Flexible PCBA processing positioning optimization method

Abstract

The invention provides a flexible PCBA (Printed Circuit Board Assembly) processing positioning optimization method, which relates to the technical field of accurate positioning control, and is characterized in that a temperature field area is divided based on temperature data, a mapping relation between a temperature gradient and pad displacement is established, and the thermal displacement of a reference pad is dynamically compensated in combination with a fitted stress-strain function. Meanwhile, the dynamic alignment control of the mounting head is optimized, the problem of pad position offset caused by thermal expansion and deformation in the flexible PCBA processing process can be effectively solved, the problem of accurate positioning of dislocation of an IC chip and a pad caused by thermal expansion and local deformation of a substrate in the COF mounting process of the flexible PCBA can be solved to a certain extent, and the accuracy of chip mounting is improved. Therefore, the processing quality and reliability of the flexible PCBA are obviously improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of precise positioning control, and more particularly, to a flexible PCBA processing positioning optimization method. BACKGROUND

[0002] With the rapid development of electronic products towards light and thin, small and high performance, flexible printed circuit board assembly (Flexible Printed Circuit Board Assembly, referred to as flexible PCBA) has been widely used in consumer electronics, automotive electronics, medical equipment and other fields. Flexible PCBA has excellent flexible and bendable characteristics, which makes it have significant advantages in complex spatial layout and dynamic application scenarios. However, compared with traditional rigid circuit boards, flexible PCBA faces more technical challenges in manufacturing and mounting processes, especially in IC chip mounting. Due to the thermal expansion characteristics of flexible substrates and the complexity of their stress state, mounting precision and stability are difficult to guarantee. Traditional PCBA mounting methods mostly rely on the fixed position of the reference pad for alignment, but flexible materials are prone to thermal expansion and deformation in high temperature environment, which causes the actual position of the reference pad to deviate from the designed position, thereby affecting the mounting precision. In addition, the high integration and miniaturization trend of flexible PCBA puts higher requirements on the precision control of mounting equipment, and the existing technology still has obvious shortcomings in dynamic compensation and high-precision alignment.

[0003] The existing flexible PCBA mounting optimization technology mainly focuses on improving the mechanical precision of the mounting equipment, optimizing the pad design layout, or using static compensation methods, but the effect of these methods is limited. For example, the improvement of mechanical precision is limited by processing technology and cost, and pad layout optimization can only play a role in the design stage and cannot dynamically adapt to the problem of thermal deformation in actual processing. Some technologies try to correct the pad position through static compensation methods, but they ignore the stress and strain distribution differences of flexible PCBA in different temperature field regions, and the compensation effect is not accurate enough. At the same time, due to the nonlinear characteristics of the thermal expansion and stress distribution of flexible PCBA, the existing technology has great limitations in establishing a thermal displacement compensation model. In addition, the lack of dynamic adjustment capability of X-Y direction position and mounting angle in the mounting process is also an important factor affecting the mounting precision. SUMMARY

[0004] In order to solve the above technical problems, the present application is proposed. The present application provides a flexible PCBA processing positioning optimization method. To a certain extent, the present application can solve the precise positioning problem of IC chip and pad misalignment caused by substrate thermal expansion and local deformation in the COF mounting process of flexible PCBA

[0005] According to one aspect of the present application, a flexible PCBA processing positioning optimization method is provided, which comprises:

[0006] Collecting reference pad temperature data and position data of a to-be-mounted area on the flexible PCBA;

[0007] Based on the temperature data, dividing the to-be-mounted area into two temperature field areas of a center area and an edge area, calculating temperature gradient values of the center area and the edge area, and establishing a mapping relationship between pad coordinates and temperature gradients;

[0008] According to the temperature gradient values, fitting stress-strain curves of the center area and the edge area into a segmented continuous function, calculating thermal displacement amounts of reference pads in each area through a stress-strain function, and continuously dynamically compensating the position data of the reference pads;

[0009] Based on the compensated reference pad position data, controlling a mounting head of a COF mounting device, adjusting X-Y direction positions and mounting angles of the mounting head, and accurately aligning pins of an IC chip with reference pads.

[0010] Further, the temperature distribution data includes real-time temperature values of each reference pad in the to-be-mounted area;

[0011] The temperature distribution data is obtained by a first detection device arranged above a work platform;

[0012] The position data is obtained by a second detection device arranged above the work platform.

[0013] Further, establishing the mapping relationship between the pad coordinates and the temperature gradients comprises the following steps:

[0014] Performing temperature field partition processing on the to-be-mounted area to obtain a temperature similarity S(P), when the temperature similarity S(P) is greater than a preset threshold, dividing any point P to the center area, otherwise, dividing it to the edge area;

[0015] Respectively establishing local temperature field descriptions of the center area and the edge area;

[0016] Based on the local temperature field expressions of the center area and the edge area, calculating a temperature gradient field to obtain a temperature gradient vector;

[0017] Based on the temperature gradient vector, constructing a mapping function of the pad coordinates.

[0018] Further, the mapping function of the pad coordinates is as follows:

[0019]

[0020] Wherein,

[0021]

[0022] wherein M(x, y) is the mapped target coordinate, δ is a displacement coefficient, is a unit direction vector determined by the temperature gradient direction, κ is a vorticity influence coefficient, represents the vorticity of the temperature gradient field, and η(x, y) is a thermal stress influence factor.

[0023] Further, the piecewise continuous function is divided according to the region, when in the central region, based on the material mechanics test data and the temperature gradient influence analysis, the stress-strain curve is divided into three characteristic intervals, including a linear elastic interval, an elastic-plastic transition interval and a plastic deformation interval;

[0024] When in the edge region, the stress-strain curve of the central region is modified by using a boundary effect correction term.

[0025] Further, the temperature gradient value of the central region and the edge region includes:

[0026] According to the temperature gradient value of the position of each reference pad, combined with the stress-strain curve at the position, the strain value of the pad under the action of the temperature gradient is calculated;

[0027] Based on the strain value, the displacement vector of the pad is calculated by using the thermoelastic equation.

[0028] Further, the continuous dynamic compensation of the reference pad position data includes:

[0029] An associated equation of the pad displacement amount and time is established;

[0030] According to the associated equation, the compensation amount at each time is calculated, and the compensation amount is superimposed on the original position data of the pad.

[0031] Further, the compensation amount is calculated as follows:

[0032] P compensated (x, y, t) = P original (x, y) + (Δx(P, t) Δy(P, t))·R(θ)

[0033] wherein P compensated (x, y, t) is the compensated reference pad position coordinate, P original (x, y) is the original design position coordinate of the reference pad, Δx(P, t) represents the thermal displacement component of point P in the x direction at time t, ΔP(t) is a time response term, Δy(P, t) represents the thermal displacement component of point P in the y direction at time t, and R(θ) is a local rotation matrix caused by thermal deformation:

[0034]

[0035] The rotation angle theta is determined by the curl of the temperature gradient field:

[0036]

[0037] Where K is the rotation sensitivity coefficient, θ is the curl of the temperature gradient field.

[0038] Further, based on the compensated reference pad position data, the center coordinates and the direction reference line of the target pad array are calculated, and then the motion trajectory planning of the mounting head is generated according to the relative deviation between the IC chip pin position and the target pad position.

[0039] Further, the center coordinates and the direction reference line of the target pad array are calculated by dividing the quadrant, selecting the weighted reference point, removing the points with large deviation, and fitting the orthogonal reference axis, so as to accurately calculate the geometric center and the direction reference line of the pad array, and finally realize the center position and the direction reference determination with strong anti-interference ability and accurate positioning.

[0040] Compared with the prior art, the flexible PCBA processing positioning optimization method provided by the application can effectively solve the problem of pad position deviation caused by thermal expansion and deformation in the flexible PCBA processing process, and can to a certain extent solve the problem of accurate positioning of IC chips and pads caused by thermal expansion and local deformation of the substrate in the COF mounting process, thereby significantly improving the quality and reliability of the flexible PCBA processing. BRIEF DESCRIPTION OF DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the application, and those skilled in the art can obtain other drawings according to these drawings without creative labor. In the drawings:

[0042] Figure 1 The flow chart of the flexible PCBA processing positioning optimization method according to the embodiment of the application. DETAILED DESCRIPTION

[0043] ​Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part but not all of the embodiments of the present application, and the present application can be implemented in many different forms. Therefore, the present application should not be limited to the embodiments described herein, but should be defined by the appended claims in conjunction with the full text of the specification.

