Fluorite-based APO vs ED vs SuperED

<p>It appears that using Fluorite-based correction for chromatic aberration (CA) is the most effective and is considered a higher-end solution (if cost is not an issue) than ED or SuperED solutions.</p>

<p>For your reading pleasure ...<br>

http://www.takahashiamerica.com/FluoriteAdv.htm<br>

http://clublexus.com/forums/digital-photography/321730-ed-vs-fluorite-lens-material.html<br>

http://www.birdforum.net/showthread.php?t=40468</p>

<p>If I understand it correctly, the main driving force for not using Fluorite-based solutions in Nikkors is cost. This seems reasonable and consistent with Nikon's corporate policy regarding mass production costs. But, my question is whether the super-teles should be the exception for this policy since they compete directly with Canon's super teles, which certainly use Fluorite, let alone Leica's APO lenses.</p>

<p>What do you guys think? What are your opinions?</p>

<p> </p>

<p>Good enough is good enough. Typically Nikon super-teles are good enough. I don't suppose Nikon photographers particualarly care what their lenses are made of.</p>

As I recall from a long ago conversation, Cost isn't the reason Nikon decided not to use fluorite elements elements in their

high end telephotos. I'm pretty sure Leica and Zeiss don't use Fluorite elements either. My general recollection is that

fluorite elements are softer and possibly more heat sensitive - and that last factor may have been the reason Canon

started using their distinctive white lens barrels in the first place so they wouldn't heat as much on sunny days.

<p>Maybe Nikon/Zeiss didn't have a choice. Back when fluorite elements first started to be used Canon had their own process for growing crystals large enough and pure enough to make the elements out of. Canon also fabricated all their own optics so once they had the material they could make the lenses. Not sure Nikon/Zeiss could have found a commercial source of the material even if they wanted to fabricate their own fluorite elements. Then there's NIH (Not Invented Here) which might have been a factor in not following Canon (though it doesn't seem to have stopped Nikon using USM motors and Optical stabilization!).</p>

<p>Fluorite is somewhat more themally sensitive and softer than most optical glasses, but as far as I know that's never been an issue with the Canon "L" lenses that use fluorite elements. I've never heard of a lens that needed that elelment replaced due to aging or environmental damage. I think you can do most of what Fluorite does if you use two "special" glass elements. Modern exotic glasses are available now that didn't exist when Canon first introduced fluorite element lenses</p>

<p>1969 is when Canon came out with their first Fluorite lenses. I first handled one in 1971. I recall they made an allowance in the focus control to go past infinity. They did mention the thermal sensitivity. The lenses were still black. I don't think they came out with the white lenses until 1976.</p>

<p>Leica uses fluorite in it's latest 50mm f/1.4 ASPH. I don't know who makes that for them. I doubt they would ever say if it was a Canon made element.</p>

<p>Fluorite lenses have been used for at least 100 years, mainly for their transparency to ultraviolet. That's fine for a lab, but fluorite is soft, brittle and the refraction changes dramatically with temperature. Fluorite also has a low dispersion index, which makes its use in long lenses attractive in order to reduce chromatic aberation. Nikon has eschewed fluorite because of its poor mechanical properties, and achieves the same results with low dispersion (ED) glass. Canon chose otherwise.</p>

<p>Several decades ago, I purchased Nikkor 400 mm F5.6 P lens that claimed to have a fluorite element. In the manual, it mentioned that the lens would focus past infinity in order to accommodate the expansion of the fluorite element when heating up. I think it is the only Nikon lens ever made with a fluorite lens element. There was no gold ring on the lens barrel. Also, Hoya has developed an ultra low dispersion glass which I believe is used in the Sigma 120-300 F2.8 and 300-800 F5.6 Sigmonster. Nikon uses a similar ULD glass in its 200 F2.</p>

<p>Fluorite is used not just because it has low dispersion, but it also has anomalous dispersion, meaning dispersion changes more non-linearly with wavelength than silca based glasses. Both of these factors can be used to minimize CA.</p>

