Jump to content


Photo

CO Madness

  • Please log in to reply
44 replies to this topic

#1 azure1961p

azure1961p

    Voyager 1

  • *****
  • Posts: 10363
  • Joined: 17 Jan 2009
  • Loc: USA

Posted 09 December 2012 - 03:54 PM

Just got the latest Sky and Tel - huge Lx600 is pictured In it with this humungous CO!! It's F/8 I'm assuming but at what cost? It's gotta be at least 50% by diameter and I wouldn't rule out 60%. I'm getting it that it's an advantage to have a faster scope for wider fields in the interest of deep sky and imaging. Still, the light in those diffraction rings has got to be bothersome on a number of fronts and lunar and planetary has to be aggrevated in terms of contrast loss .


Pete

#2 Cotts

Cotts

    Just Wondering

  • *****
  • Posts: 4827
  • Joined: 10 Oct 2005
  • Loc: Toronto, Ontario

Posted 09 December 2012 - 04:47 PM

I would think this is an Astrophotography scope so, unless you're doing crazy things with barlows and/or severely cropping the images then the somewhat extra brighter diffraction rings won't be visible at all.

It is possible that the limiting magnitude of the photos will be a bit less due to the more 'spread out' energy in the diffraction pattern, even if it is unresolved. This doesn't seem to discourage the use of RC's and other heavily obstructed scopes, sometimes very, very pricey ones and with stunning results.

I wouldn't want to compare these scopes with similar aperture Newts and Refractors on the planets, though. 'Twould be a slaughter......

Dave

#3 desertlens

desertlens

    Viking 1

  • *****
  • Posts: 839
  • Joined: 06 Dec 2010
  • Loc: 36°N 105°W

Posted 09 December 2012 - 05:07 PM

I've often thought that percentage of diameter was a bit misleading as a measure of CO. I realize that this is the conventional method but still... Area is another matter. As an example, my C6 has a diameter based CO of 36% but calculated by area of primary minus area of secondary it's only about 13%. Still, I'd agree that that Lx600 has a large CO by any standard.

#4 Eddgie

Eddgie

    Voyager 1

  • *****
  • Posts: 12702
  • Joined: 01 Feb 2006

Posted 09 December 2012 - 05:14 PM

This is an imageing system.

That does not mean that it can't be used visually, but I am not sure why anyone would want to do that really. For visual use, an obstruction this large will rob a huge amount of contrast.

That doesn't mean anything to some people though, and I am sure that somewhere someone is using one visually, but I don't think this is really how the designer intended the scope to be used.

When used for imageing, a very large obstruction has little meaningful impact. Yes, it lowers contrast, but at prime focus, the image scale is so small that in most cases, you are not looking for miute low contrast structural detail in Nebula with a scope like this.

And stars are such small points at prime focus that it really doesn't matter that the first ring is bright. It will be too small to resolve in the images.

Anyway, this is a scope design that is heavily compromised toward imageing. Not that you could not use it for general observing, but clearly that is not what the designer intended. The contrast will suffer to badly for visual observing.

#5 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 8218
  • Joined: 08 Aug 2007
  • Loc: La Union, PI

Posted 09 December 2012 - 05:41 PM

And stars are...too small to resolve in the images.


Okay, that makes sense. Didn't do the math, but I am sure that's right. All the light is captured by one pixel. Never thought of that being the reason large COs are fine for imaging. Visually, on planets, t'would be a slaughter, for sure.

Now, if I got this right, however, the central visible disc will fall of in brightness and angular diameter somewhat. This could allow a bit higher frequency (extended) objects to be resolved.

However, remember a telescope transfers contrast from the image to the focal plane. At these high frequencies (at or beyond maximum for that aperture), the contrast translated is not very high even for 100% on the target. And you need contrast between the spurious disc and the darker feature to resolve them. You could loose all advantage at some point by dimming the spurious disc too much. The resulting contrast on the focal plane could be zero or close to it. Hence, no resolution.

This condition begins to really degrade visual observation at some point despite any gains in smaller disc size. And, 50 or 60% /might/ be getting close to that point.

#6 rmollise

rmollise

    Hubble

  • *****
  • Posts: 15567
  • Joined: 06 Jul 2007

Posted 09 December 2012 - 07:00 PM

Just got the latest Sky and Tel - huge Lx600 is pictured In it with this humungous CO!! It's F/8 I'm assuming but at what cost? It's gotta be at least 50% by diameter and I wouldn't rule out 60%. I'm getting it that it's an advantage to have a faster scope for wider fields in the interest of deep sky and imaging. Still, the light in those diffraction rings has got to be bothersome on a number of fronts and lunar and planetary has to be aggrevated in terms of contrast loss .


Pete


Well, maybe. The RCXes worked pretty well on the Moon and planets from what I recall.

#7 Gord

Gord

    Surveyor 1

  • *****
  • Posts: 1686
  • Joined: 06 Jan 2004
  • Loc: Toronto, ON, Canada

Posted 09 December 2012 - 07:19 PM

This is an imageing system.

That does not mean that it can't be used visually, but I am not sure why anyone would want to do that really. For visual use, an obstruction this large will rob a huge amount of contrast.


Hey Eddgie,

I know you are very familiar with the MTF curves. For yuks, would you be able to generate a comparison set between one of these (14") and a C14?

My gut says there is going to be a noticeable difference, and I would go as far as Dave says and say these things are going to get pummelled by many other scopes for planetary duties. Of course, that isn't what they were designed for, but it is a departure from the all-around scope that Meade has traditionally sold.

Clear skies,

#8 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 8218
  • Joined: 08 Aug 2007
  • Loc: La Union, PI

Posted 09 December 2012 - 08:09 PM

Scroll down...

http://www.damianpea.../simulation.htm

More...

http://www.brayebroo...orum/c-o's.html

#9 azure1961p

azure1961p

    Voyager 1

  • *****
  • Posts: 10363
  • Joined: 17 Jan 2009
  • Loc: USA

Posted 09 December 2012 - 10:22 PM

I'm familiar with Peaches simulations on contrast versus CO and for the life of me I really believe he's greatly played it down and the results are probably twice as apparent as his sims let on. I've hot a lot of respect for the guy but I can't agree with his finds at all. Graphs fine but sims are a bit flattering of CO.

Pete

#10 azure1961p

azure1961p

    Voyager 1

  • *****
  • Posts: 10363
  • Joined: 17 Jan 2009
  • Loc: USA

Posted 09 December 2012 - 10:27 PM

I've often thought that percentage of diameter was a bit misleading as a measure of CO. I realize that this is the conventional method but still... Area is another matter. As an example, my C6 has a diameter based CO of 36% but calculated by area of primary minus area of secondary it's only about 13%. Still, I'd agree that that Lx600 has a large CO by any standard.


I don't think it matters so much which is used so long as its the common measure from one scope to another. I've got to believe tho the lx600 has the CO diameter of at least a medium sized refractors aperture .

Pete

#11 Alph

Alph

    Surveyor 1

  • -----
  • Posts: 1759
  • Joined: 23 Nov 2006
  • Loc: Melmac

Posted 10 December 2012 - 12:26 AM

probably twice as apparent as his sims let on.


I am interested to see how you would quantify it. Apparently you have a better way of measuring it. Please explain.

#12 hottr6

hottr6

    Vanguard

  • *****
  • Posts: 2406
  • Joined: 28 Jun 2009
  • Loc: 7,500', Magdalena Mtns, NM

Posted 10 December 2012 - 11:31 AM

I've often thought that percentage of diameter was a bit misleading as a measure of CO. I realize that this is the conventional method but still... Area is another matter. As an example, my C6 has a diameter based CO of 36% but calculated by area of primary minus area of secondary it's only about 13%.

I'd have to agree. Measuring CO linearly is pretty, I'm going to have to say it, stupid. Lineal measurements of square and circular CO may indicate the same, when the differential impact of these two obstructions on photon count will be enormous.

#13 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 8218
  • Joined: 08 Aug 2007
  • Loc: La Union, PI

Posted 10 December 2012 - 12:07 PM

As I understand it, the biggest impact is on contrast rather than the square of some radius. There is exponentially more area of aperture in relation to the CO, that photon count - while important - is not the primary concern.

When a large CO ~0.5D (or whatever percentage of diameter) reduces the spurious disc by nearly half and sends the other half into the rings - most of that into the first ring - well, good luck resolving a low contrast feature anywhere near it. Not only has the contrast fallen on the brighter surface, but the double whammy is light from the rings reducing contrast against nearby darker features.

Small light and dark features more rapidly blur into a more uniform shade of gray. A white picket fence would still look fine, though, from down the street. That's the importance of the MTF, diffraction describes your scope more accurately than light gathering power.

#14 Eddgie

Eddgie

    Voyager 1

  • *****
  • Posts: 12702
  • Joined: 01 Feb 2006

Posted 10 December 2012 - 05:11 PM

Norme has covered it pretty well.

The clear aperture method though (the diameter of the primary minus the diameter of the secondary obstruction = clear aperture) is only an approximation.

If fails when the obstruction is very small or very large.

And it ignores the fact that when an obstruction is present, at the very very highest spatial frequencies, the contrast transfer in an obstructed scope is almost always better than in an unobstruced scope. That's right. At the limits of angular resolving power, the obstruction improves contrast transfer vs an unobstructed scope.

But the damage cased by the obstruction is not really much to do with shading of the mirror reducing light transmission. Reduced transmission has no effect on cotrast transfer at all.

All of the damage is done because light that would normally go into the Airy Disk in an unobstruced system is now going in to the diffraction rings around the point.

84% of the light in an unobstructed scope goes into the Airy Disk, 7% into the first diffraction ring, 3% into the second ring, and the rest in the outer rings.

By comparison, a scope with a very large obstruction, rather than putting 91% of the light into the Air Disk and first only puts about 70% of the energy into the Airy Disk and first ring.

Now remember, the Airy disk may be small, but the first ring goes out to twice the diameter of the Airy disk, and the second ring goes out to three times the diameter of the Airy Disk.

And this is what kills the contrast. Low contrast details of a size that are two or three diameters of the Airy disk will be washed out but the light that is being thrown from the point on the image that created it.

And if the obstruction is very large, it can extend to four and five rings, which is why we see such a huge droop in the MTF plots around .5 in MTF plots. Small details simply get drowned out.

Anyway, the "Clear Aperture" forumla offers an approximation for contrast transfer, but is not at all accurate when the contrast is for detail that is very fine.

For example if you have two equal magnitude doubles, an obstructed system might show them better seperated than a perfect unobstructed system because the Airy Disk will appear smaller, if the stars are seperated by the widty of the gap between the edge of the Airy Disk and the first ring, they will actually appear to be seperated by a wider space in the obstructed instrument. For this reason, we say that the contrast transfer is better. The space between the stars appears wider, and hence, blacker.

The damage is very real, and the MTF plots do a very exact job of describing how contrast transfer will differ between two different scope.

Anyone that chooses to ignore MTF plots or treat them as witch-doctor stuff is simply ignoring the physics of how telescopes work. It is 100% reliable. If you accuratly model a telescope, an MTF plot describes exactly how it will behave.

That is why professional optical engineers talk in terms of MTF and encircled energy. These describe how an instrument performs with great clarity.

#15 desertlens

desertlens

    Viking 1

  • *****
  • Posts: 839
  • Joined: 06 Dec 2010
  • Loc: 36°N 105°W

Posted 10 December 2012 - 05:51 PM

Anyway, the "Clear Aperture" formula offers an approximation for contrast transfer, but is not at all accurate when the contrast is for detail that is very fine.



Thanks Eddgie. I knew there was a contrast issue and the application of MTF in this case explains that aspect to me.

#16 hottr6

hottr6

    Vanguard

  • *****
  • Posts: 2406
  • Joined: 28 Jun 2009
  • Loc: 7,500', Magdalena Mtns, NM

Posted 10 December 2012 - 06:31 PM

Thanks Norme and Eddgie, that's a good explanation.

So if CO has such a "profound" effect on image quality, it seems that we should also include the area of the spider vanes (Mak-Gregs and Schmidt-Cass' need not apply). Looking at the thickness of the vanes on some current-production visual and AP Newts, this will have a considerable effect.

