Large mirrors can't achieve theoretical resolution, due to surface flaws?
Neil B.
wide:
http://www.tmt.org/
A 30-meter telescope would be way cool, but something about the supposed
capabilities of such mammoth apertures has long bothered me. I question
how ever-larger apertures can continue to provide ever better resolution
with e.g. visible light - if they are simply accurate to within a given
standard like 1/8 wave. And this one would be a MMT, but the same
argument below applies to portions and to single piece versions too.
In theory, in the ideal, I see why ever larger apertures confer
ever-better resolution. But consider that as a mirror gets larger, the
tiny variations in orientation of portions of the mirror should
eventually cause errors greater than the ever-decreasing diffraction
spot. IOW, consider e.g. that little portions of the 30-meter mirror
have roughly random (how "random" isn't the main issue) "orientation
errors" of around 0.01 arc second. So if this mirror was idealized and
we used hard UV at 50 nm, the theoretical resolution should be about
0.0003 arcsec. But application of geometric optics to the surface
irregularities should mean, a circle of confusion 0.01 arcsec diameter.
IOW, resolution is limited by surface traits and can't be endlessly
improved with ever-shorter wavelengths. The flip side: ever larger
apertures should fail to provide ever-increasing resolution with the
*same* wavelengths.
So let's switch to visible light. The theoretical resolution should be
about 0.003 arcsec. Can that somehow be even better than the UV
resolution, because of some global treatment of the larger waves? I
find that hard to believe. The centers of where diffracted portions of
light are sent would still be directed to slightly different points at
the collection plane, per the 0.01 arcsec variations. I made up the
0.01 arcsec figure of orientation variation, but the same principle
should apply in the real world.
Maybe I misunderstand the implications of applying a "1/8 wave" etc.
standard to different apertures, but I can't imagine that as
mirrors/lenses were made larger we'd be able to proportionately reduce
the magnitude of orientation flaws. It would be absurd. We're just
making a bigger piece of the same stuff that someone must polish locally
etc. (Also, the inherent physical grain etc.)
Any scoop, especially on "who asked this first" etc? tx
Kai-Martin Knaak
> if they are simply accurate to within a given standard like 1/8 wave.
They are a lot better.
> But consider that as a mirror gets larger, the tiny variations
> in orientation of portions of the mirror should eventually cause
> errors greater than the ever-decreasing diffraction spot.
The trick is not to allow "tiny variations" on a small scale. This is
called "polishing". The result is a surface that is smooth over the whole
range of lengths from the diameter down to the size of the wavelength.
> But application of geometric optics to the surface irregularities should
> mean, a circle of confusion 0.01 arcsec diameter.
You better apply wave optics here.
BTW: The reason for large mirrors is not only resolution but also the
amount of light gathered.
---<(kaimartin)>---
--
Kai-Martin Knaak
Öffentlicher PGP-Schlüssel:
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x6C0B9F53
Neil B.
"Kai-Martin Knaak" <k...@lilalaser.de> wrote in message
news:pan.2009.07...@lilalaser.de...
> On Thu, 23 Jul 2009 12:54:41 -0400, Neil B. wrote:
>
>> if they are simply accurate to within a given standard like 1/8 wave.
>
> They are a lot better.
>
particular level of accuracy.
>
>> But consider that as a mirror gets larger, the tiny variations
>> in orientation of portions of the mirror should eventually cause
>> errors greater than the ever-decreasing diffraction spot.
>
> The trick is not to allow "tiny variations" on a small scale. This is
> called "polishing". The result is a surface that is smooth over the
> whole
> range of lengths from the diameter down to the size of the wavelength.
>
prevent "tiny variations" since nothing is perfect. It's just a matter
of how far we push the given surface standards. Given any particular
level of tiny variations, the geometric effect of variation in directing
the average pencil of light will eventually outgrow the size of the
ideal diffraction spot.
>
>> But application of geometric optics to the surface irregularities
>> should
>> mean, a circle of confusion 0.01 arcsec diameter.
>
> You better apply wave optics here.
I question whether that is good enough. There is still a variation in
the direction the light is sent, from various parts of the mirror. The
waves still have a propagation vector that must correspond to the
orientation at various places (if the variations occur at distances
further from each other than the wavelength.) Just imagine a mirror
that you would say worked fine. OK, if I tilted part of that by 0.01
arcsec, and then more, and then more .... how could it not be a
continuous shifting of some part of the concentrated spot? Then
multiply that by various portions.
IOW we are not talking about effectively resolving features on the
mirror closer together than a certain distance. Instead, it's about the
directed nature of the wave, given angular uncertainty in the
orientation of portions at various scales. Like I said, given a certain
orientation standard like "0.01 arcsec" (or whatever it is), the sending
of PVs to slightly different spots should eventually overcome the ideal
resolution as the mirror gets bigger. But if you are right, then we
really could get better resolution in some cases from using longer-wave
light! (Because, in the short-wave case, there is no possible excuse
that wave properties could compensate for the unevenness.)
Androcles
"Neil B." <neil_...@caloricmail.com> wrote in message
news:jeSdnYy0j8BACPXX...@posted.widowmaker...
http://tinyurl.com/m24thr
Phil Hobbs
Even ordinary 1-metre class telescopes fail to achieve the diffraction
limit, due to atmospheric irregularities. The big news in astronomical
instruments since Hubble has been adaptive optics, where the waveform
errors are sensed and corrected in real time.
The same systems can correct for errors that are static (mirror figure)
or slowly varying (e.g. sag under gravity as the telescope tracks).
Cheers
Phil Hobbs
--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net
Marco
each ~1m in size. Each is then aligned to form the larger mirror surface.
So when polishing, you are always dealing with a mirror of ~1m not 30m.
Marco
UCO Lick Observatory
Laboratory for Adaptive Optics
"Phil Hobbs" <pcdhSpamM...@electrooptical.net> wrote in message
news:4v-dnffVmINuOvXX...@supernews.com...
Uncle Al
>
> There are plans to build a telescope with a mirror that is thirty meters
> wide:
> http://www.tmt.org/
> A 30-meter telescope would be way cool, but something about the supposed
> capabilities of such mammoth apertures has long bothered me. I question
> how ever-larger apertures can continue to provide ever better resolution
> with e.g. visible light - if they are simply accurate to within a given
> standard like 1/8 wave. And this one would be a MMT, but the same
> argument below applies to portions and to single piece versions too.
>
> In theory, in the ideal, I see why ever larger apertures confer
> ever-better resolution. But consider that as a mirror gets larger, the
> tiny variations in orientation of portions of the mirror should
> eventually cause errors greater than the ever-decreasing diffraction
> spot.
Adaptive optics for all large mirrors - all the summed wavefront
errors are compensated all at once. This also allows for a much less
massive mirror segment since you want it to flex a very tiny bit on
command.
--
Uncle Al
http://www.mazepath.com/uncleal/
(Toxic URL! Unsafe for children and most mammals)
http://www.mazepath.com/uncleal/lajos.htm#a2
Bob May
is specified at a particular wavelength. Normally this is in the green
(500nm approdimately). This means that the mirror won't be anywhere near a
wave accurate at the UV. Also, the Airy Disc (circle of confusion) is also
dependent upon the wavelength specified.
Another item of extreme interest to the astronomers is the actual light
gathering of the telescope. Since a lot of work is done with spectrographs,
you really don't need a superaccurate surface to work with as the fiber
optic end is of a fairly finite size, the desire for more light is a lot
more inportant so as to be able to take a spectrum in an acceptable amount
of time.
--
Bob May
rmay at nethere.com
http: slash /nav.to slash bobmay
http: slash /bobmay dot astronomy.net
John Stafford
smooth out infinitudes in software.
Kai-Martin Knaak
> The way to make a larger mirror:
> http://tinyurl.com/m24thr
Close. This how the ESO does it in Paranal:
http://en.wikipedia.org/wiki/Very_Large_Telescope
--> four 8m mirrors, optically linked for better resolution.
Neil B.
"Marco" <nu...@null.net> wrote in message news:4a68c04a$1@darkstar...
care about whether we can correct for the surface irregularities except
as a practical hope we can get pretty pictures. The point of the post is
the theoretically interesting question, of whether visible resolution of
a single large mirror, as is; would start to be limited by surface
irregularities as the aperture grew larger.
Benj
> Thanks, but you miss the point (in effect, at least.) You cannot
> prevent "tiny variations" since nothing is perfect. It's just a matter
> of how far we push the given surface standards. Given any particular
> level of tiny variations, the geometric effect of variation in directing
> the average pencil of light will eventually outgrow the size of the
> ideal diffraction spot.
No, tis you who miss the point. First of all get geometric optics out
of your pointy little head! Your ideas simply do no compute! And the
mirror IS perfect. If it is a small one it is polished to a fraction
of a wavelength AT THE WAVELENGTH it is intended to be used so it IS
accurate! And it DOES have resolving power close to theoretical (but
not through the atmosphere obviously).
When mirrors are made larger the very weight, temperture expansion,
and changing forces cause it to deflect and lose accuracy even as you
move it from pointing at one part of the sky to another. So for a long
time telescopes made with mirror polished into perfect figure had a
size limit. But the new idea was the "rubber mirror". Basically you
build a mirror as good as you can and then use a series of electronic
jacks to bend it back where it belongs when it sags. Using this idea
one can actually create imaging telescopes out of multiple mirrors
that are all jacked into perfect alignment to a fraction of a
wavelength. [Which, let me note is NOT what is going on in Andro's
boiler picture which as usual has absolutely nothing to do with what
we are talking about here. ]
To elaborate, the relationship between the mirror and the far field
resolving power is as in any antenna namely where the angular pattern
in the far field is the Fourier transform of the light distribution
across the aperture. If for example the mirror were square, the far
field angular pattern that is used to resolve point sources would be
the familiar sin x/x function. If you have multiple mirrors then they
act as classic arrays. The advantage to an array is the that you get
angular patterns related to he spacing of the elements rather than the
size of a single mirror. And the is obviously more because it is
possible to adjust the far field pattern to make it sharper by
modulating the intensity on the receiving miror. And the ultimate is
the "synthetic aperture" antenna where gigantic equivalent mirrors are
created by moving smaller ones. The telescope guys can tell you more,
but I'd guess that only the miltary spooks are using that technology
at optical wavelengths presently. For Sonor it's SOP.
The bottom line is that you are SO far behing state of the art, that
it's hard to tell you where to begin to catch up.
Helpful person
> "Marco" <n...@null.net> wrote in messagenews:4a68c04a$1@darkstar...
> > Additionally, large telescopes >8m are segmented and made of many
> > mirrors each ~1m in size. Each is then aligned to form the larger
> > mirror surface. So when polishing, you are always dealing with a
> > mirror of ~1m not 30m.
