Jump to content


Photo

Where does Light Scatter from Curved Vanes?

  • Please log in to reply
34 replies to this topic

#1 Brian Albin

Brian Albin

    Seeker

  • -----
  • Posts: 869
  • Joined: 22 Aug 2006
  • Loc: Western Oregon

Posted 13 January 2013 - 04:55 AM

When we say light scattered by curved support vanes is splashed across the image, where does it go?
Does light from a single star wash over the entire field, or does the light spread only around the star, and only as far outward from the star as the rays of a straight vane would have been imaged?

#2 MKV

MKV

    Vanguard

  • *****
  • Posts: 2476
  • Joined: 20 Jan 2011

Posted 13 January 2013 - 07:15 AM

Read more about it here.

#3 Atl

Atl

    Viking 1

  • *****
  • Posts: 501
  • Joined: 13 Apr 2012

Posted 13 January 2013 - 09:43 AM

Instead of large spikes you get a smaller light blur around the object. Diffraction is still noticeable, but with very small objects in the glare of a large one like some double stars it can lessen the glare making the smaller companion more obvious. On large objects like Jupiter it works but doesn't seem to have an advantage. Last night using one I was able to view the trapezium in M42 and see black space around e and f. In the past I sort of had to look for them closely.

#4 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 8052
  • Joined: 08 Aug 2007
  • Loc: La Union, PI

Posted 13 January 2013 - 09:49 AM

And here...at bottom.
http://www.telescope....net/spider.htm

"Diffraction effect of a curved vane can be illustrated by breaking it into a number of smaller, practically straight sections, with varying orientations (FIG. 110a). While the total amount of energy produced by a curved vane is identical to that of a straight vane of equal length and thickness, it is spread out wide, making it practically invisible (it still lowers the contrast the same, on average)."

#5 Jarad

Jarad

    Fly Me to the Moon

  • *****
  • Posts: 6354
  • Joined: 28 Apr 2003
  • Loc: Atlanta, GA

Posted 16 January 2013 - 10:11 AM

On bright objects, instead of seeing 4 individual bright spikes, you see a dim surrounding glow. It makes something like Jupiter look like there is a bit of atmospheric haze scattering light around it.

Jarad

#6 Atl

Atl

    Viking 1

  • *****
  • Posts: 501
  • Joined: 13 Apr 2012

Posted 16 January 2013 - 03:57 PM

One important thing with curved vanes I have learned. The arc of the curve must be 180 degrees. With one vane a full 180 degree arc must be created, with 2 vanes each must be 90 degrees, with three vanes each vane must be 60 degrees, and with a 4 vane 45 degrees for each vane. If all of the curves in the spiders light path do not come to 180 degrees the diffraction effect will not occur. If you exceed 180 degrees the diffraction will overlap creating unwanted effects that can have a detrimental effect on viewing. The more vanes the spider has the less curve it needs for each, but there is also more diffraction with more vanes.

#7 Jarad

Jarad

    Fly Me to the Moon

  • *****
  • Posts: 6354
  • Joined: 28 Apr 2003
  • Loc: Atlanta, GA

Posted 17 January 2013 - 11:16 AM

In a straight spider, each vane produces two spikes that stick out from opposite sides of the star. In a curved spider, each vane produces two fans that stick out from opposite sides of the star. The angle of the fans equals the curvature of the vane. So if they add up to 180 degrees, the fans spread out and "connect", making an even surrounding glow in a full circle. If they add up to less than 180 degrees, there will be dark gaps between the fans. If they add up to more than 180 degrees, there will be lighter stripes where 2 fans overlap.

Jarad

#8 Brian Albin

Brian Albin

    Seeker

  • -----
  • Posts: 869
  • Joined: 22 Aug 2006
  • Loc: Western Oregon

Posted 17 January 2013 - 02:29 PM

It sounds like the light stays with the star, and does not spill all over the picture.

Thank you all for the answers.

#9 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1892
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 17 January 2013 - 03:06 PM

Might the requirement really be that the arcs must be circular arcs, that add up to an even multiple of 180 degrees?

The 360 degree secondary produces diffraction that is not noticeable, same with a vignetting aperture. Don Pensack recommends a curved support design using two 180 degree arc supports.

#10 Thomas Karpf

Thomas Karpf

    Surveyor 1

  • -----
  • Posts: 1742
  • Joined: 09 Feb 2009
  • Loc: Newington, CT

Posted 17 January 2013 - 03:45 PM

It sounds like the light stays with the star, and does not spill all over the picture.


Not really. No more so than the light from the normal four diffraction spikes stays with the star. Take those four diffraction spikes and smear them around the star and that's the effect. Because there's more total length in the vanes when you use curved vanes, you smear MORE light from the star into the bajillion spikes you get from the curved vanes.

#11 Thomas Karpf

Thomas Karpf

    Surveyor 1

  • -----
  • Posts: 1742
  • Joined: 09 Feb 2009
  • Loc: Newington, CT

Posted 17 January 2013 - 03:47 PM

Might the requirement really be that the arcs must be circular arcs, that add up to an even multiple of 180 degrees?


You could certainly look at it that way, although I don't know why you would want more than 360 total. That would be like using two comnplete circles to hold the secondary rather than two semi-circles. You could use LESS than 360 degrees, but then each star would look like a 'snow angel' with wings on opposite sides.

#12 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1892
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 17 January 2013 - 04:13 PM

That would be like using two comnplete circles to hold the secondary rather than two semi-circles.


That is in fact a design I have been considering.

#13 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 3985
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 17 January 2013 - 04:15 PM

It sounds like the light stays with the star, and does not spill all over the picture.


No, quite the opposite. It spills everywhere around the star, but because it's so uniform (spikeless) you just don't recognize it easily. It's a bargain with the devil. :lol:

Best,
Mark

#14 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1892
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 17 January 2013 - 04:55 PM

It sounds like the light stays with the star, and does not spill all over the picture.


No, quite the opposite. It spills everywhere around the star, but because it's so uniform (spikeless) you just don't recognize it easily. It's a bargain with the devil. :lol:

Best,
Mark


If you can't see it, how big a devil can it be?

#15 Sean Cunneen

Sean Cunneen

    Let Me Think

  • *****
  • Posts: 3215
  • Joined: 01 Aug 2007
  • Loc: Blue Island Illinois

Posted 17 January 2013 - 05:07 PM

I thought curved spiders were THE thing until I noticed the light-fog, after that I switched to 4 vane and never went back. Some targets(my favorites)like Jupiter and at the time, Mars, can be very distracting but I think curved vanes do help on dim diffuse objects, like planetary nebula and structure in galaxies.

#16 Brian Albin

Brian Albin

    Seeker

  • -----
  • Posts: 869
  • Joined: 22 Aug 2006
  • Loc: Western Oregon

Posted 18 January 2013 - 12:10 PM

I think I might like combination vanes - curved out where the energy is slight near the perimeter of the telescope tube, and straight in next to the mirror to avoid enlarging the apparent diameter of the star.

#17 Jarad

Jarad

    Fly Me to the Moon

  • *****
  • Posts: 6354
  • Joined: 28 Apr 2003
  • Loc: Atlanta, GA

Posted 18 January 2013 - 02:50 PM

I'm not sure it quite works that way - the transformation isnt' linear, it's a fourier transform. I think what you are describing would produce a spike with a faint fan.

Test the shapes you are considering by taping a cutout of the shape in front a small refractor, and see how it looks.

Jarad

#18 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 3985
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 18 January 2013 - 03:40 PM

If you can't see it, how big a devil can it be?


It's not whether you can see it, it's what you can't see because of it.

It obscures detail by lowering contrast. A rough analogy - I have 20/10 vision in my left eye, and at my last checkup it shows no astigmatism either, just undercorrection (I'm near sighted). Now what I can see with that eye, compared to what I could see with "ordinary" vision, is significant. Suppose I just defocus the scope a bit everytime I use it, and get used to that. Maybe after a while I wouldn't "see" it anymore. But neither would I see as well or as much as I could see if the focus were perfect.

Anything that reduces contrast transfer reduces ultimate detail. Why would I want to install a permanent defocus on my scope? Same with any other device that reduces contrast... :shrug:

Best,
Mark

#19 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1892
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 18 January 2013 - 04:22 PM

If you can't see it, how big a devil can it be?


It's not whether you can see it, it's what you can't see because of it.

It obscures detail by lowering contrast. A rough analogy - I have 20/10 vision in my left eye, and at my last checkup it shows no astigmatism either, just undercorrection (I'm near sighted). Now what I can see with that eye, compared to what I could see with "ordinary" vision, is significant. Suppose I just defocus the scope a bit everytime I use it, and get used to that. Maybe after a while I wouldn't "see" it anymore. But neither would I see as well or as much as I could see if the focus were perfect.

Anything that reduces contrast transfer reduces ultimate detail. Why would I want to install a permanent defocus on my scope? Same with any other device that reduces contrast... :shrug:

Best,
Mark


But here is the question - is the amount of diffraction light added enough to actually obscure detail that you would otherwise able to see? There is always some amount of contrast reduction so small that the human optical system cannot perceive it.

Everything in engineering involves trade-offs.

People are pretty comfortable with the fact that an illumination drop off of less than 30% at the edge of the visual field cannot be seen, and so do not demand full illumination across it.

And there is the old argument about central obstruction - any at all reduces contrast to some degree, and the bigger the greater this contrast reduction is. But there are very well respected experts (Suiter et al) who contend that this contrast reduction is genuinely negligible if not much more than 20%.

If a curved vane is the equivalent of going from a 20% CO to a 21% CO*, and the 21% CO is normally judged acceptable, why would the vanes decision not be similarly judged?

*A relative 0.21 CO with 4 vanes having a 0.0015 thickness, and a 0.20 CO with vanes that have a profile 2.4 times larger, for example.

#20 idp

idp

    Messenger

  • -----
  • Posts: 453
  • Joined: 21 Aug 2009
  • Loc: New Haven, CT

Posted 18 January 2013 - 04:27 PM

Shouldn't curved vanes be thicker too, everything else being equal? Straight vanes work under tension, I cut mines from a tin can; they are razor-thin and perform well. No way I could made curved vanes with them, I'd have to use something thicker (which scatters more light).


Ivano

#21 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 8052
  • Joined: 08 Aug 2007
  • Loc: La Union, PI

Posted 18 January 2013 - 07:31 PM

It's not whether you can see it, it's what you can't see because of it.


Mark is spot on. Try splitting a close, unequal double star. They might be well outside the Raleigh criteria and within the light grasp of a scope, but diffraction makes them very difficult. Even if the diffraction is below the visible threshold. I am not sure if difficulty arises from flux or if the interference pattern plays a role, but close uneven pairs are difficult to split. And close unequal pairs are on the same scale as some finer, low contrast planetary detail.

#22 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 3985
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 18 January 2013 - 09:32 PM

But here is the question - is the amount of diffraction light added enough to actually obscure detail that you would otherwise able to see? There is always some amount of contrast reduction so small that the human optical system cannot perceive it


Two words - Richard Suiter. :waytogo:

If a curved vane is the equivalent of going from a 20% CO to a 21% CO*, and the 21% CO is normally judged acceptable, why would the vanes decision not be similarly judged?


Not really, these two things are not equivalent in their impact on the MTF curve, inasmuch as they present quite different obstructions to the entrance pupil. And though the diffraction from thin vanes is almost impossible to model accurately, a few experiments in the field can clear these things up. When I was first looking at the effect of CO on performance I made a number of central masks and tried them out. Ditto for spider effects... Highly recommended.

Best,
Mark

#23 Dan McConaughy

Dan McConaughy

    Ranger 4

  • -----
  • Posts: 309
  • Joined: 11 Nov 2006
  • Loc: LA

Posted 18 January 2013 - 10:24 PM

Aren't curved spider vanes longer than straight ones? Wouldn't they therefore scatter more light than straight ones?

#24 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1892
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 18 January 2013 - 11:20 PM

...
Not really, these two things are not equivalent in their impact on the MTF curve, inasmuch as they present quite different obstructions to the entrance pupil. And though the diffraction from thin vanes is almost impossible to model accurately, a few experiments in the field can clear these things up. When I was first looking at the effect of CO on performance I made a number of central masks and tried them out. Ditto for spider effects... Highly recommended....


Indeed.

I haven't been able to do such test myself yet, but I am relying in part on tests by Jay Scheuerle described here:
http://www.cloudynig...&Board=refle...

He tested a very thick 180 degree baffle and commented:

"The real surprise was the baffle-spider. The vane is exposed for a single 180° curve and I made it extra thick to simulate the area blocked by a curve that acts as a light baffle as well. Though the thickness of the profile did provide extra diffraction energy, the simplicity of the 180° curve spread it out extremely smoothly and I'd put it up there with the wire spider as the most pleasing of the views. What was interesting to note was that the diffraction seemed to spread out in rings. I only really noticed one nearer the center, like a donut, but because of the symmetry, was not distracting."

This was a test of a really huge obstruction area (something like 10 times or more than the area of even a set of curved vanes).

#25 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1892
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 18 January 2013 - 11:29 PM

They are somewhat longer and significantly thicker also, and yes they do scatter more light.

The real question (to me) is now serious a reduction in contrast this scattered light really is in practice. I encounter conflicting views, conflicting evidence offered, and it seems difficult to get to an answer that can be supported by physical analysis.

One thing about straight vanes is that they scatter light in a way that is tailor made to be readily detected by the human visual apparatus - the scattered light is concentrated in a very thin, linear region against a very dark background. The eye is extremely efficient at detecting linear, high contrast features (particularly light on dark).






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics