Jump to content


Photo

Compensating For Coma Corrector Induced Tilt

  • Please log in to reply
19 replies to this topic

#1 Peter10

Peter10

    Vostok 1

  • -----
  • Posts: 106
  • Joined: 31 Mar 2007

Posted 12 July 2014 - 08:18 PM

As some of you may know, I have been fighting alignment/stability issues with my 8" f/5 Newtonian and ASA 0.73X focal reducer. Since my last post, I have made significant progress using a work-around technique that essentially gets me where I need to be. The problem is that the "technique" is not deterministic and it may not be the best solution to the problem.

Summary:

With the reducer/corrector installed, the telescope is operating at around f/3.9. Not slow, but not outrageously fast either. I align using the Catseye collimation tools and can achieve a picture perfect result as viewed through the tools. My basic collimation is unquestionably good.

Post collimation, I install the ASA corrector and take some short exposures (~20 sec) to assess imaging performance. With the corrector installed the results are always bad, with elongated and often distorted stars. The central region of the frame(s) is "almost" acceptable, but things get really dismal at the four corners of the image. Exposures taken without the corrector do not show this behavior but show lots of field dependent coma (as expected).

To get around this problem, I introduced a "Post Collimation" adjustment imparted at the primary mirror. I simply place a defocused star at the center of the field and adjust the primary mirror in tip/tilt until the image of the defocused star is as symmetric as possible. Once I complete this step, the results are dramatically better than the default state provided by the CatsEye Tools. In effect, I am decollimating the telescope to compensate for the "issues" with the ASA corrector. This is not optimal!

I sent the corrector back to the manufacturer suspecting that the optics were misaligned - but the "newly verified" corrector performed the same way. Apparently I have to live with things the way they are.

I am wondering if it might be advantageous to correct this "tilt issue" using the secondary mirror as opposed to the primary?

What I really need is a way to tilt the corrector optics with respect the focuser axis. I do not think that tilting or shimming the focuser itself will be productive since it will both de-collimate the telescope AND tilt the corrector.

A tip tilt adapter will not work since the ASA corrector is ~3" long - limiting placement options within the imaging train. The input end of the corrector has a radically curved convex lens that prevents screwing the corrector to the focuser tube.

I know what needs to be done, but this piece of hardware (the corrector) is fighting me every step of the way. I will not give up, since I have seen hints at what it can do if I can ever get it properly aligned.

Can anyone think of a way to align this beast without corrupting my "known good" telescope alignment?

Peter

#2 152ED

152ED

    Explorer 1

  • -----
  • Posts: 64
  • Joined: 04 Jun 2009
  • Loc: Tehachapi & Sacramento, Ca

Posted 12 July 2014 - 09:44 PM

the corrector may be decentered under the focuser. I have a similar issue on my dob with its built in lumicon coma corrector, and I compensate same as you, with a star test collimation that does not match the cheshire collimation but gives tight focused stars.

going off of memory, my meade ed refractor showed on axis coma when the front lens was not centered correctly before I modified the cell and added nylon setscrews to fix it, so the same thing may be happening with a decentered coma corrector with respect to the optical axis of the primary mirror.

#3 gdd

gdd

    Apollo

  • -----
  • Posts: 1390
  • Joined: 23 Nov 2005
  • Loc: N Seattle suburb, WA

Posted 12 July 2014 - 10:00 PM

Have you taken steps to make the OTA absolutely rigid? One of the problems is even the most solid focuser will tilt if the OTA gives way under it. The OTA/focuser connection will feel solid, but not according to the collimation tools.

Have you read the recent threads about collimation problems when installing ASA reducers in low cost imaging newtonians?

Gale

#4 Peter10

Peter10

    Vostok 1

  • -----
  • Posts: 106
  • Joined: 31 Mar 2007

Posted 13 July 2014 - 10:15 AM

Thanks everyone for the suggestions. I agree that decentration (in combination with) could be the possible root cause, as opposed to pure tilt. Not having a quantitative model of the system - it is difficult to establish the dominant contributor.

