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

Basic back focus question

astrophotography CMOS optics equipment
  • Please log in to reply
19 replies to this topic

#1 WalterG

WalterG

    Explorer 1

  • -----
  • topic starter
  • Posts: 51
  • Joined: 26 Dec 2011

Posted 09 July 2019 - 02:19 PM

I have what I think is a basic question but I can't find the answer.  When calculating the back focus distance to your sensor what equipment do you have to take into account?

 

backfocus distance.jpg  

 

Does the back focus distance include the distance of my focuser (A), and my filter wheel (B) from the scope back or just the distance labeled "C"  The distance behind the filters?

 

If it is the distance behind the filters how do I calculate or correctly estimate that distance from this diagram?

 

OAG_wheel_side_view.jpg  

 

Thanks

Walter 

 

 



#2 einarin

einarin

    Surveyor 1

  • *****
  • Posts: 1930
  • Joined: 23 Dec 2016

Posted 09 July 2019 - 02:31 PM

What is your scope and do you have a focal reducer in it ?



#3 ngc7319_20

ngc7319_20

    Apollo

  • *****
  • Posts: 1014
  • Joined: 25 Oct 2015
  • Loc: MD

Posted 09 July 2019 - 03:01 PM

That looks like an SCT in your picture, so you are probably OK.  They have lots of back focus available.

 

Anything between the scope and sensor counts as to increasing the required back focus.  So its A + B + distance from the camera front flange to the sensor.  I guess in your drawing B = 25 + 17mm.

 

The filter will push the focus back a tiny bit -- by about 1/3 of its glass thickness -- probably 1/2 to 1 mm.



#4 WalterG

WalterG

    Explorer 1

  • -----
  • topic starter
  • Posts: 51
  • Joined: 26 Dec 2011

Posted 09 July 2019 - 04:20 PM

Thank you so much!

#5 WalterG

WalterG

    Explorer 1

  • -----
  • topic starter
  • Posts: 51
  • Joined: 26 Dec 2011

Posted 09 July 2019 - 04:33 PM

It’s a Celestron 8” edge, no focal reducer. I use an ASI 1600 camera that needs 55mm of distance to the sensor.

#6 WalterG

WalterG

    Explorer 1

  • -----
  • topic starter
  • Posts: 51
  • Joined: 26 Dec 2011

Posted 12 July 2019 - 02:56 PM

I'm still trying to understand the need for a specific back focus distance and I have been Googling around to try to find a ray-diagram that I can look at.  I understand the need to have a focused image at the sensor, I think that is obvious, but here are my questions.  1) If the telescope you are using has enough focal travel either "in" or "out" to achieve focus on the sensor then why would we need to pay attention to a specific distance of back focus?  Question 2) I have seen posts that talk about stars that are not in focus at the edges because of incorrect back focus distance.  Is this true or are stars that are not in focus or elongated due to your optics not producing a flat image or some degree of tilt in your imaging train that causes the sensor to have tilt? 

 

 

 

 



#7 carolinaskies

carolinaskies

    Surveyor 1

  • *****
  • Posts: 1626
  • Joined: 12 Dec 2014
  • Loc: Greenville SC

Posted 12 July 2019 - 03:54 PM

I'm still trying to understand the need for a specific back focus distance and I have been Googling around to try to find a ray-diagram that I can look at.  I understand the need to have a focused image at the sensor, I think that is obvious, but here are my questions.  1) If the telescope you are using has enough focal travel either "in" or "out" to achieve focus on the sensor then why would we need to pay attention to a specific distance of back focus?  Question 2) I have seen posts that talk about stars that are not in focus at the edges because of incorrect back focus distance.  Is this true or are stars that are not in focus or elongated due to your optics not producing a flat image or some degree of tilt in your imaging train that causes the sensor to have tilt? 

1) Backfocus for telescopes using internal or external flatners, focal reducers, etc have a specific distance that a camera imaging sensor must be in order to achieve optimal focus according to that specific flatner, FR, etc.  An SCT varies the distance from primary to secondary and there is a 'sweet spot' of correction for the SCT to minimize aberrations.  This is most critical as your imaging sensor increases in size utilizing the full FOV the telescope provides. When you use smaller sensors then it is not as critical. 

