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

comet stars in the image corner ?

  • Please log in to reply
15 replies to this topic

#1 drknoppi

drknoppi

    Lift Off

  • -----
  • topic starter
  • Posts: 15
  • Joined: 20 Feb 2014
  • Loc: Germany, Brandenburg

Posted 14 November 2017 - 03:38 PM

Hello,

i am  new in astroimaging with ccdcams like ASI 1600 MC cooled. in the past i got images with a ugly

phenomenon ; )) - Comet stars i the corner of the images . I used a SCT Meade 10 " ACF System with a reducer

from AP CCDT67 with a backspace of 120 mm behind the rear flange of the reducer . The chip of the ASi is about

17x13 mm . With PI script Imageanalysis i have got the attached image .

Whats is the reason of the issue ??- The backfocus ?, tilt in the light tray ? . I analysed the image with CCD Inspector .

but i cannot found  a solution . The object is NGC 891 with 180 sec and  -30 ° CCD Temperature .

in the last month i spend a lot of time to solve the problem, but i cant ; (((

I hope that the imagers of the CN community with great experience can help me to sove the problem ?

 

CCD I values : FWHM-1,92"; Aspect-26%-Stars-1115; Scale-0,43;PixSize-3.8;Focallength-1820mm , with reducer;Exposition-180";Collimation-2.2";Curvatur-82.4%;Tilt-total13%@31°

 

Best regards

Rainer

 

Attached Thumbnails

  • L_2017_11_13_20_43_58_Bin1x1_180s__30C_c_d_r_mosaic.jpg

Edited by drknoppi, 14 November 2017 - 03:40 PM.


#2 Tapio

Tapio

    Surveyor 1

  • -----
  • Posts: 1525
  • Joined: 24 Sep 2006
  • Loc: Tampere, Finland

Posted 14 November 2017 - 03:57 PM

Lot of curvature so I would say experiment with reducer ccd distance.

#3 drknoppi

drknoppi

    Lift Off

  • -----
  • topic starter
  • Posts: 15
  • Joined: 20 Feb 2014
  • Loc: Germany, Brandenburg

Posted 14 November 2017 - 04:27 PM

Hello Tapio,

thanks for the quick answer smile.gif , i have tested from 70 mm to 134 mm , but i cannot found the right distance.

I hope you have a solution , what do you think about the distance between 70 to 90 mm . But what is the right point to measure

the distance, the rear flange of the reducer or  the middle of the reducer ? And is  the addtion of about 16 mm like AP wrote to the backspace

to the CCD Sensor neccessary , or is the 16 mm a part of the distance ????

Last question, what is  absolutely critical for round stars , the collimation  in CCD I or the curvatur ???,

my autoguiding system is MGEN II with a guidescope 345 mm .

 

Bet regards Rainer


Edited by drknoppi, 14 November 2017 - 04:47 PM.


#4 Tapio

Tapio

    Surveyor 1

  • -----
  • Posts: 1525
  • Joined: 24 Sep 2006
  • Loc: Tampere, Finland

Posted 15 November 2017 - 12:21 AM

From AP site :
"Optimized for 0.67x compression with a fully illuminated circle of 29 mm when the back spacing distance is 85 mm from the rear flange to the camera sensor is used"

#5 drknoppi

drknoppi

    Lift Off

  • -----
  • topic starter
  • Posts: 15
  • Joined: 20 Feb 2014
  • Loc: Germany, Brandenburg

Posted 15 November 2017 - 12:59 AM

Thank you, first i will try a new collimation, because
AP say Coma and Astigmatism occured in the image above,
then i will see what happens . What is your method to collimate
a SCT carefully ? Did you used CCD Inspector ?

Best regards
Rainer

#6 einarin

einarin

    Viking 1

  • -----
  • Posts: 505
  • Joined: 23 Dec 2016

Posted 15 November 2017 - 01:22 AM

I'm kind of old fashioned collimating my SC - I just use camera & laptop and tweak so that I get most concentric rings with stars.

https://starizona.co...mating_sct.aspx



#7 AstroScience

AstroScience

    Messenger

  • ****-
  • Posts: 470
  • Joined: 28 Oct 2010
  • Loc: Israel

Posted 15 November 2017 - 02:12 AM

Hi Rainer,

using any kind of reducer with this scope will break the ACF and will introduce scope's inherent curvature. Your best bet is to use lower reduction ratio of the CCDT67 eniugh to cover your ship.

Optimum reduction of 0.67 you'll get when placed reducer at distance of 85mm from the flange of the reducer to the sensor. Additional 16mm is the optical backfocus of the reducer. 85+16=101. You may try to place reducer, at distance from the flange to sensor maybe at 83-82 mm or maybe closer. 

Let's say you place the reducer at 82mm from the flange, so calculation should be as follows:

(305-82+16)/305=0.78 that would be the new reduction ratio.


Edited by AstroScience, 15 November 2017 - 02:17 AM.


#8 drknoppi

drknoppi

    Lift Off

  • -----
  • topic starter
  • Posts: 15
  • Joined: 20 Feb 2014
  • Loc: Germany, Brandenburg

Posted 15 November 2017 - 12:36 PM

Hello Einarin and Astroscience,
thank you for your replies 😊, what you think about the neccesary value of collimation in arcsec in CCDI ?
Value < 0.5 arcsec ? In summary i wil try so 😉 as you
wrote above , thank you !!!!

#9 drknoppi

drknoppi

    Lift Off

  • -----
  • topic starter
  • Posts: 15
  • Joined: 20 Feb 2014
  • Loc: Germany, Brandenburg

Posted 15 November 2017 - 01:05 PM

Hello George, that means, to reduce the curvatur i
need a flattener additionally to the reducer , right ?
Regards
Rainer

#10 Tapio

Tapio

    Surveyor 1

  • -----
  • Posts: 1525
  • Joined: 24 Sep 2006
  • Loc: Tampere, Finland

Posted 15 November 2017 - 03:30 PM

I'm not 100% sure but I thought that CCDT67 was one of the few reducers that would fit ACF type scopes because it doesn't have flattener properties.
ACF doesn't need flattener because it already have reasonable flat field.

#11 Jared

Jared

    Fly Me to the Moon

  • *****
  • Posts: 5994
  • Joined: 11 Oct 2005
  • Loc: Piedmont, California, U.S.

Posted 15 November 2017 - 04:38 PM

The ACF is NOT a flat field scope.  It is just a coma free scope.  It has a field that is "flat enough" for many uses, but it's definitely not flat.  It isn't advertised as such, either.

 

By adding the reducer, you are actually exacerbating that assuming you are using the entire field of view of your chip.  Reducers have a very slight flattening effect for a given field of view, but if you use the entire field of view that becomes visible as a result of the focal reduction, they actually make things worse--much worse if you have much reduction.  So, while your ACF scope has a field that is probably flat enough for the ASI when used at the native f/8, the field curvature becomes more much noticeable with the wider field of view you get with the reducer.

 

At 89mm of back focus, the CCD should give its intended reduction of 0.67.  I'd try something near that for your size chip and see what CCDI says.  Generally, most objects look pretty good if you can get the curvature (as reported by CCDI) down to 20% or so.  Keep in mind that CCDI's method of calculation of curvature is kind of arbitrary and does not provide an absolute value since it is tied to the camera you are using with your scope, but I can tell you from experience that once you are down to around 20% the corners will look nearly as sharp as the center.

 

In addition to the field curvature, you have two other things going on.  First, you have some off axis coma that is visible and you appear to also have quite a bit of tilt.  I'd check the collimation and see if that improves the stars in the top left corner (the off axis coma).  Then I'd go after the tilt.  Make sure everything is mounted squarely.  Use threaded adapters to attach everything--if you have any slip fit compression attachments you may want to get rid of those.  


  • jdupton likes this

#12 Jared

Jared

    Fly Me to the Moon

  • *****
  • Posts: 5994
  • Joined: 11 Oct 2005
  • Loc: Piedmont, California, U.S.

Posted 15 November 2017 - 04:47 PM

Oh, and a couple more minor recommendations:

 

- I would use much shorter exposures of a rich star field with CCDI in order to keep guiding errors out of the equation or at least minimized.  It can be hard to tell with a 180s exposure how much of, for example, the elongation of stars in the middle is due to collimation and how much from tracking.  Something more like 10s would work well.

