Preface

 

 

 

Mounting accuracy is the most basic and important process characteristic of the placement machine. It is the basis for measuring the machine's placement capability, placement quality and component placement range. Mounting accuracy, as the name suggests, refers to the accuracy of machine mounting, that is, the accuracy between the actual mounting position of the machine and the set mounting position.

 

 

 

The accuracy of a general placement machine is usually described by the following three parameters:

 

•Placement accuracy;

• Resolution;

• Replacement accuracy.

 

 

In the past, because there was no unified standard for the accuracy of placement machines, placement machines from different countries and different manufacturers defined and measured the accuracy of the machine according to their own standards. Users must understand their specific definitions when selecting and comparing the accuracy indicators of placement machines. and test conditions and methods.

 

 

 

Although the IPC-9850 standard has new regulations on the performance testing methods of placement machines, some manufacturers still use traditional methods, and a large number of placement machines produced before the publication of the IPC-9850 standard are still in use, and the second-hand mobile phone market is becoming increasingly active. , so it is still necessary to understand the accuracy of traditional placement machines. Moreover, to accurately understand and apply the IPC-9850 standard, we also need to start from the original meaning of "precision".

 

 

 

1. Analysis of precision parameters

 

 

 

1. Precision in surveying - accuracy, precision and precision

 

 

 

In surveying, accuracy, precision and precision are used to evaluate and measure the quality of measurement results. These three terms are often easily confused.

 

• Accuracy - Indicates the degree of agreement between the measurement results and the (agreed) true value of the measurement, sometimes referred to as accuracy, which reflects the degree of systematic error in the measurement results. High accuracy means that the system error is small. At this time, the average value of the measured data deviates less from the true value, but the data is dispersed, that is, the size of the repeated error is unclear.

• Precision - the degree of agreement between the results obtained when the measured object is measured multiple times under specified conditions, also known as repeat accuracy, which indicates the degree of repeat error in the measurement results. High precision means that the repetition error is small. At this time, the measurement data is relatively concentrated, but the size of the systematic error is not clear.

• Accuracy - It is a comprehensive indicator of accuracy and precision, sometimes referred to as accuracy, but it is easily confused with precision. High accuracy means that the repetition error and systematic error are relatively small, and the measurement data is relatively concentrated near the true value.

 

 

The figure below takes dart shooting as an example to illustrate the meaning of the above three words. The target value position is represented by a bullseye. The figures (a) and (b) below indicate the accuracy of shooting, that is, the size of the system error. The figures (c) and (d) below indicate the accuracy of shooting. That is, the size of the repetition error; (e) in the figure below indicates that the precision and accuracy are relatively good, which is called high accuracy. At this time, the system error and the repetition error are relatively small.

 

 

 

 

 

2. Resolution and resolution

 

 

 

Resolution has two common meanings: one refers to the minimum limit of objects that can be distinguished by instruments, meters or tools, for example, the minimum moving distance of mobile equipment (cm, mm, μm and nm), the current and voltage of electronic instruments (mV , μV, mA and μA), etc.; the second refers to the number of points or lines that display image devices (such as monitors and scanners, etc.) can distinguish within unit length. Its common units are "dpi", "lpi", "spi" and "ppi" [i.e. points (lines, samples, pixels) per inch] or "dp, lp, sp and PP)/mm", [i.e. points (lines, samples, pixels) per millimeter] etc. .

 

 

 

Resolution is an abbreviation for "resolution ability" (or "resolution ability"). Generally speaking, "resolution power" only has high and low levels, and the quantified result of resolution is called "resolution." But in practical applications, people often use "resolution" and "resolution" interchangeably.

 

 

 

3. Accuracy and resolution

 

 

 

In the performance indicators of measuring equipment, accuracy and resolution often appear as parameters of accuracy. But in fact, resolution and accuracy are not directly related. High resolution is only the basis for high accuracy. The high resolution of equipment and instruments is only a necessary condition for high accuracy, not a sufficient condition, nor a necessary and sufficient condition.