>
> > Marco
> > UCO Lick Observatory
> > Laboratory for Adaptive Optics
>
>
> - Show quoted text -
The answer is that the frequency response of the mirror is a function
of the autocorrelation of the pupil function. The pupil function is
the wavefront error of the converging (imaging) light. If a larger
mirror is made to the same specification in terms of wavefront errors
then the pupil fuction will remain constant.
Androcles
>
>> The way to make a larger mirror:
>> http://tinyurl.com/m24thr
>
> Close. This how the ESO does it in Paranal:
> http://en.wikipedia.org/wiki/Very_Large_Telescope
> --> four 8m mirrors, optically linked for better resolution.
>
The difference is the focal point is shared as I've shown it, whereas at
Paranal each individual dish has its own focal point and is used for radio
wavelengths. Don't be fooled by "optically linked", fibre optics are used
everywhere these days. My phone is "optically linked" in journalist speak.
Androcles
"Benj" <bja...@iwaynet.net> wrote in message
news:ba360ae7-f0dc-4217...@c1g2000yqi.googlegroups.com...
You wouldn't know a mousetrap from a dog's jaw if it bit you on
the arse, Jocaby.
http://en.wikipedia.org/wiki/File:USA.NM.VeryLargeArray.02.jpg
Frogwatch
wrote:
> "Benj" <bjac...@iwaynet.net> wrote in message
I think what he is talking about is surface roughness. Although the
average overall surface may be accurate to within 1/8 wave, it has
some overall surface roughness. Roughness on long spatial dimensions
becomes a variation in figure. Roughness on a much smaller scale is a
problem EVEN FOR SPATIAL WAVELENGHTS SMALLER THAN THE DESIGN photon
wavelength of the mirror. That is, this micro-roughness which may be
on the order of a few angstroms in height will scatter light. The
degree of scatter (fractions of photons removed from the desired beam)
is given roughly by the Debye -Waller factor. For visible wavelengths
it is ussually not a problem because such micro-roughness can be
reduced to below a few nanometers in height. It CAN be a problem for
very sensitive measurements where it reduces overall contrast.
Mirrors for IR military space based sensors are often superpolished to
get roughness of roughly 5 angstroms rms. There is an equation that
escapes me right now for the angle into which half of the scattefred
photons will be scattered inside of. I do know this equation is in a
paper by Susini et. al in Optical Engineering. The paper is about x-
ray optics where such roughness is a major problem.
Androcles
"Frogwatch" <dbo...@mindspring.com> wrote in message
news:b6722df7-f955-4b60...@b14g2000yqd.googlegroups.com...
========================================
Jokaby still wouldn't know a mousetrap from a dog's jaw if it bit
him on the arse.
The way to build a large mirror is a collection of small mirrors
individually actuated and computer controlled.
See: http://tinyurl.com/lvh6dd
Neil B.
"Frogwatch" <dbo...@mindspring.com> wrote in message
news:b6722df7-f955-4b60...@b14g2000yqd.googlegroups.com...
...
Thanks, but no I don't mean "roughness". I mean, that a "patch"
(macroscopic section, like cm scale) of the mirror one place is e.g.
0.01 arcsecond out of alignment with another patch and so on. I suggest,
even with wave optics, that having different parts point in slightly
wrong directions, directs the centers of the diffraction image into a
larger net region than the ideal Airy disk. I don't think you can ignore
the geometric part, it still influences where the prop. vectors of
wavefronts are directed.
It seems like no one here will deal directly with the specific issue to
show what happens. Everyone either goes on a tangent like about
practical issues of gravity, or just insists that wave optics will clear
things up - so there! etc.
I don't know just how "fractional wave" specs correspond to surface
traits, but my point is not so much about whether a 1/8 wave mirror of
any size would have predicted resolution, but more about whether such
ever-larger mirrors can have such resolution regardless of how we would
describe one that *was* perfect enough.
Androcles
> It seems like no one here will deal directly with the specific issue to
> show what happens. Everyone either goes on a tangent like about practical
> issues of gravity, or just insists that wave optics will clear things up -
> so there! etc.
>
> I don't know just how "fractional wave" specs correspond to surface
> traits, but my point is not so much about whether a 1/8 wave mirror of
> any size would have predicted resolution, but more about whether such
> ever-larger mirrors can have such resolution regardless of how we would
> describe one that *was* perfect enough.
Either you accept angle of incidence equals angle of reflection at a
mirror or you don't. Unless the mirror surface is a grating like a CD
or DVD, it ain't gonna diffract. Colour fringes are not a result
diffraction but refraction. The reason "everyone goes on a tangent"
about the practical issue of gravity is a simple one: that is the issue
that matters for large heavy discs which distort and cause colour
fringes, and not the issue for small light discs.
Bob May
the surface. I regularly make mirrors to 1/50th of a wave Peak to Valley of
surface error and test them. I'm not all that good in mirror making either.
Mirrors of that quality are essentilally perfect mirrors. Yet, there are
other problems that can occur.
I'd also suggest that you look at what the big multisegment telescopes are
designed as. In addition, the big telescopes have active shape control so
you really need to find out what that is all about. Then finally, you need
to find out what the scopes are mostly used for and the requirements for
that.
By the time you get a good understanding of all that, you'll understand the
questions you are asking as well as the futility of the way you are asking
them.
Jim Newman
>
> Either you accept angle of incidence equals angle of reflection at a
> mirror or you don't. Unless the mirror surface is a grating like a CD
> or DVD, it ain't gonna diffract.
Hey Androcles, that's a very odd statement to make.
Does a photon need to know what sort of surface it's striking before it
decides whether to diffract or reflect? How exactly does it do that?
Or are you talking your usual ignorant bullshit again?
Androcles
"Jim Newman" <j...@gmaill.invalid> wrote in message
news:7cunfdF...@mid.individual.net...
mirror or you don't. Take it up with Snell, fuckwit.
http://tinyurl.com/nb9apn
Jim Newman
Snell? Snell? - that would be refraction rather than reflection. Your
ignorance is truly remarkable
I ask again - you imply a photon knows whether it's hitting a mirror or
a grating - describe how it does this or, by omission, accept you are an
ignorant halfwit of the worst order.
Try Quantum Electrodynamics Theory
Being only 60 years old it may be a bit modern for you, but it won't be
the first time you'll have been bested by a youngster - you old sad fool.
Androcles
"Jim Newman" <j...@gmaill.invalid> wrote in message
news:7cunfdF...@mid.individual.net...
>
>>
>> Either you accept angle of incidence equals angle of reflection at a
>> mirror or you don't. Unless the mirror surface is a grating like a CD
>> or DVD, it ain't gonna diffract.
>
> Hey Androcles, that's a very odd statement
"Anything you can snip I can snip better,
I can snip anything better than you."
"No you can't!"
"Yes I can."
"No you can't."
"YES I CAN, YES I CAN, YES I CAAAAAAAN!"
Fuck you too.
Jim Newman
Is that your 'considered' response? The one you came back to post
because you thought it was BETTER than your first response?
Don't you ever wish you drank less and thought more before responding?
Androcles
"Jim Newman" <j...@gmaill.invalid> wrote in message
> Androcles wrote:
>> "Jim Newman" <j...@gmaill.invalid> wrote in message
>> news:7cunfdF...@mid.individual.net...
>>> Androcles wrote:
>>>
>>>> Either you accept angle of incidence equals angle of reflection at a
>>>> mirror or you don't. Unless the mirror surface is a grating like a CD
>>>> or DVD, it ain't gonna diffract.
>>> Hey Androcles, that's a very odd statement
>>
>> "Anything you can snip I can snip better,
>> I can snip anything better than you."
>> "No you can't!"
>> "Yes I can."
>> "No you can't."
>> "YES I CAN, YES I CAN, YES I CAAAAAAAN!"
>>
>> Fuck you too.
>
>
> Is that your considered response?
Now fuck off, cretin.
Androcles
"Jim Newman" <j...@gmaill.invalid> wrote in message
> Androcles wrote:
>> "Jim Newman" <j...@gmaill.invalid> wrote in message
>> news:7cunfdF...@mid.individual.net...
>>> Androcles wrote:
>>>
>>>> Either you accept angle of incidence equals angle of reflection at a
>>>> mirror or you don't. Unless the mirror surface is a grating like a CD
>>>> or DVD, it ain't gonna diffract.
>>> Hey Androcles, that's a very odd statement to make.
>>> Does a photon need to know what sort of surface it's striking before it
>>> decides whether to diffract or reflect? How exactly does it do that?
>>>
>>> Or are you talking your usual ignorant bullshit again?
>>
>> Either you accept angle of incidence equals angle of reflection at a
>> mirror or you don't. Take it up with Snell, fuckwit.
>> http://tinyurl.com/nb9apn
>
> Snell? Snell?
Yes.
Now fuck off.
Jim Newman
> "Jim Newman" <j...@gmaill.invalid> wrote in message
> news:7cut68F...@mid.individual.net...
>> Androcles wrote:
>>> "Jim Newman" <j...@gmaill.invalid> wrote in message
>>> news:7cunfdF...@mid.individual.net...
>>>> Androcles wrote:
>>>>
>>>>> Either you accept angle of incidence equals angle of reflection at a
>>>>> mirror or you don't. Unless the mirror surface is a grating like a CD
>>>>> or DVD, it ain't gonna diffract.
>>>> Hey Androcles, that's a very odd statement to make.
>>>> Does a photon need to know what sort of surface it's striking before it
>>>> decides whether to diffract or reflect? How exactly does it do that?
>>>>
>>>> Or are you talking your usual ignorant bullshit again?
>>> Either you accept angle of incidence equals angle of reflection at a
>>> mirror or you don't. Take it up with Snell, fuckwit.
>>> http://tinyurl.com/nb9apn
>> Snell? Snell?
>
> Yes.
> Now fuck off.
Heh heh
Where did the wabbit go?
Silly man!
For anyone interested, the Feynman Vega lectures give an interesting
introduction to QED. The first 2 lectures are particularly relevant.
Jim Newman
> "Jim Newman" <j...@gmaill.invalid> wrote in message
> news:7cuu3bF...@mid.individual.net...
>> Androcles wrote:
>>> "Jim Newman" <j...@gmaill.invalid> wrote in message
>>> news:7cunfdF...@mid.individual.net...
>>>> Androcles wrote:
>>>>
>>>>> Either you accept angle of incidence equals angle of reflection at a
>>>>> mirror or you don't. Unless the mirror surface is a grating like a CD
>>>>> or DVD, it ain't gonna diffract.
>>>> Hey Androcles, that's a very odd statement
>>> "Anything you can snip I can snip better,
>>> I can snip anything better than you."
>>> "No you can't!"
>>> "Yes I can."
>>> "No you can't."
>>> "YES I CAN, YES I CAN, YES I CAAAAAAAN!"
>>>
>>> Fuck you too.
>>
>> Is that your considered response?
> Yes.
