Guessing Lens Design Without Disassembly

Started 1 month ago | Discussions thread
Flat view
Tons o Glass 0 Class
Tons o Glass 0 Class Contributing Member • Posts: 617
Guessing Lens Design Without Disassembly

Usually an internet search is sufficient to determine the number of elements and groups in a particular lens. Lots of us are curious about these kinds of things here on this forum, and many of us have done some disassembly. But when that information isn't available, or if you're just curious, you can pretty reliably guess these specifications (and perhaps more if you're good at it) without any disassembly(!?).

You just have to stare through your lens (wait, you guys don't do that? ). By examining and counting reflections on lens surfaces (internal and external). All you need is the lens, your eyes, your hands, a single light source, and preferably an otherwise dim environment. With a bit of experience, you can get pretty good at it!


-- hide signature --


Air-to-glass surfaces are easy - each surface of a single element will have a reflection. They'll often be tinted one color or another due to lens coatings (if you want to know why, peruse this article by Rick Oleson). You need to move the lens around a lot to make sure you see them all.

Cemented surfaces (as in cemented pairs [doublets] or triplets) are a little trickier:

  • two surfaces mated with balsam / resin will appear to share one reflection, but
    • the reflection will be dimmer (sometimes to the point of being very hard to see by comparison if the balsam is optically well matched to one or both of the surfaces it's joining)
    • the reflection may be devoid of tint (cemented surfaces are often uncoated)
    • the reflection may be little bit rainbowy as you move the lens around a bit (polished lens surfaces aren't perfect - minute changes in the thickness of the balsam, or stressed balsam, are to blame).
    • these harder-to-see surface reflections can hide behind brighter surface reflections - you need to move the lens (or your light source) around quite a bit to view reflections from different angles.

If you stop the aperture all the way down, you can split up the work. By counting/observing reflections from each side individually (where you consider each half to be its own lens at that point, then sum things together after the fact), you may make things easier on yourself while also being able to determine how symmetrical the design is by knowing where the aperture is placed.

The reflections layer/line up in such a way that it can sometimes help to determine, for example, whether the four elements in two groups you have found are two doublets or a triplet and a single lens. They way they get layered/line up might also tell you the order in which those groups are placed. It's difficult!



Examine the lens with your light source while moving it around. Count the number of reflections you see (cemented surfaces are counted twice) to end up with the total number of surfaces in your lens. You should end up with an even number - if you didn't, you missed a reflection, or fudged your count in some other way.

The number of elements is your surface total divided by 2. The total number of groups is the number of elements minus the number of cemented surfaces you found.

Afterword and an Example


Did this post need pictures? Sorry, it's gonna be mostly textual. It would actually benefit from video too...

Is this witchcraft to you? Would you rather take a mystery lens apart? Let me know what you think or if you have any tips.

Here's a simple example of the front half of a lens (with a known design - RE Auto Topcor 100mm f/2.8 [five elements in three groups]) with one sloppily executed picture (that should have been at least two or three):

These reflections are all before the aperture. Red lines indicate regular surfaces. Blue lines indicate where I counted surfaces twice because they look to be cemented surfaces (there is a very dim reflection under the largest/brightest reflection, and the neutral reflection that is dimmer than the others is a bit rainbowy as you move the lens around). The bright yellow smudged reflection is actually two reflections/surfaces (moving the lens around would show this). Total surfaces in the front = 8. Total elements in the front is then four. Total groups in the front is then four minus two.

Looking at the back we'd find two more tinted reflections which implies one more lens/group. So there you have it, 5 lenses in 3 groups, with two groups up front and a single element in back. Figuring out that there is a triplet in there as opposed to two doublets by looking at reflections alone requires a bit of magic that I haven't mastered yet.

Hope someone finds this amusing or useful!

 Tons o Glass 0 Class's gear list:Tons o Glass 0 Class's gear list
Canon EOS 550D Sony Alpha NEX-3N A3000 Sony a7 II YI M1 +2 more
Flat view
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow