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

How to minimize spider vane spikes?

This topic has been archived. This means that you cannot reply to this topic.
94 replies to this topic

#76 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 41085
  • Joined: 23 Jun 2003

Posted 28 March 2015 - 12:36 AM

What are you looking for?  Less intense spikes?  How would you tell or even notice?

Vanes that stay above ambient temperature?  Could you get a reading?

The experiment is only worthwhile if you can notice a difference or even know what difference to look for.



#77 Oberon

Oberon

    Mercury-Atlas

  • -----
  • Posts: 2806
  • Joined: 24 Feb 2013

Posted 28 March 2015 - 02:00 AM

Jonathan,

You're still drawing the two arc secondary vanes (bottom row, middle) incorrectly.  See post #66, which is closer to actual.

 

OK...I've redrawn a couple your way (one with 100mm CO the other with 125mm CO), run the simulation again in the interests of completeness, and thrown in a few others for comparison.

gallery_217007_4746_98156.jpg

 



#78 Daniel Mounsey

Daniel Mounsey

    Vendor (Woodland Hills)

  • *****
  • Vendors
  • Posts: 7429
  • Joined: 12 Jun 2002

Posted 28 March 2015 - 02:45 AM

Could alway be that your vanes aren't perpendicular to the primary....if they are twisted a bit, that expands the surface area of the vane and will create bigger spikes.......

 

Yea, like this http://www.cloudynig...-curved-spider/



#79 Nils Olof Carlin

Nils Olof Carlin

    Vanguard

  • *****
  • In Memoriam
  • Posts: 2227
  • Joined: 26 Jul 2004

Posted 28 March 2015 - 04:30 AM

It is a template based on this research work - http://www.ejournal....11/RMF51116.pdf
 
I use it for my 10" Newt and is suppresses about 2/3 of the spiking...plus improves the planetary view subtly.  Attached is the template I made for my XT10.  Just cut them out and tape them over the section of the vanes that are over the mirror.

Harold Suiter and Bill Zmek wrote a "Dialogue on Spider Diffraction"in issue 11 of the ATMJ, reprinted by Willmann-Bell in "The best of Amateur Telescope Making Journal vol 1". Their discussion is worth reading if you can find the book. Their simulations include the design in #73 and others by Couder. They write:
"In ATM II, a suggestion appeared that was originally connected with a non-rotating tube telescope and an unfortunately placed double star." "...it is our firm opinion that these devices should only be employed in their original function - as a work-around. The important thing is minimizing the area covered, and in every case that area is large."

Nils Olof

#80 pstarr

pstarr

    Skylab

  • *****
  • Posts: 4316
  • Joined: 17 Sep 2004

Posted 28 March 2015 - 08:33 AM

 

Could alway be that your vanes aren't perpendicular to the primary....if they are twisted a bit, that expands the surface area of the vane and will create bigger spikes.......

 

Yea, like this http://www.cloudynig...-curved-spider/

 

Destiny spiders have less vane material in the light path than one may think. According to their link, they are close to 60* on the arcs. Also the pictures posted by Johnatan above don't allow for the tube clearance. As shown in the link, that can make a difference as to what the mirror sees of the spider.

http://www.destinyco...urved_vanes.htm I use a Destiny in my 12.5" with a 16" tube. see below. Knock 1-3/4" off the ends of what you see in the photo and what's left is what I have in the light path. Vane material could maybe be a little thinner, mine is 1/16" but I don't know if thinner would be as stable. The one I'm using doesn't change the view I see in a autocollimator when I move the scope to different positions. Are these as good as the ones Grissom makes, probably not. Outside of making your own, I think they are a decent alternative and you can actually buy one of these. If you look real close, you can see the vane arcs reflecting in the mirror. That's what the mirror sees of the spider.

 

IMG_1305s4.jpg


Edited by pstarr, 28 March 2015 - 07:22 PM.


#81 Oberon

Oberon

    Mercury-Atlas

  • -----
  • Posts: 2806
  • Joined: 24 Feb 2013

Posted 29 March 2015 - 02:43 AM

CORRECTION!

method "I" illustrated below utilising 45 degree curves is not a good design as, unlike E, it does not spread diffraction energy evenly around the field.

