NL9400638A - Measurement standard for optical attenuation measurements. - Google Patents

Description

Measurement standard for optical attenuation measurements.

The invention relates to a measuring standard for optical attenuation measurements, consisting of an optical cable with at least one artificially aged fiber-optic fiber of known damping and known length in a cable jacket filled with tough medium, which consists of a material that protects the fiber-optic fiber against external influences for a long time.

As a rule, backscatter reflectometers (English OTDR: Optical TimeDomain Reflectometer) are used for the measurement of optical attenuation. With this, attenuation coefficients, splitting and plug losses can be determined, as well as faults (fiber breaks) in optical cables can be located. During the measurement, a short intensity light pulse is radiated into a light waveguide and the damping coefficient, as well as the location and magnitude of the losses, are determined from the time course of the backscattered or the incident light power.

Monomode optical waveguides (calibration cables) are predominantly used for calibrating a backscatter reflectometer. The optical fibers are aged through temperature cycling to achieve the longest possible stability as long as the measurement standard must meet a range of requirements due to the necessary long-term accuracy and consistency. Optical calibration standards are known from DE-A-3910503.

The attenuation coefficient of the calibration cable is measured as accurately as possible over a predetermined wavelength range using the internationally recognized cut-back method. The cut-back method provides an integral damping coefficient over the total length of the fiber. The length of the light waveguide must also be known for the measurement. This is either calculated from transit time measurements of short light pulses using the effective refractive index of the fibers or determined via calibrated mechanical measuring devices. The length measurement should take place with an accuracy of 0.1% or better still of 0.03%.

The attenuation coefficients for optical cables are specified by individual telecommunications companies down to 0.01 dB / km. Therefore, values for a measurement standard to an accuracy of at least 0.003dB / km, if possible of 0.001dB / km should be known.

To protect against mechanical damage and environmental influences, as well as against shocks and temperature changes during transport, a primary coated calibrated fiber has hitherto been drawn in a gel-filled plastic tube for calibrating purposes. It is also known to use a fiber cable used with glass fiber reinforcement as a cable, which in diameter corresponds approximately to conventional inner cables.

When used as a cable with glass fiber reinforcement, the problem arises that for still easy to handle weights of the cable wound on a drum, only lengths of less than 3 km can be used. As a rule, this length is not sufficient to measure the attenuation coefficients with an accuracy of 0.001 dB / km. The necessary length is about 6 km.

Embedding the optical fiber in a gel-filled plastic tube does not provide sufficient mechanical protection for a measurement standard. The diffusion of certain ions or gas molecules inwards through the plastic tube into the fiber, as a result of which significant changes in the damping values can be effected, cannot be excluded.

Hydrogen readily diffuses into quartz glass fibers, even when it is present in the ambient atmosphere only in small concentrations. The penetration of water vapor or OH-ions can also lead to damping increases in the long term through the effect of the stress cor- rosion in microchips in the fiber sheath. Alkali or other ions will probably also affect the damping in the long term.

Both methods, so both the hitherto appropriate use as a cable and the withdrawal into a gel-filled plastic tube, do not provide sufficient protection against environmental influences.

It is the object of the invention to provide a measurement standard for optical attenuation measurements, which satisfies the requirements mentioned in the introduction with regard to accuracy and length constancy of the attenuation coefficients and long-term stability.

The object is achieved according to the invention with a measuring standard, characterized in that the light-conducting fiber consists of at least two partial strands split with one another and the splitting attenuation is known wavelength-dependent. Advantageous embodiments are indicated in the subclaims.

Only with the invention is a measurement standard provided, which ensures the long-term constancy of the damping coefficients within narrow tolerances for each sub-segment of the fiber. A purposefully introduced splice expands the scope of the calibration standard.

In the manufacture of standard monomode fibers with a core doped with germanium oxide, variations in the production of the preform and in the production of fibers produce a fluctuation in the damping coefficient at 1550 nm between 0.185 dB / km and more than 2dB / km. From the stocks of the fiber manufacturer, a fiber as long as possible is preferably selected from the aforementioned material, the damping coefficient of which is as low as possible, preferably at most about 0.002 dB / km above the production technical minimum. A length of about 50 km is advantageous. Such a fiber is largely homogeneous. The attenuation coefficient of this fiber is measured in a predetermined spectral range (e.g. 1260 to 1650 nm) with an accuracy of 0.001 dB / km and the fiber in equally long sub-segments preferably from 25 km first, then 12.5 km and finally 6.25 km distributed, which are all measured with the same accuracy. In this way, eight sections of 6.5 km in length are each obtained, and the damping coefficient thereof can be obtained with the double, quadruple and eight-fold length, as well as comparison. The damping of the total length is close to the lower technically possible limit. This ensures that the differences between the partial fibers are minimal.

