JPH0727945A - Production of optical fiber coupler - Google Patents

Abstract

PURPOSE:To provide an optical fiber coupler capable of producing an optical fiber coupler with prescribed characteristic with superior reproducibility. CONSTITUTION:Fnsion drawing is performed as measuring light output from the output terminals of optical fibers f1, f2. When light is inputted from the optical fiber f1, the fusion drawing is stopped when the light output from the output terminal of the optical fiber f1 arrives at a value in which the maximum value P2 of the light output is multiplied by a prescribed ratio r2 larger than 50%. Or, when the light with specific two wavelength are inputted from the output terminals of the optical fibers f1, f2, respectively, the fusion drawing is stopped when the difference of the light output from the output terminals of the optical fibers f1, f2 arrives at a value in which the maximum value P3 of the difference of the light output is multipled by a prescribed ratio r4 larger than 0% and also smaller than 50%.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for manufacturing an optical fiber coupler, and more particularly to a method for manufacturing an optical fiber coupler capable of suitably forming an optical coupling portion of the optical fiber coupler.

[0002]

2. Description of the Related Art Conventionally, in an optical fiber coupler, the coatings of a plurality of optical fibers are partially removed, the coating removal portions are brought into contact with each other, and then the contact portions are fused and stretched to form an optical coupling portion. Manufactured by forming.

Optical coupling between the optical fibers occurs only at the optical coupling portion formed by the above fusion drawing. The characteristics of the optical fiber coupler manufactured in this manner are the same as the shape of the optical coupling part, that is, the diameter of each core and each core, as well as the difference in the refractive index between the core and the clad of the optical fiber forming the optical fiber coupler. It is determined by the mutual gap and the total length of the fusion-stretched optical coupling portion.

Therefore, in order to manufacture an optical fiber coupler having predetermined characteristics, it is necessary to control the shape of the optical coupling portion, that is, the diameter of each core, the gap between the cores, and the total length of the optical coupling portion. is there.

However, it is impossible to actually measure the shape of the optical coupling portion during the process of fusion drawing, and in reality, the optical characteristics of the optical fiber coupler, specifically, the branching ratio and the excess ratio are excessive. Fusion-drawing is performed while measuring loss. For example, in the case of an optical fiber coupler for a 3 dB optical coupler, fusion drawing is terminated at a target characteristic, that is, a branching ratio of 50%.

More specifically, in the fusion drawing process,
Both ends of the optical fiber are fixed and held, and tension is applied to the parts of the optical fiber that are in contact with each other to heat the region that becomes the optical coupling part, and both ends are relatively separated in the same axial direction as the tension. To move. The optical fiber is softened by heating and plastically deformed (stretched) at its plastically fluidized portion. When the heating is stopped and the relative movement of both ends of the stretched part is finished, it is cooled and loses plastic flowability, and plastic deformation (stretching) is stopped.

Although the fusion-bonding and stretching step for forming the above-mentioned optical coupling portion is carried out by a skilled worker, there is a problem in that the quality of products varies and the yield is low.

Recently, in order to manufacture an optical fiber coupler with good reproducibility, automation of a fusion drawing process for forming an optical coupling portion has been attempted. For example, when manufacturing an optical fiber coupler for a 3 dB optical coupler, automation of a fusion drawing process employing a measurement system as shown in FIG. 3A has been attempted. That is, according to this method, at the time of fusion-stretching of the optical fibers f ₁ and f ₂ , one of the optical fibers, in this example, the first optical fiber f ₁ is input to the input side from the light source LD and has a predetermined wavelength, for example, 1 A monitor light of 0.55 μm is incident, and light output detecting means, for example, light receiving elements (photosensors) S ₁ and S ₂ such as photodiodes are connected to the output sides of the optical fibers f ₁ and f ₂ , respectively. Monitor the light output from f ₁ and f ₂ .

With such a construction, the optical output P ₀ monitored by the optical sensor before the start of the fusion drawing is set as an initial value, and the fusion drawing proceeds and the optical output monitored by the optical sensor is (1/2). ). It is determined that the point at which P ₀ is reached is the point at which stretching should end. At that time, the heating is automatically stopped, and at the same time, the relative movement of both ends of the stretched part is also terminated.

[0010]

However, in the above-mentioned conventional method, even when the heating is stopped and at the same time the relative movement of both ends of the stretched portion is completed, the tension applied to the both ends of the stretched portion from the outside is increased. Exists, the plastic deformation (stretching) continues. Then, the optical coupling portion is cooled and loses plastic flowability, and the plastic deformation (stretching) stops when the spring tension due to the elastic deformation of the optical coupling portion and the tension applied from the outside are balanced. For that reason, the total length of the fusion-stretched optical coupling portion becomes longer than the time when it is determined that the stretching should be finished, but the diameter of each core and the gap between the cores are correspondingly small. .

Further, the optical coupling portion is heated and volume-expanded during the fusion drawing process, but is contracted by cooling, and the diameter of each core and the gap between the cores are reduced. As a result, the coupling coefficient changes, and the optical output monitored by the optical sensor after being cooled is smaller than the optical output (1/2) · P ₀ monitored when it is determined that the stretching should be terminated. It becomes a value. That is, there is a problem that the target characteristic, that is, the branching ratio becomes larger than 50%. In particular,
In the biconical type optical fiber coupler in which the center of the optical coupling part has the most narrowly constricted shape, the region in which plastic deformation mainly occurs during the above-mentioned extra stretching is the most constricted part in the center, and the effect is more serious. .

Further, unlike the configuration of the 3 dB optical coupler in which the stretching is stopped at the point where the complete optical coupling (the state where the branching ratio becomes 50%) is first generated by the fusion splicing, higher order perfect optical coupling is performed. A broadband 3 dB optical coupler, which is a so-called wavelength-independent optical fiber, in which a light beam having a specific two wavelengths of, for example, 1.33 μm and 1.55 μm is completely optically coupled by stretching to a point Coupler (Wavelength
In Independent Coupler (WIC), the above-mentioned effects are fatal.

That is, in the case of the WIC type optical fiber coupler, since the fusion splicing is performed up to the point of complete optical coupling of higher order, the diameter of the optical coupling portion is further reduced. Therefore, the effect of changing the optical coupling coefficient is further enhanced by reducing the diameter of each core and the gap between the cores while the optical coupling portion loses plastic fluidity and is cooled. Therefore, there is a problem that the target characteristic, that is, the deviation from the branching ratio of 50% becomes larger.

Therefore, it is an object of the present invention to provide a method of manufacturing an optical fiber coupler capable of manufacturing an optical fiber coupler having predetermined characteristics with good reproducibility.

Another object of the present invention is to automate the fusion splicing / drawing process of an optical fiber to accurately determine the point at which the fusion splicing / drawing should be ended, and to form an optical coupling portion having a predetermined shape. An object of the present invention is to provide a method of manufacturing an optical fiber coupler.

[0016]

The above object can be achieved by the method of manufacturing an optical fiber coupler according to the present invention. In summary, the first aspect of the present invention is, in a method of manufacturing an optical fiber coupler in which a plurality of optical fibers are fused and stretched to form an optical coupling portion, (a) a portion where the optical fibers are in contact with each other. (B) heating the fusion region uniformly while continuously measuring the amount of light propagating through the fusion region of the optical fiber at the output end of each optical fiber. At the same time, both ends of the fusion-bonded region are moved at a predetermined speed in a direction in which they are relatively separated from each other to stretch the fusion-bonded region, (c)
The optical output from the output end of a specific optical fiber selected from the plurality of optical fibers has a maximum value of 50%.
The method for producing an optical fiber coupler is characterized in that the fusion-spreading is stopped when a value obtained by multiplying by a larger predetermined ratio is reached.

According to the second aspect of the present invention, in the method of manufacturing an optical fiber coupler in which a plurality of optical fibers are fused and drawn to form an optical coupling portion, (a) the optical fibers are brought into contact with each other. (B) heating the fusion region uniformly while continuously measuring the amount of light propagating through the fusion region of the optical fiber at the output end of each optical fiber; And, at the same time, moving both ends of the fusion-bonding region in a direction in which they are relatively separated from each other at a predetermined speed to extend the fusion-bonding region, (c) a specific 2 selected from the plurality of optical fibers. When the difference in light output from the output ends of the two optical fibers reaches a value obtained by multiplying the maximum value of the difference in light output by a predetermined ratio that is larger than 0% and smaller than 50%, the fusion drawing is stopped. A method for manufacturing an optical fiber coupler is provided.

The specific optical fiber selected from the plurality of optical fibers forming the optical coupling portion of the optical fiber coupler according to the first aspect of the present invention is an optical fiber that is made into an optical coupling portion by stretching. When measuring the propagation amount of light that passes through the fusion area at the output end of each optical fiber, the light whose maximum propagation amount of light is measured after fusion of each optical fiber before starting stretching Fiber. Normally, when there is one optical fiber that inputs light used to measure the amount of propagation of light, that one is selected. In addition, the specific two optical fibers selected from the plurality of optical fibers forming the optical coupling portion of the optical fiber coupler in the second aspect of the present invention are the optical fibers that are made into the optical coupling portion by stretching. When measuring the propagation amount of light passing through the fusion region at the output end of each optical fiber, the propagation amount of light measured after fusion of the optical fibers and before starting stretching is Secondly, there are two large optical fibers. Usually, when there are two optical fibers that input light used to measure the amount of propagation of light, the two optical fibers are selected.

Further, in the manufacturing method of the present invention, the wavelength of the input light used to measure the amount of light propagating through the fused region of the optical fiber at the output end of each optical fiber should be manufactured. It is appropriately determined according to the appropriate optical characteristics of the optical fiber coupler. For example, when the optical fiber coupler to be manufactured has an optical characteristic in which the characteristic that the branching ratio is 50% with respect to the light of the predetermined wavelength is appropriate, the predetermined wavelength can be selected. . Further, for example, an optical fiber coupler to be manufactured has a branching ratio of 50% for light of two predetermined wavelengths.
When the characteristics that satisfy the following are optical characteristics that are appropriate, it is possible to select the predetermined two wavelengths. Needless to say, in the manufacturing method of the present invention, the optical fibers that are brought into contact with each other so as to be fused by heating are optical fibers designed to obtain appropriate optical characteristics of the optical fiber coupler to be manufactured. It depends on the configuration of the coupler, and the manner of bringing them into contact with each other is also appropriately determined according to the configuration of the designed optical fiber coupler.

In addition, 50 in the first aspect of the present invention
The predetermined ratio larger than% can be appropriately determined according to the stretching conditions used in manufacturing the optical fiber coupler, that is, the stretching speed and the like. On the other hand, instead of the predetermined ratio larger than 50% in the first aspect of the present invention, if the predetermined ratio is selected to be 50% or less, at the time when it is determined that the stretching should be finished using the ratio, The amount of propagation of light already measured at the output end of the optical fiber is 50% or less of its maximum value (that is, the value at which the branching ratio is 100%). As described in "Problems to be solved by the present invention", the obtained optical fiber coupler has an extension amount increased from the time when it is determined that the extension should be terminated, and the propagation amount of the light is measured. The branching ratio for the light of the wavelength greatly exceeds the target of 50%. That is,
The optical fiber coupler to be manufactured has a proper branching ratio of 50% with respect to light having a wavelength for measuring the amount of propagation of the light, but it can be obtained by selecting a ratio of 50% or less. The characteristics of the optical fiber coupler cannot be proper, and are far from the target characteristics.

