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

An offset wire spider design

  • Please log in to reply
115 replies to this topic

#76 Oberon

Oberon

    Soyuz

  • -----
  • topic starter
  • Posts: 3,590
  • Joined: 24 Feb 2013
  • Loc: Hunter Valley NSW Australia

Posted 18 January 2017 - 11:28 PM

I have no direct experience with wire spiders, so I have to ask the question (assuming a standard non-Stewart truss configuration):

 

Wouldn't secondary collimation (or more accurately stated, Focuser Axial Error) elimination by the tuners be fairly tedious?

Well...I've only had to do it once, when I built it, and it wasn't tedious then. Not needing to use tools make things easy...but to be honest, not having to do this prior to observing means i don't really know; I don't see why it should be, but just don't know. :shrug:

Does this help?

 

 

You may have been my inspiration to use the locking guitar heads...   In collimating the wire suspended secondary I am amazed how fast and easy it is to move the secondary mirror around.  It's like you can move one of the 8 machine heads and it immediately will work or not, instant feed back!  I am at a compete loss of how to effectively string the words together to convey the ecstatic delight in adjusting the mirror.  Is it because there are no tools involved?  Is it like a 'joystick' on an airplane? 

 


Edited by Oberon, 18 January 2017 - 11:29 PM.


#77 I forge iron

I forge iron

    Explorer 1

  • -----
  • Posts: 60
  • Joined: 20 Oct 2016
  • Loc: S W Nebraska

Posted 19 January 2017 - 12:30 AM

I was thinking about this today and I think a general comparison is something like a complete amateur climbs up in a backhoe operator's seat pulls the levers and it's like you grab the correct lever every time and it either works pulling the laser collimator the direction you want it to move or it does not. The light swims around and once the assembly has been centered any one of the eight tuning machine heads will responsively move the light beam around. Since there are not tools to find, drop, loose or, insert into some small slot... it's like flying an airplane; you make the right choice or it's the wrong choice / switch to a different tuning knob.

... AND, My choice of using turning machine heads ... is under ALL your inspiration, THANK YOU!

Gifts like this don't happen every day!
Fred

Edited by I forge iron, 19 January 2017 - 12:31 AM.

  • Oberon likes this

#78 Bob4BVM

Bob4BVM

    Mercury-Atlas

  • *****
  • Posts: 2,581
  • Joined: 23 Mar 2015
  • Loc: W. Oregon

Posted 19 January 2017 - 04:16 PM

Bob, you remember correctly, there is no need for mirror adjustment screws with a wire spider configured as I have illustrated, as the tuners can take care of that. However in my case I never need to adjust the tuners either, as my Stewart Platform is all I need for collimation after initial construction set up.

Great, thanks Jonathan.

I will go ahead and build my new sec holder with no screws, easy enough to add later if needed.

At this point I am really leaning toward truss collimation as you have done, sounds too cool to not at least try !

Thanks for sharing the intricate details of  your many ingenious ATM ideas !

 

CS

Bob


  • Oberon likes this

#79 drneilmb

drneilmb

    Apollo

  • *****
  • Posts: 1,023
  • Joined: 05 Jan 2017
  • Loc: Decorah, IA, USA

Posted 30 January 2017 - 10:28 PM

Consequently these 3 spider and UTA concepts offer ideal solutions for spider geometry; full-size, compact and UL.

gallery_217007_4746_60444.png


I am very excited about the third possibility for a lightweight secondary ring to help balance a ball scope design. In the two views, it is hard for me to tell if there are a total of four or eight guitar tuners. Are all of the connections between the mirror mount and the wires rigid, with a single strand of wire going to each tuner?

Thanks for any clarification you can give.

-Neil

#80 Oberon

Oberon

    Soyuz

  • -----
  • topic starter
  • Posts: 3,590
  • Joined: 24 Feb 2013
  • Loc: Hunter Valley NSW Australia

Posted 30 January 2017 - 10:59 PM

Neil, there are 8 tuners and 8 wires in the illustration, much as per the OP.