 

 

ATM's wishing to to build the stiffest and most efficient curved spider that spreads the least possible diffraction most evenly across the entire field would be best served with either method E or method I, depending on whether they have a 6 or 8 pole truss.

The illustration below shows why methods F and H should be avoided.

gallery_217007_4746_180221.png

 



#82 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 22364
  • Joined: 22 Feb 2010

Posted 29 March 2015 - 07:30 AM

 

 

 

you can always cut something like this out of thin black plastic and tape it to all your vanes when viewing Jupiter.

 

its easy, its cheap and if you don't like it its easy to throw away.

 

attachicon.gifvane mask.jpg

Please, help an old guy out here.  What is this a pencil drawing of?  How is it applied to the vanes?  What effect does it have on the spikes?

 

Dave

 

 

It is a template based on this research work - http://www.ejournal....11/RMF51116.pdf

 

I use it for my 10" Newt and is suppresses about 2/3 of the spiking...plus improves the planetary view subtly.  Attached is the template I made for my XT10.  Just cut them out and tape them over the section of the vanes that are over the mirror.

 

 

yes apodizing masks for the vanes.

 

while looking up apodizing masks I've always saw the 3 layers of wire mesh but I never saw the one layer of scallops

 

http://stargazerslou...-1374582696.jpg

 

another inexpensive method.



#83 Cotts

Cotts

    Just Wondering

  • *****
  • Moderators
  • Posts: 9426
  • Joined: 10 Oct 2005

Posted 29 March 2015 - 08:25 AM

Wild, out of left field idea.  For newts up to 11" or so.  How about an optical window?  The same sort of glass that SCT corrector plates are made of. I believe these corrector plates start out as flat on both sides and are put under slight tension for grinding which leaves the correct figure when the tension is released.  Celestron must buy the optical windows that they grind to the correct configuration for SCT use.  What is their source?  Or does Celestron make its own optical windows?  Either way, optical windows exist already.

 

the SCT corrector plates are plenty strong if they can hold the Hyperstar/DSLR equipment with no problem.   Hooking up a secondary mirror housing would be easy...

 

They can't be that expensive.  What portion of the cost of a 8" Celestron would be tied up in the corrector plate's glass?   Even if they are $200 that's not a lot to spend if achieving a bit better contrast is worth it to you..

 

Of course you would now be dealing with a grade-A dew magnet....

 

Dave

I'll just start my own thread, then.

 

L8r

 

Dave



#84 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 22364
  • Joined: 22 Feb 2010

Posted 29 March 2015 - 09:01 AM

I have an earlier version of Tex's book that doesn't have the chapter about optical windows.

 

I doubt many here have made an optical window.

 

I breezed thru Tex's book that did have the optical window and the pic showed putting up a window against the primary to test thru the window I presume to see if it effects the wavefront.


Edited by Pinbout, 29 March 2015 - 09:04 AM.


#85 DesertRat

DesertRat

    Fly Me to the Moon

  • *****
  • Posts: 6253
  • Joined: 18 Jun 2006

Posted 29 March 2015 - 09:12 PM

Bill write:

It is a template based on this research work - http://www.ejournal....11/RMF51116.pdf

 

I use it for my 10" Newt and is suppresses about 2/3 of the spiking...plus improves the planetary view subtly. Attached is the template I made for my XT10. Just cut them out and tape them over the section of the vanes that are over the mirror.

 

 

Interesting Bill.

 

I don't believe an additional obstruction can improve overall performance.   It is true this kind of mask will shorten the diffraction spikes, however it is at the cost of spreading even more energy where it is not wanted.  I believe any perception of improvement would be far more likely due to seeing changes.

 

After a central obstruction passes some percentage, usually noted in the 16-20% range, any further apodization or obstruction is not effective, at least for the normal useage a telescope is intended for.  An unobstructed optic can benefit from a gaussian mask, in fact some kinds of experiments require it.  In that case not only are there no spikes but no diffraction rings as well, at some cost in 'resolution' as the center spot is enlarged some.

