JPH0820576B2 - Method for fusion splicing of multi-core optical fiber ribbons - Google Patents

Description

The present invention relates to a fusion splicing method for a multi-core optical fiber tape core wire and an apparatus therefor.

(Prior Art) When connecting optical fibers, the end faces of the optical fibers are melted,
The method of fusion splicing is often used. However, in fusion splicing of multi-fiber optical fiber ribbons,
When the end faces of the fibers are brought into contact with each other, the end face spacing becomes uneven as shown in FIG. 4, and this is a very problematic point in that the fusion conditions are set so that all the fibers can be fused. FIG. 4 shows a case where the fiber strands of the n-core optical fiber tape are butted against each other, and F ₁ , F ₂ , ...
, F _n , F ₁ ′, F ₂ ′, ... F _n ′ represent the fiber strands of the n-core tape fiber, and D ₁ , D ₂ , ..., D _n represent the respective end face intervals.

FIG. 5 shows the result of a butt experiment of a 10-fiber optical fiber using a multi-fiber optical fiber end treatment fixing jig. The horizontal axis D _a is a value obtained by subtracting the minimum end face distance D _min from the end face distance D _i of each fiber, and D _a = D _i −D _min (1 ≦ i ≦ 10) (8) (8) . The total number of data is 250 pairs. From this result, even if the fiber end treatment fixing jig is used, it averages 11 μm.
It can be seen that the unevenness of the end face spacing of m, maximum 50 μm occurs.

The probability density function of D _a can be approximated by the following equation.

P (D _a ) = 0.025 (D _a +1) exp [− (D _a +1) /6.4]
(9) In order to deal with such variations in end face spacing, conventionally, fusion splicing of multi-core fibers has been performed by the procedure shown in FIG.

First, after pressing the optical fibers 1 having uneven end faces with the soft clamp 2 [Fig. 6 (a)], the optical fibers 1 are abutted against the abutting plate 3 [Fig. 6 (b)], and the end faces are After forcibly adjusting the unevenness of the optical fibers 1, the optical fibers 1 are fixed to each other by the hard clamp 4 [Fig. 6 (c)], and discharge fusion is performed [Fig. 6 (d)]. The end face of the fiber is often scratched when the fiber is abutted against the abutting plate 3, and there is a drawback that bubbles are likely to enter during fusion.To avoid such a risk, the unevenness of the end face spacing is not corrected. However, as described above, it is difficult to carry out an appropriate embedding because the end face spacing varies.

FIG. 7 shows the relationship between the splicing loss α _{s and} the fiber pushing amount S at the time of fusion, when a fusion splicing experiment of single-core optical fibers was performed. The fiber pushing amount S here is the amount that the fiber is pushed after the both end faces of the optical fiber are in contact with each other.

From Fig. 7, the average connection loss ▲ ▼ can be formulated as in equation (10).

From this result, the average connection loss α _s is the fiber pushing amount S
It can be seen that it increases in proportion to.

FIG. 8 shows the average splice loss ▲ ▼ with respect to the fiber pushing amount S ₀ when the 10-fiber optical fiber is fused by using the equation (9) when the end face spacing shown in FIG. 5 varies.
And the result of calculating the connection failure rate R _f is shown.

Here, S ₀ is the pushing amount for the fiber pair with the smallest end face spacing.

Therefore, the fiber pushing amount S _i for each core wire is S _i = S ₀ −D _a = S ₀ (D _i −D _min ) where F is the sum of the fiber advance amounts of the opposing n-core optical fiber pairs. = F−D _i (7)

Here, the fiber pushing amount S ₀ ′ of the fiber pair having the maximum end face spacing can be expressed by the equation (7) ′.

S ₀ ′ = F−D _max (7) ′ where D _max is the maximum end face spacing of the n-core optical fiber pair.

Therefore, if S ₀ ′ <0, it means that the end faces of the fiber pair are not in contact with each other, and the connection of the tape core wire has failed.

From the results of FIG. 8, in order to reduce the splice failure rate of 10 optical fiber ribbons to 5% or less, S ₀ > 50 μm must be set, which is higher than that of single fiber splicing. It can be seen that there is a drawback that the loss in dB value increases by 2-3 times. In this case, the failure rate is 5% and the average connection loss is 0.06 dB.

(Problems to be Solved by the Invention) The present invention relates to fusion bonding of a multi-core optical fiber tape core wire with a small connection loss and a small connection failure rate when fusion-bonding the multi-core optical fiber tape core wires. It is to provide a connection method and its device.

(Means for Solving the Problems) In the present invention, when fusion splicing a multi-core optical fiber ribbon, the end face spacing is automatically set by using an image processing technique, and all the fiber cores are set. The distance between the end faces of the line pair is measured, and from the result, the fiber pushing amount and the optical fiber pulling amount that minimize the average splice loss are uniquely determined by calculation.

That is, in fusion splicing method of multi-core optical fiber ribbon, the end face spacing D _1, D ₂ of the n-core optical fiber pair of opposed ribbon in a fixed V-groove, ..., and D _n, the maximum in this The end face spacing D _max and the minimum end face spacing D _min are respectively measured, and thereafter, while melting the respective optical fiber end faces, when advancing and joining at the same time, after the end faces of the optical fiber pair whose end face spacing is D _max are joined, 'as, the sum F of advancement of each fiber among n-core optical fiber pair of _{opposed, F = D max + S 0} ' S 0 the fiber pushing amount being pushed and (1), after the end interview if the connection With the portion heated, the n-core optical fiber is pulled back by the pullback amount of the optical fiber determined by the calculation, and fusion is completed.

Therefore, compared to the conventional method in which only constant pushing is performed and the fiber is not pulled back in any state of the distribution of the end face spacing, the conventional push-in / pull-back is performed in the point of performing the optimum push-in / pull-back in some cases. Basically different from the method.

FIG. 1 shows the structure of an apparatus for abutting the n-core optical fiber 1 in the present invention, and 6 is an illumination lamp.
7 is an electrode rod, 8 is an objective lens, 9 is an image pickup element (CCD element), 10 is an enlarged image pickup of the n-core optical fiber 1 connected on the CCD element 9, 11 is an n-core V groove, and 12 is a fiber coating. is there.

First, the end face spacing of the n-core optical fiber 1 is set as follows. The n-core optical fiber 1 is cut, set in a fusion splicer, and fed from the left and right. At this time, the magnified image 10 of the n-core optical fiber 1 is connected to the CCD element 9 via the objective lens 8. The image 10 has a shape exactly as shown in FIG. While moving the n-core optical fiber 1 closer to each other, the data of the imaged 10 is connected to the CCD element 9
Treated by PU, end face spacing D ₁ ~ of each fiber pair
Measure D _n . While measuring D _{1 to} D _n , the end face distance D _i (1 ≦ i ≦ n) of any one pair of fibers has a specific value a μm.
When it reaches (for example, 10 μm), the feeding of the n-core optical fiber 1 is stopped. At this time, the maximum value D _max and the minimum value D _min are detected from D _{1 to} D _n , and the fiber pushing amount S ₀ of the optical fiber _ribbon is calculated as S ₀ = D _max −D _min ( 11) The indentation amount S _i of the _i -th core wire pair is S _i = D _max −D _i (7) ″. As a result, S _i ≧ 0 for all the core wire pairs, and the failure rate becomes zero. If this is left as a whole, the number of fiber core wires that have been pushed in more than necessary will increase, the strain at the connection point will increase, and the average connection loss will increase. In order to achieve this, when the optical fiber ribbons are fused, the work of pulling back the fiber ribbons is performed subsequent to the pushing in to reduce the splice loss.

FIG. 2 shows the results of measuring the connection loss α _s by changing the pushing amount S and the pulling back amount L when the single core wire is fused.
From this, it can be seen that pulling back the fiber core wire has the effect of reducing the strain at the connection part and reducing α _s, and the optimum value of L should be approximately the same as S from the experimental results. .

In addition, the relationship between the connection loss α _s after fusion bonding and S, L is C _f ,
C _g , C _h , C _i , C _c , and C _{d are} constants, and can be experimentally determined in advance from equation (5) ′ and equation (12) as in equation (5).

α _s (S, L) = G _a (S) L ² + G _b (S) L + G _c (S)
(5) Therefore, considering the case of connecting an n-core optical fiber pair whose end face spacing is D _{1 to} D _n , the sum of the connection loss of the n-core optical fiber core wires is calculated by the formula (5), ( It can be expressed as follows using 12).

With this, L may be determined so that the average connection loss {circle around (1)} can be minimized. In Expression (3) ′, the value of L that minimizes the average connection loss ▲ ▼ is obtained by differentiating Expression (3) ′ with L and is as follows.

From these results, if the results of equations (11) and (6) are used as the optimum value L ₀ of the fiber push amount S ₀ and the pullback amount when fusing the n-core optical fiber, the average splice loss It can be minimized.

FIG. 3 shows the cumulative probability of splice loss when the present invention is used in the case of connecting 10-fiber optical fiber ribbons and the variation in the end face spacing D _a is as shown in FIG. ) And the case of using the conventional method (II) are computer simulation results.

From this, it can be seen that the case of using the method of the present invention has a much higher probability of achieving a low-loss connection than the case of using the conventional method, is effective, and has a failure rate of zero. Can be

(Effects of the Invention) As described above, in the present invention, when fusion splicing of multi-core optical fiber tape cores, butting of the fiber cores is performed using an image processing technique, and further, each fiber core pair of By measuring all the end face distances and calculating the fiber pushing amount and the pullback amount at the time of fusion based on the result, the probability of connection failure is zero, and by pulling back the fiber by the optimum amount, There is an advantage that the connection loss can be reduced as compared with the conventional method.

[Brief description of drawings]

FIG. 1 is a principle diagram of an end face spacing matching method when the present invention is carried out, FIG. 2 is a diagram showing a relation of connection loss with respect to a pullback amount, and FIG. 3 is a case where a conventional method and the method of the present invention are used. Fig. 4 shows the cumulative probability of splice loss in Fig. 4, Fig. 4 shows a state in which n-core optical fiber pairs are abutted, and Fig. 5 shows the frequency distribution of end face spacing D _a when a 10-core optical fiber pair is abutted. Fig. 6, Fig. 6 (a), (b), (c), and (d) are views showing a fusion splicing procedure of a multi-core fiber core wire that has been conventionally used, and Fig. 7 is a diagram showing fusion of optical fiber core wires FIG. 8 is a diagram showing the relationship between the fiber indentation amount and the connection loss when connecting / disconnecting. FIG. 8 is a diagram showing the relationship between the average intrusion loss and the connection failure rate with respect to the fiber indentation amount when the end face spacing frequency distribution shown in FIG. 5 occurs. Is. F _{1 to} F _n , F ₁ ′ to F _n ′ ... Optical fiber strands D _{1 to} D _n …… Optical fiber end face spacing 1 …… n-core optical fiber 2 …… Soft clamp 3 …… Abutting plate, 4 ... Hard clamp 5 ... Arc discharge, 6 ... Illumination lamp 7 ... Electrode rod, 8 ... Objective lens 9 ... Imaging device (CCD device) 10 ... Enlarged imaging of multi-core optical fiber 11 ... n-core V groove, 12 ... Multi-core fiber coating

Continued Front Page (72) Toshio Oyanagi, Toshio Oyanagi, Tokai-mura, Naka-gun, Ibaraki Prefecture 162 Shirahane, Shirahane, Nippon Telegraph and Telephone Corporation, Ibaraki Telecommunications Research Institute (72) Mitsuru Miyauchi, Tokai-mura, Naka-gun, Ibaraki Prefecture Shirahone 162 Shirane, Nippon Telegraph and Telephone Corporation, Ibaraki Research Institute of Electrical Communication (56) Reference JP-A-55-130508 (JP, A) JP-A-57-135909 (JP, A) JP-A-59- 160113 (JP, A) JP-A-60-111206 (JP, A)

Claims (2)

[Claims] 1. A coating of a set of tape cores to be connected which collectively cover n optical fiber strands, wherein n is a positive integer, and the coating is removed, and n optical fiber strands in the coating stripping section are removed. And then a set of the cut n-core optical fiber ribbons are opposed to each other on the fixed V-groove, and the electrodes are discharged to collectively fuse the multi-fiber optical fiber ribbons. In the connection method, the end face intervals D ₁ , D ₂ , ..., D _n of the n-core optical fiber pairs of the tape core wires facing each other on the fixed V groove, and the maximum end face interval D _max and the minimum end face interval D among them _min , respectively, and then, while melting the respective optical fiber end faces, when advancing and joining at the same time, the fiber pushing amount further pushed after joining the end faces of the optical fiber pair having an end face spacing of D _max is S, Pull the optical fiber after pushing When the return amount is L, the relation with the splice loss α _s after fusion is experimentally obtained beforehand as α _s = H (S, L), and the splice loss of the n-fiber optical fiber estimated from this is estimated. Sum of L condition that minimizes From the above, the optimum value L ₀ of the pullback amount is obtained, and the pullback amount is set to L ₀ for each connection, and the fusion splicing method for a multi-fiber optical fiber ribbon. 2. C _f , C _g , C _h , C _i , C _c and C _d are constants, and G _a (S) = C _f S + C _g G _b (S) = C _h S + C _i G _c (S ) = C _c S + C _d , the relationship between the splice loss α _s and S and L after the fusion is completed is α _s (S, L) = G _a (S) L ² + G _b (S) L + G _c (S) for each time you connect And S _i = F−D _i are used to determine the pullback amount L ₀ of the optical fiber that minimizes the average splice loss, and the multicore optical fiber ribbon according to claim 1. Fusion splicing method.