A lense you didn't know you had

[Sparky]

Ole- can you specify the difference between spherical aberration and 'field curvature'? I'd always thought they were exactly the same phenomenon.

 

[Ole]

"Field curvature" means that the lens focuses the image on a curved field instead of a flat film.

"Spherical aberration" means that rays from the edge of the lens come to focus at a different point than rays from the center of the lens, giving blurry pictures. That's why some lenses have aspheric elements.

 

[David A. Goldfarb]

You can recognize field curvature when the center of the image is in focus and the corners are blurry. Spherical aberration produces a diffuse image over the sharp image.

 

[Sparky]

Doh! Sorry. I was assuming Ole was talking about curvature of the plane of focus - which, to me, would be a 'spherical' aberration. What you both call 'field curvature' I would have referred to as 'distortion', which I don't consider so much an aberration, since it's not a physical phenomenon so much as a result of the lens design... well - you know what I mean. I hope.

 

[Ole]

You can recognize field curvature when the center of the image is in focus and the corners are blurry. Spherical aberration produces a diffuse image over the sharp image.

That could be coma as well, or a bad case of astigmatism, or...

But if you focus in the center and then have to refocus to make the corners sharp, then it's field curvature. Especially if you get either corners or center sharp, but never at the same time.

If you have to refocus after stopping down ("focus shift"), spherical aberration is likely to be the main culprit.

 

[Ole]

"Distortion" is a change in the scale of reproduction over the field of view. If the corners are enlarged you get pincushion distortion; if they're smaller you get barrel distortion. If they get really small, you may have put on a fisheye lens by mistake...

Distortion is one of the few things that doesn't change when you stop down, BTW.

 

[Sparky]

It DOES change with magnification, though... anyway - I was referring to that as distortion following zeiss's lead (they call it 'radial distortion'). I figured they'd be fairly reputable in the nomenclature dept. http://72.14.253.104/search?q=cache...ent&hl=en&ct=clnk&cd=1&gl=us&client=firefox-a

 

[Struan Gray]

I don't have a reference, but I remember reading that the plasmat convertible Symmar was designed so that the individual cells had low coma. In a complete lens, like the more modern Symmars, the approximate symmetry of the whole lens can do that job, which gives you freedom to optimise other things. I.E. eliminating coma in the individual cells acts as a restraint on the performance of the whole lens.

I have 150 and 360 convertible Symmars. The 360 is very demanding of your mounting when converted (in fact, I use it to test tripod setups) as it requires very long extensions. Mine also has pretty bad colour fringing which is visible on the ground glass. The converted 150 looks much better, but 12-14" lenses are so common and cheap I have never used it that way for real.

 

[Struan Gray]

It DOES change with magnification, though...

True, but bear in mind that the absolute value of the distortion is not so important. It's the variation in that value over the frame that causes lines to bend. A constant distortion of, say, 10% over the whole picture area just means that your focal length was mis-specified by 10%.

 

[Ole]

The difference between the lateral magnification in the image field and the on-axis lateral magnification is referred as distortion.

As Struan said: It's the difference that is the distortion.

 

[Sparky]

well - the % distortion seems to be a bit of a meaningless figure unless plotted over the entire field. But that's the basis of photogrammetry. Known distortion so you can offset for it.

 

[Dan Fromm]

But Sparky, the good lenses sold for photogrammetry are virtually distortionless. And ones in use are periodically tested to make sure they're still on spec.

 

[AgX]

'far as I know' (as I heard) they stopped doing this - even though the lens design is identical, despite what marketing hype says, years ago because of complaints that the 'converted' lens was of 'poor quality'. In order to save their reputation - they stopped advertising that feature.

This seems convincing, on the fact that the few major makers of LF lenses were clinged to `optimum image quality´ and an Imagon, from my perspective, seemed merely to be considered a sort of stepchild.

I think (just my opinion) that the most recent modern versions of these "symmetrical" lens designs are more and more unable to function as "sharp" convertable lenses, as they are becomming more "un-symmetrical".

This implies that all the more modern lenses are more a-symmetrical. Just looking at those lens sketches, the Symmar-S and Sironar-N do not seem less symmetrical than their ancestors to me. One modern lens (Macro-Symmar?) even seems absolute symmetrical. The Apo-Sironar however is very asymmetric and the Super Symmar HM even extremely asymmetrical.

