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

Proper Backfocus for Reducer/Corrector

  • Please log in to reply
6 replies to this topic

#1 schellaj

schellaj

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 01 Nov 2018
  • Loc: Halifax, NS

Posted 28 February 2020 - 10:48 AM

Backfocus of the Celestron F6.3 Reducer/Corrector is listed as 105.0 cm but there has always been confusion on my part as to where to measure from.  I contacted Celestron and they indicated that the measure was “from the rear surface where the flange of the T-adapter meets the end of the reducer” (i.e. bottom of the threads). I had been measuring from the surface of the lenses, a difference of 10 mm.  This actual gave me a backfocus of 115.0 mm. 

Much of the online discussion was regarding the F-stop at varying backfocal distance, which did not concern me so much. I was more “focused” on the impact on coma correction.  Using the proper backfocus, my coma correction was indeed better.

What I didn’t appreciate is the impact on sharpness closer to the center of the image.  This is a side-by-side comparison of backfocal distance (left being correct, right is 10 mm too far).  The images are both essentially the same: integration time, pre-processing, and histogram stretch. Background values and S/N are nearly identical.  Even though the sub-exposure length is 5 minutes for left image and 3 minutes for the right, I would not expect too much difference in the fainter portion of the galaxy given that the total integration times are nearly identical.

As one can see, the left-hand image is much sharper even though the guiding was slight worse for that image.  I’m guessing that the focus across the image is better, allowing more light from the fainter portions of the galaxy to be gathered.

I was also testing the use of a longer sub-exposure time (5 min vs 3 min) to see if light pollution would play a role since many people seem to think that they need short exposures (30-60s) because they are under heavy light pollution.  I think the results speak for themselves. As long as you aren’t blowing out your stars or galactic center, you can get the benefit of longer sub-exposures. NO LPF WAS USED and the blue filter should be worse-case for LP.

Let me know what you think.

 

The comparison Image can be found on my astrobin page at          https://www.astrobin...hbpc8v/?nc=user

 

Jason



#2 sg6

sg6

    Fly Me to the Moon

  • *****
  • Posts: 7,219
  • Joined: 14 Feb 2010
  • Loc: Norfolk, UK.

Posted 28 February 2020 - 11:27 AM

Backfocus of the Celestron F6.3 Reducer/Corrector is listed as 105.0 cm but there has always been confusion on my part as to where to measure from.  I contacted Celestron and they indicated that the measure was “from the rear surface where the flange of the T-adapter meets the end of the reducer” (i.e. bottom of the threads). I had been measuring from the surface of the lenses, a difference of 10 mm.  This actual gave me a backfocus of 115.0 mm.

As the lens is inside the metal housing and the measurement is from the face of the housing then I would have said you were setting the back focus to 95mm not 115mm.

 

Why does "back focus" cause so much confusion ?

It seems immensly simple. You have to separate the sensor from the end of the assembly by a specified amount.

The camera will have a fixed flange/sensor distance - look it up.

You take that away from the given back focus requirement of the corrector.

What is left is the spacing you need to add to get the required "back focus".



#3 schellaj

schellaj

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 01 Nov 2018
  • Loc: Halifax, NS

Posted 28 February 2020 - 11:52 AM

As the lens is inside the metal housing and the measurement is from the face of the housing then I would have said you were setting the back focus to 95mm not 115mm.

 

Why does "back focus" cause so much confusion ?

It seems immensly simple. You have to separate the sensor from the end of the assembly by a specified amount.

The camera will have a fixed flange/sensor distance - look it up.

You take that away from the given back focus requirement of the corrector.

What is left is the spacing you need to add to get the required "back focus".

The thing was that I was setting my imaging sensor to a distance of 105 mm from the front of the lens (B on the image below), which gives 115 mm (not 95) from point A on the image.  When I contacted Celestron, they indicated that the 105 mm that is typically quoted is measured from point A.

 

Back focus should be "simple" but one has to know how the manufacturer is defining it as well. Many people assume it is measured from the end of the assembly, however, in this case, that is not so.

 

 ReducerAB.jpg



#4 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 18,922
  • Joined: 27 Oct 2014

Posted 28 February 2020 - 12:14 PM

It's not that you "need" shorter exposures in light polluted skies.  It's that the optimal subexposure, in terms of gathering data, is naturally shorter.

 

You "need" enough subexposure to "swamp" the read noise of your camera.  That means make the sky noise in the frame large enough that read noise is insignificant.  Once you reach that point, longer subs buy you nothing except more saturated, colorless stars.  You're "sky noise limited".  Then, the key to capturing more detail becomes longer total imaging time, more subs.  Signal strength increases faster than sky noise, so you win.  Especially important in light pollution.

 

More light pollution, more sky noise, less exposure needed to swamp read noise.  Equally important, less read noise, less subexposure needed to swamp it.

 

The bottom line is that people used to higher read noise cameras tend to shoot longer subs than optimal with low read noise cameras.  And it's particularly true in light polluted skies.

 

This is more complicated than can be covered in a short post here.  Three good things to look at.  The first is the optimal subexposures for a specific camera, with various optical speeds, and various skies.  Even if you don't have that particular camera, note how much things change in light pollution.  It's pretty stunning.

