M42 critique & Multi-night imaging.

[CraigD1986]

Hi All.

Attached is my first ever attempt of M42. The image comprises of 54min 20sec of data (20 seconds exposures), darks, flats and dark flats. Stacked in DSS and processed in Adobe Photoshop. Critique and suggestions are always welcome.

The clouds are clearing so I’m getting back out there tonight as I want to add some more data to this. When you add more data to an existing image, do you re-stack all the data from scratch and start again?

Also, how do I go about stacking data with different exposure lengths? I’m thinking of getting some 5 second exposures for the core and then some 2+min exposures for the darker nebulosity. Would I stack these two sets separately and then open all three images in photoshop as layers, to mask and expose different parts of the three layers? How do you go about ensuring the processing is consistent between different stacks? I hope that makes sense.

[powerlord]

No idea if its the 'proper' way, but yes  - I process as seperate stacks. then bring each into affinity photo (photoshop for you), and mask, etc to add different bits using 'add' blend more. sometime with differnt opacity.

[The Lazy Astronomer]

That's very nice for a first attempt!

For a critique, I would say the stars appear to have an elongated shape and are possibly a bit bloated - former may just be tracking errors and for the latter; do you use a UV/IR cut filter? If not, I'd recommend you look for one which cuts the lower end of the blues and upper end of the reds, such as the Astronomik L3.

Looking at this image, l don't think you need shorter exposures. The core is not blown out, and I think the trapezium would be resolved if you could get tighter stars (improve tracking, focus, and reduce star bloat).

It may also surprise you to learn you don't need longer exposures either. I've taken the equipment in your signature and made some assumptions for the calculation (72ed at native f ratio, unity gain and bortle 5 skies), and 20s is probably the ideal minimum exposure time for you (well, 19s actually). What this means is, you are already exposing properly - going longer will not net you any significant gains.

When adding new data, yes, restack all frames from scratch. Take care to ensure the framing between nights is exactly the same (or as close as you can get it) to avoid having to do some heavy cropping of stacking artifacts. 

[CraigD1986]

Thanks for the help!

I did discuss the option of a UV/IR filter with FLO when I purchased the camera and telescope but at the time I was looking at using the camera with my Sigma camera lens so the filter was more about solving problems with the camera lens. I see a very prominent blue shadow to the bottom right of each star. Is this the kind of thing a UV/IR filter will fix? I was considering the Optolong L-Extreme filter as my next purchase but this may change things.

I assumed that some longer exposures (2 min maybe) would reveal a lot more of the darker, outer nebulosity. When stretching this image, I would bring out quite a bit more but not without blowing the core.

I focused using a Bahtinov mask (Albeit a home made one) so I assume the focus is correct but I could be wrong. 

[happy-kat]

With viewing on a mobile the signature is hidden, it would be helpful to add what camera, telescope etc was used please?

Infrared is often blocked by the camera, but not reading what camera was used can't say if the camera natively blocked IR 

[CraigD1986]

Sorry, didn't realise the signature would be hidden:

 

Sky-Watcher 72ED DS-Pro, ZWO ASI533MC Pro,

Star Adventurer Pro on Manfrotto 055 Carbon fibre tripod,

ASIAir PLUS, ZWO 120mm guide scope & camera

[happy-kat]

As far as I can read the ASI533MC has only a Anti Reflective window, you might want to do your own reading to check, a IR UV filter would be needed and see what other people with the camera are doing.

The L Extreme probably includes IR UV line blocking too, check the specs.

 

Edited February 7, 2022 by happy-kat

[powerlord]

Yup u need a uv/ir with asi533.

Lextreme is great. Do remember you will need much longer exposures so make sure your mount and guiding is up to it.

Stars wise, take some short exposures and stack them for stars.

Use starnet to remove the stars from your stacked sets once they have been stretched a bit. Then you have clean sharp stars from the short exposure stacking you can use. You can then focus on stretching and combining your different exposure stacks and out the stars back at the end.

Tip: to get stars only, put starless image in layer above the one with stars. Make it as subtract blend type. You should now be seeing only stars. Save that as an image.

To put stars back, add a new layer and add that image, and make layer blend Add. You can then play with star opacity to reduce stars as u wish.

Stu

Edited February 7, 2022 by powerlord

[The Lazy Astronomer]

1 hour ago, CraigD1986 said:

I see a very prominent blue shadow to the bottom right of each star. Is this the kind of thing a UV/IR filter will fix? I was considering the Optolong L-Extreme filter as my next purchase but this may change things.

