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

Collimation shift on a Dob

  • This topic is locked This topic is locked
31 replies to this topic

#1 GShaffer

GShaffer

    Knight of Ni

  • *****
  • Moderators
  • topic starter
  • Posts: 6397
  • Joined: 28 Feb 2009
  • Loc: Bogart, Ga USA

Posted 28 December 2018 - 11:33 AM

Would appreciate some input here on potential culprits.

 

Lets say you collimate a truss dob with the scope pointing near vertical and the laser is dead on center of the mirror.

 

Insert a good laser in the focuser and as you lower the pointing angle you see the laser spot on the primary move DOWN below the center spot.

 

Which is the more likely culprit? Truss flex or spider/holder issue?

 

I lean more towards it being a spider/holder issue than truss but also realize that is not a given.....

 

The particular scope in question is a 16" f/4.5 with 1" trusses and an astrosystems type 2 spider/holder with a 3.1" secondary. Both in my mind would seem sufficient for a 16" scope.


 

#2 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 42400
  • Joined: 23 Jun 2003
  • Loc: Los Angeles

Posted 28 December 2018 - 12:06 PM

The 1" truss tubes are a bit too light (I use 1.25" poles on a 12.5" f/5 because I saw a difference), but that's not the cause of the problem.

When the scope points up, the secondary hangs below the spider and imparts no twist to the vanes.

When the scope points low, the secondary hangs beside the spider and imparts a twist to it.

A 3.1" secondary is not that light, if you count the weight of the holder.

 

The cure is either easy or complicated.

 

The easy cure is to tighten the spider vanes until they are very tight.  The tighter they are, the less laser movement you'll see.

 

The harder cures come in many forms:

1) replace the spider vanes with the heavier ones that have 2 bolts per vane.

2) move spider vanes down in scope so secondary hangs much closer to the spider vanes.

3) Shorten the secondary holder so there is no space above the upper end of the secondary in the holder.

4) replace spider with a new design that looks like this: >o< You still get 4 spikes but the stiffness is much higher.

5) add a small counterweight to the center bolt above the spider that balances the weight of the secondary + holder.

6) Make sure the spider vanes, from the front of the scope, appear like a "+" or a "X" and are not joined at an angle.  When the spider is joined at an angle, unless everything is heavy and stiff,

you will get collimation movement with altitude change.  Rob Teeter experimented with this and found this to be true.  Additionally, the collimation change has a lateral component to it, not just vertical.

 

All of this presumes you have:

--tightened the nut on the center bolt until it is very tight.

--filled up the space in the spider the center bolt passes through.  I wrapped the bolt threads with plumber's tape.  Others have used toothpicks between the center bolt and the spider.

But there should be no looseness whatsoever of the center bolt as it passes through the spider.

--made sure there is enough stuffing behind the mirror to press the mirror against the lip of the holder, even when the mirror is on its side rather than when gravity pulls it down against the lip at the zenith.

There should be no looseness here or collimation will shift as the scope points lower.

 

If the trusses bend, the UTA will drop.  The likelihood is that the secondary will rise, causing the laser beam to go UP on the primary.

Since it goes down, I vote for secondary sag.

 

In the future, since you have 1" poles, I'd vote for collimating at a 60° angle so the error at low angles is reduced.


Edited by Starman1, 28 December 2018 - 12:09 PM.

 

#3 SeaBee1

SeaBee1

    Soyuz

  • *****
  • Posts: 3808
  • Joined: 19 Mar 2015
  • Loc: Under the DFW light barrier

Posted 28 December 2018 - 12:08 PM

Hi Greg! From my experience, the most flexure problems seem to always originate in the secondary assembly. The truss structure of your scope, as long as you have true trusses, normally hold up to the different altitude angles. Most secondary assemblies are designed with flex just waiting to happen. That 3.1" mirror is stuck out on a stem suspended by 4 vanes that are normally not designed to eliminate flexing without being banjo string tight.

 

I built my 10 inch with eliminating as much flex in the secondary as possible in mind. I used an offset spider design that ensures rock steady collimation.

 

Keep in mind that even in the best designed and built reflectors, all materials have some flexure. Materials move, fact of life.

 

Barring any re-design, or retro-fit, try collimating at about 40 to 50 degrees off level and see if it improves. I say this, because so much of the time this is the elevations I observe at.

 

I do hope this helps!

 

CB

 

Edit to add: I see that Starman1 answered about the same time I was typing. He gives a much better answer!


Edited by SeaBee1, 28 December 2018 - 12:09 PM.

 

#4 GShaffer

GShaffer

    Knight of Ni

  • *****
  • Moderators
  • topic starter
  • Posts: 6397
  • Joined: 28 Feb 2009
  • Loc: Bogart, Ga USA

Posted 28 December 2018 - 12:17 PM

Excellent replies both of you......in fact I passed yours on Don as a virtual checklist to the fellow with the issue.....


 

#5 Starman47

Starman47

    Viking 1

  • -----
  • Posts: 637
  • Joined: 10 Jun 2018
  • Loc: Middle Tennessee

Posted 28 December 2018 - 03:23 PM

Problem 95% (plus) solved. I took the secondary off the spider and wrapped the bolt with several layers of plumbers tape. And Bob's your uncle. (Australian lingo for Voila, just in case you did not know)

 

I did notice that the image on the TuBlug moved while the telescope moved up and down. But in less than a second the image of the reinforcement ring returned to its original position plus or minus a few percent. But it is now so close to perfect that I would not have said anything about the situation. 

 

Thank you all for your help in this matter. bow.gif 

 

I suspect that I will have to get the telescope out to a dark sky location very soon. My next project is the Herschel 400 and then . . . snoopy2.gif


 

#6 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 42400
  • Joined: 23 Jun 2003
  • Loc: Los Angeles

Posted 28 December 2018 - 03:45 PM

Excellent!


 

#7 a__l

a__l

    Apollo

  • -----
  • Posts: 1063
  • Joined: 24 Nov 2007

Posted 28 December 2018 - 10:13 PM

There was already an idea.
Load the Secondary Cage with more or less weight and see the response. Weight closer to the bottom ring.
If yes, see tubes and clamps.
No - see secondary and holder-spider.

 

I have AstroSystems type 1 spider (the weakest spider) and 4" secondary.

No movement.
I have  1"  Moonlight tubes-clamps on the 14.5" and 18" telescopes.
No movement.


Edited by a__l, 29 December 2018 - 12:06 AM.

 

#8 a__l

a__l

    Apollo

  • -----
  • Posts: 1063
  • Joined: 24 Nov 2007

Posted 28 December 2018 - 10:48 PM

Problem 95% (plus) solved. I took the secondary off the spider and wrapped the bolt with several layers of plumbers tape. And Bob's your uncle. (Australian lingo for Voila, just in case you did not know)

 

 

Curious. Typically, a locking metal nut (which is on the bolt) clamp is better than any plastic tape.

This clamp holder to the spider.

 

If this is a poorly held collimation plate (which is on the bolt), then this is a consequence the collimation screws (or one of them) are not pressed tightly.

 

There is no need for plastic tape.


Edited by a__l, 28 December 2018 - 11:26 PM.

 

#9 a__l

a__l

    Apollo

  • -----
  • Posts: 1063
  • Joined: 24 Nov 2007

Posted 28 December 2018 - 11:34 PM

There is another option. Burst (crumbled) material (holder), which rests against the main  bolt head. This happens when the screws is tightened too much.
This is an unknown me lightweight material that Astrosystems uses. This is not metal.

It would be better to use a ball there. And the reciprocal hole in the material. But Astrosystems uses the usual bolt plus plastic washer.

But still this is unlikely.

 

I have 4 AstroSystems holders of different sizes, I studied them well smile.gif

Attached Thumbnails

  • AstroSystems (3)_z.jpg

Edited by a__l, 29 December 2018 - 01:55 AM.

 

#10 Mike Lockwood

Mike Lockwood

    Vendor, Lockwood Custom Optics

  • *****
  • Vendors
  • Posts: 1574
  • Joined: 01 Oct 2007
  • Loc: Usually in my optical shop

Posted 29 December 2018 - 12:36 AM

Greg, how much does the spot move?

 

Focuser or focuser board flex often adds a sideways component to the movement, doesn't sound like this is happening to you.  Secondary mirror or mirror cell flex is typically vertical.


 

#11 GShaffer

GShaffer

    Knight of Ni

  • *****
  • Moderators
  • topic starter
  • Posts: 6397
  • Joined: 28 Feb 2009
  • Loc: Bogart, Ga USA

Posted 29 December 2018 - 01:11 AM

Greg, how much does the spot move?

 

Focuser or focuser board flex often adds a sideways component to the movement, doesn't sound like this is happening to you.  Secondary mirror or mirror cell flex is typically vertical.

 

See the reply from Starman47 above, it is his scope.....he already per his statement pretty much cured the issue with one of Starman1's suggestions.....


 

#12 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 42400
  • Joined: 23 Jun 2003
  • Loc: Los Angeles

Posted 29 December 2018 - 01:22 AM

Curious. Typically, a locking metal nut (which is on the bolt) clamp is better than any plastic tape.

This clamp holder to the spider.

 

If this is a poorly held collimation plate (which is on the bolt), then this is a consequence the collimation screws (or one of them) are not pressed tightly.

 

There is no need for plastic tape.

The nut on my center bolt was tight as a wrench could get it, but I got some secondary movement after everything was as tight as possible.

I wrapped the centerbolt with tape and the movement disappeared.

It seems a couple thousandths of an inch of movement of the bolt in the spider was sufficient to cause the secondary collimation to change as seen in an autocollimator.

FWIW, the bolt had close to a half millimeter of back and forth movement in the hole in the spider before I wrapped it with tape.


 

#13 a__l

a__l

    Apollo

  • -----
  • Posts: 1063
  • Joined: 24 Nov 2007

Posted 29 December 2018 - 01:44 AM

Plastic tape with a bolt to push in a spider, you need space. Otherwise it will get stuck. This may be the part of the inch about which you write. The tape can wrinkle further.

 

Two nuts (photo blue) clamp perfectly! As an option, replace the plastic washers with metal ones. You can add a spring washer (metal) for add fixation.

4 collimation bolts must tightly press the collimation plate (red in the photo)

 

The second photo, the pad under the nut, is quite powerful!

The third photo , as one of the options for additional fixation (my spider type 1). Adding curly washer. I try to avoid the plastics (washers, tape), where you need a reliable metal connection.

Attached Thumbnails

  • Holder_zz.jpg
  • паук_Izz.jpg
  • Паук_2z.jpg

Edited by a__l, 29 December 2018 - 06:08 AM.

 

#14 a__l

a__l

    Apollo

  • -----
  • Posts: 1063
  • Joined: 24 Nov 2007

Posted 29 December 2018 - 03:21 AM

Focuser or focuser board flex often adds a sideways component to the movement

SIPS adds this as it increases torque smile.gif

If SIPS not, then the laser is lightweight compared Ethos-21 or Bino. For flexure focuser board.


Edited by a__l, 29 December 2018 - 07:51 AM.

 

#15 Starman47

Starman47

    Viking 1

  • -----
  • Posts: 637
  • Joined: 10 Jun 2018
  • Loc: Middle Tennessee

Posted 29 December 2018 - 06:17 AM

I have read all of the posts, and I will keep them all in mind as I use the telescope. Later, I hope to build another telescope, and all your wisdom will help.

 

P.S. I tried loading and unoading weight from the UTA. The improvement was minimal, at best. I suppose I could go up to 1.25 inch tubes, but first  I need to test the test the scope under clear skies. And then I can decide if I want to spend $200 to $300, which could go towards a good eyepiece.


Edited by Starman47, 30 December 2018 - 05:45 AM.

 

#16 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 42400
  • Joined: 23 Jun 2003
  • Loc: Los Angeles

Posted 29 December 2018 - 10:18 AM

Plastic tape with a bolt to push in a spider, you need space. Otherwise it will get stuck. This may be the part of the inch about which you write. The tape can wrinkle further.

 

Two nuts (photo blue) clamp perfectly! As an option, replace the plastic washers with metal ones. You can add a spring washer (metal) for add fixation.

4 collimation bolts must tightly press the collimation plate (red in the photo)

 

The second photo, the pad under the nut, is quite powerful!

The third photo , as one of the options for additional fixation (my spider type 1). Adding curly washer. I try to avoid the plastics (washers, tape), where you need a reliable metal connection.

You'll notice that the hole in the spider is square and the bolt is round.

And if, when the nut is loosened, the bolt can be wiggled in the hole, then there is a possible cause of angle error.

 

One experimental fellow here on CN decided to remove the looseness of the bolt by filling the corners with toothpick fragments.

For him, that worked well.

 

In my case, the round bolt was smaller than the square opening, so a pass over the threads with teflon tape made the fit snug.

The bolt still slid easily through the hole, but, with the tape, without looseness.

 

I have steel washers at both ends and also nuts and the spider-to-bolt connection is tight and secure.

But it was always secure--the only thing I modified was to add a space-filler to the bolt to more completely fill the hole, and this reduced the visible movement in the lateral pupil

of a Catseye autocollimator to zero where, before, there was slight movement there.

 

Which brings up one last point, and that is the diameter of the center bolt.  How much deflection of the bolt can occur before it is visible in the AC?

The 4 collimation screws should prevent the secondary holder from sagging on the bolt, but some deflection occurs.

 

I notice the spider is easy to misalign with very small side pressure from my finger and this produces a movement of the laser on the primary.

The real question here is how to make the entire mechanism so rigid no sag can occur on the scale of <0.03mm?  

Is it possible?  I think it may be, but I think we are still at a level on dobsonian telescopes where a bit of applied engineering can improve rigidity significantly

without adding weight.


 

#17 Starman47

Starman47

    Viking 1

  • -----
  • Posts: 637
  • Joined: 10 Jun 2018
  • Loc: Middle Tennessee

Posted 29 December 2018 - 11:42 AM

I am not an engineer by any means. Just a stubborn tinkerer.  Now I am just thinking here, if we are trying to fit a round peg into a square hole and get a tight fit, then we have a bit of a problem. But it seems there are a few ways to approach this problem. Randy (at Astrosystems) offers a bushing that goes tightly into his smaller spiders (see pic below). This bushing fits very snugly in my smaller spider.

 

Alternatively, we can change the hole or the peg.

 

2) If we make the peg square, then we can just make the upper and lower ends of the peg to have threading. Add nuts and washers as needed and there you go.  

 

3) make the hole round with its own treading. And the peg (bolt) can have threading down the complete length of the bolt.

 

If there is a machinist out there who has some time on there hands, then it seems that there is a possibility of someone coming along and making a better spider.

 

 

 

 


 

#18 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 42400
  • Joined: 23 Jun 2003
  • Loc: Los Angeles

Posted 29 December 2018 - 01:18 PM

A square bolt for the secondary holder, or a threaded hole, would not allow easy rotation of the secondary on its center bolt during preliminary mechanical alignment.

Ideally, the hole in the spider would simply be a round cylinder of close to exactly the same I.D. as the bolt O.D.

There are more rigid ways of fabrication a spider than we typically see; that reduce the necessity of high tension on the blades (vanes); that reduce the likelihood of rotation when the collimation

screws are tightened.  Many of the scopes in the ATM forum have shown different ways of doing this.  So far, I have not seen commercial versions of these stiffer spiders.


 

#19 Starman47

Starman47

    Viking 1

  • -----
  • Posts: 637
  • Joined: 10 Jun 2018
  • Loc: Middle Tennessee

Posted 29 December 2018 - 02:00 PM

Don would you please point to a forum thread where there are better spiders than those available commercially? 


 

#20 airbleeder

airbleeder

    Apollo

  • *****
  • Posts: 1452
  • Joined: 04 Jan 2014
  • Loc: Powder Springs, Georgia

Posted 29 December 2018 - 02:29 PM

You'll notice that the hole in the spider is square and the bolt is round.

And if, when the nut is loosened, the bolt can be wiggled in the hole, then there is a possible cause of angle error.

 

One experimental fellow here on CN decided to remove the looseness of the bolt by filling the corners with toothpick fragments.

For him, that worked well.

 

In my case, the round bolt was smaller than the square opening, so a pass over the threads with teflon tape made the fit snug.

The bolt still slid easily through the hole, but, with the tape, without looseness.

 

I have steel washers at both ends and also nuts and the spider-to-bolt connection is tight and secure.

But it was always secure--the only thing I modified was to add a space-filler to the bolt to more completely fill the hole, and this reduced the visible movement in the lateral pupil

of a Catseye autocollimator to zero where, before, there was slight movement there.

 

Which brings up one last point, and that is the diameter of the center bolt.  How much deflection of the bolt can occur before it is visible in the AC?

The 4 collimation screws should prevent the secondary holder from sagging on the bolt, but some deflection occurs.

 

I notice the spider is easy to misalign with very small side pressure from my finger and this produces a movement of the laser on the primary.

The real question here is how to make the entire mechanism so rigid no sag can occur on the scale of <0.03mm?  

Is it possible?  I think it may be, but I think we are still at a level on dobsonian telescopes where a bit of applied engineering can improve rigidity significantly

without adding weight.

   Wrapping the center bolt with plumbers tape works well for taking up slack in an oversized round hole also, whether the bolt is secured with one or two nuts.


 

#21 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 42400
  • Joined: 23 Jun 2003
  • Loc: Los Angeles

Posted 29 December 2018 - 02:43 PM

Don would you please point to a forum thread where there are better spiders than those available commercially? 

Example:

https://www.cloudyni...ders/?p=8228017

https://www.cloudyni...es#entry6657269

https://www.cloudyni...es#entry6565061


 

#22 Kunama

Kunama

    Aussie at large

  • *****
  • Posts: 4684
  • Joined: 22 Oct 2012
  • Loc: Australia

Posted 29 December 2018 - 04:48 PM

SIPS adds this as it increases torque smile.gif

If SIPS not, then the laser is lightweight compared Ethos-21 or Bino. For flexure focuser board.

Can you explain what you mean by this? 

Surely there is no point in collimating with a lightweight laser if you typically use heavy eyepieces/binoviewers if you suspect any flexure in the focuser board, as by doing so you might get the laser spot to stay in place but suffer collimation drift as soon as you install your binoviewers.

 

If you want to test the source of flex I suggest that you first collimate with the laser only.  Then, once collimated, install your binoviewer and put your laser into one side of your binoviewer, turn it on and watch the laser spot while moving the scope....to see if there is a drift that wasn't there before.


 

#23 Oberon

Oberon

    Mercury-Atlas

  • -----
  • Posts: 2906
  • Joined: 24 Feb 2013
  • Loc: Hunter Valley NSW Australia

Posted 29 December 2018 - 04:59 PM

Also this thread here ended up with a lot of discussion about building spiders.


 

#24 Mike Lockwood

Mike Lockwood

    Vendor, Lockwood Custom Optics

  • *****
  • Vendors
  • Posts: 1574
  • Joined: 01 Oct 2007
  • Loc: Usually in my optical shop

Posted 29 December 2018 - 10:24 PM

See the reply from Starman47 above, it is his scope.....he already per his statement pretty much cured the issue with one of Starman1's suggestions.....

Oops, I missed it.  Very interesting fix.
 

SIPS adds this as it increases torque smile.gif
If SIPS not, then the laser is lightweight compared Ethos-21 or Bino. For flexure focuser board.

Can you explain what you mean by this? 
Surely there is no point in collimating with a lightweight laser if you typically use heavy eyepieces/binoviewers if you suspect any flexure in the focuser board, as by doing so you might get the laser spot to stay in place but suffer collimation drift as soon as you install your binoviewers.

I can explain what he means - he bashes the SIPS at every possible opportunity, even when it's off-topic, and this time quoting my comment about focuser board flex just to contradict/irritate/involve me.  In reality, in a quality instrument, the focuser board should be built stiff enough to not flex significantly.

 

With a heavy eyepiece and Paracorr 2 hanging out of a focuser, the focuser itself can, and will, flex.  This is greatly reduced in the SIPS because only the eyepiece is in the focuser.

 

I'll just leave it there or else this will get off-topic.
 


 

#25 Retsub

Retsub

    Ranger 4

  • *****
  • Posts: 316
  • Joined: 08 Dec 2006
  • Loc: Houston,Tx.

Posted 30 December 2018 - 01:10 AM

Interesting answers ! You say "Insert a good laser in the focuser and as you lower the pointing angle you see the laser spot on the primary move DOWN below the center spot." I have done it this way before. As soon as you see the laser move StoP, and just by flexing different areas a little by hand I think you'll see what it takes to move the laser back to where it should be to tell you where the problem is. Please let us know. *BW*
 


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