First Attempt at M51 - how to improve?

[Sarek]

This is my first attempt at M51.  It may be a bit distant for my current set up but I thought it was worth a shot as I've managed to get my guiding working pretty well in recent sessions and skies were very clear here on Wednesday evening.

This is about an hours worth of integration using 300s exposures. I also took 25 darks, flats and dark flats to add to my bias frames. I processed it in PI using the Autointegrate script and tweaked a bit afterwards.

Other than getting more integration time I'm keen to learn if there are other thngs I can do to improve on this before looking for a scope (maybe SW150PDS) with more reach and/or investing in a dedicated astro camera (probably ASI 533MC pro).  My skies are around Bortle 4 so I havn't looked at getting pollution filters yet.

 

 

 

[Clarkey]

I think for an hour of imaging this is a good result. Star shapes look good and there is a fair bit of detail, especially for a small scope. I would question the need for 5 minute exposures, 2 or 3 is probably enough. More subs will help with the intergrated result.

Other than that, I would be pretty pleased. After the summer, your 72ed scope will be ideal for nebulae and some of the larger targets so I would stick with it for now. (My kit has now been packed up for the summer - not enough darkness now until the end of August).

[wimvb]

9 hours ago, Sarek said:

My skies are around Bortle 4 so I havn't looked at getting pollution filters

No need for a light pollution filter. They will only mess with your colour balance. As @Clarkey wrote, this is a very nice image already. Just add more of the same. Since you are using a dslr camera, you might want to reconsider the use of darks (for lights and flats). Without temperature control, darks sometimes do more harm than good. Pixinsight has a process called cosmetic correction, which helps get rid of hot and cold pixels. You can use it in the wbpp script or "manually".

[Sarek]

1 hour ago, Clarkey said:

I think for an hour of imaging this is a good result. Star shapes look good and there is a fair bit of detail, especially for a small scope. I would question the need for 5 minute exposures, 2 or 3 is probably enough. More subs will help with the intergrated result.

Other than that, I would be pretty pleased. After the summer, your 72ed scope will be ideal for nebulae and some of the larger targets so I would stick with it for now. (My kit has now been packed up for the summer - not enough darkness now until the end of August).

Thank you. It's interesting to hear what you say about shorter exposures and defintely something I'll try.  I've had some success with guiding so that was the reason for pushing out to 5 minutes.

I know what you mean about the shorter nights - I didn't get to bed until 3:30am for this session!

49 minutes ago, wimvb said:

No need for a light pollution filter. They will only mess with your colour balance. As @Clarkey wrote, this is a very nice image already. Just add more of the same. Since you are using a dslr camera, you might want to reconsider the use of darks (for lights and flats). Without temperature control, darks sometimes do more harm than good. Pixinsight has a process called cosmetic correction, which helps get rid of hot and cold pixels. You can use it in the wbpp script or "manually".

Thank you. I've heard some conflicting advice about the use of darks but there's no harm in experimenting without so I'll give that a go. That would also mean I could get to bed earlier!! Thanks for the PI insight 

[dark knight]

I thought with dslr it was better to use bias frames. Maybe give them a try.

[Coco]

Good effort! I'm assuming you heavily cropped this image, might be worth not doing it with this scope and camera combination, wide field galaxies ( use the 'rule of thirds' ) look fantastic.

[Sarek]

29 minutes ago, Coco said:

Good effort! I'm assuming you heavily cropped this image, might be worth not doing it with this scope and camera combination, wide field galaxies ( use the 'rule of thirds' ) look fantastic.

Thank you. Yes, it was heavily cropped. As you suggest I might try something a little less aggressive

[Sarek]

39 minutes ago, dark knight said:

I thought with dslr it was better to use bias frames. Maybe give them a try.

Yes, I have all the data so defintely worth a try!

[wimvb]

43 minutes ago, dark knight said:

I thought with dslr it was better to use bias frames. Maybe give them a try.

Yes, because darks don't always work with uncooled dslrs. In an ideal world, darks should exectly match in terms of temperature, iso/gain and exposure time, in order to remove the digital signature of the camera. For CCD cameras it was generally allowed to scale flats, adjusting the exposure time difference. But cmos sensors, including those in dslrs, don't always allow this. That's why it may be better to refrain from using darks if you can't get them to match the exposures they are to correct (flats and lights). The workflow then becomes:

• Integrate bias frames to create a master bias
• Calibrate flats with the master bias and integrate to create a master flat.
• Calibrate lights with master bias and master flat
• Use cosmetic correction if you see hot or cold pixels
• DeBayer lights
• Register lights
• Integrate lights to create a master light

If you use this workflow and dither at least 12-15 pixels between exposures, you will end up with a clean master image.

[Sarek]

4 hours ago, wimvb said:

Yes, because darks don't always work with uncooled dslrs. In an ideal world, darks should exectly match in terms of temperature, iso/gain and exposure time, in order to remove the digital signature of the camera. For CCD cameras it was generally allowed to scale flats, adjusting the exposure time difference. But cmos sensors, including those in dslrs, don't always allow this. That's why it may be better to refrain from using darks if you can't get them to match the exposures they are to correct (flats and lights). The workflow then becomes:

• Integrate bias frames to create a master bias
• Calibrate flats with the master bias and integrate to create a master flat.
• Calibrate lights with master bias and master flat
• Use cosmetic correction if you see hot or cold pixels
• DeBayer lights
• Register lights
• Integrate lights to create a master light

If you use this workflow and dither at least 12-15 pixels between exposures, you will end up with a clean master image.

Thank you. I've yet to try dithering although I did watch a YT video that recommended 2 pixels every 3rd frame? I guess I can try a range but I'm interested to know why you recommend 12-15  - is that related to exposure time, higher exposure time  = more pixels and higher frame frequency?

Edited April 22, 2023 by Sarek

[wimvb]

19 minutes ago, Sarek said:

Thank you. I've yet to try dithering although I did watch a YT video that recommended 2 pixels every 3rd frame? I guess I can try a range but I'm interested to know why you recommend 12-15  - is that related to exposure time, higher exposure time  = more pixels and higher frame frequency?

The recommendation originally comes from astrophotographer Tony Hallas. DSLR cameras have, what he calls "color mottle", large scale colour noise. To efficiently get rid of it, you dither with larger steps. This talk may be a bit dated, but it is still one of the best introductions

 

[alacant]

20 hours ago, Sarek said:

other thngs I can do

Hi

Excellent shot. Just keep adding frames, but if you want to give it a go...

700d? Hands on over several years with the 18Mp sensors and this model in particular:

- iso 800
- No dark frames of any type
- No bias frames
- Dither around 5-10px between each frame
- Use a clipping algorithm during stacking: sigma 5-2 works well for us

Processing:

Simply subtract the offset (2048) from the flat and light frames.
But of course, try differing values and combinations for yourself and please post back if you find anything better.

Cheers and HTH

Edited April 22, 2023 by alacant

[WolfieGlos]

For an hour of imaging, that's a good result! Especially the lack of noise in the sky. What ISO are you shooting at? I'm at ISO-400 (which suits my camera), which requires...just like you 🙂 ....  300s subs in bortle 4 Gloucestershire. I use NINA for capturing and I dither every 2 subs, since I found every 3 was giving me a bit of walking noise. But I'm not sure what my dither distance is, so I might check this - thanks Wim and alacant.

I dropped dark frames once I started dithering when I got my HEQ5 4 months ago. I found the darks were masking walking noise (which I didn't realise I was getting). I did consider using LENR on the camera to get around this but everywhere you look for astrophotography suggests not to use it, since it doubles your exposure time (and so halving your imaging time).

I use only flats and biases now, never bothered with dark-flats.

Definitely put some more imaging time into this target though, I got this with the 72ed nearly a year ago to the day. It was my first target once I got a scope. 87 x 120s subs, ISO-1600 on the star adventurer. I'm sure it could be processed better a year on, but actually even though it's a smaller galaxy it seems to do well with our scope 🙂

[Sarek]

5 hours ago, wimvb said:

The recommendation originally comes from astrophotographer Tony Hallas. DSLR cameras have, what he calls "color mottle", large scale colour noise. To efficiently get rid of it, you dither with larger steps. This talk may be a bit dated, but it is still one of the best introduction

Thanks for that. Great video which explains the requirements very clearly 

 

4 hours ago, alacant said:

Hi

Excellent shot. Just keep adding frames, but if you want to give it a go...

700d? Hands on over several years with the 18Mp sensors and this model in particular:

- iso 800
- No dark frames of any type
- No bias frames
- Dither around 5-10px between each frame
- Use a clipping algorithm during stacking: sigma 5-2 works well for us

Processing:

Simply subtract the offset (2048) from the flat and light frames.
But of course, try differing values and combinations for yourself and please post back if you find anything better.

Cheers and HTH

Thank you, that's given me plenty to work on. I'll definitely try these things. 

[Sarek]

53 minutes ago, WolfieGlos said:

For an hour of imaging, that's a good result! Especially the lack of noise in the sky. What ISO are you shooting at? I'm at ISO-400 (which suits my camera), which requires...just like you 🙂 ....  300s subs in bortle 4 Gloucestershire. I use NINA for capturing and I dither every 2 subs, since I found every 3 was giving me a bit of walking noise. But I'm not sure what my dither distance is, so I might check this - thanks Wim and alacant.

I dropped dark frames once I started dithering when I got my HEQ5 4 months ago. I found the darks were masking walking noise (which I didn't realise I was getting). I did consider using LENR on the camera to get around this but everywhere you look for astrophotography suggests not to use it, since it doubles your exposure time (and so halving your imaging time).

I use only flats and biases now, never bothered with dark-flats.

Definitely put some more imaging time into this target though, I got this with the 72ed nearly a year ago to the day. It was my first target once I got a scope. 87 x 120s subs, ISO-1600 on the star adventurer. I'm sure it could be processed better a year on, but actually even though it's a smaller galaxy it seems to do well with our scope 🙂

Thanks Chris. I'm always shooting at ISO 800 which I've read is the sweet spot for my camera. 

I really like the idea of dropping darks - I hate doing them as I'm usually ready for bed when I finish my subs!

I'm going to try dithering on my next session following the guidance in the video linked above and from alacant

As another factor to throw into the mix I'm not entirely sure my flats are great (I use the white t shirt method) so I've just ordered a cheap and cheerful light screen from Amazon

Our local council implemented a street lights off after midnight policy a couple of years ago so perhaps that helps with the sky noise?  

Vaughan

[pipnina]

Given as you are using PixInsight, I am surprised Spectrophotometric colour calibration hasn't been mentioned yet!

It works absolute wonders for colour rendition, even under limited signal conditions.

I processed this for a friend (not my data), who uses a Moded D3200 (full spectrum). The signal is clearly limited, I think moreso than your image. Yet with PixInsight SPCC the colour looks very natural and vibrant, especially in the cigar!

[WolfieGlos]

1 hour ago, Sarek said:

Thanks Chris. I'm always shooting at ISO 800 which I've read is the sweet spot for my camera. 

I really like the idea of dropping darks - I hate doing them as I'm usually ready for bed when I finish my subs!

I'm going to try dithering on my next session following the guidance in the video linked above and from alacant

As another factor to throw into the mix I'm not entirely sure my flats are great (I use the white t shirt method) so I've just ordered a cheap and cheerful light screen from Amazon

Our local council implemented a street lights off after midnight policy a couple of years ago so perhaps that helps with the sky noise?  

Vaughan

Hi Vaughan,

With Nina, I did have to increase my settling time for dithering from the default as I was getting slightly oblong stars (iirc, default was 20s and I upped it to 30).

I do the same with my flats, white t-shirt and an ipad screen zoomed into Google homepage!!! Set the camera to Aperture priority and adjust the brightness until it's suitable. I don't dismantle my imaging train or change rotation, so imaging the same target across multiple nights I use the same flats. But I then take new ones for a new target, different rotation or not. I've not noticed anything untoward so far with this method, but I would shoot more if something unusual appeared in a stack, say a new dustmote.

That's awesome of your council, I wish Stroud council would do that! I have an LED light across the road from me that blocks my view south. I enquired if they can add shields last weekend, I had a response immediately and am awaiting them to do the work...fingers crossed it will help. Looking north I have nothing but fields...

Edited April 22, 2023 by WolfieGlos

[Sarek]

20 minutes ago, WolfieGlos said:

Hi Vaughan,

With Nina, I did have to increase my settling time for dithering from the default as I was getting slightly oblong stars (iirc, default was 20s and I upped it to 30).

I do the same with my flats, white t-shirt and an ipad screen zoomed into Google homepage!!! Set the camera to Aperture priority and adjust the brightness until it's suitable. I don't dismantle my imaging train or change rotation, so imaging the same target across multiple nights I use the same flats. But I then take new ones for a new target, different rotation or not. I've not noticed anything untoward so far with this method, but I would shoot more if something unusual appeared in a stack, say a new dustmote.

That's awesome of your council, I wish Stroud council would do that! I have an LED light across the road from me that blocks my view south. I enquired if they can add shields last weekend, I had a response immediately and am awaiting them to do the work...fingers crossed it will help. Looking north I have nothing but fields...

The YT video I watched today was suggesting 1 minute for settling time but maybe that's overkill if you have no problems with 30 seconds?

Interesting to hear you're able to use same flats for several same target sessions. Another time savig tip! I must admit I wasn't thinking of using a t shirt and the light screen but no bother to do so if thats best?

I'm continually surprised by how much guiding accuracy can change during a session.  Last wednesday it was stuck over 1.5rms for ages but later went consistently below 0.8.  I guess it might be atmospheric conditions are at play. One thing I forgot to check was if there was any change after the meridian flip. I have my rig balanced slighly east heavy so wanted to know if that would cause a problem after flipping .

Good that you might have a solution from Stroud CC. I'm in a small village in South Glos and with lights off its pretty dark. 

I noticed a lot of satellite trails before midnight and was relieved to find that stacking removed them!  All this software is pretty amazing!

