JP2007248732A - Optical power monitor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an optical power monitor which can be made compact even with multiple channels and which has small optical loss. <P>SOLUTION: In the optical power monitor, the center axes of two optical fibers are subjected to offset fusion, light is leaked from a core into a clad, third or fourth order intensified light of the leaked light is reflected through a notched face provided in the clad, light traveling direction is changed nearly at 90° and the light is radiated outside the clad, the light is then detected by an optical diode. Consequently, the optical power monitor can be provided which is made compact even with multiple channels and has small optical transmission loss. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an optical power monitor mainly used in the field of optical communication.

In recent years, technological innovation in information communication has been remarkable, and in order to respond to the demand for higher communication speed and the increase in the amount of information due to the spread of the Internet, there is a shift from communication using electrical signals to communication using optical signals. Many core cables have been replaced by optical cables due to their processing capacity and processing speed because information gathers from many relay points. As communication between optical cables and user terminals has been reviewed, the demand for a cheaper and more comfortable information communication environment has been increasing.

As the optical communication network is improved, the exchange of information will be carried out at high speed, and new applications will be expanded accordingly, so the amount of information passing through the optical communication network will increase more and more. . In order to increase the amount of information that can be processed by an optical fiber, a high-frequency signal is used to increase the amount of signal per unit time, or multiple wavelength signals having different information, which is called wavelength multiplexing, can be combined into a single signal. A technique of transmitting simultaneously in an optical fiber is used. In addition, in order to form a dense and highly reliable communication network, it is necessary to secure connections in multiple directions and multiple paths, and the use of a plurality of optical fibers is essential from the viewpoint of maintenance applications.

When forming an optical communication circuit that transmits a large number of signals through an optical fiber, the wavelength-multiplexed optical signal is demultiplexed into wavelengths, or the optical signals of the respective wavelengths are combined, Wavelength Division Multipl that said to perform signal branching and insertion
An ex (hereinafter abbreviated as WDM) system is required. As the amount of information increases, the importance of the information to be handled increases. When an optical signal is lost, it is necessary to quickly grasp which optical signal is lost where. In addition to the presence or absence of optical signal connection, it is also necessary to check the signal strength. Further, when the transmission distance is increased, the optical signal intensity is attenuated only by propagating through the optical fiber. Therefore, the Erbium Doped Fiber for amplifying the optical signal is used.
An apparatus called r Amplifier (hereinafter abbreviated as EDFA) is also required. The EDFA is intended to determine the amplification rate, and it is necessary to accurately grasp the intensity of the optical signal input from the outside and the intensity of the optical signal output after amplification. In order to construct a highly reliable optical communication system, it is indispensable to have such a fine monitoring function.

As a method for monitoring an optical signal, a method of branching a part of the optical signal with an optical coupler and detecting the branched optical signal with a photodiode connected to an optical fiber has been used. In this method, each component needs to be fusion-bonded, which hinders reduction in mounting man-hours. The optical coupler has a structure in which an optical signal is branched by bringing a core that is an optical signal propagation portion of an optical fiber close to the optical coupler. Since the length of the core proximity part is an important parameter of the branching amount, it is difficult to reduce the size of the product. Recently, the amount of information that can be transmitted at once is increased by increasing the number of wavelengths to be multiplexed. Since signal detection is performed by demultiplexing each wavelength, the number of optical power monitors required for one apparatus increases. Since the storage space in the apparatus allocated to the optical power monitor is limited, it is becoming essential to reduce the size of the optical power monitor.

An example of a miniaturized optical power monitor is disclosed in Patent Document 1. The disclosed structure is shown in FIG. FIG. 12 b) is a cross-sectional view of the optical power monitor 70. FIG.
) Is an optical power monitor assembly 71 in which a plurality of optical power monitors 70 are assembled in a case 69.
It shows an example in which the upper lid of the case is removed. In FIG. 12b), two optical fibers 5
1 and 52 multicapillary glass ferrule 53 and GRIN (Gradien
t Index) lens 54 is made to face with a predetermined gap 55. A filter 56 is formed on the end face of the GRIN lens, and reflects and transmits light that has passed through the GRIN lens. The transmitted light passes through the gap 57 and is converted into an electrical signal by the photodiode 58 and is taken out from the terminal 59. The intensity value of light in the optical path can be obtained from the electrical output value of the photodiode 58. The multicapillary glass ferrule 53 and the GRIN lens 54 are positioned by a glass tube 60. The GRIN lens is a glass cylinder whose refractive index continuously changes radially from the central axis toward the outer circumferential direction. The refractive index increases from the center toward the outer periphery, and the more the light spreads toward the outer periphery, the more the light travels in the direction of the central axis.

The flow of light will be described with reference to FIG. Light entering the gap 55 from the optical fiber 51 (
Input light) passes through the GRIN lens 54 and reaches the filter 56 on the end surface of the GRIN lens.
Most of the light reaching the filter 56 is reflected, passes through the GRIN lens 54 and the gap 55, enters the optical fiber 52, and becomes output light. The light transmitted through the filter 56 passes through the air gap 57, enters the photon detector 58, is converted into an electric signal, and is output from the terminal 59 as an electric signal. The series of light paths are indicated by solid arrows. On the contrary, when light is input from the optical fiber 52, the same process as the optical path described above is followed, and the light can be extracted from the optical fiber 51 (output light). These series of light paths are indicated by broken arrows. In the description, the names described in the patent literature are used. In the present invention, the photon detector is referred to as a photodiode.

US Pat. No. 6,603,906 B2 FIG.

