What are good optics?

[koputai]

If you have to pay decent money for 'good' optics, then what's the go with production optics?
Advertisements say things like:
"guaranteed to be diffraction-limited, meeting the theoretical limits of resolution for its size"
So, if these production optics are theoretically perfect, why pay many times more for custom optics?

Cheers,
Jason.

[jase]

Jason,
I'm not sure what you mean by theoretically perfect. Just because a telescope is stated as diffraction limited doesn't mean its perfect. If we take a holistic approach, diffraction limited simply means the telescope has the ability to provide an Airy Disc. Diffraction limited optics are not particularly difficult to achieve from a manufacturing perspective.

A statement by a manufacturer indicating their optics are diffraction limited indicates that the optics are at least 1/4 wave peak to valley at the wavefront for a visible wavelength (usually 550mm yellow-green) as tested by an interferometer. A 1/4 wave peak to valley also translates to other optical calculations/formulas such as 1/27 wave RMS or an 0.80 Strehl ratio. I can provide formulas to calculate between these values if desired. So, these figures defined above are the baseline for diffraction limited optics. If the wave peak is not 1/4 wave (say 1/2) or the Strehl ratio is 0.60, the optics are not considered diffraction limited.

To answer your question, why pay for custom optics if production scopes meet the diffraction limited specification... Simply because diffraction limited doesn't indicate theoretical perfection (as previously mentioned). Theoretical perfection are optics that interferometer test at Strehl ratios of 0.97 or higher. To achieve this level of perfection takes considerable time and effort, something not viable in "run of the mill" high volume production line telescopes. If you want a higher optical quality than the basic diffraction limited specification, then you'll pay more for it.

Why would you want to pay for optics that are higher than the diffraction limited specification depends on your goals. Peter Ward has a great example of optical quality between an SCT (84% Strehl ratio that meets the diffraction limited baseline of 80%+) and RC (99% Strehl ratio) - http://www.atscope.com.au/rcos.html
As they say, a picture speaks a thousand words.

[PhilW]

You might find William Zmek's Jupiter sequence of some interest too:

http://www.rfroyce.com/compimag.htm

[Gama]

In simple terms, the better your Strehl, then more light is focused inside the airy disc to create your star image, and less is used outside to brighten your rings etc which in turn will make your image less sharp.

Theo.

[Bassnut]

Interesting Jase, could you complete your reply on the need for higher than diffraction limited optics "depending on your goals". Such as? (really, I dont know ;-)

Cheers

[Peter Ward]

Don't know that I want to wade into this debate....but like mounts ;) , optics can indeed vary in quality, and if a system is "diffraction limited" who cares if it is any better?

"Diffraction limited" can mean a lot of things, I suspect it is a variation of the Raleigh criteria...ie at this point of optical correction you'll be able to split a double star of separation "x" arc seconds (x being a function of aperture only) but for manufacturers to suggest at this level of correction in optics the image can only be bettered by increasing a telescope's aperture is, well, utter rubbish!

Nature again is rarely so kind!

Tube currents, local seeing & the earth's atmosphere collectively bugger up a telescope's image.

If the quality of that image is already hovering on barely adequate, then all of the above will make sure it is less than that.

If however the optics are near enough to "perfect", the image can be disturbed 20% or so before you know it's not looking too good....

Cheers
Peter

[Starkler]

Good optics have an accurate figure (the right curvature)
Good optics have a uniformly smooth figure without lumps and bumps
Good optics have a well polished surface.

Cheap optics will not have all of the above, even if they do satisfy the term "diffraction limited".

Is the difference visible? Heck yes! It was immediately apparent to me the first time I pointed my new dob with Suchting primary toward Jupiter.
Even in poor seeing the view was much cleaner without anywhere near the amount of scattered light lighting up the background as i see with my gso diffraction-limited dob.

[Bassnut]

Peter, OK, so 80% plus optics (diffraction limited base line) have less "headroom"" than say 97% optics. ie allow some 20% more "atmospheric disturbance" before image deteriation becomes obvious.

Hey, im not argueing, my RCOS always produces tighter images over my LX200R optics in the same conditions, much tighter ;-).

Cheers

[mick pinner]

astronomical products are not immune from the fact you get what you pay for. advertising claims are not always what they seem, do your homework.

[wavelandscott]

I won't delve into the science of the optics...many other folks are far far better qualified than me to do that.

The "science of perfection" aside (and I do beleive that there are differences), think of it in terms of cars...there is transportation that can be purchased on a budget or transporation that can be purcahsed on a BUDGET!

If you are buying your first car and have not had much driving experience a small "b" budget car can provide transportation...I liken this to the mass produced scopes. They work and they generally get the job done.

However, at some point in time, your transportation needs and interests may change and when they do, then it is time for a big "B" budget car. Similar transportation but subtley (maybe or maybe not depending on the size of the budget) different.

Is big "B" gear better than small "b" gear? Strictly speaking yes for all of the reasons that others have listed...

However, don't ever confuse "better" with "more enjoyable" depending on what your interests, budget and objectives are, small "b" gear may be perfectly adequate and there is nothing wrong with that at all. The more important thing is to get out under the stars and enjoy them.


Clear Skies...