Let us look what the industry replied (in 1993) on a letter of mine where I asked why they stopped advocating the convertibility of those lenses and asked about that possibility on their current ones (or so). I translated the letters for you:



Rodenstock:

… Rodenstock once also stated this chance of doubling the focal length. However it works only with symmetrical lenses and even then is coupled to a significant reduction in quality. This is the reason why today we do not longer consider this possibility, as the quality demands claimed to large format photography are no longer fulfilled. Thus we advise our customers to use the lenses only `complete´, in order to yield an optimum image quality and thus making the photograph fulfill the demands toward large format photograph. …




Schneider:

… With the transition of the Symmar-range to the Symmar-S we abondoned the optics feature of being convertible, because the construction of improved high-performance lenses did not permit this anymore.
The optical performance of one of the modules of the Symmar was significantly worse, especially because of the occuring color fringes and diffraction phenomenon of the slower module which had a useful aperture of 22-33.
With the revision of of the Symmar-range towards the Symmar-S an improvement of the central part was realized; with the Apo-Symmar [an improvement of] the outer part in addition to the central part.
With the Super-Symmars HM the outer part was adjusted in a way that an angle of view of 80° was achieved. …

 

[Sparky]

But Sparky, the good lenses sold for photogrammetry are virtually distortionless. And ones in use are periodically tested to make sure they're still on spec.

Not true, actually. They MIGHT be a cut-above - but they're fully 'calibrated' that means they come with a paper spelling out the known distortion in that SPECIFIC lens (NO lens is distortionless... except of course, my lovely 100mm hasssy planar!!). The known distortions for that lens are also supplied on computer disk in tabular format that the photogrammetry/CAD interface program can read, so that accurate (well, semi-accurate) measurements can be derived from the image data.

 

[Murray@uptowngallery]

color friinging question:

I picked up a big fat lens on eBay someone lost a jousting match with- I am surprised the front element wasn't broken it was so bad. I suspect it is from a monochromatic application. Auto-Logic, later aquired by Agfa, unable to find any info about it.

I removed the front element. I noted it, as well as the remander of the lens ass'y, have a yellow look. The removed element made it clear that both sides of the lens have a yellowish coating out to about 1/4" from the edges, thick enough that you can see the edge of the coating, so it's not a 'radiation browning' look.

Projecting an image with the single element or the remainder of the lens onto a wall (like a bare light bulb for example) produces a strange uneven 'rainbow' fringe in addition to the expected image. There are a number of colors visible, but the blue is very pronounced, almost iridescent.

I would have thought that a yellow filter would block blue, so maybe they are not yellow filters but some kind of stranger bandpass filter. And a yellow filter on a conventional lens doesn't seem to produce flamboyant color fringing...just a yellow image.

Murray

 

[Curt]

Murray, here's your solution to that big fat lens you got.

http://www.focalpointlens.com/fp_intro.html

Good luck, let us know how it turns out.

Curt

 

[Struan Gray]

Murray, you have two possibilities, both of which may be acting at once.

The first is that coatings for monochromatic applications can be strongly dichroic. That is, they have one colour in reflection and the complementary colour in transmission. Thus an anti-reflection coating designed to be efficient in the blue will be yellow in reflection.

Does the lens have a yellow tint if you place it on a piece of white paper, or does the coating look yellow if you look at a light source reflected off the element? If the latter, the strong blue fringe is probably the result of just such a dichroic coating, and it indicates that the lens was optimised for the blue part of the spectrum. The 'logic' in the name suggests an application like circuit board printing, where blue or near-UV light is common.

The second possibility is that the image forming the colour fringes has the right structure to produce a strong blue. It is commonly thought that a colour fringe should be a rainblow, or at least contain all the colours of the rainbow, but that is not the case. Goethe's "Farbenlehre" and Land's "Retinax" theory of vision are the two classic studies and fairly widely available on the web. In short, fringes from high contrast edges and lines can be almost any colour among the additive or subtractive primaries, although green and magenta are rare. What you see is not necessarily what a spectrometer would measure.

 

[Murray@uptowngallery]

cool info

Well Focal Point looks like a cool place...something tells me those are some seriously labor-intensive services.

Struan - I'll take another look.

The lens is too fat to mount on anything but a box. about 4" front diameter, 5" rear.

I thought I'd take it apart for the glass and iris but it's anything but easy.

Thanks.

 

[Sparky]

Take some pictures - you never know... it could well turn out to be the best lens you've ever used! Honestly - you never know 'til you try. The fringing you were getting as a result of shining a light through the back of it - may not be the same result you'll get with film at all... again... just try it, I'd say. At least have a look at the ground glass image.

 

[Murray@uptowngallery]

Yeah, it does look better on GG than projecting a light bulb onto a white wall, but until it's mounted, holding a ground glass between my eye and the lens is a pretty lousy preview.

It was 6.64" or approx. f.l. I think f/2.3-f/8, no flange, too fast for a hat, too fat for a Speed Graphic front end... The seller claimed the two halves come apart, but it's a one piece barrel (hah-ha - probably had the usual disclaimer I don't know nothin 'bout photography....) it was only $5 and they underestimated the shipping and didn't squawk.

Removing the damaged front element surely has some effect on the corrections but it had a very short back focal length before, about 1" (but the labeled f.l. didn't agree with the focusing distance???). Could have been a phototypesetter lens also. Now it's about 5".

I paid alot more for a spanner large enough to gut it than the lens and shipping...

I'm working on a shutter for it.

Ok, back to topic.