Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

How much strehl difference is actually detectable?

  • This topic is locked This topic is locked
240 replies to this topic

#26 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 92,570
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 01 May 2021 - 10:06 AM

I would be interested if these two refractors were tested for their "Strehl" by the same method under the same conditions. If you can see such a difference, then I strongly suspect they weren't...

 

Strehl is like horsepower, it depends on who tested it, how it was tested.

 

I don't believe what the manufacturers say the horsepower is for cars and such.

 

On the other hand, back in the 80s when I was working as a truck mechanic, I was reading the rebuilding specs for Cummins truck diesels, (18 wheelers)

 

Break in was on a dyno. There were four 15 minute sections. The last session was 15 minutes at full power. If the engine did not produce 96% of full power, it was to be torn down until the problem was discovered..

 

I believed Cummins ratings.

 

With apos, having two very similar scopes of different Strehls tested by the same skilled and independent tester is unlikely.  One testers 0.93 might be another's 0.98..

 

Jon


 

#27 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 18,404
  • Joined: 08 Aug 2007
  • Loc: Pampanga, PI

Posted 01 May 2021 - 10:16 AM

In my experience, about 7% was not immediately detectable, if at all, even in the best seeing and inconclusive over time. Some faint change in the star test, but not with planetary detail. My guess is you'd need 10% or more for a difference in planetary detail to become apparent. But, even then it's not necessarily detrimental. I believe our eye/brain has a contrast acuity limit near 5%, so something greater would be required to observe an increasingly apparent difference. Plus, you'd need to be closer to 1x/mm magnification to see it, with almost no chance at lower magnification because the angular scale of any effect is small. I would not worry much about it until Strehl drops below 80. I agree with Jon on testing, advertising, and on polychromatic Strehl mentioned above. Hypothetically speaking, of course.

Edited by Asbytec, 01 May 2021 - 10:19 AM.

 

#28 Mike Spooner

Mike Spooner

    Vendor (mirrors)

  • *****
  • Vendors
  • Posts: 807
  • Joined: 06 Aug 2010

Posted 01 May 2021 - 10:17 AM

Personally, I can easily see the difference between .50 and .95.

On Cloudynights that’s a big difference.

On cloudy nights they look the same!lol.gif

 

just illustrating the reference standards. cool.gif

 

Mike Spooner 


 

#29 Jeff B

Jeff B

    Cosmos

  • *****
  • Posts: 7,782
  • Joined: 30 Dec 2006

Posted 01 May 2021 - 10:55 AM

Strehl anxiety and optical quality obsession seems to be solely a disease of the refractor lovers community. I know it's true of me when I use my refractor. But the mirror scope folks seem to spend more time enjoying the night sky than worrying about optics and this is also true of me when I use my mirror scope:)

I'm an exception.


 

#30 Jeff B

Jeff B

    Cosmos

  • *****
  • Posts: 7,782
  • Joined: 30 Dec 2006

Posted 01 May 2021 - 10:57 AM

On Cloudynights that’s a big difference.

On cloudy nights they look the same!lol.gif

 

just illustrating the reference standards. cool.gif

 

Mike Spooner 

Ah, but I have a cloud filter Mike....and I'm not sharing it.


 

#31 skybsd

skybsd

    Fly Me to the Moon

  • *****
  • Posts: 5,053
  • Joined: 01 Feb 2008

Posted 01 May 2021 - 11:28 AM

For example, suppose you have 2 scopes of identical aperture (and preferable same brand and model). One tested .94 while the other tested .98 (again, just an example)

 

What I want to know is:

 

...

3. Are the differences detectable by the casual observer or did it take an experienced amateur to see it?

...

Rest assured - you can be quite certain that there are some CN members who genuinely do believe (in their own heads at least) they indeed fall into the latter category.., 

 

skybsd 


 

#32 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 92,570
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 01 May 2021 - 12:13 PM

Ah, but I have a cloud filter Mike....and I'm not sharing it.

 

I think if you live where Mike lives, a cloud filter would get relatively use...

 

Jon


 

#33 Jeff B

Jeff B

    Cosmos

  • *****
  • Posts: 7,782
  • Joined: 30 Dec 2006

Posted 01 May 2021 - 03:32 PM

Jon, do you mean relatively little use?

 

I use mine around here all the time. 