The light of the optical power monitor shown in FIG. 12b) has a structure that is emitted (radiated) once into the air. Since air has a refractive index different from that of the optical fiber, light emitted into the air diffuses. A lens typified by a GRIN lens is an essential component for condensing diffused light. For this reason, the product size of the optical power monitor depends on the size of the GRIN lens or the glass tube. Therefore, it is difficult to reduce the size of the optical power monitor assembly 71 shown in FIG.

An optical power monitor using a waveguide is described in Patent Document 2. FIG. 13a) shows a plan view of the optical waveguide module, and FIG. 13b) shows the principle of measuring the light quantity. A plurality of waveguides 82 are formed substantially parallel to the substrate 81, and a groove 83 is provided to cross the waveguides 82 at a right angle, so that the waveguides are divided into the input end 86 side and the output end 87 side. A reflection filter 84 is inserted into the groove 83,
A photodetector 85 is disposed on the input end 86 side of the reflection filter 84, and constitutes a planar waveguide type optical circuit 80. The light quantity measurement method will be described using the flow of light with reference to the cross-sectional view of FIG. The waveguide 82 is provided with an upper clad 86 and a lower clad 88 so as to sandwich the core 87. The light passing through the core 87 is emitted into the air in the groove 83, and most of the light passes through the reflection filter 84 and enters the core 87 on the output end 87 side. A part of the light (broken arrow) is reflected by the reflection filter 84 and enters the photodetector 85, where it is converted into an electrical signal, and the intensity value of the light in the optical path can be obtained.

JP, 2003-329862, A FIG.

It can be easily understood that the planar waveguide type optical circuit 80 of Patent Document 2 is an obstacle to downsizing the thickness of the substrate holding the waveguide, the holding mechanism of the photodetector, and the like. In addition, it is well known that the optical loss at the portion connecting the waveguide and the optical fiber is large, and it is difficult to reduce the loss. In order to reduce the loss, it can be easily conceived to replace the optical waveguide of Patent Document 2 with an optical fiber. However, similarly to Patent Document 2, light that enters from the input side is once emitted into the air. Since the light emitted into the air is branched into light entering the output side and light entering the photodetector by the reflection filter, it is difficult to reduce light loss.

Even in the optical power monitor 70 of Patent Document 1 considered to have a small loss, the size of the single-channel optical power monitor 70 can be reduced to φ3.0 mm × 20 mm as long as an individual pigtail fiber, a GRIN lens, and a photon detector are used. Is considered the limit. As shown in FIG. 12a), when the number of channels is increased, the product size is further increased by the amount accommodated in the case 69, and it is difficult to reduce the size.

An object of the present patent is to provide an optical power monitor that can be reduced in size even when the number of channels is increased and that has little optical loss.

The optical power monitor of the present invention is fused by offsetting the central axes of the two optical fibers, causing a part of the propagation light to leak from the core part of the fusion part to the cladding part of the optical fiber in the light propagation direction front side, The leaked light is reflected by the notch surface on the fusion side of the notch provided in front of the fused part of the optical fiber on the front side in the light propagation direction, and the reflected light is transmitted through the cladding on the opposite side of the notch. An optical power monitor that radiates out of the optical fiber, detects the emitted light with a photodiode, and measures the amount of propagating light, and the light on the cut surface on the fusion side from the fusion portion It is preferable that the reflection part is in a position where light leaked to the cladding part interferes and strengthens.

It is more preferable that the light reflection portion of the cut-out surface on the fusion side of the cut-out portion from the fusion portion is in a position where the light leaking into the clad portion interferes with the third-order to fourth-order.

An optical fiber consists of a core located at the center of the axis and a cladding covering the outer periphery. Total reflection occurs at the outer peripheral surface of the core, which is the interface between the core with a high refractive index and the cladding with a low refractive index, and light propagates only within the core. It is a structure to do. If the central axes of the two optical fibers are fused with an offset of several μm, most of the fused portions are connected to each other between cores or clads.
In the part where the cores are connected, light propagates through the cores as they are. On the other hand, in a portion where a part of the core is fusion-spliced with the clad, light reaching the fusion surface leaks to the clad portion. Although the leaked light travels forward as it is, the refractive index differs between the core and the clad, and the fused surface has a crescent shape, so that the leaked light spreads radially. Furthermore, it propagates in the cladding while repeatedly reflecting on the outer peripheral surface of the cladding. Since the reflected light causes interference, the light travels while strengthening and weakening each other during propagation. The light propagated while being repeatedly reflected in the clad is reflected by the cut-out surface on the fusion side in the cut-out portion provided in the clad, and changes its traveling direction to radiate the light outside the optical fiber. By disposing the photodiode on the extension line in the traveling direction of the radiated light, the light amount is converted into a current, and an electric signal proportional to the light amount can be obtained.

As for the shape of the fused part of the core, when the joint part of the core part of the optical fiber is seen from the front side in the light propagation direction, the core is joined with an offset of several μm. Looks like. The thickest part of the crescent moon is the offset value. The shape of the cross section of the fused part is that the core surface is substantially perpendicular to the central axis of the optical fiber in the joined state before the fusion,
Although the substantially vertical planes are offset by several μm, the bonded portion of the core is bonded with a natural curvature by heat fusion. The same applies to the cladding.

