The information technology networking of company, control and field levels requires the secure and stable transmission of high data rates. Fiber optic cables are ideally suited for this purpose. The components used in this transmission technology - such as pre-assembled connectors or patch cables - are usually tested in production or in the laboratory using attenuation measuring devices. The values determined are added to the articles in order to document the quality for the end user. So far, so good. However, if these components are now installed in the field and then measured again, the measured values often differ from the values documented by the manufacturer. It is not uncommon for the values to deviate by more than 50% from the documented values!
The reason for this discrepancy is that there are numerous methods for measuring the insertion loss (IL) of fiber optic components. If manufacturers and users want to minimize the deviation in results for multimode components, identical coupling conditions must be observed during the measurement. This can reduce the measurement uncertainty to up to 10 %.
First of all, both the transmitter - i.e. the light source - and the receiver should be guaranteed to have valid calibration certificates and allow a reliable and stable measurement process. To ensure that the transmitter and receiver understand each other, the correct wavelength must be set and matched to the other measuring equipment and the fiber optic components to be tested. Defined measurement cables and couplings ultimately enable reliable and reproducible measurement results. One of the prerequisites for this is the use of high-quality measurement cables with reference connectors that have been manufactured with low tolerances for geometries and ferrule dimensions. Couplings with a tightly toleranced ceramic guide sleeve are suitable as test couplings.

Before connecting the measuring cables to the test specimens, it is essential to check that the end faces of the connectors are clean and undamaged. Damaged or dirty end faces in particular can lead to subsequent contamination, consequential damage and even failure of the fiber optic components. For this reason, it is important to inspect the connectors of the measurement cables and the connectors of the fiber optic components to be measured and to clean them if necessary. The connectors may only be plugged in once all connectors and test couplings have been checked, are clean and undamaged. The visual inspection of fiber optic connectors is described in the standard DIN EN 61300-3-35.

Semiconductor lasers as the new standard

If the requirements described above are met, large deviations in the measured insertion loss values can still occur with multimode components. These deviations are caused by different measuring devices, as different measuring devices are generally used in the field than in production or in the laboratory. The corresponding transmitters therefore have different light sources, which in turn produce different light conditions. These different light conditions have an effect on the insertion loss measurement results, particularly in the case of multimode measurements on G50/125 µm and G62.5/125 µm quartz glass optical fibers. Important influencing variables here are the amount of light and mode distribution - they differ depending on the type of light source used.

Winding mandrels eliminate the higher order modes in multimode measurements.

© Phoenix Contact

In the early years of fiber optic measurement technology, light was coupled into the multimode optical fiber using light-emitting diodes (LEDs). Today, surface-emitting semiconductor lasers (VCSEL = vertical-cavity surface-emitting laser) are predominantly used to measure quartz glass multimode fibers G50/125 µm - for example in the OM3 and OM4 categories. While LED light sources generate a wide light beam with overfilled excitation, laser light sources generate a narrow, directed light beam with underfilled excitation. Underfilled excitation is when the light is strongly concentrated on the center of the glass core. The light is introduced into a few low-order modes. In the case of overfilled excitation, on the other hand, the light is introduced in high-order modes towards the glass cladding.

In recent years, different methods for defining the light distribution have been described and specified in standards as multimode excitation conditions. As the light propagates in different modes in multimode fibers, the elimination of the unstable higher-order modes is of particular importance for reproducible measurement results. A comparatively simple technical solution is offered by so-called glass fiber winding mandrels.

The operating principle of the winding mandrels is based on the different scattering of the light modes. While the low-order modes propagate stably along the length of the glass fiber core, the higher-order modes scatter, particularly in the case of strongly curved optical fibers. This curvature is achieved by means of a winding mandrel. The coated or jacketed multimode optical fiber is wound several times around such a mandrel with a defined diameter. Since the winding diameter is specified differently for the same fiber thickness depending on the selected standard, different light conditions can still occur at the coupling point of the fiber optic component to be measured (DUT = device under test). This means that the actual excitation at the coupling point is unknown.

New specification for light distribution

Example of an EF template for a G50/125µm fiber at 850 nm (source: DIN ISO/IEC 14763-3 Annex A): The enclosed radiant flux represents the distribution of light power over the radius of the fiber.

© Phoenix Contact

In order to eliminate this uncertainty as well, the experts from the international standardization committees have redefined the conditions for light distribution at the coupling point. The result is the 'encircled flux' (EF). The standardization does not require this excitation condition at the coupling point from the transmitter into the transmission cable, but at the coupling point to the tested fibre optic component - i.e. at the output of the transmission cable. This specified excitation condition is therefore fed directly into the fiber optic component to be measured. The specifications are described in the DINEN61280-4-1 standard and essentially define the excitation conditions for the multimode fibers G50/125µm and G62.5/125µm at the wavelengths 850 nm and 1300 nm. The enclosed radiation flux can be generated by connecting a mode controller or a so-called 'Encircled Flux Mode Conditioner' and checked by a near-field measurement with a suitable near-field measuring device.

Of the many methods for measuring insertion loss on fiber optic components, some have proven themselves over the years - and are used accordingly in practice.

The methods in practice

Comparison of light sources: LED light sources usually overfill the fiber, while lasers significantly underfill the fiber - VCSEL-capable multimode transceivers do not underfill the fiber as much as laser light sources. (VCSEL = vertical-cavity surface-emitting laser)

© Phoenix Contact

The methods according to DINEN61300-3-4 'Insertion method (C3) for patch cords' and DINEN61300-3-4 'Insertion method (C2) for connectors' are established as measurement methods in the laboratory and in production facilities. The standards describe the various methods for measuring the attenuation of fiber optic components. DINISO/IEC14763-3 is generally used for measurements in the field. This describes measuring equipment and procedures for the visual inspection and measurement of fiber optic cabling designed in accordance with ISO/IEC11801 or DIN EN 50173 or comparable standards.

In a nutshell: Measurements with different measuring devices, but the same normative structure, generate measurement deviations caused by different coupling conditions. Manufacturers of fiber optic components and measurement technicians in the field must therefore take into account the conditions that apply to the implementation of the measurement technology and the normative structure. In this way, measurement uncertainties can be minimized, reproducible results can be achieved and consistent quality of the fibre optic connection components in the field can be ensured.

Author:
Frank Kölske is a development engineer in the Field Device Connectors division at Phoenix Contact.