> Now fuck off, cretin.
>
>
>
Hey, how clever! - second time of posting he thinks to set the followups
How long before the stupid fuck thinks to learn not to assume he can
lecture people on simple physics ?
Androcles
"Jim Newman" <j...@gmaill.invalid> wrote in message
> Androcles wrote:
>> "Jim Newman" <j...@gmaill.invalid> wrote in message
>> news:7cuu3bF...@mid.individual.net...
>>> Androcles wrote:
>>>> "Jim Newman" <j...@gmaill.invalid> wrote in message
>>>> news:7cunfdF...@mid.individual.net...
>>>>> Androcles wrote:
>>>>>
>>>>>> Either you accept angle of incidence equals angle of reflection at a
>>>>>> mirror or you don't. Unless the mirror surface is a grating like a
>>>>>> CD
>>>>>> or DVD, it ain't gonna diffract.
>>>>> Hey Androcles, that's a very odd statement
>>>> "Anything you can snip I can snip better,
>>>> I can snip anything better than you."
>>>> "No you can't!"
>>>> "Yes I can."
>>>> "No you can't."
>>>> "YES I CAN, YES I CAN, YES I CAAAAAAAN!"
>>>>
>>>> Fuck you too.
>>>
>>> Is that your considered response?
>> Yes.
>> Now fuck off, cretin.
>>
>>
>>
>
> Hey, how clever!
Androcles
"Jim Newman" <j...@gmaill.invalid> wrote in message
> Androcles wrote:
>> "Jim Newman" <j...@gmaill.invalid> wrote in message
>> news:7cut68F...@mid.individual.net...
>>> Androcles wrote:
>>>> "Jim Newman" <j...@gmaill.invalid> wrote in message
>>>> news:7cunfdF...@mid.individual.net...
>>>>> Androcles wrote:
>>>>>
>>>>>> Either you accept angle of incidence equals angle of reflection at a
>>>>>> mirror or you don't. Unless the mirror surface is a grating like a
>>>>>> CD
>>>>>> or DVD, it ain't gonna diffract.
>>>>> Hey Androcles, that's a very odd statement to make.
>>>>> Does a photon need to know what sort of surface it's striking before
>>>>> it decides whether to diffract or reflect? How exactly does it do
>>>>> that?
>>>>>
>>>>> Or are you talking your usual ignorant bullshit again?
>>>> Either you accept angle of incidence equals angle of reflection at a
>>>> mirror or you don't. Take it up with Snell, fuckwit.
>>>> http://tinyurl.com/nb9apn
>>> Snell? Snell?
>>
>> Yes.
>> Now fuck off.
>
> Heh heh
Still a brainless grinagog...
zzbu...@netscape.net
> On Thu, 23 Jul 2009 12:54:41 -0400, Neil B. wrote:
> > if they are simply accurate to within a given standard like 1/8 wave.
>
> They are a lot better.
But, mirrors of all shapes and sizes are also why the people who
actally
know how optics work, work on AI, adaptive optics, optical
computers,
microcomputers, electronic books, fiber optics, usb, distributed
processing software,
GPS, HDTV, C++, Home Broadband, Holograms, PGP, Laser Disk
Libraries, Blue Ray,
On-Line Banking, On-Line Publishing, Self-Replicating Machines, and
Self-Assembling Robots,
rather than with astronomers anyway. so it might only be an issue
for Martian Subtanners.
>
> > But consider that as a mirror gets larger, the tiny variations
> > in orientation of portions of the mirror should eventually cause
> > errors greater than the ever-decreasing diffraction spot.
>
> The trick is not to allow "tiny variations" on a small scale. This is
> called "polishing". The result is a surface that is smooth over the whole
> range of lengths from the diameter down to the size of the wavelength.
>
> > But application of geometric optics to the surface irregularities should
> > mean, a circle of confusion 0.01 arcsec diameter.
>
> You better apply wave optics here.
>
> BTW: The reason for large mirrors is not only resolution but also the
> amount of light gathered.
Benj
wrote:
> You wouldn't know a mousetrap from a dog's jaw if it bit you on
> the arse, Jocaby.
> http://en.wikipedia.org/wiki/File:USA.NM.VeryLargeArray.02.jpg
Really nice "optical" array, there Andro. As usual your examples have
NOTHING to do with the points under discussion.
Idiot.
Benj
He's talking total ignorant bullshit again as usual. He's got NO idea
what diffraction is. He has NO idea that every object diffracts
because Maxwell's equations are obeyed for electromagnetic radiation.
All finite objects have edges. He's an idiot.
Androcles
Idiot.
===============================================
Subject:
"Large mirrors can't achieve theoretical resolution"
No "optical" mentioned.
Ignorant cunt.
Benj
wrote:
> On Jul 23, 1:18 pm, Kai-Martin Knaak <k...@lilalaser.de> wrote:
>
> > On Thu, 23 Jul 2009 12:54:41 -0400, Neil B. wrote:
> > > if they are simply accurate to within a given standard like 1/8 wave.
>
> > They are a lot better.
>
> But, mirrors of all shapes and sizes are also why the people who
> actally
> know how optics work,
Which is why real scientists and engineers who actually know optics
spend their time working on perpetual motion machines, free energy,
channeled schematic diagrams, UFO tracking, debunking relativity, time
travel, ghost busting, aether theory, faster than light
communications, warp drives, ...
Idiot.
Androcles
=====================================
Bwhahahahahaha!
Jocaby is worried about the edge diffraction you can see in this
photograph of a mirror and the colour fringe it leaves on the wall behind:
http://www.scienceahead.com/images/moonscope_mirror_2405.jpg
If the facts don't fit the theory, change the facts, Jocaby.
Totally ignorant fuckwit.
Neil B.
> There are plans to build a telescope with a mirror that is thirty
> meters wide:
> http://www.tmt.org/
> A 30-meter telescope would be way cool, but something about the
> supposed capabilities of such mammoth apertures has long bothered me.
> I question how ever-larger apertures can continue to provide ever
> better resolution with e.g. visible light - if they are simply
> accurate to within a given standard like 1/8 wave. And this one would
> be a MMT, but the same argument below applies to portions and to
> single piece versions too.
>
> In theory, in the ideal, I see why ever larger apertures confer
> ever-better resolution. But consider that as a mirror gets larger,
> eventually cause errors greater than the ever-decreasing diffraction
> have roughly random (how "random" isn't the main issue) "orientation
> errors" of around 0.01 arc second. So if this mirror was idealized
> and we used hard UV at 50 nm, the theoretical resolution should be
> about 0.0003 arcsec. But application of geometric optics to the
> endlessly improved with ever-shorter wavelengths. The flip side: ever
> larger apertures should fail to provide ever-increasing resolution
> with the *same* wavelengths.
>
> So let's switch to visible light. The theoretical resolution should
> be about 0.003 arcsec. Can that somehow be even better than the UV
> resolution, because of some global treatment of the larger waves? I
> find that hard to believe. The centers of where diffracted portions
> of light are sent would still be directed to slightly different points
> at the collection plane, per the 0.01 arcsec variations. I made up
> the 0.01 arcsec figure of orientation variation, but the same
> principle should apply in the real world.
>
> Maybe I misunderstand the implications of applying a "1/8 wave" etc.
> standard to different apertures, but I can't imagine that as
> mirrors/lenses were made larger we'd be able to proportionately reduce
> the magnitude of orientation flaws. It would be absurd. We're just
> making a bigger piece of the same stuff that someone must polish
> locally etc. (Also, the inherent physical grain etc.)
>
> Any scoop, especially on "who asked this first" etc? tx
Because of some confusion about my point, I summarize and clarify:
First, this is a suggestion/question and I'm not insisting the point is
right.
Second, of course wave diffraction also applies. I am suggesting that
the resolution can�t be any better than as limited by either the ray
tracing *or* the diffraction. IOW, both are limiting factors. I suspect
that resolution is limited by diffraction unless the ray tracing would
create a circle of confusion, of a substantial portion or bigger than
the diffraction spot. Otherwise we have an odd circumstance where the
resolution would be e.g. limited to 0.01 arcsec in hard UV due to ray
tracing, despite potential 0.0001 arcsec allowed by diffraction - but
then resolution limit shrinks to e.g. 0.001 arcsec in visible, then
starts going larger again as we move into the IR. If you accept that in
the limit of very short waves the ray tracing must give the distribution
of the spot, it's hard to see how to make it even smaller as wavelength
increases. It seems to me, the wave properties simply superpose various
propagation vectors (perp. to the front) and would superimpose
additional diffraction spreading onto the geometric circle of confusion.
Yet I realize that a given "portion" of the mirror wouldn't resolve by
itself as well as the whole anyway - nevertheless, if those portions
aren't directed to the required angular accuracy, how can they "work
together" to get a resolution even better than from ray tracing?
Third, I suggest that we cannot make mirrors to unlimited surface
consistency specs, of ever smaller angle-consistency from place to place
as the mirror is larger. Hence, the CoC from ray tracing will eventually
outgrow the diffraction disk and thus put a rough lower limit on the
resolution we can get from ever-larger mirrors.
zzbu...@netscape.net
Since the only thing real scienctists even do is make pretenses
that they actually know what Hilbert Space is, that's why real
engineers even work on electronic books, C++, usb, fiber optics,
holographics,
on-line publishing, microcomputers, laser disk libraries, atomic
clock wristwatches,
light, cyber batteries, self-replicatnig machines, and self-
assembling robots,
rather than idiot things like Quantum Chemistry.
>
> Idiot.
Benj
wrote:
> On Jul 25, 8:44 am, Benj <bjac...@iwaynet.net> wrote:
> > Which is why real scientists and engineers who actually know optics
> > spend their time working on perpetual motion machines, free energy,
> > channeled schematic diagrams, UFO tracking, debunking relativity, time
> > travel, ghost busting, aether theory, faster than light
> > communications, warp drives, ...
> rather than idiot things like Quantum Chemistry.
Isn't Quantum Chemisty a lot like Ghost Busting and Warp drives?
zzbu...@netscape.net
It more like Quantun Philosophy than anything else, which is
also why the engineers with non dense set brains also invented
laser-guided phasors for the string theory cranks.
Salmon Egg
Kai-Martin Knaak <k...@lilalaser.de> wrote:
> On Thu, 23 Jul 2009 12:54:41 -0400, Neil B. wrote:
>
> > if they are simply accurate to within a given standard like 1/8 wave.
>
> They are a lot better.
>
>
> > But consider that as a mirror gets larger, the tiny variations
> > in orientation of portions of the mirror should eventually cause
> > errors greater than the ever-decreasing diffraction spot.
>
> The trick is not to allow "tiny variations" on a small scale. This is
> called "polishing". The result is a surface that is smooth over the whole
> range of lengths from the diameter down to the size of the wavelength.
>
>
> > But application of geometric optics to the surface irregularities should
> > mean, a circle of confusion 0.01 arcsec diameter.
>
> You better apply wave optics here.
>
> BTW: The reason for large mirrors is not only resolution but also the
> amount of light gathered.
>
> ---<(kaimartin)>---
It has been a long while ago that I duplicated others' calculation on
strehl ratio. That the ratio of intensity a of a plane wave focused to
the focal point of a mirror compared to what it would be for a perfect
mirror. This depended only upon the rms variation of the surface from
the ideal surface,
That means that as the mirror gets larger, a focused image will not
deteriorate. The image will always "stick out" the same relative amount
compared to an ideal mirror.
The effect of spatial variation is to determine just how the energy that
is lost is distributed around the image.
Bill
--
Most people go to college to get their missing high school education.
Mike Dworetsky
news:jeSdnYy0j8BACPXX...@posted.widowmaker...
> There are plans to build a telescope with a mirror that is thirty meters
> wide:
> http://www.tmt.org/
> A 30-meter telescope would be way cool, but something about the supposed
> capabilities of such mammoth apertures has long bothered me. I question
> how ever-larger apertures can continue to provide ever better resolution
> with e.g. visible light - if they are simply accurate to within a given
> standard like 1/8 wave. And this one would be a MMT, but the same argument
> below applies to portions and to single piece versions too.
>
> In theory, in the ideal, I see why ever larger apertures confer
> consider e.g. that little portions of the 30-meter mirror have roughly
> random (how "random" isn't the main issue) "orientation errors" of around
> 0.01 arc second. So if this mirror was idealized and we used hard UV at
> surface traits and can't be endlessly improved with ever-shorter
> wavelengths. The flip side: ever larger apertures should fail to provide
> ever-increasing resolution with the *same* wavelengths.
>
> So let's switch to visible light. The theoretical resolution should be
> about 0.003 arcsec. Can that somehow be even better than the UV
> resolution, because of some global treatment of the larger waves? I find
> that hard to believe. The centers of where diffracted portions of light
> are sent would still be directed to slightly different points at the
> collection plane, per the 0.01 arcsec variations. I made up the 0.01
> arcsec figure of orientation variation, but the same principle should
> apply in the real world.
>
> Maybe I misunderstand the implications of applying a "1/8 wave" etc.
> standard to different apertures, but I can't imagine that as
> mirrors/lenses were made larger we'd be able to proportionately reduce the
> magnitude of orientation flaws. It would be absurd. We're just making a
> bigger piece of the same stuff that someone must polish locally etc.
> (Also, the inherent physical grain etc.)
>
> Any scoop, especially on "who asked this first" etc? tx
>
You may be thinking of the concept of the Strehl Ratio, which measures the
ratio of the actual peak of the image of a point source compared to a
theoretical diffraction-limited image of the same. No telescope ever
performs 100% in this regard, and the bigger they are, the harder it is to
achieve good performance without real-time correction of the wave-front.
Error in the wave-front is the only thing that matters for achieving
diffraction-limited performance. A perfect wave-front will give
diffraction-limited images.
Yes, 1/8-wave errors on a huge 30-m telescope could cause significant errors
if you didn't correct for them. But no telescope of that size is built to
work in a passive mode. The segments will have many hundreds or thousands
of actuators over the 30-m diameter working in real time (or close to it) to
improve the image by correcting for geometric errors (imperfect shape,
misalignments of segments, gravitational sag, etc) to obtain
near-diffraction-limited performance. Then, by using a laser guide star
(artificial point source generated at the telescope and projected into the
upper atmosphere) in real time the atmospheric seeing variations can be
corrected within the isoplanatic patch (an area normally a few arcsec
across). The wave-front correction is usually done in a device in the
optical path near the focus. This is now commonly used at large telescopes.
So any such errors in the mirrors won't stop the achievement of something
approaching theoretical performance, because they are removed in real time
(or, rather, fast enough to make the corrections measured a ms in the past
applicable now, or a ms in the future). There is a small time lag so the
result is not quite perfect.
One reason for choosing a site with the best possible atmospheric seeing is
that the better the seeing, the closer the actuators and corrector will come
to generating a diffraction limited performance, and, perhaps, the
isoplanatic patch will be a bit bigger.
Of course, as has been pointed out by other posters, the main aim of
building such telescopes is to do spectroscopy of fainter objects, and for
spectroscopy you don't need diffraction-limited performance, just a big
light bucket with acceptable images.
As to who first thought of the idea of correcting for seeing in real time, I
think the credit belongs to Horace W Babcock (Mt Wilson/Palomar) in the
1940s/50s. The idea of deforming a mirror to achieve better performance
also goes back a long time, possibly to astronomers at one of the
observatories in South Africa whose mirror was too thin or badly supported,
so they devised a way of "tweaking" the mirrror with hand-operated
actuators.
--
Mike Dworetsky
(Remove pants sp*mbl*ck to reply)
Neil B.
"Mike Dworetsky" <plati...@pants.btinternet.com> wrote in message
news:Rb2dnSf7Kem0gPHX...@bt.com...
of. It seems like you're agreeing with me to some extent, although you
do say
"Error in the wave-front is the only thing that matters for achieving
diffraction-limited performance. A perfect wave-front will give
diffraction-limited images."
rays larger than diffraction spot) limit resolution because the wave
front perforce wouldn't be "perfect"? I suggest, that even "good" 1/20
etc. wave standard doesn't prevent the issue of regions of the mirror
directing propagation vectors to slightly different angles as the mirror
gets larger and the desired diffraction spot gets to ever smaller angle.
REM that the alternative is for resolution to go from one size (short
wave much smaller than "fit" layer, that simulates ray tracing), to
getting better (smaller than ray spread), to getting worse (larger)
again as wave length increases.
Mike Dworetsky
That is true enough, but all such large telescopes are constructed with
corrective actuators built in, and these adjust the figure to give
performance as near to diffraction-limited as may be achievable. Perfect
performance is never achieved because even sites with the world's best
seeing fuzz things up a bit. However, they come pretty close.
A large space telescope can be adjusted once or occasionally to get nearly
diffraction-limited performance. This can be done regardless of size, and
only cost factors prevent the launching of 10, 20 and 30 m space telescopes.
Even the James Webb ST has been "de-scoped" for cost reasons, to the point
where it is getting below its original mission criteria in terms fo
collecting area.
Large interferometric arrays in space can achieve something approaching
diffraction-limited resolution, at least in principle. In fact, this is
being done on the ground, e.g., the VLT interferometer array at ESO (where
atmospheric seeing makes it harder).
Neil B.
>
> "Neil B." <neil_...@caloricmail.com> wrote in message
> news:jeSdnYy0j8BACPXX...@posted.widowmaker...
>> There are plans to build a telescope with a mirror that is thirty
>> meters wide:
>> http://www.tmt.org/
>> A 30-meter telescope would be way cool, but something about the
>> supposed capabilities of such mammoth apertures has long bothered me.
>> I question how ever-larger apertures can continue to provide ever
>> better resolution with e.g. visible light - if they are simply
>> accurate to within a given standard like 1/8 wave. And this one would
>> be a MMT, but the same argument below applies to portions and to
>> single piece versions too.
I think it's "sad", that after around 100 comments we still didn't get
IMHO a clear and definitive answer. Some people hinted I was right, but
no clear affirmation of resolution being the larger (at best) of the
ray-trace circle of confusion combined with the ideal diffraction spot.
Any good links, to clear exposition?
tx to most, for trying ...
Androcles
to the reply to yourself. Does it occur to you that you might
be raving fuckin' mad?
"Neil B." <neil_...@caloricmail.com> wrote in message
news:KuydndL39KTkJ-3X...@posted.widowmaker...
Neil B.
"Androcles" <Headm...@Hogwarts.physics_f> wrote in message
news:KX2cm.101296$qz1....@newsfe12.ams2...
> That's not bad. You reply to yourself and then answer the reply
> to the reply to yourself. Does it occur to you that you might
> be raving fuckin' mad?
>
more thoughts to a post or comment doesn't have to mean "answering
yourself" anyway. But even if it was: it's not silly to answer your own
question posed earlier - good thinkers often come up with answers to
their own questions, that they didn't think of earlier. Maybe you
contribute nothing and can't imagine what creative output consists of.
Maybe you've noticed, almost everyone in these NGs thinks you are
worthless. Most of the time you are, but don't take it too hard.
I don't fall for your Followup tricks, but am leaving alt.morons in
there too so that crowd can see my reply.
JohnH4
> There are plans to build a telescope with a mirror that is thirty meters
> wide:http://www.tmt.org/
> A 30-meter telescope would be way cool, but something about the supposed
> capabilities of such mammoth apertures has long bothered me. I question
> how ever-larger apertures can continue to provide ever better resolution
> with e.g. visible light - if they are simply accurate to within a given
> standard like 1/8 wave. And this one would be a MMT, but the same
> argument below applies to portions and to single piece versions too.
>
Neil, To expand you thoughts on the matter, I suggest that you
consider how the Very Long Baseline Array (VLBA) of radio telescopes
works. See http://www.vlba.nrao.edu/. I like to think of the VLBA as
having an apature of thousands of miles in diameter covered with a
mask with just a dozen or circular holes cut into it, each with a
diameter of a maybe a few hundred wavelengths. Rgds, JohnH
Martin Brown
> On Jul 23, 11:54 am, "Neil B." <neil_del...@caloricmail.com> wrote:
>> There are plans to build a telescope with a mirror that is thirty meters
>> wide:http://www.tmt.org/
>> A 30-meter telescope would be way cool, but something about the supposed
>> capabilities of such mammoth apertures has long bothered me. I question
>> how ever-larger apertures can continue to provide ever better resolution
>> with e.g. visible light - if they are simply accurate to within a given
>> standard like 1/8 wave. And this one would be a MMT, but the same
>> argument below applies to portions and to single piece versions too.
>>
>> In theory, in the ideal, I see why ever larger apertures confer
>> ever-better resolution. But consider that as a mirror gets larger, the
>> tiny variations in orientation of portions of the mirror should
>> eventually cause errors greater than the ever-decreasing diffraction
>> spot. IOW, consider e.g. that little portions of the 30-meter mirror
>> have roughly random (how "random" isn't the main issue) "orientation
>> errors" of around 0.01 arc second. So if this mirror was idealized and
>> we used hard UV at 50 nm, the theoretical resolution should be about
>> 0.0003 arcsec. But application of geometric optics to the surface
>> irregularities should mean, a circle of confusion 0.01 arcsec diameter.
>> IOW, resolution is limited by surface traits and can't be endlessly
>> improved with ever-shorter wavelengths. The flip side: ever larger
>> apertures should fail to provide ever-increasing resolution with the
>> *same* wavelengths.
Like all these things the mirror is never ideal although the HST mirror
when polished was one of the smoothest surfaces ever manufactured in its
day. It was a shame that it was very precisely the wrong shape.