The reason why I think I am less than happy with CO as a "specification", is that it has been too casually thrown around by well-meaning users. This casual use ignores vanes, shape of the secondary and/or holder, etc.

Am I making sense, or just being a curmudgeon?

#17 Starhawk

Starhawk

    Space Ranger

  • *****
  • Posts: 5553
  • Joined: 16 Sep 2008
  • Loc: Tucson, Arizona

Posted 10 December 2012 - 06:39 PM

Curmudgeon.

The vanes cause spikes on stars- more out there on this than there is on central obstruction effects.

-Rich

#18 hottr6

hottr6

    Vanguard

  • *****
  • Posts: 2406
  • Joined: 28 Jun 2009
  • Loc: 7,500', Magdalena Mtns, NM

Posted 10 December 2012 - 07:17 PM

The vanes cause spikes on stars- more out there on this than there is on central obstruction effects.

With a concomitant effect on contrast and image quality, similar to CO effects.

A good article here, though a little scholarly:
http://www.telescope....net/spider.htm

I'm not a curmudgeon.... yet! :grin:

#19 azure1961p

azure1961p

    Voyager 1

  • *****
  • Posts: 10363
  • Joined: 17 Jan 2009
  • Loc: USA

Posted 10 December 2012 - 07:26 PM

probably twice as apparent as his sims let on.


I am interested to see how you would quantify it. Apparently you have a better way of measuring it. Please explain.


"Alf"

Quantify it? Oh that's simple I'd say the effects are twice as apparent as the sim photos let on. And yes I have a better way of measuring it - you put down the simulation and look through the eyepiece.


I'm onboard with the graphs and such but something's lost in the translation in those sim pix. Too, Jupiter would have been a better sim photo as the contrasts are in greater variety. In the end though that would fail too. You simply can't photoshop a soft filter and call it that simple. Their are nuances in contrasts with varying to no CO that simple slide rule image softening won't address. A lot of things change a little and cumulatively it's more involved than the simplistic sim pix. Too, a nifty program to apply direct contrast attenuation per the numbers is no solution either . It is not an absolute and its as subjective as the human being developing it.

You want to do it right? Get a 6" apo and add varying central obstruction discs. Image it and process it fairly raw as wavelets applied can skew the results. Line up all the raw pix and there's a fair representation of CO to no CO differences. Attenuating fixed photos is playing. Just be careful the seeing is stable enough to allow even handed sampling.

Pete

#20 Eddgie

Eddgie

    Voyager 1

  • *****
  • Posts: 12702
  • Joined: 01 Feb 2006

Posted 10 December 2012 - 09:50 PM

So if CO has such a "profound" effect on image quality



I didn't say that at all.

I said a very large CO has a very big effect. My response was meant in the context of the scope the OP is talking about.

The effects of a 20% obstruction are so small that most observers would struggle to see it.

Even the 33% SCT obstruction does not degrade the image to the point that it is glaringly obvious, but in a direct comparsion by a good observer, it will be enough to see.

Once the obstruction gets over about 40% though, the damage becomes more apparent, and this is why this has been considered the cut-off for the maximum size allowed for visual use. Below this and the damage is not so great as to make the scope undesirable to use. As any C9.25 owner. The obstruction is 38%.

In fact, most telescopes sold with an obstruction larger than 40% are usually sold as "Imaging" telescopes.

Anyway, it is all relative to the size. A small obstruction does almost no damage, and a very large obstruction (50%) is not really all that good for high resolution observing (though it works well for most other kinds of observing).

#21 Cotts

Cotts

    Just Wondering

  • *****
  • Posts: 4827
  • Joined: 10 Oct 2005
  • Loc: Toronto, Ontario

Posted 10 December 2012 - 10:59 PM

Norme has covered it pretty well.

The clear aperture method though (the diameter of the primary minus the diameter of the secondary obstruction = clear aperture) is only an approximation.

If fails when the obstruction is very small or very large.

And it ignores the fact that when an obstruction is present, at the very very highest spatial frequencies, the contrast transfer in an obstructed scope is almost always better than in an unobstruced scope. That's right. At the limits of angular resolving power, the obstruction improves contrast transfer vs an unobstructed scope.

But the damage cased by the obstruction is not really much to do with shading of the mirror reducing light transmission. Reduced transmission has no effect on cotrast transfer at all.

All of the damage is done because light that would normally go into the Airy Disk in an unobstruced system is now going in to the diffraction rings around the point.

84% of the light in an unobstructed scope goes into the Airy Disk, 7% into the first diffraction ring, 3% into the second ring, and the rest in the outer rings.

By comparison, a scope with a very large obstruction, rather than putting 91% of the light into the Air Disk and first only puts about 70% of the energy into the Airy Disk and first ring.

Now remember, the Airy disk may be small, but the first ring goes out to twice the diameter of the Airy disk, and the second ring goes out to three times the diameter of the Airy Disk.

And this is what kills the contrast. Low contrast details of a size that are two or three diameters of the Airy disk will be washed out but the light that is being thrown from the point on the image that created it.

And if the obstruction is very large, it can extend to four and five rings, which is why we see such a huge droop in the MTF plots around .5 in MTF plots. Small details simply get drowned out.

Anyway, the "Clear Aperture" forumla offers an approximation for contrast transfer, but is not at all accurate when the contrast is for detail that is very fine.

For example if you have two equal magnitude doubles, an obstructed system might show them better seperated than a perfect unobstructed system because the Airy Disk will appear smaller, if the stars are seperated by the widty of the gap between the edge of the Airy Disk and the first ring, they will actually appear to be seperated by a wider space in the obstructed instrument. For this reason, we say that the contrast transfer is better. The space between the stars appears wider, and hence, blacker.

The damage is very real, and the MTF plots do a very exact job of describing how contrast transfer will differ between two different scope.

Anyone that chooses to ignore MTF plots or treat them as witch-doctor stuff is simply ignoring the physics of how telescopes work. It is 100% reliable. If you accuratly model a telescope, an MTF plot describes exactly how it will behave.

That is why professional optical engineers talk in terms of MTF and encircled energy. These describe how an instrument performs with great clarity.


100% right on, Eddgie.

For high resolution visual observing at medium to high powers with low contrast targets.

For prime focus photography at short to modest focal lengths the CO has virtually zero affect on contrast.

For planetary photography à la Damien Peach the CO will have a detrimental effect but it seems that post processing can negate these efeects.

Dave

#22 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 8218
  • Joined: 08 Aug 2007
  • Loc: La Union, PI

Posted 10 December 2012 - 11:08 PM

Wow, a few replies to read, including Eddgie's. I will read them in a second, so if I repeated anything, sorry.

What is, Encircled Energy? Actually, it's like an "effective" Strehl due to CO's light scattering. A scope that puts 84% light into the Airy disc has a Strehl of 100% and encircled energy of 100%. A perfect optic puts 100% (all) of the expected maximum of 84% available light into the Airy disc. Therefore, EE is identical to Strehl in unobstructed apertures.

For example, a scope with an aberrant Strehl of 0.94 puts 0.94 * 84% = 79% of the light into the Airy disc due to aberrations alone, compared to 0.83 * 84% for an unobstructed aperture with a Strehl of 0.99.

But EE accounts for the energy distribution caused by aberration and the CO (versus Strehl which only accounts for aberrations.) EE falls a bit when accounting for the CO as light is further scattered by the obstruction. This fall off can be approximated by 2(oD)^2.

For a 30% CO, that is 2*(0.3^2) = .18 approximate drop in light from a maximum of 0.84 for a perfect unobstructed optic. So, a perfect(?) scope with a 30% CO puts a mere 0.66 of the light in the Airy disc compared to 0.84 of the perfect unobstructed scope. The energy encircled in the Airy disc radius would be (actual)/(perfect) or 66/84 = 79% of 100% perfect.

Edit: This is why a .3D CO gives an equivalent MTF as 1/4 P-V LSA, both produce an Airy disc with about 80% of the light there. The resulting MTF curves are very similar. The difference is very small and probably cannot be seen at the eyepiece.