My suspicion is that decentration (of the coma corrector) is more sensitive than tilt. I was hoping that the close tolerance machining on the Moonlite focuser tube and the ASA corrector would keep decenter and tilt within acceptable bounds. This may not be the case.

From the standpoint of stability, my tube was fabricated from 0.090" thick aluminum tube supplied from Hastings. I am using a high quality commercially made PM mount and a heavy duty spider supplied by Astrosystems. The mechanicals seemingly are capable of holding the optics in any desired alignment configuration.

If I adjust the PM tilt based on star test data I can achieve acceptable results. I was hoping that I could get a "useful" alignment using the CatsEye tools without the need for additional adjustments.

Once aligned using the star test method, I removed the camera and reinstalled the Black Cat Cheshire. As expected, the Cheshire indicated that the scope was severely out of collimation - though it was then producing tight round stars.

Peter

#5 Nils Olof Carlin

Nils Olof Carlin

    Surveyor 1

  • *****
  • Posts: 1579
  • Joined: 26 Jul 2004

Posted 13 July 2014 - 12:29 PM

One thing to consider: the actual centering of the center spot. If it is decentered (reported to happen!), collimation to it will result in coma even with the corrector in place. And it is at least simple to check
BTW you have images without the corrector. Can you identify an area of minimum coma, and if so, is it decentered?

Nils Olof

#6 Peter10

Peter10

    Vostok 1

  • -----
  • Posts: 106
  • Joined: 31 Mar 2007

Posted 13 July 2014 - 01:05 PM

Hi Nils Olof :

I did check the centration of the PM center spot and found it to be quite accurate. I performed the check using both the CatsEye template and via the photographic method. The results from both procedures showed that the spot is properly placed.

I shot the images without the coma corrector a few nights ago but did not save them. As such, I can not say whether the minimum coma area is decentered. I will check this out and report back once the skies clear.

Do you have any thoughts on whether it is better to correct this problem at the primary mirror or the secondary mirror? It is easier and more precise to introduce these fine corrections using the Primary mirror mount - but is this at least as effective as using the Secondary mirror?

Thanks,

Peter

#7 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 22543
  • Joined: 23 Jun 2003
  • Loc: Los Angeles

Posted 13 July 2014 - 07:01 PM

If the area of least coma and the center of the coma corrector do not align, then it is obvious that either mirror's tilt could correct that. But only for one position of the focuser.
If you are collimating the primary to the center point of a tool with one position of the focuser, and taking an image with another position of the focuser, and the two positions are not collimated at the same time, it seems to me that the optical axis is not coincident with the focuser drawtube.
That would result in different collimation revelations at different positions of the focuser.

The first step is to make certain the focuser can move through most of its throw without changing collimation at all. This can be partially seen with a sight tube, though the tolerance you are trying for may be beyond the resolution of a sight tube.
Then follows the alignment of the primary using the cheshire to make sure the primary's optical axis points to the same focuser axis.
But, and this is critical, it's necessary to follow up with an autocollimator to eliminate the residual collimation errors.
Even after the first two tools are used, some errors are left that can be removed.
By going back and forth between the cheshire and the AC, one can pretty much be assured the focuser-secondary-primary line is coincident with the primary-secondary-focuser line to a very great degree.
I do a very careful job of using the first two tools, but I ALWAYS find that further tweaking is called for when the AC is used.

A final check, by moving the drawtube in and out, allows me to find the best average for the focuser motion. Why I say that is the small thousandth of an inch irregularities in the drawtube can be seen in the AC's lateral pupil as i move the focuser up and down. a "perfect" collimation in one place is not "perfect" in another place for the focuser.

Since I know where the focuser is located when my coma corrector is in place, I try to hit the "average" at that point.

