TWI723388B - Semiconductor laser inspection device and semiconductor laser inspection method - Google Patents

Abstract

The object of the present invention is to detect the zigzag with high accuracy without omission. The present invention includes the evaluation condition setting section 20, regulate the range distinguished waveform data F ₁ to the drive current I plural evaluation field R, the evaluation field R of the plurality of specifications are set evaluation condition; characteristic evaluation unit 40, based on the set For each of the plural evaluation areas R, calculate the index showing the degree of sharp curvature, that is, the tortuosity K, _{for each of the divided waveform data F 1, and compare it with the reference value Ks to evaluate the semiconductor laser 90} Characteristics. The evaluation condition setting unit 20 is configured to regulate the range of the plural evaluation areas R so that each of the plural evaluation areas R has a range that overlaps with at least one of the other evaluation areas R.

Description

Semiconductor laser inspection device and semiconductor laser inspection method

This application is related to semiconductor laser inspection devices and semiconductor laser inspection methods.

In the inspection of semiconductor lasers, a method is used to determine whether there is a defect (for example, refer to Patent Document 1) based on whether there is a sharp bend (also called a kink) in the IL waveform showing the drive current-light output characteristics. ). At this time, a zigzag detection device has been proposed, which divides the IL waveform into a plurality of sections by the drive current, and uses the approximate curve obtained from each divided section to detect the zigzag position, so that even small zigzags can be more accurately The zigzag position is detected (for example, refer to Patent Document 2).

Advanced technical literature

Patent Document 1: Japanese Patent Application Laid-Open No. 2-96671 (bottom right column on page 2 to bottom right column on page 4, figure 1 to figure 5)

Patent Document 2: Japanese Patent Application Laid-Open No. 8-29291 (paragraphs 0013 to 0015, Figure 1 to Figure 2)

However, when the I-L waveform is divided, if a zigzag occurs in the boundary between the divided ranges, the resulting zigzag may not be detected. Of course, it can be suppressed by reducing the variation range of the drive current to obtain small data, but in the mass production line, the time spent on inspection of each component is limited. Therefore, in practical applications, it is very difficult to detect the tortuosity with high accuracy without omission.

This application is a technology disclosed to solve the above-mentioned problems, and the purpose is to obtain a semiconductor laser inspection device and a semiconductor laser inspection method that can detect windings with high accuracy without omission.

The semiconductor laser inspection device disclosed in the present application includes: a measurement unit that injects current into the semiconductor laser to be inspected and drives it, and then measures the light output from the driven semiconductor laser; a waveform data generation unit uses the measurement The measured value output by the semiconductor laser generates waveform data showing the drive current-light output characteristics of the semiconductor laser; the evaluation condition setting section regulates a plurality of evaluation fields that distinguish the waveform data by the range of the drive current. The evaluation conditions are set for each of the plural evaluation areas; the characteristic evaluation unit, based on the set evaluation conditions, calculates an index showing the degree of sharp curvature for each of the plural evaluation areas for each of the plural evaluation areas, and the evaluation The standard value set in the conditions is compared to evaluate the characteristics of the semiconductor laser. The evaluation condition setting unit constitutes a range that regulates the plurality of evaluation fields, and each of the plurality of evaluation fields has a range that overlaps with at least one of the other evaluation fields.

The semiconductor laser inspection method disclosed in this application includes: a waveform data generation step, injecting current into the semiconductor laser of the inspection object and driving it, and then using the measured value of the light output from the driven semiconductor laser to generate a display The waveform data of the driving current-light output characteristics of the semiconductor laser; the evaluation condition setting step is to standardize a plurality of evaluation areas of the waveform data divided by the range of the driving current, and set the evaluation conditions for each of the standardized evaluation areas ; Characteristic evaluation step, according to the set evaluation condition, for each of the plural evaluation fields, calculate an index showing the degree of sharp curvature for each of the divided waveform data, and compare it with the reference value set in the evaluation condition, Evaluate the characteristics of the semiconductor laser. When the range of the plural evaluation areas is regulated, the plural evaluation areas will each have a range that overlaps with at least one of the other evaluation areas.

According to the semiconductor laser inspection device or the semiconductor laser inspection method disclosed in the present application, since the repeating field is provided and the presence or absence of the zigzag is determined for each divided evaluation area, the zigzag can be detected with high accuracy without omission. .

Implementation Type 1

Figures 1 to 4 are used to illustrate the semiconductor laser inspection device of the first embodiment and the semiconductor laser inspection method carried out by it. Figure 1 shows the semiconductor laser in which the inspection object is set in the semiconductor laser inspection device. The block diagram of the structure in the state of being fired. Figure 2 shows the operation when inspecting the semiconductor laser in the semiconductor laser inspection device, that is, the flow chart of the semiconductor laser inspection method. In addition, Figure 3 shows the specifications of the evaluation field when inspecting semiconductor lasers, the waveform data that constitutes the I-L waveform generated from the measured value, the differential curve of the waveform data, and the graph form of the approximate curve processed by the differential curve. Figure 4 is a diagram in the form of a graph to illustrate the method of calculating the differential value from the value of each minute interval of the I-L waveform. In addition, Figure 5 is a diagram for explaining the ideal I-L waveform and the actual I-L waveform including atypical twists and suspected twists in the form of a graph.

Hereinafter, the semiconductor laser device and the semiconductor laser inspection method of Embodiment Mode 1 of the present application will be described with reference to the drawings. As shown in Fig. 1, the semiconductor laser inspection device 1 has a measurement unit 50 that drives a semiconductor laser 90 to be inspected and measures the light output, and is based on the I-L waveform data showing the drive current-light output characteristics ( Waveform data F ₁ ) to check whether it is good or bad. In addition, the description of the structure of the replacement and installation of the semiconductor laser 90 is omitted, and the structure for measuring and evaluating the installed semiconductor laser 90, that is, the structure for executing the semiconductor laser inspection method, will be described below.