It is possible to use less of either while retaining the same configuration, but that results in reduced control. For example, you may choose to string the wires through the secondary support to save on wire, or you may choose to string two wires to the one tuner to save on tuners. You wouldn't do it for cost, because the cost of wire and tuners are so trivial, but on an UL you may want to reduce the number of tuners to minimise weight. In that case I would probably first look at a lighter weight open body tuner or even make my own using a modified screw, or something of the sort, and retain the ability to tension each wire individually. Individual tuners - one for each string - allows you to use a simpler stiffer and lighter structure on the secondary support with no collimation screws.

Note that an advantage of the UL design here with 8 tuners is that the tension is applied to both sides of the UTA ring, which prevents twisting the ring. This permits a lighter thinner ring; using 4 tuners only would require a stiffer (heavier) ring.


  • I forge iron likes this

#81 astrobeast

astrobeast

    Mariner 2

  • *****
  • Posts: 225
  • Joined: 26 Sep 2009
  • Loc: Milford, CT

Posted 08 February 2017 - 04:51 PM

Hi all,

 

Just starting to take a look at this forum as contemplating building a 10-12.5" reflector. 

 

While there is a lot of great info on building wire-spiders, didn't see a lot on the motivation. I'm guessing this is for reducing the amount of diffraction. If this is this correct, can someone either qualitatively or qualitatively relate how much of a difference it makes visually.

 

Thanks!

 

Rick



#82 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 9,763
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 08 February 2017 - 05:10 PM

Yes it reduces the diffraction, done right to insignificant levels.  Motivation enough?  YMMV.

 

http://www.raddobs.c.../wirespider.htm


Edited by mark cowan, 08 February 2017 - 05:10 PM.

  • I forge iron likes this

#83 astrobeast

astrobeast

    Mariner 2

  • *****
  • Posts: 225
  • Joined: 26 Sep 2009
  • Loc: Milford, CT

Posted 08 February 2017 - 05:27 PM

Well that's pretty cool!!

 

Anybody out there with comparative photos or simulations? 



#84 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 9,763
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 08 February 2017 - 06:20 PM

H.R. Suiter has some but you'll have to search them out online.  Simulations when it comes to very small aspect ratio wire spiders (i.e. wires on the order of .008" for a 15" aperture) are almost impossible to run due to the precision required.  But the test itself is easy - string a thin wire across a well equilibrated scope without it lining up with existing supports and see what sort of diffraction ensues on bright stars.



#85 Jeff Morgan

Jeff Morgan

    Voyager 1

  • *****
  • Posts: 12,716
  • Joined: 28 Sep 2003
  • Loc: Prescott, AZ

Posted 08 February 2017 - 10:49 PM

While there is a lot of great info on building wire-spiders, didn't see a lot on the motivation. 

 

Most likely you can add thermal effects to the list:

 

http://www.astrosurf...thermique_e.htm

 

If your French is rusty, the general idea is that painted vanes (the standard practice) has some very undesirable effects with respect to emissivity, chilling vanes below ambient and creating refraction effects.

 

Wire vanes are of course unpainted. And there is a lot less mass.



#86 I forge iron

I forge iron

    Explorer 1

  • -----
  • Posts: 60
  • Joined: 20 Oct 2016
  • Loc: S W Nebraska

Posted 09 February 2017 - 10:55 PM

That about sizes the problem up..  However.  ... The thermal mass of guitar wires are a mere fraction of the mass of a flat ribbon... Or... God forbid!  A curved spider!  

 

The 'Motivation' ?  A wire spider constructed with guitar machine heads  (once roughed in) is far rewarding for quickest and easiest collimation I have ever done!

 

Fred

 

 

P.S.  ... Then.... My strengths are not yours, and your strengths are not mine!


Edited by I forge iron, 09 February 2017 - 11:00 PM.


#87 I forge iron

I forge iron

    Explorer 1

  • -----
  • Posts: 60
  • Joined: 20 Oct 2016
  • Loc: S W Nebraska

Posted 09 February 2017 - 11:10 PM

Neil, there are 8 tuners and 8 wires in the illustration, much as per the OP.

It is possible to use less of either while retaining the same configuration, but that results in reduced control. For example, you may choose to string the wires through the secondary support to save on wire, or you may choose to string two wires to the one tuner to save on tuners. You wouldn't do it for cost, because the cost of wire and tuners are so trivial, but on an UL you may want to reduce the number of tuners to minimise weight. In that case I would probably first look at a lighter weight open body tuner or even make my own using a modified screw, or something of the sort, and retain the ability to tension each wire individually. Individual tuners - one for each string - allows you to use a simpler stiffer and lighter structure on the secondary support with no collimation screws.

Note that an advantage of the UL design here with 8 tuners is that the tension is applied to both sides of the UTA ring, which prevents twisting the ring. This permits a lighter thinner ring; using 4 tuners only would require a stiffer (heavier) ring.

 

I agree there would be little or difficult collimation with only 4 adjustment points.  There would be no adjustment to center secondary mirror in focus tube up and down the OTA.  Also the thing that makes 8 adjustment points such a joy is turning one tuner one way and turning the opposite the other way at the same time... the operation is self evident in it's adjustment. Look at a bicycle wheel... all the spokes pull on a tangent of the hub.... Only a child's  trike has perpendicular spokes... and there must be 8 to hold the rim stable.   If you placed such an arrangement on a secondary without  radially placing the tension spokes there would be no way to turn the secondary mirror from left to right in the ) OTA (with out a center rotating pin)  ... and that to my present memory is much more difficult to turn and adjust than the use of the guitar machine head tuners... That are so "Buttery" feeling in your hands (sorry JDob!)

 

Fred 


Edited by I forge iron, 09 February 2017 - 11:16 PM.


#88 I forge iron

I forge iron

    Explorer 1

  • -----
  • Posts: 60
  • Joined: 20 Oct 2016
  • Loc: S W Nebraska

Posted 09 February 2017 - 11:39 PM

 

While there is a lot of great info on building wire-spiders, didn't see a lot on the motivation. 

 

Most likely you can add thermal effects to the list:

 

http://www.astrosurf...thermique_e.htm

 

If your French is rusty, the general idea is that painted vanes (the standard practice) has some very undesirable effects with respect to emissivity, chilling vanes below ambient and creating refraction effects.

 

Wire vanes are of course unpainted. And there is a lot less mass.

 

 

....?   Sorry!  ... My French is somewhere in the swamp alongside my computer skills!  ... But thanks' for the effort!

 

Fred 



#89 brickbots

brickbots

    Explorer 1

  • -----
  • Posts: 67
  • Joined: 02 Oct 2010
  • Loc: Sydney, Australia

Posted 13 February 2017 - 07:13 AM

Question for everyone here, particularly Jonathan (Oberon) who seems to have the mechanics and maths down.  How workable is a three vane wire spider?  I have seen a few examples of this, but I'm not sure how to analyze the mechanics of them. 

 

I currently have have a three vane curved spider and its heavier than I like and does not hold collimation as well as it could through altitude changes. I'm really intrigued by the stability and adjustability of wire spiders, but may have to do some work to the OTA to switch to four vanes. 



#90 Oberon

Oberon

    Soyuz

  • -----
  • topic starter
  • Posts: 3,590
  • Joined: 24 Feb 2013
  • Loc: Hunter Valley NSW Australia

Posted 13 February 2017 - 07:55 AM

There is a thread somewhere on that topic where some serious attempts were made to develop such, but each attempt failed. You can build a spider that way but it will suffer from lack of torsional stiffness. You could however build a successful 4 vane wire spider with two of the vanes terminating at one point aligned behind the mirror, and so utilise your existing 3 point support structure, but it wouldn't be optimal.

