JP6101387B2 - Defect detection device and defect detection method - Google Patents

Description

The embodiment relates to a failure detection device that detects a failure using, for example, time domain reflectometry (TDR).

In the manufacturing process of an electronic device that mounts an electronic component on a mounting substrate, when mounting the electronic component on the mounting substrate, the connection portion on the mounting substrate and the connection portion of the electronic component are connected by solder, and then The connection is inspected for a short circuit or disconnection.

For example, as one form of the semiconductor device, there is a BGA type semiconductor device. In this case, in the ball grid array, solder balls are arranged in a grid on the bottom surface of the package,
Connection to a printed circuit board or the like is performed via the solder balls. Printed circuit board (mounting board)
The state of the solder ball mounted on the board cannot be visually confirmed, and, for example, a boundary scan or X-ray inspection is performed.

In addition to inspection by boundary scan and X-ray, for example, a time domain reflection measurement device (TD
A soldering inspection apparatus using R: Time Domain Reflectmetry has been proposed.

JP 2003-124851 A JP-A-9-61486

The present embodiment provides a defect detection device that can detect a defective portion of a device to be inspected with higher accuracy.

The defect detection apparatus according to the embodiment includes a measurement unit that measures a first time from when a signal is input to the inspected apparatus until a reflected signal reflected by a defective portion of the inspected apparatus is received, and a plurality of components or a plurality of parts A storage unit that stores a plurality of model data indicating conduction characteristics according to each of the parts,
Based on the conduction characteristics of each region, the range of the inspection object in the inspected device is divided into a first part and a second part that is farther from the measurement unit than the first part, Substituting the second time obtained by subtracting the time corresponding to the time taken to reflect and receive the signal after inputting the signal into the first part from the first time into the model data corresponding to the second part. The control part which calculates the 1st prediction conduction distance in 2 part, and the display part which displays the position of a defective part on image data based on a calculation result are provided.

1 is an overall configuration example of a defect detection apparatus according to a first embodiment. Sectional drawing which shows the BGA type semiconductor device which is 1 aspect of a to-be-inspected apparatus. FIG. 3 is a schematic diagram showing CAD data corresponding to a first surface of a glass-epoxy substrate 201 when an inspected device 200 is a BGA type semiconductor device. FIG. 4 is an enlarged view of S in FIG. 3. The schematic diagram which shows the image data in case the to-be-inspected apparatus 200 is a BGA type semiconductor device. The schematic diagram which shows the table regarding the characteristic data hold | maintained at the memory | storage part 30 of this embodiment. The schematic diagram of the table which shows the relationship between characteristic data and model data hold | maintained at the memory | storage part 30 of this embodiment. It is a flowchart figure which shows the test | inspection operation | movement of the defect detection apparatus of this embodiment. The flowchart figure which shows step S5 of FIG. The flowchart figure of the test | inspection operation | movement of the defect detection apparatus of this modification 1. The figure which shows the table regarding the model data hold | maintained at the memory | storage part 30 of the defect detection apparatus of the modification 3. FIG. The flowchart figure of the test | inspection operation | movement of the defect detection apparatus of the modification 3. The figure which shows the table regarding the model data hold | maintained at the memory | storage part 30 of the defect detection apparatus of 2nd Embodiment. The flowchart figure of the test | inspection operation | movement of the defect detection apparatus of 2nd Embodiment.

Hereinafter, this embodiment will be described with reference to the drawings. In the description, common parts are denoted by common reference symbols throughout the drawings. However, it should be noted that the drawings are schematic, and the relationship between the thickness and the planar dimensions, the ratio of the thickness of each layer, and the like are different from the actual ones.
Accordingly, specific thicknesses and dimensions should be determined in consideration of the following description. Moreover, it is a matter of course that portions having different dimensional relationships and ratios are included between the drawings.

(First embodiment)
[Configuration of Defect Detection Device of First Embodiment]
The configuration of the defect detection apparatus of this embodiment will be described with reference to the block diagram of FIG. FIG.
These are the block diagrams which show the defect detection apparatus and to-be-inspected apparatus of this embodiment. For convenience of explanation, a BGA type semiconductor device is used as an example of the device 200 to be inspected. However, the present invention is not limited to this, and any device can be used as long as a plurality of electronic components are electrically connected. May be. For example, an SSD or an SD card equipped with a NAND flash memory.

First, before describing the defect detection apparatus 100 of the present embodiment, the BGA type semiconductor device of the inspected apparatus 200 will be described with reference to the cross-sectional view of FIG. 2 in order to simplify the description.

As shown in FIG. 2, a BGA type semiconductor device 200 includes a glass epoxy substrate 201, a plurality of semiconductor chips 203a to 203c (first semiconductor chip 203a, second semiconductor chip 203b, and third semiconductor chip 203c), bonding wires 205, solder A ball 206 and a mold resin 209 are included. Here, the first semiconductor chip 203 a is bonded to the glass-epoxy substrate 201 through the adhesive layer 202. The second semiconductor chip 203b is in contact with the first through the adhesive layer 204a.
Bonded to the semiconductor chip 203a. The second semiconductor chip 203c is bonded to the second semiconductor chip 203b through the adhesive layer 204b. The surface (first surface in FIG. 2) of the glass epoxy substrate 201 has a plurality of first electrodes (not shown) and a plurality of first wirings connected to the first electrodes. Each first electrode is connected to a corresponding solder ball 206. Glass epoxy substrate 201
The back surface (second surface in FIG. 2) has a plurality of second electrodes (not shown) and a plurality of second wirings connected to the second electrodes. Each second electrode is connected to a corresponding bonding wire 205. The first wiring and the second wiring are connected via vias (not shown). Each bonding wire 205 has a function of connecting a corresponding second electrode and a corresponding third electrode (not shown) of the semiconductor chips 203a to 203c. A plurality of semiconductor chips 203a to 203c,
The bonding wire 205 is covered with a mold resin 209. Each solder ball 206 is assigned a pin number.

1. Overall Configuration A defect detection apparatus 100 according to this embodiment will be described with reference to FIG. The defect detection apparatus 100 according to the present embodiment includes, for example, an input / output unit 10 that exchanges data with the outside, a control unit 20 that performs desired arithmetic processing, a storage unit 30 that stores various data, and an inspected device 200. A time domain reflection measurement unit 40 for inputting a frequency signal and performing a desired measurement;

1-1. Input / Output Unit 10 The input / output unit 10 of the present embodiment will be described. As shown in FIG. 1, the input / output unit 10 is connected to a control unit 20 described later. The input / output unit 10 is a device to be inspected 2 input by a user.
For example, it has a function of transferring 00, for example, CAD data to the control unit 20. Further, the input / output unit 10 has a function of transferring various types of data input by the user to the control unit 20 when performing the defect detection inspection of the inspected apparatus 200.

When the input / output unit 10 receives the inspection result of the defect detection inspection 200 from the control unit 20, the input / output unit 10 outputs the inspection result to the outside. For example, when there is a defective part in the inspected apparatus 200, a pointer indicating the defective part is specified in the image data of the CAD data and output to an external user.

1-2. Next, the control unit 20 according to the present embodiment will be described with reference to the block diagram of FIG. As illustrated in FIG. 1, the control unit 20 includes a RAM unit 21, a library 22, a division unit 23, and a calculation unit 24. In addition, although the library 22 of this embodiment was provided in the control part 20, it is not restricted to this, For example, you may provide in the memory | storage device 30 mentioned later.

The RAM unit 21 has a function of temporarily storing, for example, CAD data input from the input / output unit 10. The RAM unit 21 is also used as a space for performing a desired calculation.
The CAD data includes image data (for example, FIG. 3) and characteristic data indicating the dimensions of each component.

The dividing unit 23 has a function of dividing the image data of CAD data into parts constituting the device under test 200. Data regarding the parts to be divided is held in the library 22.
The dividing unit 23 reads the CAD data into the RAM unit 21 and then accesses the library 22 to divide the CAD data into parts.

Specifically, the CAD data corresponding to the first surface of the glass epoxy substrate 201 shown in FIG.
An example in which 0 is received will be described with reference to FIGS. FIG. 3 shows the device under test 20.
6 is a schematic diagram showing CAD data corresponding to the first surface of the glass-epoxy substrate 201 when 0 is a BGA type semiconductor device. FIG. FIG. 4 is an enlarged view of S in FIG.

When the dividing unit 21 divides the image data S connected to the solder ball 206 (1) having the pin number 1 into parts on the first surface of the glass epoxy substrate 201, the dividing unit 21 receives the library 2.
2, based on the data relating to each component, as shown in FIG. 4, the solder ball 206 (1) of pin number 1 and the first wiring 207 (
1) Divide into vias 208 (1).

Further, the dividing unit 21 divides the part into parts for each characteristic when the characteristic of the part changes in the middle of the part. This will be specifically described using the above example. As shown in FIG. 4, the first wiring 207 (1) has a wiring width that changes in the middle of wiring or a wiring that is bent in the middle of wiring. The dividing unit 21 divides the first wiring into the first part 207 (1-1), the second part 207 (1-2), ... the thirteenth part (2- And 13).

The library 22 holds data indicating a criterion for dividing a part into parts. For example, when the component is a wiring, the wiring is divided when the wiring width changes. The wiring is bent (for example, 20
7 (1-6)) and the wiring are divided.

There are various division methods other than those described above, but the division is performed when the resistance value per unit length changes. As long as the resistance value per unit length does not change, it is handled as one part without being divided. Since the resistance value per unit length changes when the wiring width is different, it is divided. Similarly, when the wiring is bent in the middle, the resistance value per unit length is changed, so that the wiring is divided.

The control unit 20 includes image data of CAD data and each divided part or part.
The characteristic data is output to the storage unit 30. Note that the present invention is not limited to this case. For example, the image data and characteristic data of CAD data are stored in the storage unit 3 for each divided part or part.
You may output to 0.

At the time of defect detection inspection, the calculation unit 24 recognizes a component or part corresponding to the solder ball that is performing the defect detection inspection, reads out model data (described later) for each component or part to the RAM unit 21, and sets the time. It has a function of identifying a defective part in a part or part by applying data input from an area reflection measuring unit (described later) 40 to model data. Detailed description will be given later.

1-3. About Storage Unit 30 The storage unit 30 holds characteristic data of CAD data for each component constituting the device under test 200 or for each portion obtained by dividing the component into a plurality of parts. The storage unit 30 may hold the image data of CAD data as a whole, or may hold the image data of CAD data for each part or for each part.

These data are data used to specify a non-destructive defective part in a part or part, and may be data used to clearly indicate a defective part in a part or part to a user.

The storage unit 30 holds image data and characteristic data for each part or part every time CAD data is input from the input / output unit 10. For example, in the case of the BGA type semiconductor device shown in FIG. 2, as shown in FIG. 5, the storage unit 30 has five image data I1 to I5 and characteristic data for each part or part corresponding to the image data I1 to I5. Holds D1 to D5. Here, FIG. 5 is a schematic diagram showing image data when the device under test 200 is a BGA type semiconductor device.

The image data I1 is image data related to the first surface of the glass epoxy substrate 201.
The image data I1 is image data of each of the first electrode, the first wiring, and the via formed on the first surface of the glass-epoxy substrate 201. The characteristic data D1 is characteristic data of each of the first electrode, the first wiring, and the via (data indicating the arrangement position, data indicating the length, data indicating the width,
Data indicating the inclination with respect to the reference line, data indicating the shape, and material data).

The image data I2 is image data related to the second surface of the glass epoxy substrate 201.
The image data I2 is image data of each of the second electrode, the second wiring, and the via formed on the second surface of the glass-epoxy substrate 201. The characteristic data D2 is characteristic data of each of the second electrode, the second wiring, and the via (data indicating the arrangement position, data indicating the length, data indicating the width,
Data indicating the inclination with respect to the reference line, data indicating the shape, and material data).

The image data I3 to I5 are image data related to the semiconductor chips 203a to 203c, respectively. That is, it is image data such as wiring connected to the third electrode of each of the semiconductor chips 203a to 203c and electronic components in the semiconductor chips 203a to 203c.
Characteristic data D3 to D5 are characteristic data of wiring, electronic parts, etc. (data indicating arrangement position, data indicating length, data indicating width, data indicating inclination with respect to the reference line, data indicating shape,
Material data).

In the present embodiment, the storage unit 30 stores the image data I1 to I5 and the image data I1.
The characteristic data D1 to D5 for each part or part corresponding to .about.I5 are held. For example, when the image data I1 to I5 can be reproduced by the characteristic data D1 to D5, only the characteristic data D1 to D5 are stored in the storage unit. 30 may be held. In the present embodiment, as the characteristic data, when the component is a wiring, the data indicating the arrangement position of the component or part, the data indicating the length, the data indicating the width, the data indicating the inclination with respect to the reference line, and the data indicating the shape are used. Although shown, it is not restricted to this, Any data may be used as long as it is data specifying a wiring or a wiring part. As the solder ball and via characteristic data, for example, data indicating a circular shape, data indicating the diameter (data indicating the size of the first diameter, data indicating the size of the second diameter), and the material are indicated. There is data. Any data may be used as long as the data specifies a part such as a solder ball and a part of the part.

These characteristic data are held in the storage unit 30 as a table shown in FIG. 6, for example. FIG. 6 is a schematic diagram illustrating a table relating to characteristic data held in the storage unit 30 of the present embodiment.

As shown in FIG. 6, specific data indicating the width (or first diameter), length (or second diameter), barycentric coordinates, inclination with respect to the reference line, material, and shape is provided for each part or for each part. It is associated. Solder ball 206 (1), first wiring 207 (1-1) to 207 (1-13)
, Vias 208 (1) have widths (or first diameters) of a1-1, a2-1 to a2-1, respectively.
3 and a3-1, and the length (or the second diameter) is b1-1, b2-1 to b2-
13 and b3-1, and the shapes are c1-1, c2-1 to c2-13, and c3-1, respectively. The material of the solder balls 206 (1), the first wirings 207 (1-1) to 207 (1-13), and the vias 208 (1) are all X (for example, copper Cu).

The storage unit 30 holds not only the image data of the CAD data, the characteristic data for each part or part, but also model data. Here, the model data is data (for example, function data) obtained by modeling the conduction characteristics of the frequency pulse for each part or for each part. This model data is used to identify a defective part in a part or a part based on data input from a time domain reflection measuring unit 40 described later, image data, and characteristic data.

The model data is stored in advance in the storage unit 30 before performing the defect detection inspection in the time domain reflection measurement unit 40.
Retained. A method for calculating the model data will be described later.

  The storage unit 30 stores model data corresponding to characteristic data for each part or part.

Hereinafter, a specific description will be given using the table shown in FIG. Here, FIG. 7 is a schematic diagram of a table showing the relationship between the characteristic data and the model data held in the storage unit 30 of the present embodiment.

As shown in FIG. 7, for example, when the characteristic data is the material X, the width a2-1, and the shape c2-1, the model data L2-1 satisfies the formula (1).

L2-1 = f2-1 (t) (1)
Here, the model data L2-1 indicates that when the frequency pulse is input from the end of the wiring (or wiring portion) of the material X, the width a2-1, and the shape c2-1, the time pulse elapses. Indicates the predicted conduction distance conducted from the end of the wiring (or part of the wiring).

Although only three types of model data are disclosed in FIG. 7, model data corresponding to all the parts and parts used in the inspected apparatus 200 are held. However, material X, width a
Even if there are a plurality of wirings having different lengths for the wiring of 2-1 and shape c2-1 (straight line), the same model data is used if the conduction characteristics of the frequency pulse are the same. Thereby, it is possible to prevent an increase in model data to be held.

In the present embodiment, the storage unit 30 stores model data corresponding to all parts and parts used in the inspected apparatus 200. However, if the inspection target is predetermined, the inspection target part, The storage unit 30 may hold only model data corresponding to the part.

The function f2-1 (t) is a function having a positive correlation at time t. For example, f2-1 (t) = A
It is a linear function such as 2-1 * t + B2-1. Here, A2-1 (> 0) and B2-1 are
It is a constant determined based on the material X, the width a2-1, and the shape c2-1. Function f2-1 (t
) May not be a linear function, for example, a square root function of time t or a quadratic function.

The model data f1-1 (t), f2-1 (t),... Are stored in the storage unit 30 in association with the characteristic data. In the present embodiment, the storage unit 30 stores model data in association with each characteristic data. However, model data may be stored in the storage unit 30 in association with each part or part.

1-4. Time Domain Reflection Measurement Unit 40 Next, the time domain reflection measurement unit 40 of this embodiment will be described with reference to FIG. As shown in FIG. 1, the time domain reflection measurement unit 40 is connected to the control unit 20 and the device under test 200.
The time domain reflection measurement unit 40 has a probe needle (not shown), and performs a defect detection inspection by contacting the probe needle with a solder ball to be inspected.

As the time domain reflection measurement unit 40, for example, time domain reflection measurement (TDR: Time Domain Refl
ectmetry) equipment.

Upon receiving a signal for executing the defect detection inspection from the control unit 20, the time domain reflection measurement unit 40 inputs a frequency pulse to the device under inspection 200. If there is a defect in the wiring connected to the solder ball to be inspected of the device under inspection 200 (such as a disconnection or scratch in the wiring), the frequency pulse is reflected at the defective part. The time domain reflection measurement unit 40 receives this reflection pulse (reflection signal) from the device under test 200. The time domain reflection measurement unit 40 measures a time T from when a frequency pulse is input to when the reflection pulse is received.

  The time domain reflection measurement unit 40 outputs data of time T to the control unit 20.

[Inspection Operation of Defect Detection Device of First Embodiment]
Next, the inspection operation of the defect detection apparatus of this embodiment will be described using the flowchart of FIG. Here, FIG. 8 is a flowchart showing the inspection operation of the defect detection apparatus of the present embodiment.

In the inspection operation of the defect detection apparatus according to the present embodiment, the storage unit 30 holds the characteristic data of the CAD data for each part constituting the inspection target of the inspected apparatus 200 or for each part obtained by dividing the part. Then, after the storage unit 30 holds the model data for each part or part, it is executed.

Before describing a specific inspection method for a defect detection apparatus, an example of a model data calculation method will be described. The model data is determined by measuring the conduction specification for each part or part.
Specifically, for example, the conduction characteristics of a certain wiring are measured as follows.

Two wires having the same characteristics are prepared, and a defect is generated in the first wire at a distance LL1 from the end of the wire. In the second wiring, a defect is generated at a distance LL2 (> LL1) from the end of the wiring. The time domain reflection measuring unit 40 is connected to the end of the first wiring, and a frequency pulse is input to the end of the wiring. The time domain reflection measurement unit 40 measures a time TT1 from when the frequency pulse is input until the reflection pulse is received. Similarly, the time domain reflection measurement unit 40 is connected to the end of the second wiring, and a frequency pulse is input to the end of the wiring. The time domain reflection measurement unit 40 measures a time TT2 from when the frequency pulse is input until the reflection pulse is received.

Model data is determined based on the times TT1 and TT2 and the distances LL1 and LL2. Accurate model data can be determined by increasing the number of samples.

The model data for each part or part determined in this way is held in the storage unit 30.

Next, the inspection operation of the defect detection apparatus of this embodiment will be described with reference to FIG. In addition,
Already stored in the storage unit 30 is the characteristic data of the CAD data, the image data of the CAD data, and the data to be inspected 200 for each part constituting the inspected apparatus 200 or for each part obtained by dividing this part into a plurality of parts. It is assumed that model data corresponding to all the parts and parts are held.

First, in step S1, when the pin number of the solder ball to be inspected is received, the input / output unit 10
Outputs the pin number to the control unit 20. In step S1, not only the pin number but also the material of solder resist, the material of the wiring, the material and size of the solder ball, the height of the wiring (the height in the direction perpendicular to the plane in which the wiring extends), for example. Upon receipt, the input / output unit 10 outputs these to the control unit 20. Based on these data, model data may be extracted and arranged in step S3 described later.

In step S <b> 2, the calculation unit 24 of the control unit 20 extracts a range to be inspected from the CAD data with the range electrically connected to the solder ball having the selected pin number as the inspection target. For example, when the device 200 to be inspected is a BGA type semiconductor device, the calculation unit 24 identifies the first electrode, the first wiring, and the via connected to the solder ball to be inspected from the image data I1 and puts them in the range to be inspected. include.

The calculation unit 24 specifies the second wiring and the second electrode connected to the via included in the inspection target range by the image data I1 from the image data I2, and includes these in the inspection target range.
Similarly, from the image data I3 to I5, among the semiconductor chips 203a to 203, the electronic components and the like of the semiconductor chips 203a to 203c connected to the inspection target range of the image data I2 through the bonding wires 205 are specified. Is included in the scope of inspection.

In step S3, the calculation unit 24 reads the model data corresponding to the parts and portions included in the range to be inspected into the RAM unit 21 so that the frequency pulses are arranged in the order of conduction. Specifically, the calculation unit 24 stores characteristic data corresponding to parts and portions included in the inspection target range.
6 is read out from the table of FIG. 6 to the RAM unit 21, and the components and parts included in the inspection target are arranged in the order in which the frequency pulses are conducted. Then, the calculation unit 24 accesses the table of FIG. 7 in the storage unit 30 and reads model data corresponding to the characteristic data of each component and part to the RAM unit 21. Then, the calculation unit 24 arranges the model data in the RAM unit 21 in the order in which the frequency pulses are conducted.

In step S <b> 4, the control unit 20 outputs a signal for executing the defect detection inspection to the time domain reflection measurement unit 40. Thereby, a defect detection inspection is executed. Time domain reflection measurement unit 4
In 0, the probe needle is brought into contact with the solder ball to be inspected of the apparatus 200 to be inspected. The time domain reflection measurement unit 40 receives a frequency pulse, and measures a time T1 from when the frequency pulse is input until the reflection pulse is received. The time domain reflection measurement unit 40 outputs the data at time T1 to the control unit 20.

In step S5, the control unit 20 identifies a defective portion based on the model data in the order in which the frequency pulse obtained in step S3 is conducted and the time T1 obtained in step S4.

A specific method will be described with reference to the flowchart of FIG. In the following description, it is assumed that the solder ball to be inspected is pin number 1 and the defective portion is in the first wiring 207 (1-3). It is assumed that the model data in the order in which the frequency pulse obtained in step S3 is conducted and the characteristic data of the length are as follows.

