Passive Fiber Optics
2: Fiber modes
3: Single-mode fibers
4: Multimode fibers
5: Fiber ends
6: Fiber joints
7: Propagation losses
8: Fiber couplers and splitters
9: Polarization issues
10: Chromatic dispersion of fibers
11: Nonlinearities of fibers
12: Ultrashort pulses and signals in fibers
13: Accessories and tools
Part 5: Fiber Ends
Preparing Clean Fiber Ends: Stripping, Cleaving, Polishing
In most cases when a fiber is used, it is essential to prepare clean ends. A first step is usually to strip the polymer coating on the last centimeters, using a mechanical stripper. In in problematic cases, one may have to use a solvent (chemical stripping). The mantle of the glass fiber will then usually be quite clean, but the fiber end, if it simply has been broken, will still have an irregular shape. We thus need some method to obtain a nice surface – normally, a flat surface, which is perpendicular to the fiber axis, or sometimes with some other angle.
The most common method for preparing clean ends is cleaving. Essentially, this means controlled breaking of the glass of the bare fiber. One begins with making a tiny scratch on the side of the fiber, e.g. with a sharp diamond, carbide or ceramic blade, before or while some defined tension or bending is applied to the fiber. This causes the fiber to break, starting at the mentioned fracture point. Often, the resulting surface is quite smooth.
Cleaving is often done with a simple diamond blade. One slightly scratches the fiber and then breaks it e.g. by tipping an end with a finger. This procedure requires some practicing, and the results are somewhat variable. For more consistent results, one needs to cleave under more controlled conditions, using a precision fiber cleaver apparatus. Some of these devices can also be used to prepare angle cleaves (see Figure 2), with a relatively well controlled angle between the cleaved surface and the fiber axis.
Cleaving gets more difficult in non-standard situations, such as large fiber diameters or non-standard glass compositions. When cleaving fluoride fibers, for example, one at least needs to use adapted parameters for a precision cleaver.
Recleaving a fiber can be a substitute for cleaning, as it is hard to reliably clean fiber ends.
For very high-quality fiber surfaces, or when using fibers with large diameter, or when attaching a fiber connector, it may be necessary to apply some polishing procedure after cleaving. One will typically insert the fiber end into a ferrule (a hollow ceramic, glass or metal tube) and fix it there with a glue. The fiber is then polished down together with the glass tube, using a special polishing machine. This procedure allows one to produce a high-quality surface with an arbitrary well-defined orientation of the fiber surface. However, it takes substantially more time than simple cleaving, and of course it is essential to have all details of a polishing machine (e.g., the load force, speed and time) and the polishing agent well adapted to the ferrule and fiber material and size. Hand polishing is also possible, but usually leads to inferior results. See the article on polishing of fibers for details.
Polished fiber ends, other than cleaved ends, may have some convex curvature, resulting from the use of a flexible polishing pad. Such a “domed surface” facilitates a good contact e.g. between two single-mode fibers in a connector set.
Relevance of Cleave Angles
In some cases, it is important to have a cleaved fiber surface just perpendicular to the fiber axis. For example, this is often the case when a fiber is inserted into a fiber connector (see part 6), although some connectors require angle cleaves. Mechanical splices also don't work well with non-perpendicular ends (see Figure 1).
Note that due to refraction at the fiber end, a non-normal cleave causes a deviation of the output beam direction from the fiber axis (see Figure 2). Also, one then requires an appropriately tilted input beam for efficient launching. This makes the use of angle cleaves somewhat inconvenient.
The cleave angle also has an important influence on back-reflected light. If it is small, light reflected at the output surface (Fresnel reflection due to the index difference to air) will essentially travel backward in the fiber core. For large enough cleave angles, however, the light will entirely get into the cladding and will be lost there. This means that there is a very large return loss (e.g. 60 dB) despite a significant reflection, which for a normal cleave would cause a return loss of only 14 dB.
It depends on the fiber details how large the cleave angle needs to be for a high feedback suppression. For a usual single-mode fiber, for example, the mode has a beam divergence of several degrees. One may then require a cleave angle as large as 8°, for example. For a fiber with high numerical aperture, it may be even more. For large mode area fibers, however, rather small cleave angles are sufficient to suppress feedback.
Other End Shapes
In most cases, fiber ends are just flat – either perpendicularly cut or at some angle against the fiber axis as discussed above. In some cases, however, one uses different geometrical shapes of fiber ends:
- Lensed fiber ends are equipped with a strong curvature which leads to collimation or at least a reduction in beam divergence of the beam exiting the fiber. Due to the typically rather small core sizes, rather small curvature radii are required to obtain a substantial lensing effect. A particular implementation is the fiber ball lens, where a tiny glass sphere is fused to a fiber end. Special fusion splicers can be used for that purpose. The natural surface tension of the glass facilitates the fabrication of high quality fiber ball lenses.
- The above mentioned glass sphere can also be processed further; for example, it can be equipped with a reflecting flat surface which reflects the outgoing beam to the side. This is useful, for example, for some medical applications where a fiber is embedded in an endoscope.
- There are fiber axicon lenses, where near a fiber end the fiber diameter is rapidly reduced down essentially to zero. This can be achieved either by polishing (leading to a kind of pencil shape) or with the tapering technique. Only in the latter case, the core size is also reduced towards the end; this aspect, however, may not be important for the performance of the device. Light coming from the fiber and going through such an axicon end is focused down to a rather small diameter, so that it can be launched into a very small waveguide of a photonic integrated circuit, for example. Conversely, light from such a waveguide can be efficiently transferred into a single-mode fiber.
- A fiber end may be tapered down (→ tapered fibers) and then cleaved in the region with reduced fiber diameter. Such a piece can be used for a mode field converter, if the reduced mode size at the smaller end fits to that of a different kind of fiber.
- A core-less end cap is a homogeneous glass part spliced to the end of a fiber. (In case of a photonic crystal fiber, one may simply collapse the holes in the end region using a fusion splicer.) Light coming from the fiber core will expand within the core-less end cap, so that its beam radius is substantially increased (at the intensity decreased accordingly) once it reaches the glass/air interface. Such devices allow the transfer of light at very high power levels from a fiber into air or vice versa.
Questions and Comments from Users
Here you can submit questions and comments. As far as they get accepted by the author, they will appear above this paragraph together with the author’s answer. The author will decide on acceptance based on certain criteria. Essentially, the issue must be of sufficiently broad interest.
Please do not enter personal data here; we would otherwise delete it soon. (See also our privacy declaration.) If you wish to receive personal feedback or consultancy from the author, please contact him e.g. via e-mail.
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.
These sharing buttons are implemented in a privacy-friendly way!