[rally]

Jason,

This might give you some insight into quantifying good optics.
http://www.rfroyce.com/standards.htm

The current best (single) scientific criterion is the measure of the Strehl ratio of the optics, it is a measure of how well the scope is able to take all the light (energy) coming into it and focus this into the (first order) airey disc.
The ratio is a function of what the scope can achieve compared to the theoretical maximum that is possible.
The article attempts to explain it and also to correlate that back to the traditional 1/4wave to 1/10 wave optics style of measurement that most people are familiar with.
Any optical imperfections and aberrations, whatever they are, will affect and reduce the Strehl ratio, but stated simply you just can't get a high number unless the optics are very good.

Of course there is more to it than only this, but generally if the Strehl ratio is above 95 then you have a very good scope, greater than 99% and its near perfect.
This is vastly different to a statement that a scope is "diffraction limited" - whilst not totally meaningless, it says almost nothing really about the real quality of the scopes optics.

You might be horrified if you knew what the numbers are for a lot of scopes purporting to be very good by hiding behind meaningless waffle.

The sooner all scope manufacturers start putting useful numbers (like strehl ratio) behind their advertising the better - as it stands now many buyers are being mislead by the omission of this sort of information.

Hope that helps

Regards Rally

PS Another useful measure for assessment (especially for refractors) is looking at the ray fan plots of coloured light to see how well they focus - as a measure of the degree of chromatic aberration. Some are extremely well corrected, most are not. But you will pay substantially more for those that are corrected.
Once again getting this info is difficult if not impossible.

[Peter Ward]

Amen to that! :)

[Satchmo]

The star test is a much more sensitive and easier to carry out comparison between scopes . all the aspects of a superb mirror can't be quantified and contained in one strehl number just by zernicke polynomial point fitting of interferograms. Its not the only basis you would use for for comparing good mirrors from excellent ones as there is so much more involved to a great mirror than trying to fit a polynomial to represent the overall shape using a mathematical approximation.

You only have to compare the synthetic fringe output which is a reconstruction of the interferometer fringers from the polynomial modelling data. Fringe point fitting analysis is lousy at quantifying turned edges and surface roughness, but good at measuring basic spherical aberration.

[Peter Ward]

Bollocks! :)

That is the whole point of the Strehl ratio. Provided sufficient data points are measured (typically 2500+) all aberrations are indeed accounted for and quantified.

The star test is a great *qualitative* test for the optic when it is out of focus...what happens in focus is much more valuable ;)

But don't take my word for it. There is this little company just outside Chicago who seem to know a thing or two about optics....

http://geogdata.csun.edu/~voltaire/roland/startest2.html

[Satchmo]

Peter wrote : "Bollocks! :)" Well bollocks to you too !:thumbsup: and Happy New year ;)

Only 100 to 200 points ( not 2500+) are typically fitted which is not enough point density at the edge to define problems during the polynomial fitting ( count the number of points fitted on the two interferograms on your web page). Note also that the surface micro roughness in the C8 interferogram is not picked up by the fringe fitting, as it is completely invisible in the 3D wavefronty map, all that are obvious in the wavefront map are primary and higher order Seidal aberrations. .Polynomial fitting to interferograms usually ignors surface roughness and highly sloped edge problems. I can post interferograms to illustrate these issues.

I note that astigmatism has been switched off from both data sets presented . Its rare that consumer interferograms even mention if astigmatism has been subtracted from the data.

Of tests available to the end user , only a star test or ronchi test under excellent seeing will generally show how good the edge is and if the surface is smooth showing nicely defined fresnel rings with the star defoccused. Two mirrors can have equal high Strehl ratio interferograms but only star testing reveals the `gem' which will reveal the better low contrast planetary detail on a superb night. Remember that a Strehl ratio , even in its idealised form, only tells you about the light concentration in the Airy disc, and nothing about the distribution of light that doesn't make it.

I've learnt this through 20 years experience with interferometry and star testing and have tested a number of mirrors that didn't always meet their owners expectatons but looked good `on paper' . As always let your eyes be the final judge of quality .

[jase]

While I do agree with you Mark, not enough points can lead to inconsistent output of the interferogram data. I would be brave enough to say that when entering into the high quality optical market, there are few that provide only 100 to 200 point maps. Those that do, you'd surely be foolish to trust given you're probably about to hand over a considerable sum of money and you'd want quantitative evidence of the optical performance.

Star Instruments who are one of the leading manufacturers of Ritchie Chretien and Cassegrain mirror sets perform a 2500+ point test as indicated by Peter. These mirrors are deployed in the RCOS series.

Information on one of their 20" mirror sets can be seen here - http://www.star-instruments.com/interfero.html - Note fringe analysis sheet - Number of pts.

An interesting statement on the Star Instruments site:
" STAR INSTRUMENTS guarantees a minimum of 1/4 wave front, 1/20 wave r.m.s. (http://www.star-instruments.com/rms.html) on all systems.
STAR INSTRUMENTS continues to be concerned with the false advertising claims being made by amateur optical suppliers who claim 1/10 to 1/20 wave optics. These claims tend to confuse the amateur astronomer into believing you must have 1/10 wave optics, when in reality there are very few 1/4 wave 8" and larger optical systems. The fact is that Pyrex is not a zero expansion glass and, therefore, cannot keep a figure better than 1/4 wave front."
So I wonder how many commercial instrument manufactures out there that are using Pyrex and making such false claims... With the term "diffraction limited" meaning absolutely anything (well almost), it makes it difficult for a future purchaser to make the right decision.

I've read the book Star Testing Astronomical Telescope by Harold Suiter. Great read for those interesting in optical testing, however for someone after quantitative optical figures it leaves a little to be desired. Is the CCD camera not more discerning than the eye? I'm surprised by your statement that the only way to validate optical quality is the eye.

[Peter Ward]

More Bollocks & Happy New Year indeed! LOL :)

I'd happily put my money down for any Christen or Jones optic that quantifies and guarantees the figure.

Saying both Star Instruments and Astro-Physics only use a few hundred test points is patently false.

As Jase points out, they use thousands of data points....if not, then what you describe is a possibility....but clearly not so with the above two manufacturers.

The qualitative road you suggest is fraught with dangers...a local manufacturer's Dall Kirkham comes to mind...despite costing many $thousands it was so bad they had to take it back after legal action was
brought against them....

[Satchmo]

Hi Jase , Thanks for your thoughtful comments. Regards to fringe point numbers I didn't say less points leads to inconsistant results, I maintain that edge data often doesn't make it into final consideration at all due to lack of sampling density.I used the example interferograms from Peters site becasue re fringe point #, that is what is provided in the interferometry demonstration on that site.

The link to Star Instruments shows a fringes that have been analysed with phase shifting interferometer..which is why there are so many points: companies that amateurs typically buy from use standard fringe analysis . This is still only slightly more than one data point per square cm of mirror area on a 20" which is only going to be able to mathematically model gross surface roughness if present, not compare mirrors with medium and fine scale roughness.

RCOS / Star may well be right that large equatorially mounted Cassegrains using thick 6:1 ratio blanks in Pyrex might find it hard to settle within 1/4 wave if the temperature is moving around. Users of large altazimuth scopes using Pyrex in much thinner ratios of computer designed cells don't have to contend with such thermal issues. In the workshop enviroment at least , I have no trouble stabalising 12:1 ratio Pyrex mirrors to an aspheric correction of 1/10 wavefront which is the point in th eprocess I stop quantitative measurements of residual spherical aberation shadows and concentrate on achieving an excellent optical Null.

My point about the validity of evaluating good mirrors from excellent mirrors, by eye, is that the more subtle features such as edges and surface smoothness are just not necessarily getting picked up by interferometry and reflected in the calculated Strehl ratio, but will be obvious in a star test on a good night. Traditional methods of knife edge and eyepiece under optical Null are ample to give quantitative and qualitative measure of a mirrors quality. Interferometry IMHO is icing on the cake for any company that knows what it is doing and been around for along while.

[Peter Ward]

Mark...I am curious...do you actually have a phase shift rig? I spent a good deal of time a few years ago trying to get a questionable optic tested in Australia, and could not find anyone (including the CSIRO) with anything larger than a 4" Zygo.

But I digress. Both Star and AP get the required density by taking many samples over multiple orientations.

And the coup de-grace for smoothness would be Ion-Milling offered by RCOS....a process hard to replicate by pushing glass and using a null test ;)

[jase]

Thanks for the detailed explanation Mark. It has provided clarity (for me at least). It would appear there is no "value for money" principle when it comes to high end optics.

If all manufacturers tested their optics to the same capacity, then I'd suspect we'd feel more comfortable about what we are purchasing. Perhaps its best we don't know as we'd all filing law suits against the large, high volume manufacturers for not delivering what they state. Reminds me of the Parks Optical debarcle around five years ago, when they stated their optics were all 1/10 wave, but in fact most were pushed out of the factory as 1/4.

[Satchmo]

Peter, I used a Wyko 6000 Phase shifting rig at the CSIRO in Lindfield for 10 years. We had an F3 diverger lens that was quite happy testing up to 24" on the anti vibration damping bench.

Was your bum scope a C8 ?. I recall giving a quote of $400 to do the testing on the testing on the Wyko 6000 back in the early to mid 90's?

I don't agree that 1 sample point per cm is enough to characterise an edge defect and it will only diagnose very coarse figure roughness of a scale of 1 cm or more. Rotating and stacking will tend to smooth insufficient data not increase its accuracy, though it is good at removing transient atmospheric effects.



Ion beam milling still relies on feedback results from a Null test ( usually autocollimation against a large flat for Cassegrains) and the results showing where glass needed to be polished off would be just as obvious to a skilled optician using a humble knife edge at the Null focus, as placing an interferometer. As Leon Foucault discovered a knife edge is capable of revealing small period defects of less than lambda/100 ;)

[Satchmo]

Jase, I think it comes down to the old adage, 'you get what you pay for'. As an amateur astronomer I wouldn't feel conned if I bought a 12" scope for under $1000 and it was an honest 1/4 wave or 0.8 Strehl. If as a consumer I payed twice that for a 12" mirror I would have higher expectation.

I don't think there has been a better time to be an amateur astronomer. Good optics in real terms a fairly cheap, and the price point vs quality is pretty well known . There are not really many `dud' scopes or optics out there, but like the world of HI FI you pay a lot to gain another 14% signal to noise. :)

[Peter Ward]

Mark,

CSIRO's optics lab at I understand no longer has the Wyko, and gather that you no longer have access to it either ( wonder what happened to it?)