 

But I'm still looking for a rain and snow filter.  grin.gif

 

Jeff


 

#34 KBHornblower

KBHornblower

    Viking 1

  • -----
  • Posts: 819
  • Joined: 01 Jul 2020
  • Loc: Falls Church, VA (Washington DC suburb)

Posted 01 May 2021 - 07:17 PM

I don’t get into any of this meaningless ........ but it has been pointed out, mentioned on these type of threads that some fanatical astronomers have literally sent the scopes that they bought back for a refund or exchanged because their strehl number was slightly less than someone else's, still above 0.95 but lower than a friends ! They didn’t care about the views, they thought they weren’t getting their advertised moneys worth so they literally went to the trouble of ......... ! Talk about brainwashed, talk about mindless unthinking, lol !

The extremists in this field remind me of what we call "rivet counters" in model railroading.  That is, those who are pedantically fussy about accuracy of small details on model locomotives or cars, including details that are invisible from more than about 6 inches away.  Both hobbies should be fun, and extreme fussiness takes the fun out of it.


 

#35 LDW47

LDW47

    Cosmos

  • *****
  • Posts: 8,362
  • Joined: 04 Mar 2012
  • Loc: North Bay,Ontario,Canada

Posted 01 May 2021 - 07:51 PM

The extremists in this field remind me of what we call "rivet counters" in model railroading.  That is, those who are pedantically fussy about accuracy of small details on model locomotives or cars, including details that are invisible from more than about 6 inches away.  Both hobbies should be fun, and extreme fussiness takes the fun out of it.

Well said, I don’t normally get involved in this issue but to literally send something back because it is 0.96 instead of Joe’s 0.98 is unbelievable, I have read that before on some of these threads. Those posters actually thought they were getting short changed !  LOL !


 

#36 gnowellsct

gnowellsct

    Hubble

  • *****
  • Posts: 18,699
  • Joined: 24 Jun 2009

Posted 01 May 2021 - 08:18 PM

Well said, I don’t normally get involved in this issue but to literally send something back because it is 0.96 instead of Joe’s 0.98 is unbelievable, I have read that before on some of these threads. Those posters actually thought they were getting short changed ! LOL !


Which is why AP does not tell you the strehl. People don't understand the manufacturing concept of variation around a mean and still being within spec.
 

#37 gnowellsct

gnowellsct

    Hubble

  • *****
  • Posts: 18,699
  • Joined: 24 Jun 2009

Posted 01 May 2021 - 09:03 PM

The extremists in this field remind me of what we call "rivet counters" in model railroading.  That is, those who are pedantically fussy about accuracy of small details on model locomotives or cars, including details that are invisible from more than about 6 inches away.  Both hobbies should be fun, and extreme fussiness takes the fun out of it.

Welcome top the world of guys talking competitively about gear.


 

#38 DAG792

DAG792

    Vostok 1

  • -----
  • Posts: 190
  • Joined: 21 Mar 2020

Posted 02 May 2021 - 12:59 AM

Well said, I don’t normally get involved in this issue but to literally send something back because it is 0.96 instead of Joe’s 0.98 is unbelievable, I have read that before on some of these threads. Those posters actually thought they were getting short changed !  LOL !

These people need to learn about measurement uncertainty, because I don't think that any interferometer exists that could reliably and consistently differentiate between about 0.96 and 0.98 Strehl scopes. I generally group telescope quality by the following and it has served me well(it's based on Suiter's book and other sources):

 

1.Below 0.8 Strehl

Poor scopes, that give fuzzy planetary images.

2. 0.8 to 0.85 Strehl

Acceptable scopes.

3. 0.85 to 0.9 Strehl

Good scopes. This quality is already enough to see everything.

4. 0.9 to 0.95 Strehl

Excellent scopes. Telescopes in this range are generally impossible to separate from those in (5) 

5. Above 0.95 Strehl

Essentially perfect. Astrophysics and Zambuto.

 

These ranges reflect testing accuracy and what is easily visible at the eyepiece. In my opinion, as you go up in Strehl, equivalent differences in telescope quality become harder and harder to detect. 