 

 

[WolfieGlos]

26 minutes ago, Sarek said:

The YT video I watched today was suggesting 1 minute for settling time but maybe that's overkill if you have no problems with 30 seconds?

Interesting to hear you're able to use same flats for several same target sessions. Another time savig tip! I must admit I wasn't thinking of using a t shirt and the light screen but no bother to do so if thats best?

I'm continually surprised by how much guiding accuracy can change during a session.  Last wednesday it was stuck over 1.5rms for ages but later went consistently below 0.8.  I guess it might be atmospheric conditions are at play. One thing I forgot to check was if there was any change after the meridian flip. I have my rig balanced slighly east heavy so wanted to know if that would cause a problem after flipping .

Good that you might have a solution from Stroud CC. I'm in a small village in South Glos and with lights off its pretty dark. 

I noticed a lot of satellite trails before midnight and was relieved to find that stacking removed them!  All this software is pretty amazing!

Dithering time, I would say try at the default and increase if necessary. I noticed odd star shapes in some of my subs, and tracked it down to being right after a dither. I tried 30s and it worked, as we have the same scope and imaging train, in theory should work for you too. I haven't watched the video, but it might be for a heavier scope (?).

I still use the t-shirt! I read elsewhere a year ago or so now, it's something to do with how screens emit light and that maybe the t-shirt helps to diffuse it (?). Correct me if I'm wrong on that, but I just continue with the method as it works, but I'm sure others (who know more about this) can advise. I'd also be interested to hear about it, always looking to refine and improve. I think it's best to take flats with an astro cam every session, as they have electronic shutters that don't wear out. If I can save 35-50 shutter actuations every session that's a bonus for my DSLR's lifespan.

In fact, at the minute I've been imaging M81/M82 until 12:30 AM, then I slew across to Markarians Chain (once it's above my roof and past that pesky southern streetlight) and image that until 3/4 am....or mist like last night at 2am. I'm using the same flats and biases for both of these two, as there's no change in focus or to the imaging train. Initial stacks appear OK, albeit lacking integration.

Agreed on guiding, I've never set my rig up to be east or west heavy, but something I need to look into. I generally get 0.8 RMS, but last night I re-calibrated and it was between 1.3-1.5 for most of the night, sometimes dropping to 0.9. I suspected it was cable snags, but nothing looked unusual. That's interesting about sky conditions, will that affect it?

 

[pipnina]

2 hours ago, Sarek said:

The YT video I watched today was suggesting 1 minute for settling time but maybe that's overkill if you have no problems with 30 seconds?

Interesting to hear you're able to use same flats for several same target sessions. Another time savig tip! I must admit I wasn't thinking of using a t shirt and the light screen but no bother to do so if thats best?

I'm continually surprised by how much guiding accuracy can change during a session.  Last wednesday it was stuck over 1.5rms for ages but later went consistently below 0.8.  I guess it might be atmospheric conditions are at play. One thing I forgot to check was if there was any change after the meridian flip. I have my rig balanced slighly east heavy so wanted to know if that would cause a problem after flipping .

Good that you might have a solution from Stroud CC. I'm in a small village in South Glos and with lights off its pretty dark. 

I noticed a lot of satellite trails before midnight and was relieved to find that stacking removed them!  All this software is pretty amazing!

 

 

I think wind affects it more than we think. Even in winds clearoutside classes as green we can get gustsa few times in a sub that wobble things about a bit.

[edarter]

You might already know this from the YT videos etc but the dithering value you need to put in your guiding software needs to take in to account to the difference between your imaging setup and your guiding setup. I found the following calculator really useful to help me home in on the right setting for NINA.

https://www.astrohowto.com/interactive-dithering-calculator/

Hth

Ed

[WolfieGlos]

12 hours ago, edarter said:

You might already know this from the YT videos etc but the dithering value you need to put in your guiding software needs to take in to account to the difference between your imaging setup and your guiding setup. I found the following calculator really useful to help me home in on the right setting for NINA.

https://www.astrohowto.com/interactive-dithering-calculator/

Hth

Ed

Thanks Ed. I have to confess that I just ran with the defaults, Nina’s default is set to 5 pixels and that calculator tells me that it should be 6. I haven’t noticed anything untoward so far, but I will try changing it on my next target and see what difference(if any) it makes. 

[Clarkey]

On 22/04/2023 at 23:34, WolfieGlos said:

I still use the t-shirt! I read elsewhere a year ago or so now, it's something to do with how screens emit light and that maybe the t-shirt helps to diffuse it

I have used the screen on a tablet without diffusing and it is fine. The only reason I use a t-shirt is if the flats are too bright. My led flats panel has a sheet of perspex over it with about 5 sheets of paper behind that. Dims it and diffuses - so solves both problems.