The fused surface has a crescent shape and a natural curvature. The light emitted in the direction having a natural curvature changes the angle of the light emitted depending on the curvature, so that the light reflected in the clad interferes with each other, and the light strengthens and weakens. Preferably, the intensified light from the third order to the fourth order is reflected by the cut-out surface on the fusion side of the cut-out portion, and the reflected light is detected by the photodiode. For this reason, it is important that the surface having the natural curvature of the fusion part and the cut surface on the fusion side of the notch part are opposed to each other with a predetermined angle. In other words, the fusion part is orthogonal to the central axis of the optical fiber, and the notch part is also orthogonal, so the fusion part and the notch part are parallel. Since the fusion part and the notch part each have a predetermined angle with respect to the optical fiber cross section, the surface having the natural curvature of the fusion part and the notch surface on the fusion side of the notch part have a predetermined angle. Hold it and face it. On the other hand, light emitted in a direction parallel to the natural curvature exits symmetrically with respect to the central axis of the optical fiber and is reflected by the clad, so that interference does not occur.

From the results of many experiments and work, the natural curvature of the fused part is large in the offset direction (the direction in which the core cross section behind the light propagation direction looks like a crescent moon) and small in the direction opposite to the offset. I understand. In order to make it easy to understand, the description will be made by replacing with the inclination angle instead of the curvature.
The tilt angle varies depending on the fusing conditions such as the fusing temperature and the force to butt, but it is in the range of 5 degrees to 15 degrees, and there is a difference of 3 to 6 degrees in the angle opposite to the offset direction. I understand. Because the tilt angle is different, there is a gap between the period in which the light emitted in the offset direction repeats reflection and the period in which the light emitted in the opposite direction of the offset repeats reflection. But that happens. If the first strengthening point is the primary, 3
Concentrate the light leaked from the core of the welded part to the clad on the cutout surface most efficiently by arranging the cutout part on the fusion side of the cutout part at the second or fourth strengthening place. Can do.

The optical power monitor of the present invention can be a single channel optical power monitor in which an optical fiber having a notch and a photodiode are housed in a case. Further, a plurality of channels can be arranged in parallel to form a multi-channel configuration. By using a single multi-channel photodiode without arranging a plurality of photodiodes, the size can be further reduced. The optical power monitor of the present invention leaks a part of the light propagating through the optical fiber from the core to the clad at the fusion part, and changes the direction of the leaked light by about 90 degrees at the notch surface provided in the optical fiber,
Light is output from the clad on the side opposite to the cut surface to the air, and the light is converted into an electric signal by a photodiode. Since light emitted from the clad into the air is detected, light emitted from an adjacent optical fiber or light entering from outside the case becomes noise. Therefore, light shielding between adjacent optical fibers and light shielding of the case are essential.

In the optical power monitor of the present invention, it is preferable that the light reflection portion of the cut-out surface on the fusion side of the fusion portion and the cut-out portion is a distance of 4.5 mm or more and 7.5 mm or less.

The emission angle of the light leaked from the core of the fused part to the clad depends on the natural curvature shape of the fused part, so it is not easy to obtain by calculation. From the results obtained by numerous experiments,
It has been found that the light receiving sensitivity increases when a notch surface is provided at a position where the third order or fourth order strengthens each other. This is a position 4.5 mm or more and 7.5 mm or less away from the offset fusion site from the wavelength of light used. The main communication wavelength band is 1520 nm to 1620 nm.
Although the third-order or fourth-order strengthening position varies depending on the wavelength, the length of the cut-out surface is sufficiently long, so that the difference in wavelength of 100 nm can be completely ignored.

In the optical power monitor of the present invention, it is preferable that the cut-out surface on the fusion side of the cut-out portion is 38 degrees or more and 45 degrees or less with respect to the optical fiber optical axis on the front side in the light propagation direction.

Since the fusion-side notch surface of the notch functions to change the traveling direction of light propagating in the clad by approximately 90 degrees, the angle Θ1 of the notch surface is preferably 38 degrees or more and 45 degrees or less. When the angle is less than 38 degrees, the light is reflected by the fusion-side cut-out surface, but the direction of the reflected light is not perpendicular to the outer periphery of the optical fiber but is inclined obliquely. That is, most of the light is radiated to the outside of the clad, but the amount of light reflected on the outer peripheral surface of the clad and returning into the clad increases. This is because light enters at an angle of 38 degrees or less with respect to the outer peripheral surface of the clad. On the other hand, when the angle of the fusion-side cutout surface exceeds 45 degrees, the ratio of transmission through the cutout surface increases and the light receiving sensitivity is lowered. Good characteristics can be obtained when the fusion-side cut-out surface of the cut-out portion has an angle of 38 ° to 45 °.

The cut-out surface opposite to the fusion-side cut-out surface does not have any influence on the reflection of leaked light, and therefore may have any shape. Further, the shape of the bottom of the notch need not be specified. The notch is easy to form by grinding. In order not to leave processing stress on the optical fiber, the cross-sectional shape of the notch is preferably symmetrical, and particularly preferably wedge-shaped. As long as the angle of the fusion-side cutout surface can be obtained accurately, the present invention is not limited to grinding, and may be formed using a dry etching technique such as ion milling.

In the optical power monitor of the present invention, the distance d between the bottom of the notch and the outer peripheral surface of the core is 0.5 μm.
It is preferable to have an interval of m or more and 8 μm or less.

