Fiber Collimators

The beam radius of the obtained collimated beam depends on the circumstances. In some cases, the beam diameter is as small as the fiber diameter, e.g. 125 μm; the Rayleigh length can then be less than 1 cm. In other cases, one needs beam diameters of several millimeters or even more.

For calculations, the simpler case is that of a single-mode fiber. Here, the beam radius can be calculated with reasonably good accuracy using the following equation:

This assumes that the beam profile of the fiber mode has an approximately Gaussian shape, so that we can apply the corresponding formula for the beam divergence half-angle θ_fiber.

It is also assumed that the distance between fiber end and lens is close to the focal length f of the lens. If the distance is too small, the beam will diverge, and for too large distances it converges to a focus at some distance. It can be useful to get slightly into that latter regime, where a beam focus (with a beam diameter slightly below that at the collimator) is reached in a suitable working distance. The longer the focal length, the less critical is the longitudinal positioning.

Note that a smaller mode size of the fiber implies a larger beam divergence and thus a larger collimated beam for a given focal length. This also means that a shorter wavelength, which usually leads to a smaller mode size, leads to a larger output beam. This holds even more if the fiber gets into the multimode regime for sufficiently short wavelengths. For such reasons, a visible pilot beam for an infrared beam, for example, may not accurately show the size of the infrared beam. Also, the correct fiber positioning for collimation may depend on the wavelength, particularly if no achromatic lens (see below) is used.

For multimode fibers, the beam divergence at the output (and thus the collimated beam size) depends on the launch conditions, and possibly even on the condition (e.g. bending) of the fiber. Generally, the beam divergence angle will be larger than according to the estimate for the single-mode fiber – possibly even much larger.

Fiber-optic collimators are available for different collimated beam sizes, which simply means different values of the focal length. Naturally, devices for larger collimated beams need to be both longer and larger in diameter. The largest fiber collimators are those for high-power multimode fibers as used in laser material processing or for pumping high-power lasers; they also need to be optimized for reliable operation at high optical power levels.

Some fiber-optic collimators have adjustment screws for controlling the beam direction (by an integrated tilt adjustment) or possibly even for the fine longitudinal positioning (adjustment of focusing or working distance). Others don't have such adjustment options, and one may position and align the whole collimator with additional opto-mechanics.

Different kinds of lenses can be used in collimators. For standard telecom fibers and in fact many others, one mostly uses GRIN lenses (gradient-index lenses), as these are relatively cheap and small. However, they are less suitable for larger beam diameters, e.g. of more than a few millimeters. In such cases, one tends to use conventional singlet or doublet lenses, which may be of spherical or sometimes aspheric type. This is needed, for example, when a collimated beam needs to be transmitted over a large distance, such as in free-space optical communications, where a long Rayleigh length is required.

Special requirements may lead to the use of special lenses. For example, achromatic doublet lenses are used if beams with quite different wavelengths need to be handled, as otherwise proper collimation may not be achieved for all wavelengths. Aspheric lenses may be used in cases with large beam divergence from the fiber (i.e., for fibers with small mode radius) in order to eliminate spherical aberrations.

The insertion loss of a single fiber collimator can be pretty small – of the order of 0.2 dB (i.e., a few percent) or even lower. It depends on various factors, such as anti-reflection coatings and dirt on the lens. It should not matter, however, whether a bare fiber or a connectorized fiber is used.

The insertion loss for a pair of collimators, as used for fiber-to-fiber coupling, may be substantially higher than the sum of insertion losses of the two devices. Particularly for single-mode fibers, it is important to achieve good mode matching. Obviously, both collimators should have the same collimated beam size. Depending on the exact longitudinal positioning of the fibers in the collimators, some non-zero distance between the collimators may be ideal. This also allows one to insert additional optical elements, such as optical filters or polarizers.

Angled fiber ends are often used to suppress back-reflections from the fiber end face into the core, i.e., to maximize the return loss. Unfortunately, the angle leads to some deflection of the output beam.

For some fiber connectors with inclined fiber connection, this can be compensated by some tilt of the fiber fixture. Otherwise, the beam from the fiber will hit the lens at some angle. After the lens, the beam direction should nevertheless be in the direction of the fiber (assuming correct longitudinal and transverse positioning), but it will be somewhat offset from the center of the lens. That may also lead to increased insertion loss and to beam quality deterioration if some clipping, reflection or scattering occurs at the edge.

In principle, a fiber collimator can also be used “in reverse”, i.e., for launching a free-space beam into an optical fiber. However, it usually does not provide the required tools for fine adjustment (which are particularly required for single-mode fibers). That adjustment then has to be done e.g. with turning mirrors in the path of the input beam.

There are also fiber launch systems with additional adjustment tools for different degrees of freedom, allowing the launch of a non-adjustable input beam.

Your question or comment:

Spam check:

  (Please enter the sum of thirteen and three in the form of digits!)

By submitting the information, you give your consent to the potential publication of your inputs on our website according to our rules. (If you later retract your consent, we will delete those inputs.) As your inputs are first reviewed by the author, they may be published with some delay.

[1]	R. Paschotta, tutorial on "Passive Fiber Optics", Part 13: Fiber Accessories and Tools

(Suggest additional literature!)