I don't think I ever contacted you over a C-8.....but was keen to get some locally made RC's happening when the $A was at 49 cents.

The piece by Roland Christen I pointed to earlier very much echoes my experience in star testing optics. I have known an optic to be excellent (via test data) yet the star test was a little weird.

Imaging with the same set convinced the the test data was right all along.

Peter Ceravolo (who my liver knows well) is also adamant that what a telescope does out of focus (i.e. star test) is of far less of a concern than what it does in focus.

So I guess we'll have to agree to dis-agree :)




[quote=Satchmo;282671]Peter, I used a Wyko 6000 Phase shifting rig at the CSIRO in Lindfield for 10 years. We had an F3 diverger lens that was quite happy testing up to 24" on the anti vibration damping bench.

Was your bum scope a C8 ?. I recall giving a quote of $400 to do the testing on the testing on the Wyko 6000 back in the early to mid 90's?
quote]

[GrahamL]

nice scope Jason.lots to see with some dark skys and what you have got:thumbsup:

[Peter Ward]

Something didn't ring true here, so I did a quick check of the Ion Beam Figuring literature, and almost without exception IBF facilities used interferometric data to build a beam time erosion matrix....and then apply two or three iterations, (interferometrically checking after each pass) of the ion beam to achieve effectively molecular level precision.....not a mention knife edge test anywhere..... gosh

[Alchemy]

another opportunity for people to say ... my 20 000 dollar scope is better than a 600 dollar scope ..... well duh:P youd be an idiot if it wasnt

[theodog]

I agree. I grow tired of this one-up-man-ship!:sadeyes:

"Good optics have an accurate figure (the right curvature)
Good optics have a uniformly smooth figure without lumps and bumps
Good optics have a well polished surface." (Starkler 26/12):thumbsup:

I thought the origional posting was for 'good optics' -not perfection.
:hi:

[Satchmo]

Peter

An intereferometer is just a device that goes at the focus of a converging light cone and tests wavefront disparity, just like a knife edge. When testing large optics in a workshop, a Null test artificially produces aberration opposite to what has to be polished in, so spherical aberration will be appear nil when the aspheric is polished in. The only other way other than using mirrors or lenses to Null is a Holograpghic reflective null grating.

Null tests are still used in 99% of pro situations, whether or not an interferometer or knife edge is used at the focus.

What exactly is your problem with Null testing anyway ? Is it not high tech soundiung enough ?

[Peter Ward]

Mark,

I don't have a problem there at all. The point I am making is you need to quantify the error you need to be able to express it in real terms.

This seems at odds with your comment earlier that only the star test matters.

Sure, a gross error check is useful and certainly would have been a pretty good idea with the Hubble! :)

What I am saying is if you're milling optics down to molecular levels of precision, the testing needs to equally accurate.

[Peter Ward]

I have used $600 scopes that have had simply excellent optics. Price is not the issue. I would like to see however more meaningful ratings of optical quality so you can be confident the $400 or $40,000 scope being purchased is going to work well.

[Satchmo]

Peter.




1. CSIRO Center for Precision Optics still have the Wyko 6000 or an upgraded model of it and use it frequently. Where exactly do you get your information from ?

2.I've never even asked for access to it or needed it in the last 10 years...and I've never claimed that I used it other than the 10 years I worked there so I'm not reallly sure what your point is there ????

I have a very nice interferometer of my own thanks ( for the last 14 years and currently upgrading it ) , and it doesn't show me anything that the knife edge and eyepiece shows clearly and quickly, I don't use it routinely, I am far more interested in monitoring the things it doesn't measure well.

3.Rolands article is specifically not on the subject I have been discussing. I refer to mirrors of high Strehl ratio by interferometry that have obviouss surface rougness and edge problems. Roland talks about weather or not to polish out high order aberrations , that customers can see in the star test. Not quite the same thing....

4. I have lots of respect for Peter. AT the end of the day, if you are wondering *why* your scope has no contrast in focus then defocus and look at the contrast in the fresnel rings.

IMHO A knowledgeable star tester can always pick a superior mirror , and an optician that pays close attention to the star test can make a better mirror. I've got 30 years of observing and 20 years practical experience in manufacturing and testing optics. I look at star images in and out of focus , day in day out. There are certainly minor aberattions that can be seen in the star test that won't affect in focus but there are many common aberrations which do effect contrast infocus , it just takes experience to know what is significant.

[Peter Ward]

Mark,

I contacted the CSIRO, albeit 2-3 years ago now wanting to get some medium aperture certification done. In short they told me they simply didn't have the equipment to do it anymore....clearly I was talking to the wrong person! (note to self: contact them again in 2008)

I would have been surprised if you did not have an interferometer in your shop and find it interesting that you place equal weight on other methods.

As for my (current) personal telescopes, they all star test well, and the test data on them simply confirms this.

I have no doubt in your abilities as a mirror maker. Similarly I have no doubts about the expertise of other opticians in the US and Canada such as Jones, Ceravolo and Christen.....who all do a star test as well....but I know for a fact sweat over their interferograms when trying to get that last bit or correction.

Clearly there is more than one way to skin a cat here ;) and I am sure you will continue to make fine optics for some time to come, as will AP, Star Instruments and Ceravolo.