- I'm not certain where the 1,820mm focal length calculation came from.  If you have an f/10 scope and are using the reducer at its recommended back focus you would get a 1,700mm focal length.  If you are using the reducer at 120mm, your focal length would be even lower.  If you have an f/8 scope, it would be lower still.  It only matters whether you have the right focal length if you want to do comparisons to different scopes or different configurations, but you probably can't trust things like the FWHM calculation if you don't have the correct focal length.

- For collimation, I would recommend just doing it visually using a high power eyepiece that is slightly defocused.  You can do it using CCDI, but it's more work and may not be perfectly accurate if you have compounding issues like tilt.  Visual will be the most straightforward.  It also guarantees that you will be collimating for the center of the field.



#13 drknoppi

drknoppi

    Lift Off

  • -----
  • topic starter
  • Posts: 15
  • Joined: 20 Feb 2014
  • Loc: Germany, Brandenburg

Posted Yesterday, 01:43 AM

Hello Jared,

thank you very much  for your answer`s smile.gif . I understand , that i have a collimationserror and a tiltproblem . The 1,820mm focal length calculation came from PI and  the PhotometricColourCalibration Tool and my scope is Meade 10 " ACF . What would you think , useful or not ?

Next -  if the ACF is not flat as you say, i don`t understand now, why i don`t need a additionally flattener ?

 

Todo list  for me :

1. Collimation first,

2. Evaluation of focal length  with the reducer - but how to do ? AP _for CCDT67 :  305-distance + 16 mm / 305 is the F-ratio , for SCT`s  to ?  d would be than if i have the ratio d=321-305*F for CCD T67

3. Remove of any tilt in the optic train - I have a tiltadapter from ZWO - all elements in the train are threaded adapters .

4.Shorter exposure times to see what is more collimationserror / trackingerror ? - i will looking for short exposures and  show you the images .

 

I hope i have understand ?

 

Best wishes and thank you for your help !!!!!

Rainer



#14 drknoppi

drknoppi

    Lift Off

  • -----
  • topic starter
  • Posts: 15
  • Joined: 20 Feb 2014
  • Loc: Germany, Brandenburg

Posted Yesterday, 03:12 AM

Hello Jared,

Here i found  1 image Mirfak unguided without reducer at 2500 mm F 10 and  30 seconds exposure time and

1 image NGC 6946 unguided with CCD T 67 and d=134 mm 30 sec , at 1400 mm ( theoretical F0.561 (  305-134 /305 see AP CCDT67 )

the influence of autoguider MGEN II is zero .

Best regards Rainer

Attached Thumbnails

  • NGC6946_unguided_1769.35mm_30sec_CCDT67_d=134mm.jpg
  • Mirfak_unguided_2500mm_30sec_nonCCDT67.jpg

Edited by drknoppi, Yesterday, 12:19 PM.


#15 Jared

Jared

    Fly Me to the Moon

  • *****
  • Posts: 5994
  • Joined: 11 Oct 2005
  • Loc: Piedmont, California, U.S.

Posted Yesterday, 05:09 PM

Hello Jared,

thank you very much  for your answer`s smile.gif . I understand , that i have a collimationserror and a tiltproblem . The 1,820mm focal length calculation came from PI and  the PhotometricColourCalibration Tool and my scope is Meade 10 " ACF . What would you think , useful or not ?

Next -  if the ACF is not flat as you say, i don`t understand now, why i don`t need a additionally flattener ?

 

Todo list  for me :

1. Collimation first,

2. Evaluation of focal length  with the reducer - but how to do ? AP _for CCDT67 :  305-distance + 16 mm / 305 is the F-ratio , for SCT`s  to ?  d would be than if i have the ratio d=321-305*F for CCD T67

3. Remove of any tilt in the optic train - I have a tiltadapter from ZWO - all elements in the train are threaded adapters .

4.Shorter exposure times to see what is more collimationserror / trackingerror ? - i will looking for short exposures and  show you the images .

 

I hope i have understand ?

 

Best wishes and thank you for your help !!!!!

Rainer

Not certain where PI is getting the 1,820mm focal length, but it doesn't really matter as long as you are only doing comparison from before/after as opposed to comparisons with other scopes.  Is your scope the f/10 version or the f/8?  I believe Meade currently offers both.

 

In any event, the reason most people don't need an additional flattener with the ACF is that it is "flat enough".  Stars in the corners are usually pretty good as long as your camera has a small enough chip and as long as you aren't trying to use too much focal reduction.  I can't tell you, though, that you'll be fine as long as your chip is smaller than 'x' mm or as long as your focal length is greater than 'y'mm.  First, I don't own this particular scope so I would just end up looking up other people's results just as you can, and second because everyone's standards for what constitutes "good enough" is different.  None of my telescopes--not one--has stars that are as good in the extreme corners as in the center of the field.  That includes OTA's as expensive as a $20,000 Astro-Physics Riccardi-Honders.  But all of them meet the standard of 'good enough'.  My personal standard of 'good enough' is that FWHM in the extreme corners--all four--should be within about 5% of the center of the field.  That's enough for me to stop caring.  So if I can get 1.9" stars in the center of the field, I would be fine with 2" stars in the corner, but I would work on improving things if I were at 2.1" or 2.2".  Improving things might mean choosing a different focus point not in the center (what I did with an RC I had), or tweaking the spacing behind a flattener, or getting that last little bit of tilt out of the optical train.  You'll have to set your own standards.  I can tell you, though, that there is always more that could be done.  At some point you need to say, "I'm satisfied for now, and I want to start making pictures not optimizing my scope."

 

You list of "to do's" seems reasonable.  Don't stress too much on the focal length.  It only matters if you want to compare your results with other people's.  Personally, I would recommend you put the spacing back at the AP spec of 90mm or so (assuming you have the room) and assume it's working at 0.67x.  The only way to get it exact is with a plate solve, and that's overkill for now.  

 

Keep in mind I was recommending the shorter exposure times just for evaluating the optics.  Not for making your actual images.  Short exposures will remove the tracking variable from the equation so you can evaluate just the optical alignment and quality.

 

Good luck!



#16 Jared

Jared

    Fly Me to the Moon

  • *****
  • Posts: 5994
  • Joined: 11 Oct 2005
  • Loc: Piedmont, California, U.S.

Posted Yesterday, 05:27 PM

Hello Jared,

Here i found  1 image Mirfak unguided without reducer at 2500 mm F 10 and  30 seconds exposure time and

1 image NGC 6946 unguided with CCD T 67 and d=134 mm 30 sec , at 1400 mm ( theoretical F0.561 (  305-134 /305 see AP CCDT67 )

the influence of autoguider MGEN II is zero .

Best regards Rainer

Both these images look substantially better than your original post, but it's hard to say when they are reduced in resolution.  

 

With 134mm of back focus for the CCD T67 you will be running close to f/5 (as you noted).  Probably not quite because the focal length of the scope itself increases when you insert the telecompressor.  Maybe f/6?  A plate solve would tell you for sure.  Even with the relatively small 4/3" chip I'd be surprised if you didn't see at least some field curvature in the corners, but if you are happy with it who cares?  Certainly there is no obvious coma with either of these two images which is good.

 

So why is there no coma with these two?  I assume these are both images you took earlier than the edge-on galaxy image you showed earlier. Either something has changed in your scope with respect to collimation (very possible) or perhaps your focusing technique is showing some variability in collimation.  Mirror flop affects collimation.  When you collimate an SCT--even an ACF or a Celestron Edge--you are collimating for a particular tilt of the primary mirror.  Unless you have literally zero image shift as you focus (extremely rare), your scope can be perfectly collimated when you focus in one direction, and not quite perfect when you focus in the other direction.  The usual recommendation is to make sure you always finish focusing by turning counter clockwise.  That way you are always "lifting" the mirror against gravity, so it will tend to shift less as you track.  You should do this both for imaging as well as for collimation.  If you already know this, sorry for the lecture.

 

The other option, of course, is just that the telescope has shifted out of collimation a bit.  That's not unusual for an SCT.

 

As far as the apparent lack of tilt in these two samples... Not certain why you are getting a different result than with the galaxy you posted earlier.  I'm assuming the telecompressors is threaded to the camera with whatever spacers you are using to get the 134mm of back focus.  How is the camera attached to the scope?  Are you just inserting it into the visual back?  Or is it also threaded?  If you are using the visual back, that is the likely source of tilt from the first image you posted.  




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.







Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics