When one makes a piece of glass for a lens, they start with a hunk of glass and grind it into the desired shape. When a lens is aspheric, this doesn't change; merely the shape the lens is ground into is an aspheric (typically parabolic) shape. It doesn't seem like this should cause much in the way of additional labor or time on the part of the lens maker. But aspheric elements significantly increase the cost of the lens. Why is this?

  • 1
    This tread seems to cover at least part of the expense question, not sure if it covers all of it - photo.stackexchange.com/questions/10823/…
    – dpollitt
    May 30 '12 at 18:30
  • @dpollit The provided thread does not give any details on why exactly it is more expensive to make and measure an aspherical lens. I would be interested to know, as well. May 30 '12 at 20:30
  • 2
    @pau.estalella - Some quotes from the other thread "Aspheric lens surfaces are much more expensive to produce, because they can't be ground by a natural rotary motion.", "aspherical[sic]...tends to be more difficult and expensive to carry out: glass spherical shapes are simple to make and measure; aspherical ones are not."
    – dpollitt
    May 30 '12 at 21:08
  • @dpollitt: I saw and responded to that comment in that question. That doesn't make sense -- paraboloids and hyperboloids can be ground using simple circular motions too, because they're still symmetric with respect to the center axis of the lens. (Hence my confusion) May 31 '12 at 1:30
  • If you take two pieces of glass, one concave and one convex, and simply start rubbing those against each other (both rotating and sliding, in all possible directions), you will eventually have two spherical lenses. If you search for "telescope lens grinding" @ youtube, you will find plenty of videos of very low-tech approaches that produce fairly good spherical lenses. Jun 4 '12 at 7:44

A lot of it is manufacturing costs, but it's not really quite like @Matt has described it (or at least not how I understand his description).

With spherical elements, you normally take a number of blanks, and mount them all to a roughly sphere-shaped holder (with "divots", so to speak, where you're going to mount the blanks). Then you have a sphere-shaped grinder/polisher that mounts around the outside (for a convex surface) or inside (for a concave surface) and lets you grind/polish a large number of elements at a time. Th exact number depends on the diameter and curvature of the element, but we're typically talking at least a few dozen, and especially for smaller elements, can be hundreds.

There are several ways of making aspherical elements (ground, molded, hybrid), but they have one thing in common: every one of them requires that you process elements individually instead of in large groups. Until fairly recently (1980's or so) about the only way to do aspherical elements was to grind them individually by hand. Those were/are restricted to very limited production at extremely high prices (Leicas, Nocturnal Nikkors, etc.) There are ways to semi-mass produce them now, but that's still a matter of loading one blank into a machine, and automating most of processing that one blank, then taking it out and loading one more blank. In most cases, that's in addition to doing normal spherical grinding as well.


Aspherical elements are expensive due to the manufacturing costs. It's relatively simple to make a lens with a spherical profile you just spin it infront of a polishing tool. With aspherical glass its nut just a case of deviating from a sphere, its about cutting a very precise but otherwise arbitrary shape, getting a perfect circle is not a problem, but getting a perfect curve with equation x,y,z is.

You can throw a round pot by hand, the spinning of the wheel guarantees a nice circular profile. Hiya could also stop the wheel and squash the pot into a non-circular shape by hand, but throwing a pot with a precise parabolic curvature would require a computer controlled tool, there's no way to move the wheel to do.it accurately by hand.

  • Many designs have elements that are more complex than even parabolic.
    – Skaperen
    May 30 '12 at 22:14
  • 2
    Putting the glass on a lathe-like device though is sufficient -- it's not like aspheric designs are asymmetric with respect to the center axis. Sure, you could get a "round" pot by hand, but perfectly spherical? Even spherical elements are ground to much much much higher levels of precision than are possible by hand. May 31 '12 at 1:29
  • 1
    A spherical shape, though, involves a spinning lens, a spinning wheel, and a single pivot point (with normal automatically achieved by the pivot). An aspherical element involves a spinning lens, a spinning wheel, a moving pivot and a constantly-changing normal angle. And some aspherical elements are a lot more aspherical than others; the complexity of the curve (even if you are only considering half of the sagittal section) can be astounding.
    – user2719
    May 31 '12 at 21:54
  • @BillyONeal: Actually that's not true. Optics can be turned on a lathe-like machine, but this does not produce high quality lenses that you'd find in an SLR lens, even a cheap one. This process is used to make cheap plastic optics (eg. what you might find in a childs toy) or for pre-shaping a glass optic prior to polishing. Also, high quality spherical optics Can be made by hand, and often are. An experience optical manufacturer can achieve surface figures measured in nanometers when hand-polishing.
    – Colin K
    Jun 4 '12 at 20:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.