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

New saddle place to stop aluminum tube flexing.

  • Please log in to reply
15 replies to this topic

#1 photonhunter

photonhunter

    Vostok 1

  • *****
  • topic starter
  • Posts: 196
  • Joined: 24 Feb 2007

Posted 16 June 2019 - 10:05 PM

I've posted here before about the "Behemoth" my 17.5" equatorially mounted Newtonian.  It started life as a Celestron Starhopper Dob many many years ago and has evolved into an aluminum tube equatorial scope.

I've always tried to collimate this beast to the limits of accuracy using Catseye collimation tools including a custom XLKP autocollimator with COC masks attached.   I've learned how to get things dialed in to near perfection by following some of the threads here discussing all the sources of collimation error and how to lock them down.

 

The problem has always been that when the scope slews across the sky the collimation changes.  I've gone through several recommendations and could never quite identify the source of the movement as each thing I would try changed the behavior of the scope but never eliminated the collimation changing.

 

I finally decided (after consulting several folks more knowledgeable than I) that my aluminum tube was probably flexing just enough to cause the issue.   Thick aluminum plate is expensive but given what I have in the scope and mount it's worth it to improve the behavior of the scope.

 

I test the flexing by rotating the scope 180 degrees from one side of the mount horizontal to the other side, horizontal again.  This is an extreme test for a Newtonian scope but working to this level gives me the best chance of maintain collimation while pointed at different places in the sky.

 

So I ordered a 48" x 8" x 1" saddle plate and a 48" x 8" x 1/2" top plate.

 

Original scope configuration

IMG_1205_small.jpg

 

Scope loaded onto new saddle

20190616_130508.jpg

 

While it has not eliminated all the movement, it has eliminated the tube flexing element which was making it impossible to lock in on any source of movement as no matter what I changed, there was no improvement.  Now when I tweak the spider, focuser, and mirror cell I can both identify which component is contributing and selectively eliminate it.

 

The worst culprit other than the tube flexing turned out to be my spider -- which I was suspicious of all along.  I tried tightening the spider (a Protostar spider made of a combination of aluminum and some sort of plastic/polymer material).  Unfortunately the spider is simply not sturdy enough to totally lock down.  I did get it's movement down to a pretty reasonable level using of all things, gorilla tape!.  Now there is only a couple of millimeters of movement moving all the way across from one side to the other -- about the diameter of the collimation laser beam I have.

 

West side of mount

20190616_210330.jpg

 

East side of mount

20190616_210307.jpg

 

Fortunately there are several sources of heavy duty spiders that are not terribly expensive so that is next on my list.  I think that this much movement across 180 degrees is not too bad though I will keep trying to eliminate it.

 

I'm definitely reading thread about testing spider flexing.

 

Mitch


Edited by photonhunter, 16 June 2019 - 10:08 PM.


#2 TOMDEY

TOMDEY

    Aurora

  • *****
  • Posts: 4964
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 17 June 2019 - 03:15 AM

Hi, Mitch; very good!

 

I addressed that type of angular drooping by adding/inventing this thing I am calling the ~torsion Compensator~ weight / Angular Serrurier Truss >>>

 

If the effect is gravity twisting of the spider, that weight comps it out, in all orientations. Even better, it can also comp each and every such in the system, provided that the contributors all go as the cosine of Alt. Think of it as an angular (vs lateral translaion) Serrurier Truss. In that sense, it's the exact compliment of the Serrurier Truss. The beauty in this approach is that you don't need to stiffen the structure, because it is auto-compensating for all alt angles (0o-90o) and all g-clockings of the tube (0o-360o). So, all you add is a stem and a weight... with the amount and position empirically determined. Costs nearly nothing and can dial it in... in a matter of minutes; Set it, and forget it!

 

I assume other people have been doing this, before I invented it... around 40 years ago?  Works like a charm and ~collimation~ is invariant all over the sky.    Tom

Attached Thumbnails

  • 67 29 INCH Toms SM torsion compensator 78.jpg
  • 69 12.5 inch scope Tom torsion compensator.jpg

  • Eddgie, photonhunter and slavicek like this

#3 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 24898
  • Joined: 01 Feb 2006

Posted 17 June 2019 - 04:32 PM

Outstanding innovation, Tom.    Salute. 



#4 photonhunter

photonhunter

    Vostok 1

  • *****
  • topic starter
  • Posts: 196
  • Joined: 24 Feb 2007

Posted 17 June 2019 - 05:56 PM

Hi, Mitch; very good!

 

I addressed that type of angular drooping by adding/inventing this thing I am calling the ~torsion Compensator~ weight / Angular Serrurier Truss >>>

 

If the effect is gravity twisting of the spider, that weight comps it out, in all orientations. Even better, it can also comp each and every such in the system, provided that the contributors all go as the cosine of Alt. Think of it as an angular (vs lateral translaion) Serrurier Truss. In that sense, it's the exact compliment of the Serrurier Truss. The beauty in this approach is that you don't need to stiffen the structure, because it is auto-compensating for all alt angles (0o-90o) and all g-clockings of the tube (0o-360o). So, all you add is a stem and a weight... with the amount and position empirically determined. Costs nearly nothing and can dial it in... in a matter of minutes; Set it, and forget it!

 

I assume other people have been doing this, before I invented it... around 40 years ago?  Works like a charm and ~collimation~ is invariant all over the sky.    Tom

Awesome Tom!  Thank you very much -- where do I send the royalty check once I add this to my scope?  grin.gif grin.gif



#5 TOMDEY

TOMDEY

    Aurora

  • *****
  • Posts: 4964
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 17 June 2019 - 08:53 PM

Awesome Tom!  Thank you very much -- where do I send the royalty check once I add this to my scope?  grin.gif grin.gif

Heh... I got a boatload of patents... but that ain't one of them!

 

Actually, opto-mechaincal compensators are all over the place, almost entirely in ~professional systems~ I worked aerospace optics most of my life, and our Structures Group, ~Mirrors and Mounts~ handled all that kinda stuff. Passive (like that little weight on a stem) are the nicest... but active compensation is a big thing, too.

 

As you might imagine... when I see poor practices in our commercial hobby stuff - it drives me nuts! When I get a commercial scope and turn the alignment knobs/screws and they don't work properly... Arrgh!

 

Anyway, that sliding weight on a stem is nice because one can just try it and see if there is a setting that works. If not, then something else is going on. No harm done and not even any invasive re-build of the system. On a typical Newt, just a threaded coupling can screw it right on there. It doesn't block any light or get in the way.    Tom


  • photonhunter likes this

#6 photonhunter

photonhunter

    Vostok 1

  • *****
  • topic starter
  • Posts: 196
  • Joined: 24 Feb 2007

Posted 17 June 2019 - 09:08 PM

Heh... I got a boatload of patents... but that ain't one of them!

 

Actually, opto-mechaincal compensators are all over the place, almost entirely in ~professional systems~ I worked aerospace optics most of my life, and our Structures Group, ~Mirrors and Mounts~ handled all that kinda stuff. Passive (like that little weight on a stem) are the nicest... but active compensation is a big thing, too.

 

As you might imagine... when I see poor practices in our commercial hobby stuff - it drives me nuts! When I get a commercial scope and turn the alignment knobs/screws and they don't work properly... Arrgh!

 

Anyway, that sliding weight on a stem is nice because one can just try it and see if there is a setting that works. If not, then something else is going on. No harm done and not even any invasive re-build of the system. On a typical Newt, just a threaded coupling can screw it right on there. It doesn't block any light or get in the way.    Tom

Well my spider made some noise the other night that sounded like a screw popping over a thread.  I've never been really happy with the material of the center bolt and the rear plate on the secondary mirror has holes worn in it by the three collimation screws I've had them so tight with the secondary still moving a bit.  I even put metal nuts under the secondary to hold it rigid in case the foam backing was compressing, no dice..  So today I ordered an astrosystems heavy duty spider (model #4), their next to heaviest.  They have about 4" of threaded rod extending past the spider mount so I can put a weight on it if necessary.



#7 TOMDEY

TOMDEY

    Aurora

  • *****
  • Posts: 4964
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 17 June 2019 - 09:18 PM

Well my spider made some noise the other night that sounded like a screw popping over a thread.  I've never been really happy with the material of the center bolt and the rear plate on the secondary mirror has holes worn in it by the three collimation screws I've had them so tight with the secondary still moving a bit.  I even put metal nuts under the secondary to hold it rigid in case the foam backing was compressing, no dice..  So today I ordered an astrosystems heavy duty spider (model #4), their next to heaviest.  They have about 4" of threaded rod extending past the spider mount so I can put a weight on it if necessary.

Good! And, if still needed... I think there's a good chance that the weight on a stem would fix it.    Tom



#8 slavicek

slavicek

    Mariner 2

  • *****
  • Posts: 252
  • Joined: 01 Aug 2017
  • Loc: Massachusetts

Posted 17 June 2019 - 09:23 PM

This is sooo simple and so genius! If I understand it correctly all one needs is a threaded coupling to attach the threaded rod on the secondary mirror holder and piece of threaded "dead weight" which screws on the threaded rod.

I have 5.5" dia secondary. Shall I start with like +/- 2lb weight?



#9 photonhunter

photonhunter

    Vostok 1

  • *****
  • topic starter
  • Posts: 196
  • Joined: 24 Feb 2007

Posted 17 June 2019 - 09:29 PM

Good! And, if still needed... I think there's a good chance that the weight on a stem would fix it.    Tom

I showed them your picture and they said there is enough threaded center bolt left to add a weight (about 4").  We discussed making it longer but decided not to have it extend beyond the front of the tube.



#10 photonhunter

photonhunter

    Vostok 1

  • *****
  • topic starter
  • Posts: 196
  • Joined: 24 Feb 2007

Posted 17 June 2019 - 09:48 PM

Well I found a disc weight that was once a counterweight for my C-8 and it fits the protostar stalk perfectly.  It does seemed to have helped things a bit.

 

20190617_213922.jpg

 

East side:

20190617_214011.jpg

 

West side:

20190617_214040.jpg


Edited by photonhunter, 17 June 2019 - 09:52 PM.


#11 TOMDEY

TOMDEY

    Aurora

  • *****
  • Posts: 4964
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 17 June 2019 - 10:42 PM

This is sooo simple and so genius! If I understand it correctly all one needs is a threaded coupling to attach the threaded rod on the secondary mirror holder and piece of threaded "dead weight" which screws on the threaded rod.

I have 5.5" dia secondary. Shall I start with like +/- 2lb weight?

I showed them your picture and they said there is enough threaded center bolt left to add a weight (about 4").  We discussed making it longer but decided not to have it extend beyond the front of the tube.

Well I found a disc weight that was once a counterweight for my C-8 and it fits the protostar stalk perfectly.  It does seemed to have helped things a bit.

If you move that out with an extender it might help a lot! (see below) >>>

 

Hi, guys... If the effect is coming only from the secondary, then nominal comp weight would be to balance that at the center of the spider hub, so that the torques compensate out. So, that could be a long stem with little weight or closer in with more weight. On the other hand, if there are other 1/cos(Alt) effects effectives within the opto-mechanical train, the fully-comped weight setting could be different. Fortunately, just fiddling with the weight and location should find where it works best... hopefully providing complete compensation.

 

The full-blown reactions to gravity get more complicated than that... but this single gizmo usually can comp out the lion's share, especially on alt-az geometries.   Tom



#12 photonhunter

photonhunter

    Vostok 1

  • *****
  • topic starter
  • Posts: 196
  • Joined: 24 Feb 2007

Posted 18 June 2019 - 12:05 AM

If you move that out with an extender it might help a lot! (see below) >>>

I do have the rod used to hold that weight on the C-8 and its about 4" long.  But I need to figure out how to find a coupler that would connect it to the protostar bolt.   



#13 photonhunter

photonhunter

    Vostok 1

  • *****
  • topic starter
  • Posts: 196
  • Joined: 24 Feb 2007

Posted 18 June 2019 - 12:27 AM

Well necessity is the mother of invention I guess -- I realized the part that fitted to the rail for the counter weight was threaded all the way through so it works perfectly as a coupler.

 

20190618_002028.jpg

 

It may have helped some but didn't stop the last bit of movement, so I'm thinking there may be one more source of movement somewhere.  I did wedge paper beneath the flotation points on the main mirror to insure it couldn't move.  

 

Hmm...  I'm either circling the field or circling the drain...  finding subtle movement is a PITA LOL.



#14 TOMDEY

TOMDEY

    Aurora

  • *****
  • Posts: 4964
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 18 June 2019 - 08:44 AM

Well necessity is the mother of invention I guess -- I realized the part that fitted to the rail for the counter weight was threaded all the way through so it works perfectly as a coupler.

 

attachicon.gif 20190618_002028.jpg

 

It may have helped some but didn't stop the last bit of movement, so I'm thinking there may be one more source of movement somewhere.  I did wedge paper beneath the flotation points on the main mirror to insure it couldn't move.  

 

Hmm...  I'm either circling the field or circling the drain...  finding subtle movement is a PITA LOL.

Hardware stores have couplers that look like very long hex nuts.

 

On the much larger/generalized topic:

 

Telescope mounts have myriad potential sources of flexure, and they come in all flavours and manifest in any/all degrees of freedom. That is, each and every component that needs to be aligned and to maintain that alignment, can potentially drift as the g-loads (gravity vector) moves around, temp changes... and all six degrees of freedom are vulnerable, on each component (PM, SM, focuser, etc.).

 

Add to that, how the optics react to those movements and how they optically-manifest, in the final image and its location relative to the eyep/cam/instrument suite... even that gets confounding. Those functional reactions can range from disastrous to annoying to entirely transparent and benign.

 

I had analyzed all that, when I was at Kodak, awaiting my background investigation to get on ~Government Projects~ That turned into reams of math... that we eventually (successfully!) used on satellite builds.

 

BUT: The general rule is to make your entire structure deterministic, stiff, rigid, auto-compensating... to begin with. My Cave Astrola fiberglass tube was hopelessly flexible/mushy. So, I replaced it with an extremely stiff Aluminum Tube... and that vastly improved things.

 

Here's p164 (and counting) of that analysis that I did back then. I was able to show that one could take some measurements of any complete optical train using a (remote/non-obtrusive) instrument called an ~Axicon Autocollimator~  and then back-solve for all component misalignments, in all degrees of freedom (via a matrix).    Tom

Attached Thumbnails

  • 76 60 page 164 of Toms Generalized Axicon Analysis.jpg

  • photonhunter and howardcano like this

#15 photonhunter

photonhunter

    Vostok 1

  • *****
  • topic starter
  • Posts: 196
  • Joined: 24 Feb 2007

Posted 19 June 2019 - 01:26 AM

That sounds like what I need to learn, how to analyze and isolate each movement so I can understand exactly what I need to fix. I suspect that your math may be beyond me though!

Sent from my SAMSUNG-SM-P907A using Tapatalk

#16 TOMDEY

TOMDEY

    Aurora

  • *****
  • Posts: 4964
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 19 June 2019 - 03:27 AM

That sounds like what I need to learn, how to analyze and isolate each movement so I can understand exactly what I need to fix. I suspect that your math may be beyond me though!
Sent from my SAMSUNG-SM-P907A using Tapatalk

Heh! Actually, a Newtonian is a nice simple system; on the other hand... very few people have Axicon Autocollimators! We built our own at Kodak... small 2-inchers, medium 4-inchers and big 6-inchers. They were clunky and heavy... but sure fun and useful!

 

McLeod invented them for Kodak.

 

When I arrived, no one had done the analysis, except for one little "Axicon Equation" that appeared in the patent. I thought that was kinda nuts, and had nothing to do while awaiting my "Tickets"... so I decided to do the generalized applications analysis, and nobody cared.

 

It's luxurious when you get hired, awaiting access... and the boss just gives you a desk and says... "Here, just do something constructive that's optical... and we pay you for the duration. Kodak, back in the ~fat days~ was what the financial guys call "Cash Rich"... which meant they were making so much profit that management's biggest problem was deciding what to do with the excess. Ummm... that obviously changed, years later. Somehow, I survived all that turmoil!    Tom

Attached Thumbnails

  • 81 McLeod patent axicon kodak.jpg

  • photonhunter 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