US20020034364A1 - Methods and devices for reducing splice loss in an optical transmission line - Google Patents

Abstract

Techniques and devices are described for reducing splice loss in an optical transmission line. According to one technique, an electric arc is generated from an arc current, the arc current having a level and duration sufficient to produce an electric arc with an intensity and duration sufficient to achieve a desired splicing temperature at a splice point between a first optical fiber and a second optical fiber positioned within the electric arc. The electric arc is used to splice together the first and second optical fibers. After the fibers have been spliced together, the level of the arc current is ramped downward over time, thereby creating a downward ramp in temperature at the splice point from the splicing temperature to a cooler temperature, the downward ramp in temperature being shaped to reduce splice loss. The techniques and devices described herein are suitable for use with various splice combinations.

Description

• This application is a continuation-in-part of U.S. patent application Ser. No. 09/667,031, filed Sep. 21, 2000, and assigned to the assignee of the present application, the disclosure and drawings of which are hereby incorporated by reference in their entirety.[0001]
BACKGROUND OF THE INVENTION
• 1. Field of the Invention[0002]
• The present invention relates generally to improvements to optical fiber transmission lines, and more particularly to advantageous aspects of methods and devices for reducing splice loss in an optical transmission line.[0003]
• 2. Description of the Prior Art[0004]
• As optical data transmission lines increase in length and in the amount of data that they carry, there is increasing interest in the development of new types of optical fiber and in techniques that can be used to refurbish already existing transmission lines. One important parameter of an optical transmission line is the amount of signal dispersion resulting from the optical characteristics of the materials used to construct the line. A new class of fibers has recently been developed known as dispersion-compensating fiber (DCF), which has a steeply sloped, negative dispersion characteristic.[0005]
• One application for DCF fiber is to upgrade already existing fiber optic communication links. These already existing links are typically constructed using standard single-mode fibers (SMF) having dispersion characteristics that are optimized for operation at a signal wavelength of 1310 nm. However, certain applications require optimization of a communication link for operation at a longer wavelength, particularly where the communication link spans great distances. For example, one wavelength-division multiplexing (WDM) technique currently in use requires optimization of the link for operation at a wavelength of 1550 nm.[0006]
• It is possible to refurbish an already existing SMF fiber transmission line optimized for operation at a given wavelength, such as 1310 nm, by splicing a length of DCF fiber into the transmission line. The length of the DCF fiber added to the SMF fiber transmission line is precisely calculated to produce an adjustment in the overall dispersion characteristics of the line such that it is now optimized for operation at a different desired wavelength, such as 1550 nm. A suitable technique for precisely calculating a length of DCF fiber to be spliced into an already existing line in order to optimize the line for operation at a different wavelength is disclosed in U.S. patent application Ser. No. 09/596,454, filed on Jun. 19, 2000, assigned to the assignee of the present application, the drawings and disclosure of which are hereby incorporated by reference in their entirety.[0007]
• In addition to dispersion, a second important parameter for DCF fiber is the fiber's loss value, that is, the amount of excess signal loss resulting from the introduction of the DCF fiber into a transmission link. Optimally, a DCF fiber should provide a highly negative dispersion, while only introducing a small excess loss to the fiber link. A useful index of the performance of a DCF fiber is the so-called “figure of merit” (FOM), which is defined as the dispersion of the fiber divided by the attenuation.[0008]
• Another important issue arising in connection with DCF fiber is the excess loss that results when DCF fiber is spliced to a standard single-mode fiber (SMF). To obtain a highly negative dispersion, DCF fiber uses a small core with a high refractive index, having a mode-field diameter of approximately 5.0 μm at 1550 nm, compared with the approximately 10.5 μm mode-field diameter of SMF fiber at 1550 nm. The difference in core diameters results in significant signal loss when a fusion splicing technique is used to connect DCF fiber to SMF fiber. It is possible to reduce the amount of signal loss by choosing splicing parameters that allow the core of the DCF fiber to diffuse, thereby causing the mode-field diameter of the DCF core to taper outwards, resulting in a funneling effect. However, the amount and duration of the heat required to produce the funneling effect result in an undesirable diffusion of dopant in the ring of refractive material surrounding the DCF fiber core. This diffusion of ring dopant limits the amount of splice loss reduction that can be obtained using a mode-field expansion technique. For example, using DCF fiber with a FOM of 200 ps/nm/dB, the splice loss typically cannot be reduced below 0.7-0.8 dB when splicing directly to SMF fiber.[0009]
• There is thus a need for improved techniques for splicing DCF fiber to SMF fiber that reduces splice loss below current limits.[0010]
SUMMARY OF THE INVENTION
• The above-described issues and others are addressed by the present invention, aspects of which provide methods and devices for reducing splice loss in an optical transmission line. According to one aspect of the invention, an electric arc is generated from an arc current, the arc current having a level and duration sufficient to produce an electric arc with an intensity and duration sufficient to achieve a desired splicing temperature at a splice point between a first optical fiber and a second optical fiber positioned within the electric arc. The electric arc is used to splice together the first and second optical fibers. After the fibers have been spliced together, the level of the arc current is ramped downward over time, thereby creating a downward ramp in temperature at the splice point from the splicing temperature to a cooler temperature, the downward ramp in temperature being shaped to reduce splice loss.[0011]
• Additional features and advantages of the present invention will become apparent by reference to the following detailed description and accompanying drawings.[0012]
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 shows a transverse cross section, not drawn to scale, of a length of dispersion-compensating fiber (DCF).[0013]
• FIG. 2 shows a refractive index profile of the DCF fiber shown in FIG. 1.[0014]
• FIG. 3 shows an axial cross section, not drawn to scale, of a transmission line fabricated from a length of DCF fiber that has been spliced to a length of single-mode fiber (SMF).[0015]
• FIG. 4 shows a diagram, not drawn to scale, of an optical fiber transmission line according to a first aspect of the present invention, fabricated from a length of DCF fiber, a length of a bridge fiber (BF), and a length of SMF fiber that have been spliced together.[0016]
• FIG. 5 shows a transverse cross section, not drawn to scale, of a BF fiber suitable for use in the transmission line shown in FIG. 4.[0017]
• FIG. 6 shows a refractive index profile of the BF fiber shown in FIG. 5.[0018]
• FIG. 7 shows an axial cross section, not drawn to scale, of the optical fiber transmission line shown in FIG. 4.[0019]
• FIG. 8 shows a bar graph illustrating the splice loss distribution for the splice between the DCF fiber and the BF fiber in the transmission line illustrated in FIG. 4.[0020]
• FIG. 9 shows a bar graph illustrating the splice loss distribution for the splice between the BF fiber and the SMF fiber in the transmission line illustrated in FIG. 4.[0021]
• FIG. 10 shows a diagram of a suitable furnace arrangement for heating the splice between the DCF fiber and the BF fiber according to a further aspect of the present invention.[0022]
• FIG. 11 shows a table setting forth the setpoints used in heating the splice between the DCF fiber and the BF fiber.[0023]
• FIG. 12 shows a graph illustrating a technique according to a further aspect of the present invention for determining a maximum temperature to which the splice between the DCF fiber and the BF fiber is to be heated to obtain an optimal reduction in splice loss.[0024]
• FIG. 13 shows a table setting forth a comparison of the splice loss at the splice between the DCF and BF fiber before and after heating the splice in accordance with the present invention.[0025]
• FIG. 14A shows an elevation view of a length of DCF fiber and a length of BF fiber that have been spliced together and mounted into a frame.[0026]
• FIG. 14B shows a plan view of the fibers and frame shown in FIG. 14A.[0027]
• FIG. 15 shows a diagram of a system in which a laser is used to heat the splice between the lengths of DCF and BF fibers shown in FIGS. 14A and 14B.[0028]
• FIG. 16 shows a diagram of a fusion splicer.[0029]
• FIG. 17 shows a flowchart of a method for reducing splice loss according to an aspect of the invention.[0030]
• FIG. 18 shows a flowchart of a method for reducing splice loss according to another aspect of the invention.[0031]
DETAILED DESCRIPTION
• Aspects of the present invention provide an optical transmission line with reduced splice loss and methods for fabricating an optical transmission line with reduced splice loss. According to one aspect of the invention, a length of bridge fiber (BF) is introduced between a length of DCF fiber (or other suitable first transmission fiber) and a length of SMF fiber or other suitable second transmission fiber, such as, for example, True-Wave fiber or Large Effective Area fiber. As described in further detail below, the BF fiber is fabricated such that, with the use of suitable first and second sets of splicing parameters, the BF fiber can be spliced to both the DCF fiber and SMF fiber with significantly reduced loss at each splice, thereby resulting in a combined fiber line that exhibits lower splice loss than a line fabricated from DCF and SMF fibers without a BF fiber. Splice loss at the splice between the dispersion-compensating fiber and the bridge fiber is then further reduced by heating the splice to a predetermined temperature and cooling the splice using a predetermined temperature ramp.[0032]
• FIG. 1 shows a transverse cross section, not drawn to scale, of a length of[0033]DCF fiber10. As shown in FIG. 1, the design of theDCF fiber10 is based on asmall diameter core12, typically approximately 5 μm, compared with a core diameter of approximately 10 μm for SMF fiber. TheDCF fiber core12 has been doped with a high level of germanium oxide (GeO₂) to have a high refractive index. Surrounding the core is aring14, which has been doped with a high concentration of fluorine (F) to have a low refractive index. Thecore12 andring14 are encased in a layer ofovercladding16.
• FIG. 2 shows a[0034]refractive index profile20 for theDCF fiber10 shown in FIG. 1.
• As shown in FIG. 2, the[0035]refractive index profile20 includes acentral spike22, representing the high refractive index of theDCF fiber core12. On either side of thespike22 aresharp dips24, representing the low refractive index of thering14. Finally, on either side of thesharp dips24 areflat regions26, representing the refractive index of theovercladding16. TheDCF fiber10 whose structure and refractive index profile are illustrated in FIGS. 1 and 2 typically exhibits a dispersion at 1550 nm of approximately −100 ps/km/nm, with a loss of approximately 0.5 dB/km.
• Theory predicts that the splicing of this[0036]particular DCF fiber10 to a length of typical SMF fiber will result in a splice loss of approximately 2.2 dB. This loss results from the mismatch of the mode-field distribution in the splice region. However, this splice loss can be reduced by using a fusion splicing technique in which the splicing parameters are chosen to allow a tapered mode-field expansion of the core of the DCF fiber, producing a “funneling” effect that reduces the mismatch between the DCF fiber core and the SMF fiber core.
• Specifically, when DCF fiber is fusion spliced, splicing parameters may be chosen such that the amount and duration of the heat generated by the splicing process will cause the GeO[0037]₂in the core12 to diffuse, altering the refractive index profile of the fiber and the mode-field distribution in the splice region. The amount of diffusion can be controlled by optimizing the splice parameters. Thus, using optimized splice parameters, theDCF fiber core12 can be tapered outward in the splice region to better match the SMF. This is illustrated in FIG. 3, not drawn to scale, which shows an axial cross section of atransmission line30 fabricated from a length of theDCF fiber10 shown in FIG. 1 spliced to a length ofSMF fiber32. As shown in FIG. 3, theDCF fiber core12 tapers outwards as it approaches thesplice point34, such that it approximates the diameter of thecore36 of theSMF fiber32.
• Using this technique, it is possible to lower the splice loss from the theoretically predicted value of 2.2 dB to approximately 0.7-0.8 dB. It is believed, however, that the amount of splice loss reduction using a direct splicing technique is limited by the high mobility of fluorine during the splicing process. In particular, fluorine begins to diffuse at a temperature much lower than the highest temperatures reached during fusion splicing. Because of the relatively high concentration of fluorine dopant in the[0038]ring14 surrounding the core12 in theDCF fiber10, thering14 will diffuse at a faster rate than the core12, which in turn tends to increase splice loss. FIG. 3, not drawn to scale, schematically illustrates the relatively greater dispersion of thering14 relative to thecore12.
• One technique that is currently used to reduce splice loss is to introduce a bridge fiber (BF) between the DCF and SMF fibers. FIG. 4 shows a diagram of a[0039]transmission line40 incorporating this technique. The transmission line includes a length of DCF fiber50 (or other suitable first transmission fiber), a length ofBF fiber60, and a length of SMF fiber70 (or other suitable second transmission fiber). As mentioned above, other suitable second transmission fibers include, for example, True-Wave fiber or Large Effective Area fiber. A first end of theBF fiber60 is spliced to theDCF fiber50 at afirst splicing point80, and a second end of theBF fiber60 is spliced to theSMF fiber70 at asecond splicing point82. As discussed below, it has been found that the introduction of theBF fiber60 between theDCF fiber50 and theSMF fiber70 may be used to reduce the splice loss to as little as 0.4 dB, which is significantly lower than the 0.7-0.8 dB splice loss obtainable without the use of aBF fiber60.
• FIG. 5 shows a cross section of the[0040]BF fiber60 taken through the plane5-5, and FIG. 6 shows arefractive index profile90 for theBF fiber60. Thecore62 of the BF fiber is similar to theDCF fiber core12 illustrated above in FIGS. 1 and 2. It is doped with GeO₂at the same concentration as theDCF fiber core12, and has substantially the same diameter, approximately 5 μm. However, thering64 surround theBF fiber core62 differs from thering14 surrounding theDCF fiber core12. First, theBF ring64 has a larger diameter than theDCF ring14. Second, although bothrings14 and64 are doped with fluorine (F), the concentration of fluorine in theBF ring64 is lower than the concentration of fluorine in theDCF ring14. Theovercladding66 of theBF fiber60 is similar to theovercladding16 of theDCF fiber10.
• Thus, the BF[0041]refractive index profile90 shown in FIG. 6 has a different shape than the DCFrefractive index profile20 shown in FIG. 2. Because theBF core62 is similar to theDCF core12 in diameter and dopant concentration, thecentral peak92 of the BF refractive index profile is similar to thecentral peak22 of the DCFrefractive index profile20. However, because of the larger diameter and lower dopant concentration of theBF ring64, compared with the diameter and dopant concentration of theDCF ring14, thedips94 on either side of thecentral peak92 in the BFrefractive index profile90 are wider and shallower than thedips24 on either side of thecentral peak22 in the DCFrefractive index profile20. Because theBF overcladding66 is similar to theDCF overcladding16, the flatouter regions96 of the BFrefractive index profile90 are similar to the flatouter regions26 of the DCFrefractive index profile20.
• The diameter and dopant concentration of the[0042]BF ring64 are chosen such that overall splice loss can be reduced by choosing a suitable first set of splicing parameters for thesplice80 between the DCF andBF fibers50 and60 and a suitable second set of splicing parameters for thesplice82 between the BF andSMF fibers60 and70. FIG. 7 shows an axial cross section, not drawn to scale, of thetransmission line40 shown in FIG. 4. As shown in FIG. 7, because theDCF core52 and theBF core62 have similar diameters, no mode-field expansion is required. Thus, it is possible to select splicing parameters for thefirst splice80 that minimize or eliminate diffusion of the fluorine dopant in theDCF ring54. As further illustrated in FIG. 7, because of the relatively low concentration of fluorine dopant in theBF ring64, there is less diffusion of theBF ring64 when theBF fiber60 is spliced to theSMF fiber70. It is therefore possible to select splicing parameters for thesecond splice82 that allow for the full mode-field expansion required to match theBF core62 to theSMF core72 without the splice loss associated with excessive diffusion of the fluorine dopant.
• It should be noted that the particular BF fiber design described herein is one of a number of different designs for BF fiber that have proven to be suitable for use with the present invention. For example, it is possible to use a BF fiber having a core similar to the[0043]core62 of the above-described BF fiber, but without the fluorine-dopedring portion64. It will be appreciated that alternative BF fiber designs may be used without departing from the spirit of the present invention. Also, as mentioned above, the present invention can be used with other splice combinations, including, for example, combinations using True-Wave fiber or Large Effective Area fiber in place of theSMF fiber70.
• By optimizing the splicing parameters, a first end of the[0044]BF fiber60 can be spliced toDCF fiber50 with an average loss as low as 0.17 dB, and a second end of theBF fiber60 can be spliced to anSMF fiber70 with an average loss as low as 0.23 dB. Thus, using theBF fiber70 as a bridge fiber between theDCF fiber50 and theSMF fiber60, it is possible to reduce the total splicing loss to approximately 0.4 dB. FIG. 8 is agraph100 showing the splice loss distribution for the splice between theDCF fiber50 and theBF fiber60. The average splice loss is 0.169 dB and the standard deviation is 0.031 dB. FIG. 9 is agraph102 showing the splice loss distribution for the splice between theBF fiber60 and theSMF fiber70. The average splice loss is 0.228 dB and the standard deviation is 0.016 dB.
• As mentioned above, the parameters used to splice the[0045]DCF fiber50 to theBF fiber60 are different from the parameters used to splice theBF fiber60 to theSMF fiber70. Fluorine dopant begins to diffuse at a lower splice temperature than GeO₂, and also diffuses more rapidly than GeO₂. Thus, the splicing of theDCF fiber50 to theBF fiber60 must be done using a low fusion current and a short fusion time. These parameters allow thisfirst splice80 to be accomplished with minimal, if any, mode-field expansion and minimal, if any, fluorine diffusion, thus minimizing splice loss. It is also important that the electrodes of the fusion splicer be as clean as possible, as small variations in the splice conditions due to dirty electrodes can result in a significant increase in the splice loss.
• The splicing of the[0046]BF fiber60 to theSMF fiber70 is not as critical with respect to the cleanliness of the electrodes. For thissecond splice82, a higher fusion current and a longer fusion time are used to allow the GeO₂in theBF core62 to diffuse, proximate to thesplice82, such that good tapering is provided from theBF core62 to theSMF core72. As shown in FIG. 9, the splice loss has a narrow distribution with a standard deviation of only 0.016 dB.
• In accordance with a further aspect of the invention, it is possible to reduce the splice loss between the[0047]DCF fiber50 and theBF fiber60 even further. In one approach, two fibers are spliced together and removed from the splicer. As described below, the splice is then heated to a predetermined maximum temperature for a predetermined period of time, and then cooled back down to room temperature following a predetermined temperature ramp. In another approach, also described below, the controlled ramping down of the splice temperature takes place as part of the splicing operation.
• FIG. 10 shows a diagram of a[0048]tube furnace110 that may be suitably employed where a splice between two fibers is heated and cooled in an operation separate from the initial creation of the splice. Thefurnace110 includes aceramic tube112 surrounding thesplice point80 between theDCF fiber50 and theBF fiber60. Aheating wire114 is coiled around thetube112, and apower supply116 causes a heating current to flow through theheating wire114. A suitable material for theheating tube112 is Degussit ceramic, manufactured by Friatec AG (Germany). Suitable dimensions for theheating tube112 include an inner circumference of 2 mm, an outer circumference of 3 mm, and a length of 10 mm. A suitable material for theheating wire114 is 90/10 Pt/Rh, and a suitable wire diameter is 0.5 mm. The length of bare fiber is approximately 2 cm.
• After the[0049]DCF fiber50 has been spliced to theBF fiber60, the two fibers are mounted into thefurnace110 with thesplice point80 located inside theheating tube112. Thesplice point80 is heated to approximately 1100° C. and kept at this temperature for approximately 30 seconds. Heating of thesplice80 causes a measurable decrease in splice loss. The amount of the decrement depends upon the particular design of DCF andBF fibers50 and60. It is possible for this splice loss to be maintained at room temperature by ramping down the temperature over a time period of approximately 90 seconds. Because of the small diameter of the DCF andBF fibers50 and60 and theheating tube112, the relatively low specific heat of the materials used to fabricate thefibers50 and60 and theheating tube112, and the relatively large surface area of thefibers50 and60 and theheating tube112, it is possible to implement this heating ramp simply by decreasing the amount of current flowing through theheating wire114, without the need for an outside cooling mechanism. However, if needed, it would be within the spirit of the invention to add a cooling mechanism, such as a fan, to facilitate the cooling process. Best results are obtained if splice loss is monitored during the heat treatment. The loss reduction effect has been observed in all DCF designs manufactured at Lucent Technologies, including Standard DCF, WideBand DCF, Inverse Dispersion Fiber, and High-Slope DCF.
• Returning to FIG. 4, one suitable length that has been used for the[0050]BF fiber70 is approximately 3 meters. Thefirst splice90, between theDCF fiber50 and theBF fiber70 is made on a fusion splicer using a splice time, or fusion time, of approximately 0.2 sec. A splice time significantly longer than 0.2 sec may induce fluorine diffusion in the DCF, which in turn will increase splice loss. Using an Ericsson splicer, the following splicing parameters has been used to produce satisfactory results:
• Pre-fusion Time=0.2 sec[0051]
• Pre-fusion Current=10.0 mA[0052]
• Gap=50.0 microns[0053]
• Overlap=5.0 microns[0054]
• [0055]Fusion Time1=0.3 sec
• [0056]Fusion Current1=10.5 mA
• [0057]Fusion Time2=0.2 sec
• [0058]Fusion Current2=17.5 mA;
• Fusion Time[0059]3=0 sec
• Fusion Current[0060]3=0 mA.
• FIG. 11 shows a table[0061]120 illustrating a heating current profile used as setpoints to the power supply to obtain the results set forth below. It is also possible to obtain good results by manually adjusting the temperature of theheating tube112, such as by manually adjusting the current flowing through theheating wire114, while splice loss is monitored during the heat treatment. After the completion of the heating current profile shown in FIG. 11, the furnace is allowed to cool to ambient temperature for approximately 1 min. Thesplice80 and the DCF andBF fibers50 and60 are then removed from the furnace. The distal end of theBF fiber60 is now spliced to theSMF fiber70, using a second set of parameters, as described above.