Since a part of the light in the core is leaked into the clad, reflected by the cut-out side of the fusion side and detected by the photodiode, it is important that the light is reliably applied to the cut-out surface. If the depth of the notch becomes shallow, that is, if the area of the fusion notch is reduced, not all of the light hits the notch and the proportion of light propagating in the cladding near the core increases. . For this reason, it is important to keep the distance between the outer peripheral surface of the core portion and the bottom surface of the notch (the end surface on the core side of the notch) to 8 μm or less. On the other hand, although the optical fiber has a structure in which light propagates in the core, there is no strict boundary, so there is light that propagates leaking into the cladding near the core. When the depth of the notch becomes deep and the distance between the outer peripheral surface of the core and the bottom of the notch is 0.5 μm or less, the light propagating in the cladding near the core and the light propagating in the core Leaks and becomes a loss. For this reason, it is preferable that the space | interval of the bottom part of a notch part and a core part outer peripheral surface is 0.5 micrometer or more and 8 micrometers or less.

While leaking from the core of the fused part to the clad and reflected by the clad, the light strengthened by interference hits the cut-out side of the fusion side and changes the direction of the light by approximately 90 degrees. The light passes through the clad on the side of the notch, further passes through the clad on the side opposite to the core and the notch, exits the optical fiber, and enters the photodiode.

The optical power monitor of the present invention uses the distance between the central axes of two optical fibers as an offset amount h, and by changing the offset amount h, determines the amount of light leaked to the clad portion of the optical fiber in the light propagation direction, and the offset amount h Is more than 0.05k of optical fiber core diameter k 0.3
It is preferable that it is 2k or less.

When the offset amount h is changed, the amount of light leaking to the cladding can be changed. Since the intensity distribution of the light in the core follows a Gaussian distribution, the light amount change increases as the offset amount increases. In a commonly used single mode optical fiber, the core diameter k is 9.2 μm (wavelength 1310 nm). When fused with an offset amount of 2.5 μm,
The light receiving sensitivity is about 100 mA / W. Photosensitivity is the intensity of light entering the photodiode (W)
The output current (mA) of the corresponding photodiode is represented by (mA / W).

In the optical power monitor of the present invention, the surface roughness R of the light reflection portion of the cut-out surface on the fusion side of the cut-out portion.
a is preferably 2 nm or less.

The light propagating in the clad is reflected by the cut-out surface on the fusion side of the cut-out portion. The surface roughness of the reflection surface changes the amount of light entering the photodiode, that is, the light receiving sensitivity changes. . When the surface roughness becomes rough, the reflected light spreads due to light scattering, so that the light incident on the photodiode becomes weak. If Ra is 2 nm or less, high light receiving sensitivity can be obtained. Ra is J
It is a value measured according to IS B0601. In addition, since the wavelength of light used is a long wavelength in the vicinity of 1550 nm, defining not only the surface roughness Ra but also the surface waviness is effective in reducing the light reflectance. The average length AR of the roughness motif is obtained from the envelope undulation curve according to JIS B0631, and it is preferable that it is not more than ½ of the wavelength used by AR. By making the average length of waviness smaller than the wavelength to be used, the light reflection effect can be increased.

The optical power monitor of the present invention can form a film having a high light reflectivity on at least the cut-out surface of the cut-out portion on the fusion side.

Providing a high light reflection film on the notched surface is effective in increasing the light receiving sensitivity. The high light reflection film can increase the light reflection efficiency by the effect of suppressing light scattering at the time of reflection and the effect of suppressing transmission on the reflection surface. Examples of film materials include gold (Au) and silver (Au
Ag), aluminum (Al), and copper (Cu) are desirable. Ag, Al, Cu, which is easily oxidized
From the viewpoint of temporal stability, it is more preferable to use Au that is difficult to oxidize. Vapor deposition or sputtering can be used for film formation. The light reflecting film may be provided not only on the fusion-side cutout surface but also on the outer peripheral surface of the optical fiber and the cutout portion other than the portion facing the photodiode. The effect of suppressing the influence of external light can be obtained by covering the portion other than the portion facing the photodiode with the light reflecting film.

A photodiode is known to change with time in characteristics of converting light into current due to humidity. For this reason, it is preferable that the sealed case of the optical power monitor of the present invention is filled with dry nitrogen or dry argon.

Offset fusion of the central axes of the two optical fibers, leakage of light from the core into the cladding, reflection of the third or fourth intensified light of the leaked light at the notch surface provided in the cladding, To provide an optical power monitor that can be reduced in size even if the number of channels is increased by detecting with a photodiode after changing the traveling direction of the light by approximately 90 degrees and radiating it outside the cladding, and reducing optical propagation loss. Became possible.

Hereinafter, the present invention will be described in detail based on examples with reference to the drawings. In order to make the explanation easy to understand, the same reference numerals are used for the same parts and parts.

FIG. 1 is a plan view of an optical power monitor with the top cover removed, FIG. 2 is a sectional view of the optical power monitor, FIG. 3 is an enlarged sectional view of a fusion part and a notch part, and FIG. 4 is a core of the fusion part. The explanatory view of the progress of the light leaked from the clad to the clad is shown.

The structure of the optical power monitor assembly 1 of the present invention will be described in detail with reference to FIGS. FIG. 1 shows an 8-channel optical power monitor 1 in which 8 sets of optical power monitor components are assembled in one case 9. The rear optical fiber 2 in the light propagation direction and the front optical fiber 3 in the light propagation direction are drawn out from the case 9 through the protective tube 8, filled with dry nitrogen gas, and the upper lid 11 is bonded with a resin (not shown). The electrode 10 of the photodiode 7 is taken out from the outer side surface of the case 9. Illustration of the wiring of the photodiode 7 and the electrode 10 passing through the case 9 is omitted. Two optical fiber fixing blocks 4 and a photodiode 7 were fixed to the inner bottom of the case 9 with an adhesive. The optical fiber fixing block 4 has eight V-shaped grooves in order to keep the distance between the optical fibers with high accuracy. An optical fiber was fixed to the V groove of the optical fiber fixing block 4 with a resin. The photodiodes 7 have 8 channels and are arranged at positions corresponding to the notch portions 6 of the optical fiber. In the present application, the fused optical fiber 2 in the light propagation direction and the optical fiber 3 in the light propagation direction are referred to simply as optical fibers unless otherwise specified. Further, in order to prevent light interference between the optical fibers, a light shielding plate 30 is provided between the optical fibers.

With reference to FIG. 3, the dimensions of each part and the flow of light will be described. As shown in FIG. 3 a), the central axis 15 of the optical fiber 2 in the light propagation direction rear side and the central axis 16 of the optical fiber 3 in the light propagation direction front side were fused with an offset amount h of 2 μm. For the fusion, an offset fusion mode of the fusion machine was used. When a predetermined offset amount h was input, the rear optical fiber 2 in the light propagation direction was moved by a distance stepping motor having an offset amount h, and then heated and fused. The notch portion 6 was formed by grinding so that the light reflecting portion 14 of the V-shaped notch portion 6 was located at a position 6.80 mm away from the fused portion of the optical fiber subjected to offset fusion. Since the depth of the notch 6 was 55.0 μm, the distance d between the notch bottom 12 and the outer periphery of the core was 2.9 μm. The angle Θ1 with respect to the center axis of the cut-out surface on the fusion side of the cut-out portion 6 is 40.5 degrees, the surface roughness of the cut-out portion is 1.2 nm in Ra, and the average roughness motif obtained from the envelope waviness curve The length AR was 580 nm. For reference, the angle Θ2 of the notch surface of the notch 6 on the side opposite to the fusion side was 45.2 degrees. A gold light reflecting film 20 was formed to a thickness of 0.03 μm by vacuum deposition on the fusion-side surface of the notch.

The light (solid arrow) entering the rear optical fiber 2 in the light propagation direction is divided into light leaking to the clad side at the fusion part 5 (broken arrow) and light entering the optical fiber 3 in the light propagation direction (solid arrow). . Light entering the optical fiber 3 in the light propagation direction passes through the core without exiting into the air. In the conventional optical power monitor, once it goes out into the air, it returns to the optical fiber or the optical waveguide. One of the features of the present optical power monitor is that light other than leaked light entering the photodiode is never emitted into the air, and this is one of the reasons why low propagation loss is obtained. The light leaked to the clad side at the fused portion 5 is repeatedly reflected in the clad and strengthened by third-order or fourth-order interference, and the optical path is bent by about 90 degrees at the light reflecting portion 14 of the notch portion 6. The light passes through the core-cladding, exits into the air, enters the photodiode 7 and is converted into an electrical signal.

The offset fused part of the core of the A part in FIG. 3a) will be described in more detail with reference to FIGS. 3b) and 3c). FIG. 3b) is a cross-sectional view of the fused portion as seen from the X direction of FIG. 3c). The joint between the core portion of the optical fiber 2 in the light propagation direction rear side optical fiber 2 and the core portion of the optical fiber 3 in the light propagation direction front side optical fiber 3 is integrated with a natural curvature 17 created by the melted core portion. The leaked light that has passed through this natural curvature 17 enters the cladding at various angles and is repeatedly reflected. Light that has been repeatedly reflected causes interference, which intensifies and weakens the light. The natural curvature 17 plays an important role of causing light interference. FIG. 3c) shows the core of the fusion part as seen from the optical fiber 3 in the light propagation direction. The rear optical fiber 2 in the light propagation direction has a crescent-shaped core portion 18.
Is forming.

FIG. 4 shows the flow of leaked light viewed from the X and Y directions in FIG. 3c). FIG. 4a) shows the flow of light as seen from the X direction. The leakage light emitted from the crescent-shaped core portion 18 is divided in the direction in which the light is emitted at the natural curvature 17 region. Light emitted at an angle of Θu (broken arrow) and light emitted at an angle of Θd (solid arrow) are repeatedly reflected on the outer periphery of the cladding. For this reason, the part where the tip of the broken line and the solid line arrow approaches is the part where the light is strengthened. The first strengthening is the primary, and the third or fourth order hits the light reflecting portion 14 of the notch 6 and enters the photodiode 7 by changing the direction by approximately 90 degrees. The part that strengthens and the part that weakens appear alternately. FIG. 4b) shows the flow of light viewed from the Y direction. Since the light emitted from the crescent-shaped core part is symmetrical, light interference does not occur, and reflection is repeated on the outer periphery of the clad to reach the notch. In this embodiment, Θu
Was 14.0 degrees and Θd was 9.6 degrees.

Twenty completed optical power monitors 1 were irradiated with light having a wavelength of 1550 nm and a light intensity of 1.2 mW, and the optical power monitor characteristics were evaluated. Since there are 20 channels in 8 channels, a total of 160
Channel characteristics were measured and averaged. The propagation loss was 0.72 (-dB) and the light receiving sensitivity was 64.4 (mA / W), which was a good value. By placing a light shielding plate between the optical fibers, the crosstalk could be reduced to 48.3 (-dB). The outer dimensions of the case 9 and the upper lid 11 excluding the protective tube 8 and the electrode 10 are 23 mmL × 17 mmW × 2
. Compared with the conventional optical power monitor assembly of 6 mmh in FIG. 11 a), the volume can be significantly reduced to 1/5.