If you try to operate at ever shorter wavelengths then eventually the
performance will be limited by geometrical optics considerations rather
than being diffraction limited for the physical aperture by
1.22lambda/D. The Chandra X-ray telescope is not diffraction limited for
instance. And its glancing incidence mirrors are very hard to
manufacture and align.
It is just becoming possible to make surfaces accurate and smooth enough
diffraction limited optics for soft X-rays which may be very
interesting. eg.
http://aas.org/archives/BAAS/v33n4/aas199/371.htm
>>
>> So let's switch to visible light. The theoretical resolution should be
>> about 0.003 arcsec. Can that somehow be even better than the UV
>> resolution, because of some global treatment of the larger waves? I
>> find that hard to believe. The centers of where diffracted portions of
>> light are sent would still be directed to slightly different points at
>> the collection plane, per the 0.01 arcsec variations. I made up the
>> 0.01 arcsec figure of orientation variation, but the same principle
>> should apply in the real world.
Diffraction doesn't quite work like that. You get into trouble if you
try to mix together geometrical optics with a diffraction treatment in
wave optics. Essentially to the incoming photon mirror imperfections
look like a phase error depending on whether it is high or low.
>>
>> Maybe I misunderstand the implications of applying a "1/8 wave" etc.
>> standard to different apertures, but I can't imagine that as
>> mirrors/lenses were made larger we'd be able to proportionately reduce
>> the magnitude of orientation flaws. It would be absurd. We're just
>> making a bigger piece of the same stuff that someone must polish locally
>> etc. (Also, the inherent physical grain etc.)
>>
>> Any scoop, especially on "who asked this first" etc? tx
Large optical mirrors where large is about 8" are limited on Earth by
the turbulence in the Earth's atmosphere rather than by any problems in
their manufacture. Big mirrors in space are currently limited by the
dimensions of the launch vehicle and g-forces during launch.
>
> Neil, To expand you thoughts on the matter, I suggest that you
> consider how the Very Long Baseline Array (VLBA) of radio telescopes
> works. See http://www.vlba.nrao.edu/. I like to think of the VLBA as
> having an apature of thousands of miles in diameter covered with a
> mask with just a dozen or circular holes cut into it, each with a
> diameter of a maybe a few hundred wavelengths. Rgds, JohnH
Classical aperture synthesis with a simple set of E-W fixed
interferometer baselines might be a lot easier to understand.
VLBI requires cunning combinations of the baseline pair observations to
obtain good observables even though there are unknown fluctuating phase
errors over every contributing big dish. You only get N(N-1)(N-2)/6 good
phase observables with N telescopes in the network.
The larger the number of scopes in the network the better the image
reconstruction. This has led to people keeping a big scope in play for
VLBI when stormy weather conditions should have caused it to be stowed.
Regards,
Martin Brown
Benj
> I think it's "sad", that after around 100 comments we still didn't get
> IMHO a clear and definitive answer. Some people hinted I was right, but
> no clear affirmation of resolution being the larger (at best) of the
> ray-trace circle of confusion combined with the ideal diffraction spot.
> Any good links, to clear exposition?
> tx to most, for trying ...
What is sad is that after 100 posts you are still working hard to
maintain your ignorance.
You really have very little understanding of these matters. I assume
you are a science teacher.
You need to understand the difference between the "ideal" diffraction
spot (which is to say "classical" one) and the ACTUAL diffraction spot
produced by the actual figure of the mirror. Ray trace is simply an
approximate method to get optical performance and hence has
limitations. Actual performance is determined by the actual
diffraction spot.
Martin Brown has told you: "Diffraction doesn't quite work like that.
You get into trouble if you
try to mix together geometrical optics with a diffraction treatment in
wave optics. Essentially to the incoming photon mirror imperfections
look like a phase error depending on whether it is high or low."
Which is exactly what I told you. (and Gisse) And you wouldn't
believe it. Sorry Neil, your theory is nonsense. Your understanding is
superficial. Large mirrors can have higher resolution than smaller
mirrors. And there are MANY factors including atmospheric ones that
determine the total performance of a telescope. Your idea of
manufacturing ray trace angles isn't one of them. The people building
the 30 meter telescope are not morons and they do know more about what
they are doing than you do.
Is it going to take 100 more posts to get you to actually study this
question instead of just repeating your "theories"?
You can start by looking into the "keyhole" series of spy satellites.
3" resolution from earth orbit. [actually now much better and able to
read license plates as I understand it] You still think large mirrors
can't work? Idiot.
Neil B.
On Jul 29, 5:08 pm, "Neil B." <neil_del...@caloricmail.com> wrote:
>> I think it's "sad", that after around 100 comments we still didn't
>> get
>> IMHO a clear and definitive answer. Some people hinted I was right,
>> but
>> no clear affirmation of resolution being the larger (at best) of the
>> ray-trace circle of confusion combined with the ideal diffraction
>> spot.
>> Any good links, to clear exposition?
>> tx to most, for trying ...
> What is sad is that after 100 posts you are still working hard to
> maintain your ignorance.
> You really have very little understanding of these matters. I assume
> you are a science teacher.
Pretty close, BTDT. Kind of sad to make that a derogatory.
You keep forgetting, as I said, that I pose this in terms of a Socratic
question to get a discussion going. My speculation isn't really a
theory, it's something put up to see what adjustment it gets if someone
can explain. What I hear is mostly generalizations about how such and
such isn't enough, it's not a good model etc. but I was looking for more
specific answers. I mean, things like "given a 30m mirror with
aberrations per se producing a 0.03 arcsec circle of confusion (yes,
they do use that term and you should have known), the wave treatment
allows for resolution only 20% larger than the theoretical best of 0.003
arcsec" etc. That's better net than the largest CoC, but it's still a
reduction of resolution scale and that's more to the point.
I didn't hear that, just vague statements about how ray trace isn't
enough etc. Well of course it isn't enough, I wanted to hear in just
what way it interacted with the wave treatment. I did get the suggestion
from Jim Black to calculate the angular error of regions of given sizes
for 1/n wave mirrors of different sizes. he earlier implied that
orientation errors should degrade the image by reducing intensity of
wave available right at the focal point. This is what I get:
Try 1/20 wave accuracy, which for 500 nm is 25 nm. That is the "layer"
the mirror surface must fit into. Pick regions about 1m scale and see
what relative angle distortion is possible from them at a tilt within
those bounds, approx. It's simple. Multiply by four because (1) the
region could be tilted either way, (2) the angle deviation of a normal
ray is double the angle of the surface. Then divide tolerance by size to
get angle in radians.
Angle in radians = 2*25nm/1m = 100/1,000,000,000 = 0.000 000 1 radian =
0.02 arcsec. That would spread light out beyond a perfect diffraction
disk, for a mirror larger than 5 meters. So if the surface had
variations at that scale it could be a problem - maybe not limiting to
that large a resolution, but some effect. I don't know, I wish an actual
designer/mfr. would say whether this sort of thing matters. Note that
the 8-10m scopes we do have, are not classical and have adaptive mirrors
which can correct for more than just the atmosphere as some commenters
noted.
You also falsely make it appear everyone else agreed with each other,
but they didn't and that complicates things. Eric gisse doesn't think
much of your shtick, why should I believe you instead of him?
> You need to understand the difference between the "ideal" diffraction
> spot (which is to say "classical" one) and the ACTUAL diffraction spot
> produced by the actual figure of the mirror. Ray trace is simply an
> approximate method to get optical performance and hence has
> limitations. Actual performance is determined by the actual
> diffraction spot.
OK, and that is dependent in what way and to what degree on the
deviations from a perfect curve - ?
> Martin Brown has told you: "Diffraction doesn't quite work
> like that. You get into trouble if you try to
> mix together geometrical optics with a diffraction treatment
> in wave optics. Essentially to the incoming photon mirror
> imperfections look like a phase error depending on
> whether it is high or low."
OK answer as a generalization, but doesn't say to what degree, given
imperfections affect the final Airy disk. Just to hear that doesn't tell
us whether the effect on resolution is small or large.
> Which is exactly what I told you. (and Gisse)
See points as made above. The issue is to hash out how much effect
aberrations and patchy flaws have on resolution as mirrors get larger -
not whether the "toy thesis" given as initial stimulus is precisely
correct.
> And you wouldn't believe it.
I might easily believe a specific claim, of the sort shown above as a
"good" answer.
> Sorry Neil, your theory is nonsense.
Out of context - "not even wrong."
> Your understanding is superficial. Large mirrors
Posed as a challenge to answer, that isn't the point
> can have higher resolution than smaller mirrors.
Of course they can. I earlier explained how you utterly misread my
point. Your grossly misdirected caricature of my point was so absurd, it
reminded me of distortions from Glenn Beck, Limbaugh, and Faux News.
*The issue is, just how close to theoretical best is allowed by various
levels of mirror imperfection.* That's a perfectly good question,
regardless of whether or not it happens to be specifically the larger of
CoC combined with ideal diffraction or other diffraction spot.
> And there are MANY factors including atmospheric ones
Not relevant here.
> that determine the total performance of a telescope.
> Your idea of manufacturing ray trace angles isn't one of them.
Optical designers absolutely do a race trace through a scatter of points
into the aperture, and try to minimize the size of it. Would you know?
> The people building the 30 meter telescope are not morons
> and they do know more about what they are doing than you do.
And you. Note that light-gathering power is also important and the 30m
need not achieve Dawes' limit. If it achieved say, double that it would
still be very useful and sharp. And I just used the 30 meter as an
example, it could be that we can't figure a 1000 meter mirror etc. good
enough. If *they* tell me themselves that it's no harder to get near
Dawes' limit in their telescope as with a 10 meter scope, I will be
impressed - not by you.
> Is it going to take 100 more posts to get you to actually study
> this question instead of just repeating your "theories"?
Better attention from most commenters to the specific issue of the
issues of *very* large mirrors, and with "for example, ..." would have
helped. OK, I'll study it but that's part of the point of the NGs.
> You can start by looking into the "keyhole" series of spy satellites.
> 3" resolution from earth orbit. [actually now much better and able to
> read license plates as I understand it] You still think large mirrors
> can't work? Idiot.
No, I don't think that. Keyhole series is similar to Hubble, mirrors
around 2.5m. I know we can get Dawes' limit with that, I ask whether we
can actually get Dawes' limit at 30 or 1000m. It is not obvious, or from
the comments, how easy that is.
You obviously can't read "either."
You do have a sense of humor, I got a rise out of
news:1b8e24d7-617c-4496...@37g2000yqp.googlegroups.com...
Neil B.
"Martin Brown" <|||newspam|||@nezumi.demon.co.uk> wrote in message
news:1Sgcm.50427$BP6....@newsfe24.iad...