For example, a 0.9 Strehl scope is clearly superior to a 0.82 one, and the difference *can* be seen in good seeing. However, a 0.92 Strehl and a 0.99 Strehl scope are very, very hard to separate in terms of the in-focus image


 

#39 gnowellsct

gnowellsct

    Hubble

  • *****
  • Posts: 18,699
  • Joined: 24 Jun 2009

Posted 02 May 2021 - 01:21 AM

These people need to learn about measurement uncertainty, because I don't think that any interferometer exists that could reliably and consistently differentiate between about 0.96 and 0.98 Strehl scopes. I generally group telescope quality by the following and it has served me well(it's based on Suiter's book and other sources):

 

5. Above 0.95 Strehl

Essentially perfect. Astrophysics and Zambuto.

 

These ranges reflect testing accuracy and what is easily visible at the eyepiece. In my opinion, as you go up in Strehl, equivalent differences in telescope quality become harder and harder to detect. 

For example, a 0.9 Strehl scope is clearly superior to a 0.82 one, and the difference *can* be seen in good seeing. However, a 0.92 Strehl and a 0.99 Strehl scope are very, very hard to separate in terms of the in-focus image

There are a number of others that hit this exalted terrain in the refractor side.  And some very nice mirrors too.    But I take it these are "by way of example."  GN


 

#40 Star_Shooter

Star_Shooter

    Vostok 1

  • *****
  • Posts: 142
  • Joined: 13 Jan 2009

Posted 02 May 2021 - 08:43 PM

Here is what I want to say. When I look at the strehl number report, I tend to focus more on the three components of the reports: astigmatism, coma and spherical. If those three numbers are similar, which means the strehl is not dominant by one of the errors, the lower strehl (0.94) scope will look somewhat similar when compare to the higher strehl (0.98) one. However, it is possible to have a 0.94 or 0.95 strehl scope with the errors concentrate in one of the component, say the coma number is 0.18, and the other two numbers are smaller than 0.1. In this case, the coma is about 1/5-1/6 wave. If I do a star test, I can tell the diffraction rings are slightly tilted to one side, not totally concentric under high power. Same thing with the astigmatism, if most of the errors are in there, I can see the diffraction rings have a tiny bit more energy in 4 directions. The spherical errors are less obvious though, if the errors are mostly there.

 

If the errors are more evenly distributed among these three aspects, the star image will look nicer, and it will be more difficult to tell them apart when compare it to a 0.98 scope. Of course, most of the scopes have errors spread out, but there are cases where one error is more dominant.

 

For less bright stars, or deep sky objects, the lesser scope looks equally good. For planetary, under superb seeing, the better scope may look a tiny bit more sharper and contrasty.

 

This is why sometimes, the strehl number itself doesn’t tell the whole story. Star test is probably the best way to evaluate telescope optics.

 

Lei


 

#41 db2005

db2005

    Apollo

  • *****
  • Posts: 1,290
  • Joined: 23 Dec 2017
  • Loc: Living in Denmark, under Bortle 5/6 skies.

Posted 04 May 2021 - 12:23 AM

I can visually detect differences between my FC76DCU-EX (quad, which probably has a polystrehl around 0.98 or better), FC-100 (Polystrehl ~0.96), Vixen SD81S (never measured but probably around 0.94-0.95 Strehl judging from other CN'ers estimates) and ED100, ED80. The differences between the first three (all Japan-made optics) are subtle, but they get more obvious when comparing with the mass-produced optics. IME the differences between high-end and mass-produced optics begin to become subtly visible already at well below 1x per mm of aperture, and are more obvious at above 1x per aperture. To my eyes the differences manifest themselves as differences in contrast on moon and planets and in how tightly double stars are rendered. On DSOs, not so much difference.

 

As said in previous posts, Strehl is a complex thing. And optical quality is more than a mere matter of Strehl: mono/poly Strehl, add to this surface roughness, class quality, quality control, spherochromatism, control of internal reflections, differences in focal ratios, and whatnot, and you end up having a complex multivariate problem which is hard to give a simple answer to. IME the most reliable predictor of a refractor's performance is its price tag.


Edited by db2005, 04 May 2021 - 02:10 AM.

 

#42 bobhen

bobhen

    Fly Me to the Moon

  • *****
  • Posts: 5,244
  • Joined: 25 Jun 2005

Posted 04 May 2021 - 06:21 AM

This is what the optician, telescope designer and maker R. F. Royce has to say…

 

“From this optician's viewpoint a reasonable standard for telescopic objectives of high-quality would be those producing a Strehl ratio preferably .96 (1/9 wave, P-V) with .97 (1/10 wave, P-V) as representing the best that is reasonably attainable.

 

Previously it was glass type that soon turned into the glass marketing wars. If a refractor didn’t have FPL53 it was dismissed. Now we have the Strehl marketing wars. Stellarvue is promoting their “new” .99 Strehl refractors in order to justify higher price and to separate their new refractors from what they used to sell and from the competition. The question is, is “every one of the hundreds” of scopes SV produces really .99 Strehl and who is setting up the testing?

 

From Roland Christen of Astro-Physics…

 

"There are numerous methods for measuring and interpreting the results, so that testing the same optic on several different interferometer systems can result in different wavefront numbers."

 

Get a telescope from a manufacturer with a long history of excellence. The price will usually hint at the optical quality.

 

Bob


 

#43 peleuba

peleuba

    Mercury-Atlas

  • *****
  • Posts: 2,920
  • Joined: 01 Dec 2004
  • Loc: North of Baltimore, MD

Posted 04 May 2021 - 09:03 AM

As said in previous posts, Strehl is a complex thing. And optical quality is more than a mere matter of Strehl: mono/poly Strehl, add to this surface roughness, class quality, quality control, spherochromatism, control of internal reflections, differences in focal ratios, and whatnot, and you end up having a complex multivariate problem which is hard to give a simple answer to. IME the most reliable predictor of a refractor's performance is its price tag.

 

 

Actually, for amateur level optics, its not that complex.  Respectfully, what makes it complex are those folks who do not understand what a Strehl value is; how such measurements are taken; and what a high (or low) value really means to the goodness of a lens or mirror.

 

Strehl is the measure of how much monochrome (single wavelength) light an optic is putting into the Airy disk of a star delivered to the image/focal plane.  Polychromatic Strehl is the estimation of how much light all wavelengths the optic is placing into the Airy disk.  Polychromatic Strehl will always be lower (worse) then monochromatic Strehl.  For us hobbyists, the generalization can be made that the more light a lens (or mirror) can place into the Airy disk, the better, higher quality it is.

 

What does Strehl really mean?  A theoretically perfect Strehl of 1.00 means that ~84% of the available light is placed into the Airy disk with the remaining ~16% of the light placed into the diffraction rings.  The more common .950 Strehl means that 95% of the ~84% of the light is placed into the Airy disk with the rest going into the rings.  A barely diffraction limited telescope with a Strehl of .80 means only 80% of the available ~84% of the light goes into the Airy disk

 

For me, for a Strehl value to be meaningful the overarching point is that it must be generated by an honest-to-goodness interferometer and data reduction software.  You may come across official looking test reports (mostly accompanying mirrors) that use the term "Calculated Strehl".  This is a tip-off letting me know that the data comes from some qualitative test method (like Foucault or star test) and not an interferometer.  Some other  important things to know:

 

(1) Wavelength the optic was tested.

(2) Experience of the operator

(3) How many data points were used - this ranges from under 100 to many thousands

(4) Did the results "jive" with other test methods used.  Very important.  Multiple test methods should agree if any one method is to be deemed accurate.

 

Its important to note that most interferometers use a red laser.  This presents a problem for us because most (if not all) our refractors perform best in green light.  Red and green come to focus at different points so a Strehl measured in green will not be the same as one measured in red.  Red lasers are more common and if the wavelength is not listed assume its red.  And, to make matters worse, you cannot simply use algebraic math formulas to scale from red to green when evaluating a lens.  This is because of spherochromatic aberration (spherical aberration as a function of wavelength).  Mirrors do not suffer from spherochromatism because they do not refract light, so you can scale from red to green using simple math.  For mirrors, I have found red lasers to be sufficient and usually do not go to the trouble to scale to green. 

 

You mentioned some other things like surface roughness and glass quality and these are good to know.  Over the last couple of years, I have grown fond of the Root Mean Square (RMS) metric.  Basically this is an expression of the smoothness of the lens surface and is usually one of the measurements listed on the Interferogram.

 

 

 

Get a telescope from a manufacturer with a long history of excellence. The price will usually hint at the optical quality.

Yes, agree.   You often get what you pay for.  You will not get what you do not pay for.  Learn to test the scope yourself - you don't need an interferometer.


Edited by peleuba, 04 May 2021 - 09:07 AM.

 

#44 RLK1

RLK1

    Apollo

  • -----
  • Posts: 1,115
  • Joined: 19 Apr 2020

Posted 04 May 2021 - 09:24 AM

While I understand the reasons behind the discussions regarding testing methodology, that's not, in my read of the OPs initial post, directly relevant to his question. The latter being can you actually see the difference in a case where you have an "identical brand and model, one tested .94 while the other tested .98"?

In my experience, such a distinction isn't visually detectable at the eyepiece. I remain highly skeptical of claims to the contrary.


 

#45 kel123

kel123

    Surveyor 1

  • -----
  • Posts: 1,651
  • Joined: 11 May 2019

Posted 04 May 2021 - 09:41 AM

As every other poster has noted, you can never possibly see the difference visually. And as some have noted, several other factors come into play that the high value might not necessarily be better than that of lower value.
 

#46 RLK1

RLK1

    Apollo

  • -----
  • Posts: 1,115
  • Joined: 19 Apr 2020

Posted 04 May 2021 - 10:00 AM

As every other poster has noted, you can never possibly see the difference visually. And as some have noted, several other factors come into play that the high value might not necessarily be better than that of lower value.

That's not what I've read in this thread...


 

#47 Simoes Pedro

Simoes Pedro

    Viking 1

  • *****
  • Posts: 731
  • Joined: 03 Feb 2009

Posted 04 May 2021 - 10:03 AM

For example, suppose you have 2 scopes of identical aperture (and preferable same brand and model). One tested .94 while the other tested .98 (again, just an example)

 

What I want to know is:

 

1. How much magnification was needed to see the difference? For example, did it take 400x to even begin to see the difference or was it obvious at just 150x?

2. Any difference in Lunar/planetary or deep-sky contrast?

3. Are the differences detectable by the casual observer or did it take an experienced amateur to see it?

4. Anything else?

Love these questions when hardly anyone knows the answer but everyone has an opinion popcorn.gif


 

#48 peleuba

peleuba

    Mercury-Atlas

  • *****
  • Posts: 2,920
  • Joined: 01 Dec 2004
  • Loc: North of Baltimore, MD

Posted 04 May 2021 - 10:08 AM

I remain highly skeptical of claims to the contrary.

 

And, its good to be skeptical.  I think, if you're motivated, you should purchase two identical refractors and have them sent out for testing.  You will probably need to plow through quite a few samples to find two that precisely fit the OP's test criteria of a .04 delta between Strehl measurements.  Then test them out on the same objects, same night, same eyepiece/diagonal combination.  Only then will you truly know the answer to the question.    

 

I took the OP's query as more philosophical then anything as there will be only anecdotal evidence to support any single contention unless one does exactly as I prescribe above.  

 

What myself and other posters have demonstrated is that there is more then simple Strehl that can lead to differences.  And, I am certain there are some participating in this conversation that did not know what Strehl value really meant until posted it above.  We are all here to learn.  But, this type of question that deals with abstracts may not be the best teaching moment.  I and others recognized this and expanded on the topic.

 

Now, I have personally done A-B comparisons between an AP155 and my TEC160FL.  The TEC was an honest 1/8 wave optic in green and the AP was at least .984 in Strehl.  The only differences we saw were in the star test at ~300x on a star nearly overhead.  The AP had less "junk" floating within the first diffraction ring when compared to the TEC.  My friend who owns the AP155 participates in CN so maybe he will chime in.  This comparison occurred about a decade ago on farmland in suburban Maryland about 40 miles due north of Baltimore.


Edited by peleuba, 04 May 2021 - 10:18 AM.

 

#49 kel123

kel123

    Surveyor 1

  • -----
  • Posts: 1,651
  • Joined: 11 May 2019

Posted 04 May 2021 - 10:13 AM

That's not what I've read in this thread...


What have you read? That you can visually see the difference?
 

#50 bobhen

bobhen

    Fly Me to the Moon

  • *****
  • Posts: 5,244
  • Joined: 25 Jun 2005

Posted 04 May 2021 - 10:16 AM

Because it is unlikely that different manufactures use the same testing methodology to calculate Strehl, comparing one scope to another based on what one manufacture claims versus another can be opening up a can of worms.

 

However, below are a few examples of how R. F. Royce defines quality. Just keep in mind that Strehl results from some manufactures can be suspect and that there are other factors that will influence what is seen in the eyepiece.

 

.97 Strehl and above: Excellent

.94 Strehl: Very Good

.92 Strehl: Good

.82 Strehl: Diffraction-limited resolution idefined by the Rayleigh criterion

 

Bob


 


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics