I hope Qhy can help me I’ve invested so much in this upgrade 😢

The analysis was run in mode 1 as this is the mode i am having problems with 

I bought the camera second hand unfortunately, and it has taken me a couple of months to get around to testing it properly so I’m not sure where to go. 
I will check firmware first, then contact qhy 

I had a gut feeling when I startend using it there was a problem 

22 minutes ago, Richard_ said:

No problem Bryan, these things can be frustrating to resolve and the last thing you want is buyers remorse if you can't get your new purchase to work.

OK something doesn't look right with your Sensor Analysis, but this is really useful information. I wasn't sure if you did this in Readout Mode #0 or #1.

• e/ADU or Read noise isn't being reported
• The other values reported are not sensical
• The Measurement Graph looks odd

The image below is from QHY's website and is their own measurement graph for all read modes. The purple and blue lines are Mode#0 and Mode#1 respectively. Notice how the read noise pretty much decreases with increased gain. Compare back to your graph and you can see that your read noise is increasing around gain 200 which doesn't match QHY's chart.

 

Below is my sensor analysis of Readout Mode #1 in SharpCap. Notice that e/ADU and Read Noise is present in my readings. Also, notice that read noise sharply drops at Gain 56 which matches the chart in the QHY picture above.

 

There could be a couple of different reasons for this, but before we write it off as a duff camera, there are a few things you can do (and luckily you have a friendly neighbour with a QHY268C!).

To problem solve, I would highly recommend plugging your camera directly into your laptop using the cable and 12V power supply provided with your camera to rule out any third party cables you're using.

1. Run sensor analysis
   1. If you get good results, then it's likely the USB cable is the culprit, in which case see if you can get a good quality USB cable of desired length (try the Lindy ones at FLO).
   2. If you get the same weird results per your post, then perhaps it's the drivers or firmware of camera (or dodgy camera) or perhaps the sensor analysis was run incorrectly?
2. Repeat sensor anaysis with your neighbours QHY268C
   1. If you get good results with the neighbours camera, then it's likely something to do with your QHY268M (at this point I would contact the vendor and/or QHYCCD).
   2. If you get the same bad results with your neighbours camera (and they have confirmed it works fine with them), then perhaps it's the drivers installed to your computer. I wish installation of QHY drivers were as straight forward as it were for my ASI533, as I had to install a couple of times to get it to work.
3. After re-installation of drivers, re-run sensor analysis with QHY268M and QHY268C
   1. If it works for both, excellent, it was probably a dodgy driver install.
   2. If QHY268M is bad but QHY268C is bad, then it's probably a dodgy camera, in which case contact the vendor and/or QHYCCD
   3. If both cameras are bad, then it still may be driver related issues or how SharpCap is being performed. Maybe you could ask your neighbour to test your camera out as a final sanity check.

Let's see how it goes after that Fingers crossed and it's just a USB cable and/or driver issue!

I am running sensor analysis atm with neighbours camera and straight away I can see a difference 

for starters I am not struggling to get the brightness of my iPad low enough for it to start and I am getting a straight line with the green crosses whereas with mine they were all over the place! 

14 hours ago, Richard_ said:

Good evening Bryan. I had a quick look at your two files (mode0 and mode1) and compared with an existing H-alpha usb of mine (Richard_NGC6888). I have a different telescope, f-ratio and target to your images so bear that in mind. There's a couple of observations, which I'll summarise below and include more details further on. I've included some questions as a sanity check. Not sure if this will fix anything, but it's a good place to start with.

1. Your USB limit setting is different to mine
2. Something is causing your pixels to fully saturate (this is why the mode1 image looks weird per above conversation)

As a sanity check, can you confirm the following:

• What are you connecting the camera to and what connection port (e.g. Mini-PC, USB3)
  • I'm using the USB cable provided with the camera which is connected to my Pegasus PowerBox Advance USB-3 port. This works fine for me.
• What USB cable are you using and what length?
  • It's not advised to use USB-3 cables longer than 5m otherwise you'll get download issues.
• It might sound daft, but can you very the position in filter wheel of H-alpha and ensure that you weren't shooting without it?
  • If you weren't using a filter, then you're essentially collecting Luminance data which may cause the pixels to saturate very quickly (especially at ~f/2)

 

FIT Header Review - USB Limit Setting

I reviewed the FIT headers of your files in PixInsight and made some comparisons of the important information, see Table.1 below. The one setting which jumped out at me was the USB limit setting, with mine set to 0 and yours set to 10. I see you are using NINA, so this setting will show up on the "Camera" page, towards the bottom of the driver settings. The default value is 0, and the only reason to increase this value is if you are getting vertical/horizontal banding in your subs. If you have purposely set yours to 10 to avoid banding, then leave it at 10, otherwise you could try setting this to 0, capture some subs and see if the images come out OK.

Table.1

 

Difference in Pixel Values

I then compared subs using the "Statistics" in PixInsight to see what the pixel values look like. See Table.2 below.

1. For mode1, the pixel count seems to be lower than my image and your mode0 image. Unsure why
2. Because of different telescope and f-ratios, I doesn't make much sense to compare the rest of the stats between your image and mine
   1. mode1 has significantly higher values than mode0, and there is also an insanely high deviation (this is standard deviation in pixel ADU). A high standard deviation could indicate lots of saturated pixels (i.e. max readout of 65,534)
   2. It's worrying that your minimum for mode1 is just 1, whereas my sub and your mode0 show a more reasonable value
   3. PixInsight can only report maximum ADU, but doesn't count how many times it occurs. You can see this at time of acquisition in NINA per Fig.1 below. When you capture an image, it will display the maximum ADU value and in brackets, it will show the count, such as 65,534 (x120) would indicate 120 pixels are fully saturated

Table.2

 

Fig.1

 

Saturated Stars

Reviewing your subs in detail show other things. When I review mode1 sub I can see that your stars and the nebulosity of the fish head portion are reading 65,534 (ie max value) which is oversaturating the pixel, see Fig.2 below. This would explain why this sub has such a high standard deviation. Looking at the stars, you do not see a gradual profile in brightness, they just appear to be fully saturated or nothing. Refer to Fig.3 which shows a comparison of stars from my sub compared to yours (different targets and focal length, but observe the change in star profile). Also look at Fig.4 which is a comparison of mode0 and mode1 at the same star regions. Both images show fully saturated stars, with mode1 being more sever than mode0. We can ignore shapes of the stars, this is due to flatness of the image (i.e. lack of flattener with the Samyang 135mm lens).

Fig.2

Fig.3

 

Fig.4

 

Wow! 
Thank you for all your hard work 

I really appreciate the effort you have gone to to try and help me get to the bottom of this 

Sorry for the delay in responding It is driving me to distraction at the moment and I had to take a reality break, my family are getting a bit bored of my “camera issues”

I tried to do a sensor analysis last night and it crashed on the last process twice taking up most of my evening! 
I did finally get it to complete and show me a table of results but I don’t know what it all means yet 

I have borrowed my neighbours qhy268c so I can do a comparison, not like for like but it might help 

My usb cables are new and Ugreen the longest is 5m, connected direct to a laptop I have also tried 3m 

usb limit  was indeed at 10, that was just me playing with things I didn’t fully understand, I have changed it back to 0 

I’ve checked the filter positions 

After the sensor analysis I tried to print screen to OneDrive but it failed due to my OneDrive being full! 
but I have some photos I took with my phone (you can read the table if you look) Maybe it will mean something to you? 
I don’t fully trust my results as I ended up having to increase the brightness of my iPad to do the last part and I’m not sure it had already started? 
i really need to do it again and do it for my neighbours 268c as a comparison 

Your findings are troubling, I am quite concerned there is something really wrong with this camera? I might try a firmware update, if that’s a thing? 
 

Anyway, here are the photos of the sensor analysis table 

Thanks again for all your efforts to help 

Bryan 

1 hour ago, Richard_ said:

Hey Bryan, I'm out at the moment so I can't view FIT files on my phone. I'll try and take a look when I get home this evening or tomorrow morning.

If you remember from a previous thread, I use Mode#1 with Gain 56 and Offset 25 and the same Antlia filters so I'll compare your sub with mine to see how they compare. 

Thank you Richard 

That will be really useful! 😊

Kind regards 

Bryan 

4 hours ago, geeklee said:

I can only offer experience of using a CCD (460EX) and CMOS (various) on different rigs over the same time periods and all the data had similar characteristics.  Yes I exposed for longer with my CCD but a sensibly exposed mono image always had a similar histogram and appearance (both raw and stretched) etc.  

From the few things I know about the QHY 268 family, you're using the basic low gain and high gain defaults? So nothing out of the orindary? 

When we're troubleshooting issues it's so frustrating we don't have more clear nights!

If you have others filters in the wheel and could just shoot a 30s Lum image, see how that comes out?  Even an image without a filter just to see the response and FITS file.

What about a Flat file - have you taken one yet?  I'm thinking how that might look (pixel levels) Vs the above.

It's like the above images converted anything above a certain (low) level to a value of 65535.  If I take another crop of the 180s frame I have (same as above):

Raw:

AutoSTF:

Then pick an area of "bright" nebulosity in the core of the Heart  (lower rights) that's around 0.0280 in PI (between 0 and 1).  So still a very low value as you'd expect.

Run this PixelMath, which is saying "if the value of the pixel is above 0.0280, then make it 1 (clip), otherwise leave it as is".  I get:

Raw:

Looks very similar to your mode1 file.