FIG. 5 shows the relationship between the distance L from the fused part to the notch and the light receiving sensitivity of the photodiode. The light receiving sensitivity was measured by changing the distance L from the fused part to the notch part from 0.8 mm to 9.0 mm. Other than the distance L, the shape (angle) and dimensions of the notch, the light reflection film, and the like are the same as those in the first embodiment. It can be seen that the light receiving sensitivity is large when the distance L from the fused portion to the notch portion is about 1, 3, 5, 7, 9 mm, and the light receiving sensitivity is low at about 2, 4, 6, 8 mm. Where the light receiving sensitivity is increased, it is strengthened by light interference, and where the light receiving sensitivity is decreased, it is weakened by light interference. About 1 mm is a place where light is strengthened by primary interference. As the order increased, the light receiving sensitivity also increased, and the highest light receiving sensitivity was exhibited in the third and fourth orders. It has been proved that an optical power monitor having high light receiving sensitivity can be obtained by using third-order or fourth-order interference light.

FIG. 6 shows the relationship between the angle Θ1 of the notch surface and the light receiving sensitivity. The distance L from the fused part to the notch was 6.80 mm, and fourth-order interference light was used. The optical power monitor was manufactured by changing the angle Θ1 of the notch from 28 degrees to 60 degrees. The light receiving sensitivity measurement method and the light reflecting film were the same as those in Example 1. In the case where the angle Θ1 of the notch is between 38 degrees and 45 degrees, the light receiving sensitivity shows a value larger than 60 mA / W, but the light receiving sensitivity suddenly decreases at angles of 38 degrees or less and 45 degrees or more. It is falling. The cause of the decrease in the light receiving sensitivity is that if the angle is less than 38 degrees, the light is reflected by the cut-out surface on the fusion side, but the direction of the reflected light is not perpendicular to the outer periphery of the optical fiber but inclined. That is, most of the light is radiated to the outside of the clad, but the amount of light reflected on the outer peripheral surface of the clad and returning into the clad increases. This is considered to be because light enters the outer circumferential surface of the clad at an angle of 38 degrees or less. Conversely, the angle of the cut-out surface on the fusion side is 45
If it exceeds the degree, it is considered that the ratio of transmitting through the notch surface increases and the light receiving sensitivity decreases.

FIG. 7 shows the relationship between the offset amount h / core diameter k and the light receiving sensitivity. An optical power monitor was manufactured by changing the value of offset amount h / core diameter k from 0.02 to 0.32. The light receiving sensitivity measurement method, the position and angle of the notch, the light reflection film, and the like were the same as those in Example 1. It can be seen that the light receiving sensitivity increases as the offset amount h / core diameter k increases. As the offset amount h increases, the amount of light that leaks to the clad side increases, so that the amount of light entering the notch increases. For this reason, the amount of light incident on the photodiode increases, and the light receiving sensitivity shows a high value. The high light receiving sensitivity means that accurate monitoring is possible even if the intensity of light propagating in the optical fiber is reduced. However, this means that most of the light propagating in the optical fiber is extracted, which is disadvantageous from the viewpoint of propagation loss.

FIG. 8 shows the relationship between the offset amount h / core diameter k and propagation loss. It can be seen that the propagation loss increases as the ratio of offset amount h / core diameter k increases. This means that the rate at which the light intensity of propagating light in the optical fiber decreases is increased. The performance of the optical power monitor is determined by the characteristics of both light receiving sensitivity and propagation loss. By setting the offset amount h / core diameter k to 0.05 k or more and 0.32 k or less, a high-performance optical power monitor with high light receiving sensitivity and small propagation loss was obtained. The values required for light reception sensitivity and propagation loss depend on where the optical communication device is used in the network and how the amplifier that amplifies the light intensity is arranged. It depends on the device design of the communication device.

FIG. 9 shows the relationship between the surface roughness of the notch surface and the light receiving sensitivity. The surface roughness Ra of the notched surface is 0.1 n
The optical power monitor was manufactured by changing from m to 10.0 nm. Surface roughness was achieved by changing the grain size of the diamond wheel. The surface roughness Ra is a value measured with a stylus type surface roughness meter in accordance with JIS B0601. The surface roughness was measured by folding the optical fiber at the same time after processing the notch. Gold was vacuum-deposited as a light reflecting film on the notch. Since light is reflected at the interface between the clad material and the gold light reflecting film, rough reflection tends to occur if the surface roughness is rough. When the thickness exceeds 2 nm, it has been confirmed that the light receiving sensitivity rapidly decreases because of irregular reflection.

FIG. 10 shows the relationship between the average length AR of the roughness motif obtained from the envelope waviness curve of the notch and the light receiving sensitivity. The optical power monitor was manufactured by changing the AR from 100 nm to 2800 nm. The surface roughness Ra was 1.2 nm, and the others were the same as those in Example 1. The AR measurement is
According to JIS B0631, the optical fiber which was processed simultaneously as in the measurement of the surface roughness was folded at the notch and measured. When the average length AR of the roughness motif is increased, the undulation period of the notch surface is close to the wavelength of light propagating in the optical fiber, so that the surface is easily affected. The light reflected by the cut-out surface caused interference between the reflected lights, and a decrease in light receiving sensitivity was observed. It was confirmed that the average length AR of the roughness motif is preferably about 800 nm or less. This 8
00 nm corresponds to about ½ of the wavelength of light used.

FIG. 11 shows the relationship between the distance d between the bottom of the notch and the outer peripheral surface of the core, the light receiving sensitivity, and the propagation loss. The optical power monitor was manufactured by changing the distance d from 0.2 μm to 15 μm. Except for the distance d, it was the same as Example 1. When d is in the range of 0.5 μm to 8 μm, the change in light receiving sensitivity is very small. However, as d increases, the area of the notch surface decreases, and the light receiving sensitivity gradually decreases. On the other hand, when d is smaller than 0.5 μm, the bottom of the notch is very close to the core. In the optical fiber, light mainly propagates through the core portion. Actually, however, some light leaks to the cladding in the vicinity of the core, and this leaked light is reflected by the notch and monitored. Therefore, when d is smaller than 0.5 μm, the light receiving sensitivity is increased, but the propagation loss is also increased. It was verified that a stable optical power monitor with little change in light receiving sensitivity and propagation loss can be obtained when d is in the range of 0.5 μm to 8 μm.

It is the top view which removed the upper cover of the optical power monitor of this invention. It is sectional drawing of the optical power monitor of this invention. It is an expanded sectional view of a fusion part and a notch of an optical power monitor of the present invention. It is explanatory drawing of the advancing condition of the light which leaked to the clad from the core of the fusion | melting part of the optical power monitor of this invention. It is a figure which shows the relationship between the distance L from a melt | fusion part to a notch part, and the light reception sensitivity of a photodiode of Example 2 of this invention. It is a figure which shows the relationship between the notch part angle | corner (theta) 1 and light reception sensitivity of Example 3 of this invention. It is a figure which shows the relationship between offset amount h / core diameter k and light reception sensitivity of Example 4 of this invention. It is a figure which shows the relationship between offset amount h / core diameter k, and propagation loss of Example 4 of this invention. It is a figure which shows the relationship between the surface roughness of a notch surface, and light reception sensitivity of Example 5 of this invention. It is a figure which shows the relationship between the average length AR of the roughness motif calculated | required from the envelope waviness curve of the notch surface of Example 6 of this invention, and light reception sensitivity. It is a figure which shows the relationship between the distance d of the bottom part of a notch part, and a core part outer peripheral surface, and light reception sensitivity of Example 7 of this invention. It is a figure of the conventional optical power monitor. It is a figure of the conventional planar waveguide type optical power monitor.

Explanation of symbols

1 power monitor, 2 optical fiber in the back propagation direction,
3 Front optical fiber in the light propagation direction, 4 Optical fiber fixing block,
5 fused parts, 6 notched parts,
7 photodiode, 8 protective tube,
9 cases, 10 electrodes,
11 Upper lid, 12 Notch bottom,
13 core outer periphery, 14 light reflection part,
15 central axis a, 16 central axis b,
17 Natural curvature, 18 Crescent core,
20 light reflecting film, 30 light shielding plate,
51 optical fiber, 52 optical fiber,
53 Multicapillary glass ferrule, 54 GRIN lens,
55 air gap, 56 filter,
57 gap, 58 photon detector,
59 electrodes, 60 glass tubes,
69 cases, 70 optical power monitors,
71 optical power monitor assembly, 80 plane waveguide type optical circuit,
81 substrate, 82 waveguide,
83 grooves, 84 reflective filters,
85 photodetectors, 86 upper cladding,
87 core, 88 lower cladding.

An example of a miniaturized optical power monitor is disclosed in Patent Document 1. The disclosed structure is shown in FIG. FIG. 12 b) is a cross-sectional view of the optical power monitor 70. FIG.
) Is an optical power monitor assembly 71 in which a plurality of optical power monitors 70 are assembled in a case 69.
It shows an example in which the upper lid of the case is removed. In FIG. 12b), two optical fibers 5
1 and 52 multicapillary glass ferrule 53 and GRIN (Gradien
t Index) lens 54 is made to face with a predetermined gap 55. A filter 56 is formed on the end face of the GRIN lens, and reflects and transmits light that has passed through the GRIN lens. The transmitted light passes through the gap 57 and is converted into an electrical signal by the photodiode 58 and is taken out from the electrode 59. The intensity value of light in the optical path can be obtained from the electrical output value of the photodiode 58. The multicapillary glass ferrule 53 and the GRIN lens 54 are positioned by a glass tube 60. The GRIN lens is a glass cylinder whose refractive index continuously changes radially from the central axis toward the outer circumferential direction. The refractive index increases from the center toward the outer periphery, and the more the light spreads toward the outer periphery, the more the light travels in the direction of the central axis.

The flow of light will be described with reference to FIG. Light entering the gap 55 from the optical fiber 51 (
Input light) passes through the GRIN lens 54 and reaches the filter 56 on the end surface of the GRIN lens.
Most of the light reaching the filter 56 is reflected, passes through the GRIN lens 54 and the gap 55, enters the optical fiber 52, and becomes output light. The light transmitted through the filter 56 passes through the gap 57, the light
It enters the diode 58 and is converted into an electric signal, which is output from the electrode 59 as an electric signal. The series of light paths are indicated by solid arrows. On the contrary, when light is input from the optical fiber 52, the same process as the optical path described above is followed, and the light can be extracted from the optical fiber 51 (output light). These series of light paths are indicated by broken arrows .

An optical power monitor using a waveguide is described in Patent Document 2. FIG. 13a) shows a plan view of the optical waveguide module, and FIG. 13b) shows the principle of measuring the light quantity. A plurality of waveguides 82 are formed substantially parallel to the substrate 81, and a groove 83 is provided to traverse the waveguides 82 at a right angle.
Divided into input side and output side . A reflection filter 84 is inserted into the groove 83, and a photodetector 85 is arranged on the input side of the reflection filter 84 to constitute a planar waveguide type optical circuit 80. FIG.
The light quantity measuring method will be described using the flow of light with reference to the cross-sectional view of 3b). The waveguide 82 is provided with an upper clad 86 and a lower clad 88 so as to sandwich the core 87. The light passing through the core 87 is emitted into the air in the groove 83, and most of the light passes through the reflection filter 84 and enters the core 87 on the output side . A part of light (broken arrow) is reflected by the reflection filter 84 to detect the photodetector 8.
5 is converted into an electric signal and the intensity value of the light in the optical path can be obtained.

Loss even optical power monitor 70 of Patent Document 1 considered small, miniaturization of the single channel optical power monitor 70 is a separate pigtail fiber and the GRIN lens, the light
As long as a diode is used and assembled, φ3.0 mm × 20 mm is considered the limit. FIG.
As in 2a), when the number of channels is increased, the product size is further increased by the amount accommodated in the case 69, and it is difficult to reduce the size.

FIG. 1 is a plan view of an optical power monitor with the top cover removed, and FIG. 2 is a cross-sectional view taken along line j ′ of FIG.
FIG . 3, FIG. 3 is an enlarged cross-sectional view of the fusion part and the notch part, and FIG. 4 is an explanatory view of the progress of light leaked from the core of the fusion part to the cladding.

1 power monitor, 2 optical fiber in the back propagation direction,
3 Front optical fiber in the light propagation direction, 4 Optical fiber fixing block,
5 fused parts, 6 notched parts,
7 photodiode, 8 protective tube,
9 cases, 10 electrodes,
11 Upper lid, 12 Notch bottom,
13 core outer periphery, 14 light reflection part,
15 central axis a, 16 central axis b,
17 Natural curvature, 18 Crescent core,
20 light reflecting film, 30 light shielding plate,
51 optical fiber, 52 optical fiber,
53 Multicapillary glass ferrule, 54 GRIN lens,
55 air gap, 56 filter,
57 air gap, 58 photodiode ,
59 electrodes, 60 glass tubes,
69 cases, 70 optical power monitors,
71 optical power monitor assembly, 80 planar waveguide optical circuit,
81 substrate, 82 waveguide,
83 grooves, 84 reflective filters,
85 photodetectors, 86 upper cladding ,
87 core, 88 lower cladding.

The natural curvature of the fused part is large in the offset direction ( the apex direction of the arc outside the crescent moon in the direction in which the core cross section behind the light propagation direction looks like a crescent moon ), and in the opposite direction to the offset (light The apex direction of the arc inside the crescent moon in the direction in which the core cross section behind the propagation direction looks like a crescent moon
) Shows that the radius of curvature is small from many experiments and work results. In order to make it easy to understand, the description will be made by replacing with the inclination angle instead of the curvature. The tilt angle varies depending on the fusing conditions such as the fusing temperature and the force to butt, but it is in the range of 5 degrees to 15 degrees, and there is a difference of 3 to 6 degrees in the angle opposite to the offset direction. I understand. Because the tilt angle is different, there is a gap between the period in which the light emitted in the offset direction repeats reflection and the period in which the light emitted in the opposite direction of the offset repeats reflection. However, it occurs. If the first strengthening place is the primary, the third or fourth strengthening place,
By disposing the cut-out surface on the fusion side of the cut-out portion, the light leaked from the core of the fusion-bond portion to the clad can be collected most efficiently on the cut-out surface.

more than

Claims (8)

The optical axes of the two optical fibers are fused by offsetting, and a part of the propagation light is leaked from the core part of the fusion part to the clad part of the optical fiber on the front side in the light propagation direction, and the leaked light is forward in the light propagation direction. Reflected by the notch surface on the fusion side of the notch provided on the front side from the fusion part of the side optical fiber, the reflected light is transmitted through the cladding on the opposite side of the notch and radiated out of the optical fiber, An optical power monitor that detects the emitted light with a photodiode and measures the amount of propagating light. The light reflection part of the cut surface on the fusion side from the fusion part leaks to the cladding part. An optical power monitor characterized by being in a position where it interferes and strengthens. 2. The light reflection portion of the cut-out surface on the fusion side of the cut-out portion from the fusion portion is in a position where light leaking from the third-order to the fourth-order clad portion interferes and strengthens. Optical power monitor as described in 3. The optical power monitor according to claim 1, wherein a light reflection portion of the cut-out surface on the fusion side from the fusion part is 4.5 mm or more and 7.5 mm or less. 4. The optical power monitor according to claim 1, wherein a cut-out surface on the fusion side of the cut-out portion is at least 38 degrees and at most 45 degrees with respect to the optical fiber optical axis on the front side in the light propagation direction. 5. The optical power monitor according to claim 1, wherein the bottom portion of the cutout portion and the outer peripheral surface of the core portion have an interval of 0.5 μm or more and 8 μm or less. The distance between the central axes of the two optical fibers is defined as an offset amount h, and by changing the offset amount h, the amount of light leaked to the clad portion of the optical fiber on the front side in the light propagation direction is determined, and the offset amount h is 0 of the core diameter k of the optical fiber. The optical power monitor according to claim 1, wherein the optical power monitor is 0.05 k or more and 0.32 k or less. 5. The optical power monitor according to claim 1, wherein the surface roughness Ra of the light reflection portion of the cut-out surface on the fusion side of the cut-out portion is 2 nm or less. 8. The optical power monitor according to claim 1, wherein a film having a high light reflectivity is formed on at least a notch surface of the notch portion on the fusion side.

Images