Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

NGC3718 Part 5

  • Please log in to reply
8 replies to this topic

#1 pyrasanth

pyrasanth

    Soyuz

  • *****
  • topic starter
  • Posts: 3,987
  • Joined: 08 Jan 2016

Posted 18 April 2021 - 01:07 PM

Where do I begin on this target?

 

I have over 50 hours of data to work through and the image presented is one third of that data using the C14 Edge at F11 and the ASI6200. This is the RGB data collected on the C14 over many weeks and consists of  over 350 5 minute subs Using Baader LRG filters. The target is very faint from my Bortle 8/9 location and to get the faint outer galaxy edges takes very careful processing otherwise it gets lost in the background noise. I have a poweful friend in the ASI6200 which is very low noise. 

 

The image posted has been resampled from 1x1 bin to 4x4 bin to ease test processing- however I will be producing the final image at full resolution when I have all the data which will be a massive undertaking as it will run to TB's when I have to process multiple registrations.

 

I'm really looking forward to processing the L data and HA which is running to over 20 hours.

 

As always comments happily received as I begin to process the live data- the testing however looks promising- It is better to act on your comments now as the image developes!

 

Stay safe..

 

Best wishes

Mark

 

NGC3781_V5_CN.jpg


Edited by pyrasanth, 18 April 2021 - 01:50 PM.

  • Alterf, rwstanley, dswtan and 13 others like this

#2 Lord Baltimore

Lord Baltimore

    Explorer 1

  • -----
  • Posts: 83
  • Joined: 08 Jul 2018
  • Loc: Sweden

Posted 18 April 2021 - 01:46 PM

Congratulations buddy. Some hard work there.


  • pyrasanth likes this

#3 spiantino

spiantino

    Explorer 1

  • -----
  • Posts: 87
  • Joined: 24 Jun 2020

Posted 18 April 2021 - 03:04 PM

Awesome image! I'd be curious to see what the full res vs. 4x4 looks like. I've got an order in for the EdgeHD 9.25 (which I will receive who knows when) but my big worry is that even that scope with the 6200 is unnecessarily oversampling for anything except the clearest nights in my location


  • pyrasanth likes this

#4 Mert

Mert

    Cosmos

  • *****
  • Posts: 7,723
  • Joined: 31 Aug 2005
  • Loc: Spain, Pamplona

Posted 18 April 2021 - 03:20 PM

That looks awesome Mark, colors are vivid and the faint outer area is much more visible.
Hard work.. yes, worth the
effort...YES
Great job!

Edited by Mert, 18 April 2021 - 03:21 PM.

  • pyrasanth likes this

#5 pyrasanth

pyrasanth

    Soyuz

  • *****
  • topic starter
  • Posts: 3,987
  • Joined: 08 Jan 2016

Posted 18 April 2021 - 03:52 PM

Awesome image! I'd be curious to see what the full res vs. 4x4 looks like. I've got an order in for the EdgeHD 9.25 (which I will receive who knows when) but my big worry is that even that scope with the 6200 is unnecessarily oversampling for anything except the clearest nights in my location

I read an interesting story- I don't recall the name of the observatory in question. On a very large telescope the main camera broke and a replacement had to be hastily put in its place whilst the broken camera was fixed. One of the colleagues expressed a quip that the temporary camera had small pixels when compared to the camera that had broken- the replacement was a KAF16803 sensor which has 9 um pixel size- the broken camera had orders of magnitudes larger.

 

This proves that pixel size is relative- on a very long focal length telescope in professional observatories they might be using the equivalent pixel size of 36 um plus which is apparently massive to use but when you calculate the arc secs per pixel on a very long focal length telescope you can see that the equation fits very nicely- a long focal length telescope loves big pixels!- lets take a look at your C9.25. The camera at 1x1 bin has a pixel size of 3.7 um approx so at the native resolution we get an arc sec size of  ( (3.76/2350)x 206.3) = 0.33 arc secs per pixel which is a massive oversample for typical 1-3 arc secs per pixel average seeing. There is a very nice solution bin the camera a 4x4- that will give you a very nice 1.32 arc secs per pixel which is pretty much very good seeing- drop to 3x3 bin to get a tad under 1 arc sec per pixel. The camera is a win- do not be frightenened of binning- one of the greatest CCD cameras is the 16803 sensor and that has relativey big pixels but its great on long focal length telescopes especially when binned to 2x2 or 3x3.

 

You will have a great system and camera- have no fear and push the over sampling demons back to hell.


Edited by pyrasanth, 18 April 2021 - 03:53 PM.


#6 pyrasanth

pyrasanth

    Soyuz

  • *****
  • topic starter
  • Posts: 3,987
  • Joined: 08 Jan 2016

Posted 18 April 2021 - 04:00 PM

That looks awesome Mark, colors are vivid and the faint outer area is much more visible.
Hard work.. yes, worth the
effort...YES
Great job!

Thanks Mert- I'm working the RGB data sets to the full native resolution as I type this. It is a bit of a pig as on some of the subs the filter wheel stuck in odd framing positions so I'm having to use selective rejection to deal with a lot of weird edge vignetting where each time the filter wheel rotated it did not move back to the same position. I have the FLI filter wheel on order so that problem will be negated in a few weeks as well as being able to carry the full narrow band and broad band filter set at 50x50 mm in the one wheel. That is going to cut down on a lot of work as changing between RGB and narrow band wheels meant a recalibration due to the naff filter wheel. I can then get the broken wheel serviced.

 

We go through a lot of pain to capture them photons!



#7 pixlimit

pixlimit

    Mariner 2

  • -----
  • Posts: 207
  • Joined: 20 Apr 2016
  • Loc: Hamburg, Red Zone

Posted 19 April 2021 - 03:57 AM

Excellent image Mark, great job on the processing too!

 

Peter


  • pyrasanth likes this

#8 R Botero

R Botero

    Gemini

  • -----
  • Posts: 3,060
  • Joined: 02 Jan 2009
  • Loc: Kent, England

Posted 19 April 2021 - 04:32 AM

Great image Mark - there's so much detail in it.  I think you can push your processing even further as there's so much signal in it.

 

Roberto



#9 spiantino

spiantino

    Explorer 1

  • -----
  • Posts: 87
  • Joined: 24 Jun 2020

Posted 19 April 2021 - 01:02 PM

I read an interesting story- I don't recall the name of the observatory in question. On a very large telescope the main camera broke and a replacement had to be hastily put in its place whilst the broken camera was fixed. One of the colleagues expressed a quip that the temporary camera had small pixels when compared to the camera that had broken- the replacement was a KAF16803 sensor which has 9 um pixel size- the broken camera had orders of magnitudes larger.

 

This proves that pixel size is relative- on a very long focal length telescope in professional observatories they might be using the equivalent pixel size of 36 um plus which is apparently massive to use but when you calculate the arc secs per pixel on a very long focal length telescope you can see that the equation fits very nicely- a long focal length telescope loves big pixels!- lets take a look at your C9.25. The camera at 1x1 bin has a pixel size of 3.7 um approx so at the native resolution we get an arc sec size of  ( (3.76/2350)x 206.3) = 0.33 arc secs per pixel which is a massive oversample for typical 1-3 arc secs per pixel average seeing. There is a very nice solution bin the camera a 4x4- that will give you a very nice 1.32 arc secs per pixel which is pretty much very good seeing- drop to 3x3 bin to get a tad under 1 arc sec per pixel. The camera is a win- do not be frightenened of binning- one of the greatest CCD cameras is the 16803 sensor and that has relativey big pixels but its great on long focal length telescopes especially when binned to 2x2 or 3x3.

 

You will have a great system and camera- have no fear and push the over sampling demons back to hell.

Thank you! Very encouraging. Some people scoff at CMOS binning, since it's "software", but there's nothing wrong with adding up numbers!




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics