JPS6382317A - Light interference angular velocity meter and its manufacture - Google Patents

Abstract

PURPOSE:To eliminate the fused seam trace of an optical fiber, to dispense with the reinforcing process of said place and to reduce manufacturing cost, by forming each optical part necessary for a light interference angular velocity meter by one continuous optical fiber. CONSTITUTION:The annular optical path 16 of a light interference angular velocity meter is constituted of optical fibers 71, 71A and both optical fibers are drawn out as they are as incident and emitting paths 14, 15. These fibers 71, 71A are mutually welded in parallel and stretched to constitute the first light distributing coupler 13. Further, a polishing surface 51 or a metal 53 is mounted to one fiber 71 of the coupler 13 in the vicinity of the arc-shaped apex formed to a part of said fiber 71 and a polarizer 19 generating a predetermined evanescent wave is formed. A part of the fiber 71 from the polarizer 19 and a part of the optical fiber 72 from a light source 11 are welded in parallel and stretched to form the second optical coupler 18. Further, the fiber 71 of the coupler 18 is extended to be connected to a photoelectric converter to bring the light interference angular velocity meter to simple constitution.

Description

DETAILED DESCRIPTION OF THE INVENTION "Industrial Application Field" The present invention is directed to a system in which an annular optical path is constituted by an optical fiber, and clockwise light and counterclockwise light are propagated in opposite directions to this annular optical path. The present invention relates to an optical interference angular velocity meter that measures the angular velocity applied to an annular optical path by detecting the phase difference generated between the two lights, and a method for manufacturing the same.

"Prior Art" FIG. 9 is a diagram showing the basic configuration of an optical interference gyrometer. Light L1 guided from a light source 1) such as a laser oscillator to an optical fiber 12 is divided into a clockwise light L2 and a counterclockwise light L3 by an optical distribution coupler 13, and these clockwise light L2 and counterclockwise light L3 are The light L is input into the annular optical path 16 that goes around at least once through the input/output paths 14 and 15, propagates clockwise and counterclockwise through the annular optical path 16, and exits from the annular optical path 16.4. These emitted lights L4
The light L5 is combined by the optical splitting coupler 13 and interfered with each other, and is received by the photoelectric converter 17 as interference light L6, where it is converted into an electrical signal.

The annular optical path 16 is constructed by winding an optical fiber into a loop shape multiple times. When no acceleration is applied to the annular optical path 16 in its circumferential direction, the phase difference between the output lights L4 and L5 from the annular optical path 16 is zero, but the phase difference around the axis of the annular optical path 16 is When an angular velocity Ω is applied to L4. Phase difference Δφ due to Sagnac effect during L5. occurs. The optical interference gyrometer uses this phase difference Δφ. The angular velocity Ω applied to the annular optical path can be determined by measuring the interference light signal.

Figure 1O shows a conventional optical interference angular velocity meter, where the output light L4°L5
The interference light is measured and its phase difference Δφ is determined. FIG. 3 is a diagram illustrating an example of a configuration for detecting with high sensitivity. Portions corresponding to those in FIG. 9 are shown with the same reference numerals.

In this conventional example, in addition to the first optical distributor/coupler 13, a second optical distributor/coupler 18 is provided. That is, the light Ll from the light source 1) is guided to the optical fiber 12, and is passed through the second optical splitting coupler 1.
8, the light is distributed to two lights LIA and LIB. One of the distributed lights LIA is incident on the polarizer 19 and is polarized by the polarizer 19 to become linearly polarized light. This linearly polarized light enters the first optical splitting coupler 13. Note that the polarization direction of the polarizer 19 and the optical fibers 12.14, 15.16 that constitute the optical path
The principal axes of birefringence are aligned so that they match. The light L2. which is split into two lights by the first optical splitter/coupler 13. L3 propagates in directions opposite to each other through the annular optical path 16, as already explained.

In this conventional example, one of the person exit paths 1 to the annular optical path 16 is
A phase modulator 21 is disposed at 5, and the counterclockwise light L3 entering the annular optical path 16 and the clockwise output light L5 emitted from the annular optical path 16 are each subjected to phase modulation by the phase modulator 21. Outgoing light L4. L5 is combined into one light beam by the first optical splitting coupler 13 to form interference light L6, and the interference light L6 is polarized by the polarizer 19. In other words, only the previously polarized light can pass through the interference light L6. The interference light that has passed through the polarizer 19 is further divided into two beams L6A and L6B by the second optical distribution coupler 18, and one of the distributed beams L6A enters the photoelectric converter 17 and is converted into an electrical signal.

In this case, the light intensity 1o of the interference light L6A is ■. −C(1+
cosΔφ. (J, (X)+2Jx(X) cos 2
(J) t' +−・・−・+2Jzs(X)cos
2 m ωtl + −・−−−−)+ sin Δφ
o(2J+(X)sinωt'+2Jx(X)sin3
ωt' ・・−・−・+2Jz, −+(X)sin(
2m-1) ωt'+...) )...
...(2) becomes. Here, C: constant J,: n-th hesocell function B (n = 0.1.2...
...) As is also clear, the light intensity I0 of the interference light L6A is CO3Δφ. The term proportional to and! 1) nΔφ
.

It contains a term proportional to . When the photoelectric conversion signal of this interference light L6A is synchronously detected by the synchronous detection circuit 22 at the drive frequency f0 from the drive circuit 2o that drives the phase modulator 22, the synchronous detection output ■1 becomes V+ = +J+(X)sinΔφo − −
It becomes f31. Here, the phase difference Δφ9 is the phase shift amount due to the Sagnac effect that occurs while the opposing lights are propagating through the annular optical path 16, and is expressed as Δφo−(4πRL/cλ)Ω .
(4), so from equation (3), the synchronous detection circuit 22
By measuring the output (2), the angular velocity Ω applied to the annular optical path 16 can be determined.

However, R: Radius of the annular optical path 16 L: Length of the annular optical path C: Speed of light λ: Wavelength of light in vacuum To manufacture an optical interference gyrometer using such an optical fiber, In addition to forming the annular optical path 16 with an optical fiber, conventionally, other optical fibers are used to connect each optical splitter/coupler 13 .
1B, polarizer 19, etc. are manufactured individually, and the ends of the optical fibers taken out from these individual optical components 13, 18, 19, etc. are fused and spliced to form an optical interference gyrometer. The light incident from rAz goes around the annular optical path 16 and forms an optical path to reach the photoelectric converter 17.

Next, a method of manufacturing such a conventional optical interference gyrometer will be described.

First, a method for manufacturing an optical splitter/coupler will be explained.

Figure 1) is a diagram showing a method of making an optical distribution coupler by fusion-drawing two optical fibers in parallel with each other while heating them.

When the first fiber 25 and the second fiber 26 are heated by a heater 27 or the like and fused and drawn, the diameter of the core of the fiber becomes narrower, and the number of evanescent waves from which light oozes increases. Furthermore, as the core spacing of both optical fibers 25 and 26 approaches each other, for example, a part of the light propagating in the first fiber 25 leaks out from the core and is transmitted to the second fiber.
It penetrates into the core of the fiber 26, that is, the first
from the optical fiber 25 to the second optical fiber 26 or from the second
It becomes possible for light to branch from the optical fiber 26 to the first optical fiber 25. Typically, there is a method in which two optical fibers are heated and stretched while being twisted, and a method in which two optical fibers are aligned and fused and stretched without twisting.

FIG. 12 is a diagram showing a method of manufacturing an optical distribution coupler in which the optical fiber is fused and stretched while being twisted. This method uses NRL no C, A, V
ILLARRUER and others have attempted to do so by passing two optical fibers 25, 26 through one set of fiber guides 28 and rotating one of the fiber guides 28 to twist the two optical fibers 25, 26. It is heated with a heater 27 to fuse and stretch.

When using Hitachi Cable, Ltd.'s elliptical jacket type polarization maintaining optical fiber as the optical fiber, the branching ratio is 4.
5-55% (K = 0.45-0.55), excess loss 0
．． 2db, crosstalk between polarization modes is -10 to -2
We have succeeded in creating a 0db optical splitter/coupler.

FIG. 13 is a diagram showing a method of fusion-stretching an optical fiber without giving it an I-backward bend. This method was tried by Mr. Kawachi and others of Nippon Telegraph and Telephone Corporation, and first, while observing the stress applying part of the polarization-maintaining optical fiber with a microscope, one of the optical fibers 26 was rotated to align the polarization plane. Figure 13 A)
, quartz glass fine powder (glass soot) 31 is attached to the part to be fused so that the two optical fibers 25 and 26 are well fused together (Fig. 13B), and then the two optical fibers 25 and 26 are bonded together. Lightly fuse the optical fibers 25 and 26 in advance (Fig. 13C)
An optical fiber 25 is installed to enable branching and coupling of fourth-order light.
．． The optical fibers 25 and 26 are heated and stretched while applying an appropriate tension, so that both optical fibers 25 and 26 are fused and the fused portion 32 is finished in a tapered shape (FIG. 13D).

The stress-applying type polarization-maintaining optical fiber 25, 26 has a stress-applying portion 36 on the cladding 35 surrounding the core 33, 34.
．． 37 is provided, and the two polarization modes H of the light propagating through the optical fiber 25.26 are
Ez", HE1+' (hereinafter referred to as HEx, HEy)
is saved. Figure 14 shows a stress-applied type polarization-maintaining optical fiber (usually PA) developed by Nippon Telegraph and Telephone Corporation.
1 is a cross-sectional view showing an example of an optical distribution coupler made using NDA (referred to as NDA). Several examples are shown of the mutual arrangement of the stressing sections 36, 37 of the two optical fibers 25.degree. 26 in order to reduce the crosstalk between the polarization modes. By arranging the stress applying parts 36°37 in this way, the excess loss is 0.2 db and the polarization mode HEx, HEy
An optical distribution coupler with a crosstalk between 115 db and -25 db has been reported.

In any of these methods, the branching ratio of the optical splitter coupler is
It can be controlled by changing heating/stretching conditions, etc. 1) As shown in the figure, light is input from the first boat 38 at one end of the first optical fiber 25, and the second optical fiber 2
No light is allowed to enter from the second port 39 at the other end, and the third boat 36 and fourth port 37 at the other end are
Output light having intensities Is and ra, respectively, is emitted. By measuring these emitted lights, the branching ratio K can be determined from equation (1).

(4) -□ ...fil Iz+14 In the case of an optical interference gyrometer using an optical fiber, an optical splitting coupler with a branching ratio K of 1:1 is required. In this case, branching ratio K =0.5. That is, the heating/stretching process may be stopped when r=r4.

Next, the optical fiber type polarizer will be explained. Optical fiber polarizers are broadly divided into coil types and polished types. FIG. 15 is a diagram showing an example of a coil type polarizer. As shown in this figure, the coil type polarizer is constructed by winding a birefringent optical fiber 43 around a cylinder 44, and has two mutually orthogonal polarization modes (HEx , HBy).

FIG. 16 is a diagram showing polarization dispersion characteristics of a birefringent optical fiber. A normal birefringent optical fiber has two polarization modes (
Both HEx and HEy are used as single polarization fibers (see point A in FIG. 16) capable of propagation. However, the 16th
As shown at point B in the figure, the birefringent fiber has an area where only one polarization mode can propagate (absolutely single polarization head area) ΔV
c exists. Therefore, in order to create a coil-type polarizer, if an absolutely single polarization head range is used as the birefringent fiber 43, the characteristics necessary for the coil-type polarizer can be obtained.

Figures 17 and 18 are diagrams showing the structure of a polishing type polarizer and how incident light Pi is polarized into output light PO.
Examples of a crystal polarizer and a metal polarizer are shown respectively. A crystal 52 or metal 53 having an appropriate refractive index is loaded onto the polished portion 51 of the core 49.These crystals 52 or metal 53 allow the direction of the light propagating through the optical fiber 47 to be Polarized waves are strongly absorbed or radiated and cannot propagate any further in the optical fiber. On the other hand, polarized waves orthogonal to the polarized waves are hardly absorbed or radiated and are not able to propagate further in the optical fiber. It can propagate through the fiber, and therefore functions as a polarizer that only allows specific polarized waves to pass through.

Metal type polarizers have been manufactured with extinction ratios of 14 to 45 db, and crystal type polarizers have been reported to have extinction ratios as high as 60 db.

FIG. 19 is a diagram showing some examples of cross-sectional views of polishing type polarizers. Figure 19A shows an example in which a side pit type polarization maintaining optical fiber is used, with two side pits 5
4 is parallel or perpendicular to the polishing surface. FIG. 19B shows an example in which an elliptical jacket type polarization maintaining optical fiber is used. In any case, in order to increase the performance of the polarizer, that is, the extinction ratio k, the principal axis of birefringence is aligned perpendicularly or horizontally to the crystal 52 or metal 53 loaded on the polished surface 51. In order to align the principal axes of birefringence of the optical fiber 47, adjustments are generally made while observing the stress-applying portion with a microscope.

Conventionally, an optical interference gyrometer uses light s as shown in Figure 10.
t1. The ends of optical fibers led out from the optical distribution coupler 13, 18, the optical fiber polarizer 19, the annular optical path 16, etc. are fusion-spliced.

FIG. 20 is a diagram showing a method of fusion splicing optical fibers. To fusion splice optical fibers, remove a predetermined length of the coating 62 from the end of the optical fiber 61 (see Figure 20).
, two optical fibers 61a.

The ends of the fiber wires 63a and 63b of the fiber wires 61b are abutted so that their optical axes are aligned (FIG. 20B), and the abutment portion 64 is heated and fused using a heater or arc discharge (FIG. 20B).
Figure C). FIG. 21 is a diagram showing an example of a process for fusion splicing by arc discharge.

FIG. 21A shows two optical fibers 61a to be fusion spliced.

61b is set at a fixing station (not shown), and fiber wires 63a and 63b are opposed to each other. Prior to fusion splicing, the end face of the fiber is smoothed by arc discharge from the discharge electrode 65 (FIG. 21B).

Next, as shown in FIGS. 21C, D, and E, the end surfaces of the optical fibers 63a and 63b are brought into contact with each other while arc discharge is generated, and the optical fibers 63a and 63b are further pushed in from the contact state by a predetermined amount. 63b connection.

FIG. 22 is a diagram showing a method for aligning the stress applying portions of two polarization maintaining optical fibers. Right angle prism 66-2
The two short sides 66a and 66b are used as reflecting mirrors, and the prism 6
Two optical fiber wires 63a and 63b are opposed to each other with 6 in between. The end faces of the first optical fiber strand 63a and the second optical fiber strand 63b are reflected in the same direction by reflecting mirrors 66a and 66b, and while images of the end faces are simultaneously observed with a microscope, the first optical fiber strand is One of the optical fibers is rotated so that the stress applying portions 67° and 68 visible on the end faces of the second optical fiber element vA63a and the second optical fiber element vA63b are aligned in the same arrangement.

FIGS. 23A and 23B are diagrams showing the field of view of the microscope before and after the stress applying portions are aligned, respectively. After alignment (Figure 23B)
indicates that the stress applying parts of the two optical fibers are aligned.

FIGS. 24A and 24B are diagrams showing splice loss and extinction ratio when polarization maintaining optical fibers are fusion spliced, respectively. This data was obtained by roughly aligning the directions of the principal axes of birefringence using the method shown in FIG. 22, then inputting light into an optical fiber, and performing optical WL adjustment for alignment.

According to this data, it is shown that the connection loss of the optical fiber is 0.03 db on average and 0.15 db at the maximum. Moreover, the extinction ratio (equivalent to Kukostoke) is 34.2 db on average, and the range of variation thereof is 30.9 db to 37.1 db.

By the way, at the point where the optical fiber is fusion-spliced, the optical fiber wire (quartz glass) 63 is exposed and is easily broken. Figure 20 is a diagram showing an example of a method for reinforcing the fusion spliced portion.

In this example, the exposed fusion spliced portions of the optical fiber elements M63a and 63b are covered with a heat-melt adhesive 69, and the outside thereof is covered with a heat-shrinkable tube 70 or the like for reinforcement.

“Problems to be Solved by the Invention” According to the conventional manufacturing method of an optical interference gyrometer, for example, the first
As shown in FIG. 0, the light emitted from the light R1), for example, the clockwise light L6A, is connected to the fusion splicing point P1° second optical distribution coupler 18. Fusion splicing point P2. Polarizer 19. Fusion splicing point P
3. First optical distribution coupler 13° fusion splicing point P4. Annular optical path 16. Phase modulator 21, fusion splicing point P5. First optical distribution coupler 13° fusion splicing point P3. Polarizer 19. The light passes through the fusion splicing point P2 and the second optical distribution coupler 18 and enters the photoelectric converter 17. That is, the light from the light source 1) is transmitted to five optical fiber fusion points P1 that connect optical components to each other.
~P5 for a total of 7 concave diameters to reach the photoelectric converter 17. In the process of fusion splicing these optical fibers, even if advanced technology is used and the work is proceeded with great care, the fusion splicing points P1 to P5 end up being optically imperfect areas. Due to its optical imperfections, several problems remain. That is, (1) Optical connection loss occurs at the part where the optical fiber is connected.

(2) It is impossible to perfectly match the birefringent axes of two optical fibers to be fusion-spliced.

■In order to reinforce the fusion spliced part, it is necessary to cover it with a covering material or the like.

There are various problems such as.

The effects of these problems on optical interference gyrometry will be explained.

Regarding item ■: Light contact V at the fusion splicing point of optical fiber
t FM loss increases the overall light loss in the optical interference gyrometer, resulting in less light reaching the photoelectric converter 17. Therefore, the S of the photodetection signal increases.
/N ratio decreases, which adversely affects the performance of the optical interference gyrometer.

Regarding item (2): Light in two mutually orthogonal polarization modes HEx and HEy propagates through the core of the optical fiber used in the optical interference gyrometer (excluding the polarizer). If the core has a perfect circular cross section and is an ideal strain-free optical fiber, the two polarization modes HEχ and HEy are mutually degenerate, and the propagation constant βX in the X direction is equal to the propagation constant βy in the Y direction. Therefore, HEx mode and HEy mode propagate at the same speed.

Also, if the core of the optical fiber is a perfect circle, HEx mode and H
No coupling occurs with the By mode. but,
Due to the imperfection of an actual optical fiber, the degeneracy between the two polarization modes is resolved, and the respective propagation constants βX, βy
are different (βX≠βy). Therefore, the two polarization modes HEx and HEy propagate within the optical fiber at different speeds. In addition, actual optical fibers are subject to uneven lateral pressure in the longitudinal direction due to the optical fiber coating, causing bends. Therefore, two polarization modes HEX
, HEy, and energy is mutually exchanged between these polarization modes.