That is likey to be blue bloat - the 72ed is a doublet, so not perfectly corrected for all wavelengths. I would think a UV/IR cut filter would help greatly with that, but like I said, you'd probably want one that also cuts the extreme ends of the blue and red wavelengths as well (I use the Astronomik L3 on my triplet).

The L-extreme should also cure that problem as it's a narrowband filter and will cut pretty much all of the other wavelengths except those around ha and oiii, but bear in mind it is only suitable for emission nebulae, and yes, you will need longer exposures if you go down that route. 

1 hour ago, CraigD1986 said:

I assumed that some longer exposures (2 min maybe) would reveal a lot more of the darker, outer nebulosity.

This is a common misconception, and intuitively, it seems to make sense ("to capture fainter detail, I must expose longer"), but once you've reached your ideal minimum exposure time, the improvements are very marginal.

1 hour ago, CraigD1986 said:

When stretching this image, I would bring out quite a bit more but not without blowing the core.

There are a few ways around this:

1) stretch using curves and focus on stretching the lower end of the curve more than the higher end

2) create 2 copies of the image, stretch the first one for the best core and bright detail, stretch the second one for best faint detail and then blend the 2 together

3) use some kind of hdr processing algorithm to recover highlights, such as those offered in Startools, PI or other astro processing software (not sure if there's any equivalent in PS or the like)

1 hour ago, CraigD1986 said:

I focused using a Bahtinov mask (Albeit a home made one) so I assume the focus is correct but I could be wrong. 

Bahtinovs are ok, but it can be very difficult to get precise focus with them. I'm not sure what software you're using for capture, but consider looking at hfr measurements when focusing, and trying to get the smallest number (if your capture software provides that information, which it should).

[CraigD1986]

Thanks for all the responses. Looks like I’ll be adding a UV/IR filter to my wish list. FLO recommended the ZWO one to me previously but I’ll look into these a little more.

@powerlord Thanks for the detailed advice. I’ll have to do a fair bit of research for this as I’m still a novice and a lot of that is beyond my capabilities at the moment. I’ll make an effort to learn it though.

@The Lazy Astronomer Thanks for the advice. Would you recommend just capturing some more 20 second subs and building on the data that way? I’m using an ASIAir plus and the focus feature is very useful so will concentrate on star sizes rather than using the Bahtinov mask next time.

I’m pretty sure I got the backspacing correct between the 72Ed and ASI533 but just to check, I have the 72ED, then the StellaMira 2” field flattener, the WZO 21mm spacer, then the 16.5mm spacer, then the small spacer that it fitted to the camera when it comes out of the box.

[jacko61]

According to FLO the stellamira flattener needs 55mm back focus so you have that set up right   

FWIW I use the Baader UC/IR cut filter with my ASI533MC and see a lot less colour fringing on my stars than beforehand.

Graeme

[CraigD1986]

Thanks all. Will have a look into filters this evening.

[CraigD1986]

On 07/02/2022 at 12:33, The Lazy Astronomer said:

That's very nice for a first attempt!

For a critique, I would say the stars appear to have an elongated shape and are possibly a bit bloated - former may just be tracking errors and for the latter; do you use a UV/IR cut filter? If not, I'd recommend you look for one which cuts the lower end of the blues and upper end of the reds, such as the Astronomik L3.

Looking at this image, l don't think you need shorter exposures. The core is not blown out, and I think the trapezium would be resolved if you could get tighter stars (improve tracking, focus, and reduce star bloat).

It may also surprise you to learn you don't need longer exposures either. I've taken the equipment in your signature and made some assumptions for the calculation (72ed at native f ratio, unity gain and bortle 5 skies), and 20s is probably the ideal minimum exposure time for you (well, 19s actually). What this means is, you are already exposing properly - going longer will not net you any significant gains.

When adding new data, yes, restack all frames from scratch. Take care to ensure the framing between nights is exactly the same (or as close as you can get it) to avoid having to do some heavy cropping of stacking artifacts. 

Sorry to bring up an old post but can I ask how you calculated the ideal exposure time here? I’m reading back through my old threads to try and improve

[The Lazy Astronomer]

4 hours ago, CraigD1986 said:

Sorry to bring up an old post but can I ask how you calculated the ideal exposure time here? I’m reading back through my old threads to try and improve

No worries - the general idea is to expose until your light pollution "swamps" your camera read noise. To calculate your optimum minimum sub exposure time, you only need to know two things:

1. Your camera's read noise (at your desired gain/ISO setting)

2. Your light pollution rate, in electrons per pixel per second - this will vary depending on your equipment and the pattern of your local light pollution, so you'll end up with a different number with different cameras, scopes and, most likey, when pointing at different areas of the sky (e.g. my light pollution is worse in the east as I'm shooting towards London). I tend to sort of ignore this last variable and just use a generic light pollution measurement from the web.

Getting the read noise figure is easy, decide on a gain and look up manufacturers spec. For the 533, ZWO state read noise for unity gain (100) is 1.5e-

To get your light pollution rate is a little more involved. Our resident expert vlaiv has described several times how to calculate it experimentally using your own data, which is probably the most accurate way of doing it, but for life of me I can't remember how to do it! The other way is to calculate it theoretically - I use this tool: https://tools.sharpcap.co.uk/. Input your details and it'll spit out a number at the bottom.

Once you've got the numbers, the formula is then: exposure = C*RN^2/P, where RN is camera read noise, P is the light pollution rate, and C is a factor to determine how much additional noise is introduced into the image. This can basically be any number you want, and the higher the number, the lower the amount of additional noise. A C-factor of 10 equals approximately an extra 5% noise and is generally a good figure to settle on (there comes a point where visually there is no real discernible difference in the noise for different C-factors and I think after 10 it becomes diminishing returns).

When I did it last time using your details I think I just used the peak QE figure quoted by ZWO, but it might be better to approximate an average QE across all the colour channels instead, or even use the lowest figure.

And by the way, this is by no means gospel - it just sets out an exposure time that ideally you shouldn't drop below. If you want to go longer and your mount performance/guiding is up to the job, then go for it, just be aware that once you reach the optimum minimal sub time, exposing longer is very much a game of diminishing returns. 

[CraigD1986]

@The Lazy Astronomer Thanks so much for the detailed explanation. That's so helpful. I'm going to have a look into this in detail later today. So this number doesn't vary based on the target then and theoretically there isn't any reason to expose for too much longer than the result of this formula?

I don't intend to stick to the exposure times religiously, it's just good to have a rough idea. As a complete novice, I just assumed that the longer the exposure, the better but now realise that this isn't often the case. For example, many 30 second exposures can yield the same result as fewer 5 min exposures if the total integration time is the same. I also assumed that I would need longer exposure times in order to capture the darker parts of nebulosity like that surrounding the main area of M42.

I'm looking forward to this target coming back around in the winter so I can improve on the result (hopefully).

[The Lazy Astronomer]

2 hours ago, CraigD1986 said:

@The Lazy Astronomer Thanks so much for the detailed explanation. That's so helpful. I'm going to have a look into this in detail later today. So this number doesn't vary based on the target then and theoretically there isn't any reason to expose for too much longer than the result of this formula?

I don't intend to stick to the exposure times religiously, it's just good to have a rough idea. As a complete novice, I just assumed that the longer the exposure, the better but now realise that this isn't often the case. For example, many 30 second exposures can yield the same result as fewer 5 min exposures if the total integration time is the same. I also assumed that I would need longer exposure times in order to capture the darker parts of nebulosity like that surrounding the main area of M42.

I'm looking forward to this target coming back around in the winter so I can improve on the result (hopefully).

Correct, choice of target has no impact, and yes, theoretically, once the optimal minimal exposure is reached, all the faint/dark detail has been recorded to a sufficient enough level and a longer exposure won't record any further detail. 

[CraigD1986]

44 minutes ago, The Lazy Astronomer said:

Correct, choice of target has no impact, and yes, theoretically, once the optimal minimal exposure is reached, all the faint/dark detail has been recorded to a sufficient enough level and a longer exposure won't record any further detail. 

That’s so refreshing to hear. I read all the time about people trying to get subs of 5, even 10 mins so just assumed that’s what it takes. Can’t wait for the dark nights now

[The Lazy Astronomer]

3 hours ago, CraigD1986 said:

That’s so refreshing to hear. I read all the time about people trying to get subs of 5, even 10 mins so just assumed that’s what it takes. Can’t wait for the dark nights now

5 or 10 minute subs (or longer) may be needed with narrowband filters, or cameras with high read noise, or under very dark skies, but for broadband imaging under a typical suburban sky with a modern astro cam it's way overkill IMO.

[happy-kat]

Reading earlier posts I see the mount is a star adventurer which might shape the exposure lengths anyway, are you using guiding as that I gather opens up the possibility of longer subs.

[CraigD1986]

On 02/07/2022 at 10:21, happy-kat said:

Reading earlier posts I see the mount is a star adventurer which might shape the exposure lengths anyway, are you using guiding as that I gather opens up the possibility of longer subs.

Yes, guiding via ASiAir plus and 120mm mini camera. Even with a SA, I’ve found that 5 mins plus is easily achievable.