 

https://www.cloudyni...olour-versions/

 

This video by Dr. Robin Glover, author of Sharpcap, is excellent.  It's a whole hour, and it's packed with information.

 

https://www.youtube....h?v=3RH93UvP358

 

Warning, math ahead.  <grin>  This is the postgraduate course.  The pages on proper subexposure in Woodhouse's fine (but advanced) book, explain _why_ too long subs have little benefit, AND why other considerations may intrude.

 

https://www.amazon.c.../dp/1138055360/

 

A lot of very smart people have been carefully considering what subexposure is optimal for a long time, and there are things to be learned.  In my opinion the key one is that you must use site specific data, what others use for subexposure is likely of no importance to you, whatsoever.


Edited by bobzeq25, 28 February 2020 - 12:25 PM.


#5 schellaj

schellaj

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 01 Nov 2018
  • Loc: Halifax, NS

Posted 28 February 2020 - 12:33 PM

It's not that you "need" shorter exposures in light polluted skies.  It's that the optimal subexposure, in terms of gathering data, is naturally shorter.

 

You "need" enough subexposure to "swamp" the read noise of your camera.  That means make the sky noise in the frame large enough that read noise is insignificant.  Once you reach that point, longer subs buy you nothing except more saturated, colorless stars.  You're "sky noise limited".  Then, the key to capturing more detail becomes longer total imaging time, more subs.  Signal strength increases faster than sky noise, so you win.  Especially important in light pollution.

 

More light pollution, more sky noise, less exposure needed to swamp read noise.  Equally important, less read noise, less subexposure needed to swamp it.

 

The bottom line is that people used to higher read noise cameras tend to shoot longer subs than optimal with low read noise cameras.  And it's particularly true in light polluted skies.

 

This is more complicated than can be covered in a short post here.  Three good things to look at.  The first is the optimal subexposures for a specific camera, with various optical speeds, and various skies.  Even if you don't have that particular camera, note how much things change in light pollution.  It's pretty stunning.

 

https://www.cloudyni...olour-versions/

 

This video by Dr. Robin Glover, author of Sharpcap, is excellent.  It's a whole hour, and it's packed with information.

 

https://www.youtube....h?v=3RH93UvP358

 

Warning, math ahead.  <grin>  This is the postgraduate course.  The pages on proper subexposure in Woodhouse's fine (but advanced) book, explain _why_ too long subs have little benefit, AND why other considerations may intrude.

 

https://www.amazon.c.../dp/1138055360/

 

A lot of very smart people have been carefully considering what subexposure is optimal for a long time, and there are things to be learned.  In my opinion the key one is that you must use site specific data, what others use for subexposure is likely of no importance to you, whatsoever.

Thanks for taking the time to reply.  I'll have a look at the information.

 

I've been trying to figure out what testing I can do to determine how much sky noise i typically have to deal with. My concern is not trying to reduce the length of my subexposures (I rarely have to throw any out even at 10-30 minutes (when I do NB imaging).  I just don't want to be losing any image detail by being overexposed.

 

Cheers,

 

Jason



#6 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 18,922
  • Joined: 27 Oct 2014

Posted 28 February 2020 - 12:46 PM

Thanks for taking the time to reply.  I'll have a look at the information.

 

I've been trying to figure out what testing I can do to determine how much sky noise i typically have to deal with. My concern is not trying to reduce the length of my subexposures (I rarely have to throw any out even at 10-30 minutes (when I do NB imaging).  I just don't want to be losing any image detail by being overexposed.

 

Cheers,

 

Jason

Here's the short version.  Kind of a rule of thumb.

 

Take a light, take a bias.  Measure the ADU in something.  I use Pixinsight, the free IRIS works.  Subtract the bias from the light.

 

Using ZWO data convert the ADU to electrons.  Also using that data compute read noise squared in electrons (it will be electrons squared, doesn't matter).  Note that those calculations are gain specific.

 

Divide the 1st value by the second.  You're looking for something like 5-10.  That's in the ballpark.

 

That incorporates the necessary site specific data.  The average adu characterizes skyfog, the read noise in in there.  Skyfog equals 5-10XRN^2 is in the ballpark for "swamping".  <smile>


Edited by bobzeq25, 28 February 2020 - 12:48 PM.


#7 schellaj

schellaj

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 01 Nov 2018
  • Loc: Halifax, NS

Posted 28 February 2020 - 02:39 PM

Here's the short version.  Kind of a rule of thumb.

 

Take a light, take a bias.  Measure the ADU in something.  I use Pixinsight, the free IRIS works.  Subtract the bias from the light.

 

Using ZWO data convert the ADU to electrons.  Also using that data compute read noise squared in electrons (it will be electrons squared, doesn't matter).  Note that those calculations are gain specific.

 

Divide the 1st value by the second.  You're looking for something like 5-10.  That's in the ballpark.

 

That incorporates the necessary site specific data.  The average adu characterizes skyfog, the read noise in in there.  Skyfog equals 5-10XRN^2 is in the ballpark for "swamping".  <smile>

Thanks.  I watched the video by Dr. Glover and it was very informative.  Nice to know that I am not hurting my images by using longer exposures (3-5 minutes for LRGB) as long as I'm not blowing out the stars.

 

Cheers,

 

Jason




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