The invention relates to methods and devices for determining an optimal fusion current to be used in splicing single fibers to each other using an electric arc generated between electrodes.
Background
In fusion splicing optical fibers using an electric arc generated between electrodes one of the most important parameters to be selected in the best possible way is the electrode current or fusion current passing between the electrodes in the arc. The electrode current must be determined correctly in order to obtain a low loss and high strength of the splice and an accurate estimation of the optical loss in the splice, see the article W. Zheng, "Real time control of arc fusion for optical fiber splicing", IEEE J. of Lightwave Technology, Vol. 11, No. 4, pp. 548-553, April 1993. The fibers obtained from different vendors, fibers manufactured for different purposes such as single-mode or multi-mode fibers, fibers processed using different treatments, such as fibers coated with titanium or carbon, etc., often need different fusion currents. Even for the same type of fiber, the fusion current has to be increased up to 70% when the place where the fusion welding is made varies from 100 m to 5000 m above sea level owing to the difference in the ambient air pressure. Moreover, since after a number of splices the particles evaporated from the silica fibers will gradually be plated on the surface of the electrodes used for the fusion welding, the current should often be readjusted for the changing conditions of the welding electrodes. To find out a suitable fusion current, many different methods have been developed, such as the well-known melt-back method, see G. Kiss, "High yield fusion splicing in the outside plant: using fiber meltback to monitor electrode condition," National Fiber Operation Engineering Conference, Denver, U.S.A., Sept. 1996, methods using barometric sensors, the offset-splicing method, see the Swedish patent No. 9501589-7, "Auto-fusion temperature control with an offset-splicing method for optical fiber splicing", etc.
Summary
It is an object of the invention to provide a method and device for determining an optimal fusion current to be used in welding single fibers to each other, the method and device having a simple structure and being capable of an easy implementation in commercially available splicing devices.
A determination of an optimal fusion current to be made to pass between electrodes for welding the ends of two optical fibers to each other is based on the assumption that the light intensity emitted from the fibers is dependent on the current used in the heating the fibers in an electric arc between the welding electrodes and that an optimal splice between the fibers is obtained when the temperature of the fiber ends during the welding process is equal to a fixed value, which is independent of various ambient conditions such as the height above sea level. The intensity of light emitted from the fibers is measured when the fibers are heated by means of currents between the electrodes which are significantly lower than the range, where the optimal fusion current can be expected to be located. From the relation of the measured light intensity values and the different heating currents a functional relationship is obtained, which is used for extrapolating to find a value of the current intensity which will produce a higher light intensity value equal to the desired one during the actual welding. An exponential type relation can be used requiring only two parameters to be measured for each splice and thus two different currents for heating the fiber in the arc is sufficient for determining the optimal fusion. Such a determination is easily implemented in automatic splicers having advanced image processing and logic and arithmetic facilities.
Using the method described above the fiber ends will not be destroyed as in the melt-back method, the result will be more reliable than that using barometric sensors and the adaption to the actual splicing situation is much easier than in the offset method.
Additional objects and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objects and advantages of the invention may be realized and obtained by means of the methods, processes, instrumentalities and combinations particularly pointed out in the appended claims.
Brief Description of the Drawings
A complete understanding of the invention and of the above and other features thereof may be gained from a consideration of the following detailed description of non-limiting embodiments presented hereinbelow with reference to the accompanying drawings, in which:
FIG. 1a is schematic picture of the setup in a optical fiber splicer for heating an optical fiber using an electrical discharge and for capturing a light intensity profile of the heated fiber using an optical system,
FIG. 1b is a diagram illustrating light intensity profiles of a single mode fiber heated with different fusion currents from 9 to 18 mA,
FIG. 2 is a diagram illustrating the fusion current as a function of the luminescence intensity from the fiber ends,
FIG. 3 is a diagram illustrating the light intensity as a function of the fusion current for different fiber types both as measured and calculated,
FIG. 4 is a diagram illustrating calculated fusion current values as a function of the luminescence intensity from the fiber ends,
FIG. 5 is a flow chart of a procedure for automatic fusion current selection, and
FIG. 6 is a schematic picture illustrating the electric control of an optical fiber splicer.
Detailed Description
1. Basic Concept
In FIG. 1a the basic setup in an automatic optical fiber splicer is shown. The optical fiber 1 has its end region located between the points of electrodes 3, between which an electrical glow discharge is generated for heating the fiber end, the intensity of the electrical discharge being controlled by the intensity of the electrical current between the electrodes 3. An optical system symbolized by a lens 5 depicts the fiber end region on the light sensitive area 7 of a camera, e.g. a plate carrying CCD elements. In most automatic fiber fusion splicers of this type digital imaging processing systems, compare FIG. 6, are provided for processing the electric signals from the light sensitive area 7 and thereby for monitoring the fibers used and the splicing procedure by controlling fiber positioning devices, see FIG. 6. The imaging system of such an automatic optical fiber splicer is actually one of the best means for also monitoring the fiber temperature and for controlling the fusion current, as will be apparent from the following description.
When an optical fiber is heated, the thermal radiation emitted from the fiber can be observed using a video-camera (symbolized by 7 in FIG. 1a) and analyzed using a digital image processing system of a splicer, as e.g. is described in the paper W. Zheng, O. Hulten and R. Rylander, "Erbium-doped fiber splicing and splice loss estimation", IEEE J. of Lightwave Technology, Vol. 12, No. 3, pp. 430-435, March 1994, which is incorporated by reference herein. During splicing, the radiant emissivity, consequently also the luminescence intensity from the heated fiber, is a function of the temperature at the fiber ends, see the paper T. Katagiri et al., "Direct core observation method using thermal radiation of silica fibers with dopants", Elec. and Comm. in Japan, Vol. 71, No. 11, pp. -85, 1988, and also FIG. 1b, in which a diagram illustrates the light intensity along a line perpendicular to the fiber axis for different fusion currents. Thus, the fiber temperature is a function of the fusion current. The relation between the fusion current and the light intensity can be measured in several different ways, such as peak value detecting, averaging over a part of or over the whole fiber, etc.
In FIG. 2 the measured average light intensity at the fiber central region taken as having a width about half the fiber diameter is shown as a function of the fusion current, e.g. here also taken along a line perpendicular to the fiber axis, for seven different fiber types. The unit of the abscissa axis is in grey scale units obtained from the camera system. The light intensity values are computed by averaging readings of a perpendicular line having a length of 60 .mu.m at the central part of fiber (from -30 .mu.m to +30 .mu.m, compare FIG. 1b). The abbreviations/designations in the figure mean: SM: single mode fiber; DS: dispersion shifted fiber; Titanium: titanium coated fiber; MM: multi-mode fiber. From the figure, we can clearly observe the Automatic Gain Control (AGC) function of the camera which is activated when the average intensity (not the maximum intensity) of the captured image is higher than 150 grey scales (GS). Normally, if the maximum intensity in one picture is higher than about 200 grey scale units for a 16-bit grey-scale resolution having 255 different grey scale values or units, the AGC function of such a digital camera will be triggered on, and the average intensity in the captured image will be decreased. In this way, such a 16-bit camera can increase its dynamic range.
In the region before the AGC is triggered on, the relation between the average light intensity, I, of the hot fiber and the fusion current, i, can be approximated by:
where the amplitude A is related to characteristics of the imaging system and the decline rate .alpha. and the initial current i.sub.0 might be fiber-type related, respectively. The coefficients can be found by using curve fitting of Eq. (1) to the measured data. It is then found that the parameters .alpha. and A are fiber type independent for all seven fiber types. In the testing system used, the decline rate .alpha.=0.03 and the amplitude A=18 are consistently valid. The initial current parameter i.sub.0 is the only parameter which is dependent on the fiber type. The parameter values obtained from curve fitting of Eq. (1) to measured values are given in Table 1 for the seven fiber types.
A comparison between curves (1) calculated from the curve fitted parameters and the actual measured values of four fiber types is shown in the diagram of FIG. 3 where the light intensities are plotted as functions of the fusion currents used, the different marks denoting measured values and the graphs showing calculated curves. Good agreement can be observed before the AGC is active (i.e. in the case where the light intensity is lower than 160 grey scale units in the system used). The calculated relations between the fusion current and the average light intensity of the seven fiber types of Table 1 are illustrated in the diagram of FIG. 4. Optimal fusion currents are achieved when the light intensity reaches a threshold shown by the vertical dotted line for all the fiber types used.
For the calculated relations as illustrated in FIG. 4, we can observe that the optimal fusion currents (about 16 mA, 15.5 mA, 13.5 mA, and 16.5 mA for OVD, MCVD, multimode fiber, and silica core fiber types, respectively) are obtained for almost the same light intensity I.sub.th of about 300 grey scale units. This phenomenon indicates that a preset light intensity threshold can be employed to determine the optimal current for all fiber types at any altitude. However, the threshold might not be physically feasible because of the limited dynamic range of the camera.
Thus, for every splice, the current can start from a rather low value (say 10 mA) and then increase step by step. Meanwhile, the light intensity of the fiber end can be measured in real time. Whenever the intensity reaches the preset threshold, the current will be locked until the end of splicing.
However, in practice the fusion time is often not allowed to be so long as is required for waiting that such a fusion current ramp using a plurality of measurements is run. And also very often, when the current has arisen to a high level and part of the fiber is rather bright, the automatic gain control (AGC) will be triggered on and the relation of Eq. (1) will be ruined. Thus, a method which uses only a few steps of fusion current at a lower fiber brightness to determine the correct final fusion current is required and will be described hereinafter.
2. Fusion Current Calculation
From Table 1 we observe that the decline rate .alpha. and the amplitude A are constants for all fiber types. If the values of these two parameters are known in advance, there is only one unknown parameter, the start current i.sub.0. Then it would be sufficient to make only one measurement of the light intensity for a suitably chosen current and then the light intensity function of Eq. (1) would be completely known and the optimal fusion current could be determined by finding the current i that according to the calculated relation (1) gives the light intensity value I.sub.th. Making at least three measurements for different values of the fusion current would make a determination of all parameters .alpha., A and i.sub.0 possible.
However, the amplitude A is obviously directly related to the set-up of the optical system, e.g. of the brightness adjustment thereof and could easily vary if some adjustment was made between different measurements or splices. Therefor the constantness of the amplitude cannot always be ensured. However, the amplitude A can be taken not to vary for the procedure comprising first measuring light intensity for some current values and then setting the fusion current accordingly for making a splice. Thus, there will always be necessary to determine two parameters, the amplitude and start current, the value of the decline rate being considered to be previously known, and thus at least two measurements must be made.
Thus two measurements are made for two different currents i.sub.1 and i.sub.2, where i.sub.1 <i.sub.2. The two corresponding intensities I.sub.1 and I.sub.2 are then, as given by Eq. (1):
and
From (2) over (3), we obtain ##EQU1##
And from (2), we have
Thus, for a given theoretic intensity threshold I.sub.th, which might not be physically feasible because of the AGC or the dynamic range of the CCD camera used, we should be able to obtain a corresponding current i from the knowledge of i.sub.1, i.sub.2, I.sub.1, and I.sub.2. ##EQU2##
Inserting (4) and (5) in (6), we have ##EQU3##
If we have i.sub.1 +1=i.sub.2, i.e. that the higher current intensity I.sub.2 is 1 mA larger than the lower current intensity I.sub.1, then Eq. (7) will degenerate to ##EQU4##
Using Eq. (8), we can calculate a desired fusion current i from a preset intensity threshold I.sub.th by measuring the light intensity I.sub.1 at the current intensity i.sub.1, and the light intensity I.sub.2 for the current i.sub.2 =i.sub.1 +1, where i.sub.1 and i.sub.2 are chosen to be both located in a suitable working range. Of course other values than the step of 1 mA can between the currents can be used and in that case the general Eq. (7) is used.
3. Intelligent Fusion Current Selection
Using Eq. (8) or possibly Eg. (7) an intelligent fusion current selection can easily be implemented in a conventional, automatic fiber fusion splicer. The method has been tested in such a fusion splicer and very stable and encouraging results were obtained even for different test (start) currents as listed in Table 2. Seven different fiber types, the same as in Table 1, were used for a number of splices. For each fiber type, an optimized fusion current was found with a standard deviation less than 0.5 mA for different start currents.
Table 2. Fusion currents automatically selected by a conventional splicer for seven different fiber types at different starting currents i.sub.t. The letters "NA" mean that the measured is not applicable for the cases, i.e. they are used in the cases where either I.sub.1 <I.sub.min or I.sub.2 >I.sub.max. The parameters used to obtain the results are: I.sub.th =300 grey scale units, I.sub.min =15 grey scale units and I.sub.max =170 grey scale units.
Using Eqs. (8) and (9), an intelligent fusion current selection can easily be implemented in a fiber fusion splicer of automatic type. A flow diagram of the procedure is shown in FIG. 5.
The procedure starts in a block 501, in which the ordinary fusion procedure is initiated. In this step the necessary rough alignment of the fiber end is made and in the next block 503 it is awaited, until the fiber end surfaces have been moved close to each other. Then in the block 505 the test current i.sub.t corresponding to i.sub.1 above is chosen in some suitable way, for example to a rather high value with some reasonable interval. If there is some value already stored, it can be used. Then in a block 507 an image is taken and is analyzed to find the intensity I.sub.1, for example as suggested above it can taken to be the average light intensity along a line perpendicular to a fibre end, the line segment being located centrally with respect to the fibre longitudinal axis and symmetrically thereto, e.g. encompassing half the fiber width. In a next block 509 it is determined whether the measured intensity I.sub.1 is smaller than a minimum light intensity value I.sub.min. The minimum value I.sub.min is chosen to be the smallest value for which an acceptable measurement can be made and is thus chosen to be well located above the noise level of the measuring system. If the decision of the block 509 is that the measured and analyzed light intensity is good for the measurement purpose, i.e. not smaller than the minimum light intensity I.sub.min, a block 511 is executed. There the measured and analyzed light intensity I.sub.1 is compared to a maximum light intensity value I.sub.max. The maximum light intensity value I.sub.max is selected, that for light intensity values equal to or larger than it probably the AGC function of the camera would be activated and distort the measurements. If it is found in the block 511 that the measured light intensity I.sub.1 is not larger than the maximum light intensity value I.sub.max a block 513 is executed and there the test current value i.sub.t corresponding to i.sub.2 to be used for the next measurement is determined and set to the previously used test current intensity value i.sub.1 increased by a fixed, predetermined current intensity increment of e.g. 1 mA, as suggested above, i.e. i.sub.t =i.sub.t +1 (mA). Then a new image is taken in a block 515 for the test current i.sub.t (=i.sub.2) and a second measured and analyzed intensity I.sub.2 is determined from the captured picture. In a block 517 it is then first determined whether the second measured intensity value I.sub.2 is larger than the maximum light intensity value I.sub.max. In the block 517 also the start current i.sub.0 is calculated according to Eq. (9) using the previously determined, stored value of the decline rate .alpha. and it is determined whether the calculated start current value i.sub.0 is larger than the first or previously used test current value i.sub.1. If the two determinations in this block 517 give the answer no, the test currents used (i.sub.1 and i.sub.2) are suitable for determining an optimal fusion current. Thus the fusion current i is calculated according to Eq. (8) in a block 519 using a stored value of the decline rate .alpha.. The first used test current value i.sub.1 is then stored in the processor memory in the block 521 for a possible future use for another splice. Finally, in a block 523 the determined fusion current i is used in the splicer for making the actual splice between the two fiber ends.
If it was determined in the block 509, that the first measured intensity i.sub.1 was smaller than the minimum light intensity value I.sub.min a block 525 is executed. A new first test current value is here obtained by incrementing the first used test current value by some suitable increment, e.g. 1 mA, in order to obtain a greater light intensity in the captured image. Then in a block 527 the new first test current value i.sub.t (=i.sub.1) is checked to have a value within a valid interval, e.g. as suggested between the values of 5 mA and 30 mA. If the new first test current value is located within this range the block 505 is again executed for the new first current value. If the new current value is outside said range, obviously a valid test current cannot be selected as is announced or displayed in a suitable way in a block 529 and then the procedure is ended in a block 531.
If it was decided in the block 511 that the first measured and analyzed light intensity value I.sub.1 was larger than the maximum light intensity value I.sub.max, above which the AGC can distort the measurements, a block 533 is executed. Obviously the light intensity is too high and the test fusion current must be reduced and this is made by decrementing the first used test current by a suitable increment, e.g. as suggested comprising 3 mA, i.e. a new test current i.sub.t is obtained from i.sub.t =i.sub.t -3 (mA). After the block 533 the new test first current value is tested whether it is within the valid range and the procedure thus proceeds to the block 527, as described above, in which a decision is made and after which the procedure continues in the way as already described.
If it was determined in the block 517 that either one of the conditions in this decision block was fulfilled, i.e. that the second measured light intensity value I.sub.2 was larger than the maximum light intensity I.sub.max or the start current i.sub.0 is larger than the second test current value i.sub.2, the latter condition signifying that the measurements were made on a wrong portion of the exponential curve of Eq. (1), obviously the used test currents have been too high and therefore must be reduced. Therefore the block 533 comprising a reduction of the second test current by a predetermined suitable amount is executed in the same manner as has been described above. After the block 533 the procedure continues to the block 527 as above.
The method described above is advantageously performed by an automatic image processing and positioning device used for splicing optical fibers, the components of which are schematically illustrated in FIG. 6.
In the schematic picture of FIG. 6 a fiber splicing device of the automatic type is shown having retainers 9, in which the fiber ends are placed and are retained during the positioning and the splicing. The retainers 9 are movable in three orthogonal coordinate directions both in parallel to the longitudinal direction of the fibers and in two directions perpendicular to this direction which then also are perpendicular to the illuminating directions of light emitted from the light sources 11. The retainers 9 are operated along suitable mechanical guides, not shown, by control motors 13. Electric conducting lines to the electrodes 7, the motors 13 and the lamps 11 are provided from an electronic circuit module 15 and in particular from driver circuits 17, 19 and 21 respectively arranged therein. From the TV camera 7 an electric line is arranged to a video interface 23 in the electronic circuit module 15, from which a suitable image signal is delivered to an image processing and image analysis unit 25 producing among other things pictures which can be displayed on a monitor 26. The various procedural steps for welding the fibers 1 to each other are controlled by logic circuits 27, e.g. in the shape of a suitable microprocessor. The logic circuits 27 comprises submodules for executing the procedural steps as described above for determining an optimal fusion current and for splicing the fibers 1 using this current. It thus comprises a general control submodule 29 for controlling the displacement of the fiber 1 ends in relation to each other by an activation of the motors 13 in suitable displacement directions, provides a signal to the image processing and image analysis unit 25 for starting an analysis of an obtained picture and a determination of the light intensity. Further, the control circuit 29 controls the activation of the light sources 11 by sending suitable signals to the driver circuit 21 thereof and the application of a high electric voltage to the electrodes 43 and the time period during the electric voltage is to be delivered and then setting the electric current between the electrodes to the various required values, by providing appropriate signals to the driver circuit 17. Further, a memory 31 is used for storing previously determined parameters, such as the decline rate .alpha., and values for previously used test currents. Calculation circuits 33 perform the calculation of the required parameters, such as calculation the function values of Eqs. (8) and (9), and can then use parameters previously stored in the memory 31. Finally, determination circuits 35 are provided for determining the current intensity which gives the threshold light intensity value I.sub.th.
Additional advantages and modifications will readily occur to those skilled in the art. Therefore, the invention in its broader aspects is not limited to the specific details, representative devices and illustrated examples shown and described herein. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents.