10th of a wave?

[billhinge]

When I was building an 8.5" pyrex telescope mirror in the 70's the books e.g. Howard and Fullerscopes used to say aspirational optics should be 1/8 to 1/10th of a wave. Looking at modern optics you see 1/4 of a wave and it often seems 1/10th is premium grade.

Are things measured differently now or were the old 1/10th wave claims just bs? (I think the 1930's telescope books used to say 1/4 was OK but I think I saw Planewave quote 1/20)

Do the figure work the same with refractors? (just looking at APM

"For systems of diameter up to 152mm Wave front error RMS is no more than 0.035 of wave-length at 532 nm. Wave front error p.t.v. is no more than 0.24 of wavelength at 532 nm" ) )

 

[johninderby]

All depends if they were claiming 1/10 wave or 1/10 peak to valley.   

RF Royce’s article is always worth a read.

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

[vlaiv]

If you see something like 1/10th of a wave - that should stand for PV wavefront quality.

Sometimes RMS figure is used and it should be much lower - like 1/10.

Sometimes RMS figure is used and it should be much "higher" like 1/20.

1/4th wave is diffraction limited - and these days mass produced Chinese scopes are often better than that - Strehl of 0.8 or higher.

I would say that 1/6th of wave is good optics, 1/8th is very good optics and 1/10th of wave and above is exceptional optics. These are all PV values (peak to valley).

Two scopes having 1/6th of PV won't have same other measures - like RMS and Strehl. They will be comparable in other measures but there is no direct "conversion" factor between the two. Look at this table:

1/8 of spherical aberration equates to 0.95 strehl but same error from turned edge results in 0.993 strehl - which is much better.

It also tells you that 1/4th wave PV is 1/13.4th RMS and 1/8th PV equals 1/27th RMS - again for spherical aberration.

Edited August 13, 2020 by vlaiv

[billhinge]

Thanks for the answers and link.

So for my APM example "For systems of diameter up to 152mm Wave front error RMS is no more than 0.035 of wave-length at 532 nm. Wave front error p.t.v. is no more than 0.24 of wavelength at 532 nm" ) )

0.035 rms is slightly better than 1/8 or 0.95 Strehl which seems a common quoted figure, OK so far, but ptv is quoted 0.24 which is slightly better than 1/4 or 0.745 rms which seems  twice what the rms value quotes for the same lens (in the wrong direction)

This is from an APM - LZOS Telescope Apo Refractor Astrograph spec https://www.apm-telescopes.de/media/files_public/typiegbv/APMApo-Linsen-Spezifikationen.pdf

It also has specs for scratches and bubbles but I don't expect any scratches or bubbles on an expensive lens 🙂

 

[rl]

 

1 hour ago, billhinge said:

When I was building an 8.5" pyrex telescope mirror in the 70's the books e.g. Howard and Fullerscopes used to say aspirational optics should be 1/8 to 1/10th of a wave. Looking at modern optics you see 1/4 of a wave and it often seems 1/10th is premium grade.

Are things measured differently now or were the old 1/10th wave claims just bs? (I think the 1930's telescope books used to say 1/4 was OK but I think I saw Planewave quote 1/20)

Do the figure work the same with refractors? (just looking at APM

"For systems of diameter up to 152mm Wave front error RMS is no more than 0.035 of wave-length at 532 nm. Wave front error p.t.v. is no more than 0.24 of wavelength at 532 nm" ) )

 

Sadly a lot of these older measurements have to be taken with a large pinch of salt. In many cases the measurement technique would not have been capable of measuring consistently to 1/10 wave. And you have to specify the wavelength of the test...and the mirror has to be realistically supported to avoid sag of 1/10 wave...A Foucault check with the mirror propped up on a few books in a room with air currents just won't cut it.

I've got an 14" AE optics mirror by Jim Hysom made in 1977 with the documentation quoting tested to 1/10 wave. I've had it checked on a Zygo and it comes out at about 1/4 wave at 632nm. It's a very good mirror in general use because the atmosphere is the limiting factor, but 1/10 wave it ain't. 

Unqualified numbers like this without explanations are a salesman's dream. The Strehl number tells you more about the mirror quality but for some reason the wavelength error has stuck as the criterion. Beware of knaves setting traps for fools..ot at least the uninitiated. 

Edited August 13, 2020 by rl
more info

[johninderby]

Mirror testing has it failures.

http://www.loptics.com/ATM/mirror_making/foc_hall/foc_hall.html

[John]

Not long ago a certain brand of dobsonian was being advertised as having a "1/12th wave" mirrors without specifying whether that was a RMS or PV value.

When I queried it the vendor said it was RMS and the PV was "probably" around 1/4 wave. 

 

[andrew s]

44 minutes ago, rl said:

Unqualified numbers like this without explanations are a salesman's dream. The Strehl number tells you more about the mirror quality but for some reason the wavelength error has stuck as the criterion. Beware of knaves setting traps for fools..ot at least the uninitiated. 

Strehl ratio also has issues. It refers to actual v ideal, however ideal is a function of both wavelength,  and aperture (and possibly design depending on the definition you use) . I think it's fine if you compare like with like say 5 " APOs  but comparing a 1" APO with a 5" APO would make no sense. 

Regards Andrew 

Edited August 13, 2020 by andrew s

[johninderby]

If Arthur Daley had sold telescopes.😁😁😁

I remember not long ago some manufactures tried to confuse customers who perhaps weren't knowledgeable about optics with purposely misleading terms that made their optics sound better than they were. Perhaps they were snake oil salesmen in a past life. 😁

 

[vlaiv]

9 minutes ago, andrew s said:

Strehl ratio also has issues. It refers to actual v ideal, however ideal is a function of both wavelength,  and aperture (and possibly design depending on the definition you use) . I think it's fine if you compare like with like say 5 " APOs  but comparing a 1" APO with a 5" APO would make no sense. 

Regards Andrew 

I don't think you are right here. Strehl is measure of wavefront accuracy and is ratio of two values for single telescope: (How much energy ends up in airy disk of actual telescope) / (how much energy would end up in airy disk for perfect telescope of same aperture).

As such - it does not depend on aperture size, nor on wavelength used, or rather - It is just a number and as number it is comparable across scopes.

What it can't tell you is different aspects of performance of the telescope. It won't tell you what sort of detail telescope should be able to resolve - that depends on aperture. It does not tell you light gathering capability of telescope, and so on.

Edited August 13, 2020 by vlaiv

[andrew s]

13 minutes ago, vlaiv said:

I don't think you are right here. Strehl is measure of wavefront accuracy and is ratio of two values for single telescope: (How much energy ends up in airy disk of actual telescope) / (how much energy would end up in airy disk for perfect telescope of same aperture).

As such - it does not depend on aperture size, nor on wavelength used, or rather - It is just a number and as number it is comparable across scopes.

What it can't tell you is different aspects of performance of the telescope. It won't tell you what sort of detail telescope should be able to resolve - that depends on aperture. It does not tell you light gathering capability of telescope, and so on.

Sorry I intended to say the Airy disk depends on aperture so you can have a larger aperture with a small PSF than a smaller one even if its Strehl ratio is not as good.

Hence I don't think you can use it other than comparing like with like if real world performance is the consideration. 

It does depend on wavelength indeed Takahashi quote it as a function if wavelength for some of its scopes. You can easily have a higher Strehl ratio in the IR compared to the UV.

Regards Andrew 

[vlaiv]

12 minutes ago, andrew s said:

Sorry I intended to say the Airy disk depends on aperture so you can have a larger aperture with a small PSF than a smaller one even if its Strehl ratio is not as good.

Hence I don't think you can use it other than comparing like with like if real world performance is the consideration. 

Indeed - it can only be used to compare same type scope of same aperture. Even if aperture is the same in size - strehl will not be meaningful between scopes that have different aperture "shape" - like unobstructed vs obstructed aperture. Both can have same quality wavefront but like two different apertures - MTF will be different between them.

13 minutes ago, andrew s said:

It does depend on wavelength indeed Takahashi quote it as a function if wavelength for some of its scopes. You can easily have a higher Strehl ratio in the IR compared to the UV.

I would put it like this:

Strehl ratio does not depend on wavelength - but there are scopes that have different optical characteristics with respect to wavelength and hence they will have different wavefront and strehl at different frequencies - namely telescopes with refracting elements.

You could similarly say that SCT will have strehl that depends on back focus used? Spherical correction depends on mirror separation and position of instrument measuring wavefront accuracy (and hence mirrors separation) will change this.

 

[andrew s]

21 minutes ago, vlaiv said:

Strehl ratio does not depend on wavelength - but there are scopes that have different optical characteristics with respect to wavelength and hence they will have different wavefront and strehl at different frequencies - namely telescopes with refracting elements.

You could similarly say that SCT will have strehl that depends on back focus used? Spherical correction depends on mirror separation and position of instrument measuring wavefront accuracy (and hence mirrors separation) will change this.

 

Your first sentence quoted above seems to be a logical contradiction.  Even a mirror can have a Strehl that depends on wavelength. 

However, I think both are true. Since frequency is related to wavelength Strehl depends on wavelength.  

I suspect we agree but for wording

Regards  Andrew