The predetermined ratio larger than 0% and smaller than 50% in the second embodiment of the present invention can be appropriately determined according to the stretching conditions used during the production of the optical fiber coupler, that is, the stretching speed. It is a thing. On the other hand, instead of the predetermined ratio larger than 0% and smaller than 50% in the second aspect of the present invention, the predetermined ratio is set to 0% or less (0%
Or a negative value), the difference in the amount of propagation of the light measured at the output ends of the two optical fibers at the time when it is determined that the drawing should be ended by using the ratio is It becomes 0% or less of the maximum value. As described as "problems to be solved by the present invention", the obtained optical fiber coupler has an extension amount increased from the time when it is judged that the extension should be ended, and the output of the two optical fibers is increased. The difference in the amount of propagation of light measured at the edge is much smaller than 0% of the maximum value, and the maximum value is multiplied by a negative value. That is,
The optical fiber coupler to be manufactured has a branching ratio (a ratio of the amount of propagation of light) to a light of a wavelength for measuring the amount of propagation of the light.
Is proper, and the characteristic that the difference in the amount of propagation of light measured at the output ends of the two optical fibers is zero is proper. If selected, the characteristics of the obtained optical fiber coupler cannot be proper, and are far from the target characteristics. Alternatively, if the predetermined ratio is selected to be 50% or more, the propagation amount of the light measured at the output ends of the two optical fibers is determined at the time when it is determined that the stretching should be ended using the ratio. The difference has not been reduced to 50% of its maximum.
In the obtained optical fiber coupler, the amount of extension is increased from the time when it is determined that the extension should be terminated, but the difference in the amount of propagation of light measured at the output ends of the two optical fibers is The value is smaller than a value obtained by multiplying the maximum value by a selection ratio of 50% or more, but is still a value not reaching zero. That is,
The optical fiber coupler to be manufactured has a branching ratio (a ratio of the amount of propagation of light) to a light of a wavelength for measuring the amount of propagation of the light.
Is proper, and the characteristic that the difference in the amount of propagation of light measured at the output ends of the two optical fibers is zero is proper. If selected, the characteristics of the obtained optical fiber coupler cannot be proper, and are far from the target characteristics.

In the manufacturing method of the present invention as described above, the both ends of the regions where the fusion-stretched optical fibers are in contact with each other are uniformly heated, so that the fusion of the clad occurs uniformly in the heating region, and the fusion occurs. The cross-sectional shape after wearing becomes uniform. Furthermore, since the heating is performed uniformly in the drawing step as well, the temperature of the heating part during the drawing is maintained to such an extent that the drawing gradually progresses, and the temperature does not rise at a specific place in the heating region. . Therefore, it exhibits uniform plastic flowability, and when stretched by applying tension, the cross-sectional shape of the stretched region can be kept uniform.

In the above-described manufacturing method of the present invention, during the step (a) of heating and fusing the portions where the optical fibers are in contact with each other in the first step, the inside of the heating region is heated by heating. By softening the optical fiber of, the optical fiber is displaced from the state of being extended in the desired uniaxial direction,
In order to prevent the occurrence of bending and the like, tension is appropriately applied in the axial direction in which the optical fiber is stretched. Depending on how to select the heating amount used in this fusion step, tension is appropriately applied in the axial direction of stretching, so that not only fusion but also stretching is inevitably advanced. Further, it goes without saying that both end portions of the heating region may be appropriately moved in a direction in which they are relatively separated from each other in order to apply a desired tension during the fusion bonding process. From this point of view, the step of stretching the fusion-bonded region described in (b) as the second step and the fusion-bonding step which is the first step are separated when they are carried out with time continuity. It becomes a process that cannot be done. However, in the present invention, the two steps described separately from the first step (a) and the second step (b) do not necessarily need to be performed with temporal continuity, and both steps A time discontinuity may occur between them, and the heating amount and the applied tension can be individually selected and implemented. Furthermore, the step (c) of the third step, which is to determine the time point at which the stretching should be ended, and to end the fusion-bonding stretching is
Naturally, it is carried out with time continuity with the second step (b), and it is carried out regardless of the embodiment of the first step (a) and the second step (b). I will get it.

According to the method of the present invention, the cross-sectional shape of the stretched region is kept uniform, and the local constriction having a narrow cross-sectional shape as compared with other portions of the stretched region does not occur. For this reason, it is possible to prevent a vicious circle in which a locally constricted portion, which is conventionally found, is more susceptible to plastic deformation due to heating and further constricted.

If there is a local constriction, a slight difference in the stretching time and a slight difference in the stretching amount are accumulated in the variations in the length and the thinness of the constricted portion. In the constricted portion, the core diameter and the gap between the cores are reduced in proportion to the reduction in the cross-sectional shape. Therefore, the optical coupling coefficient and the loss increase in inverse proportion to the cross-sectional shape. As a result, since the optical coupling coefficient of the constricted portion is large, the variation in the length of the constricted portion is
It gives a large variation in the overall optical power transfer coefficient, mainly in the branching ratio. Further, variations in the narrowness of the constricted portion cause variations in loss.

In the method of the present invention, the generation of the local constriction can be prevented in advance, so that the variations of the optical power transfer coefficient, the branching ratio and the loss are reduced. The unavoidable stretching amount generated during the cooling after the heating stretching (the time required for the finishing operation and the cooling time) can be similarly reduced in variation. At the same time, by eliminating the variation in the cross-sectional shape of the optical coupling portion, it is possible to eliminate the variation in the variation of the optical coupling coefficient due to the heat shrinkage.

As a result of many research experiments, the present inventor obtained a final determination by setting a criterion for the termination of stretching in consideration of the above-mentioned unavoidable stretching amount and the change in optical coupling coefficient due to thermal contraction. It has been found that the drawn amount can be made equal to the optimum drawn amount designed based on the operating principle of the optical fiber coupler. In addition, since there is little variation in the optical power transfer coefficient of the optical coupling portion in the same amount of stretching, the above criteria for determining the termination of stretching can be easily determined by experimentation, and especially when automating the process operation, the above effect becomes unclear. It was found that the variations in the optical power transfer coefficient, the branching ratio and the loss can be significantly reduced.