In the semiconductor laser inspection apparatus 1, in order to set measurement conditions and evaluation conditions in accordance with the specifications of the semiconductor laser 90 to be inspected, a model discriminating unit 10 for discriminating the model of the semiconductor laser 90 to be installed is provided. Then, a waveform data generation unit 30 is provided, and the above-mentioned waveform data F ₁ is generated based on the measurement value output from the measurement unit 50. In addition, an evaluation condition setting unit 20 is provided to set the evaluation conditions corresponding to the model determined by the model determining unit 10. The characteristic evaluation unit 40 is also provided to process the waveform data F ₁ according to the set evaluation conditions to evaluate the semiconductor laser 90. The model discriminating unit 10 discriminates the model of the semiconductor laser to be measured based on the information from the input I/F not shown, or the shape and ID number of the semiconductor laser 90 set, and then displays the discriminated model information It is output to the evaluation condition setting unit 20.

Evaluation condition setting unit 20 has a measurement condition setting unit sets the driving current of the semiconductor laser 90 of the scope of the measurement conditions of 21, to regulate the driving current range specification section to distinguish field ₁ Evaluation Evaluation of R F waveform data field 22, setting The evaluation criterion setting unit 23 of the evaluation criterion for each evaluation area R. The measurement condition setting unit 21 sets measurement conditions such as the drive current range, current change pattern, and measurement interval in accordance with the model information obtained from the model discrimination unit 10, and outputs the measurement conditions to the evaluation field specification unit 22 and the laser drive unit 51.

The evaluation area specification unit 22 will specify a plurality of evaluation areas R that distinguish the I-L waveform by the range of the drive current I, and treat them as areas for calculating the tortuosity K. At this time, by the repetitive range generating unit 22a, in order to prevent the partition from being interrupted, a repetitive range is generated so that there is a repetitive range between the evaluation areas R. The evaluation criterion setting section 23 sets evaluation criteria such as evaluation methods and thresholds for each evaluation field R regulated by the evaluation field specification section 22, and outputs information on the set evaluation criteria to the characteristic evaluation section 40 (waveform data plus announcement 41, zigzag Detection section 42).

The measurement unit 50 has a laser drive unit 51, and drives the semiconductor laser 90 to be inspected by injecting current into it. The measuring unit 50 also has a light output measuring unit 52, which measures the light output from the driven semiconductor laser 90. The laser driving unit 51 drives the installed semiconductor laser 90 based on the measurement conditions output by the measurement condition setting unit 21. The light output measuring section 52 measures the light generated by the driven semiconductor laser 90. The value of the drive current I output by the laser drive unit 51 and the measured value of the light output P output by the light output measurement unit 52 are output to the characteristic evaluation unit 40 (waveform data generation unit 30) to generate the _{The waveform data F 1} of the I-L waveform of the semiconductor laser 90.

The waveform data generating unit 30 generates information about the driving current I output by the laser driving unit 51 and the information about the measurement timing, and the information about the measured value of the light output P output by the light output measuring unit 52, and generates the information that constitutes each semiconductor laser 90 _{Waveform data F 1 of} I-L waveform.

The characteristic evaluation unit 40 includes a waveform data processing unit 41 that performs calculation processing and processing _{on the waveform data F 1 generated by the waveform data generation unit 30.} The characteristic evaluation unit 40 includes a zigzag detection unit 42 that calculates the zigzag rate K based on the processing data and detects the zigzag. Waveform data processing unit 41 generates the waveform data generation section 30 waveform data F _1, 20 according to the set evaluation condition evaluation condition setting unit, by the evaluation of each field R, and calculation processing to distinguish between the waveform data F _1, there is performed the evaluation Necessary data processing. The zigzag detection unit 42 calculates the zigzag rate K based on the evaluation criteria set by the evaluation condition setting unit 20 for the data of each evaluation area R processed by the waveform data processing unit 41 and determines whether there is a zigzag (curvature detection).

According to the above-mentioned structure, the operation will be described with reference to the flowchart in FIG. 2. After the semiconductor laser 90 that is the inspection target is set in a predetermined position of the semiconductor laser inspection apparatus 1, the mobility determination unit 10 determines the semiconductor laser 90 to be set (step S100 ). The measurement condition setting unit 21 sets the measurement conditions such as the drive current range and measurement interval based on the information of the model such as the rated current value, the maximum current value, and the threshold current Ith (step S110), and outputs the information of the set measurement conditions to the mine The radio drive unit 51 and the evaluation field specification unit 22. According to the measured measurement conditions, the laser drive unit 51 drives the semiconductor laser 90, and the light output measurement unit 52 measures the light output P from the semiconductor laser 90 (step S120).

The value of the driving current I and the measured value of the light output P are output to the waveform data generating unit 30 to generate the waveform data F ₁ (step S130). For example, the rated current of 150mA, the drive current is assumed to be set in the range of 0 ~ 200mA, the measurement interval is set at 0.5mA measured measurement conditions, it will produce a waveform as shown in FIG. 3 of data F _1. The horizontal axis of the graph is the drive current I (mA) of the semiconductor laser 90, and the vertical axis on the left is the light output P (mW). In addition, the vertical axis on the right is the differential value D (W/A), and the upper side of the frame indicates the range of the standard evaluation area R, which will be explained later.

On the other hand, in the evaluation condition setting unit 20, the measurement conditions are set and the evaluation conditions are also measured. In the first stage of the evaluation conditions, the evaluation area specification unit 22 regulates a plurality of evaluation areas R that _{distinguish the waveform data F 1 in the range of the drive current I.} First, the upper limit U and the lower limit L of the object of specific evaluation are set. The lower limit L is set to a value slightly exceeding the threshold current Ith obtained from the model information of the semiconductor laser 90, and the upper limit U is set to the maximum value of the driving current range set by the measurement condition setting unit 21.

In this example, the threshold current Ith plus the current value of 3 mA is set as the lower limit L, and 200 mA, which is 50 mA higher than the rated current value (for example, 150 mA), is set as the upper limit U. The lower limit L is set to a value higher than the critical value current Ith because there is noise in the I-L waveform near the critical value current Ith, and this prevents the noise from being detected as a tortuous defect.

Next, set the distinguishing point S, which is used to standardize the plural evaluation areas R. In this embodiment, as the distinguishing point S for determining the evaluation area R, it is set at 110 mA which is equivalent to 73% of the rated current value. If only the division point S is set, the evaluation area R is divided into only the area from the lower limit L to the division point S and the area from the division point S to the upper limit U, as disclosed in Patent Document 2. However, in this embodiment, after the area is divided by the division point S, the overlapped area is generated by the overlapped area generating unit 22a, so that the divided areas overlap with each other.

Specifically, as shown in the upper part of the frame in FIG. 3, the area from the lower limit L to the division point S is directly standardized as the evaluation area R _N. On the other hand, the field for distinguishing between a point S up to the upper limit U, and the overlapping areas will produce duplicate the full range of evaluation R _N art, and standardized as to limit the expansion of the field until the evaluation L R _W. That is, the lower limit L of Ith+3mA to the distinguishing point S of 110 mA is standardized as the evaluation area R _N , and the lower limit L of Ith+3 mA to the upper limit U of 200 mA is standardized as the evaluation area R _W (step S200 ).

Next, the evaluation criterion setting unit 23 sets the evaluation method and evaluation criterion for detecting the tortuosity for each evaluation area R. Type in the present embodiment, the two R together with the field evaluation, the waveform data is calculated differential curve F ₁ F _2, F differential curve approximation method of least squares curve F _{2. 3,} curve according to the differential value of the approximation curve F ₂ and F Calculate the tortuosity rate K (indicating the sharp curvature on the _{waveform data F 1 ) with the value on 3.} Then, as a critical value (reference value Ks) for determining whether it is a zigzag, an evaluation condition is set, and the evaluation condition adopts a value that differs depending on the evaluation area R (step S210).

According to the evaluation conditions set by the evaluation condition setting section 20, the waveform data processing section 41 calculates and processes the waveform data generated by the waveform data generation section 30 for each evaluation area R. For the evaluation of the field of R _W, the lower limit shown in FIG. 3 of U L ~ upper portion among the waveform data F as _a calculation target to calculate the differential curve of the waveform data F ₁ F _2, and depicted in the differential curve F _{2 The approximate curve F 3} of the approximate value Va produced by the least square method. For the evaluation of the field of R _N, the partial waveform data F ₁ which the lower limit L ~ dividing point S as a calculation target to calculate the waveform data F differential curve _1, F _2, and in differential curve F least squares _method of generating an approximate curve F ₃ . Further, the differential curve F ₂ and an approximate curve F _3, in particular approximate curve F ₃ shows because the current range of the field of evaluation R being different waveforms, but in FIG. 3 in order to avoid complication, are omitted for the evaluation of the field R _N is Description of the differential curve F ₂ and the approximate curve F ₃ .

The calculation method of the differential curve F ₂ is explained using Figure 4 _{of the waveform data F 1} showing the typical I-L characteristic of the semiconductor laser 90. Here, it is assumed that the light output change amount when the current of ΔI _{1 is increased from the drive current I 1 is ΔP 1} , and the light output change amount when the current of ΔI ₂ is increased from the drive current I _{2 is ΔP 2} , which increases from the drive current I ₃ The amount of change in light output at a current of _{ΔI 3 is ΔP 3} . That is, from the drive current I _i to the drive current I _i+1 , assuming that the amount of change in the light output when the drive current I changes by ΔI _{i is ΔP i} , the differential value D _i with respect to the drive current I _{i is} as follows: As shown in (1), it can be calculated by difference calculation. D _i =ΔP _i ／ΔI _i …（1）

This is performed for _{each drive current I i} in the calculation target of the evaluation area R of the _{waveform data F 1} , and the differential curve F ₂ can be calculated. In this way, the differential curve F ₂ of each evaluation area R is calculated, the approximate curve F _{3 of the differential curve F 2} of each evaluation area R is calculated, and the waveform data of each area necessary to calculate the tortuosity is processed (step S220 ).

It is assumed that the light output is increased when the amount of change from the driving current ΔI _i I _i is the premise of formula (1), ΔP _i, the differential value D _i according to the value of the drive current for (I _i), but is not limited Here. For example, the differential value D _i may be a value for the drive current (I _i + ΔI _i), or the driving current for an intermediate position located at varying current _{(I i + ΔI i / 2} ) a.

When the data necessary for calculating the tortuosity rate is obtained for each evaluation area R, the tortuosity rate Ki _{for each value of the driving current I i in the calculation target is calculated for each evaluation area R in the tortuosity detection unit 42} , Compare with the set benchmark to determine whether there are sharp bends on the I-L waveform, that is, whether there are twists or not.

Next, a calculation method for calculating the tortuosity K based on the calculated data will be described. _Taking the value of a certain drive current I i on the differential curve F ₂ and the approximate curve F ₃ as the differential value D _i and the approximate value Va _i , respectively, by formula (2), the differential curve F ₂ is calculated relative to the approximate _{The tortuosity K i} , which is the degree of deviation of the curve F ₃ , is used as an indicator of abrupt bending. K _i =|((D _i -Va _i )/Va _i )|×100(%)...(2) This is calculated for each measurement point in the current range of the evaluation area R.

