The "No EQ" DSO Challenge!

[The Admiral]

Well, I guess it's time to add my contribution . A bit of an experiment this. First off, I haven't had a go at M45 before, so this is a first. Knowing that we are dealing with a lot of bright stars here, I thought I'd use ISO400 rather than my usual ISO1600 to improve my DR, and I retained the 30s exposure to maximise the photon count to get the nebulosity. The usual flats, lights, darks and bias. Also, because it's a bit of a pain using DSS with my Fuji RAFs as I first have to convert all the files to DNG, I've bought a copy of AstroArt. This will happily accept my RAFs, but of course it is a new application so I'm still feeling my way (the manual rather assumes that you know what you are doing!). I started off with AA, but I got some peculiar artefacts around the bright stars. OK, now try with DSS. That turned out OK, but I normally use Kappa-Sigma on my lights in DSS so I thought I'd try the equivalent in AA again. Hey presto, that got rid of the artefacts . So I've now got a direct route to stacking my Fuji files. Great .

I processed the O/Ps of DSS and AA in StarTools. Then I thought I'd have a go in AA, and to be honest, pretty comparable to ST. But then again, the target probably isn't the most challenging. I'm sure that there is a lot of improvement here, but still, it's a start. All three O/Ps needed polishing in Lightroom.

This is the result using AA, but as I say, they all look much the same.

Looking at the numbers that DSS throws up, the maximum score ~400, which is way less that I'd normally get with my reducer/flattener, so perhaps the conditions weren't that conducive to good imaging. Or perhaps I was a bit more cr** this time! I'll have another go on Monday/Tuesday, Met Office permitting.

An interesting (dare I say, "cool"?) feature of AA is that you can look at a 3-D image of the star field:

From that I'm guessing that a lot of the stars are saturated (hence the flat tops), so I'd need rather less exposure, but then either I I'd have to take an awful lot of short exposure subs, or do two sets with different exposures. Hmm

Ian

 

[The Admiral]

12 minutes ago, Nigel G said:

This is a little work with photoshop express on the original .

I have another stack to process tomorrow,  last night's 245 subs were 50 % before street lights out and 50 % after, it's just the way it worked out. 

That last stack is the latter 50 %  . Hopefully it will be better to process. 

Nige.

 

Is this the image where you've tried to reduce the star field Nige? I think that somehow you need fewer of them rather than just less intensity. I'm not sure how you'd do that though, unless you can construct a mask to include only the smaller/dimmer ones and then remove those stars. Haven't a clue about doing that though, I'm not good with masks!

Ian

[Filroden]

15 minutes ago, The Admiral said:

This is the result using AA, but as I say, they all look much the same.

You've just started to capture where the reflection transitions into the dust with the reddish/brown area to the right of Merope! Some of your stars could have their colour intensity boosted. The colour is there (the yellow/amber ones) just a little pale. I find they add a little balance to what can be a very over powering white/blue image otherwise. How many subs was this?

I also found my stars a little more bloated last night. It was partly the falling temperatures and I had to refocus many times (and even then I think I didn't do it often enough) but I think there might have been a very slight haze.

[Nigel G]

Ian, good image of M 45, I must get more data for mine. Good to see you experimenting too . Plenty of nebula and dust there.

the Rosette I posted is just a dimming of the highlights from the original.  I think it will improve with the  last stack ( processing tomorrow ) I will try to reduce star numbers or there size somehow 

There was a lot of moisture in the air last night. I think that's half the trouble with processing both my images. My scope had dew after about 40 minutes, I had to demist twice in 1.5 hours. The flower pot dew shield worked a treat with the lens though, 2.5 hours no problem . But seeing wasn't that good.

Nige

[Nigel G]

M1 is being a pig to process,  just no detail and loads of noise that the offsets didn't remove,  literally a white fuzzy blob in the middle of a noisy star field. 

[alacant]

1 hour ago, Nigel G said:

Can I do that with data from a dslr? 

Absolutely. The LRGB module will pull only the channel(s) you choose, be it separate files or a colour stack. HTH.

[parallaxerr]

Hello, Alt/Az’ers, it’s nice to see some of you had a productive evening yesterday. For me, last night was a tale of two halves. I have nothing to show for my efforts, however, I believe I learnt a LOT about my set-up and the direction I need to go in. This may turn out to be a long post so apologies in advance as I’m finding it hard to get all these thoughts straight in my head, but I find sharing my findings here to be rewarding as someone always responds with a little nougat of previously unknown information!

So what went wrong?

Well....first off, due to it being the first clear night in a while, I thought I’d make the most of it and hit several targets one after another as Alt and/or FOV limits came in to play. First off was M31, then M33, then I decided to have a pop at the horse head. What this meant, was that I never really got enough subs on any of my targets to produce anything decent with.

Secondly, I decided to try a change of ISO, choosing 3200 because Sensorgen.info suggests than read noise for my camera drops off at higher ISO. But, and this is a BIG BUT...despite claiming files are recorded in RAW, Nikon apply an internal de-read noise to RAW files. In years past, this process was known as the “Nikon star eater” filter in the AP world, due to the de-noise process mistaking stars for hot-pixels. However, this is reported not to be the case with more recent Nikons.

Finally a flat battery! I have previously been impressed with the battery life of my Nikon, but last night it gave up half way through taking darks.

I came away with a load of subs on various targets but after inspection, many were binned due to thin cloud and satellite passes and ultimately I did not record enough darks.

Moving on.....The good news is that the tweaks I made to the mount and balancing of the OTA has done away with star trailing on the alt axis that I experienced during my last session. So, I decided to process a few images anyway, without flats/dark flats to see what I had and made some pretty important observations, I think.

What I learned...

ISO - Previous attempts at M31 have been at ISO 1600. On processing in ST, I struggled with a lot of background noise which I thought was LP. As such, I fitted my Baader semi-apo filter (an LPfilter + fringe killer) to try to combat it, however, I don’t think LP was the issue as I saw exactly the same when processing last nights data, in fact worse. As well as a decrease in read noise, higher ISO brings with it reduced saturation capacity and thus dynamic range – I “think” the noise I’m seeing is a result of the noise floor being too high, i.e. the dynamic range being too small. Histograms aren’t clipped but stretching the image brings out terrible grainy red background sky, which, I did not see when imaging M42 at ISO800 where I have a few more stops of DR to play with. So, whilst I’m not saturating the sensor, maybe I’m not allowing enough to capture the full DR of the target and the noise floor is being brought up in ST.

Semi-apo filter – If LP isn’t the cause of the background noise then this isn’t going to cure it. Proven last night. With such a small scope I also think cutting any light wavelengths out is counterproductive and the green tint it casts is a pest in colour processing.

Dark frames – I don’t think I need them! Darks frames are dark from my DSLR, as in completely black. No noise, not a jot. I have stacked last nights M31, with & without darks and the difference is ZERO. Recent Nikons are famed for very low read noise sensors and the in-built noise-reduction negates the need for me to take dark frames. (Note – there is an option to turn long exposure noise-reduction off, which I have done, but read noise reduction is still applied). There is a Nikon hack program available called “dark current enabler” which does away with the NR, I have tried it and got darks as we’re used to seeing with the purple/red noise, but it’s a PITA to use and requires a PC. But, if the in-built NR isn’t eating stars, then what’s the point anyway?

Aperture – I’m struggling with this. Having tried the ST120 a few weeks back, it’s clear it gathers photons far faster than the ZS66 – half the exposure time produced twice the detail on M33, notwithstanding some heavy CA. Exposure time vs CA? On a bit of a lose lose here as far as my available scopes are concerned, but I found the CA objectionable in the achromat – I wonder if reducing the ISO reduces blue bloat? Either way, not having the alt limit (once I get my new intervalometer cable) with the ZS66 is nice and I’ve ordered a second battery for the camera to allow longer sessions...If only the weather would allow. More data is always a plus, regardless of aperture, it’s just going to take me a little longer to gather the photons.

So, I feel like I have come to these conclusions (until I have another great learning experience):

·         Plan imaging sessions better – one target at a time.

·         Maximum ISO – 800.

·         Read noise for my camera is irrelevant.

·         Semi-apo filter is not required.

·         Don’t waste exposure time on darks.

For any of you that have read this far, I thankyou! Maybe this will give others something to think about and please feel free to correct me where I am mistaken, probably everywhere.

Jon

 

 

 

[Filroden]

Ok, I know the background on this one is very messed up! 

Here's my first version of M42 on my alt-az mount using the Skywatcher Esprit 80 and the ZWO ASI1600MM-cooled. 

All subs at -20C, 300 gain and 50 offset:

30 x 1s x L

2 x 45s x L

58 x 30s x L

15 x 30s x R, G & B

All calibrated with bias, darks and flats and processed in PixInsight

For comparison, here's my M42 for a year ago with the 9.25" SCT

[Filroden]

4 minutes ago, parallaxerr said:

Plan imaging sessions better – one target at a time.

You've written quite a lot above but I thought it easier to pick the points off in parts!

I actually find it useful to take short runs of subs on a few targets better than intensive sessions on a single target. I build up files of data for each target and try to hit the targets many times over the course of a month. That way I spread the seeing across all the targets. If one night is a bust I'm losing 60 subs from the target and not 400!

Also, I image in runs of about 45 minutes (which is 60 subs). That way I can monitor altitude of the target and avoid the camera hitting the mount. It's surprising how quickly some targets rise! Swapping between targets lets me stay in a nice zone of 20-40 degs.

[The Admiral]

49 minutes ago, Filroden said:

You've just started to capture where the reflection transitions into the dust with the reddish/brown area to the right of Merope! Some of your stars could have their colour intensity boosted. The colour is there (the yellow/amber ones) just a little pale. I find they add a little balance to what can be a very over powering white/blue image otherwise. How many subs was this?

I also found my stars a little more bloated last night. It was partly the falling temperatures and I had to refocus many times (and even then I think I didn't do it often enough) but I think there might have been a very slight haze.

Yes I know what you mean Ken. I did try the colour module in a quick run-through in ST and the stars took on a lot of colour, but the nebulosity didn't respond well. May be with a bit more finessing it could come good. I've yet to discover how I'd do that in AA!

I took 180 x 30s subs, most stacked by AA, about 150 in DSS, if I remember correctly.

Ian

Edited November 26, 2016 by The Admiral

[happy-kat]

That is such a nicer image from your Espirit, more subtly, more definition, clarity.

I've been checking back over several pages you all seem to favour imging to the East, is this because it is just coincidence that you can all only see East, only from that chart pages back shared I thought East and West were the same for imaging benefit?

[The Admiral]

In my case, it's just that East and South are the only directions not affected by town lights. That, and that targetting towards the East allows me to capture objects in the evening as they rise, and before they reach too high an altitude.

Ian

Edited November 27, 2016 by The Admiral

[The Admiral]

10 hours ago, Filroden said:

You've written quite a lot above but I thought it easier to pick the points off in parts!

I actually find it useful to take short runs of subs on a few targets better than intensive sessions on a single target. I build up files of data for each target and try to hit the targets many times over the course of a month. That way I spread the seeing across all the targets. If one night is a bust I'm losing 60 subs from the target and not 400!

Also, I image in runs of about 45 minutes (which is 60 subs). That way I can monitor altitude of the target and avoid the camera hitting the mount. It's surprising how quickly some targets rise! Swapping between targets lets me stay in a nice zone of 20-40 degs.

That's a very laudable approach Ken, but with the frequency of usable nights we get I think I'd worry I'd end up with a pile of unfinished targets!

Ian

[Filroden]

10 hours ago, happy-kat said:

is this because it is just coincidence that you can all only see East,

I only have a view to the east and even then it's more north east. Orion was fast moving beyond the line of the house at midnight. 

[Filroden]

6 minutes ago, The Admiral said:

That's a very laudable approach Ken, but with the frequency of usable nights we get I think I'd worry I'd end up with a pile of unfinished targets!

Ian

I'm just too impatient! I can't stand the idea of spending five hours imaging and only getting one image. 

[The Admiral]

11 hours ago, parallaxerr said:

What I learned...

ISO

Dark frames

For any of you that have read this far, I thankyou! Maybe this will give others something to think about and please feel free to correct me where I am mistaken, probably everywhere.

Jon

You raise all the vexing questions that I guess we've all been through, though we may still haven't come to any conclusions. The issue of image noise and ISO are very pertinent here.

I don't know if you've read Craig Stark's article " Profiling the Long-Exposure Performance of a Canon DSLR ", but if not I think it's worth doing so. It seems that manufacturers get up to all sorts of tricks in order to give decent conventional images, which play against our use for astrophotography. Every image is affected by dark current and read noise, which manifest themselves, if you like, as a feint background glow. By taking bias frames (for the read noise) and dark frames this 'glow' can be subtracted out. The problem is, both the read noise and dark current have a random component of noise which can't be subtracted out. The best one can do is to take multiple frames and stack them, so that the random noise averages out to a smaller and smaller component. It seems that what Canon does is reduce the effect of the dark current (i.e. 'glow') before it is written to raw, but what it can't do is reduce the noise from the dark current, which continues to grow as the ISO is increased. So all is not what it seems. I must admit from that perspective I'm debating the worth of doing darks, but that said, I'm not sure to what extent it corrects for things like fixed pattern noise. You say that the darks you take are completely black. Is that after severe stretching?

+1 for getting a spare battery. I'm sure we've all be caught out by that one and a spare is essential. Don't forget also that in the very cold nights their capacity will be reduced, so keep your spare warm before use.

Ian

[The Admiral]

13 minutes ago, Filroden said:

I'm just too impatient! I can't stand the idea of spending five hours imaging and only getting one image. 

Me to, but in my case I couldn't bare to spend so many hours on one target .

Ian

[The Admiral]

11 hours ago, Filroden said:

Ok, I know the background on this one is very messed up! 

Here's my first version of M42 on my alt-az mount using the Skywatcher Esprit 80 and the ZWO ASI1600MM-cooled. 

All subs at -20C, 300 gain and 50 offset:

30 x 1s x L

2 x 45s x L

58 x 30s x L

15 x 30s x R, G & B

All calibrated with bias, darks and flats and processed in PixInsight

For comparison, here's my M42 for a year ago with the 9.25" SCT

Nice one Ken. Good to see the complete loop of dust and nice tight stars. The Trapezium is 'just' resolvable I think, though not as clear as with your SCT. It's hard to get a handle on the difference in scale.

I'm led to understand that the bright area around the Trapezium should be greenish when in true colour, an unusual colour to find, from the oxygen luminescence. This I think is more evident in your SCT image than the Esprit one.

Ian

[Filroden]

4 minutes ago, The Admiral said:

Nice one Ken. Good to see the complete loop of dust and nice tight stars. The Trapezium is 'just' resolvable I think, though not as clear as with your SCT. It's hard to get a handle on the difference in scale.

I'm led to understand that the bright area around the Trapezium should be greenish when in true colour, an unusual colour to find, from the oxygen luminescence. This I think is more evident in your SCT image than the Esprit one.

Ian

Thank you. And yes, I lack colour resolution in the core which means I can't really pull it out. After I packed up that night I realised I hadn't taken 1s exposures in RGB so I only had the overexposed core from the 30s subs to draw colour. The good news is it will only take 2 minutes of captures to put that right  

I'm also wondering whether to also take a sequence at about 10s to help the transition between core and outer areas. 

I wonder if I can blend the SCT image into this?

Edited November 27, 2016 by Filroden

[Nigel G]

Jon. 

I'm with Ken on this, after Friday nights imaging session and a a few other long sessions I end up cropping huge amounts from my images unless I rotate my camera which I don't like to do. Also I keep thinking there's other targets I want to get. 

By gathering upto 1 hour of data on a target in a night there are many pluses, less rotation,  more targets, more chances to change settings and experiment, less chance of getting fed up looking at the same image for hours at a time, and you will see the difference adding more subs does to your image's by processing after each session. 

There are down sides too, the main one being British weather.

Experimenting is quite exciting for me, finding out the unknown, each setup is different an acts differently,  the whole Alt-Az imaging subject is unknown,  nobody knows the limits, we are searching and experimenting every night we image, since Ken got his cooled CCD I have been gob smacked by his results, taking Alt-Az a big step further forward, all of us are advancing with our techniques in imaging and processing.

Your personal kit needs to be tweaked and tuned to each nights imaging, iso settings, exposure times lens or scope choice and so on.

So your finding out what you need to know.

Now with darks and flats my theory is..... DSS makes a master dark frame from many frames and subtracts it from your light frames. The trouble is each light frame has moved slightly from the last one so the camera noise is not in the same place. Taking loads of darks doesn't seem to be profitable in my mind, to be honest if I use 15 or 50 it doesn't seem to make a difference,  others will disagree with me but that's the unknown. Flats too, a master flat is produced by DSS and the effects are removed from the final image, the same thing has happened,  your lights have rotated  and the lens effects have rotated too, so I use just 12 to 15 flats with same effect that 50 flats has, they are worth taking though.