Thus, there are two polarization states in the optical fiber,
When coupling occurs between these polarization modes, the polarization state of the emitted light changes due to frequency fluctuations of the light source, fluctuations in the temperature of the optical fiber itself, and the like. In the optical interference gyrometer, this variation in polarization state also causes a phase variation between the clockwise light L2 and the counterclockwise light L3 propagating through the annular optical B16, and therefore the angular velocity Ω is measured by the optical interference gyrometer. This also causes a decrease in accuracy.

To this end, the optical path of an optical interference gyrometer using an optical fiber uses a polarization-maintaining optical fiber in which the difference Δβ (=1βX-βyl) between the propagation constants of the two polarization modes HEx and HEV is intentionally increased. . However, even when using a polarization-maintaining optical fiber, due to alignment errors between the incident linearly polarized light and the principal axis of birefringence of the optical fiber, the propagation mode (polarization mode HEx or HEy that passes linearly polarized light)
In addition to this, optical energy may also exist in a mode orthogonal to the propagation mode (referred to as a non-propagation mode).

Also, due to the imperfection of birefringence, the double wave mode HE
Mode coupling occurs between x and HEy. Therefore, when optical fibers are configured by fusion splicing, there is a mismatch in the birefringence axes between the optical fibers, so the optical energy of the non-propagating mode is As a result, the energy due to the coupling from the non-propagating mode to the propagating mode increases, and the phase fluctuation between the clockwise light L2 and the counterclockwise light L3 propagating around the annular optical path 16 increases, resulting in optical interference. This is a cause of impediments to increasing the accuracy of angular velocity measurements.

The data shown in FIG. 24B is data indicating the magnitude of crosstalk between two polarization modes at the fusion splicing location of the optical fiber. This data is obtained by inputting linearly polarized light along the birefringence axis from one end of a fusion-spliced optical fiber.
This was determined by rotating the analyzer at the other end and measuring the intensity of the transmitted light. Maximum light i1m at this time
From ax and the minimum light amount I n+in , the crosstalk is determined by the following equation.

Crosstalk η = 10 log(I 1) in/I
max) According to this data, Cucostalk is -30
, 9 to -37.5 db.

When considered as an optical interference gyrometer using optical fiber,
In the conventional example shown in Figure 10, there are five fusion splice points for optical fibers, so the crosstalk is at least an additional 7 dB.
It is thought that the severity will worsen.

The crosstalk of the type of polarization-maintaining optical fiber used here is less than -40 db for a length of several meters. Therefore, as an optical interference gyrometer using an optical fiber, 5
Overall crosstalk is reduced to 1/2 by fusion splicing in several places.
It can be seen that 20 (30 db) or more is also worse. This deterioration of losstalk is a major problem for optical interference gyrometers that aim to improve measurement precision.

Regarding item (2): The fusion spliced portion of the optical fiber becomes weak, so it is necessary to cover it with a protective coating to strengthen it. For example, when using a currently developed fusion splicer for polarization-maintaining optical fibers, one side of the fusion splicer, centered around the fusion splice location, is used.
5c+n, the optical fiber coating must be removed over a length of 3cn on both sides for fusion splicing. Therefore, after the fusion splicing operation, a protective coating must be applied to the optical fiber over a length greater than that which was removed.

As shown in Figure 20, for this protective coating, a heat shrink tube 70 or the like is used, which impairs the flexibility of the optical fiber 61, making it impossible to bend the optical fiber in the coating 62 part with a large curvature. , it is not possible to construct an optical interference gyrometer in a small size.

"Means for Solving the Problems" This invention is intended to solve the above-mentioned problems.In this invention, by eliminating the fusion splicing points for connecting optical fibers, the fusion splicing process itself is improved. This also eliminates the process of reinforcing the fusion spliced area.

That is, in the present invention, the manufacturing process of each of these optical components is performed for continuous optical fibers including the optical splitting coupler, polarizer, annular optical path, and optical fibers that connect these optical components. It is manufactured by applying Therefore, the configuration is such that there are no fusion splicing points for optical fibers for connecting each optical component.

"Action of the invention" By eliminating fusion splicing points, the optical connection loss caused by connecting optical fibers is reduced, and crosstalk between the two polarization modes HEx and HEy does not increase. Furthermore, there are no restrictions on miniaturization that arise from reinforcing fusion splices. Therefore, according to the present invention, it is possible to provide a compact, high-performance, and low-cost optical interference gyrometer.

"Example" Next, the optical interference angular velocity meter of the present invention will be described. 1st
The figure is a diagram showing an embodiment of the optical interference gyrometer according to the present invention. 9 and 10 are shown with the same reference numerals. In this invention, the annular optical path 16 is constituted by an optical fiber, and the optical fiber that constitutes the annular optical path 16 is pulled out as it is (71, 71A), forming an entrance/exit path 14.15 to the annular optical path 16. These 2
Optical fiber 71゜7 formed with two input and output paths 14.15
1A are placed parallel to each other and fused and stretched to form the first optical distribution coupler 13. A part of the optical fiber 71 on which the first optical distribution coupler 13 is formed is formed into an arc shape with a predetermined curvature, and the fiber crund is shaved near the apex of the arc shape so that a predetermined evanescent wave is generated. Polished to A polished polarizer 19 is formed by loading crystal 52 or metal 53 onto the polished surface 51. Optical fiber 7 further extended from polarizer 19
1 and the optical fiber 72 that guides the irradiated light from the light source 1) are fused and drawn so that their parts are parallel to each other, thereby forming the second optical distribution coupler 18. The optical fiber 71 forming the second optical distribution coupler 18 is further extended and coupled to the photoelectric converter 17. Thus - a circular optical path 16 using a continuous optical fiber. First optical distribution coupler 1
By forming the optical components such as the 3° polarizer 19 and the second optical splitter/coupler 18, there is no need to fusion splice the optical fibers 71 and 72 that connect these optical components. It is now possible to construct an optical interference gyrometer without any optical interference.

In forming the optical distribution coupler 13, in order to make the optical branching ratio K a predetermined value, light is input from one end of one of the optical fibers to be fused and stretched, and the annular optical path 16 is formed.
The process of forming the optical splitter/coupler is carried out while observing the interference light that circulates around the optical fiber and returns to the end of the other optical fiber. The detailed manufacturing method will be explained later.

By the way, in the optical interference angular velocity meter of this invention, the polarizer 19
Therefore, a coil type polarizer as shown in FIG. 15 is not used. This is because the optical fiber used to improve the polarization performance of a coil-type polarizer is not used to form the annular optical path 16 due to the need to increase polarization dispersion between the HEx mode and the HEy mode. It is necessary to use an optical fiber with a structure different from a birefringent fiber with low transmission loss through which the two polarization modes HEχ and HEy can pass. Therefore, a coil-type polarizer and a continuous This means that the optical distribution coupler 13 and 18 formed of optical fibers must be fusion-spliced. This deviates from the purpose of this invention. Therefore, in the present invention, it is inevitable that the polarizer 19 be a polishing type polarizer.

FIG. 2 is a diagram showing a modification of the invention. This modified example is an example in which there is only one fusion splicing point when workability in the manufacturing process is taken into consideration.

In this modification, the optical fiber 71 that formed the annular optical path 16 is extended as it is to form the first optical distribution coupler 13 and the polarizer 19, and the optical fiber 71 further extended from the polarizer 19 and the photoelectric conversion Optical fiber 73 leading to device 17
are aligned in parallel and fused and stretched to form the second optical distribution coupler 18. That is, the optical fiber 71 extended from the polarizer 19 forms the second optical splitter/coupler 18 and is further extended toward the light source 1). In this modified embodiment, the extended optical fiber 71 and the optical fiber 72 that guides the light from the light source 1) are fusion-spliced (Pl) so that the light is guided to the annular optical path 16.

FIG. 3 is a perspective view showing an example of the structure of a modular light source. The light source 1) is generally modularized in order to make it smaller, have a longer lifespan, and improve environmental resistance. This light source module 1) includes a light emitting element 75, a Peltier element 76 for cooling the light emitting element 75, a temperature control device 77 for monitoring and controlling the temperature of the light emitting element 75,
A photodiode 78 for monitoring and controlling the output light of the light emitting element 75 is incorporated.
Usually, a small lens is used for optical coupling between the light emitting element 75 and the optical fiber 72, or the tip of the optical fiber 72 is made spherical to improve the optical coupling efficiency. FIG. 4 is a diagram showing an example of optical coupling using an optical fiber 72 with a rounded tip. The tip of the optical fiber 72 is plated with metal and its end is shaped into a sphere. This optical fiber 72 having a spherical tip is made to face the light emitting element 75 using a solder 79 or the like so that the light from the stator and the light emitting element 75 is directly introduced into the optical fiber 72. The optical fiber 7 can be made using either of these lenses or by processing the tip of the optical fiber 7.
2 and the light source module 1) are integrated.

However, if the optical fiber 72 with a spherical tip is optically coupled to the light source module 1), there is no need to use a lens.
The light source module 1) can be made smaller. Therefore, it is advantageous in making the optical interference gyrometer small.

By the way, in order to efficiently guide the light emitted from the light emitting element 75 into the optical fiber 72, the tip of the optical fiber 72 is processed into a spherical shape. An optical fiber 71 extending from the fiber 19 with a rounded tip is attached so as to face the light emitting element 75 in the light source module 1). However, attaching the optical fiber 71 extended from the polarizer 19 to the front surface of the light emitting element 75 is very inefficient. It is better to make a light source module 1) by attaching it so as to face the element 75, and then fusion splice the optical fiber 72 drawn out from this light source module 1) and the optical fiber 71 extended from the polarizer 19. Excellent workability.

Therefore, when placing emphasis on normal workability, light source 1)
and the second optical distribution coupler 18, there is one fusion splicing point P1.
Even if this happens, it is unavoidable.

However, even in this case, if the poor workability of manufacturing the light source module 1) is not to be overcome, the tip of the optical fiber 71 extended from one side of the first optical splitter/coupler 13 should be rounded. It may be attached to the light emitting element 75 surface of the light source module 1).

Therefore, considering the workability and the absence of fusion splicing parts, it can be said that the configuration shown in FIG. 1 is the most practical and favorable mounting method for the optical interference gyrometer.

"Method for Manufacturing Optical Interference Angular Velocity" Next, a specific method for manufacturing the optical interference angular velocity meter in which the constituent optical components are integrated as described above will be described.

FIG. 5 is an optical path diagram for explaining the manufacturing method of the optical interference gyrometer according to the present invention. Light I from light source 81 (first
In the figure, the light LIA (corresponding to the light that has passed through the polarizer 19) is split by the light distribution/coupling unit 82, and the clockwise light 1 cw (corresponding to L2) and the counterclockwise light 1 cc propagate through the annular optical path 16.
Both lights w (corresponding to L3) are combined by the optical distribution/coupling section 82 to form interference light, and are divided into two by the optical distribution/coupling section 82, and one of the two divided interference beams is sent to the photoelectric converter 83.
is converted into an electrical signal by Now, assuming that the phase difference between the two lights 1cw and Iccw due to propagation through the annular optical path 16 is Δφ, the light intensity I0 of the interference light received by the photoelectric converter 83 can be expressed by the following equation. That is, Ior
++Iz 2(It)””・(Iz)””cosAφ
++-141 However, (1) I: Intensity of one branched light of 1 cw of counterclockwise light I2: Intensity of one branched light of 1 ccw of clockwise light.

Next, the branching ratio of the optical splitter/coupler 82, that is, the relationship to the optical coupling coefficient will be explained.

As shown in Fig. 1), if the strength of each branched light branched by the optical splitting/coupling unit 82 is 13.14, then the optical coupling coefficient of this optical splitting/coupling unit is K = In/(Is
+ It)...It is expressed as (5). In FIG. 5, the light output I emitted from the light source 81 is the intensity of clockwise light Icw and the intensity I of counterclockwise light, if losses in the optical distribution/coupling section 83 and the annular optical path 16 are ignored.
ce- is respectively expressed as Icw=(1k) Iccw=K·■. This clockwise light re- and counterclockwise light Iccw
After propagating through the annular optical path 16, it is divided into 2
The intensities of the branched lights are 1+ and Iz, respectively, so 1,=K”l ・・・・・・・・・・・・・・・
(6) lz=(I K)”I ・・・・・
...(7) holds true. By substituting these (6) and (7) into equation (4), the light intensity of the interference light reaching the photoelectric converter 83 is obtained. can be found. That is, 1, = + [1-2K (1-K) (1+cosΔφ
) ) −...flil From this equation (8), the intensity of light ■. It can be seen that is related to the branching ratio K and the phase difference Δφ between the two lights 1cw+Icc-.