[Alchemy]

it would have been nice to have perhaps discussed the benefits of say BAK glass vs pyrex, or different types of mirror coatings, triplet vs doublet arrangements( if you want to go to refractors), thermal management via fans before or after the mirror, perhaps cardboard tube vs metal tube, baffling and its advantages,secondary mirror size, effects of scintilation on mirror size, adaptive optics benefits,closed cell vs open cell, mirror thickness and cooling times etc etc etc

without beeing disrespectful, can we move forward.... i grant it would be nice if there was a standard to compare at point of sale but there's not it seems.

[netwolf]

Is there no IEEE or similar organization for Astronomy optics that provides some standards? If not perhaps the peers of the industry need to consider coming together and forming a body to do so.

[rally]

Trying to stay on topic, as I do feel a little sorry for the original poster whose innocent question has precipitated the current thread.

"What are good optics?
If you have to pay decent money for 'good' optics, then what's the go with production optics?
Advertisements say things like:
"guaranteed to be diffraction-limited, meeting the theoretical limits of resolution for its size"
So, if these production optics are theoretically perfect, why pay many times more for custom optics?"

What we were trying to do was to correct the general misconception that continues to be promoted or at least inferred by many manufacturers and some vendors that "diffraction-limited" means the scopes optical quality has reached the theoretical limits.
This is simply false and an error by (almost) an order of magnitude to theoretical perfection.

Unfortunately the amateur field is influenced by this marketing hype and it affects people's buying decisions resulting in a lot of wasted money and time. Not everyone consults a forum or is capable of getting a credible opinion before buying a particular scope, and so these poor souls buy a scope that based on their best ability to judge - should be good, and get let down, they either discard the scope or lose interest entirely or they then go and buy further scopes eventually ending up with something that is good - at considerable expense and lost opportunity.
During this process they will have expended far more than buying a 'half decent' scope in the first place.

The use of Strehl Ratio as a term to evaluate the quality of optics effectively takes into consideration all of the various optical problems raised in this thread. (although it doesn't define any of them specifically)
This is a term used scientifically and commercially.
For those that haven't read the linked article I posted - it really is worth reading.
http://www.rfroyce.com/standards.htm

However a given strehl ratio doesn't tell you what particular problem a scope has, or help you to decide which is a better scope if two scopes have equal strehl ratios. As has been stated there is more to it than just this alone.
Scopes having a lower end Strehl ratio - say below 0.80 might need to have additional information to explain their aberrations and faults more specifically.

Until amateurs force the issue and demand that the vendors of scopes provide a meaningful measure of the optics they sell, there will never be any drive for them to do so.

If every amateur new scope buyer said - "What is the "Strehl ratio" of this scope and if you won't tell me I wont buy this particular scope", the world would change in an instant and we would all be much better off.
Resellers would be going back to the manufacturers asking for useful technical information, those that can't (or more likely won't) provide it simply wont sell scopes.
I wont even attempt to address the reasons why scope manufacturers wont provide this sort of info.

The argument can continue about what is the best measure, but the Strehl ratio is really the only one that considers the entire optical train's ability to focus light where it is supposed to be focussed.
We might then also ask for a ray fan plot to see how well the scope is able to focus light of different colours (but at what scale and what wavelength)
Others may still want to know the peak/valley wave values etc etc

But with just these two things we would be able to make a better buying/comparison decision.
Forums would then fill with comparisons of scopes by Strehl ratios (and other factors) and the amateur market would have a better benchmarking system to go by.

Strehl ratio is by no means the 'be-all and end-all' measure but it is the best currently available method for quantifying the scopes entire optical train. It also allows the comparison of quite different types of scopes - reflectors and refractors.

As it stands now - there is currently speculation, highly subjective comparisons , hyperbole, outright misleading information and a small amount of technical information that (is often hard to come by) and not always necessarily useful.

At the end of the day the buyer will have a budget to keep within and price/value decision to make.
But at least he/she will have a better idea what he/she is getting and a better method for evaluating scope optics by using a simple number between 0.0 and 1.0 (even if they don't understand what it means)

Hope that helps

Rally

PS - I was reminded by someone much wiser than I, that the ultimate goal of the scope is to resolve images and that one of the good measures of this is the modulation transfer function or MTF - used by most camera lens manufacturers.
MTF is usually presented as the optics ability to resolve a grid of closely spaced black and white lines, the closer the line spacing a lens can resolve at a given distance and the better the contrast of the final image the better the lens.
This has its own set of problems as different manufacturers perform the MTF analysis with different line spacing and omit to include this most relevant fact on the graphs in their colour glossy !
But its a good start and helps for more meaningful comparisons.
Once again on its own it doesn't specifically identify what problems exist but it does show what the entire optical system can do warts and all.

[Satchmo]

Every one agrees that Strehl ratio is a step ahead, unfortunately some the traditional methods of aquiring that value are just not sensitive enough to measure a good from an excelent optic as there are so many more subtle features seperating mirrors at the hifgher end.

You'll note that Royce himself does not measure 'Strehl ratio' or supply interferometry data. Its simply not adequate to seperate wheat from the chaff amongst all the vendors supplying Strehl certificates. He simply guarantees, that if push comes to shove, his optics, if measured by conventional techniques ( ie fringe fitting interferometry, will be better than 0.94 for 14" and above.

