ND filters are used when there is a risk for overexposed images. Mostly they are favored for long-time exposure, because especially at good light conditions and slower shutter speeds, overexposed images can occur. An ND filter reduces the amount of light and allows us to capture a turbulent river as a smooth surface.
In addition, they help achieving a shallow depth of field by allowing the use of an open aperture, even in sunny conditions. Playing with the depth of field is a very important factor in photography, as a shallow depth of field is used to put a strong emphasis on the main subject by blurring the background.
On a sunny day there is sufficient light available to capture subjects. The shortest shutter speed may still be too long, despite the lowest sensitivity. Without further corrections, too much light would reach the sensor and the taken image would be overexposed.
How can we prevent or resolve this overexposure? To reduce the incoming amount of light it may help to stop down the aperture and thus, reduce the incidence of light. However, this is not always possible or maybe not wanted. Should we accept an overexposure and try to correct it later with an imaging editor? Theoretically, it would be possible, but basically it is not advisable. The brightness could be adopted, but the structures of bright elements would remain missing. Because extremely overexposed images loose details in bright parts of the image, which are mostly recorded as white planes.
N - Neutral
If a desired combination of ISO, aperture and shutter speed doesn't allow a correct exposure, the incoming amount of light must be reduced somehow - already before the light enters the optical system.
This is done by using a neutral density filter (gray filter), short ND filter. These are semi-transparent pieces of glass, cased in a plastic frame, which are mounted in front of the lens. The assembling varies and the filters can be screwed on or plugged on the lens.
Depending on the gray tone, ND filters reduce the incident light very evenly.
But why gray? As the name implies, it is a neutral filter, whose gray coloring ensures pure color recording. Using a colored filter, the image would have a corresponding color cast.
Gray filers are typically made of glass or plastic. Depending on the intensity of the gray tone (darker/brighter) light is absorbed more or less. It is: the brighter the gray tone, the more light passes through, and vice versa. The level of transmission of an ND filter is specified in optics as optical density. These can be calculated using the formula:
D = log (1/T)
T stands for the transmittance and indicates the proportion of the passing light. The factor can be calculated by the ratio of outgoing to incoming light intensity.
T = IOUT / IIN
A filter that lets the light pass to 100% has an optical density of 0. If only half the amount of light passes through, an optical density of 0.3 results.
T = 100/100 = 1 → D = log (1) = 0
T = 50/100 = 0.5 → D = log (1) = 0.3
In photographic relation, the density value can also be expressed in f-stops. As a guideline, an optical density of 0.1 corresponds to a third aperture stop. Logically, considering that halving the amount of light at constant sensitivity and shutter speed, equals a full f-stop (see article Exposure Value). A filter that halves the incoming amount of light has a density of 0.3 (0.1: 1/3 f-stop → 0.3: 1 f-stop)
f-stop = 10-D
Exposure factor ND X
The declaration on neutral density filters not always describes the density. Some manufacturers provide the so-called exposure factor „ND X“. For example, ND X2 or ND X4. The number represents the factor by which the shutter speed is extended.
The exposure factor allows an inference to the transmittance, thus the ratio of outgoing to incoming light. Take ND X2 for example. Using such an ND filter would result in doubling the shutter speed. This means, only half of the light will reach the sensor. The transmittance, which is halved, now allows calculating the density.
T = 0.5 because IOUT / IIN = 50/100
D = 0.3
Differentiation of ND filter declarations
How do you know whether the information on ND filters is the density or the exposure factor? Usually, the spelling reveals it. Is the density written on an ND filter, you will find a decimal behind the indication ND. Have you found the description ND X, with X as a capital letter instead, the exposure factor is given.
ND0.3 → density 0.3
ND X4 → exposure factor 4
Single, combined or variable
There is a vast choice of ND filters with different densities. This can make it difficult to determine which density is best suited for various subjects and illumination levels. So e.g. when you are travelling, probably none of your filters reduces the amount of light sufficiently. For such purposes, it is possible to combine multiple ND filters – filter stacking. Of course the filters have to be connectable with each other, either over a screw- or click mechanism. If it is possible to combine the filters, a new level of transmittance results. To determine this, it applies that densities/apertures are added with each other and the exposure factors are multiplied.
Filter A: density 1 / exposure factor: ND X10 / f-stops: 3 1/3
Filter B: density 2 / exposure factor: ND X100 / f-stops: 6 2/3
Combination of filter A and filter B
density = 1 + 2 = 3
exposure factor = 10 * 100 = 1000
f-stops = 3 1/3 + 6 2/3 = 10
Besides the classical ND filters that offer a fixed density, variable ND filters exist. Here, the density can be adjusted continuously to match the situation. Especially for traveling this it is advantageous, because controlling the incident light for the different situations with only one filter is possible. Most variable ND filters have no indications for the different densities. Thus, it is difficult to read out the density itself and therefore tricky to determine the needed shutter speed/aperture.