Ira Tiffen explains where lens reflections and flare come from, how lens coatings were discovered and how cinematographers control reflections on set.
Think you’re doing the right thing by cleaning your lenses? Ordinarily, that’s true. But once, about 80 years ago, as I heard it, a researcher in England did that and got a surprise that paved the way for the lighter, faster, more capable camera lenses you use today. Measuring light transmission through a series of individual lenses, he noticed that they were marked with fingerprints. So, logically, he cleaned them. Just as logically, he expected to find more light being transmitted. In fact, the opposite occurred. Having instead measured a reduced amount of light, most of us might have assumed that something was wrong with our readings and moved on. Fortunately, that’s not what happened here.
Trying to explain the phenomenon took an understanding of why light reflects and refracts. After all, have you ever wondered just exactly why it is you can see things, like glass, that are otherwise transparent? Well, usually, what we’re seeing are reflections off the transparent surface. Remove the reflections, and the object can actually appear to disappear. Let’s take a closer look at what’s happening here and why it is important to modern filmmaking.
Light traveling along in a certain direction strikes a piece of glass, say, a window; in doing so, its speed changes. The act of reducing speed when it enters the different medium throws off energy in the form of reflected light. While not intuitive, this is the case both when it enters the glass from the air and when it exits back into air again. Each time, there is a reflection.
To help us understand this further, and apply it to optical design and reflection control, we can measure the difference in the speed of light through the medium and relate it to the speed of light in a vacuum (which, as it happens, is very close to that in air). The ratio of the speed in a vacuum to the speed in the medium is a constant for any given material known as its absolute index of refraction. The higher the refractive index, the slower the speed of light through that medium.
In plain window glass, the usual amount of light lost to reflection is about four percent per side, totaling about eight percent. Now, this may not seem like much, but take a camera lens, for example, which has eight different glass lens elements assembled in a barrel. Not unusual. Light will be lost through each one to the tune of about eight percent. By the time the light has passed through to the camera, about half that light will be lost simply to reflections. That’s a lot, but it doesn’t end there.
Think about where all that reflected light goes. Light reflecting from the front of the first element goes back from whence it came, to the outside. However, light bouncing off the second, or internal side of that element bounces back into the glass and some of that is then reflected back into the camera from the first surface. This gets increasingly more complicated when the ninety-two percent of the light that makes it through the first element loses some energy to reflections off both the first and second surfaces of the second element, and then of the third element, and so on through the entire array.
What happens is that you start getting light bouncing back and forth, with an increasing portion being directed back into the camera, and showing up in your image as secondary (and potentially tertiary, etc.) reflections. These are generally referred to as internal reflections. This situation is made worse when you have bright lights, say street lamps, in several places in the image, and each one produces a separate reflection or series of reflections in various parts of the image. This can be exceedingly disturbing.
Now we can get back to our optical researcher. He correctly deduced that the increase in light transmission from the fingerprint-covered lenses was due to a similar reduction in light lost to reflection. Reflected light was reduced because the light wasn’t just passing through an air-to-glass or a glass-to-air interface, but also through the intermediate medium of the fingerprints. With the refractive index of the fingerprints falling between that of the air and that of the glass, the effect was to reduce the light lost overall to reflection. And that accounted for the increase in transmission.
Once this phenomenon was understood, applying it to lens production involved the development of what we call anti-reflection coatings. Various materials having specific refractive indices are applied in a vacuum chamber to glass surfaces; these materials can be counted on to affect light in a pre-determined manner. In particular, the more numerous the coatings, the more capable they can become. Today, multi-layer coatings can reduce light lost to reflection to a mere fraction of a percent.
The impact of this technology on lens design is huge. Instead of losing about half the light through those eight lens elements in our earlier example, coatings reduce that to a relatively minor amount. So to transmit the same amount of light as the uncoated lens, the coated lens doesn’t have to be as large, nor as heavy. This makes for lower cost, faster speed, and smaller, more compact lightweight designs.
Another effect of refractive index is that it is also a relative measure of the amount of bending of light when it passes the interface. The higher the index, the more the light will bend. Lenses have curved surfaces, so designed to bend light to our needs which, unlike flat filters or optical panes, obtain more light bending capability when larger arcs define the surfaces. The larger the arc, the thicker the glass, and the heavier it is.
When the refractive index is higher, the arcs of the outer lens curves can be smaller, resulting in thinner lenses. This, in turn means further reductions in weight.
Now what about those pesky internal reflections? Reduce reflectivity overall from each lens element surface and you reduce them substantially as well. However, when the light is bright enough, often when the sun is in or just outside the image frame, there is still a modicum of reflection that can be problematic. That is the reason lens shades were created. Minimize the extraneous light entering the lens and you reduce reflection problems. So use lens shades especially when encountering bright light that needs to be controlled in its ability to enter the lens.
Internal reflections are still potential problems when filters are placed in front of the lens. Filters are typically uncoated, both due to cost and to durability issues, although many manufacturers will offer some with coatings at additional cost. Not having coatings, then, filters can be a chief source of internal reflections, especially when more than one are stacked together, and bright lights are in the image.
With coating filters not often a practical solution, there is another that you can employ more readily. With filters of sufficient optical quality, you can divert the internal reflections away from the path into the lens by simply tilting them, usually anywhere from about five to fifteen degrees from being parallel to the lens. Some mounting systems, notably certain matte boxes, have this ability built-in. Otherwise, it will require some custom adaptation, and perhaps the use of duct tape. Only use this technique when reflections are a problem.
At the start of this article, I mentioned lens cleaning. As a finale, let’s look at ways of doing that effectively. First off, always check with the manufacturer’s recommendations for any optical surface care. Usually, these amount to careful use of a suitable lens cleaning cloth and/or a particular type or class of liquid cleaner.
Lens cleaning cloths typically exhibit the ability to trap small dirt particles without causing them to be pressed back against the glass. Most popular are the microfiber cloths having a fine mesh for doing this, and are made of a material, often a polyester, that also won’t, on its own, scratch glass or the (usually softer) coating.
Liquid cleaners will generally contain a water-based solution containing a lubricant, to let dirt slide off the glass without scratching, and an ingredient that will aid fast evaporation without streaking. Be sure that the cleaner you use has been recommended for optical lenses and filters as some cleaners also contain solvents that can damage lens or filter assembly cements.
Finally, if you still somehow find yourself on location not suitably equipped, rather than shoot through a dirty lens, you can carefully try a soft cotton cloth and enhance it with breath-applied condensation on the surface. This tends to be relatively pure, even after garlic pizza. Just don’t offend your AC!
Dirty optics may have you trying to explain why there’s a soft foggy flare on the screen when the director had explicitly demanded crisp clear images. Keeping dirt off the lens will keep egg off your face. In over 30 years of making optical filters, Ira Tiffen created the Pro-Mist, Soft/FX, Ultra Contrast, GlimmerGlass, and others, netting him both a Technical Achievement Award from the Academy of Motion Picture Arts and Sciences and a PrimeTime Emmy Award. Elected a Fellow of the SMPTE in 2002, he is also an Associate member of the ASC, and the author of the filter section of the “American Cinematographer Manual.”
Keep learning filmmaking
Want more lessons like this? Join StudentFilmmakers free for live webinars with working pros, and stream on-demand filmmaking classes in cinematography, directing, screenwriting, sound and post.