Encircled energy for my scope with 28% CO (42mm/150mm = .28) is 0.68/0.84 = 0.81 light in the Airy disc (less than it's estimated 0.94 Strehl) compared to 0.83/0.84 = .99 for a refractor (identical to it's Strehl.) In other words, due to the CO alone, my spurious disc is ~18% dimmer. That light has to go to the rings. Factor in some SA and that light falls to about 79% for the combined effect according to Suiter's charts.

BUT! This is for equal apertures. That 20% can be made up by considering a smaller aperture unobstructed scope. This is why a larger obstructed aperture can match the performance of a smaller unobstructed scope in terms of contrast across the whole MTF.

Edit: And larger aperture can more than make up for photon loss due to the obstruction.

So, really the 66% rule holds for obstructions of 33%, a 150mm obstructed scope can match the contrast transfer of a .66 * 150 = 99mm (~4") refractor. In theory, and closely in practice with good optics. With smaller obstructions closer to 25%, the ratio grows to closer to 85% of the aperture. In this case, a (150mm * .85) can equal a 127mm refractor in terms of contrast. But this is at moderate to lower frequencies, say from 2 to 4" arc for 5" scopes (not doing the math, just an example.) At higher frequencies near the size of the Airy disc at 1" arc, aperture wins (resolution and light gathering) and a CO actually gives it more of a boost in fine resolution reducing the spurious disc ~10% in some cases.

A CO of 0.5D is much worse, it looses 1.9 * 0.25 = 0.48 of the 0.84 possible and that means (0.84 - 0.48 = 0.36), or 0.36/.84 = 43% of the maximum light is in the Airy disc due to the obstruction - about half of that for an equal diameter unobstructed aperture at 84%. The rest is in the rings. Edit: According to some MTF charts, this amount of light loss closely resembles 1/2 P-V LSA. So, if some simulations show Saturn as being pretty sharp, well 1/2 PV LSA is enough for most folks to recognize degradation in an image (Suiter says this visual limit is about 1/3 P-V.)

But, that percentage of encircled energy INCREASES with radius, if you include the first bright ring, for example. So at the size of the angular diameter captured by a pixel it's quite possible nearly all the light is encircled within that pixel. Therefore the CO has little effect.

Edit: Thats a lot of numbers being floated around in theory, but it practice with all other variables held constant actual observing can closely approximate theory. That's the importance of MTF, it describes diffraction pretty well, in theory and in practice. Even average seeing changes everything. But in those good moments when a good scope approaches theoretical MTF performance the views can be very stunning. That's why we wait for those calm moments.

Man, my posts are getting longer than Eddgie's. :)

#23 azure1961p

azure1961p

    Voyager 1

  • *****
  • Posts: 10363
  • Joined: 17 Jan 2009
  • Loc: USA

Posted 11 December 2012 - 12:09 AM

Yeah but...

(And this is a good thread)

The whole MTF thing seems to breakdown for progressively larger aperture as more and more detrimental effects of seeing are resolved. In the vacuum of space MTF would seem to rein uncontested but neath average sky's it would appear it becomes more theory than reality the larger the scope is. It'd be interesting if some kind of qualifier or adjustment could be made to account for this in formula. With sky's that generally but nt always tap out at 500x how could a 20" reflector have an MTF that could be considered practical in defining its performance of its prematurely capped?

I'd like to learn more about MTF but in practice it would seem to incur detrimental effects not always accounted for.

Pete

#24 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 8218
  • Joined: 08 Aug 2007
  • Loc: La Union, PI

Posted 11 December 2012 - 12:21 AM

So if CO has such a "profound" effect on image quality



I didn't say that at all.

I said a very large CO has a very big effect. My response was meant in the context of the scope the OP is talking about.

The effects of a 20% obstruction are so small that most observers would struggle to see it.

Even the 33% SCT obstruction does not degrade the image to the point that it is glaringly obvious, but in a direct comparsion by a good observer, it will be enough to see.

Once the obstruction gets over about 40% though, the damage becomes more apparent, and this is why this has been considered the cut-off for the maximum size allowed for visual use.


