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

Celestron EDGE HD reducer backfocus

  • Please log in to reply
19 replies to this topic

#1 Janco

Janco

    Explorer 1

  • -----
  • topic starter
  • Posts: 89
  • Joined: 21 Aug 2017
  • Loc: Johannesburg, South Africa

Posted 17 November 2019 - 11:08 AM

Hi Everyone, 

 

So this weekend I started to fiddle around a bit with my telescope. 

 

Over the last couple of months I've been having issues in my photos where it seems the stars in the upper left corner were slightly elongated towards the center of the image. So I decided I am gonna ignore the 146mm backfocus as stated by Celestron and went on a trail and error venture, basically setting the moonlite focuser to an increasingly racked out position and refocus with the stock focuser. I eventually reached a point where I was happy with the stars in all four corners of my images and measured the backfocus and what this was achieve. Turns out, the backfocus distance was 149.5 mm... 

 

I figured since these items our bulk produced one cannot expect the backfocus of these reducers to by 146mm spot on every time? I don't know... 

 

Does anyone here maybe share my experience?


  • jdupton and pyrasanth like this

#2 pyrasanth

pyrasanth

    Surveyor 1

  • *****
  • Posts: 1,984
  • Joined: 08 Jan 2016

Posted 17 November 2019 - 12:06 PM

This is interesting. I think you will always find mechanical differences in the stated lengths. The problem gets worse as you add more components into the connecting train. I found on my C14 that the 146 mm was pretty much as it should be but who knows if I just got lucky as there were a lot of adaptors required to connect the IFW filter wheel to the C14.

 

I've put the 11" RASA on the mount and taken the whole C14 camera assembly with filter wheel attached and moth balled it for next time use. The spacing of the components is so critical I dare not loose one of the fittings.



#3 jdupton

jdupton

    Vanguard

  • *****
  • Posts: 2,106
  • Joined: 21 Nov 2010
  • Loc: Central Texas, USA

Posted 17 November 2019 - 12:20 PM

Janco,

 

   I have had exactly the same experience. I found the best back-focus distance for me EdgeHD 11" & EdgeHD 0.7X reducer the same way you did -- set my Moonlight at some point, Focus with stock focuser, take and image, Inspect the corners for round stars, set a new Moonlight position 100 steps away and repeat. I have not tried to measure the actual back-focus on mine but my Plate Solve of the clean optimized images shows 1974-1976 mm focal length rather than the "specified" 1960 mm focal length for 146 mm back-focus with reducer.

 

   I agree with your conclusion. No mass produced manufacturing process will yield all copies of a product at exactly the same design specification. Tolerances are to be expected.

 

 

John


  • Janco likes this

#4 Janco

Janco

    Explorer 1

  • -----
  • topic starter
  • Posts: 89
  • Joined: 21 Aug 2017
  • Loc: Johannesburg, South Africa

Posted 17 November 2019 - 01:34 PM

Janco,

 

   I have had exactly the same experience. I found the best back-focus distance for me EdgeHD 11" & EdgeHD 0.7X reducer the same way you did -- set my Moonlight at some point, Focus with stock focuser, take and image, Inspect the corners for round stars, set a new Moonlight position 100 steps away and repeat. I have not tried to measure the actual back-focus on mine but my Plate Solve of the clean optimized images shows 1974-1976 mm focal length rather than the "specified" 1960 mm focal length for 146 mm back-focus with reducer.

 

   I agree with your conclusion. No mass produced manufacturing process will yield all copies of a product at exactly the same design specification. Tolerances are to be expected.

 

 

John

 

I am glad to hear I am not alone John,

 

I must say I was more relieved than anything else when I made the discovery. I would've been a lot more frustrated if I was using the Celestron T adapters and only be able to adjust the sensor through spacers.


  • jdupton likes this

#5 Benni123456

Benni123456

    Mariner 2

  • -----
  • Posts: 288
  • Joined: 20 Apr 2017
  • Loc: Munich, Germany

Posted 17 November 2019 - 01:47 PM

as i appear to see something similar with my edge 8 and reducer,

 it would perhaps be interesting if all users try several distances, note their edge model and show a foto that proves everything is round.

 

On my edge 8 it just appears to be more massive. advertised is 105. i come close to round at 136 and im still finding whether i should go out further.

 

Perhaps when we have many measurements we can find out whether all of these reducers need the same distance at a point different from the celestron spec....

 

On the reducer for my edge 8, it reads "made in china". It maybe that during the transition to a different manufacturing process in another country (originally i guess these things were made in california), things can happen...

 

Also, it may even be that they never devoted much time when measuring these specs.

 

Perhaps these are for an entirely outdated specimen and the factory has improved and tweaked the optics since then, and just the specs were not upgraded....


Edited by Benni123456, 17 November 2019 - 01:51 PM.


#6 Janco

Janco

    Explorer 1

  • -----
  • topic starter
  • Posts: 89
  • Joined: 21 Aug 2017
  • Loc: Johannesburg, South Africa

Posted 19 November 2019 - 01:20 PM

I hope my pictures are of good enough quality for you to see. 

 

Below is a picture of the small cluster NGC2100, close to the Tarantula Nebula placed in the top right corner of my photos at 200% zoom in SGP. The one picture shows the stars with the distance between the sensor and reducer measuring 146mm exactly. As can be seen the stars are slightly flared towards the center of the image. The second photo is the stars in the same corner at a backfocus of 149.5 mm. In my opinion they look much better here. I can probably tweak the spacing a little bit more, but the improvement is evident. 

 

Backfocus.jpg

 

 



#7 Benni123456

Benni123456

    Mariner 2

  • -----
  • Posts: 288
  • Joined: 20 Apr 2017
  • Loc: Munich, Germany

Posted 19 November 2019 - 02:01 PM

I hope my pictures are of good enough quality for you to see. 

 

Below is a picture of the small cluster NGC2100, close to the Tarantula Nebula placed in the top right corner of my photos at 200% zoom in SGP. The one picture shows the stars with the distance between the sensor and reducer measuring 146mm exactly. As can be seen the stars are slightly flared towards the center of the image. The second photo is the stars in the same corner at a backfocus of 149.5 mm. In my opinion they look much better here. I can probably tweak the spacing a little bit more, but the improvement is evident. 

 

 

It may become interesting to compare the different colors.

 

Usually, the images with the reducer at factory distance have elongations mostly in blue in the too far in elongation style (eggs pointing to the center). Going out a bit too far reduces or removes elongations in blue but then red becomes elongated in the too far out elongation style (eggs pointing around the center).

 

The question is whether there is some sweet spot where all the colors render into point like star shapes. That should then be the correct distance.

 

As for the length of the distance:

The thing may react to spacing differences of 0.5 or even 0.1 mm with the reducer. 

 

For me, it is actually quite annoying to search the optimal distance via spacers. But with filter wheel, oag, tilting adapter and ao in the imaging train, a focuser would not fit...

 

If the image with the reducer is better at a certain backfocus, one may also test whether the starshapes are better natively without reducer at this distance.

 

 

Interesting is that for my scope (an edge 800) 2mm out of the native 134mm is still a bit too short.

I wonder what happens at 138mm....

 

The distances seem to be similarly off as with your c14

 

If all these scopes have a backfocus that is around 4mm longer, it would be interesting.

 

At least it would reduce the annoying stress of fiddling around with spacers and focusers.



#8 Astrojedi

Astrojedi

    Skylab

  • *****
  • Posts: 4,138
  • Joined: 27 May 2015
  • Loc: SoCal

Posted 20 November 2019 - 08:37 PM

It may become interesting to compare the different colors.

 

Usually, the images with the reducer at factory distance have elongations mostly in blue in the too far in elongation style (eggs pointing to the center). Going out a bit too far reduces or removes elongations in blue but then red becomes elongated in the too far out elongation style (eggs pointing around the center).

 

The question is whether there is some sweet spot where all the colors render into point like star shapes. That should then be the correct distance.

 

Seriously... every EdgeHD thread you are complaining about the EdgeHD and keep bringing up the color issue. How many times have we had this discussion. Give it a rest. Go buy a different scope if this does not meet your standards - It is a mass produced scope, there will be some variation.


  • Janco likes this

#9 Cfreerksen

Cfreerksen

    Viking 1

  • *****
  • Posts: 703
  • Joined: 18 Aug 2018
  • Loc: Tooele, Ut

Posted 21 November 2019 - 12:19 AM

Seriously... every EdgeHD thread you are complaining about the EdgeHD and keep bringing up the color issue. How many times have we had this discussion. Give it a rest. Go buy a different scope if this does not meet your standards - It is a mass produced scope, there will be some variation.

Or maybe we should just to narrow band imaging. Focus for each filter. no more color issue. After all the reducer is but a $300 chunk of glass and metal. Not really high end pricing for f7.

 

Chris



#10 Janco

Janco

    Explorer 1

  • -----
  • topic starter
  • Posts: 89
  • Joined: 21 Aug 2017
  • Loc: Johannesburg, South Africa

Posted 21 November 2019 - 03:30 AM

Seriously... every EdgeHD thread you are complaining about the EdgeHD and keep bringing up the color issue. How many times have we had this discussion. Give it a rest. Go buy a different scope if this does not meet your standards - It is a mass produced scope, there will be some variation.

Hi Astrojedi, completely agree. My post was not about false color but more about field flatness.

 

We have to be reasonable in our expectations. I was just glad to have found a solution to my problem and wanted to share my findings in the hope that it might help someone and doing a sanity check to see if others have experienced similar things. I do not buy Celestron's claim that these reducers have a backfocus of exactly 146 mm for one second on a mass produced scope. 


Edited by Janco, 21 November 2019 - 03:33 AM.


#11 jhayes_tucson

jhayes_tucson

    Fly Me to the Moon

  • *****
  • Posts: 7,232
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 21 November 2019 - 04:48 AM

I hope my pictures are of good enough quality for you to see. 

 

Below is a picture of the small cluster NGC2100, close to the Tarantula Nebula placed in the top right corner of my photos at 200% zoom in SGP. The one picture shows the stars with the distance between the sensor and reducer measuring 146mm exactly. As can be seen the stars are slightly flared towards the center of the image. The second photo is the stars in the same corner at a backfocus of 149.5 mm. In my opinion they look much better here. I can probably tweak the spacing a little bit more, but the improvement is evident. 

 

attachicon.gif Backfocus.jpg

 

Changing the back working distance will have the biggest effect on field performance and changing the BWD by 3 mm is a pretty modest change.  How much of the field is shown in your images?  From what I can see, the star shapes are constant over the field indicating that the biggest issues that you have in each of these images is soft focus and possibly guide or vibration errors.  In order to accurately assess field performance, you have to have perfect focus, zero vibration, and no guide errors.  The best way to minimize the effects of vibration and guiding issues is with a well aligned mount using short exposures on moderately bight stars.  I don't think that the differences that you are seeing in your test images have much, if anything, to do with the change in the back working distance.

 

John



#12 Janco

Janco

    Explorer 1

  • -----
  • topic starter
  • Posts: 89
  • Joined: 21 Aug 2017
  • Loc: Johannesburg, South Africa

Posted 21 November 2019 - 05:08 AM

Changing the back working distance will have the biggest effect on field performance and changing the BWD by 3 mm is a pretty modest change.  How much of the field is shown in your images?  From what I can see, the star shapes are constant over the field indicating that the biggest issues that you have in each of these images is soft focus and possibly guide or vibration errors.  In order to accurately assess field performance, you have to have perfect focus, zero vibration, and no guide errors.  The best way to minimize the effects of vibration and guiding issues is with a well aligned mount using short exposures on moderately bight stars.  I don't think that the differences that you are seeing in your test images have much, if anything, to do with the change in the back working distance.

 

John

Hi John, 

 

The field shown in the photos is a very small part of the photo in the very top left corner of the frame, using an ASI1600MM. As mentioned before it is NGC2100 in the photo, which measures about 2.8 arcmin The crops were made in SGP using a snipping tool at 200% zoom. What I did with each of these shots was to move the camera back with the moonlite focuser, do rough focus with stock focuser and ran an auto focus routine in SGP. Then the images were taken. It was 3AM in the morning that I was messing around with these things so I can possibly spend a bit more time on getting the best possible performance but what was clear to me is that 146mm is not the optimal backfocus distance in my instance. It may be possible that the optimal back focus is >150mm. I'll test it a bit more when I have a clear night again. 

 

The difference that is being observed in the two photos were repeatable and not mere variations in tracking performance. Stars were round everywhere else in the field almost. Below is a screenshot of some stars near the center of the field (once again very zoomed in and cropped). Sorry I don't have the correct tools to make these kind of assessments easier. Each exposure was 15 minutes long and tracking was averaging around 0.4-0.6" total RMS. 

 

Untitled-2.jpg


Edited by Janco, 21 November 2019 - 05:49 AM.


#13 Benni123456

Benni123456

    Mariner 2

  • -----
  • Posts: 288
  • Joined: 20 Apr 2017
  • Loc: Munich, Germany

Posted 21 November 2019 - 07:21 PM

Seriously... every EdgeHD thread you are complaining about the EdgeHD and keep bringing up the color issue. How many times have we had this discussion. Give it a rest. Go buy a different scope if this does not meet your standards - It is a mass produced scope, there will be some variation.

well, what im seeing is something like this:

 

The unsharp picture is from the edge 8 with reducer at 105mm (as advertised), the sharper picture is from the edge with reducer at 136mm...

Maybe its just a feeling, but look at the small stars. They are looking blured in the 105mm image, and they are coming out more pinpoint in the 136mm image.

 

In the entire image, one sees that the scope at 136mm is now looking almost as good as my refractor. Simply because the stars are somehow sharp..

 

And they are less donuts at the edges....

 

I will try it at 138 mm tomorrow and then look at the results again... 

Attached Thumbnails

  • m27blue.JPG


#14 Benni123456

Benni123456

    Mariner 2

  • -----
  • Posts: 288
  • Joined: 20 Apr 2017
  • Loc: Munich, Germany

Posted 21 November 2019 - 07:22 PM

this is at 136mm

 

Many, even small stars now burn out. Even though the exposure time was the same....

Attached Thumbnails

  • bubbleblue.JPG

Edited by Benni123456, 21 November 2019 - 07:27 PM.


#15 Benni123456

Benni123456

    Mariner 2

  • -----
  • Posts: 288
  • Joined: 20 Apr 2017
  • Loc: Munich, Germany

Posted 21 November 2019 - 07:31 PM

maybe one can see it better in the entire image. The unsharp image at 105mm backfocus is this

 

76695180_2346051928851680_81276839521744

 

 

 

and here is the sharper one at 136mm

 

73546658_2334478066675733_85067738632568


 

 

here is one at 139mm but it has miscollimation. because of miscollimation there are donuts over the entire field. So i can not know if this distance were better...

 

74651322_2317409905049216_74240761562713



#16 Benni123456

Benni123456

    Mariner 2

  • -----
  • Posts: 288
  • Joined: 20 Apr 2017
  • Loc: Munich, Germany

Posted 21 November 2019 - 07:37 PM

So i hope you see that im not just complaining...  I just wondered why i did get better pictures with my 65/420 refractor than with my more expensive edge with the reducer. Going to around its native distance seems to have resolved much of the mystery. 

I guess i now need to optimize the spacing and then I have a usable scope. In a climate where clouds are common, this is just a bit painful. 

When they collimate these scopes at factory, they really could make a field test and write the optimum spacings on a sheet or on the tube itself if they differ for every specimen. Which I do not believe.

 

I think that all these scopes are mass produced and have the same optimum spacings up to 0.1 mm....



#17 Benni123456

Benni123456

    Mariner 2

  • -----
  • Posts: 288
  • Joined: 20 Apr 2017
  • Loc: Munich, Germany

Posted 21 November 2019 - 07:42 PM

to compare, this is the native edge. For M27 and the bubble:

 

gallery_273095_11730_31379.jpg

 

 

gallery_273095_11730_569088.jpg



#18 Astrojedi

Astrojedi

    Skylab

  • *****
  • Posts: 4,138
  • Joined: 27 May 2015
  • Loc: SoCal

Posted 21 November 2019 - 08:02 PM

I always prefer imaging at F10 since you are getting the full diffraction limited performance of the design. But if I am honest with you I think your F7 images don't look so bad either. I think you are expecting too much from a mass produced scope at this price point.



#19 Benni123456

Benni123456

    Mariner 2

  • -----
  • Posts: 288
  • Joined: 20 Apr 2017
  • Loc: Munich, Germany

Posted 21 November 2019 - 09:16 PM

actually, since im around 134mm with the reducer, it comes now close to what it should be.

 

numerically, the unsharp m27 at 105mm has in the ha narrowband channel

 

fwhm 10.95 excentricity 0.449

 

the bubble at 135mm has 

fhwm 7.43 and excentricity 0.377

 

one should note that the bubble was made from my balcony with much seeing. the unsharp m27 was made in the field without much seiing.

 

At first sight it seems m27 shows less eggs. but that is just because it is so unsharp that the blue eggy smear does only get noticed in the raw images. 

 

 

I think the reducer, at the correct distance, should not affect resolution in any way. It has many lenses to correct its own color error, coma or field curvature.

 

the reducer has no way, of course, to correct the sphaerochromatism and remaining coma of the aspheric schmidt corrector.

 

The edge 8 is optimized to minimize these errors at a backfocus of around 134mm.

 

If it has blur at other distances, the reducer will show blur there too. 

 

So the claim from celestron to go to 105mm for the reducer of the edge 8 seems to make physically no sense to me.

 

 

 

So yes, what i believe is this:

 

If my edge with the reducer shows an optimum distance at say 138mm or something at f7, then 138 mm will also be the correct backfocus when using it at f10...

 

 

My guess is also that at the correct backfocus, the edge with and without reducer is diffraction limited or close to it.

 

With reducer at 136mm, my ao and phd guiding seem to get the stars bumping around within a circle of 0.5 arcsecond, with seiing on my balcony. That is the scopes resolution.

 

What is interesting is: At 105mm with the reducer, i have difficulties finding a guide star because everything is so blurred.

 

At 136mm the stars burn out in the guiding cam.

 

This means i can now regularly use 0.05 sek exposures and so on, which is nice if you have an active optics.

 

Its also good for guiding performance if the stars are small dots....

 

Finally, I guess that these scopes are all mass produced and precise.

 

I therefore think they have tolerances of 0.1mm. So if celestron writes a wrong backfocus for one scope, the advertisement sheets may have it wrong for all scopes.

 

These things can happen when a company outsources manufacturing to somewhere else and then some parts have a bit different specs, no one notices the change and the advertisement sheet stays in place..

 

So if one has here people with measurement devices like optical benches and zygos...

 

Perhaps it maybe funny to measure with testing devices whether the backfocus is really as advertised or not, with and without reducer for several edge hd specimen.

 

Then one can put up the correct backfocus values here and people then can look it up.

 

Testing every distance in milimeters simply wastes clear nights....


  • rosalino likes this

#20 Benni123456

Benni123456

    Mariner 2

  • -----
  • Posts: 288
  • Joined: 20 Apr 2017
  • Loc: Munich, Germany

Posted 21 November 2019 - 09:56 PM

here is a comparison again of a stack of the same edge (top right) in the ha channel from m27 and bubble with 105 and 136mm both with reducer.

 

I guess it is evident which one is sharper. A difference like night and day....

 

 

Focus was always with motor and first with bahtinov, then manual check with low intensity stars...

 

I like that it resolves the triple system on the right image.... thats a bit tricky and shows the precision of the thing.... as it should be.

 

And if anybody asks why I am posting this repeatedly:

Well, I think that this finding may hold for other scopes and other users of the edge 800 may get better pictures from the reducer when they use it with around 134mm.

 

Additionally, well, the spot diagram of the edge in the whitepaper shows that the edge is very backfocus sensitive. Apparently 0.1 mm can make a difference. But that means I am unable to find the correct backfocus alone.

 

1cm means I would have to take 100 stacks of images. And after each one I would need to recollimate.

This would take a year with the few clear skies at my location.

 

It may help if we do this collectively. Several people with edges and reducers testing different distances....

 

Or we need someone, e.g, a dealer with a test equipment like an interferometer, to roam through different backfocuses in steps of 0.1mm or even 0.01 mm and then take measurements of the sharpness at focus with and without reducer for a few edge telescopes for all possible backfocus positions....

 

Even with an interferometer, this measurement would probably take much time.

 

This may be why the numbers that Celestron gives are not accurate. They may never have really taken a month to measure it.

 

To do that on the sky is a bit daunting....

 

I hope now that soon the clouds will vanish....

Attached Thumbnails

  • Unbenannt100.JPG

Edited by Benni123456, 22 November 2019 - 11:14 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