One of my best images was 30 minutes total exposure, I think 10 minutes of dark and a hand full of flats, I have imaged the same target for 3 hours, added 70 darks and had a hell of a job processing the outcome, reducing the lights by 50 % and the darks by about 65 % to get a better result, easier to process and less noise, this is probably due to seeing conditions improving through the night. 

It's all unknown, that's a great thing.

Your findings are important to your imaging  keep it up.

Cheers

Nige.

[Filroden]

2 minutes ago, Nigel G said:

Now with darks and flats my theory is..... DSS makes a master dark frame from many frames and subtracts it from your light frames. The trouble is each light frame has moved slightly from the last one so the camera noise is not in the same place. Taking loads of darks doesn't seem to be profitable in my mind, to be honest if I use 15 or 50 it doesn't seem to make a difference,  others will disagree with me but that's the unknown. Flats too, a master flat is produced by DSS and the effects are removed from the final image, the same thing has happened,  your lights have rotated  and the lens effects have rotated too, so I use just 12 to 15 flats with same effect that 50 flats has, they are worth taking though.

I don't think field rotation has any impact on dark removal. The darks are removed from each individual light before they are rotated for integration, so the hot pixels are always in the same place.

The problem with darks and noise is more, I think, to do with the DSLR chips being very warm. It's not possible, even in a quickly taken stack of images, to maintain a steady enough temperature at the chip, so the noise is varying with temperature and this cannot be matched with the lights, which are also being affected by temperature. The only solutions I know are to either not use darks (using just bias as darks) or cooling the chip (serious modification territory). I stopped using darks with the DSLR for that reason.

[Nigel G]

11 hours ago, happy-kat said:

That is such a nicer image from your Espirit, more subtly, more definition, clarity.

I've been checking back over several pages you all seem to favour imging to the East, is this because it is just coincidence that you can all only see East, only from that chart pages back shared I thought East and West were the same for imaging benefit?

Personally I have London 25 miles to my West, 2 large towns to my South,  my bungalow to the north and the dark North sea 5 miles to my east with 1 small village in between.  

Nige

[Nigel G]

Ken , I take darks between taking the lights to get full benifit of them, so I'll take 30 light the 10 dark, then 30 light 10 dark and so on to try and capture the camera and outside conditions at different temperatures etc. 

Thanks for pointing out about stacking darks though, your right. The master dark is subtracted from each frame before being stacked.

Nige.

[SteveNickolls]

12 hours ago, parallaxerr said:

So, I feel like I have come to these conclusions (until I have another great learning experience):

·         Plan imaging sessions better – one target at a time.

·         Maximum ISO – 800.

·         Read noise for my camera is irrelevant.

·         Semi-apo filter is not required.

·         Don’t waste exposure time on darks.

Hi, we have different DSLR's but through experience I agree with the conclusions you have come to-

• You need to stick to one target per session and get at least an hours worth of light frames, hopefully more to get a decent final image. If you need more frames carry out consecutive night imaging to cut down the effects of field rotation. The exceptions to an hour of images can be clusters and globular clusters.
• Iso 800 is fine to keep your signal more to the left hand side of your histogram for later stretching.
• Due to the different dark noise routines built into DSLR's (and they are probably all commercially secret and different), plus our ability to not be able to control the sensor temperature then dark frames aren't worth taking. 
• I haven't had chance (due to the bad weather here) to use my semi-apo filter but I do hope it has a benefit with the bloated stars etc.

Cheers,
Steve

[Filroden]

Just now, Nigel G said:

Ken , I take darks between taking the lights to get full benifit of them, so I'll take 30 light the 10 dark, then 30 light 10 dark and so on to try and capture the camera and outside conditions at different temperatures etc. 

Thanks for pointing out about stacking darks though, your right. The master dark is subtracted from each frame before being stacked.

Nige.

I've just realised there is one other solution, which is to take a dark after each light (effectively using most DSLR's ability to do noise reduction) but this wastes precious imaging time. I guess what you're doing is somewhere between the two extremes, taking fewer darks between runs of lights so any change in temperatures are accounted for. Another alternative is to use a hot pixel map rather than a dark to correct pixels.