Lee

 

 

Thank you 

I have been fixing my sons integrated washer/drier! Bloody awful machine to get access, but now working! (If only I could apply my skills to understand the intricacies of CMOS cameras!! 
I had to read your post a couple of times for it to sink in but I understand what you are saying, it’s like the camera is over processing the bright areas somehow? 
I need to take the camera off the filter wheel and complete a sensor analysis, that’s my next job to try and understand what’s going on 

Oh and flats, yes, that was the first issue I had! I have tried to take flats and failed miserably, even in mode 0

Thanks again 

Bryan

1 hour ago, geeklee said:

You wouldn't think 120s Ha subs even at F2.8 through 4.5nm filters should look this should they?  Crop of the Heart from mode1 FIT file with no stretching.

It's the way both images have such stark clipping/saturation.

Hope you managed to make use of the clear skies 😊

Thanks again 

I completely agree but then, this is my first qhy 268 camera 

I might email QHY with the subs to ask their opinion? I have a sinking feeling that something is wrong with the camera 😬

9 minutes ago, ONIKKINEN said:

Are these narrowband, or broadband captures? Pretty sure its broadband, since the background levels are quite high in both.

The mode1 shot looks a little bit too saturated to my eyes, plenty of areas in the actual nebula itself are clipped to white, which is obviously not great. But the mode0 shot also has these areas, just less of them. Both could use a little less exposure time so that you dont saturate the target in question. Stars will saturate to some degree always, so no point in peeping those pixels.

There is a formula you could use (if you want to) to figure out an exposure time that i sufficiently long to bury read noise by a certain factor, for example 5x which is already very good. You multiply your read noise in electrons by the desired swamp factor, and square that result to get the target number of electrons per pixel of background.

So for the mode1 image with 2.07e of read noise and a desired swamp factor of x5 you get: 2.07x5 = 10.35. 10.35^2 = 107e of background signal. Im not sure how much offset you have in these images, but i am guessing it is 250ADU with 25 offset, you will find better technical advice in the QHY268 specific thread for sure but that is my guess.

Subtracting that from an area of the image with no nebulosity i get 655ADU, which with an e-/ADU conversion rate of somewhere around 0.5 (judging from the graphs, its not clear exactly what the number is) gets you 327e- of background signal so you could expose for a third of the time you currently do and still swamp read noise x5. Shortening the subs will help or remove the saturation issue but not have any real downsides, apart from having to take loads more of the subs since they are shorter. Storage is cheap these days and most stacking software are fast so not a big deal in my opinion.

Thanks Onikkinen, time to get my calculator out!

They are Ha subs! I am surprised at how much signal is getting through, they are Antlia HD 4.5nm filters 

2 hours ago, geeklee said:

Is that with SharpCap and have you been testing the camera in SharpCap?

If so, the top right I think has the capture settings.  Ensure its 16bit raw and not 8bit. 

Sorry, was busy with family 

I’m using Nina for captures (which is new to me too) I’ve checked and looks like I’m capturing 16bit fits 

Camera analysis crashed on dark frames so I’ve abandoned it for now, I don’t want to waste this short clear spell this evening so I’m capturing data but using mode 0 photographic mode at gain 26 as that seem work for me with the Samyang 135 

Thanks for your help 

Bryan

17 minutes ago, geeklee said:

The example was an older ZWO 183MM at unity gain (111 - no relation to the QHY numbers).  So a camera with a smaller full well than a 268/2600 camera.

The FITS header says they're 16bit, but is there anywhere that may have set it to only capture in 8bit, but a 16bit FITS output?

👍 Maybe adjust the title to read "QHY 268M subs help - do these look right?" or similar.

Good idea 

I’ll change the title and check the output of the camera 

I’m trying to run a sensor analysis atm but the light is fading 🙄 I’ll probably have to do it again with a light panel 

Thanks again

Bryan 

8 minutes ago, geeklee said:

I'm not familiar with using the QHY camera Bryan but agree with your assessment.

To my eye Mode1 has a lot saturated, even the stronger nebulosity!  Mode0 looks better but some of the bright nebulosity (like the tip of the "fish head") is saturated.  Very odd - I may be way off but it seems like the full well is very small but that wouldn't be right in either of those modes (from what I do know!)

If I remember, this is your Samyang lens in front of it?  Here's that fish head on a mono CMOS, ~F2.6, Samyang 135, 180s sub.

How long were your exposures?  These were raw straight off the camera?

Sorry, I've not been much help.

You’ve been a great help, just by answering my question the best you can 

Both subs were 120s @ f2.8 

I get the feeling somethings not right, but I cannot compare it to another camera, and I haven’t got the experience with cmos yet

That 180s sub you have added, do you know what camera and/or mode used? 

I’m hoping someone with a qhy268 might be able to add a comment 

Thanks again 

Bryan

Hi Stargazers 

I was wondering if someone with a better skillset than I currently have could cast an eye over these 2 subs 

My research has led me to consider using high gain mode,gain 56, offset 25 with my qhy268m but I think the result of these settings look weird to my ccd trained brain 

Is this what I should expect?

mode1.fit

Here is a sub at mode 0 gain 26 offset 25 which looks more like what I would expect but is it just that I need many more of the mode 1 subs?

mode0.fit

Any help would be greatly appreciated 

Thanks 

Bryan