I recently captured the Helix planetary nebula during a rare run of clear nights under my City suburban B8 Skies during the New Moon period with mixed results.
Acquired the following data over 5 nights -
160 x Ha 3 min dithered subs
152 x Oiii 3 min dithered subs
106 x Sii 3 min dithered subs
Total integration approx 21 hours
Full calibration suit -
Darks
Flats
Flat Darks
8” f5 Klaus Helmerich Carbon Newt
Skywatcher EQ6-R pro mount
Imaging camera ZWO2600MM , cooled to -10C , Gain 100
PHD2 Multistar guiding ( 0.45 to 0.60 arc sec )
Tracking and Goto EQMOD and Ascom Stellarium
Captured with APT
Analysed , calibrated ,stacked and aligned in ASTAP
Post processed in Startools V1.8
The end result was a bit disappointing considering the amount of data acquired as I expected to expose more outer detail which ended up being just noise.
Faint objects like the Helix definitely require substantially more data to resolve fine detail under heavy light polluted B8 skies.
I decided to capture Sii data which is extremely faint but ended up just stacking noise which didn’t help.
In hindsight I should have captured significantly more Oiii data as this filters signal is severely compromised with noise under B8 , 10 times worse if the Moon is up.
Also conditions were average with poor seeing from 8pm to 10pm on each night.
Anyway processing was very conservative as I didn’t want to expose the background noise , gradients and undulations etc.. so background does appear clipped but it’s not as Startools won’t allow the blackpoint to be 0.
What were your NB filter bandwidths, 6/7 nm or 3nm? I live in a similar city area in Sydney under 7 or 8 bortle scale near Macquare uni. I imaged Helix a couple of times before, with 12nm and 6nm using a Canon 400/f2.8 and Skywatcher Quattro 10 with ASI1600. Regardless how many hours I did (over 50), outer edges near side of the galaxy-like structure wouldn't show up. Last week though, with 3nm filters and quattro 12/ASI2600mm these area would be visible in a couple of hours. I'm having problem with my CEM60 after changing to a pier tripod at the moment with ultra low DEC guiding errors but much higher in RA. I was hoping to find the cause and post an image.
Hi Martin, great to see your results from our run of good weather. I certainly wouldn't be upset with that result. - with some very nice detail revealed from your heavy LP location, especially in the brighter core. I agree, with 21hrs I would hope to see some more of the outer extents, but it's very very hard when the noise floor is so high. I'd love to see what the Ha b/w image looked like - perhaps that would show more faint detail without the noise from the Sii / Oiii affecting the signal?
On a side note - the jet stream has been running wild the last couple of weeks and that has really affected the seeing and stability.even at shorter fl.
It certainly seems a bit disappointing regarding the faint outer extensions given the time you've put in, although there's still some nice detail in the brighter parts despite the seeing.
I'm wondering if longer subs would help? Perhaps the Helix is a dark-sky object only if you're hoping to go really deep.
I had plans to set up the larger scope for a few days given the clear weather we've had, but with the wind spoiling things I decided to just use the small refractor (apart from one night where the wind died down).
This HOO image is based on best 90% of 11 hours Ha (+6h 45m O3) but to be honest I was seeing everything I was going to see after half that time.
Cheers
Gary
Thanks Gary,
Based on simple SNR maths under B8 skies one would expect slightly more fine detail in those outer shells with say approx 35 to 40 hrs of integration but diminishing returns beyond that point.
I experimented with a stack of around 12 hours of data and the eyebrow was just visible, obviously the noise floor was significantly higher as well.
I did image the Helix with my 2600MC OSC about 2 years ago with around 8 hours of data and did manage to expose the lower eyebrow. I do believe Sydney’s Skyglow is getting worse each year. How much I don’t know but it’s definitely getting worse.
Cheers
Martin
What were your NB filter bandwidths, 6/7 nm or 3nm? I live in a similar city area in Sydney under 7 or 8 bortle scale near Macquare uni. I imaged Helix a couple of times before, with 12nm and 6nm using a Canon 400/f2.8 and Skywatcher Quattro 10 with ASI1600. Regardless how many hours I did (over 50), outer edges near side of the galaxy-like structure wouldn't show up. Last week though, with 3nm filters and quattro 12/ASI2600mm these area would be visible in a couple of hours. I'm having problem with my CEM60 after changing to a pier tripod at the moment with ultra low DEC guiding errors but much higher in RA. I was hoping to find the cause and post an image.
Hi Martin, great to see your results from our run of good weather. I certainly wouldn't be upset with that result. - with some very nice detail revealed from your heavy LP location, especially in the brighter core. I agree, with 21hrs I would hope to see some more of the outer extents, but it's very very hard when the noise floor is so high. I'd love to see what the Ha b/w image looked like - perhaps that would show more faint detail without the noise from the Sii / Oiii affecting the signal?
On a side note - the jet stream has been running wild the last couple of weeks and that has really affected the seeing and stability.even at shorter fl.
Thanks Dave,
Yeh really faint objects at my focal length and aperture under 88 skies is hard work but we do our best.
I imaged the Dolphin a while back with 5min subs ( around 8 to 10 hrs of data ) and it too was a struggle when trying to stretch and expose more detail. The noise gradient just hits you and puts the brakes on the stretch.
I might just process the Ha alone on the Helix and see how much I can pull out before it gets grainy.
I reckon 35 to 40 hours on these dim faint targets in Narrowband under B8 skies is about the limit where anything beyond that is diminishing returns. The trouble is we just don’t get a window of opportunity during new moon to gather that amount of data and generally after a few months the target moves out of range as well. We do what we can.
Cheers
Martin
It certainly seems a bit disappointing regarding the faint outer extensions given the time you've put in, although there's still some nice detail in the brighter parts despite the seeing.
I'm wondering if longer subs would help? Perhaps the Helix is a dark-sky object only if you're hoping to go really deep.
I had plans to set up the larger scope for a few days given the clear weather we've had, but with the wind spoiling things I decided to just use the small refractor (apart from one night where the wind died down).
Thanks Pete,
I’ve tried longer subs (5min) on the Dolphin a while back , same rig plus 10 hrs of data and again it was a struggle to expose fine detail, still a very high noise floor.
As I mentioned above I reckon 35 to 40 hrs of data under B8 skies on my rig is about the max integration to expose detail and beyond that it would be diminishing returns.
In Sydney when it’s windy I just hang up and stretch some Bunnings polytarps to protect the rig , seems to work well and in addition shields my neighbours floodlights which spill over into my yard like Las Vegas
If it makes you feel better - I am currently sitting on 30+hrs of Ha/OIII data of the helix that I just can't seem to get 100% happy with... I keep adding more data thinking that at some point, I'm going to like it, but I fear that may not be the case... It's certainly not a target meant for a 400mm focal length...
New scope should be here tomorrow - maybe I abandon the 400mm data and try again at ~900mm
Well, overall, it's a good looking image really, in a classic colour palette, nice
Yes that amount of exposure does seem to suggest a little more faint stuff should be visible, maybe longer subs..?
Just for comparison re depth, this is 2hrs worth of 5min subs with my FLI Proline 16803 and 12" F3.8 Newt and a 12nm Astronomik Ha filter, in what was the usual crap seeing (at 1.6"/pix), from light polluted inner suburban Newcastle.
Mike
Last edited by strongmanmike; 09-09-2024 at 06:50 PM.
I've attached a smaller screenshot of my starless Ha from that image- I think the Ha definitely shows more when viewed on its own. I don't know if you've seen this, but there's a paper by O'Dell in 2004 which looked at the Helix structure in detail and pretty much I can see all the things that are described in Figure 20 (as you probably are seeing too), but interestingly he uses a SII image in Figure 6. I cant upload it as its too big.
Not a bad result Martin. There is more helix structure there if you stretch further (though the result isn't pretty here), so the detail is there. The trick will be balancing that weaker signal from those outer regions of the helix with the background noise (and that balance will be easier for you when you're working with the original image at full bit-depth rather than with the compressed 8 bit jpeg that is posted online)
Well, overall, it's a good looking image really, in a classic colour palette, nice
Yes that amount of exposure does seem to suggest a little more faint stuff should be visible, maybe longer subs..?
Just for comparison re depth, this is 2hrs worth of 5min subs with my FLI Proline 16803 and 12" F3.8 Newt and a 12nm Astronomik Ha filter, in what was the usual crap seeing (at 1.6"/pix), from light polluted inner suburban Newcastle.
Mike
to be fair to Martin, based on pixel size, optic speed, and sensor QE (see this old thread here) a KAF16803 sensor with 9micron pixels on a f3.8 optic with a 12nm NB filter is going to be nearly 40x more sensitive than a IMX571 sensor (with 3.76 micron pixels) on a f5 optic with 3nm filters, so its not surprising that the 16803/f3.8 combo was such a weapon for this application - even from a light polluted location.
If it makes you feel better - I am currently sitting on 30+hrs of Ha/OIII data of the helix that I just can't seem to get 100% happy with... I keep adding more data thinking that at some point, I'm going to like it, but I fear that may not be the case... It's certainly not a target meant for a 400mm focal length...
New scope should be here tomorrow - maybe I abandon the 400mm data and try again at ~900mm
Alex,
Thanks for the comments
You maybe imaging at shorter focal lengths but your skies are 50% darker than mine , makes a “HUGE” difference when imaging faint targets.
My focal length is only 1050mm by the way , not exactly long in the general scheme of things.
Cheers
Martin
Martin - have you tried software binning your data to increase SNR (would be at the expense of resolution of course, but you'd theoretically get a 4 fold increase in system sensitivity)
Well, overall, it's a good looking image really, in a classic colour palette, nice
Yes that amount of exposure does seem to suggest a little more faint stuff should be visible, maybe longer subs..?
Just for comparison re depth, this is 2hrs worth of 5min subs with my FLI Proline 16803 and 12" F3.8 Newt and a 12nm Astronomik Ha filter, in what was the usual crap seeing (at 1.6"/pix), from light polluted inner suburban Newcastle.
Mike
Thanks Mike for comments and comparisons
Much appreciated
Newcastle back in 2002 was probably a Bortle 6 compared to Sydney in 2024 at Bortle 8 to 9 , only a guess by the way
Next time your in Sydney take a drive along Canterbury road around Petersham and look up at the night sky , you’ll wonder why we bother imaging.
My LP reading is 18.22 mag/arcsec2 and getting worse each year
I’ve tried 5min subs on the Dolphin a while back and the noise floor was still fairly high which inhibited me from stretching too far to expose fine detail.
Catch 22 scenarios
Martin - have you tried software binning your data to increase SNR (would be at the expense of resolution of course, but you'd theoretically get a 4 fold increase in system sensitivity)
The Wipe module cleans up most of your images uneven gradients but noise is noise and you can only suppress so much until you hit a black point close to clipping.
Plus there’s a load of modules to help you pull out more detail whilst keeping noise at bay
I’m actually gathering more Ha on the Helix tonight , see how we go with more Ha data
Maybe try longer subs over 10 min. I looked at my shots last week with 3nm Antlia (same B8 skies I think). With 24 10m subs (4 hours, gain 200) outer details were there but need strong tweaks in Photoshop to bring out. Here is a results with 95 subs (cropped and downsized by 50% to bring within file size limit). Outer details can still be improved with more data but I'm giving up for now as I kept getting 2-3 times more guiding errors in RA compared to DEC resulting in oblong stars and less sharp details in the centre compared to my earlier images with Quattro 10.
Maybe try longer subs over 10 min. I looked at my shots last week with 3nm Antlia (same B8 skies I think). With 24 10m subs (4 hours, gain 200) outer details were there but need strong tweaks in Photoshop to bring out. Here is a results with 95 subs (cropped and downsized by 50% to bring within file size limit). Outer details can still be improved with more data but I'm giving up for now as I kept getting 2-3 times more guiding errors in RA compared to DEC resulting in oblong stars and less sharp details in the centre compared to my earlier images with Quattro 10.
Cheers,
John
Thanks John,
Your image look impressive !!
Your Quattro scopes are at f4 so you have significantly more aperture and faster light gathering capability than my smaller slower 8” f5 Carbon ie: more detail for less integration time.
I’m currently gathering more Ha data so I can just do more experimenting with processing, hopefully I can improve SNR even if it’s only up to 5% or thereabouts
See how we go ……,