[0028]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A method for manufacturing an optical fiber coupler according to the present invention will be described below in more detail with reference to the drawings. In each of the embodiments described below, a case of manufacturing an optical fiber coupler having a configuration of two inputs and two outputs will be described, but it should be understood that the present invention is not limited to this.

Example 1 In this example, the optical fiber coupler manufactured is 3 dB.
An optical fiber coupler having an optical coupler configuration, and the optical fibers f ₁ and f _{2 used} have a core diameter of 10 μm and a cladding diameter of 12
It is a 5 μm single mode optical fiber.

FIG. 1 shows the construction of an automated optical fiber coupler manufacturing apparatus for carrying out the manufacturing method of the present invention.
FIG. 2 is a control block diagram of this device.

In this embodiment, the optical fiber coupler manufacturing apparatus comprises a computer (CPU) 100 and an optical fiber f.
₁ , a fusion splicing device 10 capable of holding both ends of f ₂ and applying axial tension to the optical fiber as required, and a fusion splicing device 10 installed so as to be detachable from and attachable to the fusion splicing device 10. Also, for example, a heating device 20 equipped with a hydrogen flame torch 22
Have and. In this embodiment, the fusion splicing device 10 has stretching moving ends 10A and 10B which are fixed ends of the optical fiber, and the stretching moving ends 10A and 10B are moved by the computer 100 via the controller 30. Controlled,
Further, in the heating device 20 including the hydrogen flame torch 22, the torch drive motor 24 is controlled by the computer 100 via the relay 26.

Further, the optical fiber coupler manufacturing apparatus inputs the monitor light to the optical fibers f ₁ and f ₂ held at the extension moving ends 10A and 10B which are the fixed ends of the optical fibers of the fusion extension table 10. Light source LD for fusion, and the fusion drawing stand 10
The optical measuring device 40 includes optical sensors S ₁ and S ₂ for measuring the output light from the optical fibers f ₁ and f ₂ . The signals from the photosensors S ₁ and S ₂ are amplified by the amplifier 44 and transmitted to the computer 100. A recorder 50 is connected to the computer 100, and the control status of each device is output.

That is, a driving device (not shown) for the drawing moving ends 10A and 10B of the fusion drawing device 10 and the heating device 2.
0 operates by an electric signal from the control computer 100. That is, the moving distances, moving speeds, and directions of the stretching moving ends 10A, 10B, and the positions, moving speeds, and directions of the holders that support the hydrogen flame torch 22 are also measured by the sensors respectively provided, and the computer 100 It is taken in and is automatically controlled. Also, hydrogen flame torch 2
The flow rate of the gas supplied to 2 is controlled by the computer 100 so that a predetermined amount of heat can be obtained by the flow rate monitor / control system.
Is set and adjusted by.

Further, the computer 100 also adjusts the light quantity of the light source LD of the optical measuring device 40 and the adjustment of the optical sensors S ₁ and S _2.
This is automatically performed by reading the measurement data from and setting / changing the drive current of the light source LD and the voltage applied to the photosensor.

Next, a method of manufacturing the optical fiber coupler used by the manufacturing apparatus having the above structure will be described.

First, as in the conventional case, the optical fiber f
The coatings of ₁ and f ₂ are partially removed, and the removed coating portions are brought into intimate contact with each other and fixed to the fusion drawing apparatus 10.
The optical fiber f ₁ is provided substantially at the center between the extension moving ends 10A and 10B, which are the fixed ends of the optical fiber of the fusion drawing apparatus 10.
The f ₂ fusion-spreading region is located. The optical fibers f ₁ and f ₂ are moved in opposite directions by the extension moving ends 10A and 10B of the fusion drawing apparatus 10 to apply a predetermined initial tension in the axial direction.

As shown in FIG. 3A, in order to monitor the branching ratio and the excess loss during the fusion drawing process, in this embodiment, the light source LD is connected to the optical fiber f ₁ , and the optical fiber f is connected. Optical sensors S ₁ and S ₂ are connected to the output terminals of ₁ and f ₂ . At this time, the light source LD preferably has a small wavelength distribution, and a semiconductor laser is particularly preferable. Also, the optical sensor S
_It is preferable that ₁ and S ₂ are a combination having substantially the same photosensitivity. At the same time, the light amount of the light source LD is adjusted and the optical sensor S
_The light output from the light output end should be obtained within the range where the linearity of the photosensitivity of ₁ and S ₂ is good. This adjustment is performed based on the calibrated photosensitivity characteristics by separately calibrating the photosensitivities of the photosensors S ₁ and S ₂ in advance. When the adjustment of the optical measuring device is completed, the optical output at the output end of the optical fiber f ₁ is set to the initial value P _i before fusion. FIG. 4 shows the transition state of the optical output at the output end of each optical fiber.

The hydrogen flame torch 22 of the heating device is ignited in a state of being pulled out in the rearward direction apart from the position where the fusion drawing is manually performed in advance.

Next, the fusion process is started. The hydrogen flame torch 22 is automatically moved to a position immediately below the fusion-spreading region of the optical fiber pair f ₁ and f ₂ fixed to the fusion-spreading device. At this position, the vertical position of the hydrogen flame torch holder is fixed. Further, the stretching movement ends 10A and 10B are fused in a state where they are stopped at the initial positions. Flame torch holder, as can also be moved in the axial direction which is the extension of the optical fiber f _1, f _2, mobile axis and the optical fiber f ₁ of a flame torch holder in the lateral direction, elongation of f ₂ It is preset in parallel with the axial direction, but remains fixed at this point.

The computer 100 sets and adjusts the flow rate of the gas supplied to the hydrogen flame torch 22 so that a predetermined amount of heat required for the fusion process can be obtained. In this fusion process, the regions of the optical fibers f ₁ and f ₂ to be fused and stretched are locally heated and gradually fused. The optical output at the output end of the optical fiber f ₁ at the time when fusion is completed is defined as the initial value P ₁ after fusion.

Subsequently, the flow rate of the gas supplied to the hydrogen flame torch 22 is set and adjusted so that a predetermined amount of heat required for the fusion drawing process can be obtained, and the drawing moving ends 10A, 10B.
Is moved at a predetermined speed in a direction in which they are relatively separated from each other, and stretching is started. During the stretching, the computer 100 calculates the center of the fusion-spreading region based on the moving distance of the stretching moving end portion at each time so that the fusion-spreading region is uniformly heated. A movement signal is sent to the drive device of the holding table. During the stretching, the optical sensors S ₁ and S ₂ are arranged at predetermined time intervals.
The optical outputs at the output ends of the optical fibers f ₁ and f ₂ are measured at, and this signal is sent to the control computer 100 together with the relative separation distance (equivalent to the amount of extension) signal of the extension moving ends 10A and 10B. Send and store. Then, the optical output at the output end of the optical fiber f ₂ is a value r ₁ · P obtained by multiplying the initial value P ₁ after fusion by a predetermined ratio r ₁ , for example, r ₁ = 10%.
_The time point exceeding ₁ is determined from the output signal of the optical sensor loaded into the computer 100, and this time point is called the "optical coupling start point".

Optical fiber f ₁ up to the last time of the optical coupling start point
The maximum value P _MAXO of the optical output measured values at the output end of is calculated from the data stored in the computer 100, and is set as the reference value P ₂ before the start of optical coupling. After the optical coupling start point, the process moves to the latter half of the stretching process and, if necessary, the stretching moving end portions 10A, 1
The moving speed of 0B is changed to a predetermined value. Along with this, the gas flow rate supplied to the hydrogen flame torch 22 is also adjusted so that a predetermined amount of heat necessary for the latter half of the drawing step can be obtained. The above-mentioned change of the stretching condition is automatically executed in a predetermined order within a certain time. While the stretching is continued and the optical coupling progresses during this period, by appropriately selecting the time required for changing the stretching conditions and the stretching conditions, the change of the stretching conditions can be completed before reaching the stretching amount for complete optical coupling.

The stretching in the second half step, performed light output measured at closely time intervals, the ratio of light output to the optical coupling before starting reference value P ₂ of a larger predetermined than 50% at the output end of the optical fiber f ₁ r ₂ , For example, the time when it becomes smaller than the value r ₂ · P ₂ multiplied by 55% is detected. This time point is referred to as a "drawing end determination point", and thereafter, the drawing ending step is performed.

That is, immediately after the stretching end determination point is detected, the hydrogen flame torch holder is moved in the vertical direction and the hydrogen flame torch is retracted to the front. At the same time, the movement of the stretching movement end is stopped. Further, the gas flow rate supplied to the hydrogen flame torch 22 is changed to the state before the start of the fusing process. Optical fiber f
The fusion-stretching regions ₁ and f ₂ are cooled, during which the hydrogen flame torch holder is automatically returned to the position before the start of the fusion-bonding process.

When the fused and stretched regions of the optical fibers f ₁ and f ₂ are cooled, the optical outputs at the output ends of both optical fibers f ₁ and f ₂ are finally measured. An in-process evaluation of the branching ratio is performed on the basis of the final measurement values of the light output at the both output ends. At the same time, the reference value P ₂ before the optical coupling is started and both optical fibers f ₁ and f ₂
The excess loss in the stretched portion is estimated from the final measurement value of the optical output at the output end of the.

If the values of the branching ratio and the excess loss obtained by the in-process evaluation are within the predetermined control range, it is judged as a passing product. If either the branching ratio or the excess loss falls outside the control range, re-evaluation will be performed automatically. In the re-evaluation, products that are within the control range are also judged as acceptable products. In addition, other in-process evaluations will be conducted as necessary. Further, the quality is selected based on the result of the in-process evaluation. The pass / fail determination and other predetermined process control data are processed by the computer 100 and stored in an external storage device (medium) to be used for subsequent process control.

After the above series of steps, the next optical fiber is fixed to the fusion drawing apparatus 10 and the fusion drawing is continued.

As described above, according to the present invention, the pre-optical coupling start reference value P _{2 is set} to 5 as a reference for detecting the stretching end determination point.
By using a predetermined ratio r ₂ larger than 0%, for example, a value r ₂ · P ₂ multiplied by 55%, it is inevitable that the stretching end judgment point will occur after the stretching end judgment point during the actual completion of the stretching end step. The optimal stretching amount can be controlled in consideration of the effective stretching and an increase in the optical coupling coefficient due to the volume contraction due to cooling.

Therefore, according to the present invention, as shown in FIG. 4, the unavoidable stretching amount (Δ
L) is added to the actual drawing amount L, and at the stage where the fusion-bonded drawing region of the optical fiber is cooled, an optical coupling portion having a target branching ratio of 50% is formed.

Further, the ratio r ₂ is experimentally tentatively set to r ₂
The value of 50% is set to 50% and the above fusion-drawing process is performed.
Determine by determining the actual amount of stretching and the change in light output at both output ends, and estimating the unavoidable stretching that occurs after the stretching end determination point and the increase in the optical coupling coefficient due to volume contraction due to cooling. You can

In the above-mentioned steps, the reference value P ₂ before the start of optical coupling is multiplied by a predetermined ratio r ₂ larger than 50%, for example, a value r ₂ which is 55%, as a reference for detecting the stretching end judgment point.
Although P ₂ was used, focusing on the difference between the optical outputs at the output ends of the optical fibers f ₁ and f ₂ , the difference in the optical outputs at the above-mentioned “stretching end determination point” is about { r _2- (1
−r ₂ )} · P ₂ = (2r ₂ −1) · P ₂ . Further, when the maximum value of the difference between the optical outputs at the output ends of the optical fibers f ₁ and f ₂ is observed in the range where the amount of extension is smaller than the “optical coupling start point”, the value is P ₂ . As a reference for detecting the drawing end determination point, a predetermined ratio smaller than 0% and smaller than 50% to the maximum value P ₂ '(= P ₂ ) of the difference between the optical outputs at the output ends of the optical fibers f ₁ and f ₂ r ₂ '(= 2r ₂
When the "stretching end determination point" is determined using the value r ₂ '· P ₂ ' multiplied by -1), the determination is made using r ₂ · P ₂ as the reference for detecting the stretching end determination point. The stretching amount is the same as the "end judgment point". That is, the optimum amount of stretching can be controlled by any method of detecting the stretching end determination points.

The optical fiber coupler having the 3 dB optical coupler structure, the manufacturing method of which has been described in the first embodiment, has the maximum intensity in the wavelength distribution of the light source LD used, that is, the wavelength distribution of the optical output intensity. An optical fiber coupler having appropriate optical characteristics has a branching ratio of 50% with respect to light having a wavelength (referred to as "center wavelength").
The optical fiber coupler can be applied to a 3 dB optical coupler for the light of the center wavelength of the light source LD to be used, but it is naturally expected based on its proper optical characteristics, for example, the wavelength dependence of the branching ratio. obtain. It goes without saying that it can be applied to other purposes. For example, an optical fiber coupler generally called a wavelength demultiplexing / multiplexing coupler can demultiplex and combine lights of two specific wavelengths, but its proper optical characteristic, for example, wavelength dependence of branching ratio is seen. And the branching ratio is 50% for light of other specific wavelengths.
It will be. That is, an optical fiber coupler having a 3 dB optical coupler configuration is used for other specific wavelength light with a specific wavelength of 2 dB.
By applying light of one wavelength as an input, it functions as a wavelength demultiplexing / multiplexing coupler.

Example 2 In this example, the optical fiber coupler manufactured is a WIC.
The optical fiber coupler to be used and the optical fiber f to be used
₁ and f ₂ are single mode fibers having a core diameter of 10 μm and a cladding diameter of 125 μm.

Also in this embodiment, the optical fiber coupler manufacturing apparatus shown in FIG. 1 described in Embodiment 1 was used in the same manner.

First, as in the conventional case, the optical fiber f
The coatings of ₁ and f ₂ are partially removed, and then the cladding diameter of at least one of the optical fibers is adjusted to a predetermined outer diameter by means such as hydrofluoric acid etching. After that, they are brought into intimate contact with each other and fixed by applying a predetermined tension to the fusion-stretching apparatus shown in FIG.

In this embodiment, as shown in FIG. 3B, _two light sources LD ₁ and LD ₂ having different wavelengths are connected to the optical input ends of the optical fibers f ₁ and f ₂ , respectively, and the optical output ends are connected. Optical sensors S ₁ and S _{2 of the same type} are connected to the. At this time, it is preferable that the light sources LD ₁ and LD ₂ have a small wavelength spread, and a semiconductor laser is more preferable. In addition, the optical sensor S ₁ ,
S ₂ is the center wavelength of the light sources LD ₁ and LD ₂ , which are 1.32 μm and 1.55 μm, respectively, in the present embodiment, and it is preferable to use a combination having little wavelength dependency of photosensitivity and substantially equal photosensitivity.

The optical measuring device 40 is adjusted by the same method as in the first embodiment. The light outputs at the two output terminals at the end of this adjustment are set to initial values P _i (A) and P _i (B) before fusion, respectively. Usually, P _i (A) and P _i (B) are equal. Figure 5
Shows the transition state of the optical output at the output end of each optical fiber.

The optical fibers are fused by the same process as in the first embodiment. The fusion conditions are selected according to the cladding diameter of the region of the optical fiber to be fused and stretched. The optical outputs at the two output ends at the time when the fusion is completed are set as initial values P ₁ (A) and P ₁ (B) after the fusion, respectively.

Subsequently, the fusion drawing process is started. The procedure of the process is the same as described in Example 1. With stretching,
First, light having a wavelength of 1.55 μm starts optical coupling. Therefore, the optical output P (B) at the output end of the optical fiber f ₂ measured by the optical sensor S ₂ starts to decrease. As the stretching proceeds,
Next, light having a wavelength of 1.32 μm starts optical coupling.

Two wavelengths of 1.32 μm and 1.
When shown to separate the contributions state to light output at the light source LD _1, the individual output terminals of the LD ₂ of 55 .mu.m, is shown in FIG. Actually, the sum of the contributions shown in P (A) and P (B) of FIG. 6 is observed as the optical output at each output terminal, and
The transition shown in is obtained. However, during the fusion-drawing step,
The initial value P ₁ after fusion is adjusted by adjusting the optical measuring device 40.
Although (A) and P ₁ (B) are almost equal, the difference is exaggerated in FIG. 5 so that the two can be easily identified.

By appropriately selecting the cladding diameter of the optical fiber, as shown in FIG. 6, the second extension amount (L ₃ ) at which the light with a wavelength of 1.55 μm has a branching ratio of 50% is obtained.
And the first stretching amount (L ₃ ) at which the light having a wavelength of 1.32 μm has a branching ratio of 50% can be matched.

In FIG. 5, when the light having a wavelength of 1.55 μm starts the optical coupling and the light having a wavelength of 1.32 μm starts the optical coupling, for example, the stretching amount L ₁ , the optical output P is obtained.
(B) is only contribution of light having a wavelength of approximately 1.55 μm. Therefore, the light output P (B) is a value r ₃ · P ₁ (B) obtained by multiplying the initial value P ₁ (B) after fusion by a predetermined ratio r ₃ , for example, 90%.
The time point at which the light output decreases can be identified from the light output data captured by the computer 100. This time point is called an "optical coupling start point". Of the optical output measured values at each output end up to the previous time of the optical coupling start point, the maximum value P of each
_MAXO (A) and P _MAXO (B) are obtained from the stored data and are set as reference values before optical coupling start P ₂ (A) and P ₂ (B).

Thereafter, the second half of the drawing process is carried out, and if necessary, the drawing conditions are changed in the same order as the process described in Example 1. In the latter half of the drawing process, the optical fiber f ₁ ,
The optical output at the output end of f ₂ is measured at close time intervals, and the computer 100 calculates the difference (P (A) −P (B)) between the optical outputs at both output ends.

As shown in FIG. 5, a stretching amount exceeding the stretching amount at which light having a wavelength of 1.55 μm starts optical coupling, for example, L ₂
In, the difference (P (A) -P (B)) shows a maximum. The maximum value of this difference (P (A) -P (B)) is P ₃ . As described above, when the cladding diameter of the fusion-spreading region of the optical fiber to be used is appropriately selected, as shown in FIG. 6, when the stretching amount L ₃ exceeds the stretching amount L ₂ , a wavelength of 1 .32
A branching ratio of 50% is obtained with both light of μm and 1.55 μm. As a result, the difference (P (A) -P (B)) becomes zero.

The stretching amount exceeds L ₂ and the difference (P (A)-
P (B)) is, the P ₃ to greater than 0% and less than 50% a predetermined ratio r _4, detecting when for example be less than the value r ₄ · P ₃ multiplied by r ₄ = 15%. This time point is called a "stretching end determination point". After that, the process proceeds to the stretching ending step. The steps of terminating the heating and drawing are performed in the same order as the steps described in Example 1. At the drawing end judgment point, the drawing amount is smaller than the drawing amount L ₃ by ΔL, but as described in Example 1, after the drawing end judgment point, the unavoidable occurrence occurs while the optical fiber is actually cooled. The final stretching amount can be the stretching amount L ₃ .

The above-mentioned ratio r ₄ is obtained by carrying out the above fusion-stretching step with the value of r ₄ being 0%, and changing the actual stretching amount and the change in the optical output at both output ends shown in FIG. It can be determined by experimentally estimating the increase in the optical coupling coefficient due to the unavoidable stretching that occurs after the stretching end determination point and the volume contraction due to cooling.

Further, in the present Example 2, considering that the stretching amount L ₃ is basically the same as the first stretching amount in which the branching ratio is 50% with the light having the wavelength of 1.32 μm, the above-mentioned Example 2 1
Even if the ratio r ₂ is selected by estimating the increase in the optical coupling coefficient due to the unavoidable stretching and the volume contraction due to cooling that occur after the stretching end determination point in the same manner,
The error may be within an acceptable range.

Specifically, the stretching amount L ₃ in the second embodiment is in principle the same as the first stretching amount in which the branching ratio is 50% with the light having the wavelength of 1.32 μm. In No. 1, when the optimum stretching amount is the first stretching amount at which the branching ratio is 50% with the light having the wavelength of 1.55 μm, the light having the wavelength of 1.55 μm is input as the reference for detecting the stretching end determination point. In accordance with the procedure for determining a value r ₂ · P ₂ obtained by multiplying the reference value P ₂ before the start of optical coupling by a predetermined ratio r ₂ greater than 50%, by inputting light of wavelength 1.32 μm, The value corresponding to the reference value P ₂ before the start of optical coupling in Example 1 is determined, and the “value corresponding to the value r ₂ · P _{2 serving} as the reference for detecting the stretching end determination point in Example 1” is determined. You can That is, the "value corresponding to the predetermined ratio r ₂ larger than 50% in the first embodiment" can be set. Using a "value corresponding to the value r ₂ · P ₂ used as a reference for detection of the stretching end determination point in the Example 1" as defined in this method to detect the "stretching end decision point", it said at its drawing amount The difference (P (A) −P (B)) is calculated.
The calculated difference (P (A) -P (B)) is the above P ₃
A predetermined ratio r ₄ greater than 0% and less than 50%
In principle, it agrees with the value r ₄ · P ₃ multiplied by. Therefore, the calculated by difference of (P (A) -P (B )) by dividing the P _3, it is possible to determine the ratio r _4.

It is to be noted that only the light having the wavelength of 1.32 μm is inputted and the above-mentioned “value corresponding to the value r ₂ · P ₂ which is the reference for detecting the stretching end judgment point in the above-mentioned Embodiment 1” is judged. When the optical output at the output end of the optical fiber becomes smaller than "a value corresponding to the value r ₂ · P _{2 used} as the reference for detecting the drawing end determination point in the first embodiment", which is used as the reference for detecting the point. Needless to say, it is possible to control the amount of stretching to be the optimum even by detecting. Further, in view of the operation principle of the optical fiber coupler, the optimum extension amount is such that when only light of a specific wavelength is input, the optical output at the output end of the optical fiber becomes a branching ratio of 50%. It is needless to say that the amount of stretching can be controlled to an optimum amount by the method according to the method of Example 1 in the case where the amount of stretching basically matches.

The optical fiber coupler the manufacturing method of which is described in the second embodiment is a so-called wavelength-independent optical fiber coupler. Specifically, the wavelengths are 1.32 μm and 1.55.
It is a proper optical characteristic that the branching ratio is 50% for light of two wavelengths of μm, and is appropriate at the first stretching amount where the branching ratio is 50% for light of wavelength 1.32 μm. The structure is designed to obtain various optical characteristics. Depending on its designed configuration, "optical coupling starting point"
Although the maximum value of the difference (P (A) -P (B)) is determined in the stretching amount exceeding 1.0, when the method of the second aspect of the present invention is applied to the optical fiber coupler described in Example 1. Is a difference (P (A) −P) in the stretching amount before reaching the “optical coupling start point”.
The maximum value of (B)) will be determined. At this time, the difference (P (A) -P
The maximum value P ₂ 'of (B)) may not match P ₂ ,
There may be a plurality of maximum values of the difference (P (A) -P (B)). However, in the method of the second aspect of the present invention, the maximum value of the difference (P (A) -P (B)) is appropriately determined according to the situation of its implementation, and is described in, for example, Example 1. When applied to an optical fiber coupler, the difference (P (A) -P) existing before reaching the "optical coupling start point"
By setting the maximum value of the maximum values of (B)) to P ₂ , it is possible to match P _{2 ′,} and the maximum value of the difference (P (A) −P (B)) can be determined by the procedure. It goes without saying that the stipulation is reasonable and appropriate. Further, in the detection of the "stretching end determination point", it is needless to say that the "stretching end determination point" is determined only when the stretching amount is reasonable and appropriate according to the implementation situation. That is, in view of the optimal stretch amount that is naturally expected from the design of the designed optical fiber coupler, in the stretch amount that cannot be an appropriate "stretch end determination point", the "stretch end determination point" detection criterion Even if the above condition is satisfied, the procedure for detecting the “stretching end determination point” is not determined in advance so as not to be determined as the “stretching end determination point”, for example, to prevent erroneous determination due to inevitable fluctuations in the intensity of the optical input. Of course.

According to the present invention, since the optical fiber coupler is manufactured by the manufacturing method as shown in the first and second embodiments, the variation of the drawing end point which occurs in each manufacturing is reduced, and therefore the yield is improved. Is significantly improved. In addition, the criteria for determining the extension end point can be uniquely determined so that the optimum designed extension amount is finally obtained based on the operating principle of the optical fiber coupler. Even when the manufacturing conditions are changed due to the change, the judgment standard can be easily optimized.

Further, in general, when the target characteristic of the optical fiber coupler and the operating principle used to achieve the characteristic are determined,
Appropriately select the configuration of the optical measurement device for monitoring during the fusion drawing process, the type of data to be automatically measured during the process such as the optical output at the output end, the sequence of the fusion drawing process, and the conditions. be able to. Further, based on the operation principle of the optical fiber coupler, it is possible to design an optimum amount of extension that has a predetermined target characteristic. At the same time, the end point of a series of steps, particularly the above-mentioned extension end determination point, is detected automatically by calculation processing by the computer 100 in accordance with a separately determined extension end determination standard using measurement data such as light output at the output end. Can be done on a regular basis.
Optimum design based on the operating principle of the optical fiber coupler, which is based on the operating principle of the optical fiber coupler, which experimentally estimates an increase in the optical coupling coefficient caused by volume contraction due to unavoidable stretching and cooling that occurs after the stretching end determination point. It can be uniquely determined so that a desired stretching amount can be finally obtained.

[0073]

As described above, the method of manufacturing an optical fiber coupler according to the present invention takes into consideration the increase of the optical coupling coefficient due to the volume contraction due to the unavoidable extension and the cooling which occur after the extension end judgment point in advance. Then, in order to set the judgment criteria of the end of the drawing, the finally obtained drawing amount can be made equal to the optimum drawing amount designed based on the operating principle of the optical fiber coupler, and the optical fiber having a predetermined characteristic. The coupler can be manufactured with good reproducibility.

Further, according to the present invention, the fusion drawing process of the optical fiber is automated to accurately determine the point at which the fusion drawing should be ended, and the optical fiber having the optical coupling portion having a predetermined shape. Couplers can be manufactured.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram showing an embodiment of a manufacturing apparatus for carrying out a method for manufacturing an optical fiber coupler according to the present invention.

FIG. 2 is a schematic control block diagram of the manufacturing apparatus in FIG.

FIG. 3 is a diagram showing two types of measurement systems of the optical measurement device.

FIG. 4 is a diagram showing a transition state of an optical output at an output end of each optical fiber when manufacturing a 3 dB optical coupler type optical fiber coupler.

FIG. 5 is a diagram showing a transition state of an optical output at an output end of each optical fiber when a WIC type optical fiber coupler is manufactured.

FIG. 6 is a diagram showing the state of transition of the optical output at the output end of each optical fiber when manufacturing the WIC type optical fiber coupler, separated for each wavelength.

[Explanation of symbols]

 10 Fusion Stretching Device 20 Heating Device 22 Hydrogen Flame Torch 40 Optical Measuring Device 100 Computer

Claims (2)

[Claims] 1. A method of manufacturing an optical fiber coupler, wherein a plurality of optical fibers are fused and drawn to form an optical coupling part,
(A) Heating and fusing the portions where the optical fibers are in contact with each other, and (b) Continuously measuring the amount of light propagation through the fused region of the optical fibers at the output end of each optical fiber. While uniformly heating the fusion-bonding region and simultaneously moving both ends of the fusion-bonding region in a direction in which they are relatively separated at a predetermined speed, stretching the fusion-bonding region, (c) Stop the fusion drawing when the optical output from the output end of a specific optical fiber selected from a plurality of optical fibers reaches a value obtained by multiplying the maximum value of the optical output by a predetermined ratio larger than 50%. And a method for manufacturing an optical fiber coupler. 2. A method of manufacturing an optical fiber coupler, wherein a plurality of optical fibers are fused and stretched to form an optical coupling portion,
(A) Heating and fusing the portions where the optical fibers are in contact with each other, and (b) Continuously measuring the amount of light propagation through the fused region of the optical fibers at the output end of each optical fiber. While uniformly heating the fusion-bonding region and simultaneously moving both ends of the fusion-bonding region in a direction in which they are relatively separated at a predetermined speed, stretching the fusion-bonding region, (c) The difference between the optical outputs from the output ends of two specific optical fibers selected from the plurality of optical fibers is 0% at the maximum value of the difference between the optical outputs.
A method for manufacturing an optical fiber coupler, characterized in that fusion splicing is stopped when a value obtained by multiplying a predetermined ratio that is larger and smaller than 50% is stopped.