FIG. 6 is a diagram showing how the intensity Io of the interference light changes with respect to the branching ratio K in relation to the phase difference Δφ between the two lights 1cw and Iccw. When the branching ratio K is zero, only the clockwise light ■C1) circulates around the annular optical path 16, so
Intensity of light incident on the photoelectric converter 83■. is 1 regardless of the phase difference Δφ axis. It is shown that =1. However, when the branching ratio is 0.1, and the phase difference Δφ between the two lights, 1cw and Iccw, is Δφ=π, they just strengthen each other! . -1
However, when it deviates from Δφ−π, both beams IB and Ic
cw becomes destructive, and when the phase difference Δφ=0,
Io becomes the smallest. In this way, both light 1cse + I
ccw changes with a period of 2π according to the phase difference Δφ. When the optical fiber 71 constituting the annular optical path 16 is extended as it is and fused and stretched to form the light distribution coupling part 82, both lights propagating through the annular optical path 16 are 1cw+Ic.
Since the phase difference Δφ between cw is y° zero, in order to control the branching ratio K of the optical distribution/coupling unit 82, the intensity Io of the interference light is observed and the observed intensity Io is minimized. It turns out that it is better to do this.

In the optical interference gyrometer, the branching ratio of the optical splitting coupler 13.18 is l:1. In other words, it is considered ideal to set it to =0.5. Therefore, as is clear from FIG. 6, the interference light ■. It is sufficient to proceed with the optical fiber fusion/drawing process while observing the above, and proceed with the formation of the optical distribution/coupling portion 82 so as to stop the process when the intensity Io of the interference light becomes zero.

Next, we will examine and explain the manufacturing accuracy required for the optical distribution coupler when manufacturing the optical interference gyrometer according to the present invention.

If no angular velocity is applied to the annular optical path 16, the clockwise light, the counterclockwise light, few, which propagates through this annular optical path 16,
Almost no phase difference φ occurs between Iccw.
, cosΔφ#1. Therefore, Ni (8) is 1, =
Iil-4K(1-K)) ・・・・・・(9)
becomes.

By the way, in FIG. 5, when two lights that are branched at the optical splitting/coupling unit 82 and whose intensity toward the light source 81 is I+'+12'' (corresponding to light L6 in FIG. 1) interfere, the intensity of the interference light is Io' is 1o' = 1+' + If' + 2(II')""
(12'ν/ 2 ・・・・・・(Can be expressed as 101.■," and 1 are respectively the following expressions αυ, 0~1+' = K (l K) I ・・・・・・α
Since υ1 is expressed as '=K(1-K)I --------Q2), substituting these 2〇〇 and 2 for 20 gives 1o' = 2K (1-K)I( 1+cosΔφ)
-・----It becomes 031.

Generally, it is considered that the best configuration for an optical interference gyrometer is to set the branching ratio K to 0.5, but as a practical matter, the branching ratio K of the optical splitting coupler should be strictly equal to 0.5. This is extremely difficult from a manufacturing technology perspective. So let's consider this point. Showing the relationship between the loss α caused when the branching ratio K deviates from the desired value of 0.5 and the branching ratio, the loss α is = 101og4 K(1-K) −・=0
It is represented by 41. In the case of an optical interference gyrometer, in addition to the loss due to the branching ratio K of the optical splitter/coupler deviating from 0.5, there is an excessive loss of the optical splitter/coupler and a transmission loss of the optical fiber constituting the annular optical path 16. is also added.

In FIG. 5, if α is the total loss that the light from the light source 81 receives from all the optical paths from when it enters the optical fiber 71 until it reaches the photoelectric converter 83, then equations (6) and (7) can be expressed. The maximum light amount (■) shown in is reduced to 7 lights.■.

Figure 7A shows the intensity of the interference light when the branching ratio K in equation (9) is changed. FIG. In other words, FIG. 7A shows that when the phase difference Δφ=0 as shown in FIG.
Intensity of interference light■. This shows the relationship in which the minimum value of is changed depending on the value of the branching ratio.

In addition, FIG. 7B shows the amount α, which is a substantial loss caused by deviation of the branching ratio from 0.5, calculated using equation 04). For example, branching ratio K = 0.5
should be, but if there is an error of 2%, the branch ratio will be =
It is 0.49. The intensity of the interference light at this time ■. It has been shown that under the condition that the phase difference Δφ=0, the intensity is 4×10−′ times as large as the intensity Io=■ when K=O, that is, when there is no interference at all.

Currently, the accuracy of the branching ratio of optical splitting couplers formed using single mode optical fibers in practical use is 2.
~5%. Therefore, from the above, in order to keep the branching ratio K within 0.5 ±2%, the seventh
As shown in FIG. B, it is necessary to measure and discriminate a light amount as weak as 4×10−′ compared to the maximum light amount I at K−0. Therefore, first of all, the dark current of the photoelectric converter 83 for measuring the optical filtration, the back Rayleigh scattering from the optical fiber 71 forming the annular optical path 16, the offset voltage of the amplifier forming the photoelectric converter 83, etc. It is necessary to consider whether it can withstand the measurement and discrimination of the amount of light. Below, the results of examining these problems will be explained.

■ Regarding the influence of dark current of the photoelectric converter 83: In Fig. 5, the amount of light incident on the optical fiber 71 is 500 μW, and the length of the optical fiber constituting the annular optical path 16 is 1llJ+
If the transmission loss of this optical fiber is 3 db/+*, the interference light reaching the photoelectric converter 83 will be 250 μW at maximum (corresponding to light 1 in equation (7)). Here, if a silicon photodiode is used as the light receiving element of the photoelectric converter 83, when the maximum light amount 1 (= 250 μW) is converted into a current, the photocurrent is I x 10-'AA
It depends on the degree.