Correct, the CO does not form the image and induce aberration. It induces diffraction. Diffraction has no effect on the wavefront correction, only in the pattern of diffraction as seen. The latter is responsible for sending light to the rings, so the wavefront correction remains quite unaffected. This is why a scope with a 50% CO can still have a Strehl of 95%.

However, the resulting image can be blurred a bit due to light diffracted into the rings. This happens noticeably below about 1/4 P-V LSA and has the equivalent MTF of a ~33% CO, as Eddgie says. L/4 LSA is still at least good according to Raleigh, and better than many can detect any degradation. Such degradation does not begin to happen until about 1/3 P-V, according to Suiter, and that would be about 40% CO /equivalent/. A 50% CO approximates, roughly, 1/2 P-V LSA across most of the MTF.

That's pretty bad, visually, but due to diffraction not the wavefront. You're Airy disc "image quality" will still be quite clean with a 95% Strehl. And a cleaner, smaller spurious disc means better high frequency resolution compared to the blurred disc image with increased aberration. At this point, errant light rays are reducing contrast across the in-focus diffraction pattern where the CO does not, as I understand it. So, the CO and wavefront MTF effects are not quite identical.

I believe that's at least close to being correct.

The whole MTF thing seems to breakdown for progressively larger aperture as more and more detrimental effects of seeing are resolved.


The /actual/ MTF performance of the scope under the stars at night incorporates seeing, even if the "perfect" theoretical charts do not. But, you can find charts that plot seeing affects on the MTF. When comparing anything, you have to keep as many variables constant as possible. So, we assume no seeing. However, the MTF takes no prisoners. Seeing does affect it and differently with aperture, if memory serves.

So, if the MTF 'breaks down," it breaks down because of seeing to become a new, degraded MTF curve in the real world. But, full blown contrast transfer is still happening. In any case, a scope is always performing to some level of it's MTF. Contrast transfer is precisely the function of a telescope - that's how it shows us things.

The actual, real world, under the stars MTF will not resemble the nice looking charts we see all the time. It will include seeing and other effects, and they can get ugly.

#25 freestar8n

freestar8n

    Soyuz

  • *****
  • Posts: 3923
  • Joined: 12 Oct 2007

Posted 11 December 2012 - 03:45 AM

MTF is useful but it is not sacrosanct in characterizing the performance of an optical system, or in comparing one to another. The main things it lacks are phase and an accurate description of the coherence of the source and the coherence properties of the imaging system. For the case of visual or planetary work at high magnfication in the presence of atmospheric turbulence, it ignores the time variation of the distortion and the statistics of forming a "lucky" image. Finally, for imaging with a real detector, the loss of light does impact the realized contrast because the SNR of a given spatial frequency on the detector will be reduced if the amplitude goes down - even if the "contrast" at that frequency is theoretically 100%.

This stuff is not typically addressed in simplified descriptions of telescopes, but it is or should be in a text dedicated to Fourier or statistical optics. Born and Wolf go into great detail about the Optical Transfer Function (which includes phase) and discussions of coherence - and never even mentions MTF. MTF is an engineering convenience for describing the general imaging properties of a system, but shouldn't be used as "proof" of how bad one system is compared to another. And simulations of imaging in the presence of turbulence are particularly misleading because they greatly oversimplify the process - particularly when going by glimpses of clarity or lucky imaging, both of which involve the statistics of the turbulence.

A lot of this is analogous to judging speakers based on plots of their frequency behavior. There is no information of phase, which can be very important in making either music or an image ugly - even if the frequency transmission is 100%. Also, it depends on the type of music being listened to - just as the effectiveness of an optical system depends on what it is imaging and the nature of the measurement.

The CO is large in this sct because it is faster at f/8, and it's possible the primary is slower than the normal f/2 - but I'm not sure. It would make sense to make the primary slower and incur a larger secondary because it would help flatten the field. This would be ok for deep sky imaging, but there would be a loss of light in the center, although less drop off due to vignetting. For visual work the main problem would be the hole in the exit pupil at low powers. For planetary and double star work, the larger CO would not be good, but it would be hard to characterize just how bad it would be. And for viewing or imaging a small and detailed object like Io in white light, it would be particularly challenging to assess its impact.

Frank






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics