JP2010139445A - Solder printing inspection device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide technology easily obtaining a reference position when calculating height of soldering parts on the basis of two-dimensional light amount distribution of positive reflective light and scattering reflective light obtained from each point of a printed circuit board. <P>SOLUTION: A solder printing inspection device includes a displacement sensor 2 for receiving scattering reflective light by vertically irradiating near-infrared light on a substrate 1, and receiving positive reflective light by irradiating near-infrared light at an oblique angle. A frequency calculation means 4 generates a histogram regarding a scattering reflective light amount and a positive reflective light amount as two dimensions. A parameter determination means 5 selects reference distribution having frequency distribution being substantially peak on a side in which the scattering reflective light amount and the positive reflective light amount are high on the histogram, and determines two-dimensional range in the neighborhood of a skirt of the reference distribution as parameters. A measurement inspection part 100 obtains height of soldering by regarding a position near to solder parts to try to obtain the height of soldering as the reference position on the position on the substrate in a range of parameters of the scattering reflective light amount and the positive reflective light amount. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention irradiates light onto a printed board having cream-like solder printed on its surface for surface mounting electronic components, etc., and measures the height (displacement) of the solder based on the reflected light. The present invention relates to a printed solder inspection apparatus and method for inspecting a formation state. The printed solder inspection apparatus according to the present invention sets a reference position as a reference for the height of the solder near the solder spot to be measured, and obtains the solder height with respect to a predetermined height at the reference position. However, the present invention particularly relates to a technique that makes it easy to search the reference position from the light quantity distribution of reflected light.

  On the printed board, a circuit pattern (pad) is exposed where solder is printed (hereinafter referred to as “solder location”), and a pad portion on which a component can be attached and a circuit pattern There is a resist part coated with a resist that can transmit near-infrared light (where solder is not printed). Conventionally, the actual printed solder height is the pad surface (circuit pattern surface) at the bottom of the resist part. ) Is a technique that can be measured as a height from the above (Patent Document 1 with the same applicant as the present invention).

  The technique of Patent Document 1 is, for example, a pad 1c (also a circuit pattern surface) on a substrate 1 (hereinafter referred to as a “substrate”) shown in FIG. By irradiating near infrared light obliquely to the substrate 1 having the solder 1b and grasping the specular reflection light, the surface height of the resist 1a can be detected as shown in FIG. The height of the solder 1b is measured from the surface of the pad 1c using the fact that the surface height of the pad 1c can be detected as shown in FIG. 16B by irradiating light vertically and grasping the scattered reflected light. To do. More specifically, in the configuration shown in FIG. 14, near infrared light having a wavelength transmitted by the resist 1a is irradiated perpendicularly to the measurement point on the surface of the substrate 1 having the solder 1b and the resist 1a printed on the pad 1c. The first sensing means (the first near-infrared light source OS1, the first light receiving means D1, the second light receiving means D2) that receives the scattered reflected light from the measurement point and the same measurement point perpendicular to the predetermined point. Displacement sensor including second sensing means (second near-infrared light source OS2a, third light-receiving means D3) that receives specularly reflected light that is irradiated with near-infrared light at an oblique angle and specularly reflected at a measurement point 2 is provided. And the data processing part 11 calculates | requires the displacement in each measurement point in the surface of the board | substrate 1 based on the output of a 1st sensing means. The measurement point determination unit 12 identifies at least whether the measurement point is a solder location or a resist location based on the light amount data based on the output of the second sensing means. Further, the measurement calculation unit 13 receives the result of identification of the measurement point determination unit 12 and the displacement obtained by the data processing unit 11, and solders the pad 1c on the bottom of the resist 1a when the resist 1a is used as the measurement point. Find the height of 1b.

  In Patent Document 1, since the height of the solder 1b with respect to the pad 1c is obtained, the pad surface 1c serves as a reference surface (a position on the substrate 1 having the reference surface is referred to as a “reference position”). It is recommended that the surface be determined near the periphery of the solder 1b to be measured. This is because, if a place far from the solder 1b is used as a reference, it is easily affected by the warp of the substrate 1 and the like, so that the influence is reduced.

Japanese Patent Application No. 2008-209923

  However, in Patent Document 1, there is a resist 1a in the vicinity of the solder 1b whose height is to be measured, and a pad 1c is present under the resist 1a as a place corresponding to a reference position with a reference surface. Detection was performed using a threshold value or the like for the amount of light (brightness). However, with this method, depending on the situation of the surrounding luminance distribution, severe adjustment is required in setting the threshold value and the like, which is not easy.

  On the other hand, as in Patent Document 1, the reference position where the reference surface necessary for obtaining the height of the solder 1b with respect to the pad 1c at the bottom of the resist 1a is the vicinity of the solder spot to be measured, It can be understood that 1a and the pad 1c are located.

  Therefore, the tendency of the amount of regular reflection light (brightness) and the amount of scattered reflection light (brightness) with respect to near infrared light on the substrate 1 was examined. The result is shown in FIG. FIG. 13A is a diagram showing a cross section of the substrate 1 and a positional relationship among the resist 1a, the solder 1b, and the pad 1c. When the tendency of the reflection amount of such a substrate 1 is examined with the displacement sensor 2 shown in FIG. 14, the amount of specular reflection tends to be as shown in FIG. 13B, and the amount of scattered reflected light tends to be as shown in FIG. 13C. showed that. From FIG. 13B and FIG. 13C, the position where the resist 1a and the pad surface 1c are located is the position where both of them exist (position as shown by a two-dot chain line in the figure), and the amount of specular reflection light. In addition, both the amount of scattered and reflected light tended to be high.

  Therefore, the regular reflection light quantity and the scattered reflection light quantity at each position of the substrate 1 are acquired, and a two-dimensional histogram (frequency distribution) is calculated with one dimension being the regular reflection light quantity and the other being the dimension of the scattered reflection light quantity. From the above, it has been found that a parameter for finding the reference position can be obtained based on a region where both the amount of regular reflection light and the amount of scattered reflection light are high.

  It is an object of the present invention to reliably and easily set a reference position for obtaining a solder height based on a two-dimensional light amount (brightness) distribution of specularly reflected light and scattered reflected light obtained from each point of a printed board. It is to provide the required technology.

In order to achieve the above object, the invention according to claim 1 is directed to a displacement sensor (2) for irradiating light on a surface of a printed board on which solder is printed, and detecting reflected light from the surface; A reference position is set on the printed board, and based on the reference position, the solder height of the solder portion of the printed board is obtained, and the quality of the printed solder is determined based on the solder height. In a printed solder inspection apparatus comprising a measurement inspection unit (100) for
The displacement sensor is the same as the first measurement means (OS1a, D1, D2) that irradiates near infrared light perpendicularly to the measurement point of the printed board and receives scattered reflected light from the measurement point. A second sensing means (OS2a, D3) that receives the specularly reflected light that is irradiated with the near-infrared light at a predetermined angle oblique to the measurement point and is specularly reflected at the measurement point; and A frequency calculation means (4) for receiving a light amount output from the displacement sensor and generating a histogram having a frequency in a direction orthogonal to two dimensions, one being the amount of the scattered reflected light and the other being the amount of the regular reflected light. And a reference distribution including a frequency distribution that is substantially peaked on the side where the light amount of the scattered reflected light and the light amount of the specularly reflected light are high on the histogram, and the two-dimensional range near the skirt of the reference distribution As a parameter Parameter determining means (5) for determining, wherein the measurement and inspection unit is a position on the print where the amount of scattered reflected light and the amount of specularly reflected light are within the range of the parameters, It is characterized in that the height of the solder is obtained using the position close to the solder location where the height of the solder is to be obtained as the reference position.

  According to a second aspect of the present invention, in the first aspect of the invention, the parameter determining means is a partial region of the two-dimensional range on the side where the light amount of the scattered reflected light and the light amount of the regular reflected light are high. A selection function (5a) for selecting the reference distribution from the histogram, and approximating a distribution near the peak of the selected reference distribution by the selection function from the histogram. Parameter calculating means (5b) for calculating an approximate Gaussian distribution and determining, as the parameter, a range in which the approximate Gaussian distribution and a threshold value near the skirt of the approximate Gaussian distribution stored in advance intersect. To do.

  According to a third aspect of the present invention, in the second aspect of the present invention, the selection means includes a selection function having a Gaussian distribution as the selection function of the estimation having the predetermined frequency distribution, and the selection function Multiply the histogram and the selection function while changing the position at least in the range where the light quantity of the scattered reflected light and the light quantity of the regular reflected light is higher, and the peak value of the distribution obtained by multiplication is the maximum. The distribution obtained by the multiplication at the position when the position becomes is determined as the reference distribution.

  According to a fourth aspect of the present invention, in the invention of the third aspect, the display means (7c), the operation means (7b), the display means displays the histogram and the selection function, and the operation means Display control means (7a) for movably displaying the position of the selection function by means of the selection function, the selection means receiving the instruction to change the position of the selection function from the operation means, and the selection function. It is characterized by multiplying with a function.

  According to a fifth aspect of the present invention, in the first aspect of the invention, the display unit (7c), the operation unit (7b), and the display unit display a marker that can be moved by the operation unit together with the histogram. Display control means (7a) for causing the parameter determination means to select a reference means (5a) for selecting the distribution specified by the marker by the operating means as the reference distribution, and a peak vicinity of the selected reference distribution A parameter calculation means (5b) for calculating an approximate Gaussian distribution that approximates the distribution of the two and determining a parameter within a range where the approximate Gaussian distribution and a threshold value near the skirt of the approximate Gaussian distribution stored in advance intersect. It is characterized by providing.

  Parameters are determined from a two-dimensional histogram (light intensity (brightness) distribution) based on the amount of specularly reflected light and scattered reflected light at each position on the printed board, and the amount of specularly reflected light and scattered reflected light within the parameters is determined. From the corresponding reference positions, a reference position close to the position of the solder spot to be measured can be selected to measure the height of the solder spot. In addition, since a desired reference distribution is selected using a selection function, other unnecessary distributions can be removed (the selection function can also be referred to as a mask function). Also, since the selection function is a Gaussian distribution and the selection function is multiplied with the histogram to obtain the desired reference distribution, that is, the weighted distribution is extracted with the Gaussian distribution, so there is no need to be near the higher frequency distribution. The reference distribution can be obtained in a form close to a Gaussian distribution with a reduced distribution. Furthermore, the approximate Gaussian distribution is obtained by approximating the vicinity of the peak of the reference distribution to the Gaussian distribution, and the parameters are determined by comparing the vicinity of the tail of the approximate Gaussian distribution (where the frequency is low) with a desired threshold value. The parameter can be determined by removing an unstable distribution due to noise or the like in the vicinity of the tail.

  An embodiment of a printed solder inspection apparatus according to the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing a functional configuration of an embodiment of a printed solder inspection apparatus. FIG. 2 is a diagram illustrating a functional configuration of the displacement measurement unit of the measurement inspection unit of FIG. FIG. 3 is a diagram showing a functional configuration of the inspection unit in the embodiment of FIG. FIG. 4 is an example of a histogram in which the regular reflected light amount (X axis) and the scattered reflected light amount (Y axis) are two-dimensional. FIG. 5 is a diagram illustrating an example of a selection function on the two-dimensional coordinates for the regular reflection light amount (X axis) and the scattered reflection light amount (Y axis). FIG. 6 is a diagram for explaining the position change of the selection function. FIG. 7 is a diagram illustrating the reference distribution selected by the selection function. FIG. 8 is a view showing a cross section along the axis (X axis) of the regular reflection light quantity of the reference distribution of FIG. FIG. 9 is a diagram showing a cross section along the axis (Y axis) of the amount of scattered reflected light in the reference distribution of FIG. FIG. 10 is a diagram for explaining the approximate Gaussian distribution. FIG. 11 is a diagram for explaining the relationship between parameters and frequency thresholds. FIG. 12 is a diagram for explaining a parameter range. FIG. 13 is a diagram showing the tendency of the amount of regular reflected light (luminance) and the amount of scattered reflected light (luminance) with respect to near-infrared light (described above).

  1 to 12, the same reference numerals as those in FIGS. 13 to 16 described above perform the same function operations. The functional operations of the displacement sensor 2, the displacement measuring unit 10, the inspection unit 20, the scanning mechanism 30 and the control unit 40 of the present invention are as follows, unless otherwise specified. Performs basically the same operation as that of 209923. Moreover, a part may describe the same content as Japanese Patent Application No. 2008-209923.

  In FIG. 1, a displacement sensor 2 is connected to a first plate-like printed board 1 (hereinafter referred to as “substrate 1”) or a base on a flat plate (hereinafter referred to as “base”, not shown). The near infrared light source OS1a, the first light receiving means D1, the second light receiving means D2, the second near infrared light source OS2a, and the third light receiving means D3 are fixed in mutual positional relationship. Is formed. The first near-infrared light source OS1a generates any light having a wavelength within a range of 0.76 to 0.9 μm, and light (hereinafter referred to as a measurement point on the surface of the substrate 1 along the first optical axis from the vertical direction). , Which is a beam of light but is referred to as “light”). The wavelength of the first near-infrared light source OS1a is set to 0.76 to 0.9 μm because the wavelength is transmitted through the resist 1a when the resist 1a is vertically incident (approximately 0.76 μm or more). This is because it is easily available or inexpensive (approximately 0.9 μm or less).

  The first light receiving means D1 is disposed on a first scattered reflection light path formed at a measurement point at a first angle θ1 with respect to the first optical axis. The second light receiving means D2 is disposed on a second scattered reflection light path formed at a second angle θ2 on the opposite side of the first optical axis with respect to the first optical axis. The first light receiving means D1 and the second light receiving means D2 receive the scattered reflected light (the above is the “first sensing means”). The second near-infrared light source OS2a generates any light (beam) having a wavelength in the range of 0.76 to 0.9 μm, and forms a second light having an oblique angle θ3 with respect to the first optical axis. The measurement point is irradiated with the light of the axis. The third light receiving means D3 is disposed on a specular reflection optical path formed at a third angle 2θ3 with respect to the second optical axis of the second light source at the measurement point, and receives the specular reflection light. (The above is the “second sensing means”.)

  Each of the first light receiving means D1 and the second light receiving means D2 has a light receiving surface with a predetermined length, and is located at the light receiving position in the predetermined length direction when the scattered reflected light from the measurement point is received. Accordingly, a position detector (PSD: Position Sensitive Detector) that detects optical displacement information is used. In the PSD, light detection elements are arranged in the length direction, and output A and output B, which are electric quantities corresponding to the position, are output from each end as information indicating positions from both ends in the length direction. Then, the displacement information is indicated by (A−B) / (A + B). The PSD is an element that changes the light receiving position according to the displacement of the measurement point and outputs an electric quantity corresponding to the light receiving position, that is, an element that outputs the displacement expressed in the electric quantity. Instead of the PSD, the light receiving position may be directly output as information indicating the displacement by a photodiode array, a CCD, or the like.

  When the measurement point is the solder 1b, each of the first light receiving unit D1 and the second light receiving unit D2 receives the reflected light from the surface thereof to detect the displacement, and the measurement point is a resist 1a having transparency. In this case, the displacement of the bottom, that is, the displacement of the surface of the substrate 1 itself without the pad 1c or the pad 1c at the bottom is detected. The outputs A1 and B1 of the first light receiving means D1 and the outputs A2 and B2 of the second light receiving means D2 are used by the displacement measuring unit 10 to calculate height data.

  Each of the first light receiving means D1 and the second light receiving means D2 also outputs the light amount information (luminance data) at the measurement point to the displacement measuring unit 10 and the light amount storage means 3. These are used by the displacement measuring unit 10 to determine whether or not the pad 1c is present below the resist 1a (if not, the surface is the surface of the substrate 1 itself). The light quantity information stored in the light quantity storage means 3 is used to determine parameters for searching for a reference position to be described later.

  Further, the third light receiving means D3 measures the amount of reflected light because the obliquely irradiated light is reflected on the surface (not transmitted) regardless of whether the measurement point is the solder 1b or the resist 1a. Then, the light quantity information (luminance data) is output to the displacement measuring unit 10 and the light quantity storage means 3. This light quantity information is used by the displacement measuring unit 10 to determine whether the measurement point is the resist 1a or the solder 1b. The light quantity information stored in the light quantity storage means 3 is used to determine parameters for searching for a reference position to be described later.

  In FIG. 1, the scanning mechanism 30 has a drive mechanism (not shown) and scans the substrate 1 or the displacement sensor 2 or both of them by relatively moving them. In FIG. 1, the displacement sensor 2 measures displacement while performing main scanning in a direction perpendicular to the arrangement direction in which the respective elements such as the first light receiving means D1 are arranged (depth direction perpendicular to the paper surface of FIG. 1). When the main scanning is completed, the displacement is measured by moving (sub-scanning) in the arrangement direction (direction parallel to the paper surface of FIG. 1), changing the position, and performing the main scanning. By repeating this operation, the displacement of the substrate 1 in the desired range is measured. Note that the scanning direction is an example, and other directions may be used.

  In response to an instruction from the user interface 6 or the like, for example, the control unit 40 receives an instruction of a desired measurement range of the substrate 1 on which the solder is printed, and sends a scanning instruction for the desired range to the scanning mechanism 30 as described above. The irradiation position (measurement position) of light is scanned. In scanning, when the first near-infrared light source OS1a and the second near-infrared light source OS2a use light sources having the same wavelength, if there is no problem of mutual interference between specularly reflected light and scattered reflected light, these Simultaneous irradiation with a light source may be used. When this mutual interference becomes a problem, the first near-infrared light source OS1a and the second near-infrared light source OS2a are alternately switched at the same scanning position (measurement point) on the substrate 1. When light sources having different wavelengths are used, they can be distinguished and detected by the wavelength selection filter, so that they do not necessarily need to be switched and may be configured to output simultaneously.