Thanks to both of you. I note that "processing" is required in
interferometry, which is not quite the same as having one big mirror
just focusing light to a point. I still pose the challenge and question,
unless someone has examples and figures and not just generalities: is
the resolution limited to roughly the larger of ray-trace circle of
confusion, and Dawes' limit, and if not then to what degree do
aberrations degrade resolution esp. in large mirrors? BTW forget
atmosphere etc, it's not for dinner.
Marco
> You keep forgetting, as I said, that I pose this in terms of a Socratic
> question to get a discussion going.
Socratic method: A pedagogical technique in which a teacher does not give
information directly but instead asks a series of questions, with the result
that the student comes either to the desired knowledge by answering the
questions or to a deeper awareness of the limits of knowledge.
So are you the teacher who is trying to lead to otheres to knowledge
> My speculation isn't really a
> theory, it's something put up to see what adjustment it gets if someone
> can explain.
Or are you trying to gain knowledge?
You don't seem willing to listen, but you also don't seem willing to do more
than handwaving counter arguements. How about some worked out examples.
Marco
Androcles
"Marco" <nu...@null.net> wrote in message news:4a725330@darkstar...
That's Standard Operating Procedure for newsgroups. The argument
takes precedence over the knowledge, that then reduces to handwaving
and then flaming.
Benj
> This is what I get:
>
> Try 1/20 wave accuracy, which for 500 nm is 25 nm. That is the "layer"
> the mirror surface must fit into. Pick regions about 1m scale and see
> what relative angle distortion is possible from them at a tilt within
> those bounds, approx. It's simple. Multiply by four because (1) the
> region could be tilted either way, (2) the angle deviation of a normal
> ray is double the angle of the surface. Then divide tolerance by size to
> get angle in radians.
Boy you sure are persistent! I guess that's good. Consider a large
mirror. Say it has a curvature that is "perfect". In that case the ray
trace CoC has a diameter of? Right, ZERO! But the "classical" area
of the spot size will be determined by ? Right, the mirror diameter
and it's shape and the wavelength. [We are assuming circular mirrors
here] Now consider a "layer" upon that perfect mirror 25 nm thick.
Now we have two perfect figures spaced 25 nm apart. Now assume that
your "real" mirror has a figure that is ground to 1/20 wave which
means that no part of that mirror deviates more than 1/20th wave from
perfect. OK? This means that the figure of the mirror must "fit"
within that layer we've put on our perfect mirror. Then the question
becomes as you have asked, "how much does that "real" surface "tilt"
from the perfect figure?" Now the part that is interesting (draw this
out) is that if the variations of figure are very slowly varying the
length of your possible angle lines keeps them from tilting very much.
But if the mirror has rapid variations (like say sawtooth lines) your
ray trace can angle off at quite steep angles. Hence we can see that
actually the diameter of the mirror is irrelevant to the ray trace
results. In fact if the figure is slowly varying, a larger mirror will
have LESS tilt than a smaller one! And did you notice that the focal
length creates a longer or shorter "lever arm"?
> Angle in radians = 2*25nm/1m = 100/1,000,000,000 = 0.000 000 1 radian =
> 0.02 arcsec. That would spread light out beyond a perfect diffraction
> disk, for a mirror larger than 5 meters. So if the surface had
> variations at that scale it could be a problem - maybe not limiting to
> that large a resolution, but some effect. I don't know, I wish an actual
> designer/mfr. would say whether this sort of thing matters. Note that
> the 8-10m scopes we do have, are not classical and have adaptive mirrors
> which can correct for more than just the atmosphere as some commenters
> noted.
Lets not worry about adaptive optics at this point. Lets continue with
the above arguments. What if the mirror surface had a nice sawtooth
pattern (say it was diamond turned). What about the large COC then?
The problem is that as you make these fine highly tilted grooves on
the mirror, it's no longer a mirror it's a diffraction grating. Hint:
Ray trace calculations do not work for diffraction gratings. Well what
do diffraction calculations give for this "mirror"? Well, in this case
the angular pattern of the mirror is the Fourier transform of the
amplitude and phase of the wavefront upon the mirror. When there are
high spatial frequencies it results in high angular patterns. This is
obvious as gratings send light off at fairly high angles. Now the
"classical" diffraction limit of a mirror assumes only the low spatial
frequencies generated by the aperture are present. In other words the
transform is of a single spatial pulse the width of the mirror. An
actual mirror will not be that good as higher spatial frequencies are
present and do degrade the performance.
So now what about your fixation on ray trace? Well, if one assumes
that our mirrors are nearly perfect and high frequency variations are
not present (the figure varies slowly across the mirror) then sure,
you can calculate a COC that is reasonably useful if you have the
actual figure of the mirror to plug in. And we've seen that generally
speaking that angular deviation will be due to local variations on the
mirror. Now if the mirror is reasonably perfect wavelength wise, we
know that as the diameter increases the "classical" resolution also
increases. This means that as you make a mirror larger and larger, the
spot size gets smaller and smaller and pretty soon the ray trace COC
is larger than the "classical" angular resolution. Which was your
point. However, we've just seen that the REAL angular diffraction
limit is not set by the "classical" value. The classical value ignores
the variations in the phase of the mirror. So the real diffraction
limit is larger and closer to the COC.
But I know what you are going to say. What if we make this mirror
HUGE so that it's diffraction limit is very tiny! But our ray trace
data shows this large COC. So isn't the COC the actual performance of
the mirror? Sorry bunky. This is where you are stuck. Light is not
perfect rays. Light is not perfect mathematical lines and functions.
Mathematics is not more real than reality. The true angular resolution
will be given by the diffraction calculation and not the ray trace
one.
I know this is hard for you to swallow but angular resolution has to
do with larger mirrors providing "support" on the sides of the light
beam. It is a WAVE phenomenon. If you use the wrong formula you get
the wrong answer. The "light rays" model used in ray trace is simply
not a correct one. Thus if you push it too hard, it falls apart as a
number of us have told you. Blindly applying formulas is not
scientific understanding.
> You also falsely make it appear everyone else agreed with each other,
> but they didn't and that complicates things. Eric gisse doesn't think
> much of your shtick, why should I believe you instead of him?
Eric says I'm not even a good amateur, so why should you believe Me?
Or anyone? Since when is science a matter of faith? Since when does
"belief" enter into this? Eric and I are hopefully not your ONLY
sources of information. Cripes if you are trusting the Internet to
answer your scientific questions you are already in deep doo doo! But
hopefully you will have obtained enough hints here to give you some
places to start digging. What you believe is up to you. I'm not here
to "convert" anyone.
Skywise
_XnZ2dnUV...@posted.widowmaker:
> You keep forgetting, as I said, that I pose this in terms of a Socratic
> question to get a discussion going.
That could be interpreted as trolling.
Brian
--
http://www.skywise711.com - Lasers, Seismology, Astronomy, Skepticism
Seismic FAQ: http://www.skywise711.com/SeismicFAQ/SeismicFAQ.html
Quake "predictions": http://www.skywise711.com/quakes/EQDB/index.html
Sed quis custodiet ipsos Custodes?
Helpful person
> "Neil B." <neil_del...@caloricmail.
>
> I think it's "sad", that after around 100 comments we still didn't get
> IMHO a clear and definitive answer. Some people hinted I was right, but
> no clear affirmation of resolution being the larger (at best) of the
> ray-trace circle of confusion combined with the ideal diffraction spot.
> Any good links, to clear exposition?
>
As already stated the autocorrelation of the pupil function
(wavelength) gives the MTF response. Ray tracing is irrelevent for
well corrected systems. In fact ray tracing can show excellent
correction whereas the true resolution calculated from the wavefront
may be poor.
Simply put, in general, reducing the wavefront rms improves the image.
Neil B.
> news:vtqdnW_dyL0ao-_X...@posted.widowmaker...
>
>> You keep forgetting, as I said, that I pose this in terms of a
>> Socratic
>> question to get a discussion going.
>
> Socratic method: A pedagogical technique in which a teacher does not
> give information directly but instead asks a series of questions, with
> the result that the student comes either to the desired knowledge by
> answering the questions or to a deeper awareness of the limits of
> knowledge.
>
> So are you the teacher who is trying to lead to otheres to knowledge
>
>> My speculation isn't really a
>> theory, it's something put up to see what adjustment it gets if
>> someone
>> can explain.
>
> Or are you trying to gain knowledge?
>
that out through "challenge-response." A Socratic _dialog_ can involve
mutual learning. Many would say, that's actually the best application -
not the one-sided "teacher" version. I want readers to take the query
and work the issues into their response. They try, but mostly put out
reflex answers that, even if true in a broad sense, don't deal with the
specific query "what if there are patchy irregularies that ray-trace to
worse resolution than the Dawes' limit?" We already know we need to deal
with ray trace or the spherical abberation wouldn't matter, or we
wouldn't need to ray-trace to make sure a Ritchey-Chretian had the
correct combined curvature of mirrors. And how about the coma at the
edge of field of a parabolic mirror? Would that ever be made *less* by
wave optics? So that has to be correct, the issue here is "patchy"
irregularities.
*In particular, does the *local* scale have to get better to keep up
with finer Dawes' limit in ever-larger mirrors?* The circumambulatory
things I've heard aren't as good as more direct answers and examples.
> You don't seem willing to listen, but you also don't seem willing to
> do more than handwaving counter arguements. How about some worked out
> examples.
>
Suggesting that resolution is limited also by ray-trace is not
"hand-waving" because it already is minimized in optical design. The
critics also "hand-wave." They describe the purported alternative in
generalities that don't show us what the net resolution would be. If I
suggest such and such, wouldn't it be more effective for someone to say
"OK, it hashes out to the following level of interaction or combination:
________" They didn't directly. I keep needling them because they won't
do that per se AFAICT.
I did work out an example in my 7/30 reply to Jim Black.
How about some worked examples by commenters? You? tx if you do.
Neil B.
"Skywise" <in...@oblivion.nothing.com> wrote in message
news:l0tcm.108546$Sn5....@newsfe26.ams2...
> "Neil B." <neil_...@caloricmail.com> wrote in news:vtqdnW_dyL0ao-
> _XnZ2dnUV...@posted.widowmaker:
>
>> You keep forgetting, as I said, that I pose this in terms of a
>> Socratic
>> question to get a discussion going.
>
> That could be interpreted as trolling.
>
this is clearly not the first and shouldn't turn into the latter. SM is
a venerable way to get discussions going and it doesn't have to flow
one-way. I clarify, per Marco's question:
> give information directly but instead asks a series of questions,
> with the result that the student comes either to the desired knowledge
> by answering the questions or to a deeper awareness of the
> limits of knowledge.
> So are you the teacher who is trying to lead to otheres to
>
Repeated: I'm trying to do both. I don't *know* who's right, I'm trying
to hash that out through "challenge-response." A Socratic dialog can
application - not the one-sided "teacher" version. I want readers to
Neil B.
BTW this often happens to your comments, are you using an odd news
program or settings?
Spoiler: my basic point is in a few sentences at the end!]
"Benj" <bja...@iwaynet.net> wrote in message
news:6719f918-c9c5-4486...@o6g2000yqj.googlegroups.com...
On Jul 30, 8:44 pm, "Neil B." <neil_del...@caloricmail.com> wrote:
"> This is what I get:
>
> Try 1/20 wave accuracy, which for 500 nm is 25 nm. That is the "layer"
> the mirror surface must fit into. Pick regions about 1m scale and see
> what relative angle distortion is possible from them at a tilt within
> those bounds, approx. It's simple. Multiply by four because (1) the
> region could be tilted either way, (2) the angle deviation of a normal
> ray is double the angle of the surface. Then divide tolerance by size
> to
> get angle in radians."
Benj:
"Boy you sure are persistent! I guess that's good. Consider a large
mirror. Say it has a curvature that is "perfect". In that case the ray
trace CoC has a diameter of? Right, ZERO! But the "classical" area
of the spot size will be determined by ? Right, the mirror diameter
and it's shape and the wavelength. [We are assuming circular mirrors
here] Now consider a "layer" upon that perfect mirror 25 nm thick.
Now we have two perfect figures spaced 25 nm apart. Now assume that
your "real" mirror has a figure that is ground to 1/20 wave which
means that no part of that mirror deviates more than 1/20th wave from
perfect. OK? This means that the figure of the mirror must "fit"
within that layer we've put on our perfect mirror. Then the question
becomes as you have asked, "how much does that "real" surface "tilt"
from the perfect figure?" Now the part that is interesting (draw this"
- OK, thanks for restarting an intelligent conversation. You get my
geometric framing of the issue to that point, even if we won't concur on
the effect of wave optics
"out) is that if the variations of figure are very slowly varying the
length of your possible angle lines keeps them from tilting very much."
- Yeah, but as in my calculation for Jim Black: if the variations are
the tilt of a *given patch size* like around one meter - instead of the
mirror as a whole, or instead of proportionately defined regions - then
the allowed tilt angle is of course a given angle. That's what bothers
me, as per below. I am not asking about overall figure, I am not asking
about surface roughness, but the effects of mid-scale lumpiness.
"But if the mirror has rapid variations (like say sawtooth lines) your
ray trace can angle off at quite steep angles. Hence we can see that
actually the diameter of the mirror is irrelevant to the ray trace"
- You jump too fast to micro-scale stuff and miss my mid-scale concerns.
The diameter matters because a mirror should keep getting better and
better resolution with size. But a given scale "lump" tilted within a
given "layer" like 25 nm, equates to a *constant* angle discrepancy.
"results. In fact if the figure is slowly varying, a larger mirror will
have LESS tilt than a smaller one! And did you notice that the focal
length creates a longer or shorter "lever arm"?"
- but 'slowly varying' is the bug. Given variation over a scale, like
one meter, instead of a percentage of the mirror, it's the same angle
error range.
- Focal length: it all cancels out. A longer fl creates a longer lever
arm which is worse, but the image scale is bigger too. It goes right
into the image scale. That's why the angle, itself, tells you the story!
The angle error displays directly at the focal plane as is, regardless
of focal length. An error (circle of confusion) of .01 arcsec smears
right over a 0.01 arcsec double star, at any image scale.
- Earlier, I wrote:
"> Angle in radians = 2*25nm/1m = 100/1,000,000,000 = 0.000 000 1 radian
=
> 0.02 arcsec. That would spread light out beyond a perfect diffraction
> disk, for a mirror larger than 5 meters. So if the surface had
> variations at that scale it could be a problem - maybe not limiting to
> that large a resolution, but some effect. I don't know, I wish an
> actual
> designer/mfr. would say whether this sort of thing matters. Note that
> the 8-10m scopes we do have, are not classical and have adaptive
> mirrors
> which can correct for more than just the atmosphere as some commenters
> noted.
" Lets not worry about adaptive optics at this point. Lets continue with
the above arguments. What if the mirror surface had a nice sawtooth
pattern (say it was diamond turned). What about the large COC then?
The problem is that as you make these fine highly tilted grooves on
the mirror, it's no longer a mirror it's a diffraction grating. Hint:
Ray trace calculations do not work for diffraction gratings. Well what"
- Sure, but that's microscale roughness. Out of context.
"do diffraction calculations give for this "mirror"? Well, in this case
the angular pattern of the mirror is the Fourier transform of the
amplitude and phase of the wavefront upon the mirror. When there are
high spatial frequencies it results in high angular patterns. This is
obvious as gratings send light off at fairly high angles. Now the
"classical" diffraction limit of a mirror assumes only the low spatial
frequencies generated by the aperture are present. In other words the
transform is of a single spatial pulse the width of the mirror. An
actual mirror will not be that good as higher spatial frequencies are
present and do degrade the performance."
- OK, stuff like that shows you know this subject. Maybe it's offbeat
examples that throw you (or me)?
" So now what about your fixation on ray trace? Well, if one assumes
that our mirrors are nearly perfect and high frequency variations are
not present (the figure varies slowly across the mirror) then sure,
you can calculate a COC that is reasonably useful if you have the
actual figure of the mirror to plug in. And we've seen that generally
speaking that angular deviation will be due to local variations on the
mirror. Now if the mirror is reasonably perfect wavelength wise, we
know that as the diameter increases the "classical" resolution also
increases. This means that as you make a mirror larger and larger, the
spot size gets smaller and smaller and pretty soon the ray trace COC
is larger than the "classical" angular resolution. Which was your
point."
- Wow! So you get that, at least. tx for that and your effort here. Good
grief it's tough to get commenters to chill out - was it Obama's "beer
summit" that inspired everyone? ;-)
"However, we've just seen that the REAL angular diffraction
limit is not set by the "classical" value. The classical value ignores
the variations in the phase of the mirror. So the real diffraction
limit is larger and closer to the COC."
- Oops, IMHO this is where you go wrong. You don't say why the
diffraction could make the resolution *even better* (smaller) than
ray-trace issues. Forget micro stuff, it's the wrong tree. Think about
coma, spherical aberration etc.
"But I know what you are going to say. What if we make this mirror
HUGE so that it's diffraction limit is very tiny! But our ray trace
data shows this large COC. So isn't the COC the actual performance of
the mirror? Sorry bunky. This is where you are stuck. Light is not
perfect rays. Light is not perfect mathematical lines and functions.
Mathematics is not more real than reality. The true angular resolution
will be given by the diffraction calculation and not the ray trace
one."
- Maybe, but see my final challenge. I am wary of that "and not"
wording, the RT has an input and must have "an effect" of some kind -
just consider coma etc!
"I know this is hard for you to swallow but angular resolution has to
do with larger mirrors providing "support" on the sides of the light
beam. It is a WAVE phenomenon. If you use the wrong formula you get"
- Sure, the edge effect etc. but in textbooks they pretend the mirror is
either "perfect" or has overall aberration such as spherical - and in
that case, the wave optics sure as hell don't cure the spherical
aberration etc!
"the wrong answer. The "light rays" model used in ray trace is simply
not a correct one. Thus if you push it too hard, it falls apart as a
number of us have told you. Blindly applying formulas is not
scientific understanding."
- > You also falsely make it appear everyone else agreed with each
> other, but they didn't and that complicates things. Eric gisse
> doesn't think much of your shtick, why should I believe you
> instead of him?
"Eric says I'm not even a good amateur, so why should you believe Me?
Or anyone? Since when is science a matter of faith? Since when does
"belief" enter into this? Eric and I are hopefully not your ONLY
sources of information. Cripes if you are trusting the Internet to
answer your scientific questions you are already in deep doo doo! But
hopefully you will have obtained enough hints here to give you some
places to start digging. What you believe is up to you. I'm not here
to "convert" anyone."
- OK, sure, but then it wasn't stupid not to fold and agree with you and
the others earlier. It can't be both ways. OK.
- So then the "final" scoop (for now!): I will look into the effects the
wave optics have in conjunction with the ray-trace issues, especially
the "support" on the sides of the light beam. But here is what motivated
my question, and it's still a good "point" (no pun intended) IMHO: Find
the ray-trace circle of confusion from the mis-match of mid-scale
portions about one-meter size on the mirror (like you would find the
coma blob but with a random element instead), and you appreciate the
ray-trace implications to that point. Then I suggest: If wave effects
can't make spherical aberration, coma etc, any less, then why would it
make mis-matched light from patchy imperfections any narrower?
- I suggest, only suggest: wave optics can't make the image *any better*
than it would be from rays and aberrations (whether over-all like SA and
coma, or from lumpiness.) If you think the lumpiness is "cured" by WO,
then I question why it doesn't help other aberrations too - why not
reduce coma etc? I don't think it ever makes things better except for
the case of micro variations. And this is harder for large mirrors since
they have a higher diffraction standard to achieve. This in brief was my
whole point.
- Thanks for giving it another try.
Benj
> - I suggest, only suggest: wave optics can't make the image *any better*
> than it would be from rays and aberrations (whether over-all like SA and
> coma, or from lumpiness.) If you think the lumpiness is "cured" by WO,
> then I question why it doesn't help other aberrations too - why not
> reduce coma etc? I don't think it ever makes things better except for
> the case of micro variations. And this is harder for large mirrors since
> they have a higher diffraction standard to achieve. This in brief was my
> whole point.
>
> - Thanks for giving it another try.
Actually the question as to whether or not ray trace aberrations
calculated to be large can actually produce a lower spread of light on
the actual mirror (obviously a diffraction "calculation") is a good
question. I don't think I know the answer and I'm not about to try to
calculate it! One thing I'm certain, is that this comparison very much
depends upon the actual situation.
What I'm getting at is that ray trace represents a calculation of
optical performance with imaginary light of "infinite" frequency.
Because of this, wavelength is zero and no diffraction or other wave
effects appear. So we sort of expect that calculation to be the "best"
we can do. And if the mirror is nearly perfect, diffraction represents
a degradation of performance from the ray trace value. A perfect
mirror gives a zero CoC size, a real image from the SAME "perfect
mirror" gives a spread of light limited by diffraction.
But what happens if the mirror is sort of lousy? We know that when
mirrors are built the usual thing is to figure the surface to some
small spec deviation from a perfect curve such as 1/20th wave. Now
here's where it gets tricky. If you've got a 1/20th wave mirror all
the waves from each part of it will be adding up. And they do it
nearly in phase which means on the sine waves where the slope is
nearly zero. Hence we find the situation where if we have a 1/20th
wave mirror and a 1/40th wave mirror we seen next to no difference in
performance! But on the other hand if you were to ray-trace those two
mirrors the CoC for each would differ by a factor of two because in
ray trace the angles are all perfect, the lever arms as long as you
like and the frequency infinite. In other words with ray trace in
effect the additions are taking place at a place where slopes are
maximum rather than zero.