 

Not sure what people are after with these spider questions.  Seems like we have solutions chasing a mostly harmless problem.  The performance of a reflector has many issues that need  addressing, with spiders far down the list.

 

I don't know if its important but I note the paper you referenced has zero citations.

 

Glenn  

 

 



#86 GeneT

GeneT

    Ely Kid

  • *****
  • Posts: 15401
  • Joined: 07 Nov 2008

Posted 29 March 2015 - 10:09 PM

A lot of people like the curved spider vanes, but I prefer the four vane variety. I know how to tweak them, and I don't know how the three vane, curved ones work. Anyway, just look past the spikes; concentrate on the sky object, and over time, you will almost forget that the spikes are there.



#87 Oberon

Oberon

    Mercury-Atlas

  • -----
  • Posts: 2806
  • Joined: 24 Feb 2013

Posted 30 March 2015 - 02:32 AM

Unless you're using a binoscope and the spikes are misaligned (which they will if you have a rotating UTA for adjusting IPD). Basically your brain can't work out which orientation the spikes are supposed to be and flickers between the choices.



#88 BillP

BillP

    Hubble

  • *****
  • Posts: 18448
  • Joined: 26 Nov 2006

Posted 30 March 2015 - 10:22 AM

Bill write:

It is a template based on this research work - http://www.ejournal....11/RMF51116.pdf

 

I use it for my 10" Newt and is suppresses about 2/3 of the spiking...plus improves the planetary view subtly. Attached is the template I made for my XT10. Just cut them out and tape them over the section of the vanes that are over the mirror.

 

 

Interesting Bill.

 

I don't believe an additional obstruction can improve overall performance.   It is true this kind of mask will shorten the diffraction spikes, however it is at the cost of spreading even more energy where it is not wanted.  I believe any perception of improvement would be far more likely due to seeing changes.

 

 

You may believe that if you wish, but I do not since the testing protocol and quantity of tests mitigated that possibility.  On numerous occasions of testing and comparing, to suggest that seeing consistently was worse when the mask was off would be quite a statistical unlikelihood.  It was subtle, but it was there, and it was consistent across months of testing.  What I did, fwiw, is create a pop-on mask to fit the Dob's opening, so I could just pop the mask on and off in a flash.  What was also interesting to do with this device, was to rotate it so the pattern was no longer over the spider vanes.  Was fun to watch the spikes get longer then shorter :lol:   I certainly would not go through the trouble for the very subtle contrast improvement the planetary image showed, but well worth it for suppression of the annoying spikes.


Edited by BillP, 30 March 2015 - 10:26 AM.


#89 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 22364
  • Joined: 22 Feb 2010

Posted 30 March 2015 - 11:14 AM

from vlads site

 

"In all, apodizing mask can produce effects beneficial for observing extended details larger than roughly twice the stellar resolution limit. It is not dramatic: generally less than the benefit of correcting 1/4 wave P-V of spherical aberration, but it is easy to employ and deploy, and every little bit can count. Apodizer can also be helpful in resolving close unequal doubles, with the faint companion situated in the first bright ring of the principal star. These beneficial effects, however, come at a price of contrast loss with details close and at the limit of resolution. In general, benefits of apodizing are greater for larger apertures, which are due to the larger seeing error typically able to resolve only details in their mid-to-low frequency range. Finally, apodizing mask will have little effect in apertures with central obstruction significantly larger than 0.2D"

 

http://www.telescope...dizing_mask.htm


Edited by Pinbout, 30 March 2015 - 11:16 AM.


#90 Nils Olof Carlin

Nils Olof Carlin

    Vanguard

  • *****
  • In Memoriam
  • Posts: 2227
  • Joined: 26 Jul 2004

Posted 30 March 2015 - 01:04 PM

Danny,
what Vlad talks about here is the mesh-type pseudo-Gaussian mask that cuts down diffraction rings (at the cost of a larger Airy disk), not the odd-shaped Couder type masks for spider vanes we've talked about here.
The basic reason why it works so poorly with obstructed telescopes is that the relatively strong first diffraction ring is largely caused by the obstruction itself, which cannot really be apodized against.
(If it appears to work well, as some say, it just may be because it shades a poorly corrected edge...)
Nils Olof

#91 DesertRat

DesertRat

    Fly Me to the Moon

  • *****
  • Posts: 6253
  • Joined: 18 Jun 2006

Posted 30 March 2015 - 02:27 PM

Hi Danny,

 

I did not visit that website, but I am pretty sure Vlad was addressing apodization of the entire pupil.  A gaussian apodization will control diffraction rings, but at the expense of enlarging the central spot.  When the central obstruction passes some mark, that being dependent on what one is after, apodization is mostly ineffective. 

 

As I mentioned earlier, an incoming planar wavefront shaped into a gaussian profile over a circular aperture will focus to another gaussian, with no diffraction rings.

 

Bill,
I know some people here don't trust simulations by Suiter and others so I wont aggravate the situation.  However one can predict what the effect of that apodization is simply by inspection.  The higher frequencies are captured by zones of increasing radius.   This apodization not only increases the obscuration, which spreads energy about the central diffraction spot, but as you go out from the center you find information being removed from the middle spatial frequencies.  It is effective in making the diffraction spikes smaller but at the cost of spreading more light about the central spot.

 

My analysis shows a contrast loss of approximately 2-3% across the middle frequencies.  So its not a huge effect, but I hardly see the utility of it.

 

Not everything published in a journal or website is true.  As a reader of various journals I can state that with the explosion of desktop publication even some respected journals release stuff which is a little dodgy.  The peer reviewed process is not what it once was.  That may be good sometimes, as good stuff in the past failed publication because it was truly new and went against conventional wisdom.  But i don't think this paper is in that category.

 

Glenn



#92 BillP

BillP

    Hubble

  • *****
  • Posts: 18448
  • Joined: 26 Nov 2006

Posted 30 March 2015 - 03:23 PM

My analysis shows a contrast loss of approximately 2-3% across the middle frequencies.  So its not a huge effect, but I hardly see the utility of it.

 

Not everything published in a journal or website is true. 

 

Is the contrast loss across the entire image, or is your analysis restricted to only the contrast loss where the spike of light is intersecting an image in the FOV.  This is critical because the measurement of the loss must be at the central point of the extended image where the spikes are intersecting.

 

But any way...very true one should not trust without verify.  But I did verify on many occasions so not a matter of "believing".  Empirically it has slight advantage on a planetary image.  One needs to remember that when theory does not match results, and one is certain they adequately mitigated confounding factors, then the theory is simply incomplete.  In our realm though, when folks suggest that something must be a certain way because x, y, or z states so, it is more likely that it is not that the theory is technically lacking, but that the application of the single theory to the complex "system" of things going on is just inappropriate.  To adequately describe what is going on with a telescope requires much more than any single-focused component of optical theory.  Everything happening in a complex optical system needs to be modeled completely so that the interaction of all the components can be determined.  This interaction is what is never addressed. 

 

As far as the paper...what is not to trust?  They had a contention, built and experiment, and demonstrated the results.  I rebuilt their experiment, and replicated their results.  What's not to love in that process!

 

Bottom line though (always), is we have eyes to tell us what we see, and continue to see.  Best to trust them until some adequate system model can correctly inform the experiences the eyes are relating.  So as long as the theory is not predictive, cannot put much stock that the theory is being correctly applied to the complex system, or is near complete enough to be predictive of the system  :grin:


Edited by BillP, 30 March 2015 - 03:28 PM.


#93 DesertRat

DesertRat

    Fly Me to the Moon

  • *****
  • Posts: 6253
  • Joined: 18 Jun 2006

Posted 30 March 2015 - 07:02 PM

Hi Bill,

 

I would certainly agree that when someone says they see something, its hardly productive to argue the point.

 

I can refute the papers results, mostly because its incomplete.  For one thing, it ignores the angular dependency that results from this apodization.  A simple cross section of the point spread function does not demonstrate this.  In fact the apodized spider creates 4 peaks in the first diffraction ring more evident than a normal spider yields.  I don't know what kind of spider they compared against, they say a wire spider.  A wire spider does almost nothing to the contrast transfer of the optic.  Secondly they do not examine the imaging properties in detail, which would show the contrast being knocked down ~2.5% across the first 2/3 of the spatial frequency axis.  This is not a trivial obscuration being added.  Its consequence is not earth shattering but it does not seem called for, unless you really truly hate diffraction spikes.  Finally they sort of dodge around the extra energy released about the central diffraction spot.

 

I cannot explain your results.  However one can have certain aberrations, along with defocus and in various combinations that will result in perceived contrast features, primarilly in the higher spatial frequencies.  But most of that percieved contrast results in false detections, and in fact contrast reversals.  This can and does happen.  Additional obscurations can amplify these effects.  To be clear I am not saying this is your observation.

 

My analysis assumed perfect optics.  I can add effects of various perturbations but don't see what that would tell me that is not already known, explained in the preceding paragraph.

 

This is not 'theory' using the common parlance definition - that is something proposed but not proven.  Its the prediction of diffraction physics as we understand it going back at least one hundred years, which is proven.

 

I know this apodization reduces the extent of the diffraction spikes.  There is no doubt about that.  But it does not and cannot improve the imaging characteristics of an optic unless it is covering some flaw in the optic.

 

The objective here is only to inform.  I am an experimentalist, as you are apparently.  Keep it up!  I enjoy trying different things as well.  And I did say the paper was interesting. 

 

Glenn


Edited by DesertRat, 30 March 2015 - 07:04 PM.


#94 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 76160
  • Joined: 16 Jun 2004

Posted 15 April 2015 - 06:36 AM

I was looking for an image of a Kenneth Novak mirror cell and stumbled on this PDF of a 1992 Novak catalog that includes a "Twin Semi-Circle" spider...

 

These things have been around a while.. 

 

Jon



#95 wh48gs

wh48gs

    Surveyor 1

  • *****
  • Posts: 1840
  • Joined: 02 Mar 2007

Posted 19 April 2015 - 10:21 AM


I can refute the papers results, mostly because its incomplete. 

 

I'd say it's plain flawed. The PSF that according to it represents a 29% (linear) obstructed aperture with 4-vane spider is nowhere close to what it should be. With the obstruction and no vanes, the first bright ring is some 22 times fainter than the central maxima. Assuming that the two lined up vanes do not exceed 1% of the annulus area - which is a lot - the central maxima of the spike is some 100 times fainter than that of the aperture (in incoherent light; twice brighter in coherent). Since a vane is, in effect, an inverse aperture, this energy is of opposite sign, lessening both, central maxima and bright rings of the aperture, while brightening the dark rings. The PSF graph shows second maxima (1st bright ring) as bright as 70% of the peak, and even third maxima (2nd bright ring) at 45%. The actual second maxima should be below 5% of the peak. Even assuming they've got it right with their graphs, this bright rings would be very pronounced in the PSF image, yet there's no ring structure detectable.

 

The PSF is radially asymmetric, which it shouldn't be etc.

 

Shape of the apodising vane is broken into a triple aperture, which means that it would produce roughly twice wider spike than a rectangular vane of the same length. At the same time, since up to several times wider than the standard thin vane, its spike would also be up to several times shorter. It would have made it noticeably less pronounced, and relatively roundish. But its central maxima - which is where most of its energy is - is roughly 10 by 20 times the aperture central maxima. With its area being up to several times the area of the standard thin vane, it takes as much more of the energy from telescope's central maxima to the rings. Cannot be advantageous in terms of contrast, only the opposite.

 

As for what we "see" (or, for that matter, "feel"), there is a reason why blinding to the status is a must in any serious study, or why they have placebo group when testing medications. One cannot assume to be the exception to such a common human imperfection as it is having a bias.

 

Vla


Edited by wh48gs, 19 April 2015 - 10:22 AM.



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