2)  Building on answer 1 - for larger imaging sensors, especially full frame, if the focus point is moved beyond the 'sweet spot' the stars at the edge will elongate somewhat. This is due to the secondary mirror being a 5x spherical or aspherical(depending on telescope) design.  So it is bending the light from the stars off-axis the further away from the sweet spot.   Tilt in an imaging train will cause a specific type of elongation, usually seen as non-uniform in the direction away from tilt.  so that one sector may be perfectly round but as you move to the opposing sector the elongation becomes worse and worse.  Most often tilt is the result of either poor or weak attachment methodology or a camera which has tilt adjustment being out of 'flat'.   Often the best systems use as few bits between the sensor and telescope attachment point and all attachements are done in very secure manner via screw together or bolt together components to achieve the optimal back focus.  

BTW, camera makers give a focal distance number, this is the distance from the sensor to the flat outside point where the body will attached to the imaging train.  If for instance it is 20mm, that means if you were to measure from that external flat surface to the front of the sensor it would be 20mm and this number is added to the calculation of total back focus between imaging sensor and back of telescope.  Cameras like the ZWO120MC have a distance of 12.5 while  a Canon EOS is 44mm.  An ASI1600 is 6.5mm!    Remember that the sensor in the camera is recording whatever light is hitting it. It is the job of the telescope optics to produce the in-focus disc at the point of that sensor face. 



#8 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 24442
  • Joined: 01 Feb 2006

Posted 12 July 2019 - 04:17 PM

I'm still trying to understand the need for a specific back focus distance and I have been Googling around to try to find a ray-diagram that I can look at.  

Page 5.

 

https://s3.amazonaws...paper_final.pdf

 

Not specific to the EdgeHD 8", but I will explain why the back focus is critical and it helps to look at the ray trace on this link.

 

Note that off axis the correction is not perfect.  Just about no telescope made will give a perfect off axis performance.

 

Also, not that Celestron does not say the field is flat. They say it is "flatter" than the standard SCT.

Now for visual use, none of this would be important, but we now have CCD cameras that can resolve the smallest deviation from a perfectly focused star.  When Celestron gives back focus, the figure they give you is going to produce the smallest average abberated blur diameter across the field of the telescope

 

What this means when field curvature is present is that you pick a point somewhere between the furthest focus in the field and the closest focus in the field as your best focus point.  If you picked a point all the way to the outside of the field, stars on the inside of the field would be out of focus. If you picked a point a the center of the field, stars at the outside of the field would be out of focus and the more out of focus, the more likely the camera will see any aberration present.  The spacing also provides the smallest astigmatic blur. 

 

The goal is to get the smallest average aberrated blur diameter for all of the stars in the field and the back focus Celestron provides is intended reduce the size of any aberration to its lowest level and to place the curve of the field at the ideal point to keep the aberrated blur diameter anywhere in the field too small to resolve. 


Edited by Eddgie, 12 July 2019 - 04:23 PM.


#9 WalterG

WalterG

    Explorer 1

  • -----
  • topic starter
  • Posts: 51
  • Joined: 26 Dec 2011

Posted 12 July 2019 - 04:19 PM

Thank you for that explanation.

 

Here is my issue/problem.  I use a Celestron 8" Edge. According to ZWO 55mm of back focus is needed for my 1600mm.   If I measure the "distance from the back of the scope to my sensor I have over 140mm of distance.  The only way I could achieve the 55mm back focus is to not use the focuser.  What am I to do?

 

Here is a link to a very old CN post that talks about Celestron mentioning that 5.25" is the optimal illumination of back focus for the Edge http://www.cloudynig...focus-question/

 

If this is true does this mean that I need to place the CCD sensor at 5.25 inches (133.35 mm) from the back of the scope? If so, I still have a problem ... Yes?



#10 carolinaskies

carolinaskies

    Surveyor 1

  • *****
  • Posts: 1626
  • Joined: 12 Dec 2014
  • Loc: Greenville SC

Posted 12 July 2019 - 04:24 PM

Thank you for that explanation.

 

Here is my issue/problem.  I use a Celestron 8" Edge. According to ZWO 55mm of back focus is needed for my 1600mm.   If I measure the "distance from the back of the scope to my sensor I have over 140mm of distance.  The only way I could achieve the 55mm back focus is to not use the focuser.  What am I to do?

 

Here is a link to a very old CN post that talks about Celestron mentioning that 5.25" is the optimal illumination of back focus for the Edge http://www.cloudynig...focus-question/

 

If this is true does this mean that I need to place the CCD sensor at 5.25 inches (133.35 mm) from the back of the scope? If so, I still have a problem ... Yes?

Where did you get this information? "According to ZWO 55mm of back focus is needed for my 1600mm."

As I state in my post above, the camera doesn't care about backfocus distances. All it does is record light, it doesn't focus anything.  Only the optical system determines backfocus distance necessary to have a fully illuminated field. 



#11 WalterG

WalterG

    Explorer 1

  • -----
  • topic starter
  • Posts: 51
  • Joined: 26 Dec 2011

Posted 12 July 2019 - 04:28 PM

Page 5.

 

https://s3.amazonaws...paper_final.pdf

 

Not specific to the EdgeHD 8", but I will explain why the back focus is critical and it helps to look at the ray trace on this link.

 

Note that off axis the correction is not perfect.  Just about no telescope made will give a perfect off axis performance.

 

Also, not that Celestron does not say the field is flat. They say it is "flatter" than the standard SCT.

Now for visual use, none of this would be important, but we now have CCD cameras that can resolve the smallest deviation from a perfectly focused star.  When Celestron gives back focus, the figure they give you is going to produce the smallest average abberated blur diameter across the field of the telescope

 

What this means when field curvature is present is that you pick a point somewhere between the furthest focus in the field and the closest focus in the field as your best focus point.  If you picked a point all the way to the outside of the field, stars on the inside of the field would be out of focus. If you picked a point a the center of the field, stars at the outside of the field would be out of focus and the more out of focus, the more likely the camera will see any aberration present.  The spacing also provides the smallest astigmatic blur. 

 

The goal is to get the smallest average aberrated blur diameter for all of the stars in the field and the back focus Celestron provides is intended reduce the size of any aberration to its lowest level and to place the curve of the field at the ideal point to keep the aberrated blur diameter anywhere in the field too small to resolve. 

Thank you for the link to the White Paper



#12 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 24442
  • Joined: 01 Feb 2006

Posted 12 July 2019 - 04:39 PM

My pleasure.  As I said, it is not specific to the 8", but it clearly shows how the field is not perfect, The goal of the back spacing Celestron provides is to give the smallest average blur diameter. 

 

(There is also the issue of mirror spacing.  The system in theory is only perfectly corrected for SA when the primary is an exact distance from the secondary, but a few millimeters one way or the other does not change the SA enough to be an issue.  It would appear that it is a combination of CA and Astigmatism that the very specific spacing is trying to address).  


  • WalterG likes this

#13 WalterG

WalterG

    Explorer 1

  • -----
  • topic starter
  • Posts: 51
  • Joined: 26 Dec 2011

Posted 12 July 2019 - 04:45 PM

I just measured the distance that focuser is out from being completely retracted.  It is about 16mm.  So I should be able to bring the focuser in almost 10mm then use the Celestron focus knob to refocus the mirror. This would give me room to play around to get the ~133mm that I need.  Do you think that this would be a work-around solution? 



#14 WalterG

WalterG

    Explorer 1

  • -----
  • topic starter
  • Posts: 51
  • Joined: 26 Dec 2011

Posted 12 July 2019 - 04:55 PM

Where did you get this information? "According to ZWO 55mm of back focus is needed for my 1600mm."

As I state in my post above, the camera doesn't care about backfocus distances. All it does is record light, it doesn't focus anything.  Only the optical system determines backfocus distance necessary to have a fully illuminated field. 

I found this on their website

https://astronomy-im...tions-55mm.html



#15 jhayes_tucson

jhayes_tucson

    Fly Me to the Moon

  • *****
  • Posts: 7100
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 12 July 2019 - 06:37 PM

I'm still trying to understand the need for a specific back focus distance and I have been Googling around to try to find a ray-diagram that I can look at.  I understand the need to have a focused image at the sensor, I think that is obvious, but here are my questions.  1) If the telescope you are using has enough focal travel either "in" or "out" to achieve focus on the sensor then why would we need to pay attention to a specific distance of back focus?  Question 2) I have seen posts that talk about stars that are not in focus at the edges because of incorrect back focus distance.  Is this true or are stars that are not in focus or elongated due to your optics not producing a flat image or some degree of tilt in your imaging train that causes the sensor to have tilt? 

 

When a telescope is designed, the predicted imaging performance is only valid when the components are properly spaced.  The Celestron telescopes use moving primary mirror to focus the system, which means that the spacing between the components can vary.  That's generally not a big problem for image quality near the axis, but it can be a big problem for off-axis image quality.  The change in spherical aberration varies very slowly when you change the mirror spacing but the off-axis aberrations have large sensitivity to the mirror spacing.  The only way to get the spacing right is to set the image plane at a known location, which is the specified back working distance (AKA back focus distance.)  When the scope is set to focus on this "reference plane," the mirror spacing will be correct.  So, the accuracy required in setting the BWD depends strongly on the size of the sensor that you use.  The bigger the sensor, the more careful you need to be.  As I recall, Celestron states that the tolerance is within 0.5 mm in order to achieve the specified field for the Edge scopes.  I've done a fair amount of analysis on this and they are being pretty conservative with this recommendation.  Even for a large sensor, getting the spacing to within a millimeter is usually more than sufficient.

 

When you compute the BWD, you need to include all the physical distances and account for any glass in the path.  Any plane-parallel window (or filter) will increase the physical distance by roughly one third of it's thickness.  On Celestron scopes, the reference plane is at the rear surface of the baffle nut so there where everything is measured from.

 

John


  • WalterG likes this

#16 WalterG

WalterG

    Explorer 1

  • -----
  • topic starter
  • Posts: 51
  • Joined: 26 Dec 2011

Posted 13 July 2019 - 10:01 AM

When a telescope is designed, the predicted imaging performance is only valid when the components are properly spaced.  The Celestron telescopes use moving primary mirror to focus the system, which means that the spacing between the components can vary.  That's generally not a big problem for image quality near the axis, but it can be a big problem for off-axis image quality.  The change in spherical aberration varies very slowly when you change the mirror spacing but the off-axis aberrations have large sensitivity to the mirror spacing.  The only way to get the spacing right is to set the image plane at a known location, which is the specified back working distance (AKA back focus distance.)  When the scope is set to focus on this "reference plane," the mirror spacing will be correct.  So, the accuracy required in setting the BWD depends strongly on the size of the sensor that you use.  The bigger the sensor, the more careful you need to be.  As I recall, Celestron states that the tolerance is within 0.5 mm in order to achieve the specified field for the Edge scopes.  I've done a fair amount of analysis on this and they are being pretty conservative with this recommendation.  Even for a large sensor, getting the spacing to within a millimeter is usually more than sufficient.

 

When you compute the BWD, you need to include all the physical distances and account for any glass in the path.  Any plane-parallel window (or filter) will increase the physical distance by roughly one third of it's thickness.  On Celestron scopes, the reference plane is at the rear surface of the baffle nut so there where everything is measured from.

 

John

John, how would you suggest I go about getting to within 1-mm that seems daunting. I added up the thickness of the focuser including the number of steps I am currently "out", the filter-wheel housing and the CCD offset and I'm well within the 133.35mm distance that I need to reach.  But I know that this is not true because I'm not taking into account the distance I lose because of connection points between the scope and focuser, the focuser and the filter-wheel and the filter-wheel and the camera.  I have a set of digital calipers but the jaws are not long enough to touch the surface of the Edge rear thread and the CCD camera at the same time.  I looked into buying some extra long jaw calipers and they are very expensive but I'm not confident that they would be able to reach past the focuser and filter wheel.  How do people go about getting this exact back focus offset? When dealing with 1mm tolerances I can't imagine trying to fabric some DIY solution. Question 1) From the Celestron White Paper, page 5, it seems to suggest that even if you are 5mm off-axis distortion is still very similar to being on-axis is this true?  

 

EdgeOffset.jpg

 

2)  Am I overly concerned about all this since none of this takes into account seeing conditions? 



#17 nicklin

nicklin

    Lift Off

  • -----
  • Posts: 5
  • Joined: 16 Sep 2018
  • Loc: Canterbury, UK

Posted 14 July 2019 - 07:09 AM

how would you suggest I go about getting to within 1-mm that seems daunting.

Baader have a system of extensions called VariLock with an accurate printed scale that are adjustable to sub millimeter, e.g.

 

https://www.firstlig...nsion-tube.html

 

I have both a 20-29mm one and a 29-46mm as I wasn't sure what I'd need and they work well, though they do let light in very slightly, I suspect due to the indented scale, so I avoid stray light hitting the tube.


Edited by nicklin, 14 July 2019 - 07:10 AM.


#18 WalterG

WalterG

    Explorer 1

  • -----
  • topic starter
  • Posts: 51
  • Joined: 26 Dec 2011

Posted 14 July 2019 - 08:20 AM

Baader have a system of extensions called VariLock with an accurate printed scale that are adjustable to sub millimeter, e.g.

 

https://www.firstlig...nsion-tube.html

 

I have both a 20-29mm one and a 29-46mm as I wasn't sure what I'd need and they work well, though they do let light in very slightly, I suspect due to the indented scale, so I avoid stray light hitting the tube.

I did not know that such an adapter was made, thank you for the heads up.  I am still not sure how I can accurately measure the total distance that I currently have.  To recap, the problem that I think I have are the connection points between the scope, focuser and filter-wheel and camera are adding length to my system.  If I add up the theoretical length of all the three components I have 117.5 mm while I need 133.35 so theoretically I should be fine, but I'm sure the connection points are adding more than 15.85 mm.  Question 1) Do you have any suggestions on how I can measure accurately to within mm's the total distance I currently have?  With my crude measurement, I think I'm at 145 mm.  So I have too long of an imaging train by ~ 11mm. 



#19 nicklin

nicklin

    Lift Off

  • -----
  • Posts: 5
  • Joined: 16 Sep 2018
  • Loc: Canterbury, UK

Posted 14 July 2019 - 03:56 PM

Do you have any suggestions on how I can measure accurately to within mm's the total distance I currently have? 

Simply running a tape measure from the back of the scope to approximately where the sensor is and getting as close to what you believe might be the right distance should do as a starting point. Assuming you can focus, get an image with stars in the corners, zoom in to see how they look and try to refine from there if needed.  Even if you're 10mm or so off, I wouldn't worry because maybe the results will still be acceptable.

 

The 105mm distance everyone quotes for the Celestron f6.3 reducer gives a 0.67 reduction not 0.63 for my OTA, so is wrong if 0.63 is what one was expecting, and the best distance for star shape is unclear to me at the moment - about 105mm worked fine when I had a small sensor, but switching to the ASI294 is revealing a can of worms in this area now as I'm getting a larger field of view, and I doubt 105 is what I'll need to end up with.


  • WalterG likes this

#20 jhayes_tucson

jhayes_tucson

    Fly Me to the Moon

  • *****
  • Posts: 7100
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 14 July 2019 - 04:35 PM

John, how would you suggest I go about getting to within 1-mm that seems daunting. I added up the thickness of the focuser including the number of steps I am currently "out", the filter-wheel housing and the CCD offset and I'm well within the 133.35mm distance that I need to reach.  But I know that this is not true because I'm not taking into account the distance I lose because of connection points between the scope and focuser, the focuser and the filter-wheel and the filter-wheel and the camera.  I have a set of digital calipers but the jaws are not long enough to touch the surface of the Edge rear thread and the CCD camera at the same time.  I looked into buying some extra long jaw calipers and they are very expensive but I'm not confident that they would be able to reach past the focuser and filter wheel.  How do people go about getting this exact back focus offset? When dealing with 1mm tolerances I can't imagine trying to fabric some DIY solution. Question 1) From the Celestron White Paper, page 5, it seems to suggest that even if you are 5mm off-axis distortion is still very similar to being on-axis is this true?  

 

attachicon.gif EdgeOffset.jpg

 

2)  Am I overly concerned about all this since none of this takes into account seeing conditions? 

 

1)   I simply measure all of the components in the path and use part specs to compute the distance from the reference surface to the sensor.  Just be sure that you account for the addition spacing required by any glass windows in the path.

 

2)  The diagram that you are referencing shows the field performance of the system when the back working distance is properly set.  The spots will become larger and more aberrated as you move off axis when the BWD is not set properly.

 

John


  • WalterG likes this


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