Next, on the basis of the tortuosity K _i of each calculated drive current I _i , the tortuosity detection for each evaluation area R is performed. For the evaluation of the field of R _W, the winding ratio for each measuring point (driving current I _i) Field Evaluation of the calculated R _W K _i, Ks value for the comparison with the reference field R _W evaluation set. For the evaluation area R _W , the reference value Ks of the tortuosity K is set to 20%. If the maximum value of the tortuosity K of each measurement point in the area is greater than 20%, the semiconductor laser 90 of the inspection object will be It is judged to have twists, that is, the twists are bad.

Next, using the same method of calculating the evaluation field winding of R _N K. For the evaluation area R _N , the reference value Ks of the tortuosity K is set to 12%. If the maximum value of the tortuosity K of each measurement point in the area is greater than 12%, the semiconductor laser 90 of the inspection object will be It is determined to have a tortuosity, that is, the tortuosity is bad (step S230).

If no tortuosity is detected in any of the plurality of evaluation areas R, the semiconductor laser 90 to be inspected is evaluated as a good product with no tortuosity defects, and if a tortuosity defect is detected in any of the evaluation areas, it is evaluated It is a defective product with a tortuous defect (step S300). The evaluation result is linked to the ID of the semiconductor laser 90 to be inspected, and displayed or recorded through an I/F (not shown). Then, while changing the inspection object, this evaluation method was repeated in order.

As described above, since the evaluation area R of the semiconductor laser 90 is standardized to a plurality of overlapping ranges (evaluation area R _W , evaluation area _RN ), there is no boundary between the evaluation areas R. Thereby, like the problem pointed out in Patent Document 2, the situation that the boundary portion becomes the start point or the end point of the bend determination does not occur, and therefore, it is possible to prevent erroneous determination of the presence or absence of the bend in the boundary portion.

Further, in order to evaluate the evaluation FIELD R _W R _N-wide field covering into the area of evaluation than the evaluation set R _W R _N wide field, the field will be used to evaluate a narrow range R _N is set to the reference value Ks ratio The reference value Ks of _{R W} used in a wide range of evaluation areas is low. Therefore, even when no tortuosity is detected in the evaluation area R _W , it is conceivable that a tortuosity will be detected in the evaluation area R _N.

On the other hand, in actual inspections, ideal waveforms are not necessarily obtained. For example, the semiconductor laser 90 I-L of the waveform, typically due to failure such as meandering waveform data F ₁ of FIG. 3 as comprising a large portion of the consecutive points P _k, which can be detected. Considering the comparison with the ideal I-L waveform F _1I shown in Fig. 5, it can be easily understood.

However, in the semiconductor laser 90, a zigzag with an atypical waveform or a suspicious zigzag with a wave like a zigzag may sometimes appear. In the present embodiment, it is assumed in advance that a tortuosity with an atypical waveform or the occurrence of a suspected tortuosity is generated, and the above-mentioned evaluation conditions are set.

The atypical zigzag is called the "initial slight zigzag K _M ", shown in the realistic I-L waveform F _1R in Fig. 5, which is close to the critical value current Ith, and the driving current I is in the range of about 40 mA. The segment offset of the I-L waveform. The suspected twists and turns are called "fluctuation _{zone K P} ", because the temperature rise causes electrons to overflow from the active layer and do not contribute to the light emission, but it is caused by the carrier overflow phenomenon that promotes heat generation. Gentle wave fluctuations are generated in the range near the rated output (150mA).

This fluctuation area K _P is prominently displayed in the differential curve F ₂ , and if a strict reference value Ks is set, there may be cases where a good part is judged as a bad part. Therefore, as in Patent Document 2, using the tortuosity calculation method or keeping the reference value as it is, using the divided areas to change the accuracy of the quadratic approximation curve, it will be difficult to distinguish between the initial small twists K _M and the undulating area K. _P. If it is applied to this embodiment, for the initial small twists K _M and the undulating area K _{P , for example, if the tortuosity K i} is calculated by the above formula (2), there may be situations where the value is the same as the original twists. . However, for the fluctuation area K _P , there are many cases in which there is no problem in characteristics, so it is desirable to avoid the tortuous determination.

Thus, patterns in embodiment 1, for the evaluation of the possibility of high field K _P waviness covers the vicinity of the rated current of the field R _W, evaluation than using near field does not contain the higher rated current reference value R _N Ks. In other words, it does not include the vicinity of the rated current, which is expected to produce small initial twists K _M , and only covers the low current range that does not include the high current range of more than 70% of the _{rated current. For such an evaluation area R N} , it will use more than the rated current The benchmark value Ks of the nearby evaluation area R _{W is lower.}

Accordingly, even if the start winding micro _M K _P and K to the same degree of fluctuation calculated field value of the winding ratio K, K _P field fluctuation error is not detected and turns, can be detected without omission initial slight meandering K _M. In other words, it is possible to detect the zigzag with high accuracy without omission.

In addition, in the first embodiment, an example in which one evaluation area (evaluation area R _W ) completely includes another evaluation area (evaluation area R _N ) is generated, but it is not limited to this. For example, when the division point S is used as a reference to divide into two areas (not shown, but described as R ₁ and R _{2 ), the area R 2 on the} higher current side than the division point S can also be expanded to not include the area R _{A part of 1} is a region lower than the discrimination point S. Or expand R ₁ and R ₂ to a range that does not include another area at all. In addition, one may completely include the other, and the other may also expand beyond the point of distinction.

In addition, in the first embodiment, there is no mention of the processing interval or calculation accuracy of the current data in each of the plural evaluation areas R. However, in addition to the above-mentioned difference in the reference value Ks, the interval of the drive current I at the time of waveform data processing, the number of times of curve approximation, etc. can also be appropriately changed. Further, using a description of the flowchart in FIG. 2, for convenience of explanation, after performing the described measurement condition setting step and the step of generating the waveform data F ₁ performs the evaluation condition setting step, but not limited to This can also be executed simultaneously. Implementation Type 2

In the first embodiment described above, an example is shown in which the differential curve of the I-L waveform and the approximate curve of the processing curve are used to calculate the tortuosity. In the second embodiment, an example of calculating the tortuosity rate by using the moving average line obtained by performing the moving average processing of the differential curve as the processing curve will be described. In addition, in the semiconductor laser inspection method of the semiconductor laser inspection device and the operation of the semiconductor laser inspection device in the second embodiment, the parts other than the calculation of the tortuosity are the same as those in the first embodiment, and the same parts are omitted. Description.

Regarding the operation of the semiconductor laser inspection device 1 and the semiconductor laser inspection method of the second embodiment, the evaluation field specifications (step S200) described in Fig. 2 of the first embodiment are the same as those of the first embodiment. S210 begins to explain.

The evaluation criterion setting unit 23 sets an evaluation method and evaluation criterion for detecting the tortuosity for each evaluation area R. In the second embodiment, it is set to calculate the differential curve F _{2 of the waveform data F 1} together for the two evaluation areas R, and plot the average value (moving average D _MA ) of the minute interval of each differential curve F ₂ Calculate the tortuosity rate K from the value on the differential curve F ₂ and the value on the moving average line with the same calculation formula. Then, as a reference value Ks for judging whether it is a zigzag, an evaluation condition is set, which uses a value that differs depending on the evaluation area R (step S210).

In accordance with the evaluation conditions set by the evaluation condition setting section 20 including the evaluation criterion setting section 23, the waveform data processing section 41 calculates and processes the waveform data generated by the waveform data generation section 30 for each evaluation area R. For the evaluation of the field of R _W, among the waveform data F _1, the description portion between the lower limit L to upper limit U of FIG. 3 as the calculation target, the waveform data is calculated differential curve F ₁ F ₂ and F moving average differential curve ₂ line. For the evaluation of the field of R _N, the waveform data F ₁ among the lower limit of the portion between L and S as the dividing point computing objects, the waveform data is calculated differential curve F ₁ and F ₂ moving average differential curve F _2.

In the calculation of the differential curve F ₂ , the formula (1) explained in the first embodiment is used. On the other hand, the moving average used in this embodiment 2 is calculated as follows. For example, the five consecutive driving current I _i ~ I _{i +} measuring point (or the calculated target point) ₄ do moving average, will be in the case where the differential value D _i the driving current I _i smoothing constituting moving average The moving average D _{MAi of the} line can be calculated as shown in equation (3). D _MAi =(D _i +D _i+1 ＋D _i+2 ＋D _i+3 ＋D _i+4 )/5 …(3) In addition, regarding the numerator on the right side of formula (3), as long as it is relative to the driving current I _i , continuous data and i _i, it can be in four previous information i _i, i _i hanging out before and after the two materials, etc., can take any form. Also, the formula (3) gives an example of a simple moving average, but it is also possible to apply an accentuated moving average or other moving averages.

For calculation of differential data by the field R of the curve F ₂ Evaluation is performed on the object type, the moving average can be calculated. In this way, the differential curve F _{2 for} each evaluation area R and its moving average are calculated, and the waveform data processing of each area necessary for calculating the tortuosity is performed (step S220).

If the data necessary for calculating the tortuosity rate is obtained for each evaluation area R, the tortuosity rate K _{i for each value of the driving current I i} in the calculation target is calculated for each evaluation area R in the tortuosity detection unit 42, and The set reference value Ks is compared to perform zigzag detection.

Specifically, in a driving current value Ii differential curve F ₂ and the moving average, as each differential value D _i, the moving average D _MAi, by the formula (4) is calculated tortuosity K _i. K _i =|((D _i -D _MAi )/D _MAi )|×100 (%)… (4) This is carried out for each measurement point within the current range in the evaluation area R.

The subsequent tortuosity detection step (step S230) in each field is the same as that described in the first embodiment. However, the reference value Ks is not limited to the same as in the first embodiment. In the second embodiment, the evaluation area R of one semiconductor laser 90 is set to a plurality of (evaluation area R _W , evaluation area _RN ) so that the overlapping range exists, so there is no boundary between the evaluation areas R. Thereby, the part that becomes the boundary part does not become the start point or the end point of the meandering determination, and it is possible to prevent the erroneous determination of meandering at the boundary part.

Further, for the evaluation of the possibility of high field K _P waviness covers the vicinity of the rated current of the field R _W, the evaluation does not include the use of more than the rated current of the field near the higher reference value R _N Ks. Thereby, even if the initial minute twists K _M and the value of the tortuosity K calculated in the undulating area K _P are the same degree, the undulating area K _P does not erroneously detect the twists, and the initial minute twists K _M can be detected without omission. That is, tortuosity can be detected with high accuracy without omission.

On the other hand, in the calculation of the tortuosity K used in the second embodiment, the moving average D _{MA of} the differential value D is used, because the tortuosity will be calculated to be lower in the range where the inclination of the I-L waveform is gentle. Value. Here, regarding the undulation area K _P caused by the aforementioned carrier overflow phenomenon, because the tilt is gentle, and the tortuosity of the equation (2) is different, a lower value is calculated in the equation (4).

At present, the initial minute twists K _M generated in the low current range may also be generated in the high current range in semiconductor laser components developed in the future. At this time, it is necessary to use the tortuosity K to determine the initial minute twists K _M and the undulating area K _P that exist in the same or close current range. At this time, a moving average is calculated tortuosity K words, by R _W in the art that the evaluation reference value Ks R _N field close evaluation reference value Ks, the art is possible to avoid the fluctuation of misidentification K _P, detected high electric The initial small twists and turns in the watershed K _M.

Therefore, if the calculation method of the tortuosity K used in the second embodiment is used, it goes without saying that even if the plural evaluation area R is not standardized, the tortuosity can be detected with high accuracy without omission. However, in order to smooth without detecting fluctuations, it is necessary to increase the number of moving average objects to, for example, a number greater than 5 as shown on the scale. When applied to the entire range, the number of calculations will increase. In this regard, by dividing the evaluation area into overlapping ranges, the parts that produce fluctuations are smoothed with emphasis, and the degree of smoothing is reduced in other areas, which can reduce the computational burden. In particular, in the inspection of mass production steps, the reduction of the calculation burden is also an important effect. Implementation Type 3

In the above-mentioned embodiments 1 and 2, the example in which two evaluation areas are provided is described, but it is not limited to this. In the third embodiment, an example in which three evaluation areas are set will be described. Figure 6 is used to illustrate the semiconductor laser device of Embodiment 3 and the semiconductor laser inspection method implemented by it. It is used to illustrate the specifications of the evaluation field when inspecting semiconductor lasers, including atypical twists and suspected twists. The waveform data of the I-L waveform and the graph form of the differential curve. In addition, the semiconductor laser device of the third embodiment and the semiconductor laser inspection method of the operation of the semiconductor laser inspection device can be applied to the above-mentioned embodiment 1, except for the specifications related to the evaluation field. 2 The description of the same content and the same part is omitted.

In the operation of the semiconductor laser inspection device 1 of the third embodiment and the semiconductor laser inspection method, as shown in Fig. 6, two division points S ₁ and S _{2 are used} to standardize three evaluation areas R (No. Step S200 in Figure 2). The lower limit L and the upper limit U are the same as those described in Figure 3 of the first embodiment, and are set to the value of the threshold current Ith+3mA and the value of 200mA, respectively.

Among the two division points S ₁ and S ₂ , the division point S ₁ is set to be 110 mA corresponding to 73% of the rated value in the same manner as the division point S of the first embodiment. On the other hand, the division point S _{2 is} set to 70 mA corresponding to 47% of the rating. Then, after dividing into _{the areas of the two division points S 1} and S ₂ , the overlapping range generating unit 22 a generates overlapping ranges in which the divided areas overlap with each other.

Specifically, as the upper portion of the frame shown in FIG. 6, the lower limit of the field until the dividing point S ₁ L to directly regulate the art to evaluate the R _N1. In addition, the area from the division point S ₂ to the upper limit U is directly standardized as the evaluation area R _N2 . On the other hand, regarding the area between the _{two division points S 1} and S _{2 , there will be a repetition range that overlaps the entire range of the two evaluation areas R N1} and R _N2 , and it is set as the evaluation area R from the lower limit L to the upper limit U _W. In other words, the range from the lower limit L of Ith+3mA to the discrimination point S ₁ of 70 mA is standardized as R _N1 , and the range from the discrimination point S _{2 of} 110 mA to the upper limit U of 200 mA is standardized as R _N2 , and from Ith+3mA The range from the lower limit L of 200mA to the upper limit U of 200mA is standardized as R _W (corresponding to step S200 in Fig. 2).

Then, for example, regarding the evaluation area R _N1 , in order to detect the initial minute twists K _M without omission, when the evaluation criterion is set, the criterion value Ks is lowered to be lower than the evaluation area R _W. Then, regarding the evaluation area R _N2 , in order not to detect the undulation area K _P erroneously, when the evaluation conditions are set, the tortuosity rate K is calculated using the moving average described in the second embodiment.

As shown in the third embodiment, the number of evaluation areas R is larger than that of the example described in the first embodiment, so that tortuosity can be detected more reliably, without omission, and with high accuracy. In particular, only the evaluation area R _N1 that covers the area less than 50% of the rated current, and only the evaluation area R _N2 that covers the area of 70% or more of the rated current, optimizes the respective evaluation conditions, and can reliably distinguish the initial micro Judgement of twists and turns K _M and K _{P in the undulating area.}

In addition, the number of evaluation areas R can be appropriately changed according to the characteristics of the semiconductor laser 90 to be inspected, or the number can be increased. In addition, for example, in _{the case of standardizing three evaluation areas R with two distinguishing points S 1} and S ₂ , as described in the implementation pattern 1, it is not necessary to set one evaluation area R to completely include other evaluation areas R, and it is appropriate Just set the range of repetition.

In addition, the standard evaluation conditions for each evaluation area R are not limited to the above examples. For example, it is also possible to use the differential curve F ₂ and its approximate curve F _{3 for the evaluation areas R N1} and R _N2 , and use the differential curve F ₂ and the moving average for the evaluation area R _W. That is, it can be set appropriately according to the inspection target (model information).

In addition, in each of the above-mentioned embodiments, the differential curve F ₂ is calculated in order to detect the sharp bend in the I-L curve, but it is not limited to this. For example, for the waveform data F ₁ separated in the current domain, a moving average or approximate curve can be calculated, and other calculation methods can also be applied.

In addition, in the semiconductor laser inspection apparatus 1 of each embodiment described above, the parts (model discriminating unit 10, evaluation condition setting unit 20, waveform data generating unit 30, characteristic evaluation unit 40) that are responsible for calculation processing, for example, the seventh As shown in the figure, it can be expressed as a control unit 100 including a processor 101 and a memory device 102. Although the memory device 102 is not shown, it includes a volatile memory device such as a random access memory and a non-volatile auxiliary memory device such as a flash memory. In addition, a supplementary memory device such as a hard disk can also be provided to replace the flash memory. The processor 101 executes a program input from the memory device 102. In this case, the program is input to the processor 101 from the auxiliary memory device through the volatile memory device. In addition, the processor 101 may output data such as calculation results to the volatile memory device of the memory device 102, or may store the data in the auxiliary memory device through the volatile memory device.

In addition, although this application describes various exemplified implementation types and examples, the various features, aspects, and functions described in one or plural implementation types are not limited to be applied to specific implementations. Types can be applied to implementation types individually or in various combinations. Therefore, countless modified examples that are not illustrated are also assumed to be within the technical scope disclosed in the specification of this application. For example, the present application includes cases where at least one component is modified, added, or omitted, and even when at least one component is extracted and combined with components of other implementation types.

As described above, the semiconductor laser inspection device 1 according to each embodiment includes: a measuring unit 50 that injects current into the semiconductor laser 90 to be inspected and drives it, and then measures the light output from the driven semiconductor laser 90 ; The waveform data generating section 30 uses the measured value output by the measuring section 50 to generate waveform data F ₁ showing the drive current-light output characteristics of the semiconductor laser 90; the evaluation condition setting section 20 standardizes the waveform data F ₁ to The plurality of evaluation areas R divided by the range of the drive current I set evaluation conditions for the standardized plural evaluation areas R; the characteristic evaluation unit 40, based on the set evaluation conditions, for each of the plural evaluation areas R, Calculate the index (tortuosity K) showing the degree of abrupt bending from the waveform data F ₁ , and compare it with the reference value Ks set in the evaluation conditions to evaluate the characteristics of the semiconductor laser 90. The evaluation condition setting unit 20 constitutes a range that regulates plural evaluation areas R, and each of the plural evaluation areas R (e.g., evaluation area R _W ) has a range that overlaps with at least one of the other evaluation areas R (e.g., evaluation area R _{N ).} Therefore, the portion forming the boundary portion in the current range does not become the starting point or the end point of the meandering determination, and it is possible to determine whether there is a meandering in the boundary part by mistake, and it is possible to detect the meandering with high accuracy without omission.

In addition, the index (tortuosity K) is calculated as the sharper the bend is, the larger the value is calculated, and the evaluation condition setting unit 20 for the plural evaluation areas R includes the evaluation area of the rated current value of the semiconductor laser 90 (for example Evaluation area R _W , evaluation area R _N2 ), set a value larger than other evaluation areas (e.g. evaluation area R _N , evaluation area R _N1 ) as the reference value Ks. In this case, even because of the fluctuation area K _P Also, if the tortuosity rate K is calculated to be the same as the true tortuosity, it is possible to detect the tortuosity with high accuracy without false detection.

And, 20 to distinguish the evaluation condition setting unit waveform data F differential curve ₁ F ₂ with respect to similar or after the _second processing curve smoothing differential curve F of the deviation is set to index (tortuosity K), it is possible to objectively characteristics Evaluation.

At this time, the characteristic evaluation unit 40 uses the differential curve F ₂ to move _{the evaluation area (for example, evaluation area R W} , evaluation area R _N2 ) that includes the rated current value of the semiconductor laser 90 among the plural evaluation areas R The average moving average is used as the processing curve. In this way, it is possible to suppress _{the tortuosity calculated in the undulating area K P} by smoothing, and it is possible to detect the tortuosity with high accuracy without erroneous detection.

In addition, the semiconductor laser inspection method according to each implementation type includes: waveform data generation steps (steps S120 to S130), injecting current into the semiconductor laser 90 of the inspection object and driving it, and then using the semiconductor laser 90 from the driving The measured value of the light output of the semiconductor laser 90 produces the waveform data F ₁ showing the driving current-light output characteristics of the semiconductor laser 90; the evaluation condition setting step (steps S200 ~ S210), the standardization of the waveform data F ₁ to the driving current I For the plural evaluation areas R divided by the scope, the evaluation conditions are set for the standardized plural evaluation areas R; the characteristic evaluation step (steps S220 to S300), according to the set evaluation conditions, for each of the plural evaluation areas R, The divided waveform data F _{1 is} calculated, and an index showing the degree of sharp bending (tortuosity K) is calculated, and compared with the reference value Ks set in the evaluation conditions, the characteristics of the semiconductor laser 90 are evaluated. When the range of the plural evaluation areas R is regulated, each of the plural evaluation areas R (for example, the evaluation area R _W ) has a range that overlaps with at least one of the other evaluation areas R (for example, the evaluation area R _{N ).} Therefore, the portion forming the boundary portion in the current range does not become the starting point or the end point of the meandering determination, and it is possible to determine whether there is a meandering in the boundary part by mistake, and it is possible to detect the meandering with high accuracy without omission.

In addition, the index (tortuosity rate K) is calculated as the sharper the bend is, the larger the value is calculated. In the evaluation condition setting step, the evaluation area R including the rated current value of the semiconductor laser 90 is included in the evaluation condition setting step ( For example, evaluation area R _W , evaluation area R _N2 ), set a value larger than other evaluation areas (e.g. evaluation area _RN , evaluation area _RN1 ) as the reference value Ks. In this case, even because of the fluctuation area K _P and calculate the tortuosity K which is the same level as the true tortuosity, without erroneous detection, and it is possible to detect the tortuosity with high accuracy.

Further, evaluation condition setting step, differentiates the waveform data differential curve _{F. 1} of F ₂ with respect to the approximated or smoothed differential curve F deviation processing curve after the _2, is set to index (tortuosity K), it is possible to objectively Characteristic evaluation.

At this time, in the characteristic evaluation step, for the evaluation area (for example, evaluation area R _W , evaluation area R _N2 ) that includes the rated current value of the semiconductor laser 90 among the plural evaluation areas R, the differential curve F _{2 is used} as The moving average of the moving average is used as the processing curve. In this way, it is possible to suppress _{the tortuosity calculated in the undulating area K P} by smoothing, and it is possible to detect the tortuosity with high accuracy without erroneous detection.

1: Semiconductor laser inspection device 10: Model determination section 20: Evaluation condition setting section 21: Measurement condition setting section 22: Evaluation area specification section 22a: Repeat range generation section 23: Evaluation criterion setting section 30: Waveform data generation section 40: Characteristic evaluation section 41: Waveform data processing section 42: Zigzag detection section 50: Measurement section 51: Laser drive section 52: Light output measurement section 90: Semiconductor laser 100: Control section 101: Processor 102: Memory device D: Differential Value D _MA : Moving average F ₁ : Waveform data F ₂ : Differential curve F ₃ : Approximate curve I: Drive current K: Tortuosity K _S : Reference value R: Evaluation area S: Differentiating point Va: Approximate value

FIG. 1 is a block diagram for explaining the structure of the semiconductor laser inspection apparatus according to the first embodiment. FIG. 2 is a flowchart for explaining the operation of the semiconductor laser inspection device of Embodiment 1, that is, the semiconductor laser inspection method. Figure 3 is used to illustrate the waveform data and processing curves generated in the semiconductor laser inspection method of Embodiment 1. Figure 4 is used to illustrate the calculation of the differential curve from the generated waveform data in the semiconductor laser inspection method of Embodiment 1. Figure 5 is used to illustrate the ideal I-L waveform and the actual I-L waveform that contains suspected twists. Figure 6 is used to illustrate the waveform data and processing curves generated in the semiconductor laser inspection method of the third embodiment. FIG. 7 is a block diagram showing the structure of the hardware part responsible for the calculation processing of the semiconductor laser inspection device of each implementation type.

1: Semiconductor laser inspection device

10: Model Judgment Department

20: Evaluation condition setting section

21: Measurement condition setting section

22: Evaluation Field Specification Department

22a: Repeating range generation part

23: Evaluation Criteria Setting Department

30: Waveform data generation part

40: Characteristic Evaluation Department

41: Waveform Data Processing Department

42: Zigzag detection department

50: Measurement Department

51: Laser Drive

52: Light output measurement department

90: Semiconductor laser

Claims (8)

A semiconductor laser inspection device, including: The measuring part injects current into the semiconductor laser of the inspection object and drives it, and then measures the light output from the driven semiconductor laser; The waveform data generating part uses the measured value output by the measuring part to generate waveform data showing the driving current-light output characteristics of the semiconductor laser; The evaluation condition setting unit standardizes a plurality of evaluation fields that distinguish the waveform data by the range of the drive current, and sets evaluation conditions for the standardized plurality of evaluation fields; The characteristic evaluation unit calculates an index showing the degree of sharp curvature for each of the plurality of evaluation areas based on the set evaluation condition, and compares it with the reference value set in the evaluation condition to evaluate the The characteristics of semiconductor lasers, The evaluation condition setting unit regulates the range of the plurality of evaluation fields so that the plurality of evaluation fields respectively have a range that overlaps with at least one of the other evaluation fields. The semiconductor laser inspection device as described in item 1 of the scope of patent application, in which: This index will be calculated as the greater the value of the sharper the bend, The evaluation condition setting unit sets a value larger than the other evaluation areas as the reference value for the evaluation area including the rated current value of the semiconductor laser among the plurality of evaluation areas. The semiconductor laser inspection device described in item 1 or 2 of the scope of patent application, in which: The evaluation condition setting unit sets the deviation of the differential curve of the divided waveform data from the processing curve obtained by approximating or smoothing the differential curve as the index. The semiconductor laser inspection device as described in item 3 of the scope of patent application, in which: The characteristic evaluation unit uses a moving average line obtained by taking a moving average of the differential curve as the processing curve for the evaluation area including the rated current value of the semiconductor laser among the plurality of evaluation areas. A semiconductor laser inspection method, including: The waveform data generation step is to inject current into the semiconductor laser of the inspection object and drive it, and then use the measured value of the light output from the driven semiconductor laser to generate a waveform showing the driving current-light output characteristics of the semiconductor laser data; The evaluation condition setting step is to standardize a plurality of evaluation fields that distinguish the waveform data by the range of the drive current, and set evaluation conditions for the standardized plurality of evaluation fields; In the characteristic evaluation step, according to the evaluation condition, for each of the plural evaluation fields, an index showing the degree of sharp curvature is calculated for each of the divided waveform data, and compared with the reference value set in the evaluation condition, the semiconductor is evaluated The characteristics of lasers, When the range of the plural evaluation areas is regulated, the plural evaluation areas will each have a range that overlaps with at least one of the other evaluation areas. The semiconductor laser inspection method described in item 5 of the scope of patent application, in which: This index will be calculated as the greater the value of the sharper the bend, In the evaluation condition setting step, for the evaluation area including the rated current value of the semiconductor laser among the plurality of evaluation areas, a value larger than other evaluation areas is set as the reference value. Such as the semiconductor laser inspection method described in item 5 or 6 of the scope of patent application, in which: In the evaluation condition setting step, the deviation of the differential curve of the differentiated waveform data from the processing curve obtained by approximating or smoothing the differential curve is set as the index. The semiconductor laser inspection method described in item 7 of the scope of patent application, in which: In the characteristic evaluation step, for the evaluation area including the rated current value of the semiconductor laser among the plurality of evaluation areas, a moving average line obtained by taking the differential curve as a moving average is used as the processing curve.

Images