Air bubbles

[miran]

I have a couple of medium format wide angle lenses which have what I can only guess are air bubbles in the glass at some point (not the front element, I think, somewhere inside as far as I can see). Might also be tiny drops of some oily substance on one of the lens elements or something else but I don't think so as both are mechanically perfect and very clean. Either way, looking through these two lenses I can see some sort of tiny imperfections in the glass (plus some dust particles here and there of course). How do such imperfections affect image quality? Should I let these two lenses go and look for better copies or can I just forget about this issue? As far as I can see the IQ is good but I haven't tested thoroughly yet. Both have some problems with flare, but that's expected with wide lenses and both are non MC. Sharpness seems to be good.

[alexthekid]

Somewhere i found a statement that air bubbles are proof of good quality glass (something about production process). Don't know if this is true, but a few days ago i bought a Pentacon auto 135/2.8, that has an air bubble in one of the elements. Previously i had two copies of the same lens and i wasn't impressed with them, but after a discussion here i decided to try another one. Don't know if it's due to different glass, but the lens with the bubble is much, much better than the previous ones.

[andya147]

i have a contax 35 70mm 3.4 with lots of air bubbles and it is pin pin sharp.this link im sending is a superb read all about it,hope it helps.i was allso told serial numbers starting with 8 are sharper

http://www.zeisscamera.com/services_lens.shtml

[miran]

Thanks, I'll sleep easier now. Actually one of the two mentioned lenses (a 50mm CZJ Flektogon) only has three or four very tiny bubbles and the other (a 30mm Arsat fisheye) has two slightly larger ones but as I said, the image quality seems good to me though I've only tested on a 1.5 crop digital so far.

[Mos6502]

The process for producing optical glass without bubbles IIRC was invented by Nikon after WWII. For lenses made in the 1950s and earlier, bubbles in the glass was seen as a sign that the glass was of good quality.

For later lenses bubbles within the glass are less common, but they do occur. I wouldn't worry too much about a random bubble in an element unless it was right in the center of the lens. As I understand it QC counts the bubbles present in a specific volume of glass, so the larger an element is the more likely it will have bubbles but still be considered acceptable.

[Lightshow]

The info I came across said:
The materials for the crucible which holds the melting glass would break down after a period of time and would contaminate the product,
therefor there was not enough time for the bubbles to rise out of the molten glass, Better materials for the crucible were developed after the war.

As for what those bubbles do to IQ?
My guess is that they will behave much like a symmetrical lens, the light rays will bend at the glass/air transition, then bend back at the air/glass transition cancelling themselves out, like the bubble wasn't there.

[dnas]

[Lightshow]

I may be wrong, I'm not an optical engineer.
Another possibility/way of looking at it:
If you could focus the lens close enough to get the plane of focus(DOF) where the bubble is you would see it in the images, but since the plane of focus is no where near the bubble you don't see it.
ie. your eyes are focusing on the bubble, therefor you see it.

[miran]

Yes, I think so too. This is much like dust particles stuck on an element. They are there but are so far out of focus they are blured out of vision. They lower the amount of light that gets through the lens a little bit but as they're so small in relation to the entire surface, that effect is negligible. Again, in my case at least there doesn't seem to be any issue with IQ.

[Lightshow]

The design of the lens seems to dictate how much a flaw will be transferred to the final image.
eg. The Lensrentals blog post about the shattered front element and how little it effects IQ, and some test shots I seen from a lens that had a scratch on the rear element that did visibly effect IQ.
In the end, trust your eyes, if you don't see something, don't sweat it.

[JohnBar]

Air bubbles are cool.

I was surprised to find one in a brand new product Nikkor 1.8/35mm AF-S G lens. Needless to say its as sharp as a razor.

Out of interest, it would be fascinating to compute the light path through the bubble as the refractive index changes for a few photons as they pass through the air inside the bubble.

[dnas]

Yes that would be interesting. The fact that you can see an air bubble very easily means that it scatters the light quite a lot, due to both refraction and total internal reflection. However, the effect on the final image would be a function of location of the bubble (e.g. front element of back element, etc) , the size, and how many there are. One small bubble may not affect the sharpness that much, but I'd say that the contrast would be affected due to scattering of light.

[Lightshow]

Flare would be the worst issue I think.

[Tesselator]

I see what look like air bubbles every once in a while in older (pre 80's and 80's) lenses. So far without exception experts § have told me that those are due to separation. Basically flaws in the glue which hold doublets or triplets together. Sometimes separation looks like circular white wipe marks and sometime it looks like a distribution of tiny air-bubbles. It can happen with new lenses and newer glue types and it can happen with older lenses which use balsam based glues.

Either way of course it's not good and will affect the IQ adversely. How much I guess just depends on conditions like: Angle and strength of side or bounced light entering the lens. Focus distance. Scene brightness and contrast. Which element group (placement) is the one actually affected. and so on. I guess a standard park shot with your back to Sun focused 20 to 40 meters away and you can't tell any difference.

I think it's needless to say that if you see one in a lens which is still under warrantee you should return it immediately. It's not like a double-die penny or something where it's worth more because of flaws in manufacturing.



§ People I deal with professionally face to face who demonstrate repeatedly they know what they're talking about.

[miran]

Well, what I'm talking about are two or three small individual bubbles inside a single large element. At least that's what it looks like to me.

[iangreenhalgh1]

[Tesselator]

Those are massive! And of course they are going to affect the image. It'd be impossible for it not to. I haven't encountered bubbles like those before tho. Are you sure that's what he's talking about? Here's an image of a lens I owned with bubble-separation - a fairly common type:

http://tesselator.gpmod.com/Images/Temporary/Bad_9120478/_1060465_Bad_Voigtlander.JPG (voigtlander apo lanthar 125/2.5)

The rainbow (dispersion) kind you're talking about also exists too. I've seen that a few times too actually but didn't think to mention it - because I don't see that NEARLY AS often.

Here's the circular wipe mark type I was referring to:


That's also my lens, my image. And I repaired it too. That looks like it's on the surface but it was between those two cemented elements (you can see the overhanging lip where they glue together - and actually a third element glued to the bottom of those too (it's a triplet - have a cigar!).


Here's some websites if you need more information:
http://www.largeformatphotography.info/forum/showthread.php?66763-Separation-in-1940s-WA-Dagor

http://www.zeisscamera.com/services_overhaul-cIIa-lenses.shtml
"The following picture is of a 50mm f1.5 Sonnar that spent too much time in a closed car in Summer: [image] The group of bubbles in the center of the lens is not fungus. It is cement failure. "

There's a pretty good (extreme) example of bubble separation at that last one there.

And I've never seen bubbles actually in the glass elements. I've seen what looks like it could be in the glass - but on disassembly it became obvious it was in the cement. That's not to say there couldn't be bubbles in the glass. But I think if I saw that I'd never buy another model of that manufacturer's. A lens company that can't make glass properly... pretty weird! Too weird for me anyway...

I've handled and closely inspected about 650 lenses in the past 18 to 20 months. In the past year or two I've acquired 450 lenses and sold off about 150. That's rough tho - I currently have 384 (I think) lenses at last count. I've used and tested all of them too. I've averaged about 1000 shouts per prime and about 300 shots per zoom. I've opened and cleaned maybe 10% or 15% of that number and repaired about 120 lenses this year - tho mostly not from my own collection.

[iangreenhalgh1]

Bubbles in the glass don't affect IQ, period. Otherwise Zeiss would not have sold the lenses.

That rainbow type separation I showed is common.

Bubbles are not from separation, they are within the glass.

[sichko]

TIE-28 : Bubbles and Inclusions in Optical Glass - Technical paper from SCHOTT AG.

The paper includes a discussion showing how bubbles can affect image quality.

[miran]

What I see in my lenses is not even close to what's in those photos. If I were to take photos of my lenses on roughly the same scale, the "bubbles" wouldn't even be visible. You have to look really close to see anything. And as I said, it's only two or three tiny ones spread over a large element.

[JohnBar]

Interesting paper, thanks.

Most of my old lenses have the odd transparent bubble, I have a couple with semi transparent black bubbles which I initially thought was dust, but under high magnification confirmed to be bubbles.

Overall I think bubbles are there to be loved, and like art, is the acceptance of imperfection.

[sichko]