The 6 lengths with the lowest damping are chosen from the 8 sub-segments, because they guarantee the best homogeneity. The differences in attenuation coefficients of these sub-segments should preferably be between 1260 nm and 1350 nm, as well as between 1450 nm and 1620 nm, above 0.01 dB / km.

The 6 part lengths are drawn in a stainless steel tube filled with a tough medium, which is seamlessly welded with the laser welding method. As an extension paraffin, thixotropic gel or also powder can be provided.

In such a steel tube with an outer diameter of 2 mm and a wall thickness of 0.2 mm, 6 primary-coated standard monomode fibers with an outer diameter of 0.25 mm can be drawn. With a length of about 6km, which is sufficient for the measurement accuracy of 0.001 dB / km, this mini cable weighs about 70 kg. That is to say, including the coil and a transport container, a total weight of less than 100 kg is achievable. The measuring standard is correct in terms of weight in the laboratory and during transport.

The advantage of the stainless steel jacket is that, despite its small diameter, it guarantees sufficient mechanical protection for the measuring standard and allows greater cable lengths at the same weight. A further decisive advantage, which exists both over the usual use as a cable and the use of gel-filled plastic tubes, lies in the fact that the steel jacket forms a complete diffusion block against hydrogen, water vapor and other ions, atoms or molecules. It can therefore be assumed that the damping values of the potash fibers in the steel tube remain stable in the long term.

After retraction into the steel jacket, the damping coefficients of the individual cores are checked once more. When the fibers are retracted with the correct additional length measured for this application, identical values as before should be established. The fibers must be pre-measured on suitable measuring coils without curvature and micro-curvature losses. I The steel jacket cable is then subjected to a cyclic temperature change program (500 changes between -40 ° C and + 75 ° C for one hour each). . The subsequent control measurement should yield unchanged attenuation values. Since a comparison length of 2 to 3 meters must be cut off for measurements on the fibers used as a cable, a special tool for burr-free separation of the steel tube should be used.

After these process steps, a first with a second fiber segment is split. The splitting attenuation, which has a typical value of 0.1 dB, is determined accurately and wavelength-dependent after the cut-back method via the difference of the splitting segment attenuation and the sum of the values for the individual lengths at 0.003 dB. If required, further segments can be split and the damping determined as in the first split. The split is thus introduced in a targeted manner and its attenuation is determined depending on the wavelength. His damping should also not be negligible. The measurement standard stands out for its knowledge of all wavelength-dependent attenuation values at each location of the fiber and at the location of the splits.

The unprotected fiber segments in the splitting range are protected by a steel tubular jacket with slightly larger inner diameter, which is slid over the steel jacket of the first and second fiber segments and joined to it with a diffusion-tight connection. The curvature in the splitting region is chosen to be so large that no measurable curvature losses occur. This is ensured with a bend diameter of more than approximately 60 mm.

When using multiple fiber partial standards, the transition from the protective jacket, which surrounds all light-conducting fibers, to the plugs of the individual light-conducting fibers must also be protected against external influences. For this purpose, a steel tube jacket with a slightly larger inner diameter is used, from which the number of light-conducting fibers, such as fingers from a glove, thinner steel tubes emerge correspondingly. The thicker tube is diffusion-tightly connected to the stainless steel jacket and the thinner steel tubes, as already described, diffusion-tightly connected to the sockets. The splices and / or sockets are preferably permanently secured in a diffusion-tight split housing. Measures can still be taken to keep the humidity in the split house low, for example with silica gel as a drying agent.

After the splice connection and the protective sheathing have been established, the cable is again subjected to a temperature change program and the damping is then checked again, in order to ensure that sufficient long-term stability can also be expected for the splitting damping. Subsequently, the entrances and exits of the 12.5 km long segment and the four 6.25 km long segments are closed with optical sockets, the houses of which are connected diffusion-tight with the steel jacket. Plug-in connections with high reflectance are preferred. -ping applied, for example, plugs with beveled, convex end faces with physical contact when inserting (English: physical contact plug, PC plug). Preferably the plug sockets are built into a holder and then attached to a side flange of the cable drum.

For the calibration of an OTDR, the measurement standard offers the following application cases: 1. Calibration of the damping coefficient on 4 individual fibers. By cross-comparison, the certainty of the calibration can be increased and its influences can be recognized and eliminated by defective plug end faces. Long-term changes are also recognizable, insofar as they do not have the same effects for all four segments.

2. On the split 12.5 km segment, both the local location and the measured value for the split loss can be checked and, in addition, the damping coefficient can be calibrated to a length of 12.5 km.

3. By connecting all fiber-part standards one after the other via short connection cables with optical plugs, a measuring range of a total of 37.5 km is available with one split, which can be either at the beginning or at the end of the measuring range. and with four plug connections in known place.

If required, other combinations can also be constructed, such as 2 times 2 split individual lengths and 2 segments of 6.25 km each.

The invention is further described in the figures, in which: Fig. 1 schematically shows a measuring standard wound on a drum; Fig. 2 shows an embodiment of a socket housing; Fig. 3 shows the detail of the protective jacket end and Fig. 4 two examples of a switch connection of the fiber part standards.

Fig. 1 shows a cable 2 wound on a drum. The cable 2 (see fig. 3) contains several light guide fibers 4, which are fixed with a tough liquid in the cable jacket 6 of stainless steel. One end of the optical cable 2 is guided on one side of the cable drum 20 (flange 22) and into a splicing housing 18. The ends of the light guide fibers of the other end of the cable 2 are preferably laid pre-assembled on optical plugs. The plugs are deposited in a second housing on one of the flanges 22 of the cable drum 20, from where they are brought into contact with the measuring device. Since there is not yet a unit with optical plugs, it is recommended to leave the externally accessible fiber ends with an extra length exposed. The free fiber ends can then be welded to appropriate plugs as required or fed into the equipment used.

The splice housing 18 is shown open in Fig. 2. For example, a splice housing customary for optical cables can be used, which can be closed durably (diffusion-tight) against external influences. The light guide fibers 4 emerging from the end of the protective jacket 6 are placed in the splice housing 18 around supporting elements 19, without falling below the critical radius of curvature. .

In fig. 3 the protective jacket 6 of the cable 2 is flanged to avoid sharp edges towards the outlet side. The light guide fibers 4 are protected against damage in the region of the jacket outlet with a short, sliding protective sleeve 5 of plastic (for example Teflon). protected.

Fig. 4 shows two different circuits of the fiber strand 8. Various measurement partial standards can be realized. In both examples, six optical fiber conductor fiber sub-strands 8 of not unconditionally equal length are used. In Fig. 4a, two-part strands 8 are connected by a durable splice 10. The fiber ends 9 each terminate in an optical plug socket 14. The fiber ends 9 can be closed with optical plugs 14 or remain free for splitting. The plugs 14 can optionally be brought into contact with each other or with the measuring device. Fig. 4b shows a measuring standard that can be manufactured from six partial strands 8 with a maximum length. Fiber partial strands 8 are connected in series via unsustainable splice connections 10a (e.g. plugs) and with at least one durable splice 10.

Claims (6)

Measurement standard for optical attenuation measurements, consisting of an optical cable with at least an artificially aged fiber optic fiber of known attenuation and known length in a tough medium-filled cable sheath, which consists of a material that protects the fiber optic fiber against external influences for a long time, characterized in that the the light guide fiber (4) consists of at least two part strands (8) which are split together and the splitting attenuation is known depending on the wave length. Measuring standard according to claim 1, characterized in that the partial strand ends (9) are accommodated diffusion-tightly and that the housing is diffusion-tightly connected to the cable jacket (6). Measuring standard according to claim 1 or 2, characterized in that the plugs (14) sealing the fiber optic fiber are connected to the cable sheath (6). Measuring standard according to one of the preceding claims, characterized in that several plug sockets (14) are housed in a common housing. Measuring standard according to one of the preceding claims, characterized in that the socket housing is attached to a cable drum (20) receiving the measuring standard (2). Measuring standard according to one of the preceding claims, characterized in that the cable drum (20) is housed in a holder.