The "No EQ" DSO Challenge!

[alacant]

Here's a side by side. Both have moon and skyglow which removes the muddy orange. The better image is with the uv-ir I mentioned. Under €25 for both on AliExpress. HTH.

**Taken on an eq so not strictly on topic but I thought my experiment findings -all by coincidence of choosing the wrong filter in the dark- may help other achro ap users. If you want to eliminate the ca completely you can use a 495nm long pass (wratten 12) along with the uv-ir -useful for e.g. the horsehead where there is a bright star in the fov.

[parallaxerr]

That's one worth remembering alacant, thanks. If I can remove some of the orange/brown to start with, I'll be happy!

[alacant]

1 minute ago, parallaxerr said:

orange/brown to start with

...just the moon and skyglow then. No need to spend a fortune!

[Nigel G]

A comparison of  30s ISO 1600. With and without . .Different targets but same light pollution. 

 

 

Edited November 8, 2016 by Nigel G
resize

[SteveNickolls]

3 hours ago, parallaxerr said:

I've had a look back through the thread for posts about filters and came across two posts of interest.

@SteveNickolls, you mention using the semi-apo, have you tried it yet? I rate the 1.25" one I have highly for visual and am considereing the 2" for AP.

Hi,

The filter is a birthday present and I shouldn't be using it before next week. I will of course let you know how it performs :-)

Cheers,
Steve

[parallaxerr]

1 hour ago, Nigel G said:

A comparison of  30s ISO 1600. With and without . .Different targets but same light pollution. 

 

 

Impressive NIge, thats's the result I'm after. Let's hope the Baader comes up with the goods with a bit of fringe killing added in.

[Nova2000]

I was shooting the Nan with 4sec exposure at my camp site shot about 500.when I reached home I realized I didn't shoot in raw but in jpeg. Wasted my time. 

I'll still check if pixinsight will process it and upload the result. Will there be any problem if jpeg images are stacked? 

Edited November 8, 2016 by Nova2000

[Stub Mandrel]

Y0ou will lose detail, mostly from JPEG being 8-bit as much as from the compression, but you should still get something usable from 500 frames.

[parallaxerr]

Here's the result of Monday nights session with the ST120. M33, about 1hr of 30s subs, x50 darks, x50 flats and x50 dark flats.

The light frames suffered heavy LP so I'm not suprised there's still significant gradient in the image. I'm still learning startools so maybe more could be removed? Due to the noisy light frames I used x2 drizzle in DSS, which left me with a 12068x8024 image @ 1.08GB  I then binned to 25% in startools which did seem to smooth things out a bit.

The CA is there, as expected, but I don't find it overly offensive. Hopefully the semi-apo filter will help address it and the skyglow when it arrives.

Most of my attempts so far have been with about an hours worth of data, early in the evening. It's becoming clear I really need to capture more data, later at night, when the lights have gone off, target altitude allowing. Oh for a clear night on a weekend...and a new moon..and a target between 30° - 60°!!!!

Edited November 9, 2016 by parallaxerr

[happy-kat]

That's come out well.

Startools has great CA removal capability.

[parallaxerr]

Just now, happy-kat said:

That's come out well.

Startools has great CA removal capability.

Thanks Kat, I have readt that ST is good at removing CA but I don't know how. Will have to read up on it!

[happy-kat]

Somewhere on this long thread I did a brief process flow on doing it. My 135mm lens has CA.

Thinking about it I think Steve's processing workflow has moved on from that now.

It is on page 7 comment 2895088

Edited November 9, 2016 by happy-kat

[DorsetBlue]

Not as impressive as some others here but below are a couple quick experiments I did with a basic setup Monday night.  Both taken using a Canon 100D, prime focus via a T-mount 2x Barlow on a SW 100 Heritage Reflector (yes, the little Dobsonian) mounted on a SW SynScan AltAz mount.  Both 10x 30s exposures at max ISO and then stacked in DSS and lightly processed in StarTools.

M31

M92

A little soft on focus (I was in a hurry, experimenting with my GPCAM v2 at the same time).

[Filroden]

4 hours ago, parallaxerr said:

Here's the result of Monday nights session with the ST120. M33, about 1hr of 30s subs, x50 darks, x50 flats and x50 dark flats.

The light frames suffered heavy LP so I'm not suprised there's still significant gradient in the image. I'm still learning startools so maybe more could be removed? Due to the noisy light frames I used x2 drizzle in DSS, which left me with a 12068x8024 image @ 1.08GB  I then binned to 25% in startools which did seem to smooth things out a bit.

The CA is there, as expected, but I don't find it overly offensive. Hopefully the semi-apo filter will help address it and the skyglow when it arrives.

Most of my attempts so far have been with about an hours worth of data, early in the evening. It's becoming clear I really need to capture more data, later at night, when the lights have gone off, target altitude allowing. Oh for a clear night on a weekend...and a new moon..and a target between 30° - 60°!!!!

 

As you say, there is noticeable CA which hopefully the filter will reduce. There is also a colour balance issue with a definite green tint across the whole image. You can see it most strongly on the galaxy core and on the yellow/orange stars which all have a slightly acid tint! I could remove the green tint in Pixinsight but I don't know how you'd adjust colour in StarTools. Other than the CA, your stars are nice and tight, so adding more data should be a breeze. You've done a good job with any light pollution, as I can't detect any gradient (other than a flat colour shift).

I used to drizzle and reduce but I found it just added more processing time and little benefit. I think it's usually only recommended if you're seeing square stars. There are probably other, more efficient, ways of smoothing backgrounds within StarTools, allowing you to skip the drizzle steps.

[parallaxerr]

Thanks for the encouraging words Fil. The green seems to be a problem for me, in both scopes. When I use the colour tool in ST the image has a heavy green tint to it, same with my M31 attemp. I cap green to yellow and reduce green further with the sliders but then then struggle to balance red an blue.

I just reprocess the same data in an attempt to remove more noise and address the ca, but I think the image is too soft now.

I also didn't drizzle in DSS and only binned 50% which I think has proven to remove the noise a bit better.

 

Edited November 9, 2016 by parallaxerr

[Filroden]

31 minutes ago, parallaxerr said:

Thanks for the encouraging words Fil. The green seems to be a problem for me, in both scopes. When I use the colour tool in ST the image has a heavy green tint to it, same with my M31 attemp. I cap green to yellow and reduce green further with the sliders but then then struggle to balance red an blue.

I just reprocess the same data in an attempt to remove more noise and address the ca, but I think the image is too soft now.

I also didn't drizzle in DSS and only binned 50% which I think has proven to remove the noise a bit better.

I don't think it is soft. I think it's just a problem with the colour balance and CA. Otherwise it's the start of a nice project to add more data (especially with the new filters).

I had a quick play with the jpg in Photoshop. As it's very black clipped I had trouble manually adjusting the colours. I did find (for the second image) that reducing the red a lot and the green a little using curves helped. I then tried PS' autocolour and it did a much better job with one click!

StarTools remains a mystery to me so I don't know the equivalent way to do the above. 

[alacant]

6 hours ago, parallaxerr said:

The CA is there

Hi. We've a thread on the StarTools forum. Basically we decon the blue channel back onto the central star then pull the RG separately on top of the de-convoluted blue. It sounds complicated but do it a few times and it becomes second nature. Helping the weak Canon uv-ir filter -it records them as visible- at source with a proper uv-ir block filter really helps too. If you just want a quick fix, put the fat stars in a mask and the hit filter > fringe killer. The problem with the latter is that the halos are still there, just in a shade of grey. But hey, it sure beats having spikes sticking out from every star!  HTH.

[Nigel G]

2 hours ago, parallaxerr said:

Thanks for the encouraging words Fil. The green seems to be a problem for me, in both scopes. When I use the colour tool in ST the image has a heavy green tint to it, same with my M31 attemp. I cap green to yellow and reduce green further with the sliders but then then struggle to balance red an blue.

I just reprocess the same data in an attempt to remove more noise and address the ca, but I think the image is too soft now.

I also didn't drizzle in DSS and only binned 50% which I think has proven to remove the noise a bit better.

 

Nice image, good stars, just the slight colour issue, I use to have trouble with the colour module in ST but sussed the problem out.

Set ANY colour balance to no or off in DSS, then open image in ST with the second tab, Linear, was bayered is not white balanced.

I found that if I turned off the bayer arrangement under fits in DSS before registering I had terrible trouble with colours. so I set the camera to Canon Rebel  which has the same arrangement.

This seemed to sort the colour problem.

Nige.

[Nigel G]

This was direct from Ivo of StarTools on processing M33, It worked well for processing my M33. worth a try. You might have to change settings slightly or maybe not.

The image this processing was used on was from a refractor scope originally, mine was with a reflector but still had a good result.

Nige.

--- Auto Develop
To see what we got. Default settings. We see light pollution, noise, oversampling. Coma and/or camera not sitting quite flush.
--- Bin
To make use of oversampling and reduce noise.
Parameter [Scale] set to [(scale/noise reduction 50.00%)/(400.00%)/(+2.00 bits)]
--- Lens
To mitigate coma (do this before cropping!). Unfortunately there is no exact science to this... YMMV.
Parameter [Curvature Linked] set to [126.03 %]
--- Crop
Crop remaining artefacts and frame object a bit better.
Parameter [X1] set to [10 pixels]
Parameter [Y1] set to [17 pixels]
Parameter [X2] set to [1105 pixels (-407)]
Parameter [Y2] set to [920 pixels (-250)]
--- Wipe
Masked out the galaxy using Lasso tool.
Parameter [Dark Anomaly Filter] set to [8 pixels] to catch any Dark Anomaly pixels.
--- Auto Develop
Like M31, region of Interest over a slice of the disc, representing a good sample of the dynamic range that we're interested in.
Parameter [Ignore Fine Detail <] set to [6.0 pixels]
Parameter [Outside ROI Influence] set to [8 %], reducing dynamic range allocated to non-RoI area a little.
--- HDR
To dig out a bit more detail. Reveal preset.
Parameter [Detail Size Range] set to [1293 pixels]
Parameter [Strength] set to [1.1]
--- Color
Used MaxRGB mode to balance the green a little. We're looking for a yellower inner disc, bluer outer disc, pink/purple HII areas dotted around, as well as a good representation of all star colors in the foreground stars.
Parameter [Dark Saturation] set to [6.00] to introduce color in the shadows.
Parameter [Saturation Amount] set to [129 %]
Parameter [Blue Bias Reduce] set to [1.10]
Parameter [Green Bias Reduce] set to [1.04]
Parameter [Red Bias Reduce] set to [1.29]
--- Life
I used the Isolate preset (no particular mask set, just everything selected) to push back the noisy background. This technique is also good for busy star fields in very/extreme wide fields.
--- Contrast
To dig out a bit more detail from the core. There is a slight spiral structure to M33, it just hard to make visible.
Parameter [Compensate Gamma] set to [Yes]
Parameter [Precision] set to [256 x 256 pixels]
Parameter [Dark Anomaly Filter] set to [7 pixels]
--- Wavelet De-Noise
Parameter [Grain Size] set to [15.2 pixels]
Parameter [Smoothness] set to [82 %]
 

Edited November 9, 2016 by Nigel G

[The Admiral]

9 hours ago, DorsetBlue said:

Not as impressive as some others here but below are a couple quick experiments I did with a basic setup Monday night.  Both taken using a Canon 100D, prime focus via a T-mount 2x Barlow on a SW 100 Heritage Reflector (yes, the little Dobsonian) mounted on a SW SynScan AltAz mount.  Both 10x 30s exposures at max ISO and then stacked in DSS and lightly processed in StarTools.

M31

M92

A little soft on focus (I was in a hurry, experimenting with my GPCAM v2 at the same time).

That's a good start DorsetBlue. When I imaged M13, for one reason and another I only had 26 x 30s subs from a 102mm 'scope to stack, but that still gave me a reasonable result. I think that your 10 subs need to be increased somewhat. But experimenting is what it's all about (or would be if only we had a lot more usable nights!)

Ian

Edited November 9, 2016 by The Admiral

[The Admiral]

10 hours ago, parallaxerr said:

Here's the result of Monday nights session with the ST120. M33, about 1hr of 30s subs, x50 darks, x50 flats and x50 dark flats.

The light frames suffered heavy LP so I'm not suprised there's still significant gradient in the image. I'm still learning startools so maybe more could be removed? Due to the noisy light frames I used x2 drizzle in DSS, which left me with a 12068x8024 image @ 1.08GB  I then binned to 25% in startools which did seem to smooth things out a bit.

The CA is there, as expected, but I don't find it overly offensive. Hopefully the semi-apo filter will help address it and the skyglow when it arrives.

Most of my attempts so far have been with about an hours worth of data, early in the evening. It's becoming clear I really need to capture more data, later at night, when the lights have gone off, target altitude allowing. Oh for a clear night on a weekend...and a new moon..and a target between 30° - 60°!!!!

Jeez, I've not tried drizzle for the very reason you give, mammoth file sizes from a DSLR!

But it doesn't seem to be an easy target; although it's large it's not very bright and I think much competition from sky-glow doesn't help. I ended up taking over 300 subs and only stacking about 200, but even then I found it hard to process, to give at best a just about acceptable result. Still, I would guess that with the right manipulation of ST you could get a bit more out of that.

Ian

[Nigel G]

10 hours ago, DorsetBlue said:

Not as impressive as some others here but below are a couple quick experiments I did with a basic setup Monday night.  Both taken using a Canon 100D, prime focus via a T-mount 2x Barlow on a SW 100 Heritage Reflector (yes, the little Dobsonian) mounted on a SW SynScan AltAz mount.  Both 10x 30s exposures at max ISO and then stacked in DSS and lightly processed in StarTools.

Using a barlow will shorten the possible exposure times by about 50% and also reduce the amount of photons reaching the sensor, I think your better off at prime focus and then cropping to required size.

Good to get 30s with a x2 barlow WD

Still good images though  

Nige.

Edited November 9, 2016 by Nigel G

[The Admiral]

16 minutes ago, Nigel G said:

Using a barlow will shorten the exposure times by about 50% and also reduce the amount of photons reaching the sensor, I think your better off at prime focus and then cropping to required size.

Nige.

I like to think of a Barlow as spreading the photons over a larger area of the sensor, and therefore giving rise to magnification, but each pixel will receive fewer photons and hence the exposure will need to be longer. I don't know what percentage of photons are 'lost' in the optics, but with good coatings I shouldn't think it will amount to much. Or am I missing something here?

Ian

Edit. Ah, perhaps you mean that the exposure times you can get out of the mount when using a Barlow are shorter!

Edited November 9, 2016 by The Admiral

[The Admiral]

6 hours ago, parallaxerr said:

....The green seems to be a problem for me, in both scopes. When I use the colour tool in ST the image has a heavy green tint to it,....

 

Try adjusting the colour bias sliders using the "Max RGB" option. You may still not be able to get rid of the green, but you can get an overall impact of the colour changes you are making. See page 42 of the manual.

Ian

[Nigel G]

8 minutes ago, The Admiral said:

I like to think of a Barlow as spreading the photons over a larger area of the sensor, and therefore giving rise to magnification, but each pixel will receive fewer photons and hence the exposure will need to be longer. I don't know what percentage of photons are 'lost' in the optics, but with good coatings I shouldn't think it will amount to much. Or am I missing something here?

Ian

You have a better way of explaining this  My experience with barlows, tracking is harder as twice the magnification will double the errors of the mount compared to prime focus ? roughly twice the exposure time needed due to loss though optics and reduced field of view ?

Something like that. I may be completely wrong of course  

Nige