• An important parameter is the maximum temperature for the heat treatment illustrated in FIG. 10. This parameter may be determined empirically. A[0062]test splice80 between aDCF fiber50 and aBF fiber60 is mounted into thefurnace110 and splice loss is monitored during heating. Several heating trials are now made, in which the maximum heating current is varied (i.e., in which the maximum temperature of the splice is varied) and in which the cooling ramp is kept constant. The ramp can be linear, or may also be non-linear, such as the ramp set forth in the table120 shown in FIG. 11. If the maximum current is increased for each new trial, thesame splice80 can be used for the experiment. FIG. 12 shows agraph122 illustrating the trend of splice loss, after cooling, as a function of maximum heating current. From thisgraph122, it will be seen that there is an optimal value of maximum current at which the desired splice loss is obtained. After determining the maximum current, the ramp can now be optimized by monitoring splice loss during cooling.
• FIG. 13 shows a table[0063]124 containing data for splice loss reduction by applying the above-described heat treatment to a WideBand DCF sample manufactured at Lucent Technologies Denmark A/S. The table124 compares the amount of splice loss at a signal wavelength of 1550 nm before and after heat treatment. At higher wavelengths, the loss reduction is even greater. For example, at a wavelength of approximately 1600 nm, splice loss is reduced several dBs for some DCF designs.
• According to a further aspect of the invention, the above-described heat treatment of the splice between the DCF and BF fibers is accomplished using a laser. The use of a laser, as described below, allows the heat treatment to be performed without making physical contact with the splice point. This is useful, for example, in the construction of a high-strength optical transmission line. Another advantage of using a laser is that a very compact heating zone can be obtained. Thus, when the heating zone is provided by a laser, a smaller length of bare fiber (for example, 1 cm) may easily be used.[0064]
• FIGS. 14A and 14B show, respectively, elevation and plan views of a length of[0065]DCF fiber130 and a length ofBF fiber132 that have been spliced together at asplice point134 and mounted into aframe136. FIG. 15 shows a diagram of alaser heating system140, in which the frame-mounted, spliced DCF andBF fibers130 and132 shown in FIGS. 14A and 14B are positioned proximate to a C0₂laser142, or other laser of suitable power, such that thelaser beam144 emanating from thelaser142 provides the required heating of thesplice point134. The temperature of thesplice point134 is regulated by adjusting the power of thelaser beam144. According to a further aspect of the invention, the splicedfibers130 and132 and theframe136 are housed in achamber146 that is filled with a protectant gas, such as nitrogen. Alternatively, the splice point may be purged with a stream of protectant gas during the heat treatment. In this case, the heat treatment may be conducted in a normal ambient atmospheric environment.
• It should be noted that the graphical technique illustrated in FIG. 12, described above, for empirically determining an optimal maximum temperature for heating the splice between the DCF and BF fibers can also be used in connection with a laser-heated system, such as the[0066]system140 illustrated in FIG. 15. However, instead of graphing splice loss as a function of maximum heating current, splice loss is instead graphed as a function of maximum laser beam intensity. In all other significant respects, the technique is the same.
• According to a further aspect of the invention, splice loss is reduced as part of the splicing operation. Specifically, a fusion splicer is used to splice together a first and a second optical fiber. After the fibers have been spliced together, the temperature of the splice point is ramped downward from the splicing temperature to a cooler temperature by ramping down the arc current used to drive the fusion splicer's electric arc.[0067]
• FIG. 16 shows a diagram of the main components of a[0068]fusion splicer150 according to the present invention. Thefusion splicer150 includes achassis152, upon which is mounted a pair of optical fiber routing guides154 and156. The optical fiber routing guides154 and156 are used to hold a pair ofoptical fibers158 and160 that are to be spliced together at a splice point162.
• The splice point[0069]162 is positioned to lie in the path of an electric arc164 that is generated between twoelectrodes166 and168. A variablecurrent source170 is used to provide power to theelectrodes166 and168. The intensity of the electric arc164 and the amount of heat that is delivered to the splice point162 is a function of two parameters: the level of the current delivered by the current source168 to the electric arc164, and the amount of time that the splice point162 is exposed to the electric arc164. Suitable parameters are chosen based upon the particular fibers being spliced together.
• As shown in FIG. 16, the[0070]fusion splicer150 is provided with acontroller unit172 connected to thecurrent source170 that allows the level and duration of current delivered by thecurrent source170 to thearc electrodes166 and168 to be controlled automatically. Thecontroller unit172 may suitably be driven by a microprocessor or other suitable programmable device. It should be noted, however, that once suitable time and current parameters have been determined, it would also be possible for the arccurrent controller172 to have fixed settings. It should further be noted that the present aspect of the invention can also be practiced using currently available fusion splicers, by manually adjusting the arc current to achieve the desired ramping.
• According to the prior art, once the two[0071]fibers158 and160 have become fused at the splice point162, the electric arc164 is typically cut off abruptly, resulting in a precipitous drop in temperature. However, according to the present aspect of the invention, after the twofibers158 and160 have become fused at the splice point162, the current from the variablecurrent source170 is instead ramped downward in a controlled manner, thereby producing a controlled ramping down of the temperature at the splice point162 from the splicing temperature down to a cooler temperature. By ramping the temperature of the splice downward in a controlled manner, it is possible to achieve a reduction in splice loss similar to that achieved by using the other techniques described herein.
• The parameters used to construct a suitable arc current ramp are determined empirically, using techniques similar to those described above in conjunction with FIGS. 11 and 12. It should be noted that, because of the intense heat generated by a fusion splicer, it may be difficult to create a controlled, gradual ramping of the temperature of the splice from the splicing temperature all the way down to room temperature. However, even ramping at higher temperatures will result in a reduction of splice loss.[0072]
• It will be appreciated that, according to this aspect of the invention, the ramped cooling of the splice is now a continuous part of the splicing process. Thus, this aspect of the invention may be more suitable for certain applications, such as high-strength applications, as no additional treatment is required after splicing. This aspect of the invention may suitably be used, for example, with the new Inverse Dispersion Fibers (IDF) recently introduced by Lucent Technologies.[0073]
• FIGS. 17 and 18 are flowcharts of a[0074]first method180 and asecond method200 for reducing splice loss in an optical transmission line according to aspects of the present invention. Thesemethods180 and200 may suitably be used in conjunction with theoptical transmission line40 shown in FIG. 4, and discussed above. However, thesemethods180 and200 may also suitably be employed with other combinations of optical fibers. For example, these methods may be applied in splicing together two lengths of Inverse Dispersion Fiber (IDF×IDF).
• In the[0075]first method180, two optical fibers are spliced together instep182. Instep184, the spliced fibers are removed from the splicer. Instep186, the splice is then heated to a predetermined maximum temperature. As described above, the splice may suitably be heated using a furnace, such as that shown in FIG. 10, or a laser, such as that shown in FIG. 15. Instep188, the splice is then cooled using a predetermined ramp. The shape of the ramp may suitably be determined using the empirical techniques described above.
• In the[0076]second method200, two fibers are spliced together instep202. As described above, the fibers may suitably be spliced together using a fusion splicer. In step204, the arc current of the fusion splicer is then ramped downward to achieve a controlled ramp of the temperature of the splice down from the splicing temperature to a cooler temperature. As described above, this ramping of the temperature of the splice results in a reduction of splice loss. Instep206, the spliced fibers are removed from the splicer.
• In an optical transmission line with multiple splices, it would be possible to perform the[0077]first method180 or thesecond method200 on more than one splice. Further, it would also be possible to perform thefirst method180 on a first splice in the optical transmission line and thesecond method200 on a second splice in the optical transmission line. Where thesecond method200 is performed on multiple splices in an optical transmission line, the arc current is adjusted between splices in preparation for the next splice. As mentioned above, depending upon the particular fibers being spliced together, the splicing parameters may vary from splice to splice.
• While the foregoing description includes details which will enable those skilled in the art to practice the invention, it should be recognized that the description is illustrative in nature and that many modifications and variations thereof will be apparent to those skilled in the art having the benefit of these teachings. It is accordingly intended that the invention herein be defined solely by the claims appended hereto and that the claims be interpreted as broadly as permitted by the prior art.[0078]

Claims (20)

I claim:

1. A method for reducing splice loss in an optical transmission line, comprising:

(a) generating an electric arc from an arc current, the arc current having a level and duration sufficient to produce an electric arc with an intensity and duration sufficient to achieve a desired splicing temperature at a splice point between a first optical fiber and a second optical fiber positioned within the electric arc;

(b) using the electric arc to splice together the first and second optical fibers; and

(c) ramping the level of the arc current downward over time, thereby creating a downward ramp in temperature at the splice point from the splicing temperature to a cooler temperature, the downward ramp in temperature being shaped to reduce splice loss.

2. The method ofclaim 1, wherein in step (c), the ramping of the arc current is performed automatically by an arc current controller.

3. The method ofclaim 2, wherein the arc current controller is programmable, and wherein the method further includes the following step performed before step (a):

programming the arc current controller to create a downward ramp in temperature at the splice point from the splicing temperature to a cooler temperature after the first and second fibers have been spliced together, the downward ramp in temperature being shaped to reduce splice loss.

4. The method ofclaim 1, wherein the first optical fiber is dispersion-compensating fiber.

5. The method ofclaim 4, wherein the second optical fiber is a bridge fiber.

6. The method ofclaim 1, wherein the first optical fiber is inverse dispersion fiber.

7. The method ofclaim 6, wherein each of the first and second optical fibers is inverse dispersion fiber.

8. A method for reducing splice loss in an optical transmission line, comprising:

(a) generating an electric arc from an arc current, the arc current having a level and duration sufficient to produce an electric arc with an intensity and duration sufficient to achieve a desired first splicing temperature at a first splice point between a first optical fiber and a bridge fiber positioned within the electric arc;

(b) using the electric arc to splice together the first optical fiber and the bridge fiber; and

(c) ramping the level of the arc current downward over time, thereby creating a downward ramp in temperature at the first splice point from the first splicing temperature to a cooler temperature, the downward ramp in temperature being shaped to reduce splice loss;

(d) removing the first optical fiber and the bridge fiber from the electric arc;

(e) adjusting the arc current to produce an electric arc with an intensity and duration sufficient to achieve a desired second splicing temperature at a second splice point between the bridge fiber and a second optical fiber positioned within the electric arc;

(f) using the electric arc to splice together the bridge fiber and the second optical fiber; and

(g) ramping the level of the arc current downward over time, thereby creating a downward ramp in temperature at the second splice point from the second splicing temperature to a cooler temperature, the downward ramp in temperature being shaped to reduce splice loss.

9. The method ofclaim 8, wherein the first optical fiber is dispersion-compensating fiber.

10. The method ofclaim 8, wherein the first optical fiber is inverse dispersion fiber.

11. The method ofclaim 10, wherein each of the first and second optical fibers is inverse dispersion fiber.

12. A splicer, comprising:

a chassis;

a pair of arc electrodes mounted to the chassis for generating an electric arc;

a variable current source connected to the pair of arc electrodes, the variable current source providing as an output a current for driving the pair of arc electrodes and creating an electric arc of sufficient intensity and duration to achieve a desired splicing temperature at a splice point between first and second optical fibers positioned within the electric arc;

first and second fiber routing guides mounted to the chassis for holding first and second fibers to be spliced together;

a controller connected to the variable current source for automatically creating a downward ramp of the arc current after the first fiber has been spliced to the second fiber, thereby creating a downward ramp in temperature at the splice point from the splicing temperature to a cooler temperature, the downward ramp in temperature being shaped to reduce splice loss.

13. The splicer ofclaim 12, wherein the controller is programmable.

14. An optical transmission line comprising:

a first optical fiber spliced to a second optical fiber at a splice point, the first optical fiber being spliced to the second optical fiber using an electric arc generated from an arc current having a level and duration sufficient to produce an electric arc with an intensity and duration sufficient to achieve a desired splicing temperature at the splice point, the level of the arc current being ramped downward over time after splicing, thereby creating a downward ramp in temperature at the splice point from the desired splicing temperature to a cooler temperature, the downward ramp in temperature being shaped to reduce splice loss.

15. The optical transmission line ofclaim 14, wherein the first optical fiber is dispersion-compensating fiber.

16. The optical transmission line ofclaim 15, wherein the second optical fiber is a bridge fiber.

17. The optical transmission line ofclaim 14, wherein the first optical fiber is inverse dispersion fiber.

18. The optical transmission line ofclaim 17, wherein each of the first and second optical fibers is inverse dispersion fiber.

19. An optical transmission line comprising:

a first optical fiber spliced to a first end of a bridge fiber at a first splice point and a second optical fiber spliced to a second end of the bridge fiber at a second splice point,

the first optical fiber being spliced to the bridge fiber using an electric arc generated from an arc current having a level and duration sufficient to produce an electric arc with an intensity and duration sufficient to achieve a desired first splicing temperature at the first splice point, the level of the arc current being ramped downward over time after the first optical fiber is spliced to the bridge fiber, thereby creating a downward ramp in temperature at the first splice point from the first splicing temperature to a cooler temperature, the downward ramp in temperature being shaped to reduce splice loss,

the second optical fiber being spliced to the bridge fiber using an electric arc generated from an arc current having a level and duration sufficient to produce an electric arc with an intensity and duration sufficient to achieve a desired second splicing temperature at the second splice point, the level of the arc current being ramped downward over time after the second optical fiber is spliced to the bridge fiber, thereby creating a downward ramp in temperature at the second splice point from the second splicing temperature to a cooler temperature, the downward ramp in temperature being shaped to result in a reduction in splice loss.

20. The optical transmission line ofclaim 19, wherein the first optical fiber is dispersion-compensating fiber.

Images