 

 

 

4. Accuracy, precision and precision of mechanical equipment

 

 

 

In mechanical equipment, it is customary to use precision and repeatability to evaluate the accuracy of machine positioning and processing, but at different times, the meaning of "accuracy" is different: when discussing "accuracy" and "repetition accuracy" specifically, " "Accuracy" refers to the "accuracy" of surveying, and "repeat accuracy" refers to the "precision" of surveying; when speaking of machine "accuracy" in general, it actually refers to the "accuracy" of surveying. The placement machine is a processing machinery and equipment, and people call it the same. Although it is unscientific, habits become natural and are not easy to change for a while.

 

 

 

Obviously, the definitions and titles of accuracy, precision and precision in surveying are more scientific. In fact, the same name as measurement is now used in teaching and research on placement machine technology, but people are accustomed to it and cannot ignore it. However, with the development and improvement of technology, especially the promotion and application of international universal standards, for example, regarding the application of the process capability index Cp and Cpk of placement machines, the unification of names and the clarity of concepts should be a matter of course.

 

 

 

2. Resolution of the placement machine

 

 

 

There are usually two types of resolutions of placement machines, one refers to the optical resolution of the identification and correction camera, and the other refers to the mechanical resolution of each movement system of the machine.

 

 

 

Resolution is the physical basis to ensure that the machine has a certain accuracy.

 

 

 

1. Resolution and accuracy of the placement machine

 

 

 

The resolution of the placement machine usually refers to the resolution of the placement machine positioning system, specifically the minimum limit of mechanical displacement, which includes the minimum distance of X-Y movement and the minimum angle of Z-axis rotation. Its value depends on the servo mechanism and axis drive mechanism. and the resolution of its measurement and control systems. In fact, each system of the placement machine, such as the optical system, various control and measurement sensors, has its own resolution.

 

 

 

The high resolution of the placement machine is the basis for ensuring the accuracy of the placement machine, just like a high-resolution instrument is the basis for ensuring the measurement accuracy. However, resolution is not directly related to accuracy. High resolution of some parts of the placement machine or high resolution of all mechanisms does not guarantee high accuracy of the placement machine; high resolution is only a necessary condition for high precision, not a sufficient condition, and more It is not a necessary and sufficient condition. Generally, resolution does not appear in the accuracy index of the placement machine. The resolution index is only used when comparing the overall performance of the placement machine in depth.

 

 

 

At present, the Z-axis movement and rotation of high-precision placement machines have reached resolutions of 0.001mm and 0.0024° respectively, and the camera resolution of the visual system has reached 0.038mm per pixel. With the development of precision manufacturing technology, these indicators will be further improved, thereby ensuring the development of placement machines with higher precision.

 

 

 

2. Identify and correct the optical resolution of the camera

 

 

 

The optical resolution of the identification and correction camera is an inherent characteristic of the camera. The resolution of the placement machine camera mainly refers to the size of the feature that can be identified by one pixel of the camera. For example, 2.3 MPP (Mil Per Pixel) refers to 2.3 thousandths in (58 μm) per pixel. The size of the pixel determines the size of the smallest feature that the machine can recognize. Generally, components with pins, such as SOIC, QFP and SOT, require 4 pixels to identify the span between each pin. If 2.3 in is used /pixel camera, then the minimum span it can recognize is: 4 × 2.3=9.2 (in) = 234 (μm), which is 0.234mm.

 

 

 

Another parameter of the placement machine camera is the pixel, which will determine the size of the camera's single field of view. Earlier cameras mostly used 640×480 pixels. If the resolution of the camera is 2.3 MPP, the size of a single field of view is 37.3mm×28.0mm. Now the pixels of newer digital cameras can reach 1024×1024 pixels, and the size of a single field of view of the 2.3 MPP camera can reach 59.8mm×59.8mm.

 

 

 

3. Mechanical resolution of the machine

 

 

 

The mechanical resolution of the machine is an inherent characteristic of the machine. It is a parameter that measures the precision of each motion axis of the placement machine. It is the basis for achieving machine accuracy. It is the minimum equivalent of mechanical displacement and the tolerance between machines. Mechanical factors that affect machine accuracy include the machining accuracy of the installation plane of the machine workbench, the resolution of the workbench drive motor and drive screw, the resolution of the encoder, the accuracy of the slide rails for axial movement of each axis, and the torsional deformation of the screw. wait. Since there are many factors that affect machine accuracy and their mechanical resolutions are different, the machine's mechanical resolution generally does not appear in the machine's technical specifications.

 

 

 

The installation plane of the machine workbench (as shown in the figure below) is generally made of a steel plate on the plane of the machine base frame that is integrally welded and milled by a machining center. The precision of the milling will determine the flatness of the platform. During milling, the machining center also needs to drill positioning holes and threaded holes at predetermined positions, and install the slide rails, drive shafts, motors and encoders of the fixed workbench. The flatness of the platform and the accuracy of the positioning holes and threaded holes will affect the parallelism of the slide rail and the accuracy of the encoder. Nowadays, the flatness and positioning accuracy of higher-end equipment can reach micron level or above.

 

 

 

 

 

The resolution of the encoder is the main factor affecting the accuracy of the machine. Common encoders include rotary encoders and linear encoders. The rotary encoder is generally installed behind the machine's motion shaft or drive motor, and feeds back the distance the table moves through the rotation of the screw, or directly feeds back the angle of rotation. The resolution of the rotary encoder is also called the resolution of the encoder. Now the resolution of the rotary encoder can reach 36000 lines per revolution, which is a resolution of 0.01 degrees. Linear encoders, also called grating rulers, can directly feedback the actual position of the worktable movement and are not affected by torsional deformation of the transmission mechanism and wear of the roller screw. The accuracy of the linear encoder of the placement machine can reach 1μm or even higher.

 

 

 

The drive and control system of the workbench is also an important factor affecting the accuracy. For the two moving axes of the workbench X and Y, there are three common drive control methods, which are introduced below.

 

 

 

① The servo motor drives the screw to drive the worktable to move in X and Y, and provides position feedback with the encoder that is coaxial with the motor or coaxial with the screw (as shown in the figure below). This control method belongs to semi-closed loop control, and its accuracy mainly depends on the resolution of the encoder and the accuracy of the lead screw. The advantages of this control method are that the control system and mechanical structure are relatively simple, the cost is low, and the stability is good. But its disadvantage is that what the encoder measures is not the actual position of the workbench, but the rotation angle of the motor or screw, and then the displacement value of the workbench is calculated. This method can only indirectly calculate the displacement of the workbench, and cannot compensate for mechanical errors and wear in the transmission link, such as belt drive errors, screw torsion deformation, and pitch errors, clearances and between the screw and balls. wear and tear etc.

 

 

 

 

 

② The servo motor drives the screw to drive the worktable to move in X and Y, and the speed measuring unit coaxial with the motor provides speed feedback. The position detection grating ruler is installed on the plane of the worktable, and a read/write head is installed on the moving part to read the movement of the worktable. Actual position and feedback (as shown in the figure below). This control method is also called fully closed-loop control. Its advantage is that it can eliminate the gaps in the mechanical transmission, compensate for the manufacturing errors of the mechanical transmission parts, and obtain higher positioning accuracy. However, its disadvantage is that the structure is relatively complex. During operation, the temperature rise of the screw shaft will cause the screw to elongate due to heat and reduce the positioning accuracy.

 

 

 

 

 

③ The linear motor is directly driven, and the position detection grating ruler provides position feedback (as shown in Figure 2.29). This control method is also a fully closed-loop control. Compared with the fully closed-loop servo motor drive, its advantages are direct drive, which eliminates the mechanical intermediate transmission link from the motor to the workbench. It has no wear, high response speed, and the drive positioning system is relatively simple. , high control accuracy.

 

 

 

 

 

3. Mounting accuracy of the placement machine

 

 

 

1. Mounting accuracy

 

 

 

Analytical placement accuracy (i.e. placement deviation), also known as positioning accuracy, describes how accurately a component is placed at a predetermined position on the PCB. The accuracy of the placement machine refers to the maximum deviation between the actual position of the placed components and the predetermined position, which reflects the degree of consistency between the actual position and the predetermined position. From the perspective of data analysis, it is equivalent to the concept of accuracy in surveying. But in terms of working characteristics and mechanism, the placement machine is closer to the CNC machine tool.

 

 

 

Since the movement of the placement machine includes the positioning accuracy of the X and Y guide rail movements, the Z-axis rotation accuracy includes two error components. In fact, when discussing placement accuracy, it is also divided into translation error and rotation error, as shown in the figure below.

 

 

 

 

 

(1) Translation error

 

The translation error is mainly caused by the movement error of the X-Y positioning system, which includes positioning error and coordinate axial error. The component alignment system cannot accurately keep the center of the component consistent with the center line of the spindle of the placement head, which is also a cause of translation errors.

 

The translation error is theoretically specified as the radius T of the placement error range centered on the target position. In fact, the translation error measurement of the placement machine is expressed by the error of the X-Y axis coordinates. Therefore, as shown in the figure below, the error radius T can be obtained by the following equation

 

 

 

 

 

 

 

In the formula, T is the true error radius caused by translation error; Xt is the error along the X axis; Yt is the error along the Y axis.

 

 

 

(2) Rotation error

 

The rotation error is caused by the error of the component positioning mechanism and the rotation error of the Z-axis. The definition of rotation error is the angular deviation between the theoretical position and the actual position, such as the angle θ in the figure below.

 

 

 

Since the vertex corresponding to the diagonal of a rectangular component is farthest from the center of the component, the rotation error at this point is the largest. For the convenience of analysis, as shown in the figure above, when the rotation error is θ, the distance R that the diagonal vertex of the component moves is called the angular displacement, which can be given by the following formula

 

 

 

 

 

In the formula, R is the angular displacement; L is the diagonal length of the rectangular component; θ is the rotation error. Sometimes it is necessary to use the error in the X-axis and Y-axis directions to represent the rotation error. The X-Y error component can be obtained by the following equation

 

 

 

In the formula,

 

 

 

(3)Total error

 

When actually mounting components, rotation errors and translation errors exist at the same time. In this case, a combined cumulative effect is produced. The total error is obtained by the vector addition of these two components. The total error components of the X-axis and Y-axis It can be obtained by the following formula

 

 

 

In the formula, Tx, Ty are the total error components of the X-axis and Y-axis; Xt, XT, Yt and YT. Then use the following formula to find the total error

 

 

 

In the formula, TPR is the total error; Tx and Ty are shown in the above two formulas.

 

 

 

Since the effect of rotational error depends on the size of the component, translational and rotational errors must be determined separately. When the type of component to be mounted is selected, the total placement accuracy can be calculated from these two values.

 

 

 

2. About IPC standardsMounting accuracy requirements

 

 

 

(1) Classification of electronic products

 

In the IPC standard, electronic products are divided into the following three levels.

 

Level 1: General consumer electronics. The use environment of this type of product is relatively stable, requiring high performance price ratio and low cost, but the requirements for reliability are not very high.

Level 2: Industrial and commercial electronics. The use environment of this type of product changes greatly, requiring both a high performance-price ratio and a long service life, and relatively high reliability requirements.

Level 3: High reliability electronic equipment. The continuous performance of this type of product is critical, downtime and other failures cannot be tolerated, and the user's environment may be harsh. Such products include high-reliability products such as life-saving equipment and military equipment, and the impact of each product is different.

 

(2) IPC-A-610 related standards

 

During SMT mounting, the ideal state is that the pins of the component completely coincide with the pads of the circuit board. However, during the placement process, this ideal state is difficult to achieve due to the positioning deviation of the circuit board, the deviation of the suction material, the deviation of the correction, the deviation of the machine positioning system and environmental factors.

 

 

 

IPC-A-610 is the acceptable standard for offset errors between components and pads after assembly and welding of printed circuit boards for three levels of electronic products. Taking chip components and SOP components as examples, it is briefly summarized as follows.

 

 

 

Level 1 and Level 2: The overlap (C) between the solder terminal width or lead width (W) of the component and the pad width (P) of the circuit board accounts for at least 50% of the component or pad width (as shown below) shown).

 

 

 

 

 

Level 3: The overlap (C) between the component solder terminal width or lead width (W) and the circuit board pad width (P) accounts for at least 75% of the component or pad width (as shown in the figure below).