[KenGee]

I'm going with Mark on this one, The companies suppling high end scopes are just as liable to marketing hype as any other. I often get the same feeling as I get with audiophiles. What's the point of ion milling if in the real world you will never recreate the conditions again to see that prefect figure? It's a bit like buying wooden volumeknob for your amp, pointless. But as long as their are people willing to pay, and it's away to different yourself in the market then it will be done.
As for the benefits of a "standard" way of measuring and quoting telescope performance. Even if there was one how many of us would use it? How many of us use the fuel economy of a car in our choice? Those that do how much weight do you put on it. How many of us have bought a scope because of is price, looks, or simple one up manship.
Now to answer the thread starters question. IMHO it's a bit like camera lens, the run of the mill lens will do just as good a job as the really costly ones however the dear ones will always be just that little bit sharper, have just that little bit more contrast. When that becomes important enough to you to pay the extra you'll pay it (if you can afford it). But do think that having good glass will magically turn you into a great imager, nor will "poorer" glass keep you from making great imagers.
Having just spent a small fortune on a refractor, mount and observatory I find it’s of now more use to me when it’s hot or cloudy as my much cheaper setup. :thumbsup:

[Peter Ward]

With regard to Ion Milling.... its simple, on many nights you can see the difference, get less scatter, tighter stars and can go deeper with the same aperture/exposure.

As to whether that level of performance is valuable to you, is a bit like asking: why buy a Ferrari when the speed limit is 100km/hr? To most, an extravagant waste. And there is some truth there to be sure.

Then again, I occasionally get away from the kindergarden roads in Oz and onto an Autobahn. SLK500's are pretty sweet at speeds our PC brigade insist cause instant death at home ;)

[allan gould]

Peter
Thanks for that link - most interesting.
You do learn something every day.
Allan

[ausastronomer]

Peter,

I don't agree with this in any way shape or form.

Unless the numbers for each different optic are calculated by the same person using exactly the same test equipment under identical test conditions with identical adjustments and numerical manipulation, the numbers themselves are meaningless. Similarly, comparing one set of numbers against another is pointless.

eg: OMI tests their mirrors on a Zygo, anything over .90 strehl is an excellent mirror. Zambuto uses zonal focault testing and hasn't sent anything out of his shop with an "estropalated" strehl less than .97. Which mirrors are better? The numbers tell you nothing because of the different test methods and conditions. Suffice to say both are excellent. Similarly, which scope is better ? One Peter Ceravolo has tested with a strehl of .96 or one which Roland Christen has tested with a strehl of .98? Again, no way to say because of different testers, test equipment and test conditions. Again suffice to say both will be excellent. The raw numbers IMO are pretty meaningless for comparison purposes, but, if the test methods and conditions are known, they do give you an indication as to the overall quality of the scope. ie. bad, fair, good, excellent.

Cheers,
John B

[koputai]

Thanks to everyone for their replies. I didn't intend it to be the sh*tfight it turned out to be! I've learned a bit, and need to learn a lot more.

Thanks,
Jason.

[Starkler]

As long as you had your question answered Jason. ;)

[Peter Ward]

John,

So are you saying test data provided by (skilled) opticians is some sort of random number that in no way correlates to the quality of the optic?

As I mentioned earlier there are many ways to test optics, and I'd certainly agree some are more subjective than others, but I would not suggest that various quantitative methods are in some way divergent.

An excellent optic makes itself apparent no matter how you look at it.

To paraphrase Ceravolo, interferometry is not necessary to make excellent mirrors, he in fact uses the Foucault test to actually make the optics, but then quantifies the results with interferometry to ensure the optic meets the required performance as it is less subjective than the star test and easily quantified (by him at least)

In fact Ceravolo did a piece for Sky and Telescope (March 92?) where telescopes with mirrors of varying (but known) quality were used side by side to see if anyone could tell the difference visually.

In this case the numbers were not just relevant, they demonstrated perfectly the difference between "bad, fair, good and excellent optics"

[Aster]

One of the funniest and most useless threads I have read since the comparing of the EQ6 to the Losmandy Mounts.

Jason, How good is your eyesight, or anyone elses for that matter ?

Unless you want to see your name in print and tell everyone within hearing distance how good your equipment and DSImages are, even if that is so, then any of the Known Manufacturers Optics will be OK. Yes, there is a chance that you could get a lemon but most will exchange or give you are refund.

Mind you, there are also lemons produced by Manufactures who charge ten times of what the everage person spends on his hobby.

More then 90% of the populace wouldn't know the difference between 1 wave optics or 1/50 wave optics. More then 90% of the time you can't even use the full potential of 1/4 wave optics due to atmospheric seeing. Then there are heaps of people who believe anything some people say. Then there are other people who have heaps of money to waste and will tell you the only car to drive is a Rolls.

I don't even know why Mark, who has more experience in mirror making and testing then 99% of the people in this forum, even bothers to get involved in discussions like this. It's a nowin useless exercise. . Unless you have actually tested a mirror yourself, with home made or the most modern equipment, you have no idea what you are talking about. Reading about it makes you an armchair expert, which is worse then a nonsmoker who gave up smoking.

Mirror making was a hobby of mine for near on 30 years, in that time I made somewhere around 250 plus mirrors from 4 1/4" to 16" including 16" RC and SC. I had also the priveledge to meet some of the top mirror makers in that time, and even work with one. All my mirrors were tested in this order, Ronchi, Knife Edge, Null. The only test which gives the same results, temperature controlled, is the Null Test. Any other test is subject to the interpretation of the person and equipment used doing the testing. I have had the priveledge to be present when three highly experienced opticians tested the same mirror with the same equipment in the same locallity and finished up with three different results, mind you, the were still still within % points of each other but still different. The Null Test gave the same result by all three. Strehl is a good rating for a mirror but nowhere near perfect. Mirror with a slight turned edge still had a Strehl of better then .9. It was clearly visible with a Null Test and rejected.

There are that many variables in optical bench testing, setup, quality of equipment, person doing the test, etc. that the only real test is the instrument performance in the field, and that is a Star Test AND How Good Is Your Eyesight.

[casstony]

Probably true, but if you set up a sharp scope beside an observer who is accustomed to using his soft scope, he will be able to see the difference.

[Stephen65]

Depends on where you draw the lines between good/OK and OK/lemon. I've read a few horror stories on CN about people buying SCTs from the big two manufacturers only to discover by star testing that they received scopes are that are, generously, very mediocre optically. But provided the scope passes the low threshholds set by those manufacturers then they will not exchange them.

Higher end scopes are not completely immune from optical defects but the variability in quality is much less and accordingly your chances of getting one that's subpar are considerably lower.

[ausastronomer]

I am saying the numbers are a lot less relevant and meaningful than many people perceive. They are good to compare optics tested by the one person, under identical test conditions. They are not good to compare optics made and tested by different people under different conditions, because of inconsistency in test methods and test conditions.



The article did appear in the March 1992 edition of Sky and Telescope. It was in fact a joint article by Peter Ceravolo, Terence Dickinson and Douglas George. I have read it many times.



Of course they demonstrated things perfectly. All the mirrors were made and tested by the same skilled person (Ceravolo) on the exact same equipment, using the same methods and conditions. This is the exact point I am trying to make and that Aster, Mark and others also referred to, and you are missing.

Cheers,
John B

[Peter Ward]

John,

I clearly am missing the point, as it seems like you are having an each way bet here. :)

The fact there is variation comes as no surprise, but a skilled practitioner will know from one of his measurements what to trust and in what circumstances.

I would not trust an optic that was only subject to one test. The opticians who I deal with make as many tests as possible....which is a very different regime to mass produced optics which I have seen first hand to be a cursory inspection at best.

I suspect therein lies the rub: testing=time=money.

Skilled observers *can* see the difference between average and excellent optics, and simply saying an optic is "diffraction limited" or 1/4 wave will shed no light on whether an optic is excellent or not.

When Jones, Ceravolo Christen et al. tell me an optic has a Strehl of 0.96, while you may not accept it, I tend to believe them.

Plus it beats flying to Chicago or Atlanta to perform a star test on the instrument before they ship it out :)

[ausastronomer]

Peter,

I also believe them as it relates to a relative measure with all the other items they each have tested and certified. My point is this. Is a Ceravolo optic of strehl .98 better than a Christen Optic of .97 strehl? Be rest assured you don't know, I don't know, and most importantly, we are unlikely to get seeing conditions good enough to tell them apart more than once a decade, if we are lucky. The only way to tell them apart is if the one optician was to test both optics under the same conditions and on the same equipment. Consequently, as a relative measure the numbers don't mean much.

I am about to order a mirror from Mark and I won't have a clue what the strehl number is; and I don't care. Suffice to say that I have looked through enough of his mirrors and mirrors from a lot of other premium opticians, to know that he doesn't make any "bad" ones. In fact he is easily in the top 10 glass pushers on the planet and that is more comfort to me than a piece of paper with a number certified under variable test conditions. Similarly, if I buy an optic made by Christen, Zambuto, Steve Kennedy or Peter Ceravolo, I don't need to look at the strehl number to know that it's as good as hands can make it and I am happy to buy it.

Cheers,
John B

[Peter Ward]

No argument from me there! You gave me the impression much larger values would go un-noticed.

To ruminate on a 0.01 difference would be ludicrous.

BTW Ceravolo just e-mailed me he never Star tests (!) ....does a whole bunch of other stuff... but commercial constraints require him to produce certified optics even when its raining or the seeing is crappy!

Agreed there are many producers of excellent optics out there, with well deserved reputations....and, Strehl ratios aside, perhaps that is how a novice might best find his or her dream telescope.

Have a good one!

[ausastronomer]

Funnily enough I was led to believe that the people that cared enough about the quality of their optics to order large thin parabolic mirrors from people like Mark Suchting, Steve Kennedy and Carl Zambuto without strehl certification, had progressed past the novice stage. Each to their own.

John B

[Bassnut]

This has been a very interesting thread, I have learnt much, thank you gentlemen.

If a University/Gov dept/Professional Obs/Military purchased a premium scope, surely they would require absolute certification of some sort for an approval process. Seems from yr discussion, organisations such as these would not rely on reputation or a relative/subjective test, so then wouldnt a many 1000 point strehl test be the most comprehensive of all of them, that can be described in absolute terms. It seems to me a glass pusher, such as some here, would have access to/intuatively understand/require different specifications to a commercial maker selling to a competative global commercial market.