<p>As far as I know fluorite is still the best route to the elimination of chromatic aberration, though some of the newer silica based glass formulations are pretty good (and cheaper).</p>

<blockquote>

<p>Nikon has eschewed fluorite because of its poor mechanical properties, and achieves the same results with low dispersion (ED) glass.</p>

</blockquote>

<p>As the saying goes, one fluorite glass element can be replaced by a few ED elements and possibly by one single SuperED element. If using more ED elements can reduce CA is the way to go, have they thought about more aberrations in the n-th tier that are present just by introducing additional elements?<br /> Additionally, isn't the clarity and brightness of fluorite elements, in generally, higher than ED elements.</p>

<p>Marc. B said:</p>

<blockquote>

<p>Leica uses fluorite in it's latest 50mm f/1.4 ASPH. I don't know who makes that for them. I doubt they would ever say if it was a Canon made element.</p>

</blockquote>

<p>I am not so sure if there's a fluorite element in there but there's certainly a very exotic element in that 50mm Summilux ASPH, whose cost is higher than all 7 elements put together from its predecessor lens it replaced.</p>

<p>The source of that element is from Schott (Zeiss).</p>

<p>It's all here ... http://www.imx.nl/photo/leica/lenses/lenses/page57.html</p>

<p> </p>

<p>From leica:</p>

<p>"Every innovation currently available in lens technology – aspherical lenses, glass with anomalous partial dispersion, glass with a high refraction index and a floating element – has been combined to create a lens that sets the new standard in this focal length class."</p>

<p>It does look like Schott is the source.</p>

<p>http://www.us.schott.com/advanced_optics/english/our_products/lithography/calcium_fluoride.html</p>

<p><em>As the saying goes, one fluorite glass element can be replaced by a few ED elements...</em></p>

<p>What "saying"? Please cite a source.</p>

<p>Has anybody seen a reference to the scientific merits of this that's not from a sales brochure of some kind? Like, how about an executive summary from an engineering report by ASTM or something like that?</p><p>I'm sure there are some merits, but the references don't look too good. So far, it looks like we've got some sales brochures. One of the links above is to a bulletin board from a Lexus fan website! Wouldn't a Lexus owner be the type of person who might have the disposable income to spend $1000+ on a lens? If you were in the marketing department of a company trying to sell those lenses, wouldn't you slip a press release to one of those guys who ran an internet site which harbored some of your potential customers?</p><p>One of the other links was to an ad run by a telescope company. Their most compelling graphic wasn't a chart summarizing lab results. It was a cartoon of two lenses. Effectively, the lens picture is a testimonial, from the person trying to sell us a lens! </p><p>Hey, they didn't even find a moderately fat person for "before" and "after" photographs. They just told you they decided it was great. </p><p>Come on now, how do we know this stuff works that well? And, even if it does, what's wrong with the other glass?</p><p>I suspect that there's a thin coating of BS on that sales brochure; it's just thin enough to be extremely transparent, but strong enough to bend somebody's mind. </p><p>Who has seen a report on these optics that wasn't written by someone who stands to make a buck off this stuff?</p>

<p>Here are links to more substantial readings ...</p>

<p>http://toothwalker.org/optics.html<br>

http://en.wikipedia.org/wiki/Chromatic_aberration<br>

http://hyperphysics.phy-astr.gsu.edu/hbase/geoopt/aber2.html</p>

<p> </p>

<p>An excerpt from an optical master - Yuri from Telescope Engineering Company. He utilizes fluorite in his designs as well as ED glass in his slightly cheaper scopes. Mind you, FPL53, FPL52 and FPL51 are considered types of ED glass. ED glass is simply glass mixed with fluorite. Advantages of ED are that it's more durable due to the glass component. Disadvantages is that there is less color correction when compared to crystalline fluorite.<br>

More on this to follow. But first, a quote from Yuri...<br>

"The main differences between CaF2 and FPL53 are:<br /> 1. CaF2 is availabe in sizes up to 300mm diameter, the FPL53 is limited to <160mm<br /> 2. The CaF2 has dispersion characteristic that let make faster focal ratio vs. FPL53 and still have good color correction<br /> 3. CaF2 has no scatter, all ED glasses do have, see the pictures of laser beams going through ED triplet objective and through Fluorite objective (the CaF2 is the second lens - no trace of laser beam)<br /> 4. The cost of high quality CaF2 is >3 times more than FPL53<br /> After all the CaF2 is an ideal material to work with!<br /> <br /> Best regards, Yuri"</p>

<p> </p><div></div>

<p>Many of you may have heard the terms "SD, UD, ED" and so forth from camera lens manufacturers. Simply put, they are just different grades of ED glass. SD might be equivalent to FPL52 or FPL53. UD may be FPL51. Technically and composition wise, they are all ED glass. <br>

The only source that I know of for crystalline fluorite is Optron, a wholly owned subsidiary of Canon. Now does this mean it's a "canon" lens? Not technically if you're looking for the "L" lenses packaging. But yes, Canon does own the company that produces the crystalline fluorite. If I recall correctly, they also make a fair amount of the ED glass there - you have to get the fluorite from somewhere. <br>

For an optical formula stating that they have fluorite or ED is no guarantee on the level of color correction. They can use FPL53 or even a fluorite crystal in the formula, but if it's not properly compensated for with the proper mating glass, then the color correction will be horrible. Even the term "APO" is misleading. Is it truly color correct from 400nm to 700nm? Is it a single, two or three crossing design relative to the zero mark? <br>

Probably one of the most color correct optics come from Takahashi. The reason is simple - there is lots of color correction going on and they are using a fluorite element in front, mating glass in center and the rear element is FPL53 if I remember correctly.<br>

http://images.google.com/imgres?imgurl=http://www.grandeye.com.hk/tak/new/toa130_chart2.jpg&imgrefurl=http://www.grandeye.com.hk/tak/new/tsa130_fct250.htm&usg=__ThQiBpzfv7FkCC-w_yUhTL394ik=&h=500&w=309&sz=15&hl=en&start=18&um=1&tbnid=wzRjfkKDXbm-cM:&tbnh=130&tbnw=80&prev=/images%3Fq%3Dtoa130%2Bcolor%2Bcorrection%26hl%3Den%26safe%3Doff%26um%3D1<br>

Probably the next best color corrected optic that I've seen would be from Astro Physics. Their 160EDF. Note the 3 crossings to the zero plane. Many so called "APO" lenses are lucky to have 1 or 2 crossings on the zero plane. If you've noticed the previous graph, the Takahashi resides mostly on the zero plane - near absolute perfect color correction.<br>

http://www.astro-physics.com/products/telescopes/160edf/160colorcurve3.jpg</p>

<p>John - the Lexus owner that uses the optic went for it since it provided the best color correction, contrast, saturation and sharpness out of any optic that he's tested. That includes the Canon 800mm f/5.6L EF IS, 600mm f/4L EF IS, 500 f/4L EF IS, 300 2.8L EF IS. Canon received a very long letter on how to redesign and improve their optics after a long field test of the 800mm 5.6L EF IS. They definitely have a flaring issue and this can be resolved by using a different type of baffle rather than the reflective rings and inadequate flocking methods currently implemented. He's very much a huge fan of Canon, Nikon, Leica, Hasselblad, and Mamiya. But the proof is in the pudding and the results.<br>

http://photography-on-the.net/forum/showthread.php?t=534639&page=3<br>

The first picture of the cormorant is with the 800. Fresh from the factory and microfocus calibrated. Gitzo CF1325 for the tripod - a very stable mounting. If you look at the image you'll see that there is a muddy look to it. It's not from focusing - it's from poor flare control. <br>

The next picture down is with the Takahashi FSQ106N, a 530mm f/5 telescope. Canon has NOTHING on one of these telescopes, optical quality wise. Two of the four elements are crystalline fluorite, the other two are the mating glass. The specialized extender was used to make the focal length 850mm at f/8. Extender itself has 5 elements, none fluorite though.<br>

"cwphoto" later admitted that he did see quite a bit of flaring and took further conversation off the board. He also later admitted that for his use an intents, he really needs the AF and the IS, but in terms of optical quality, he agrees about the flaring issues and the lack of contrast due to the flaring.<br>

As for "color correction" in Canon lenses?<br>

http://www.astropix.com/HTML/I_ASTROP/EQ_TESTS/C300MM.HTM<br>

http://www.astropix.com/HTML/I_ASTROP/EQ_TESTS/C400MM.HTM<br>

And in Nikon lenses with ED glass?<br>

http://www.astropix.com/HTML/I_ASTROP/EQ_TESTS/N400MM.HTM<br>

NOT SO GOOD. At least not for critical astro work or for anyone demanding the finest in optical quality.<br>

Remember that camera lenses are mass produced. Batches WILL differ. You can purchase the same type of FPL53 with the "melt data" and have it measure differently on the optical bench. Each batch differs. I've had camera companies that will say "our process is consistent", which to me, means absolutely nothing if they do not account for the melt data of each glass type being used. Camera companies really can't do this since it's a HUGE added expense. You would have to make master test plates for each batch and type of glass. HUGELY expensive, but the telescope people have to do this since they realize that it will greatly affect color correction. Most of the camera people aren't going to blow up their prints to 13x19, 24x36 or even 30x40. With a precision optic (telescope - high end) you can still extract the most undistorted images out of them. With the production lenses (Canon/Nikon/Leica/Zeiss, etc) you will see chromatic aberration and other nasty distortions.<br>

Further information can also be found in Nature Photographer, winter 2008 issue. There's an article there called "high resolution optics".<br>

As for the Lexus owner, you're talking to him. Feel free to ask away on any questions relating to the ED vs Fluorite. I'm also working with many people involved in the optical field. The military/classified optics are fun. :)</p>

 

<p>What about the Zeiss Superachromats. Are they close to Takahashi?</p>

<p>Carl Zeiss indicates that their UV Sonnar 105mm f/4.3 lens makes use of Fluorite; and their Superachromats are supposed to (at least according to some sources, including Wikipedia.)<br /> P Mui, thank you for the in-depth information!</p>

<p> </p>

<p>Arthur - the Zeiss telescopes during the early 1990s were exceptional pieces. In terms of color correction, I would say that they edge out AP, especially the Zeiss 6 inch f/8. Now AP will say differently (Astrophysics) but they are two very fine optics. If you compare the TOA Takahashi series now to the AP or even the Zeiss from back then, the TOA has better color correction due to two things. The first being that the TOA is air spaced and allows several more degrees of freedom in design than an oil spaced optic. Both the AP160 and the TOA130/150 are air spaced, which explains their extreme level of color correction. The second is that the Takahashi utilizes TWO elements in a three element design that have fluorite or fluorite mixtures in them. In other words, fluorite crystal up front, mating glass in center, ED glass at the back. NO other telescope company will do this as of this writing.<br>

Unfortunately, Zeiss produced very very few batches of their telescopes. The accessories that they use were quite proprietary, even for attaching an eyepiece. Zeiss , according to one of the telescope companies, did a "massive sour grapes exit" and didn't know how to manage their telescope lineup. Zeiss only made enough scopes from the one single batch of glass that they had and probably didn't realize the practical and lengthy process it takes to make a ultra precision optic. <br>

The 6 inch f/8 model that I'm talking about now sells for mid 20 thousand dollars to low 30 thousand. Highly collectible item, but not too practical in photographic usage.<br>

Bradley- I'm not surprised that Zeiss uses a fluorite element for their UV lens. ED glass has issues with UV since ED has a glass component. Fluorite corrects down to 360nm to 380nm when properly implemented with a well chosen mating glass.<br>

There's MUCH more to color correction than just saying "ED" or "fluorite". It depends on the spacing (air or oil?), the level of ED glass being used, the level of mating glass being used, the figure of the lens, etc etc. To this date, there is NO lens that offers "perfect" color correction with ultra high contrast, saturation, etc.<br>

Don't even get me started on mirror based optics. They have their own set of problems. Loss of contrast, huge amount of light scatter, etc. Yes, even the famed Hubble (which so many people seem to use as a "reference") has these same problems. Lots of photoshop helps out Hubble images!</p>

 

<p>I think you mistook the Superachromat for an astronomy scope. It's a lens for the Hasselblad 500-series body and Zeiss made this for the ultimate color corrected lens. It's corrected across 7 colors.</p>

<p>http://www.zeiss.com/C12567A8003B8B6F/EmbedTitelIntern/Tele-Superachromat5.6_350mm_CFE_104549_e/$File/Tele-Superachromat5.6_350mm_CFE_104549_e.pdf</p>

<p>Some shots from this lens<br /> http://www.horolezec.cz/galery/a_gal_66/namibia/mnama24.html<br /> http://www.horolezec.cz/gallery.htm<br /> http://www.christopherburkett.com/photographs/large/rlslg.jpg</p>

<p>I am surprised you did not consider this lens ... you should talk to your friends in the military and high resolution satellite lenses and see what they tell you how this lens is relevant.</p>

 

<p>Arthur - thanks for the link. Quite interesting. Do you have any specs where they corrected for the colors, and at which nanometer wavelength? <br>

I noticed that the spec sheet isn't saying which type of glass or ED they're using. Or if they're even using fluorite crystal. <br>

About the most overkill lens (in terms of color correction) that I've seen would be from Coastal Optics, their 60mm focal length, f/4 model. They can, for an additional charge, provide a color correction and lab test for the optic. Mind you, this is just in terms of color correction.<br>

I already know what my optical friends would say. They'd look at the lens quite critically and say "eh". What they're concerned about is not just color correction alone. They're worried about strehl ratio, contrast, what the lens is corrected and designed to in nanometer wavelength, melt data, knife edged baffles (not the cheap reflective rings that Canon uses in their 800), amount of scatter for the entire optical formula, the works. <br>

They don't impress easily. I showed them the AP and they remarked "not bad". Showed them the Tak (FSQ106N) - they were actually smiling. From their reaction, you'd think they they knew the designer of the lens... ;)<br>

Would be intersting to find out how well corrected the Zeiss lens is!</p>

<p>removed duplicate postings</p>

<p>>Do you have any specs where they corrected for the colors, and at which nanometer wavelength?</p>

<p>As far as I know, the original superachromats were used for satellite images used by you-know-who. The ones sold to the public are pretty watered down but they are still one of the most corrected, if not the most corrected for medium format lenses.<br /> I believe they are all fluorite based.<br /> http://en.wikipedia.org/wiki/Superachromat<br /> http://www.zeiss.de/c12567a8003b58b9/Contents-Frame/ff8a7b7a369c1549c12570fb0048e952</p>

<p> </p>

<p>The wiki article mentions "fluorite glasses". This is ED glass. True fluorite elements with no glass components are called crystalline fluorite. <br>

From the spec sheet, Zeiss did a GOOD job for color correction, AND they took into consideration the melt data for each batch. If the color chart holds true, they originally designed it for 550nm as a center point, which is what many telescope makers go for. Why the 550nm mark? Because it represents the green color, the color where the human eye is the most sensitive. <br>

But lets take a closer look. The AP graph that I posted earlier, http://www.astro-physics.com/products/telescopes/160edf/160colorcurve3.jpg , has THREE crossings from 400nm to 700nm. ALL of which are visual to the eye. Now if we take a look at the wiki definition of "superachromat", http://en.wikipedia.org/wiki/Superachromat , you'll see that it only has TWO crossings from 400nm to 700nm, with respect to the focal plane. Now if you're using a camera that has sensitivity out to 1000nm (infrared), then the Zeiss lens will fit the bill. But in terms of just attaching a standard digital back (phase one, leaf, etc) then it's very very overkill. <br>

So why did Zeiss go with "superachromat"? My guess is that there are only two crossings from 400nm to 700nm. APO chromat says THREE crossings, from 400nm to 700nm, if memory serves me correctly.<br>

In terms of color correction alone, it may be possibly be the best corrected commercially available camera lens out there. I'd actually like to see one of these lenses perform under a critical star test and get some strehl numbers, wavefront numbers, etc from it. I'm just tossing a number out there, but it's possibly better than .80 strehl, a number difficult to obtain with camera lens manufacturers. The only other one that claims "diffraction limited" would be Leica in SOME RARE cases. A "1.00" strehl means a perfect optic in figure, with all of the energy concentrated into the airy disk on a point source. The Takahashi FSQ106N typically gets an average of 0.95 or better, which is extraordinary in terms of a 4 element design. The TEC and AP telescopes usually get .99 on the average. Canon, and I'm just tossing out a number out there, would probably get a 0.40 or lower due to the amount of aberrations depicted on the links in previous postings. Yes, I realize that the postings were for the 300 and 400 lenses, but the distortions would be scaled up in the 800 lens. Trying to correct for 18 elements in the 800 f/5.6 lens is an extremely difficult task. Even if they took all the tricks from the t-scope manufacturers, I doubt that the strehl would get above the 0.8 mark, simply due to the amount of elements. Compounded errors with more elements.<br>

The higher the strehl ratio number, the higher the contrast and intensity there is when rendering point sources. Rendering point sources isn't easy - it's actually one of THE toughest things to render for a lens. ALL the flaws come out, especially in multi element designs. Don't expect a Canon 800 f/5.6L EF IS to have "perfect" color correction, or for that matter, halfway close to perfect rendering of an image. It does QUITE poorly on microcontrast and other ultra fine details, something which a high end ultra precision optic would THRIVE in. The poor photons are just fighting with WAY too many air to glass interfaces which rob the microcontrast and other optical goodness.<br>

If a lens can't render a point of light accurately, then it won't render an entire scene with the utmost in accuracy.<br>

Zeiss didn't mention pitch polishing or any other types of fine tuning. To put it in perspective of how critical the figure is on a lens, a typical lens will take a few minutes to hog out and grind on a multi million dollar machine. It's then ready to be coated. A lens from Takahashi or AP or TEC will take several hours to several DAYS, PER SIDE of an element to be polished to an extreme critical accuracy. They perform the extra work since the lens will be used to magnify a point source many hundreds to many thousands of times and each imperfection of a lens will add up significantly, something not really expected of a camera lens.<br>

Now just for grins and giggles - here are some strehl charts for a typical optics set. Note the roughness in the figure and the zoning errors. I would also expect this of many commercially made lenses. Note that this is considered "diffraction limited", about a .80 strehl ratio.<br>

http://geogdata.csun.edu/~voltaire/roland/sct.html<br>

Now lets take a look at this other one. Note that the optics is EXTREMELY smooth, no zoning errors. This will provide a much higher contrast and saturation than 99.9% of the optics out there!<br>

http://geogdata.csun.edu/~voltaire/roland/130edf.html<br>

Now if you change out the lens elements to include crystalline fluorite, the contrast and saturation will increase, mainly due to the fact that crystalline fluorite will have ZERO internal scatter. Remember that ED glass and fluorite GLASS (they're the same) has scatter and this will rob contrast and saturation.<br>

Still, to the credit of Zeiss, they did a good job on color correction.</p>

<p>What was the cost of the Zeiss superachromat lens?<br>

And I'm not surprised that the "you know who" guys would like the lens. Infrared surveilance would be quite ideal with it. Not meant for the ultimate in image quality mind you, but it would definitely get the job done, especially in the IR range!<br>

What doesn't make sense is that the overall resolving power isn't that great, at least not in terms of raw aperture. A 62.5mm front diameter lens (350mm / f/5.6) can only resolve a certain amount of detail. Roughly 2 arc seconds. A 4 inch diameter (~101mm) will resolve about 1 arc second of detail. Larger the diameter, more resolving power, all things being equal.</p>