Solder ball 206 (1): L1-1 = f1-1 (t), b1-1
First wiring 207 (1-1): L2-1 = f2-1 (t), b2-1
First wiring 207 (1-2): L2-2 = f2-2 (t), b2-2
First wiring 207 (1-3): L2-3 = f2-3 (t), b2-3
...
First wiring 207 (1-12): L2-12 = f2-12 (t), b2-12
First wiring 207 (1-13): L2-12 = f2-12 (t), b2-13
...
At this time, the control part 20 specifies a defective location in the following steps. As shown in FIG.
In step S5-1, the calculation unit 24 substitutes time T1 for the top model data among the model data arranged in order.

  In step S5-2, the calculation unit 24 sets i = 1 and j = 1.

In step S5-3, the calculation unit 24 calculates the distance L using the model data into which the time T1 is substituted.
i−j (T1) is calculated.

In step S5-4, the calculation unit 24 calculates the distance Li-j (T1) and the length characteristic data bij.
Compare

When the distance Li-j (T1) is larger than the length characteristic data bi-j (step S5-4)
, Yes), j is incremented. However, when j is the maximum value, i is incremented (step S5-5). When i is incremented, j is set to an initial j = 1. Note that i and j are set when the image data is divided by the dividing unit 23. These i and j are held inside the defect detection apparatus 100. Further, the calculation unit 24
Calculates a time T (i) (j) that satisfies the following equation (2) (step S5-5).

Li−j (T (i) (j)) = bi−j Equation (2)
The calculation unit 24 calculates the time T1-ΣT (i) (j) (that is, T (1) (
j), the sum of T (2) (j) for all j, and T (i for all j)
-1) The sum total of (j) and the total sum of T (i) (1) to T (i) (j-1) are calculated. Note that T11 = 0. The computing unit 24 sets time T1-ΣTij instead of time T1 (step S5-5), and returns to step S5-3.

When the distance Li-j (T1) is smaller than the length characteristic data bi-j (step S5-4)
, No), the calculation unit 24 detects the distance Li-j (T1) as a defective location (step S5).
-6), and proceeds to step S6.

The operation after step S5-3 will be described using the above example.
−1 (T1) is calculated, and the distance L1-1 (T1) is compared with the length characteristic data b1-1 (
Step S5-4). Since the distance L1-1 (T1) is larger, the calculation unit 24 increments i to i = 2, j = 1, and calculates T (1) (1) (step S5-5).
). And it returns to step S5-3, and the calculating part 24 calculates distance L2-1 (T1-T (1) (1)) with the model data which substituted time T1-T (1) (1). Step S5-3). The computing unit 24 compares L2-1 (T1-T (1) (1)) with b2-1 (step S5-4). Since the distance L2-1 (T1-T (1) (1)) is larger, the calculation unit 24 increments j to i = 2, j = 2, and T (1) (1) + T ( 2)
(1) is calculated (step S5-5). And it returns to step S5-3 and the calculating part 24
Is the distance L2-1 (T1-T (1) (1) -T (2) (1)) according to the model data into which the time T1-T (1) (1) -T (2) (1) is substituted. Is calculated (step S5-3).

The computing unit 24 compares the distance L2-1 (T1-T (1) (1) -T (2) (1)) with b2-2 (step S5-4). Distance L2-1 (T1-T (1) (1) -T (2) (1)
) Is smaller, the calculation unit 24 calculates the distance L2-1 (T1-T (1) (1) -T (2) (
1)) is detected as a defective part (step S5-6), and the process ends.

In step S6, the control unit 20 reads the image data of CAD data in the storage unit 20 to the RAM unit 21, and marks the position of the defective portion detected in step S5 on the image data. The control unit 20 outputs the image data on which the defective portion is marked to the input / output unit 10.

In step S7, the input / output unit 10 displays the image data in which the defective portion is marked.
[Effect of the first embodiment]
As described above, the defect detection apparatus according to the present embodiment can detect a defect portion of the inspected apparatus with higher accuracy. This will be specifically described below. Time domain reflectometry (TDR)
The effect of the present embodiment will be described as an example of a comparative example in which the apparatus identifies a defective portion of the inspected apparatus.

In the comparative device, when identifying a defective part of the device under test, even if the frequency pulse is input to the device under test and the time from receiving the frequency pulse to receiving the reflected pulse is measured, this time It is not possible to determine where the defective part is. When the device under test is a BGA type semiconductor device, it is difficult to determine whether there is a defect in the wiring of the glass-epoxy substrate or whether there is a defect in the wiring inside the semiconductor chip.

However, in the present embodiment, the control unit 20 divides each part or every part based on the CAD data, reads out model data for every divided part or every part, and performs a desired calculation to obtain a defective part. Is identified. As a result, it is possible not only to detect which part or part of the divided part has a defective part, but also to detect that there is a defective part in a part or part away from the part or part. From the above, the defect detection apparatus of the present embodiment is
It is possible to detect a defective portion of the device to be inspected with higher accuracy.

(Modification 1)
In the defect detection apparatus according to the first modification of the present embodiment, only the model data and the table of FIG. 7 shown in the first embodiment are held in the storage unit 30 with respect to the configuration of the defect detection apparatus of the first embodiment. The storage unit 30 is different in that image data and characteristic data of CAD data are not held, and other configurations are the same, and detailed description thereof is omitted.

Next, the inspection operation of the defect detection device according to the first modification will be described with reference to FIG. It is assumed that the storage unit 30 has already stored model data corresponding to all the parts and parts used in the device under test 200.

First, in step S1, upon receiving the pin number and CAD data (image data and characteristic data) of the solder ball to be inspected, the input / output unit 10 sends the pin number and CAD data to the control unit 2.
Output to 0. In step S1, not only the pin number and CAD data, but also, for example, solder resist material, wiring material, solder ball material and size, wiring height (the height in the direction perpendicular to the plane in which the wiring extends) The input / output unit 10 outputs these to the control unit 20. Based on these data, model data may be extracted and arranged in step S4 described later.

In step S2, the calculation unit 24 of the control unit 20 holds the CAD data in the RAM unit 21,
The range to be inspected is extracted from the CAD data with the range electrically connected to the solder ball of the selected pin number as the inspection target. The method for extracting the range to be inspected is the same as step S2 in the first embodiment.

In step S <b> 3, in the RAM unit 21, the dividing unit 23 divides the range to be inspected into parts and parts based on the data held in the library 22. The dividing unit 23 divides only the range to be inspected into parts or portions, and does not divide the range outside the inspection target.

In step S4, the calculation unit 24 accesses the table of FIG. 7 in the storage unit 30 and arranges model data corresponding to the parts and parts included in the range to be inspected in the order in which the frequency pulses are conducted.

In step S <b> 5, the control unit 20 sends a signal for executing the defect detection inspection to the time domain reflection measurement unit 4.
Output to 0. Thereby, a defect detection inspection is executed. The time domain reflection measuring unit 40 brings the probe needle into contact with the solder ball to be inspected of the device under test 200. Time domain reflection measurement unit 4
0 inputs a frequency pulse, and measures the time T2 from when the frequency pulse is input to when the reflected pulse is received. The time domain reflection measurement unit 40 outputs data at time T2 to the control unit 20.

In step S6, the control unit 20 identifies a defective portion based on the model data in the order in which the frequency pulse obtained in step S3 is conducted and the time T2 obtained in step S5.

In step S <b> 7, the control unit 20 marks a defective portion on the image data of CAD data in the RAM unit 21. The control unit 20 outputs the image data on which the defective portion is marked to the input / output unit 10.

In step S8, the input / output unit 10 displays the image data in which the defective portion is marked.
[Effect of Modification 1]
As described above, the defect detection apparatus according to the first modification can detect the defect portion of the apparatus to be inspected with higher accuracy as in the first embodiment.

Moreover, the defect detection apparatus of the modification 1 can make the size of the memory | storage part 30 small with respect to 1st Embodiment. Furthermore, in the defect detection device of the first modification, in step S3 of the inspection operation, the dividing unit 23 divides only the range to be inspected into parts or parts, and does not divide the range outside the inspection target. Therefore, the defect detection inspection can be performed without dividing all the parts or parts of the device to be inspected. As a result, the inspection operation of the first modification can be inspected in a shorter time than the inspection operation of the first embodiment.

(Modification 2)
The defect detection device according to the second modification of the present embodiment has the same configuration as the defect detection device according to the first embodiment. This modification 2 is different from the inspection operation of the first embodiment in that steps S2 and S3 of the inspection operation of the first embodiment are processed in parallel with step S4, and other inspection operations are the same. Detailed description is omitted. After detecting that both step S3 and step S4 are completed, the control unit 20 proceeds to step S5.

[Effect of Modification 2]
As described above, the defect detection apparatus according to the second modification can detect the defect portion of the apparatus to be inspected with higher accuracy as in the first embodiment.

Moreover, the defect detection apparatus of the modification 2 processes step S2 and step S3 of the inspection operation of the first embodiment in parallel with step S4. As a result, the inspection operation of Modification 2 can be inspected in a shorter time than the inspection operation of the first embodiment.

Note that Modification 2 can be applied to Modification 1, and in this case, inspection can be performed in a shorter time than Modification 1.

(Modification 3)
[Configuration of Defect Detection Device of Modification 3]
In the defect detection device of the third modification of the present embodiment, the frequency pulse is transmitted from one end of the component or part to the other end in the storage unit 30 for each component or part with respect to the defect detection device of the first embodiment. However, the other configurations are the same and the detailed description is omitted.

The time for which the frequency pulse is conducted from one end of the component or part to the other end will be specifically described with reference to FIG. Here, FIG. 11 is a diagram illustrating a table related to model data held in the storage unit 30 of the defect detection apparatus of the third modification.

As shown in FIG. 11, for example, when the characteristic data is the material X, the width a2-1, and the shape c2-1, the model data L2-1 satisfies the formula (1).

L2-1 = f2-1 (t) (1)
Here, the model data L2-1 is the wiring of the material X and the shape c2-1 (or the wiring portion).
When the time t elapses after the frequency pulse is input from the end of the
Or the predicted conduction distance conducted from the end of the wiring portion).

The function f2-1 (t) is a function having a positive correlation at time t. Model data f1-1 (t
, F2-1 (t)... Are associated with the characteristic data and stored in the storage unit 30. In the present embodiment, the storage unit 30 stores model data in association with each characteristic data. However, model data may be stored in the storage unit 30 in association with each part or part.

In the present embodiment, for example, corresponding to the model data L 2-1, the time tt 2-1 in which the frequency pulse is conducted from one end to the other end of the component or part is also held in the storage unit 30.

[Inspection Operation of Defect Detection Device of Modification 3]
Next, the inspection operation of the defect detection apparatus of the third modification will be described using the flowchart of FIG. The storage unit 30 already uses the CAD data characteristic data, the CAD data image data, and the device under test 200 for each part constituting the device under test 200 or for each part obtained by dividing the part into a plurality of parts. It is assumed that model data corresponding to all the parts and parts that have been stored are held.

First, in step S1, when the pin number of the solder ball to be inspected is received, the input / output unit 10
Outputs the pin number to the control unit 20. In step S1, not only the pin number but also the material of solder resist, the material of the wiring, the material and size of the solder ball, the height of the wiring (the height in the direction perpendicular to the plane in which the wiring extends), for example. Upon receipt, the input / output unit 10 outputs these to the control unit 20.

In step S <b> 2, the calculation unit 24 of the control unit 20 extracts a range to be inspected from the CAD data with the range electrically connected to the solder ball having the selected pin number as the inspection target.

In step S3, the calculation unit 24 reads the model data corresponding to the parts and portions included in the range to be inspected into the RAM unit 21 so that the frequency pulses are arranged in the order of conduction. At this time,
The time tt during which the frequency pulse is conducted from one end to the other end of the part or part corresponding to the model data is also read into the RAM unit 21. The times tt are arranged in the order in which the frequency pulses are conducted.

In step S <b> 4, the control unit 20 outputs a signal for executing the defect detection inspection to the time domain reflection measurement unit 40. Thereby, a defect detection inspection is executed. Time domain reflection measurement unit 4
In 0, the probe needle is brought into contact with the solder ball to be inspected of the apparatus 200 to be inspected. The time domain reflection measurement unit 40 receives a frequency pulse, and measures a time T3 from when the frequency pulse is input until the reflection pulse is received. The time domain reflection measurement unit 40 outputs data at time T3 to the control unit 20.

In step S5, the control unit 20 identifies a defective portion based on the model data in the order in which the frequency pulse obtained in step S3 is conducted, the time tt, and the time T3 obtained in step S4.

At this time, the control unit 20 compares the time tt1-1 at which the frequency pulse is conducted from one end to the other end of the part or part corresponding to the top model data with the time T3. Time T3 is time tt
When larger than 1-1, the control unit 20 reads the time tt2-1 corresponding to the next model data to the RAM unit 21, and compares the time T3 with the time tt1-1 + tt2-1.

When the time T3 is larger than the time tt1-1 + tt2-1, the control unit 20 further reads the time tt2-2 corresponding to the next model data in the RAM unit 21. For example, when the time T3 is added to the time tt1-1 + tt2-1 +.
Is the model data L corresponding to the time tt (2-i) to the time T3-tt1-1-tt2-1.
-...- tt (2- (i-1)) is substituted to identify a defective part.

In step S6, the control unit 20 reads the image data of CAD data in the storage unit 20 to the RAM unit 21, and marks the position of the defective portion detected in step S5 on the image data. The control unit 20 outputs the image data on which the defective portion is marked to the input / output unit 10.

In step S7, the input / output unit 10 displays the image data in which the defective portion is marked.
[Effect of Modification 3]
As described above, the defect detection apparatus according to the modification 3 can detect the defect portion of the apparatus to be inspected with higher accuracy as in the first embodiment.

Further, the defect detection device of the modification 3 is compared with the inspection operation of the first embodiment in step S5.
The arithmetic processing is greatly reduced. As a result, the inspection operation of Modification 3 can be inspected in a shorter time than the inspection operation of the first embodiment.

Note that Modification 2 can be applied to Modification 3, and in this case, inspection can be performed in a shorter time than Modification 3.

(Second Embodiment)
Next, the defect detection apparatus of the second embodiment will be described using the table of FIG. Here, FIG. 13 is a table showing model data in the storage unit 30 of the defect detection apparatus according to the second embodiment.

The defect detection device of the second embodiment differs from the configuration of the failure detection device of the first embodiment in that the model data is different in the storage unit 30 and other configurations are the same, and detailed description thereof is omitted.

[Configuration of Defect Detection Device of Second Embodiment]
2-1. Storage Unit 30 The storage unit 30 of the defect detection device of this embodiment will be described with reference to FIG. The storage unit 30 of the present embodiment also stores model data. Here, the model data is data (for example, function data) obtained by modeling the conduction characteristics of the frequency pulse for each part or for each part.
This model data is used to identify a defective part in a part or a part based on data input from a time domain reflection measuring unit 40 described later, image data, and characteristic data.

The model data shown in FIG. 13 is not only a function of time t but also a function of width. As shown in FIG. 13, model data is associated with each material and shape. For example, when the characteristic data is the material X and the shape c2-1, the model data L2-1 satisfies the formula (1).

L2-1 = f2-1 (t, width) (1)
Here, the model data L2-1 is the wiring of the material X and the shape c2-1 (or the wiring portion).
When the time t elapses after the frequency pulse is input from the end of the
Or the predicted conduction distance conducted from the end of the wiring portion).

The function f2-1 (t) is a function having a positive correlation at time t and a function having a negative correlation with the wiring width.

The model data f1-1 (t), f2-1 (t),... Are stored in the storage unit 30 in association with the characteristic data. In the present embodiment, the storage unit 30 stores model data in association with each characteristic data. However, model data may be stored in the storage unit 30 in association with each part or part.

The variables of the model data in this embodiment are time and width characteristic data. However, the present invention is not limited to this, and when there is characteristic data that can be entered as variables in the model data, the model data may be a function of these characteristic data.

[Inspection Operation of Defect Detection Device of Second Embodiment]
Next, the inspection operation of the defect detection apparatus of this embodiment will be described using the flowchart of FIG. The storage unit 30 already uses the CAD data characteristic data, the CAD data image data, and the device under test 200 for each part constituting the device under test 200 or for each part obtained by dividing the part into a plurality of parts. It is assumed that model data corresponding to all the parts and parts that have been stored are held.

First, in step S1, when the pin number of the solder ball to be inspected is received, the input / output unit 10
Outputs the pin number to the control unit 20. In step S1, not only the pin number but also the material of solder resist, the material of the wiring, the material and size of the solder ball, the height of the wiring (the height in the direction perpendicular to the plane in which the wiring extends), for example. Upon receipt, the input / output unit 10 outputs these to the control unit 20.

In step S <b> 2, the calculation unit 24 of the control unit 20 extracts a range to be inspected from the CAD data with the range electrically connected to the solder ball having the selected pin number as the inspection target.

In step S3, the calculation unit 24 reads the model data corresponding to the parts and portions included in the range to be inspected into the RAM unit 21 so that the frequency pulses are arranged in the order of conduction. At this time,
Model data is selected from the association between the characteristic data that is not a variable of the model data and the model data, and is read out to the RAM unit 21.

In step S <b> 4, the control unit 20 outputs a signal for executing the defect detection inspection to the time domain reflection measurement unit 40. Thereby, a defect detection inspection is executed. Time domain reflection measurement unit 4
In 0, the probe needle is brought into contact with the solder ball to be inspected of the apparatus 200 to be inspected. The time domain reflection measurement unit 40 inputs a frequency pulse, and measures a time T4 from when the frequency pulse is input until the reflection pulse is received. The time domain reflection measurement unit 40 outputs data at time T4 to the control unit 20.

In step S5, the control unit 20 identifies a defective portion based on the model data in the order in which the frequency pulse obtained in step S3 is conducted, and the time T1 and wiring width data obtained in step S4.

At this time, the wiring width data, which is a variable of the model data, is also read from the characteristic data of the CAD data into the RAM unit 21 and sequentially substituted into the model data. If the wiring width varies from part to part or from part to part, but other characteristic data such as material and shape does not change, the defective part is calculated using the same model data.

In step S6, the control unit 20 reads the image data of CAD data in the storage unit 20 to the RAM unit 21, and marks the position of the defective portion detected in step S5 on the image data. The control unit 20 outputs the image data on which the defective portion is marked to the input / output unit 10.

In step S7, the input / output unit 10 displays the image data in which the defective portion is marked.
[Effects of Second Embodiment]
As described above, the defect detection apparatus according to the second embodiment can detect a defect portion of the inspected apparatus with higher accuracy as in the first embodiment.

Further, in the defect detection device of the second embodiment, the control unit 20 determines the defect based on the model data in the order in which the frequency pulse obtained in step S3 is conducted, the time T1 obtained in step S4, and the wiring width data. Identify the location. As a result, the control unit 20 calculates the defective portion using the same model data when the wiring width changes for each part or for each part but other characteristic data such as material and shape does not change. Therefore, this embodiment is compared with the first embodiment.
The size of the storage unit 30 can be reduced.

Note that the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the scope of the invention in the implementation stage. Furthermore, the above embodiments include inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some constituent requirements are deleted from all the constituent requirements shown in the embodiment, the problem described in the column of the problem to be solved by the invention can be solved, and the effect described in the column of the effect of the invention Can be extracted as an invention.

DESCRIPTION OF SYMBOLS 10 ... Input-output part 20 ... Control part 30 ... Memory | storage part 40 ... Time domain reflection measurement part 100 ... Defect detection apparatus 200 ... Inspected apparatus

Claims (9)

A measurement unit for measuring a first time after receiving a reflected signal reflected at a defective portion of the device to be inspected after inputting a signal to the device to be inspected;
A storage unit for storing a plurality of model data indicating conduction characteristics according to each of a plurality of parts or a plurality of parts;
Based on the conduction characteristics of each region, the range of the inspection object in the inspected device is divided into a first part and a second part farther from the measuring unit than the first part. A second time obtained by subtracting, from the first time, a time corresponding to the time taken to divide and input the signal to the first part and then reflect and receive it corresponds to the second part. A controller that calculates the first predicted conduction distance in the second portion by substituting into the model data;
A display unit for displaying the position of the defective portion on the image data based on the calculation result;
A defect detection apparatus comprising: The defect detection device according to claim 1, wherein the control unit divides each of the regions having different resistance values per unit of wiring. The defect detection device according to claim 1, wherein the characteristic data of the first portion is different from the characteristic data of the second portion in a wiring width, material, or shape. The characteristic data of the first part includes the characteristic data of the second part and the circular shape of the solder ball or via, the size of the diameter of the solder ball or via, or the material of the solder ball or via. The defect detection device according to claim 3 which is different. The controller is
Further dividing the range of the object to be inspected into a third portion that is further away from the measuring unit than the second portion;
A third time obtained by subtracting, from the first time, a time corresponding to the time taken to reflect and receive the signal after inputting the signal to the first part and the second part is the third part. The defect detection device according to claim 1, wherein the second predicted conduction distance in the third portion is calculated by substituting into the model data corresponding to. The control unit reads out model data arranged in the order in which the signals are conducted within the range of the inspection target, substitutes the first time into the model data, and the first predicted conduction distance or the second prediction. The defect inspection apparatus according to claim 1, wherein the conduction distance is calculated. A step of measuring a first time after receiving a reflected signal reflected from a defective portion of the device to be inspected after inputting a signal to the device to be inspected;
Extracting a range of an inspection object in the inspected apparatus; and
Dividing the range to be inspected into a first portion and a second portion farther from the measuring unit than the first portion, based on the conduction characteristics of each region within the range to be inspected; ,
A second time obtained by subtracting a time corresponding to the time taken to reflect and receive the signal after inputting the signal to the first portion from the first time is a plurality of parts of the inspection apparatus or a plurality of times Substituting into the model data corresponding to the second part among a plurality of model data indicating conduction characteristics corresponding to each part, and calculating a first predicted conduction distance in the second part;
A defect detection method comprising: Before the step of measuring the first time,
Measuring a first reflection time taken to reflect and receive the signal between a first location separated from the first position by the first distance;
Measuring a second reflection time taken to reflect and receive the signal between a second location away from the first position by the second distance;
A first distance from the first position of the device to be inspected to a first location, the first reflection time, a second distance longer than the first distance from the first position, the first Determining one model data of a plurality of model data based on two reflection times;
The defect detection method according to claim 7. Dividing the range of the inspection object into a third part further away from the measuring unit than the second part in the dividing step; and
A third time obtained by subtracting, from the first time, a time corresponding to the time taken to reflect and receive the signal after inputting the signal to the first part and the second part is the third part. The defect detection method according to claim 7, wherein the second predicted conduction distance in the third portion is calculated by substituting into the model data corresponding to.

Images