 

 

 

 

 

(3) Relevant provisions of IPC9850 standard

 

IPC-A-610 only stipulates the acceptability of the offset between component leads and PCB after welding is completed, and does not involve offset errors and welding process methods after placement. In fact, if the reflow soldering process is used, since the components are in a "suspended" state during the soldering process, the reflow soldering has a "self-calibration" effect due to the wetting force, that is, the components can reduce the offset error after welding. . According to the principle of taking the most unfavorable state data in error measurement and calculation, for the evaluation of placement accuracy, we can only consider the offset error during placement.

 

 

 

Among the requirements for the accuracy of the placement machine in the IPC9850 standard, in addition to the accuracy requirements for the three axes of the machine degree of influence. The basic rationale for this approach is that many errors are caused by a combination of appropriately sized X, Y, and theta placement errors, never by any one of them. This method considers the combined impact of three types of errors, which is called Overhang or Superposition. In fact, it has been found that when the X, Y and θ axis deviations are considered separately, the specified requirements can be met, but when combined together, sometimes they do not form a qualified welded connection.

 

 

 

The benefits of this method are obvious. It is closer to the actual assembly process than the traditional belief that the performance of each axis is considered independently, because the placement accuracy is ultimately reflected in the product quality evaluation after welding is completed.

 

 

 

In the comprehensive consideration of the three-axis errors of X, Y and θ in IPC9850, two quantities are used: the total error and the lead pad coincidence degree. The total error is the maximum amount of error at the outer end of the lead. It is a quantity related to the overlap of the lead pad. For specific calculations, please refer to the IPC9850 standard "3.3.3.2.2 Use Cpk to describe the excess amount of the component lead pad (Cpk for Termination-to- Land Coverage)”. When actually measuring the accuracy of the placement machine, the total error corresponds to the lead pad overhang. For example, for a QFP with a lead width of 0.2mm, the total error (MLTE for leaded components) is required to be ≤0.100, which is equivalent to the requirement that the lead pad coverage rate is ≥75%. A total error limit of 0.15 corresponds to LTL ≥ 50%. For detailed regulations and calculations, please refer to the relevant IPC standards.

 

 

 

3. The accuracy requirements of the placement machine for mounting micro-small components

 

 

 

The accuracy requirements for placement machines for mounting micro-small components vary depending on the product type. When products are required to meet Level 3 standards, assuming that the pads or pins of the component are the same width as the pads of the circuit board, and the 01005 (0.4mm×0.2mm) component is mounted, the machine accuracy must reach 0.05mm; placement For components with a pin pitch of 0.4mm and a lead width of 0.2mm, the machine accuracy must also reach 0.05mm.

 

 

 

4. Repeatability of the placement machine

 

 

 

Repeatability describes the ability of the placement head of the placement machine to return to a certain set position repeatedly. It can also be defined as a standard deviation when placing different components at the same position on different circuit boards, sometimes also called repeatability. It reflects the degree of convergence between deviations when the placement head reaches a placement position multiple times, which is equivalent to the concept of precision in metrology. However, as mentioned earlier in terms of placement accuracy, the characteristics and working mechanism of the placement machine are closer to CNC machine tools. In the relevant digital machine tool repetition accuracy assessment method GB101131-811, the concept of repetition accuracy in the American NMTBA standard is quoted, which stipulates that when approaching in one direction, it means approaching a given point multiple times under the same conditions, and it is obtained that The dispersion with the average position μ as the center; when approaching in both directions, it means that under the same conditions, the positive and negative directions approach a given point multiple times, and the dispersion with the average position as the center of μ is obtained, with ±3σ Representation, as shown in the figure below, regarding the meaning of μ and σ, please follow the official account for subsequent explanations.

 

 

 

 

 

Fundamentally speaking, the X guide rail, Y guide rail, Z movement and Z-axis rotation of the placement machine's motion system all have their own repeat accuracy. The combined result of them and the placement accuracy determines the placement accuracy and ultimately affects the post-processing process. The process quality of process welding, therefore, in the IPC-9850 standard, the combination of various factors is used as the standard for evaluating the accuracy of the placement machine.

 

 

 

Currently, high-precision placement machines can provide repeat positioning accuracy up to micron level (0.001mm).

 

 

 

The repeatability of placement can also be understood by the degree of dispersion between the actual placement position and the target position. The relationship between accuracy and repeatability is shown in the figure below.

 

 

 

 

 

5. Determination of accuracy of placement machine

 

 

 

The measurement of accuracy will occur after the factory inspection of the placement machine and after the machine is installed at the end user's site. However, due to limitations of test conditions, the test methods are slightly different.

 

 

 

1. Inspection and accuracy verification of the placement machine before leaving the factory

 

 

 

In the chip placement machine manufacturing plant, the department responsible for quality and reliability will conduct a series of inspections and acceptance of the placement machine after the assembly is completed and before the machine leaves the factory to ensure that the placement machine can meet the machine specifications. quality standards. Generally, factory inspection is divided into the following four stages.

 

 

 

(1) Component trial placement

 

After all assembly work, including software assembly and machine calibration, is completed, the machine will first perform trial placement of components. The person responsible for machine inspection will now load the program for trial placement into the machine, install the feeder and components for trial placement onto the machine, and then run the machine for trial placement of components to verify the various functions of the machine. The products tested include some passive components and active components, such as 0402, 0603, 0805, 1206, SOT23, SOT89, PLC20, SO24, QFP100 and BGA225, etc. (as shown in the picture below), which can test all functions of the machine.

 

 

 

 

 

(2) Idling operation

 

The machine runs in dry cycle mode and can use the same procedure as when testing components. Idling operation requires at least 40 hours without interruption to ensure the stability of the machine's mechanical and electrical components.

 

 

 

(3) Correction of component placement

 

When the machine is calibrated, it only checks and corrects the hardware of the machine, such as the relative positions of each mounting axis, downward-looking camera, upward-looking camera, the magnification of each camera, the focal length and the brightness of the light, etc., while the components are inspected multiple times. There is still a certain deviation in the comprehensive average value between the actual placement position and the set position. Component placement correction (Trim) can correct the comprehensive deviation of each placement axis of the placement head, further improving the placement accuracy and capability of the machine, thereby achieving long-term stability of surface component placement performance.

 

 

 

The general method of correcting component placement is to mount two components in each of the four mounting directions (0°, 90°, 180° and 270°) according to each mounting axis, and then pass them through the lower part of the machine. The position of each mounted component is detected using a camera or coordinate measuring machine (CMM). The comprehensive deviation in each direction of each mounting axis is calculated and then compensated into the parameters of the mounting axis.

 

 

 

In order to minimize the influence factors of circuit boards and components, glass substrates are generally used as circuit boards. The length and width of the glass substrate are 400mm x 200mm. The four sides and back are reinforced with metal to prevent damage. Some fixed dots are engraved on the glass substrate as calibration and reference points (as shown in the figure below).

 

 

 

 

 

When used in the placement and correction of a multi-function placement machine, the component uses a glass standard piece equivalent to QFP100 (Pitch=0.5mm). There is a reference point on each side of the glass standard piece as a reference point for identification of the placement position (as shown in Figure 1 below) shown). The glass substrate will be pre-attached with double-sided tape. If the placement head has 7 placement axes, and each placement axis mounts two components in each of the 4 directions, then a total of 56 components are mounted (as shown in Figure 2 below).

 

 

 

Figure 1 QFP100 glass standard components

 

 

 

 

 

Figure 2 Distribution of QFP100 components during mounting and decoration

 

 

 

When used in the placement and correction of high-speed placement machines, the components are 0201 standard components with good appearance. Double-sided tape will be pre-attached to the glass substrate. If the placement head has 30 placement axes, and each placement axis mounts two components in each of the four directions, then a total of 240 components are mounted (as shown in the figure below).

 

 

 

 

 

After the standard component is mounted, run the placement inspection program and specify the placement machine configuration and standard component model. The machine's downward-looking camera will first identify 6 reference points on the glass substrate to determine the position of the glass substrate, and then identify the two opposite reference points on each glass standard piece or the opposite corners of the 0201 component. By calculating the positions of two points on each standard component, the deviation in the X and Y directions and the angle deviation of the mounting position of this standard component can be obtained, as shown in the figure below.

 

 

 

 

 

For the same mounting axis, the average deviation of two standard components mounted in the same direction is the deviation of the mounting axis in this direction. After the positions of all standard components are verified, the software will automatically calculate and compensate the deviation of each placement axis into the parameters of the placement axis.

 

 

 

(4) Verification of placement accuracy and capability

 

Before the machine leaves the factory, the placement accuracy and capabilities of the machine need to be verified to ensure that the machine can meet the parameter indicators in the specification and obtain a factory certificate.

 

 

 

The method used to verify the placement accuracy and capability is similar to the correction method for component placement. They both use the same glass bottom plate as the circuit board and use glass standard components as components. The method and quantity of component placement can also be the same. The correction of component placement is the same, as shown in Figure 2.40. What is different from the verification of mounting accuracy and capability before leaving the factory is that the mounting position of glass standard components will be measured using special coordinate measuring equipment, as shown in the figure below, which is often called CMM (Coordinate Measurement Machine). The generally used CMM has a stable workbench that is not easily affected by the environment and a high-resolution downward-looking camera. The measurement accuracy is generally several times higher than the accuracy of the placement machine (the accuracy of commonly used CMMs is below 2.5 μm, which is less than 2.5 μm for placement machines. The accuracy of the chip machine is more than ten times).

 

 

 

 

 

The data required for CMM measurement and calculation, such as the size of the glass substrate, the position of the reference point on the substrate, the mounting position of the glass standard component, and the data of the glass standard component, can be transmitted to the CMM system in a specified format. Run the automatic detection software, and the CMM will detect the reference points on the substrate and the two diagonal reference points on each glass standard component. After the inspection is completed, the CMM software can automatically calculate the accuracy and capability of the entire glass test component placement according to the settings, that is, the average deviation (Mean), standard deviation σ (Standard Deviation) and process capability of the component placement measurement values. Index Cpk, as shown in the figure below.

 

 

 

 

 

The standard component tested is QFP100, and the upper and lower limits of the placement accuracy of this machine are ±65μm. The actual measurement accuracy and capabilities of this machine's X-axis are described below.

 

 

 

Average deviation of placement measurement values (Mean)

 

 

 

 

 

Standard deviation σ (Standard Deviation)

 

 

 

 

 

Process Capability Index (Cpk)

 

 

 

 

 

2. The accuracy of the placement machine on site

 

 

 

After the inspection and placement machine is installed on-site on the production line, the accuracy and capabilities of the machine can also be inspected on-site. Due to limited conditions, there is generally no coordinate measuring equipment on site, and the same inspection method as before the placement machine leaves the factory cannot be used. There are generally two methods to test machine accuracy and capabilities on site.

 

 

 

(1) Use the machine’s downward-looking camera for inspection

 

Generally, the placement machine has the function of using the machine's downward-looking camera to inspect the placement accuracy. The method is roughly the same as the inspection method before leaving the factory.

 

 

 

The glass substrate and standard glass original or standard components used in the test, as well as the mounting method and component distribution are the same as those used in the accuracy inspection before leaving the factory. After the standard components are mounted, do not transfer the glass substrate out of the machine. Run the inspection program that comes with the machine. The machine's downward-looking camera will automatically detect the reference points of the glass substrate and the standard components. After the inspection is completed, the machine's software can also automatically calculate the accuracy and capabilities of the placement machine.

 

 

 

This method is relatively simple, easy to operate, and is not restricted by conditions. However, since the same machine is used for the placement and testing of standard components, errors in the machine itself cannot be ruled out. Therefore, the final accuracy and capability values of the placement machine are not very accurate.

 

 

 

(2) Detection using a microscope

 

In order to conduct more precise accuracy inspection using simpler conditions at the production line site, visual inspection using a microscope can also be used. The test uses a glass substrate with graduations on the mounting position, as shown in Figure 1 below. There are 28 mounting positions with scales on the glass substrate, which is suitable for testing the accuracy and capability of mounting heads with less than 7 mounting axes. There are two circles of scales at each mounting position, and data such as −15, −10, 5, 0, 5, 10 and 15 are marked in 4 directions. There is a reference point at each of the 4 corners of the mounting position. like