When you add weight to the focuser, it sags. Unless your scope is ultra-rigid, which it isn't given your quoted spec, or the focuser is ultra-rigid (which the Moonlite is not--I use one, but it does sag with a lot of weight in the focuser), there will be some miscollimation that occurs when a lot of weight is put in the focuser.

And if your coma corrector decenters a bit when it is in the focuser, and if we assume the lenses are well-centered in that piece, then the focuser centerline is probably no longer corresponding to the optical axis.

Since movement in both mirrors essentially, during collimation, takes the focuser axis as the "given" to which collimation is achieved, having this axis move relative to the optical axis will throw the correction off and move the point of best correction away from the center.

What I suggest is to "lock" the focuser at the position where your camera is focused the next time out. LEAVE it locked in that position, and then re-collimate the scope using all 3 Catseye tools, but especially the XLK autocollimator. Then try another image. If the star images are still not best in the center, then you have too much sag in the focuser or tube to result in good images with the corrector and camera in place.
Moving the primary OR secondary to produce perfect images just proves this, since you are miscollimating the scope to compensate for the sag.

I see nothing wrong with that, however, as an interim step. Beefing everything up until there is no sag is the preferable, long-term, solution.

Or, collimate for visual use and recollimate for photographic each time you switch back and forth. Your choice.

#8 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5271
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 14 July 2014 - 12:57 AM

0.09" isn't strong enough. That's about 2mm.

My steel tube newt with 2mm aluminium epoxied under the focuser still showed a bit of flex with the camera etc attached to the Feathertouch.

I know this isn't what you want to hear but after 8 months of fighting the Keller I sold it and got a Paracorr. For a now glacial f4.6 but the stars are absolutely refractor like.

#9 Peter10

Peter10

    Vostok 1

  • -----
  • Posts: 106
  • Joined: 31 Mar 2007

Posted 14 July 2014 - 08:59 AM

I would like issue a sincere thanks for taking the time to fully address my problem with telescope collimation/stability. The information you provided resonates with me and provides me with a clear path forward. I have been fighting these issues to various degrees for over a year without a systematic plan forward. The ASA corrector/reducer is difficult to setup, but I must say it is impressive when operating at full potential. My goal is to transition directly to imaging after using the CatsEye collimation tools.

On a related topic, can you tell me your thoughts on telescope tube stability. I selected a Hastings aluminum tube with a wall thickness of 0.90”. At the time, it seemed to be the best compromise between cost and stiffness among the commonly available options. Even with this tube, I suspect that there are load induced distortions that compromise collimation. Would I benefit going from my current tube to one fabricated from Carbon fiber? I need a stiff telescope tube but I am trying to avoid creating a 200lb 8” Newtonian.

Peter

#10 Peter10

Peter10

    Vostok 1

  • -----
  • Posts: 106
  • Joined: 31 Mar 2007

Posted 14 July 2014 - 09:10 AM

You are correct, this is not good news. I have managed to get good results with the corrector but it requires extensive work after the baseline calibration with the CatsEye tools. It is good to hear that you are getting quality images now!

Peter

#11 gdd

gdd

    Apollo

  • -----
  • Posts: 1390
  • Joined: 23 Nov 2005
  • Loc: N Seattle suburb, WA

Posted 14 July 2014 - 10:40 AM

Would I benefit going from my current tube to one fabricated from Carbon fiber? I need a stiff telescope tube but I am trying to avoid creating a 200lb 8” Newtonian.



I believe one of the reasons carbon fiber is so stiff is that it is really a double walled tube with a composite filling the space between the walls. The composite can be very soft, so care must be taken not to crush it when tightening down screws.


Gale

#12 famax

famax

    Sputnik

  • -----
  • Posts: 49
  • Joined: 01 Jul 2007

Posted 15 July 2014 - 07:11 AM

Hello peter
i'm using the 0.73 on my FD4.5 newton, taking it down to FD3.3.
Had a lot of fight with it , but it is almost ok now.

Things to do :
-reinforce the focuser
-Make evrything square and centered (focuser/secondary/primary) if needed spot the secondary at optical point to place it correctly under the focuser.
-Collimate with catseyes at prime focus
-Use the "atomic" spot from catseye.
-Calculate the proper back focus of the corrector and lower it a bit (let say -0.5 or 1 mm)

that did the trick for me.

regards

francois

#13 Peter10

Peter10

    Vostok 1

  • -----
  • Posts: 106
  • Joined: 31 Mar 2007

Posted 16 July 2014 - 06:04 PM

Hi francois:

As you suggested, I am planning to reinforce the tube behind the focuser. Can you elaborate a bit on the corrector backfocus recommendation? Are you talking about the suggested 65mm spacing between the corrector's mounting flange and the camera's focal plane?

Peter

#14 famax

famax

    Sputnik

  • -----
  • Posts: 49
  • Joined: 01 Jul 2007

Posted 18 July 2014 - 10:15 AM

Yes i do.
you can even lowering it till 3 mm.
do not forget to add your filter/3 space
example for 65.2 back focus recommanded by Keller, add 1 mm
if you have 3mm thick astrodons filters.

To widden the corrected field you substract 3 mm from the 65+1 mm value.

#15 CatseyeMan

CatseyeMan

    Vendor (Cats Eye Collimation)

  • *****
  • Vendors
  • Posts: 550
  • Joined: 16 Dec 2004
  • Loc: Madison, AL USA

Posted 19 July 2014 - 08:08 AM

.... I align using the Catseye collimation tools and can achieve a picture perfect result as viewed through the tools. My basic collimation is unquestionably good.

You didn't mention if you are using an extension tube when using the CATSEYE tools. The native focal plane for most Newtonian imaging scopes (incorporating a corrector) is well above the focuser drawtube. When collimating (without corrector in place), an extension tube of appropriate length is needed to bring the tools' active elements (Cheshire ring, AC mirror, sight tube pupil) up to the focal plane, else false collimation confirmation in the visual queues can result due to induced parallax error.

#16 gdd

gdd

    Apollo

  • -----
  • Posts: 1390
  • Joined: 23 Nov 2005
  • Loc: N Seattle suburb, WA

Posted 19 July 2014 - 08:54 AM

How close to the focal plane do you need to be?

Gale

#17 Peter10

Peter10

    Vostok 1

  • -----
  • Posts: 106
  • Joined: 31 Mar 2007

Posted 19 July 2014 - 02:37 PM

Hi Jim:

I did use an extension tube for the reason you stated. I selected a tube height that maintained parity in the apparent size of the reflections as viewed through the autocollimator.

Since my last post, I increased the tension in the spider vane supports for my secondary mirror. This improved but did not eliminate my oval stars. I seem to incrementally improve star symmetry without ever getting to where I need to be. At times, I think it is impossible to perform quality imaging with a fast newtonian.

PP

#18 famax

famax

    Sputnik

  • -----
  • Posts: 49
  • Joined: 01 Jul 2007

Posted 21 July 2014 - 04:53 AM

Peter , if you use a kaf 8300 ccd, having bad stars shapes
at the corner is somehow expected , as the corrected field
is 20mm.
As said before, try to lower a bit the back focus...
try with 3 mm less.

#19 Peter10

Peter10

    Vostok 1

  • -----
  • Posts: 106
  • Joined: 31 Mar 2007

Posted 21 July 2014 - 05:59 PM

Hi famax:

I will give it a try my next night out.

PP

#20 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5271
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 22 July 2014 - 06:38 AM

Jim is correct. You need about 2" of extension to get close to the focal plane with the cats eye with the corrector not in place. This introduces another point of flexure - the interface between the focuser and the extension tube. Plus the existing point between the extension tube and the cats eye.

This is hard to solve. Suggest use only one set screw on the focuser and tube. And rotate the compression ring so that the used set screw pokes through the gap in the compression ring and bears directly on the cats eye tube or extension tube.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics