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

Correct me if I am wrong, I am a newbie,lol

  • Please log in to reply
25 replies to this topic

#1 Waterobert

Waterobert

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 04 Mar 2017
  • Loc: Laguna Niguel, California , USA

Posted 21 October 2019 - 08:01 PM

I was just reading up about matching sensor with optics. It is a general rule that best image scale is around 1-1.5 arc second. So, for ZWO ASI 1600 camera (3.8 micron pixel) using telescope with focal length greater than 783mm would result in oversampling. In order to get the best resolution out of the camera, we want to avoid oversampling and unless we have bortle 1-2 skies, getting scope with focal length greater than 783mm would be pointless. Am I correct or is it one off those rules which can be challenged?



#2 Gipht

Gipht

    Surveyor 1

  • *****
  • Posts: 1952
  • Joined: 12 Nov 2016
  • Loc: Prescott Valley, AZ.

Posted 21 October 2019 - 08:15 PM

If skies remain at a constant level of seeing,  then your point would be important.  Tonight my skies looks a little dingy, and I expect some transperancy issues.  3 weeks ago, I had a night where seeing was spectacular.

 

For me, the imaging scale  I choose is related mostly to field of view.  What is the right telescope and camera that will make a  for a nice picture with plenty of stars around the target.  This will depend more on the targets you wish to image then the image scale.


  • Cfreerksen likes this

#3 Cfreerksen

Cfreerksen

    Viking 1

  • *****
  • Posts: 517
  • Joined: 18 Aug 2018
  • Loc: Tooele, Ut

Posted 21 October 2019 - 08:22 PM

It's not pointless. A longer focal length would give you a larger image. If that what you desire, it's not pointless. The higher pixel ratio just will not give you any added detail. Your just kinda leaving pixels on the table. This is better than leaving detail on the table by undersampling.

 

It won't ruin your images.  

 

Chris


  • TerryD23 likes this

#4 SilverLitz

SilverLitz

    Mariner 2

  • -----
  • Posts: 271
  • Joined: 17 Feb 2018
  • Loc: Louisville, KY

Posted 22 October 2019 - 02:19 AM

It will depend on a few things:

1) How good is your mount?  (AP Mach1, NO PROBLEM; better the mount the smaller number of pixel scale capable)

2) The aperture of your scope determines the Dawes Limit; e.g. 100mm => 1.16, 142mm => 0.82

3) Your sky's seeing conditions, in arcsec

 

Generally, ideal sampling for 2D photos are about 1/3 your limiting factor.  Even more accurate would be = sqrt(Dawe^2 + Seeing^2 + Factor3^2 + ...)

 

My home has typical seeing ~1.5, and I have an Esprit 100, so my Nyquist limit would be ~0.5 arcsec'ish.  My Esprit 100 with my ASI183mm-Pro is about 0.9 arcsec, definitely NOT oversampled.  I should be able to get a larger scope for small galaxies, as my G11 can handle it.



#5 Stelios

Stelios

    Cosmos

  • *****
  • Moderators
  • Posts: 7865
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 22 October 2019 - 02:41 AM

Here in Southern California, I've got very nice images (most of which I haven't finished, but some are in my Astrobin near the top) with my ASI1600MM-C and Edge800 with reducer (1421mm F/L and image scale of 0.552). 

 

So don't limit yourself unnecessarily. 

 

Incidentally, Bortle level has nothing to do with it. It's seeing that matters, not darkness for image scale purposes. 


  • happylimpet likes this

#6 Stelios

Stelios

    Cosmos

  • *****
  • Moderators
  • Posts: 7865
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 22 October 2019 - 02:45 AM

I'm actually going to try and get my courage up to mate my new ASI183MM-Pro with the Edge, if the run with the 115/805 proves promising. That'll be a 0.348 image scale--I hope to shoot Stephan's Quintet with it. OTOH, it may end up a smudged up mess. We'll see :)



#7 Waterobert

Waterobert

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 04 Mar 2017
  • Loc: Laguna Niguel, California , USA

Posted 22 October 2019 - 11:37 AM

Sounds like even newbie like me would be ok with image scale under 1 arc second. Maybe I should limit myself to something like .75 arc second?

#8 kathyastro

kathyastro

    Viking 1

  • *****
  • Posts: 899
  • Joined: 23 Dec 2016
  • Loc: Nova Scotia

Posted 22 October 2019 - 11:44 AM

Why limit yourself at all?  Oversampling is not a bad thing.  It just means that you don't have quite as much data as the image dimensions would suggest you do.  If that bothers you, you can resample in processing.



#9 Stelios

Stelios

    Cosmos

  • *****
  • Moderators
  • Posts: 7865
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 22 October 2019 - 01:01 PM

What telescope are you lusting after?

 

With the Mach1, you should be able to handle an Edge 11 or a 6" F/6 refractor without problems (although the image scale of the Edge 11 would be better served with a larger pixel camera than the ASI1600).

 

Make sure that you balance properly (using as much weight close to the top of the CW shaft as possible--this may mean that you need more CW's than the total weight of your OTA + imaging train). Do *slightly* CW-heavy and *slightly* camera-heavy balancing. If you own a used Mach1, download PemPro and get a new PEC curve. Run your cables through the mount. Use an OAG if imaging with the ASI1600 for best results (necessary with SCT's, but best with *any* scope).



#10 the Elf

the Elf

    Surveyor 1

  • *****
  • Posts: 1816
  • Joined: 06 Sep 2017
  • Loc: Germany

Posted 22 October 2019 - 02:19 PM

There is no general rule that the scale shoud be whatsoever. There is a theorem by Shannon and Nyquist that sampling (spatial when imaging) must be done at twice the maximum frequency of the signal if you want to restore the original signal from the sampling correctly. (See wikipedia for detail. Note the restoration filter story....) The resolution you have in the sensor plane depends on your optics resolution and the seeing. It is relevant for small objects only, because for a large one you reduce maginfication until you can pack the whole thing into the field of view. In many cases the resolution is neither limited by the optical resolution nor by the seeing but by the tracking capability of your mount. Last but not least an OSC camera has a low resolution in general, there are arguments for 1.4 and for 2 depending on if you consider any color or only red an blue. DSLRs come with a spatial lowpass (= blur filter) to avoid alias artifacs. This may or may not reduce resolution. Looking at resolution only does not tell the truth about an image. Signal to noise eats up the small structures first. So producing a clear image of a small detail needs the high SNR and the resolution. Focal reducers can increase signal to noise while decreasing spatial resolution and in some cases you see more when the resolution is less.

Making it even more complicated: there are some image processing steps that simply do not work well when the sampling rate is too low. If you have single pixel stars there is no point in deconvolution.

Are you heading for the monochrome or the color version of the 1600? A worst case calculation: you go for color. You like Ha objects. You have to double the pixel scale because only one out of 4 pixels is red. You correctly understand that 2x the maximum is the theoretical limit of the sampling theorem, in reality 3x is a good idea. Say you have a 2 arcsec seeing. This leads you to 0.33 arcsec per pixel.

Other end of the scale: you go for mono. Your average seeing is 3 arcsecs. 1 arcsec per pixel is fine in theory but you live under bortle 9 skies and the noise eats up everything. You put a reducer behind the scope and all of a sudden there is more detail at 1.5 arcsec because you have more signal in the same over all imaging time.


  • caseyfinn likes this

#11 Waterobert

Waterobert

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 04 Mar 2017
  • Loc: Laguna Niguel, California , USA

Posted 22 October 2019 - 07:16 PM

Thank you guys, as I expected best image scale depends on many factors and mostly weather, lol.

I will be getting mono version of ASI 1600 and right now I will be using it with 80mm/f6. refractor. However, next year I want to get refractor with much longer focal length. There are many scopes to chose from. My mount can handle up to 40 pounds, so I am not limited that way. However, since I will be traveling  with my set up, it has to be reasonably light. What focal length is most versatile? I know it depends on what I like to take pictures off, and there are personal brand preferences. Tell me what would be your ideal scope.



#12 View2

View2

    Vanguard

  • *****
  • Posts: 2203
  • Joined: 20 May 2016
  • Loc: Vancouver, WA USA

Posted 22 October 2019 - 07:41 PM

Don't forget your electronic filter wheel and filters!

#13 View2

View2

    Vanguard

  • *****
  • Posts: 2203
  • Joined: 20 May 2016
  • Loc: Vancouver, WA USA

Posted 22 October 2019 - 07:44 PM

If your F/6.6 80 mm refractor is Apochromatic, you already have a quite versatile scope. I've used the ASI1600 on f/5.5 72mm 400mm and f/7.5 127mm 952mm(666mm f/5.25) refractors. I found the 1600 a pleasure to use in any configurationšŸ”­šŸ‘

Edited by View2, 22 October 2019 - 07:50 PM.


#14 Marcelofig

Marcelofig

    Messenger

  • -----
  • Posts: 465
  • Joined: 21 Jan 2015

Posted 22 October 2019 - 07:48 PM

Thank you guys, as I expected best image scale depends on many factors and mostly weather, lol.

I will be getting mono version of ASI 1600 and right now I will be using it with 80mm/f6. refractor. However, next year I want to get refractor with much longer focal length. There are many scopes to chose from. My mount can handle up to 40 pounds, so I am not limited that way. However, since I will be traveling  with my set up, it has to be reasonably light. What focal length is most versatile? I know it depends on what I like to take pictures off, and there are personal brand preferences. Tell me what would be your ideal scope.

 

You can go to https://astronomy.to.../field_of_view/ (> image mode) select a camera (the ASI 1600 in your case), the telescopes you are interested in and different DSOs.

 

Play with the different alternatives.



#15 imtl

imtl

    Explorer 1

  • *****
  • Posts: 54
  • Joined: 07 Jun 2016

Posted 22 October 2019 - 08:56 PM

Why limit yourself at all?  Oversampling is not a bad thing.  It just means that you don't have quite as much data as the image dimensions would suggest you do.  If that bothers you, you can resample in processing.

Resampling post-processing introduces more noise. There is no way around real time SNR. Not saying a person shouldn't do in post-processing. Just know all the story behind it.


Edited by imtl, 22 October 2019 - 09:21 PM.


#16 Stelios

Stelios

    Cosmos

  • *****
  • Moderators
  • Posts: 7865
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 23 October 2019 - 12:51 AM

I like the versatility--and the quality--of the Edge800 HD. It's relatively light, and it offers the possibility of imaging at three F/ratios: F/2 (with Hyperstar), F/7 (with 0.7x reducer) and F/10 (native). Even with the cost of Hyperstar and reducer, and throwing in a Moonlite CHL 2.5" focuser with stepper mottor and controller, you are at about $3,400 for a *complete* solution. A big quality refractor will have far less versatility and won't let you image at F/2, or image small galaxies, not to mention be far more expensive.

 

I'm not just spouting theory--I have one, and love it. I haven't used Hyperstar, but the reducer and Moonlite focuser, both of which I have, work just great. 



#17 Waterobert

Waterobert

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 04 Mar 2017
  • Loc: Laguna Niguel, California , USA

Posted 23 October 2019 - 07:38 AM

I like the versatility--and the quality--of the Edge800 HD. It's relatively light, and it offers the possibility of imaging at three F/ratios: F/2 (with Hyperstar), F/7 (with 0.7x reducer) and F/10 (native). Even with the cost of Hyperstar and reducer, and throwing in a Moonlite CHL 2.5" focuser with stepper mottor and controller, you are at about $3,400 for a *complete* solution. A big quality refractor will have far less versatility and won't let you image at F/2, or image small galaxies, not to mention be far more expensive.

 

I'm not just spouting theory--I have one, and love it. I haven't used Hyperstar, but the reducer and Moonlite focuser, both of which I have, work just great. 

I will definitely get Edge HD scope, probably 9.25, however that purchase is a year or two away for me. Long focal length will have to wait until I am skillful enough to take advantage of it. lol.gif lol.gif lol.gif



#18 the Elf

the Elf

    Surveyor 1

  • *****
  • Posts: 1816
  • Joined: 06 Sep 2017
  • Loc: Germany

Posted 24 October 2019 - 06:49 AM

 

I will definitely get Edge HD scope

Hm. This makes my post obsolete. OK, in case someone else reads this and is looking for a long focal length scope: the top two scopes on my personal list are the Skywatcher Explorer MN190 and the GSO RC8 carbon. I found some people who narrowed it down to these two as well. I went for the RC and do not regret. I see the images taken with the MN190 and sometimes wonder if I get that one as well. I'm a bit scared of the correction plate in the edge and the MN because I often have damp conditions and don't want to add dew heaters. I'm concerned about the back focus of the MN when using a DSLR and an OAG. The RC never fogs up and has got plenty of back focus and the carbon tube accepts temperature changes up to 5Ā°C without the need to refocus (rule of thumb, depending on the level of detail of course). I think the choice also depends on the sensor size. The Edge is made for pinpoint stars on large sensors. If you use a small sensor the other two might serve you well. Of course the hyper star option is unique. There are never ending horror storys about collimating an RC but so far it did not kill me. The RC can also be used with a reducer, so two in one. It is a reducer only, not flattening, and thus spacing can be changed in a wide range: you can dial in where you want your focal plane to be. I am at 1100 mm focal lenght with the shortest possible setup: the focuser is half way out (15mm) and no spacer rings between focuser and camera or focuser and scope. I like the compact setup. I think it is the compactest you can get at the focal length. Any refractor will be far longer.


Edited by the Elf, 24 October 2019 - 06:55 AM.

  • Stephen Kennedy likes this

#19 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 17082
  • Joined: 27 Oct 2014

Posted 24 October 2019 - 10:11 AM

I was just reading up about matching sensor with optics. It is a general rule that best image scale is around 1-1.5 arc second. So, for ZWO ASI 1600 camera (3.8 micron pixel) using telescope with focal length greater than 783mm would result in oversampling. In order to get the best resolution out of the camera, we want to avoid oversampling and unless we have bortle 1-2 skies, getting scope with focal length greater than 783mm would be pointless. Am I correct or is it one off those rules which can be challenged?

It's complicated, depends on a lot of things like seeing.  But beginners are usually too concerned about image scale. 

 

I use everything from 1.0 to 2.7.  The people doing a _very_ serious H alpha map of the sky are using 3.1.

 

Smaller numbers (like 1.0) can get you better resolution and rounder stars.  But a _lot_ of things have to go right to get it, and it comes at the cost of signal to noise ratio, also important for good images.  Larger numbers (like 2) get you better signal to noise ratio, no matter what.  My suggestion for a beginner is very approximately 2.

 

Worry about scope quality and your desired field of view first.  Image scale just needs to be decent.


Edited by bobzeq25, 24 October 2019 - 10:14 AM.


#20 the Elf

the Elf

    Surveyor 1

  • *****
  • Posts: 1816
  • Joined: 06 Sep 2017
  • Loc: Germany

Posted 24 October 2019 - 04:25 PM

Hmmm. Image scale needs to match the object. For the ring nebula or for jupiter's ghost a 2 arcsec scale does not help. For a beginner the advice is start with large and bright targets. Then indeed the scale does not matter. If down the road you want ot specialize on small objects be prepared to get better equippment. Maybe I'm on the transition from beginner to a bit more experienced after 2.5 years but I still pick the medium sized objects to avoid frustration, imaging at 0.85 arcsec/pix. Right now I don't have any desire to go smaller. I enjoy wide fields at 200mm, almost 5 arcsec per pixel. It really depends on the object. Sooner or late you will have more than one scope and one will be a short focal length.



#21 Waterobert

Waterobert

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 04 Mar 2017
  • Loc: Laguna Niguel, California , USA

Posted 24 October 2019 - 07:45 PM

Thank you Elf for your reply. I am an absolute beginner, I took my first image just few weeks ago. At this point I am perfectly happy with 480mm focal length, however I am slowly looking into expending my horizons. Perhaps, in the second half of the next year I will invest in 750-850mm focal length refractor. I do like to research my options early on, so went the shopping time arrives I will know what I need to get. As a newbie I want to avoid frustration associated with learning this challenging hobby. Even thought I want to go after those small galaxies, for next year or two I will limit myself to focal length below 1000mm. After all, those galaxies will still be there patiently waiting for me. lol.gif lol.gif lol.gif  

So, in your opinion imaging below 0.85 arches/pix will lead to frustration ?


Edited by Waterobert, 24 October 2019 - 07:48 PM.


#22 the Elf

the Elf

    Surveyor 1

  • *****
  • Posts: 1816
  • Joined: 06 Sep 2017
  • Loc: Germany

Posted 25 October 2019 - 05:48 AM

No, did not intend to say the 0.85 leads to frustration in general. It would not work in my place and with my equippment. You have got a high quality mount, so you should be able to track it well. For my EQ6-R it is probably the limit. I have only one or two really clear nights with steady air per year. Plus about 20 with either thin clouds, only one hour without clouds or with the moon out or some thin fog at the ground. There is just no point in trying a higher resolution. My best FWHM ever was 2 arcsecs, most of the time it is 3 or worse. When I look out of the window these days Capella is blinking like the mirror balls in a disco.

If you are in a place with better conditions and the stars are not twinkling that much and you have enough clear nights to collect 20+ hours on one target you can go even further. Some pros from CA claim to toss any sub that has got a higher FWHM than 2.0. Lucky them. Assuming they are not talking BS 0.5 or 0.6 may make sense.

My option was to use the RC with or without reducer. I don't get any extra detail with 0.55 arcsec/pixel but the signal is weaker and the thermal noise is the same.

The list of difficulties with long focal lenght is this:

- if goto is poor, the object may not be in the image. Cure: a good mount and/or computer help and plate solving

- you need auto guiding. Period. Flex between main scope and guide scope or internal flex like mirror flop will kill you. Cure: Off Axis Guiding (find a video on my YouTube)

- The system is slow like f/7 to f/10. You need a lot of exposure time. No Cure, only time.

- The mount must track well and react well to guide commands. You have a Mach 1, no problem for you.

- The tubes get longer and small changes in temperature move you out of focus. Cure: refocus manually every hour or when temperature changes. Manual focus consumes imaging time. You may loose the guide star in the OAG.

- You need a quality focus that either does not need a locking screw or does not move at all when locking it. From what I read Moonlite does the job.

- When you walk near the mount the image is ruined. You need a good tripod like a Berlebach and stay away from the setup or a permanent setup with a concret foundation and something you can walk on not transfering vibrations and movement to the rig. Railway and big trucks nearby is also a nogo.

- Little heat from a concrete pad nearby, from your house, even you walking in front of the scope causes air movment and kill the resolution.

 

Right now I made a decision not to go further and pay my house dept first. When this is done in a few years I can rethink. A 10micron is on my wishlist. If you can and want to invest in the hobby there is no reason why you have to stop near 1 arcsec. Right now the list of issues might be a bit too long and you don't get anything at all. 2 arcsec per pixel is sunday walk. No troubles. Ask yourself how you feel about it, I can enjoy to do something that is within my abilities and so leads to a predicted result as opposed to being a hit or miss or a lottery. Others might have a desire to push their own limits every day. When you have more than 20 clear nights a year you can to both. Don't sell the scope you have! Add a longer one.



#23 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 17082
  • Joined: 27 Oct 2014

Posted 25 October 2019 - 09:44 AM

Thank you Elf for your reply. I am an absolute beginner, I took my first image just few weeks ago. At this point I am perfectly happy with 480mm focal length, however I am slowly looking into expending my horizons. Perhaps, in the second half of the next year I will invest in 750-850mm focal length refractor. I do like to research my options early on, so went the shopping time arrives I will know what I need to get. As a newbie I want to avoid frustration associated with learning this challenging hobby. Even thought I want to go after those small galaxies, for next year or two I will limit myself to focal length below 1000mm. After all, those galaxies will still be there patiently waiting for me. lol.gif lol.gif lol.gif  

So, in your opinion imaging below 0.85 arches/pix will lead to frustration ?

It's not so much that it leads to frustration.  It's that it requires significantly more total imaging time to get the same signal to noise ratio, requires excellent tracking, excellent skies, and has little benefit for almost all beginners.  A numerically larger image scale is easier to deal with, and, all things considered, in most cases, the resulting images from the beginner will be at least as good.  Likely better.

 

The cost (involves more than money)/benefit ratio is poor.  Yet another example where the best equipment for an experienced imager may be significantly suboptimal for a beginner.

 

Your plan of stepping up to the 750mm range is good.  Concentrate on other things than image scale in selecting the scope; if the image scale is 1-2, that's plenty "good enough".



#24 the Elf

the Elf

    Surveyor 1

  • *****
  • Posts: 1816
  • Joined: 06 Sep 2017
  • Loc: Germany

Posted 26 October 2019 - 04:08 AM

 

A numerically larger image scale

A good way to put it. It avoids any confusion that sometimes occured in other posts.

Indeed, if the over all image time for one object is limited by wheather, job or whatever the result wil be better with a numerically larger scale. That is because the fine detail is eaten up by noise and more integration time is required to get the noise level down, just as Bob says. Here is a nice demonstration by Robin Glover:

https://www.youtube....3UvP358&t=1470s (the link supposed to start at 24:30)

It is a simple demonstration but it shows why the detail that was optically resolved does not appear in the image unless you get the noise down. Conclusion: go for 0.5 arcsecs when you are retired, you have way more time then!


  • bobzeq25 likes this

#25 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 17082
  • Joined: 27 Oct 2014

Posted 26 October 2019 - 11:11 AM

I'm actually going to try and get my courage up to mate my new ASI183MM-Pro with the Edge, if the run with the 115/805 proves promising. That'll be a 0.348 image scale--I hope to shoot Stephan's Quintet with it. OTOH, it may end up a smudged up mess. We'll see smile.gif

I tried something like that with my 183 OSC and my Telescop Services 130mm F7.  Image scale 0.5.  The results were bad (although not just a "smudged up mess").  I clearly would have been better off with my 4.5 micron pixeled Atik 460 at 1.0.

 

A _lot_ of things have to go right before such numerically small image scales give you better results on DSOs.  Numerically larger image scales are much more "forgiving" of all kinds of things.  More photons per pixel is a very good thing.  Why people bin.


Edited by bobzeq25, 26 October 2019 - 11:15 AM.



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