[Aster]

Sure they need certification and usually will require the optics to be of a certain specification. But it doesn't matter which way you look at it, the are still only a set of numbers. It's like "The average wage, etc. is....! Numbers are Numbers, there to be manipulated.

The top opticians didn't get their reputation by producing .99999 Strehl Test on a testbench but by the actually use of their optics in the field.

How good the optics - equipment will actually perform can only be tested in the field, Not on a testbench, which only will give you an indication of how well they are supposed to perform.

[Peter Ward]

So are you saying there is absolutelty no correlation between test bench figures and field performance?

[Aster]

Read the last paragraph of my post.

[Peter Ward]

O.K. the numbers give an indication of how they are *supposed* to perform.

How much of an indication?

[Aster]

What do you want Peter, a definition of the word indication?
I am afraid you have to pick your own, the same applies to numbers.

If you can't get it through your head that numbers alone are not the final result of good optics then I am wasting my time.

[Peter Ward]

Similarly, if you think a bad optic will somehow pop up up with a 0.99 Strehl then I am wasting mine.

I have star tested only a few scopes with the above spec. They were indeed wonderful.

[Alchemy]

who would have thought this would be still running.

what would happen if the perfect telescope were taken out in poor- average seeing conditions.

is it true that the worst part of a scope is the atmosphere?

[wavelandscott]

I have followed this thread with some interest and I am glad it has remained civil.

It seems from the discussion that there is no "single" consistent measure currently taken that can be compared across a variety of manufacturers. Due to differences in testing and testing equipment direct comparison can be difficult.

I'll ask a slightly different set of questions that I think is still in line with the thread...

Can you have a bad mirror with a .99 strehl (or .95 or .96 etc. etc.)?

Is it reasonable to assume that anyone that could/would create a mirror with such a high strehl mirror would automatically check for these other defects that would cause such a high strehl mirror to be "bad"?

Just looking for some helpful guidelines/rules of thumb on this as I now beleive that this may have been the original intent of the first post...and I do recognize that this may be a situation where "if you have to ask, you aren't experienced enough to see the difference"...

I stand by the implication in my earlier post that not everyone "needs" (desires/deserves/can afford) ultra premium gear but that should not disqualify them from enjoying 97.8734% of the things to see in the hobby. Like an automobile, sometimes a Kia is sufficient...

[Alchemy]

I couldnt resist posting this quote from Alan Adler (us sky and telescope issue jan 2002) a respected contributor to sky and telescope both US and AUS for many years i have enlarged a particularly interesting section relevant to this discussion

The full potential of your Newtonian can be realized with only a small investment of time and money. By Alan Adler interest in thermal management began when S&T associate editor Gary Seronik played me a video made by Bryan Greer that showed heat waves slowing rising from the front surface of a 6-inch mirror. F or years reflecting telescopes have taken a back seat to refractors for high-magnification views of planets and double stars. The reflector's central obstruction is most often blamed for this shortfall. However, a strong body of theoretical and experimental evidence has shown that central obstructions of 15 percent (perhaps even as high as 20 percent) of the diameter of the primary mirror are not visually detectable. Another often-cited scapegoat is surface accuracy. Although errors on reflector optical surfaces must be one-fourth those of refractors to achieve the same results, such accuracy is quite common in good reflectors. So what's holding back well-made reflectors? I am convinced that it is not a central obstruction, and it's not optical quality -- the problem is heat waves off the mirror surface. This layer of warm air behaves like a weak lens of very poor quality right in front of the mirror. I believe this is the main reason reflectors have always been regarded as poor cousins to refractor telescopes. Taking the Heat from Your Mirror Thermal management in your reflecting telescope can yield astonishing improvements in resolution and contrast.

[Peter Ward]

This has been a fun thread!

Mark Sutching has quite correctly pointed out the possible shortcomings of interferometric testing, and I'd put Peter Ceravolo (my optical testing mentor), has also stated provided you are aware of the pitfalls Mark mentions, quantitative testing can be equally valid. (i.e numbers!!)

The only way you are going to know a mirror has a Strehl of 0.99 is if someone actually took the time to apply some sort of quantitative test...ie
measure the bugger!

I am not saying, nor have ever said, an excellent optic can not be made without said test.

But now after the measurements have been taken you have a few possibilities....

1) They are falsifying the data
2) They are incompetent
3) The optic is indeed excellent

Assuming 1) and 2) are unlikely with reputable optical artisans (hey, you know the names) then it's reasonable to also assume the optic is indeed excellent.

[Gama]

This was describing comparisons of refractor over a newtonian/mirror in terms of visual quality.Indeed this is what causes the most prolific problems with mirror based scopes.
But what is being discussed here is different.
As an ex Test Equipment Calibration Manager, i have seen my share of numbers. Further, what some say about definitive numbers over relative numbers is correct.
The test equipment we used for out testing is calibrated and tested against some one elses known standard. But if i test and get a value of "X" for a piece of equipment, this does not mean it will produce the same numbers if the unit was tested by, lets say H.P Calibration Labs.
There needs to be a reference point to be able to establish a common point.

Until then, you say tomatoe and i'll say tamatoe !!.

Theo