[Configuration and operation for calculating parameters for determining the reference position]
In FIG. 1, the light quantity storage means 3 receives the amount of scattered reflected light at the measurement point from the first light receiving means D1 or the second light receiving means D2 or both while the displacement sensor 2 is scanning the substrate 1. The amount of specular reflection light at the same measurement point is received from the third light receiving means D3, and is stored in correspondence with each measurement position information of the measurement point when the scanning mechanism is scanning. The amount of scattered reflected light is preferably stored by adding both of the first light receiving means D1 and the second light receiving means D2. Either one of them may be used, but in the following example, the description will be given as the amount of scattered reflected light including both.

  The frequency calculation unit 4 receives the regular reflection light amount and the scattered reflection light amount from the light amount storage means 3 and calculates, for example, a two-dimensional histogram (frequency distribution) with the X axis as the regular reflection light amount and the Y axis as the scattering reflection light amount. . For example, the same series of addresses can be obtained by combining a combination of both an address determined by the value of the amount of specularly reflected light (upper address) and an address determined by the value of the amount of scattered reflected light (lower address). This can be achieved by configuring the number of times (frequency) to be updated and stored each time a signal comes to the address. Then, the X axis (regular reflected light amount) -Y axis (scattered reflected light amount) (hereinafter referred to as "XY") means X axis (regular reflected light amount) -Y axis (scattered reflected light amount). A histogram can be created by recording the frequency on the Z axis orthogonal to the plane of An example is shown in FIG. In FIG. 4, it can be seen that there are four large and small distributions having peaks in a mountain shape, and other hill-shaped distributions.

  The parameter determination unit 5 mainly includes a selection unit 5a that selects a basic distribution that is a desired distribution and a parameter calculation unit 5b that calculates a parameter based on the basic distribution. Here, the basic distribution is a distribution obtained based on a histogram, and is desirable for calculating a parameter (corresponding to the amount of light) necessary for obtaining a reference position where the resist 1a and the pad 1c exist. The frequency distribution above.

  The selection unit 5a stores a selection function in advance in the selection function storage unit 5a1. For example, FIG. As for the selection function, a distribution having a high probability that the resist 1a and the pad 1c exist (that is, can be a reference position) has a high value on each of the X axis (regular reflection light amount) and the Y axis (scattered reflection light amount). 4 is likely to appear as a distribution with a high Z-axis (frequency) value (which can be said to be a desired basic distribution before being selected later) (the distribution at the right end of the four chevron distributions in FIG. 4 is Therefore, the distribution function has a size that almost covers the distribution. Although it may be a distribution function in the form of a prism or cylinder, it is preferably a Gaussian distribution function. The reason why the Gaussian distribution is adopted is that when the desired basic distribution is selected by multiplying the histogram and the selection function by the multiplication means 5a2, weighting is performed by the selection function. This is because when there is an excessive distribution around the peak due to sharpening, they can be reduced. The selection function is determined empirically.

  The selection of the basic distribution by the selection means 5a includes the following methods (I) to (III), and at least one of them is adopted in the present invention. The structure used by the method which combined them may be sufficient. Moreover, the structure which comprises each and uses it by switching may be used. In any case, since the printed solder inspection apparatus handles a wide variety of substrates 1 such as substrates 1 having different sizes and substrates 1 having different positions of pads 1c and solder 1b, reflected light for each type of substrate 1 is reflected. The possibility of different distribution is great. Therefore, it is desirable to adopt a configuration that can cope with them.

(I) When selecting automatically In this case, it is convenient because it is performed automatically. This is effective when a desired distribution estimated from the reflected light amount distribution of the substrate 1 is easily separated from other distributions.
(I-a) As described above, the basic distribution is a distribution having a high probability of the presence of the resist 1a and the pad 1c. Therefore, as described above, the basic distribution is the position of the larger amount of scattered light on the histogram. And the position of the larger amount of specularly reflected light. Therefore, when the position is empirically almost deterministic and the distribution that can be the basic distribution among the plurality of peaks shown in FIG. 4 is assumed to be distant from other distributions, that is, unnecessary. When separation from a simple distribution is easy, the multiplying unit 5a2 multiplies the histogram (example; FIG. 4) calculated by the frequency calculation unit 4 by a selection function (example; FIG. 5) fixed in advance at the definite position. Thus, the basic distribution may be selected. An example of the selected basic distribution is shown in FIG.

  (Ib) If the position of the assumed basic distribution on the histogram is not deterministic and is fluid within a certain range, the distribution shape and distribution range of the selection function over the fluid range in advance, That is, the selection function storage means 5a1 stores the route to change only the position and the fineness of the distance to be changed without changing the area occupied by the tail of the selection function distribution on the histogram (the range of circles in FIG. 6). For example, as seen from the upper part of the XY plane in FIG. 6, the change route and position of the selection function are stored. Then, the multiplication means 5a2 changes the position of the selection function at the stored distance interval along the change route stored in the selection function stored in the selection function storage means 5a1, and multiplies it with the histogram. Let Then, every time the distance interval is changed and multiplied, the largest value of the frequency of the multiplication result is obtained, and when the multiplication is completed in all the change routes, the highest value of the frequency of each multiplication result is obtained. Among them, the multiplication result at the position of the selection function when the maximum value is shown is selected as the basic distribution (example: FIG. 7).

(II) When selecting by changing the position of the selection function manually In this case, when the distribution of the reflected light quantity of the substrate 1 is complicated, the operator separates the desired distribution while visually recognizing with the display means 6c. It is certain because it can be selected.
(II-a) In (1-b) above, the position of the selection function is automatically changed, but this is performed manually. In this case, the histogram calculated by the frequency calculation unit 4 and the selection function stored in the selection function storage unit 5a1 are sent to the display control unit 6a of the user interface 6, and the same histogram coordinates are displayed on the display screen of the display unit 6c. Display at the same time. That is, FIG. 4 and FIG. 5 are displayed simultaneously. Then, the display control means 6a displays the position of the selection function so that it can be changed by an instruction from the operation means 6b. Then, the operator overlays the selection function on the distribution desired to be the basic distribution while looking at the display screen of the display means 6c. Then, the multiplication means 5a2 receives an instruction to determine the position and multiplies the selection function at the position by the histogram to obtain a basic distribution.

  (II-b) The display control means 6a causes the display means 6c to display only the multiplication result of the multiplication means 5a2 on the same coordinates as the histogram (the histogram may be displayed together). Then, the display control means 6a receives an instruction to change the position of the selection function from the operation means 6b and transmits it to the multiplication means 5a2, and updates the display screen with the multiplication result of the selection function and the histogram at the changed position. When the operator recognizes a desired basic distribution as shown in FIG. 7 by repeating this change, the selection is completed.

(II-c) There is also a method of operating by employing both the display methods (II-a) and (II-b).
(II-d) The display control means 6a has a configuration for generating and displaying a movable marker for specifying the position of the display screen by the operation means 6b, and the display control means 6a displays the display screen of the display means 6c. Display a histogram. When the peak of the desired distribution is designated by the marker by the operating means 6b, the multiplication means 5a2 moves the position of the selection function to the position on the histogram designated by the marker, and multiplies the selection function and the histogram at that position. To calculate the basic distribution.

(III) When no selection function is used This is effective when the distribution of the amount of reflected light on the substrate 1 is simple and weighting by the selection function is unnecessary. In this case, similarly to the above (II-d), the display control means 6a has a configuration for generating and displaying a movable marker for specifying the position of the display screen by the operation means 6b. 6a displays a histogram on the display screen of the display means 6c. When the peak of the desired distribution is designated by the marker by the operation means 6b, the selection means 5a outputs the desired distribution as the basic distribution.

  The parameter calculation means 5b includes a distribution approximation means 5b1, a threshold storage means 5b3, and a comparison means 5b2, and a reference having a frequency distribution that is substantially peaked in a region where the scattered reflected light amount and the regular reflected light amount are high on the histogram by the selecting means 5a In response to the result of selecting the distribution, an XY two-dimensional range near the skirt of the reference distribution is determined as a parameter. The specific configuration / operation is as follows.

  The distribution approximating means 5b1 calculates a Gaussian distribution that approximates the upper part of the reference distribution received from the selecting means 5a. In other words, multiple points are selected from the reference distribution in the frequency range from the frequency peak near the upper part of the reference distribution to, for example, the frequency that is about 20% lower than the peak (preferably 50% or more). Approximate Gaussian distribution is calculated by the method of least squares. In calculating the approximate Gaussian distribution, the distribution near the tail of the reference distribution is not considered.

  The reason is as follows. For example, the reference distribution generally has a shape as shown in FIG. 7, but there may be a hill portion at the bottom of the distribution. FIG. 8 shows a cross section along the X axis of the reference distribution, and FIG. 9 shows a cross section along the Y axis of the reference distribution. As can be seen from FIG. 8, the base distribution has an asymmetrical bottom due to the influence of the hills. Accordingly, the vicinity of the skirt of the reference distribution has different distribution shapes in FIGS. Accordingly, the distribution approximating means 5b1 trusts the upper data of the frequency of the reference distribution and obtains an approximate Gaussian distribution based on the upper data, thereby eliminating the instability of the skirt of the reference distribution. The distribution is calculated and contributes to the determination of a highly reliable parameter. An example of the cross section of the reference distribution and the cross section of the approximate Gaussian distribution along the X axis is shown in FIG. As is clear from FIG. 10, the approximate Gaussian distribution is a distribution excluding the unstable tail of the reference distribution.

  The comparison unit 5b2 compares the frequency threshold value stored in the threshold value storage unit 5b3 in advance with the approximate distribution calculated by the distribution approximation unit 5b1, and the two-dimensional XY in which the frequency threshold value crosses the approximate Gaussian distribution. Determine the range as the parameter range. For example, FIG. 11 shows a cross section of the approximate Gaussian distribution and the frequency threshold along the X axis, but the positions H-Lo to H-Hi on the X axis where the approximate Gaussian distribution intersects the frequency threshold (regular reflection light quantity). Range) is a parameter range on the X-axis. Similarly, the range V-Lo to V-Hi (the range of the amount of scattered reflected light) on the Y axis is also determined. FIG. 12 shows this state from the top of the XY plane of the histogram. The range determined by H-Lo to H-Hi on the X-axis and V-Lo to V-Hi on the Y-axis in FIG. 12 is the obtained parameter range. This is because the frequency threshold value stored in the threshold value storage unit 5b3 is removed even in the approximate Gaussian distribution because the tail is gentler. This frequency threshold is determined empirically.

  The reference position inquiry means 50 of the measurement / inspection unit 100 includes parameter ranges H-Lo to H-Hi, V-Lo to V-Hi calculated by the parameter calculation means 5b (or H-Lo, H-Hi, V- The four parameters of Lo and V-Hi may be stored in advance, and the solder 1b (hereinafter referred to as “determining (measuring) the height”) for obtaining the height from the displacement measuring unit 10 is stored. The position information of the “solder 1b” is referred to as “solder location to be measured”), and from the light amount information stored in the light amount storage means 3 corresponding to the position of the substrate 1, the regular reflection light amount is H-Lo to H. A reference position closest to the position of the solder 1b at the position of the substrate 1 in the range of −Hi and the amount of scattered reflected light in the range of V−Lo to V−Hi is obtained and notified to the displacement measuring unit 10. The measurement calculation unit 13 (see FIG. 2) in the displacement measurement unit 10 knows the reference position corresponding to the solder 1b (measurement target solder location) for which the height obtained from the parameter calculation means 5b is to be measured. Based on the height of the pad 1c at the position and the height of the resist 1a, the height of the solder 1b (solder location to be measured) is calculated. At this time, if the calculation is based on the average value of the resist 1a, the reference position inquiry means 50 is a position closer in each direction around the solder 1b (solder to be measured) for which the height is to be obtained. Thus, it is necessary to obtain a plurality of reference positions where the regular reflected light amount is in the range of H-Lo to H-Hi and the scattered reflected light amount is in the range of V-Lo to V-Hi. Details of the displacement measurement unit 10 and the measurement calculation unit 13 will be described later.

[Displacement measurement unit]
As shown in FIG. 2, the displacement measurement unit 10 includes a data processing unit 11, a measurement point determination unit 12, and a measurement calculation unit 13. The data processing unit 11 and the measurement calculation unit 13 calculate the displacement (data) at the measurement point based on the output from each light receiving means. Then, the resist 1a and the solder 1b are identified by the measurement point determination unit 12, and the solder height from the bottom (pad 1c) of the resist 1a at the reference position near the measurement target solder point received from the reference position inquiry means 50 Is calculated.

  When measuring a plurality of substrates 1 having the same layout of solder locations, resist locations, and pad locations, the above-mentioned “operation for determining parameters” is performed by preliminarily measuring a representative substrate 1 once, It is desirable to store the parameters and to measure the displacement of a plurality of other substrates 1 of the same type in the main measurement based on the parameters. However, it is also possible to adopt a configuration in which “parameter calculation operation” and displacement measurement are performed for each substrate.

Specifically, in the displacement measuring unit 10, the adder 11a, the adder 11b, the calculation unit 11c, the memory 11d, the measurement point determination unit 12, and the measurement calculation unit 13 perform the following operations (1) to (4). Do.
(1) The displacement sensor 2 relatively scans the substrate 1 regardless of the solder 1b and the resist 1a under the control of the scanning mechanism 30. The first light receiving means D1 and the second light receiving means D2 receive the light. The adder 11a receives the output A1 of the first light receiving means D1 and A2 output from the second light receiving means D2, and calculates and outputs Ax = A1 + A2. The adder 11b receives the output B1 of the first light receiving means D1 and B2 output from the second light receiving means D2, and calculates and outputs Bx = B1 + B2.

  (2) The calculation unit 11c calculates a displacement output Lx = (Ax−Bx) / (Ax + Bx) from Ax = A1 + A2 and Bx = B1 + B2, and scans the displacement sensor 2 according to an instruction from the control unit 40. Corresponding to the measurement point (scanning position) of the scanning mechanism 30, it is stored in the memory 11d. At this time, the displacement output when the measurement point is the resist 1a indicates the displacement at the bottom of the resist 1a. If there is a pad 1c at the bottom, it is a displacement of the pad 1c, and if it is the substrate 1 itself, it is a displacement of its surface. That is, as described above, when the wavelength of the first near-infrared light source OS1a is 0.76 to 0.9 μm and the resist 1a is vertically incident, the resist 1a is transmitted and the pad 1c or the substrate 1 is transmitted. This is because the displacement is measured in response to the reflected light reflected by. However, at this time, it is not distinguished whether the measurement point is the displacement output of the solder 1b or the displacement output of the bottom of the resist 1a. In this way, the displacement calculated by the calculation unit 11c and stored in the memory 11d is measured based on the height of a known specific point of the substrate 1 or is calibrated at that height (for example, specific The displacement (height) of the displacement output Lx from the displacement sensor 2 at the point is calibrated to 0). As a specific point, there is a bare copper foil pattern (pad). Hereinafter, a displacement that does not indicate a reference and a displacement that does not indicate a comparison target indicate a displacement (height) based on the specific point.

  When determining the presence or absence of the pad 1c at the bottom of the resist 1a, the light quantity information of the scattered reflected light from the first light receiving means D1 and / or the second light receiving means D2 is also stored for each measurement point.

  (3) On the other hand, the measurement point determination unit 12 receives the output (brightness data) of the third light receiving means D3, and confirms that the amount of reflected light (brightness) is different between the solder 1b and the resist 1a. By using this, it is identified whether the measurement point (scanning position) of the displacement sensor 2 is the solder 1b or the resist 1a. That is, the measurement point determination unit 12 stores in advance a luminance threshold (or threshold range) for specifying the solder 1b and a luminance threshold (or threshold range) for specifying the resist 1a. The outputs of the three light receiving means D3 and their threshold values are compared to identify the solder 1b and the resist 1a. In response to the identification result, the measurement point, the displacement output Lx at the measurement point, and the identification result of the measurement point are stored in the memory 11d. Therefore, in other words, the displacement output Lx is stored separately from the displacement output of the solder 1b as Lh and the displacement output of the pad 1c at the bottom of the resist 1a as Lr.

  (4) The measurement calculation unit 13 receives the reference position information of the periphery from the reference position inquiry means 50 based on the position information of the solder 1c (measurement target solder location) whose height is to be measured. Then, the displacement output Lh when the measurement target solder point is the measurement point and the resist when the reference position in the vicinity of the measurement target solder point (where the resist 1a and the pad 1c are) are the measurement points. The displacement output Lr at the bottom of 1a (in this case, the displacement output at the surface of the pad 1c) is read from the memory 11d, and the displacement output Lr for the pad 1c at the bottom of the resist 1a at the reference position at the solder location to be measured is obtained. The height L = Lh−Lr of the reference solder 1b is obtained.

  In addition, it is preferable to obtain | require the average value of the displacement output Lr when setting it as the measurement point in the some reference | standard position of the vicinity of the solder 1b of the measurement object solder location which should obtain | require height for the displacement output Lr.

A series of operations in the embodiment will be described.
`` Preliminary measurement stage ''
Step S01: For the measurement point on the surface of the substrate 1 that is representative of the same type of substrate 1 as in the actual measurement and has the solder 1b on which the solder is printed on the pad 1c surface and the resist 1a on which the resist is applied. The first near-infrared light source OS1a having a wavelength within the range of 0.76 to 0.9 μm is vertically irradiated with light (near-infrared light), and the scattered reflected light is irradiated to the first light receiving means D1. And D2, and a displacement sensor 2 is prepared, in which the second near-infrared light source OS2a irradiates light (near-infrared light) obliquely and receives the specularly reflected light by the third light-receiving means D3.

  Step S02: The scanning mechanism 30 scans the measurement point by relatively moving either the substrate 1 or the displacement sensor 2 in accordance with an instruction from the control unit 40. The light quantity storage means 3 acquires the output of the regular reflection light quantity and the scattered reflection light quantity from the first light receiving means D1, the second light receiving means D2, and the third light receiving means D3, and stores them in correspondence with the position of the measurement point. To do.

  Step S03: The frequency calculation unit 4 generates a two-dimensional frequency distribution from the regular reflection light amount and the scattered reflection light amount at each measurement point, and displays a histogram on the display means 6c.

  Step S04: The selection unit 5a multiplies the histogram and the selection function and displays the multiplication result on the display unit 6c on the same coordinates as the histogram.

  Step S05: The operator observes the histogram and the multiplication result displayed on the display means 1c, and if OK, gives an instruction to confirm. If No, an operation for adjusting the position of the selection function is performed. Then, step S04 and step S05 are repeated until it becomes OK.

  Step S06: When the multiplication result is determined, the parameter calculation means obtains an approximate Gauss that approximates the top of the reference distribution by using the multiplication result as a reference distribution, and compares the approximate Gaussian distribution with a frequency threshold value stored in advance. Then, the value of the regular reflection light amount and the value of the scattered reflection light amount whose frequency distribution threshold crosses the approximate Gaussian distribution are calculated as parameters.

  Step S07: The reference position inquiry means 50 stores this parameter.

`` Main measurement stage ''
Step of inspecting solder shape of substrate 1 of the same type as the substrate in the preliminary measurement Step S04: The scanning mechanism 30 moves either the substrate 1 or the displacement sensor 2 in accordance with an instruction from the control unit 40. The measurement point is scanned. The displacement measuring unit 10 acquires the outputs of the first light receiving means D1, the second light receiving means D2, and the third light receiving means D3 in correspondence with the positions of the measurement points.

  Step S10: The adder 11a and the adder 11b receive one output A1 and the other output B1 from the first light receiving means D1, and receive one output A2a and the other output B2 from the second light receiving means D2. In response, Ax = A1 + A2 and Bx = B1 + B2 are calculated. At this time, the adder 11a and the adder 11b receive and calculate the output from each light receiving means for each measurement point to be scanned, and do not distinguish between the data of the solder 1b and the resist 1a. .

  Step S11: The computing unit 11c receives the outputs Ax = A1 + A2 and Bx = B1 + B2, calculates Lx = (Ax−Bx) / (Ax + Bx), and stores it in the memory 11d in association with the measurement point.

  Step S12: The measurement point determination unit 12 receives the output of the third light receiving means D3, and compares it with a prestored threshold value to determine whether the scanning position (measurement position) of the displacement sensor 2 is the solder 1b. Whether the resist is 1a is identified, and the identification result is stored in the memory 11d in association with the measurement point (measurement position) together with the displacement output Lx obtained by the calculation unit 11c.

  Step S13: The measurement calculation unit 13 notifies the reference position inquiry means 50 of the position of the solder 1b at the measurement point whose height is to be obtained from the memory 11d. The reference position inquiry means 50 detects the reference position in the vicinity of the solder 1b from the light quantity storage means 3 and informs the measurement calculation unit 13 of the detected reference position. The measurement calculation unit 13 stores the displacement output Lh when the solder 1b to be measured is the measurement point and the displacement output Lr of the pad 1c at the bottom of the resist 1a when the measurement point is the reference position. The height L = Lh−Lr of the solder 1b based on the displacement output Lr of the pad 1c is obtained by reading from 11d. This is performed for each solder 1b.

  As described above, in the present invention, the optical system of the scattering reflection displacement measurement system using near infrared light and the optical system of the regular reflection displacement measurement system are provided, and the height of the solder 1b from the bottom of the resist 1a is set. Can be sought.

  However, when obtaining the height of the solder 1b from the pad 1c at the bottom of the resist, in step S12, the measurement point determination unit 12 determines the light quantity information from the first light receiving means D1 and the second light receiving means D2, And whether the scanning position (measurement position) of the displacement sensor 2 is the solder 1b or the resist 1a in comparison with the threshold value stored in advance by receiving the output (light quantity information) of the third light receiving means D3. And the pad 1c at the bottom of the resist 1a is identified, and the identification result corresponds to the measurement point together with the displacement output Lx (Lh and Lrp are stored so as to be readable) obtained by the calculation unit 11c. In step S13, the measurement calculation unit 13 determines the displacement output Lh when the measurement point is the solder 1b and the solder 1b. Reads the displacement output Lrp pad 1c when the reference position near the vicinity is the measurement point determines the height L = Lh-Lrp solder 1b relative to the displacement output Lrp pad 1c.

  Next, the functional configuration of the inspection unit 20 will be described with reference to FIG. In FIG. 2, the image processing unit 21 receives the layout information (arrangement drawing) of the substrate 1 from the control unit 40 and the position information when scanning and measuring, and the height in the position information from the measurement calculation unit 13. (Displacement output) In response to L, a three-dimensional image representing the quantitative shape of the solder 1b is formed on the layout in accordance with the displacement, and is reproduced and output as an image.

  For example, the image processing unit 21 measures an area (measurement point collection region: for example, the substrate 1) measured from the displacement output L of the measurement calculation unit 13 and the measurement points (coordinates of the measurement points) of the substrate 1 from the control unit 40. Image data representing the area (for example, solder printed area) and volume (for example, the amount of solder) in the cream solder 1b surface printed thereon is generated. The comparison unit 22 receives a design value or the like in the area from the control unit 40 as a reference (area or volume), and calculates and outputs a difference between the image data and the reference. The difference between the displacement (height: for example, the height of the solder 1b) measured at the measurement point and the reference (in this case, for example, the design height at the measurement point) without conversion into image data. May be output. In the above, the example of the area and volume of the solder 1b has been described, but it is also possible to determine the positional deviation of the solder 1b.

  The determination unit 23 receives the allowable value corresponding to the reference from the control unit 40 and compares it with the output from the comparison unit 22. If the output of the comparison unit 22 is within the allowable value, the determination unit 23 passes, and if the output is outside the allowable value. If it is defective, it is determined as bad.

  The user interface 6 displays the determination result of the determination unit 23. In addition, layout information (for example, a layout diagram of solder locations on the printed circuit board) is received from the control unit 40 and displayed so that the position of the solder 1b in the layout is defective (not) and can be identified. May be. Further, separately or in combination, an image based on the image data generated by the image processing unit 21 can be displayed, and it can be displayed in an identifiable manner as to which part is in a poor solder state and is in a pass state. Note that the user interface 6 performs an operation as described in [Configuration and operation for calculating parameters for obtaining a reference position].

In addition, [Modification of displacement measuring unit] described in Japanese Patent Application No. 2008-209923
And [another embodiment of a displacement sensor] etc. are applicable also to the present invention.

It is a figure which shows the function structure of embodiment of a printed solder test | inspection apparatus. It is a figure which shows the function structure of the displacement measurement part of the measurement test | inspection part of FIG. It is a figure which shows the function structure of a test | inspection part among embodiment of FIG. It is an example of the histogram which makes a regular reflected light quantity (X axis) and a scattered reflected light quantity (Y axis) two-dimensional. It is a figure which shows the example of one selection function on a two-dimensional coordinate with respect to a regular reflection light quantity (X axis) and a scattered reflection light quantity (Y axis). It is a figure for demonstrating the position change of a selection function. It is a figure which shows the reference | standard distribution selected by the selection function. It is a figure which shows the cross section along the axis | shaft (X-axis) of the regular reflection light quantity of the reference | standard distribution of FIG. It is a figure which shows the cross section along the axis | shaft (Y-axis) of the amount of scattered reflected light of the reference | standard distribution of FIG. It is a figure for demonstrating approximate Gaussian distribution. It is a figure for demonstrating the relationship between a parameter and a threshold value. It is a figure for demonstrating the range of a parameter. It is a figure which shows the tendency of the regular reflection light quantity (luminance) and the scattered reflected light quantity (luminance) with respect to near-infrared light. It is a figure which shows the function structure of the conventional displacement sensor and a displacement measurement part. It is a figure which shows the cross-section of a printed circuit board. It is a figure for demonstrating the reflected light quantity in a printed circuit board.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Board | substrate (printed board), 2 Displacement sensor, 3 Light quantity memory | storage means, 4 frequency output part, 5 Parameter determination means, 5a Selection means, 5a1 Selection function storage means,
5a2 multiplication means, 5b parameter calculation means, 5b1 distribution approximation means,
5b2 comparison means, 5b3 threshold storage means, 6 user interface,
6a display control means, 6b operation means, 6c display means, 10 displacement measuring part, 11 data processing part, 11a, 11b adder, 11c calculation part,
11d memory, 20 inspection unit, 30 scanning mechanism, 40 control unit,
13 measurement calculation unit, 21 image processing unit, 22 comparison unit, 23 determination unit,
30 scanning mechanism, 40 control unit, 50 reference position inquiry means D1, first light receiving means, D2, second light receiving means, D3 third light receiving means,
OS1a first near infrared light source, OS2a second near infrared light source

Claims (5)

Displacement sensor (2) for detecting light reflected from the surface by irradiating light on the surface of the printed board on which the solder is printed, and setting a reference position on the printed board, and based on the reference position, In a printed solder inspection apparatus provided with a measurement inspection unit (100) for determining the quality of the printed solder based on the height of the solder obtained from the solder portion of the printed board.
The displacement sensor is the same as the first measurement means (OS1a, D1, D2) that irradiates near infrared light perpendicularly to the measurement point of the printed board and receives scattered reflected light from the measurement point. A second sensing means (OS2a, D3) for irradiating the measurement point with the near-infrared light at an oblique angle with respect to the vertical and receiving the specularly reflected light at the measurement point; A frequency calculation means (4) for receiving a light amount output from the sensor and generating a histogram having a frequency in a direction orthogonal to two dimensions, one being the amount of the scattered reflected light and the other being the amount of the regular reflected light; ,
On the histogram, a reference distribution including a frequency distribution that substantially peaks on the side where the amount of scattered reflected light and the amount of specular reflected light are high is selected, and the two-dimensional range near the skirt of the reference distribution is set as a parameter. Parameter determining means (5) for determining as
The measurement / inspection unit is a position on the print where the amount of scattered reflected light and the amount of specularly reflected light are within the range of the parameter, and is close to the solder location where the height of the solder is to be obtained. A printed solder inspection apparatus characterized in that a height of the solder is obtained using a position as the reference position.   The parameter determination means includes an estimated selection function having a predetermined frequency distribution that occupies a partial region of the two-dimensional range on the side where the light amount of the scattered reflected light and the light amount of the specularly reflected light are high, and from the histogram The selection function (5a) for selecting the reference distribution and an approximate Gaussian distribution approximating the distribution near the peak of the selected reference distribution are calculated by the selection function, and the approximate Gaussian distribution stored in advance is calculated. 2. The printed solder inspection apparatus according to claim 1, further comprising: parameter calculation means (5 b) that determines, as the parameter, a range where a threshold value in the vicinity of the bottom of the approximate Gaussian distribution intersects.   The selection means has a Gaussian distribution selection function as the estimation selection function having the predetermined frequency distribution, and the position of the selection function is determined by at least the amount of the scattered reflected light and the amount of the regular reflected light. Multiplying the histogram and the selection function while changing in the high range, the distribution obtained by the multiplication at the position when the peak value of the distribution obtained by multiplication becomes the maximum The printed solder inspection apparatus according to claim 2, wherein the printed solder inspection apparatus determines the reference distribution. Display means (6c), operating means (6b),
A display control means (6a) for causing the display means to display the histogram and the selection function and to display the position of the selection function movably by the operation means;
The printed solder inspection apparatus according to claim 3, wherein the selection unit multiplies the histogram and the selection function each time an instruction to change the position of the selection function is received from the operation unit. Display means (6c), operating means (6b),
Display control means (6a) for displaying a marker that can be moved by the operation means together with the histogram on the display means,
The parameter determining means calculates a selection means (5a) for selecting the distribution specified by the marker by the operation means as the reference distribution, and an approximate Gaussian distribution that approximates the distribution near the peak of the selected reference distribution. The parameter calculating means (5b) for determining the parameter within a range where the approximate Gaussian distribution and a threshold value near the skirt of the approximate Gaussian distribution stored in advance intersect. The printed solder inspection apparatus described.

Images