Fiber optic connector

Types of Fiber Splices

Optical fibers can be joined together so that light is efficiently transmitted from one fiber to another. There are various possibilities:

• Mechanical splicing refers to tightly fixing two fiber ends together by some mechanical means. This is typically used for permanent connections, but also allows removal of the splice without damaging the fiber end.
• Another technique is fusion splicing, in which fibers are fused together, for example using an electric arc. This results in particularly low insertion loss and very high return loss if the two fiber cores are similar.
• For non-permanent connections, fiber optic connectors can also be used (see below).

Figure 1: Microscopic visualization of a fusion splice between a photonic crystal fiber (PCF, left) and a conventional fiber (right). The hole pattern of the PCF can be seen. Photo courtesy of NKT Photonics.

Coupling Loss of Imperfect Fiber Connectors

A common question is how much is the coupling loss, e.g. at a mechanical joint, when there is some kind of imperfection such as:

• Parallel offset of the core,
• Deviation between fiber axis directions,
• Core size mismatch, or
• Air gap between fiber ends.

It turns out that some of the answers are quite different for singlemode and multimode fiber.

single mode fiber

Calculating the coupling loss of a single-mode fiber is relatively easy. Essentially, the guided mode from the first fiber (input) creates some amplitude distribution in the second fiber, which may be somewhat displaced, for example, due to imperfect splicing. The coupling efficiency can now be calculated as the overlap integral between this amplitude profile and the guided mode of the second fiber. (Numerical beam propagation is not required.)

For the case of Gaussian mode profiles with different mode radii and some parallel offset, the equations we have already discussed in Section 3 can be used:

A similar equation can be used for angular mismatch:

This suggests that angular alignment is more critical for single-mode fibers with large mode areas. For standard pattern regions, angular alignment is usually easier to achieve than positional alignment.

The figure below is based on the above equation.

the

Figure 2: Insertion loss at a mechanical splice of a single-mode fiber due to mode radius mismatch.

Figure 3: Insertion loss at a mechanical splice of a single-mode fiber due to core parallel offset.

Figure 4: Insertion loss at the mechanical splice of a single-mode fiber due to angular error, which may be due to non-perpendicular cleaving. This has been calculated for different ratios of mode radius to wavelength.

With regard to angle-cleaved fiber ends, it is often of interest what the cut angle needs to be to avoid significant reflections into the core mode. The equation works fine for this; just remember that the angular deviation of the reflected beam is twice the cut angle. For example, a fiber with a standard mode area of ​​100 μm2 and w = 5.64 μm requires a cut angle of at least 7.4° in order to have a back reflection below 10 -4 , i.e. a return loss of at least 40 dB at a wavelength of 1.5 μm. For a large mode area fiber of 1000 μm 2 , 2.3° is sufficient. Note that longer wavelengths require larger cut angles as they result in larger beam divergence.

Note, however, that the above equations are only valid for modes with a Gaussian distribution. However, in the case of highly suppressed back reflections from angled fiber ends, the results are sensitive to deviations from the Gaussian mode distribution. There, one has to calculate the back reflection numerically from the calculated mode distribution, for example using our RP Fiber Calculator PRO software.

It’s interesting to think about more details. For example, if the mode sizes are different, does the loss of which fiber the input light comes from matter? According to the formula above, it doesn’t. This is true, though perhaps surprising: conceivably going from a smaller core to a larger core would result in lower losses than going in the other direction. Note however that smaller modes have larger beam divergence, i.e. wider field distribution in spatial Fourier space, which is too large for other fibers with larger modes. Therefore, the proportion of power lost at the joint does not really depend on the direction. Only the distribution of the lost light over the cladding modes differs.

Conceivably, when switching from a fiber with a smaller mode to one with a larger mode, if the second fiber has the same NA (albeit with a larger core), coupling losses can be avoided. (If the V number is low enough, it might still be single-mode.) After all, the angular range should only be limited by NA. However, this expectation is wrong; if both fibers are single-mode fibers, a mismatch in mode size will inevitably lead to coupling losses.

multimode fiber

For multimode fibers, losses cannot be specified as a single number: they are usually mode-dependent. This means that for any input optical field, the total loss incurred will depend on how the power is distributed across the modes. For example, one can imagine that light can only be launched into low-order modes by lasing, which would lead to low splice losses. If the fiber is strongly bent before splicing, light may be redistributed to higher-order modes and the splice loss will become large.

Using coupling loss as an example, consider a perfect mechanical splice between two step-index multimode fibers with equal NAs of 0.2 (calculated from the maximum index difference), but the first fiber has a core diameter of 62.5 μm , while the core diameter of the second fiber is only 50 microns. We can calculate the coupling loss for each mode of the first fiber by summing its modulus squared with the overlap integral of all modes of the second fiber. (Alternatively, numerically simulated beam propagation can be used, but this requires more computation time and is temporarily less accurate.) Figure 2 shows the loss versus the m-value of the mode. These losses are highest for low m but high l values.

Figure 5: Mode-dependent coupling loss at a multimode fiber connector. The horizontal coordinates reflect the m value for each mode, while the color depends on l.

It might come as a surprise that the coupling loss for the LP 14,3 mode is so high – about 10 dB, much higher than that based on the mode area ratio (1.94 dB). However, the mode has a significant fraction of its power outside the 25 μm radius, and its intensity distribution in Fourier space is also far away. The calculated results have been confirmed by calculations based on numerical beam propagation, which is a very independent check.

If the two fibers are swapped, i.e. the input is from the smaller core, the coupling loss for all modes becomes smaller:

Figure 6: Same as Figure 3, but with light input into a fiber with a smaller core.

So for multimode fiber, except for singlemode fiber (see above), the coupling loss is much smaller when coming from a fiber with a smaller core. However, for some modes, these losses are still significant – for example, LP 55 mode has a loss of 2.8 dB. Numerical beam propagation confirms this result. It shows that the field starts to expand when it enters a fiber with a larger core, and that the expansion later does not stop completely at the new core/cladding boundary. This suggests that not every field distribution and limited angular content within the core is well matched by guided modes.

This effect is less pronounced for fibers with multiple modes. Basically, one has to realize that the modes of the smaller core span a mathematical space, not a subspace of the larger core.

Effect of air gap

In mechanical splices and connections based on fiber optic couplers, a tiny air gap can be formed between the two end faces. One might think that this would result in high insertion loss and low return loss through Fresnel reflections from the end face. However, this is not the case if the size of the air gap is much smaller than the wavelength of light. In this case there is almost a π phase shift between the two reflections, so they largely cancel each other out by destructive interference.

Even for slightly larger air gaps, the distance between the fiber ends is usually at least well below the Rayleigh length, so no appreciable wavefront curvature occurs over this path length. The total transmittance and reflectance can then be approximated using a simple formula for the Fabry-Perot interferometer. Figure 4 shows the results for silica fibers.

Figure 7: Insertion loss and return loss due to air gaps at mechanical fiber optic connectors.

In the worst case, the insertion loss is 0.6 dB. For less than 0.1 dB, the air gap width should be less than 0.07 wavelengths—for example, for 1.5-μm wavelengths, the air gap width should be less than 105 nm. This would also allow for very low coupling losses if distances close to half a wavelength or one wavelength could remain stable.

The optical fiber connector

Fiber optic connectors are commonly used at the ends of fiber optic cables to provide non-permanent connections between fiber optic coupled devices. In principle, they are used in a similar way to electrical connectors. However, their use usually requires more care because the fiber ends are relatively sensitive and fiber optic connectors are not always easy to attach to the fiber ends.

Be aware that even tiny dust particles on the fiber core can cause significant damage. (For this reason, fiber optic connectors are often protected with dust caps when not inserted.) In addition, small imperfections in the fiber ends can cause small air gaps between the fiber ends, causing reflection losses (see above).

A variety of fiber optic connectors have been developed, for example, for applications in fiber optic communications. Some common types are ST, FC, SC and LC connectors. Different connector types differ in various aspects such as cost, size, ease of use, insertion loss and return loss, suitable fiber size, number of mating cycles allowed, multimode, singlemode and polarization maintaining fiber applicability and various other details.

See our encyclopedia article on fiber optic connectors for more details. Section 13 on fiber optic accessories and tools may also be helpful.

The optical fiber connector

Various optical components such as fiber couplers and laser diodes are often sold with fiber “pigtails”. This means that some fiber hangs out of the device, and the user can splice it to some other fiber, or attach a fiber optic connector to it.

Pure fiber pigtails can also be purchased, ie without optical components. In this case, one gets a fiber optic connector on one end of the (usually short) fiber but not on the other. For example, open ends can be integrated into some devices, avoiding the work of assembling the connectors yourself. Of course, it is also possible to take some jumper cables and cut them into two braids.

Some fiber pigtails only have some polymer buffer, but no thick jacket like fiber optic patch cords. And jacket braids.

With regard to angle-cleaved fiber ends, it is often of interest what the cut angle needs to be to avoid significant reflections into the core mode. The equation works fine for this; just remember that the angular deviation of the reflected beam is twice the cut angle. For example, a fiber with a standard mode area of ​​100 μm2 and w = 5.64 μm requires a cut angle of at least 7.4° in order to have a back reflection below 10 -4 , i.e. a return loss of at least 40 dB at a wavelength of 1.5 μm. For a large mode area fiber of 1000 μm 2 , 2.3° is sufficient. Note that longer wavelengths require larger cut angles as they result in larger beam divergence.