 

med_gallery_217007_4746_97634.png


Edited by Oberon, 13 February 2017 - 08:29 AM.

  • maxmercy and brickbots like this

#91 Mad Matt

Mad Matt

    Vanguard

  • *****
  • Posts: 2,438
  • Joined: 20 May 2003
  • Loc: Frankfurt, Germany

Posted 13 February 2017 - 10:08 AM

Simply a little food for thought. I have not built this but if I do get around to building another dob I would give it a try. 

 

This is an 8 wire spider design that distributes all individual diffraction spikes evenly with no overlapping. I think this would only work well in a alt/Az configuration as the wire set on bottom half of the tube are providing considerable tangental stability but little radial stability.

 

Green is the upper, blue the lower or the other way around, I don't think it matters Edit: Just remembered, it does matter. As the tube moves towards the horizon, the weight of the secondary applies more torque so green needs to be at the top, blue bottom so that force is taken up by the wires with maximum radial stiffness.

 

 

The advantage to this design (i think) is that the diffraction pattern can "slip" under the eyes detection level as no diffraction spikes overlap or are close to overlapping and as a result are fainter. Of course that also depends on the gauge wire that is used. 

 

Asymetrical spider angle mockup v4.png


Edited by Mad Matt, 13 February 2017 - 10:35 AM.


#92 jtsenghas

jtsenghas

    Fly Me to the Moon

  • *****
  • Posts: 5,334
  • Joined: 14 Sep 2014
  • Loc: The flatlands of Northwest Ohio 41.11N --Bloomdale

Posted 13 February 2017 - 11:44 AM

Simply a little food for thought. I have not built this but if I do get around to building another dob I would give it a try. 

Hmmm....

 

I wonder how something like that angle arrangement would do on Jonathon's third option shown in post 71, and again in post 79, on an ultralight build with a single ring.

 

In that case the upper and lower lines would be reversed so that the more radial lines would again take up more load at low scope altitude angles, but I can see how that could work.

 

Diffraction angles, if visible, would be very faint eight pointed stars. Interesting.


  • Mad Matt likes this

#93 Mad Matt

Mad Matt

    Vanguard

  • *****
  • Posts: 2,438
  • Joined: 20 May 2003
  • Loc: Frankfurt, Germany

Posted 14 February 2017 - 04:45 AM

 

Simply a little food for thought. I have not built this but if I do get around to building another dob I would give it a try. 

Hmmm....

 

I wonder how something like that angle arrangement would do on Jonathon's third option shown in post 71, and again in post 79, on an ultralight build with a single ring.

 

In that case the upper and lower lines would be reversed so that the more radial lines would again take up more load at low scope altitude angles, but I can see how that could work.

 

Diffraction angles, if visible, would be very faint eight pointed stars. Interesting.

 

Yes, agreed, if you are using a single ring then the blue should be mounted to the upper side and green to the lower with the wires crossing the pane of the ring. 



#94 TonyStar

TonyStar

    Viking 1

  • *****
  • Posts: 584
  • Joined: 25 May 2013

Posted 14 February 2017 - 11:23 AM

Hi Jon,

 

first of all, let me praise your nicely executed secondary wire spider!

As you mentioned (and this is a general issue with building portable telescopes) the designer has to compromise between size of the UTA and spider stiffness.

I was curious, how stiff is your spider against torques caused by the weight of the secondary/holder assembly? Can you actually see the spot of a laser collimator shift slightly when the scope moves from zenith to horizon? 


  • Oberon likes this

#95 jtsenghas

jtsenghas

    Fly Me to the Moon

  • *****
  • Posts: 5,334
  • Joined: 14 Sep 2014
  • Loc: The flatlands of Northwest Ohio 41.11N --Bloomdale

Posted 14 February 2017 - 11:34 AM

I was curious, how stiff is your spider against torques caused by the weight of the secondary/holder assembly? Can you actually see the spot of a laser collimator shift slightly when the scope moves from zenith to horizon? 

Before Jonathan reassures us yet again that the wire spider currently residing in Merope is very stable for collimation, let me offer the following observation:

 

This configuration has the secondary holder somewhat nested within the highly angled wires. As a result the center of gravity of the secondary is not cantilevered very far from the supports, which further reduces the loads and torques you are concerned about. 


Edited by jtsenghas, 14 February 2017 - 11:37 AM.

  • Oberon likes this

#96 Oberon

Oberon

    Soyuz

  • -----
  • topic starter
  • Posts: 3,590
  • Joined: 24 Feb 2013
  • Loc: Hunter Valley NSW Australia

Posted 15 February 2017 - 01:44 AM

Hi Jon,

 

first of all, let me praise your nicely executed secondary wire spider!

As you mentioned (and this is a general issue with building portable telescopes) the designer has to compromise between size of the UTA and spider stiffness.

I was curious, how stiff is your spider against torques caused by the weight of the secondary/holder assembly? Can you actually see the spot of a laser collimator shift slightly when the scope moves from zenith to horizon? 

Thats a good question.

I do see the laser collimator spot shift as I move from zenith to horizon (from centre of paper circle to edge of the hole), but I don't think it is the wire spider at fault. Perhaps a small component of the movement may be the spider, but if so, it is in any case dominated by a shift associated with the load shifting on the trusses, and that in turn most likely because Merope's UTA is a little light, not quite as rigid as it would be if I was to build it again.

I spent a good hour today trying to identify the source:-

1. The focuser could flex but was easily ruled out because the likely shift was in the reverse direction.
2. The secondary mirror could be shifted a little by hand in any direction, but it was difficult to find a convincing relationship.
3. I clamped the altitude on the horizon and loaded/unloaded the trusses by lifting and weighting the UTA. This reproduced the motion seen in the laser. However the action of loading the UTA could potentially distort the UTA and thereby alter tension on the wire spider.
4. I flexed the truss tubes sideways by grabbing two adjacent tubes in the middle and pressing them together and pulling apart. This had an interesting effect. If the two tubes were joined at the UTA there was absolutely no effect. However if the two tubes were joined at the LTA (and thus separate at the UTA) then a significant shift occurred. Again this points me to flexure in the UTA, which in turn may alter tension on the spider.

5. One of the spider strings had a lower looser sounding note than the others, so I tightened it, and of course then had to tighten all of the strings to bring it back to collimation. The string was in a position that I felt made it a likely culprit. However the consequence was that the magnitude of the shift was not improved, but if anything now very slightly increased (the laser spot now touched the edge of the paper ring at horizon). Again, this probably points to flexure in the UTA (although I haven't figured out how yet).

Bottom line: yes Merope has some flexure that affects the spider, but it isn't the wire spider itself at fault, except to say that a wire spider is intrinsically more exposed to flexure in the UTA than a solid vaned spider because (unless you use a single ring) it is loaded at more points. In other words, the design, construction and stiffness of the UTA is more critical if you use a wire spider.

Be aware too that the UTA is a critical component of the truss system, and it will flex as the load shifts on the truss system. I've used 25mm aluminium trusses for Merope and consider them to be borderline adequate, and the UTA to be on the light side. In particular the red tube material is only attached with thin foamed double sided tape and performs very little by way of being a structural element; that was a mistake. I should have epoxied it into place, and I should have laminated the 9mm thin ply rings with carbon.

Next time...

gallery_217007_5817_320420.jpg


Edited by Oberon, 15 February 2017 - 01:57 AM.


#97 jtsenghas

jtsenghas

    Fly Me to the Moon

  • *****
  • Posts: 5,334
  • Joined: 14 Sep 2014
  • Loc: The flatlands of Northwest Ohio 41.11N --Bloomdale

Posted 15 February 2017 - 07:09 AM

Thank you, Jonathan, for the excellent writeup on that experimentation.  I've been playing around with a few analyses of forces on such structures for future builds.  The single ring UTA solves some signification problems, but creates others that must be solved. 

 

One thing I think well worth doing on such experiments is using a barlowed laser (or a collimation tool that uses that method like Howie Glatter's tools or my homemade "conimator") to isolate the portion of that shift that is from primary axial error.  Perhaps you did these experiments with such a tool, but if not, repeating them with one may be worthwhile. 

 

That Primary Axial Error can indeed be increased by flexure of the UTA, but due to the relative distances involved, such flexure far more affects the secondary axial error- -which has a considerably looser tolerance on such fast scopes since this error tolerance depends only on aperture. As a result, the shifts you observe may not be as significant as you may fear--again, depending on what collimation tool you may have used.

 

Back to the subject of this thread, I'm convinced that such spiders are inherently robust. Good job. . 


Edited by jtsenghas, 15 February 2017 - 07:58 AM.


#98 Oberon

Oberon

    Soyuz

  • -----
  • topic starter
  • Posts: 3,590
  • Joined: 24 Feb 2013
  • Loc: Hunter Valley NSW Australia

Posted 15 February 2017 - 07:58 AM

I used the 2" Glatter with the fine aperture for a tight spot. I wasn't paying attention to the return, I was only interested to isolate the cause/s of deviations from the primary mirror center marker with altitude.


  • jtsenghas likes this

#99 Bob4BVM

Bob4BVM

    Mercury-Atlas

  • *****
  • Posts: 2,581
  • Joined: 23 Mar 2015
  • Loc: W. Oregon

Posted 15 February 2017 - 03:41 PM

For my Dob rebuild, I am building a new UTA mockup (which may or may not become the actual UTA) in order to do thorough and precise measurement of ring structure flexure and angular shift of a heavy secondary on a wire spider. Also to compare performance of very thin wires.

Hopefully I will end up with a much lighter and stiffer UTA.  One thing I will know is that any collim. shift will not be coming from my UTA.

I am sticking with a 2-ring design for its advantages of stiffness and ease of mounting focusers and finders, and it's easier for me to experiment with a tilted focuser for low rider design testing.

 

Jonathan- Your test has me wondering if the flex is in any way due to your hexapod (vs octapod) truss. All else being equal, would 8-25mm tubes be stiffer ?

 

CS

Bob



#100 Oberon

Oberon

    Soyuz

  • -----
  • topic starter
  • Posts: 3,590
  • Joined: 24 Feb 2013
  • Loc: Hunter Valley NSW Australia

Posted 15 February 2017 - 05:17 PM

Not likely because it isn't a simple matter of the UTA load sitting on 3 or 4 points. The UTA is an integral part of the truss, forming the 3rd leg of each triangle, and so a weak UTA makes for a weak truss, and adding more trusses won't fix that.

 

But lets keep it in context; my problem arose because I was on a weight cutting exercise and I abandoned my usual approach and went with methods that seemed "tried and true" because everybody else was doing it. Merope, remember, was always an experiment, a test bed for construction methods and design. So I looked at the methods other atm's used for UTA and decided that I was being too conservative and could afford to go thinner and to not treat the tubular section as structural. In hindsight I should have stuck with my intuition. Well OK you can and in practice Merope performs perfectly well at all the altitudes that matter...but it could be better, and I know why.

 

A perfectly engineered Merope would use carbon truss tubes and a carbon monocoque UTA. Anything less is a compromise, and life is full of compromises. Stick with your hexapod, you will love it. And remember that it eliminates the need for adjustable components in your mirror cells.


  • Jeff Morgan and jtsenghas like 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