But the situation is even more complex than this. Because if a mirror
is much less than perfect with spherical aberration, coma, astigmatism
or the like, ray trace computes a CoC. The tendency would be to
compare that to a "classical" value of diffraction. But that is wrong.
Because the mirror is distorted a "true" diffraction calculation would
take into account that distortion with the actual function that
represents the actual mirror surface. The mirror in fact does this
"calculation" when producing an image. So the true comparison is
between a ray trace and the actual image produced by the mirror no
matter what it's shape. So then your question is which is larger: The
ray trace CoC or the actual image produced by the mirror from a point
source? We DO know that as the mirror surface approaches perfect the
diffracted image will be much larger no matter what the mirror size.
But what about when the mirror is full of distortions? You suspect
that in that case the CoC may in fact be larger. I don't know if it
is.
But the two situations are not really equivalent and really depend on
the actual values involved. As for a determination of the actual
values for which the CoC is larger or smaller than the diffraction
spread... Heh. We'll leave that as an exercise for the interested
student...
Neil B.
On Jul 31, 10:40 pm, "Neil B." <neil_del...@caloricmail.com> wrote:
>> - I suggest, only suggest: wave optics can't make the image *any
>> better* than it would be from rays and aberrations
>> (whether over-all like SA and coma, or from lumpiness.)
>> If you think the lumpiness is "cured" by WO,
>> then I question why it doesn't help other aberrations too - why not
>> reduce coma etc? I don't think it ever makes things better except for
>> the case of micro variations. And this is harder for large mirrors
>> since they have a higher diffraction standard to achieve.
>> This in brief was my whole point.
>>
>> - Thanks for giving it another try.
> Actually the question as to whether or not ray trace aberrations
> calculated to be large can actually produce a lower spread of light on
> the actual mirror (obviously a diffraction "calculation") is a good
> question. I don't think I know the answer and I'm not about to try to
<snip - still not getting auto->>s>
> But the two situations are not really equivalent and really depend on
> the actual values involved. As for a determination of the actual
> values for which the CoC is larger or smaller than the diffraction
> spread... Heh. We'll leave that as an exercise for the interested
> student...
Benj, - tx, that's a terrific answer even if you couldn't finally clear
up that issue. I'm going to post a similar question, make sure it's
clearly posed, to sci.physics.research soon. I thought I'd never say
this:
Regards, ...
Neil B.
"Helpful person" <rrl...@yahoo.com> wrote in message
news:83aab74e-4db7-479b...@v2g2000vbb.googlegroups.com...
On Jul 29, 5:08 pm, "Neil B." <neil_del...@caloricmail.com> wrote:
>> "Neil B." <neil_del...@caloricmail.
>>
>> I think it's "sad", that after around 100 comments we still didn't
>> get
>> IMHO a clear and definitive answer. Some people hinted I was right,
>> but
>> no clear affirmation of resolution being the larger (at best) of the
>> ray-trace circle of confusion combined with the ideal diffraction
>> spot.
>> Any good links, to clear exposition?
>> tx to most, for trying ...
>>
> (wavelength) gives the MTF response. Ray tracing is
> irrelevent for well corrected systems.
> In fact ray tracing can show excellent correction whereas
> the true resolution calculated from the wavefront
> may be poor.
than the ray trace spot?
> Simply put, in general, reducing the wavefront rms improves the image.
I guess increasing it, degrades the image.
> www.richardfisher.com
Yeah, dig the site
I read some on those topics but can't yet pull off an adequate
application. Also, a lumpy mirror is not a nice undergrad optics
exercise IMHO, but a tough nut. You and others say I need to do such
calculations but no one else will do them, so if I can cut a corner here
and just ask: do those considerations, mean that the wave optics can
actually "cure" the ray abberations? I mean, can they make the net final
light spot even smaller than from the geometric spread caused by a lumpy
mirror? We know diffraction can and does make the image worse, but (not
counting micro stuff) can it make the image better than otherwise? I
find that hard to believe, it sure doesn't "cure" coma or other
abberations.
Helpful person
>
> Sure, but can that work in reverse - can the resolution be even better
> than the ray trace spot?
As already stated ray tracing is an approximation at best. Ray
tracing can show a pinpoint spot with the true resolution being
poor. The true resolution can be excellent with ray tracing giving a
poor spot.
>
> > Simply put, in general, reducing the wavefront rms improves the image.
>
> I guess increasing it, degrades the image.
Yes
>
>You and others say I need to do such
> calculations but no one else will do them.
You need to understand the basic optics. Without this understanding
you are unable to phrase a meaningful question. Find a good treatise
on aberrations and wave optics.
Neil B.
On Aug 1, 9:36 pm, "Neil B." <neil_del...@caloricmail.com> wrote:
>
>> Sure, but can that work in reverse - can the resolution be even
>> better
>> than the ray trace spot?
> As already stated ray tracing is an approximation at best.
> Ray tracing can show a pinpoint spot with
> the true resolution being poor.
Sure, that much I knew, but then you say:
> The true resolution can be excellent with
> ray tracing giving a poor spot.
That's just the sort of direct, brazen (;-) answer to my query I was
looking for! I find it hard to beleive, but at least it's the right
"kind" of answer! tx for that much ... You must have already seen I
"ruled out" (pardon the pun) micro-scale issues. I meant patchy
(mid-scale) or overall shape. So wave optics can "cure" abberations
somehow? Coma? Have an *example* of a real scope that otherwise might
have poor resolution from mismatch of mirror surface types, spherical
abberation, etc., that actually performs better due to wave optics? How
about if the mismatch was like a 10:1 ray CoC larger than the final
diffraction-produced image spot?
...
>
>> You and others say I need to do such
>> calculations but no one else will do them.
> You need to understand the basic optics. Without this
> understanding you are unable to phrase a meaningful question.
> Find a good treatise on aberrations and wave optics.
OK, but I'm not sure merely "basic" optics will enable to find wave
effects on image formation for a "lumpy" mirror - that looks advanced to
me. And I'm hoping someone (not necc'ly you) will let me be lazy for
awhile and present it here for the edification of me and the readers. tx
again for your efforts.
BradGuth
That's actually not being very helpful, now is it.
In theory, what kind of tightly packed and carefully electro-vacuum-
deposited atoms would it take in order to create the one nm rms
mirror?
~ BG
BradGuth
You're supposed to be like Einstein and already know all there is to
know, and right off the top of your head to boot Therefore few if any
are going to help another soul figure anything out. Consider yourself
lucky that you even got this far.
~ BG
Neil B.
"Salmon Egg" <Salm...@sbcglobal.net> wrote in message
news:SalmonEgg-ABE0A...@news60.forteinc.com...
> In article <pan.2009.07...@lilalaser.de>,
> Kai-Martin Knaak <k...@lilalaser.de> wrote:
>
>> On Thu, 23 Jul 2009 12:54:41 -0400, Neil B. wrote:
>>
>> > if they are simply accurate to within a given standard like 1/8
>> > wave.
>>
>>
>>
>> > But consider that as a mirror gets larger, the tiny variations
>> > in orientation of portions of the mirror should eventually cause
>> > errors greater than the ever-decreasing diffraction spot.
>>
>>
>> > But application of geometric optics to the surface irregularities
>> > should
>> > mean, a circle of confusion 0.01 arcsec diameter.
>>
>>
>> BTW: The reason for large mirrors is not only resolution but also the
>> amount of light gathered.
>>
>> ---<(kaimartin)>---
>
> It has been a long while ago that I duplicated others' calculation on
> strehl ratio. That the ratio of intensity a of a plane wave focused to
> the focal point of a mirror compared to what it would be for a perfect
> mirror. This depended only upon the rms variation of the surface from
> the ideal surface,
>
> That means that as the mirror gets larger, a focused image will not
> deteriorate. The image will always "stick out" the same relative
> amount
> compared to an ideal mirror.
>
> The effect of spatial variation is to determine just how the energy
> that
> is lost is distributed around the image.
>
Late follow-up, if you're looking: does RMS variation per your
stipulation, include taking the scale of the lumps into account? I was
thinking of the effect of a given scale of "lumps", like one meter, not
a percentage of the mirror size. It might make a difference.
Neil B.
"BradGuth" <brad...@gmail.com> wrote in message
news:a74a506c-1206-4efe...@t11g2000prh.googlegroups.com...
WTF? Bullshit. I didn't brag about myself. You look pretty cocky
yourself in your arguments with OG. I asked as a question, I gave some
arguments and mostly got crap back. I went this far because of the very
fact I didn't get clear, direct answers; no calculations, etc. You waste
your time with putdowns and could have at least said if you agreed with
him.
But maybe your RMS question sheds a little "light" on the issue.
BradGuth
> "BradGuth" <bradg...@gmail.com> wrote in message
I was kidding. On the other hand, those others intentionally toying
with you are acting deadly serious.
>
> But maybe your RMS question sheds a little "light" on the issue.
Why yes, of course it does.
~ BG
Salmon Egg
"Mike Dworetsky" <plati...@pants.btinternet.com> wrote:
> You may be thinking of the concept of the Strehl Ratio, which measures the
> ratio of the actual peak of the image of a point source compared to a
> theoretical diffraction-limited image of the same. No telescope ever
> performs 100% in this regard, and the bigger they are, the harder it is to
> achieve good performance without real-time correction of the wave-front.
I do not know if you caught my earlier post on this subject. My point is
that the quality of the mirror finish as opposed to the quality of its
figure DOES NOT limit telescopic performance.
Strehl for a perfectly figured (and I know that is not possible) mirror
is determined only by the rms deviation of the surface from the perfect
figure. Thus, as you make the mirror larger for the same rms deviation
from a perfect figure, the central intensity will go up as the square of
the mirror diameter. This means that the diameter of the focusses spot
changes inversely as the diameter of the mirror. That is, the focal spot
area changes inversely to the square of the mirror diamter.
The spatial spectral distribution of the deviation from perfect figure
will determine just how light lost from the peak spreads out at the
focal surface. In radar terminology, this light would be called "grass."
Bill
--
Private Profit; Public Poop! Avoid collateral windfall!
Neil B.
ray-trace circle of confusion larger than the ideal diffraction spot.
Salmon Egg
"Neil B." <neil_...@caloricmail.com> wrote:
> How about patchy mirror irregularity about one meter size, with
> ray-trace circle of confusion larger than the ideal diffraction spot.
IIRC it makes no difference. The strehl ratio depends only upon the rms
deviation from perfect figure. Remember, however, that rms means
averagingh across the entire aperture. All the light diverted by a patch
reduces the strehl. For small rms deviation, a simple mathematical
approximation can give the answer. In the end,however, the strehl will
be a function of the rms error even for fairly large error. What this
does not tell you is how the off-peak light is distributed. That does
depend on more than just the rms error.