On the other hand, the dark current is about 5×10-”A. This value is
This value is about 10<-4> smaller than the measurement level 4XIO-11A required to measure the branching ratio K with an accuracy of 0.5±2%. Therefore, it can be seen that the dark current of the photoelectric converter 83 can be ignored.

■Regarding the backward Rayleigh scattering from the optical fiber reaching the photoelectric converter 83: The magnitude of the backward Rayleigh scattering is lXl
It is about 0-'. This value is also the signal magnitude lXl0-' of the maximum light jll. This value I It is possible to sufficiently measure and determine the ratio K.

(2) Regarding the offset of the photoelectric converter 83: If the offset voltage of the amplifier is appropriately adjusted, it is easily possible to reduce it to 10 μv or less.

Intensity of interference light in equation (9)! . If the voltage lOv is output when is maximum (that is, K=0), then the branching ratio is equal to 0.
．． 5 ±2% recognition level 4xl'O-'I is 4mV
become. Therefore, the offset voltage of 10 μι can be ignored.

As discussed above in sections (1), (2), and (2), it has been shown that it is sufficiently possible to manufacture an optical splitting/coupling device with a branching accuracy of 0.5 ±2%. Note that among the above items, the one that has the most influence is backward Rayleigh scattering, and when converted to branching ratio accuracy, its value is +/-0.5%.

In addition, the actual loss when there is an error in the branching ratio of ±2% is a very small amount of -0,0017 db, as shown in Figure 7, and can be ignored for an optical interference gyrometer. It's fine, and it doesn't really pose a problem.

Furthermore, the average value of the fusion splice loss shown in FIG. 23A is 0.03.
1db is converted into an error in the branching ratio, and from Figure 7B, ±
It turns out that it is 10%. On the other hand, since there are five fusion splicing points in a conventional optical interference gyrometer, the conversion error is ±
It can be as much as 20%.

FIG. 8 is a diagram showing another modification of the method for manufacturing an optical splitter/coupler. In this manufacturing method, the light incident on the optical fiber 71 is blinked by a chopper 85 driven by a drive circuit 84 and is applied via a lens 86 .

The electrical signal that circulates around the annular optical path 16 and is photoelectrically converted by the photoelectric converter 83 is supplied to a synchronous detection circuit 87 and synchronously detected by a chopper signal from the drive circuit 84 . Its synchronous detection output■. is Vo=C(1+2K(I K)(1
+cosΔφ))...θ C=constant.

This equation α9 has the same form as equation (8). Therefore, when manufacturing an optical distribution coupler, the light from the light source 81 is blinked and supplied by the chopper 85, the photoelectric conversion signal of the interference light is synchronously detected by the synchronous detection circuit 87, and the output signal
. It is shown that the value of the branching ratio of the optical distribution/coupling section 82 may be controlled while observing the value at the output end 88.

Even with this manufacturing method, in order to set the branching ratio to 0.5,
As in the example shown in FIG. 5, the process of fusing and stretching the optical distribution coupler may be stopped when the output voltage becomes zero.

This chopper type manufacturing method has the advantage that it is not affected by the dark current or offset voltage of the photoelectric converter 83.

Also, instead of blinking the light incident on the optical fiber 71 using the chopper 85, the amount of light emitted from the light source 81 (light output) is directly controlled, and this modulation control signal is supplied to the synchronous detection circuit 87 to allow the light to be transmitted from the photoelectric converter 83. It may be configured to synchronously detect the electric signals of.

``Effects of the Invention'' Since each optical component necessary to construct an optical interference gyrometer is formed using a single continuous optical fiber, there is no need for fusion splicing of optical fibers, which eliminates the need for time-consuming fusion. The process of splicing and reinforcing the fusion spliced parts is eliminated.
Manufacturing costs are reduced.

Since there are no fusion splices, performance does not deteriorate due to fusion splicing. Further, since there is no need for a reinforcing coating, the flexibility of the reinforcing portion of the optical fiber is not lost, and therefore, the optical interference gyrometer can be constructed in a small size.

In this way, it is possible to make a low-cost, high-performance, and compact optical interference gyrometer, so it is possible to make not only an inexpensive optical interference gyrometer that can be used in robots, cars, etc., but also a high-precision inertial navigation system (INS) for flying objects. It can be applied and manufactured to a wide range of optical interference gyrometry, including high-precision optical interference gyrometry that can also be used for satellite attitude control.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing a configuration example of an optical interference gyro meter according to the present invention, and FIG. 2 is a diagram showing a modified configuration example of the optical interference gyro meter according to the present invention. FIG. 3 is a perspective view showing an example of the configuration of a light source module, FIG. 4 is a diagram showing the relationship between a light emitting element and a spherical fiber in the light source module, and FIG. 5 shows a method of manufacturing an optical interference gyrometer according to the present invention. FIG. 6 is an explanatory diagram, and FIG. 7A is a diagram showing the relationship between the phase difference and the intensity of interference light with respect to the branching ratio in the manufacturing method of the optical distribution coupler used in the optical interference gyrometer according to the present invention. , B relate to the method of manufacturing an optical distribution coupler used in an optical interference gyrometer according to the present invention, and show the relationship between the branching ratio and the intensity of interference light, and the relationship between the branching ratio and the substantial loss due to a branching ratio error, respectively. table,
FIG. 8 is a diagram for explaining the manufacturing method of an optical interference angular velocity meter according to the present invention, FIG. 9 is a diagram showing the basic configuration of an optical interference angular velocity meter, and FIG. 10 is a diagram showing the configuration of a conventional optical interference angular velocity meter. Figures illustrating an example. Figure 1) is a diagram showing a method for manufacturing an optical distribution coupler by fusion-stretching optical fibers, and Figure 12 is a diagram showing a method for manufacturing an optical distribution coupler by fusion-stretching optical fibers while twisting them. Figure 13 is a diagram illustrating the manufacturing method of an optical distribution coupler in which an optical fiber is fused and stretched without twisting, and Figure 14 is a diagram illustrating a method of manufacturing an optical distribution coupler in which an optical fiber is fused and stretched without twisting. 15 is a diagram showing the structure of a coil type polarizer, Figure 16 is a diagram showing the polarization dispersion characteristics of a birefringent fiber, and Figure 17 is a diagram showing the polarization dispersion characteristics of a crystal type polarizer. Figure 18 is a diagram explaining the configuration of the metal polarizer, Figure 19 is a diagram explaining the configuration of the metal polarizer.
20 is a process diagram showing a method for fusion splicing optical fibers, FIG. 21 is a process diagram showing a method for fusion splicing optical fibers by arc discharge, Fig. 22 shows a method for aligning the principal axes of birefringence in fusion splicing of polarization-preserving optical fibers, and Fig. 23 shows the alignment of stress applying parts before and after alignment of the principal axes of birefringence of polarization-maintaining optical fibers. FIGS. 24A and 24B are diagrams respectively showing the contact VttR loss and extinction ratio accompanying fusion splicing of polarization maintaining optical fibers. 1): light source, 12: optical fiber, 13: (first) optical distribution coupler, 14, 15: input/output path, 16: annular optical path, 1
7: Photoelectric converter, 18: Second optical distribution coupler, 19: Polarizer, 21: Phase modulator, 22: Synchronous detector, 23:, 2
4: , 25: First optical fiber, 26: Second optical fiber, 27: Heater, 28: Fiber guide, 29: Microscope, 31 Niglass fine powder, 32: Fusion part, 33, 34
: Core, 35: Clad, 36: Stress applying part, 37
: Stress applying part, 38, 39.40.41 : Port, 4
3: birefringent optical fiber, 44: cylinder, 45 glass block, 46: groove, 47: optical fiber, 48: cladding,
49: Core, 51: Polished portion, 52: Crystal, 53: Metal, 54: Stress applying portion, 61: Optical fiber, 62: Coating, 63: Fiber wire, 64: Butt portion, 65: Discharge electrode, 67 .68: Stress applying part, 69: Heat melt adhesive (protective coating), 70: Heat shrink tube (protective coating),
71.72.73: Optical fiber, 75: Light emitting element, 7
6: Peltier element, 77: Samista, 78: Photodiode, 81: Light source, 82: Optical distribution/coupling unit, 83: Photoelectric converter, 84: Drive circuit, 85: Tift bar, 86: Lens, 87: Synchronous detection circuit, 88: Output end, P1 to P5:
Fusion joints. Patent applicant: Japan Aviation Electronics Industry, Ltd. Representative: Takuyuki Kusano 3
Record! 74 Fig. O 5 Fig. O 6 Fig. 1 τ At O, phase difference Δφ is off. > Figure 19 (A) (B) Figure 20 Figure 21 Figure 23 (A) (B), +I''F Figure 24 (A) (marked

Claims (4)

[Claims] (1) a light source; an annular optical path made of an optical fiber that goes around at least once; the light from the light source is distributed to the annular optical path as clockwise light and counterclockwise light, and the light is propagated through the annular optical path; a first optical distribution/coupling means for combining the clockwise light and the counterclockwise light into one light and causing interference; and a photoelectric conversion means for converting the interference light from the first optical distribution/coupling means into an electrical signal. In the optical interference gyrometer having at least the following, parts of the optical fibers constituting the two input/output paths to the annular optical path are aligned, fused and stretched to constitute the first light distribution/coupling means. Optical interference gyrometer. (2) a light source; an annular optical path made of an optical fiber that goes around at least once; the light from the light source is distributed to the annular optical path as clockwise light and counterclockwise light, and the light is propagated through the annular optical path; a first light distribution/coupling means for combining the right-handed light and the left-handed light into one light and causing interference; a second optical distributing/coupling means for distributing the interference light from the optical distributing/coupling means; a photoelectric conversion means for converting the interference light distributed from the second optical distributing/coupling device into an electrical signal; and the first optical distributing/coupling means. In the optical interference gyrometer, the optical interference gyrometer includes at least a polarizer disposed between the second optical distribution and coupling means and which attenuates the polarization of the elliptically polarized wave in a predetermined direction. A portion of the optical fibers constituting the optical fibers are aligned, fused and stretched to constitute the first optical distribution coupler, and one of the optical fibers constituting the input/output path is passed through the first optical distribution coupling means. The optical fiber is further extended, a part of the extended optical fiber is made into an arc shape, the apex of this arc-shaped optical fiber is shaved off, and the shaved portion is loaded with metal or crystal to form the polarizer. , a portion of an optical fiber further extended from the polarizer and an optical fiber that guides light from the light source or the photoelectric conversion means are aligned, fused and stretched to constitute the second light distribution and coupling means. Optical interference gyrometer. (3) a light source; an annular optical path made of an optical fiber that goes around at least once; the light from the light source is distributed to the annular optical path as clockwise light and counterclockwise light, and the light is propagated through the annular optical path; a first light distribution/coupling means for combining the clockwise light and the counterclockwise light into one light and causing interference; and a photoelectric conversion means for converting the interference light from the first light distribution/coupling means into an electrical signal. , In a method for manufacturing an optical interference gyrometer having at least the following, a part of the optical fibers constituting two entrance/exit paths to the annular optical path are aligned with respect to an annular optical path made of optical fibers that goes around at least once. The optical distribution coupler is fused and stretched to form an optical distribution coupler, and in the fusion and stretching process of the optical distribution coupler, the interference light obtained from the optical distribution coupler is converted into an electrical signal, and the electrical signal is converted into an electrical signal. 1. A method for manufacturing an optical interference gyrometer, characterized in that the fusing/stretching process of the light distribution/coupling means is stopped when the light becomes zero. (4) a light source; an annular optical path made of an optical fiber that goes around at least once; and a light source that distributes the light from the light source into the annular optical path as clockwise light and counterclockwise light and propagates the annular optical path. a first optical distribution/coupling means for combining the clockwise light and the counterclockwise light into one light and causing interference; and a photoelectric conversion means for converting the interference light from the first optical distribution/coupling means into an electrical signal. A method for manufacturing an optical interference gyrometer having at least the following steps, wherein the light from the light source is blinked by blinking or amplitude modulating means and is supplied to the annular optical path via a light distribution/coupling means, and the photoelectric conversion signal is A light characterized in that synchronous detection is performed using a flickering signal from the blinking or amplitude modulating means, and the fusing/stretching process of the optical distribution/coupling means is stopped when the synchronous detection output becomes approximately zero. A method for manufacturing an interferometric gyrometer.