[0044] Figure 1 A flow chart of a flexible PCBA processing positioning optimization method according to an embodiment of the present application. As shown in FIG. 1, in the flexible PCBA processing positioning optimization method, it includes: Figure 1

[0045] S1: Collecting the reference pad temperature data and position data of the area to be mounted on the flexible PCBA, real-time monitoring the temperature distribution of the reference pad by an infrared thermometer, and acquiring the actual coordinates of the reference pad by a CCD camera;

[0046] Placing the flexible PCBA to be measured on the working platform of the COF mounting equipment, wherein the first detection device and the second detection device are arranged above the working platform, the first detection device includes an infrared thermometer for detecting the temperature distribution of the reference pad on the flexible PCBA, the infrared thermometer uses an infrared thermal imaging sensor with a wavelength range of 8-14 μm, the field of view angle of the infrared thermal imaging sensor is 32°×24°, the spatial resolution is 1.1 mrad, and the temperature measurement accuracy is ±0.5℃; the second detection device includes a CCD camera for detecting the position of the reference pad on the flexible PCBA, the resolution of the CCD camera is 2448×2048 pixels, the field of view is 15 mm×12 mm, and the imaging frequency is 60 fps; first, starting the infrared thermometer to collect the temperature distribution data of the reference pad with a sampling period of 0.1 s, wherein the temperature distribution data includes the real-time temperature values of each reference pad in the area to be mounted; then, starting the CCD camera to collect the image data of the reference pad, processing the image data by a sub-pixel edge detection algorithm, extracting the contour features of the reference pad, and calculating the actual coordinate values of the center point of the reference pad in the working platform coordinate system, wherein the calculation accuracy of the coordinate values is better than 1 μm; finally, storing the temperature distribution data and the actual coordinate values in the data buffer for subsequent temperature field division and position compensation.

[0047] S2: Based on the temperature data, dividing the area to be mounted into two temperature field regions of a center region and an edge region, calculating the temperature gradient values of the center region and the edge region, and establishing a mapping relationship between the pad coordinates and the temperature gradient;

[0048] ​Firstly, based on the above temperature distribution data, the improved region growing algorithm is used to divide the temperature field of the region to be mounted. Considering that the traditional region growing algorithm only relies on temperature threshold, which can easily lead to over-segmentation or under-segmentation, an improved region growing criterion based on temperature and spatial distance double constraints is proposed. Specifically, for any point P in the region to be mounted, its temperature value is denoted as T(x, y), and a comprehensive similarity judgment function is introduced, which not only considers the temperature difference, but also introduces a spatial distance attenuation term, which ensures the continuity and smoothness of the judgment through the exponential function form, as shown in the following formula:

[0049]

[0050] Where S(P) is the temperature similarity, T c is the temperature value of the growing starting point, d(x, y) is the Euclidean distance from point P to the starting point, σ is the temperature standard deviation, R is the characteristic distance parameter, and α is the weight coefficient and takes the value range [0, 1]. When S(P) is greater than the threshold θ (value 0.85), point P is divided into the center area, otherwise it is divided into the edge area.

[0051] It should be noted that in the temperature similarity judgment, the influence degree of temperature difference is described by a Gaussian type exponential function, where the temperature standard deviation σ is obtained through pre-experiment, and the value is 1 / 6 of the temperature range of the region to be mounted. At the same time, the spatial distance attenuation term is introduced, and the characteristic distance parameter R is set to 1 / 4 of the diagonal length of the region to be mounted, which can effectively avoid the situation that the far distance point is wrongly divided. The weight coefficient α is used to balance the relative importance of temperature difference and spatial distance, and through a large number of experiments, when α takes the value of 0.7, the best partition effect can be obtained.

[0052] After completing the region division, the temperature field accurate description of the center area and the edge area needs to be established respectively. For the center area, considering that the temperature distribution has strong regularity and continuity, the improved two-dimensional temperature field Taylor expansion expression T c (x, y) is adopted. Unlike the traditional Taylor expansion, this scheme introduces an additional nonlinear correction term in the expansion expression, which adopts a Gaussian function form and can better describe the local fluctuation characteristics of the temperature field, as shown in the following formula. The selection of expansion order n is based on the complexity of the temperature field, and usually 3-4 orders can meet the accuracy requirements. The determination of coefficient a ij uses the weighted least squares method, and the weight has an exponential decay relationship with the distance from the center.

[0053]

[0054] Where T c (x, y) is the temperature field expression of the center area, T0 is the temperature value of the center point, a ijis the Taylor expansion coefficient matrix, with dimension of [T] · [L] -i , where i represents the expansion order, j represents the term number at this order, β is the nonlinear correction coefficient with dimension of [T] · [L] -2 , γ is the decay factor with dimension of [L] -2 , r is the distance to the center point, x, y are the coordinate values in the local coordinate system.

[0055] For the edge zone temperature field, due to its more complex and irregular distribution, an improved radial basis function network is used for description. The multi-quadratic radial basis function is selected as the basic function form, which has the advantages of good smoothness and numerical stability. The selection of control points adopts an adaptive strategy, increasing the density of control points in the area with large temperature gradient. In particular, a boundary effect correction term is introduced, which describes the temperature gradient mutation characteristics at the boundary through an exponential decay function, where the parameters λ and μ are optimized and determined by the boundary temperature sampling points:

[0056]

[0057] where T e (x, y) is the edge zone temperature field expression, is the multi-quadratic radial basis function, w k is the weight coefficient with dimension of [T], P k is the control point coordinate, P b is the nearest boundary point coordinate, λ is the boundary effect correction parameter with dimension of [T], μ is the boundary decay coefficient with dimension of [L] -1 , m is the total number of control points, and P is the coordinate of the point to be calculated.

[0058] Based on the above temperature field expression, the temperature gradient field of each region is calculated. The analytical derivation method is used for temperature gradient field calculation, which has higher accuracy and continuity compared with the finite difference method. That is, the temperature gradient vector at point P(x, y) is:

[0059]

[0060] where, represents the gradient of the center zone temperature field T c (x, y), represents the gradient of the edge temperature field T e (x, y).

[0061] ​In establishing the mapping relationship between the pad coordinates and the temperature gradient, the displacement superposition form is adopted, wherein the displacement direction is determined by the temperature gradient direction, and the displacement amount is proportional to the thermal stress influence factor. In particular, the thermal stress influence factor η(x, y) is introduced, which not only considers the size of the temperature gradient, but also introduces the influence of the curl of the temperature gradient field, which makes the description of the thermal stress distribution more comprehensive and accurate. That is, the mapping function of the pad coordinates is defined as:

[0062]

[0063] wherein,

[0064]

[0065] wherein, M(x, y) is the target coordinate after mapping, δ is the displacement coefficient, is a unit directional vector determined by the temperature gradient direction, k is a curl influence coefficient, and the dimension is [L] 2 ·[T] -1 , represents the curl of the temperature gradient field, and η(x, y) is the thermal stress influence factor.

[0066] It should be noted that the determination of the displacement coefficient δ adopts an iterative optimization method, and the dimension is [L] 2 ·[T] -1 , which is obtained by minimizing the mean square error between the predicted position and the measured position. In order to improve the stability of the optimization, an L2 regularization term is introduced, and the regularization coefficient ρ is determined by cross-validation:

[0067]

[0068] wherein, θ represents the set of parameters to be optimized, ρ is the regularization coefficient, N is the number of sampling points, M measured is the measured pad position coordinate.

[0069] The entire mapping calculation process adopts a parallel computing architecture, and tasks such as region division, temperature field calculation, and gradient field solution are distributed to multiple computing units to perform simultaneously, thereby ensuring the 0.1s update period requirement. The mapping relationship established is stored in a dynamically updated mapping table, which adopts a sparse matrix form and records the original coordinates, temperature gradient values, thermal stress influence factors, and target coordinates after mapping of each reference pad position. The update of the mapping table adopts an incremental update strategy, and only the areas that have changed significantly are recalculated, which not only ensures the calculation efficiency, but also ensures the mapping accuracy.

[0070] S3: fitting the stress-strain curves of the center region and the edge region as a piecewise continuous function according to the temperature gradient value, calculating the thermal displacement of the reference pads in each region by a stress-strain function, and continuously dynamically compensating the reference pad position data, wherein the coefficients of the stress-strain function change nonlinearly with the temperature gradient value;

[0071] First, according to the temperature gradient field distribution, the stress-strain response model of the center region and the edge region is constructed. Since the flexible PCBA substrate material will exhibit obvious nonlinear mechanical properties under the action of different temperature gradients, and the characteristics change with the change of temperature gradient, an accurate stress-strain description model needs to be established. Considering the difference in mechanical properties of the material in the low strain and high strain regions, and the influence of temperature gradient on the mechanical properties of the material, a piecewise continuous function is used to describe the stress-strain relationship, ensuring the continuity of the first derivative of stress and strain at the strain segmentation point, and avoiding stress mutation.

[0072] For the center region, based on the material mechanics test data and temperature gradient influence analysis, the stress-strain curve is divided into three characteristic intervals: linear elastic interval (0≤ε<ε1), elastic-plastic transition interval (ε1≤ε<ε2) and plastic deformation interval (ε2≤ε≤εmax). In each interval, the stress-strain function needs to consider the basic elastic modulus, the influence of temperature gradient and the nonlinear hardening effect, and its expression is:

[0073]

[0074] Wherein, the first interval mainly shows linear elastic characteristics, and the influence of temperature gradient is corrected by the quadratic term; the second interval begins to appear plastic deformation, and the cubic term is introduced to describe the strain hardening effect; the third interval shows obvious nonlinear characteristics, and a larger strain hardening coefficient is needed to describe the limit behavior of the material. In practical application, the values of E1, E2, E3 need to be obtained by material mechanics test, α1, α2, α3 are determined by temperature gradient calibration test, β1, β2 are obtained based on strain hardening curve fitting, ε1, ε2 are strain segmentation points, and εmax is the maximum allowable strain value. max is the modulus value of temperature gradient.

[0075] For the edge region, due to the existence of obvious boundary effect, the stress-strain relationship will be affected by the boundary constraint. In order to accurately describe this influence, a boundary effect correction term is introduced on the basis of the stress-strain function of the center region:

[0076]

[0077] The boundary distance attenuation function λ(d) adopts the form of Gaussian function:

[0078] ​

[0079] where ω represents the strength of the boundary effect, and d0 is a characteristic decay distance. The temperature gradient modulation function The hyperbolic tangent function is adopted:

[0080]

[0081] where d is the distance to the boundary, and γ1 and γ2 are temperature gradient modulation parameters with dimensions of [L]·[T] -1 and [L] 2 ·[T] -2 .

[0082] On the basis of the stress-strain curve, the thermal displacement of the reference pads in each region is calculated. Specifically, first, according to the temperature gradient value of the position where each reference pad is located, the strain value of the pad under the action of the temperature gradient is calculated in combination with the stress-strain curve at the position; then, based on the strain value, the displacement vector of the pad is calculated using the improved thermoelastic equation, wherein the calculation of the displacement vector needs to consider the non-uniformity of the temperature field and the cumulative effect of the deformation path; for the pads in the edge region, an additional displacement correction amount caused by the boundary effect also needs to be considered:

[0083]

[0084] where K(s) is a path integral kernel function:

[0085]

[0086] where E eff (s) is the equivalent elastic modulus on the path, ξ is a temperature field divergence influence coefficient, is the divergence of the temperature gradient field, Δx(P) is the thermal displacement component of the pad in the x direction, Δy(P) is the thermal displacement component of the pad in the y direction, and σ(ε) is the stress response function of the material.

[0087] It should be noted that the selection of σ(ε) is based on the position of point P and uses σ e (ε) or σ c (ε) as needed. That is, when point P is in the center region, σ c (ε) is used, and when point P is in the edge region, σ e (ε) is used.

[0088] Finally, the position data of the reference pads is continuously and dynamically compensated. In the compensation process, firstly, an association equation of pad displacement and time is established, which contains a quasi-static displacement term and a dynamic response term. The quasi-static displacement term is directly determined by the above-mentioned thermal displacement, and the dynamic response term considers the transient response characteristics caused by the change of temperature gradient; then, the compensation amount at each time is calculated according to the association equation, and the compensation amount is superimposed on the original position data of the pads; in the compensation process, each parameter is updated in real time with the change of the temperature gradient field, and the update period is synchronized with the temperature field sampling, so as to ensure the continuity and accuracy of the compensation. More specifically, the association equation of the pad displacement and the time is as follows:

[0089]

[0090] Wherein, ΔP(t) is the time response term, ΔP0(t) is the quasi-static displacement, τ is the first-order time response coefficient, and η is the second-order time response coefficient.

[0091] The position compensation amount of the reference pad is calculated by the following expression:

[0092] P compensated (x,y,t)=P original (x,y)+ΔP(t)·(Δx(P,t)Δy(P,t))·R(θ)

[0093] Wherein, P compensated (x,y,t) is the compensated reference pad position coordinate, P original (x,y) is the original design position coordinate of the reference pad, and R(θ) is the local rotation matrix caused by thermal deformation:

[0094]

[0095] The rotation angle θ is determined by the curl of the temperature gradient field:

[0096]

[0097] Wherein, κ θ is the rotation sensitivity coefficient, and its dimension is [L] 2 ·[T] -1 .

[0098] The verification of the compensation result is completed by the online measurement system. When the position deviation of the compensated pads is less than the preset threshold, it is considered that the compensation result meets the process requirements; if the deviation exceeds the threshold, the parameters of the stress-strain curve need to be fine-tuned, and the compensation calculation process is executed again until the accuracy requirements are met. The whole compensation process fully considers the nonlinear deformation characteristics, boundary effect and dynamic response characteristics of the flexible PCBA, and ensures the accuracy and reliability of the position compensation.

[0099] S4: Based on the compensated reference pad position data, control the mounting head of the COF mounting equipment, adjust the X-Y direction position and mounting angle of the mounting head, so that the pins of the IC chip are accurately aligned with the reference pads.

[0100] First, the reference pad position data after thermal stress compensation is transmitted to the main control system of the COF mounting equipment, which includes a motion control module and a pose adjustment module. Among them, the motion control module is responsible for controlling the translational motion of the mounting head in the X-Y plane, and the pose adjustment module is responsible for controlling the rotational motion of the mounting head around the Z axis. The mounting head uses vacuum adsorption to fix the IC chip, and the bottom of the mounting head is provided with a high-precision force sensor for real-time monitoring of the contact force during mounting.

[0101] Before performing the mounting action, first obtain the actual position information of the IC chip pins by the visual positioning system. The visual positioning system includes two orthogonally arranged high-speed CCD cameras, respectively used to obtain the X direction and Y direction images of the IC chip pins. Based on the obtained image information, the profile features of the IC chip pins are extracted by sub-pixel edge detection algorithm, and the geometric center coordinates and direction angle of the pins are calculated. At the same time, based on the compensated reference pad position data, the center coordinates and direction reference line of the target pad array are calculated.

[0102] It should be noted that the calculation process adopts the partition weighted positioning method. First, the compensated reference pads are divided into four quadrant regions according to their distribution positions. Three reference pads closest to the to-be-measured pad array are selected as positioning reference points in each quadrant region. When the number of reference pads in a certain quadrant is less than three, the reference pads in the adjacent quadrant closest to the quadrant are selected to supplement. For each reference point in each quadrant, a weight coefficient is set according to the distance from the reference point to the center of the to-be-measured pad array. The closer the distance, the greater the weight. If the position deviation of a certain reference pad is detected to exceed a preset threshold, the weight coefficient of the pad is automatically reduced, and the reference point is directly excluded when the deviation is too large. Then, the compensated position coordinates of the remaining reference pads are used to fit a set of orthogonal reference axes by the least square method, wherein the X-axis direction is parallel to the long side of the pad array, and the Y-axis direction is parallel to the short side. When the pad array presents a non-rectangular distribution, the principal component analysis method is used to determine the reference axis direction. Next, in the orthogonal coordinate system, the geometric centers of all reference pads are calculated, and the actual center position of the target pad array is calculated according to the design parameters of the pad array with the geometric centers as the starting points. Finally, the determined orthogonal reference axes are taken as the reference to extract the direction reference line of the array. The reference line is usually the connecting line of the outermost pads of the pad array, and when the outermost pads are unevenly distributed, the innermost pads are selected as auxiliary references. The center coordinates and direction reference line calculated in this way have strong anti-interference ability and positioning accuracy, and provide reliable position references for subsequent accurate alignment.

[0103] Then, the host system generates a motion trajectory planning of the mounting head according to the relative deviation between the IC chip pin position and the target pad position. The trajectory planning adopts a five-order polynomial interpolation algorithm to ensure the continuity of the acceleration of the mounting head in the motion process. The motion in the X-Y plane is driven by a double-shaft servo motor, and the position accuracy is better than ±1 μm, and the repeat positioning accuracy is better than ±0.5 μm. The rotating motion of the mounting head is driven by a high-precision voice coil motor, and the angle resolution is better than 0.001 degrees.

[0104] During the mounting process, the host system adopts a feedforward-feedback composite control strategy to adjust the motion state of the mounting head in real time. The feedforward control is based on the position instruction calculated by the trajectory planning, and the feedback control is corrected based on the real-time data of the vision system and the force sensor. When it is detected that the mounting force exceeds the preset range, the system automatically adjusts the pressing speed and the pressure size. At the same time, the relative position between the IC chip pin and the pad is monitored through a real-time image processing algorithm to ensure the alignment accuracy.

[0105] Finally, after completing the precise alignment, the mounting head presses the IC chip against the surface of the flexible PCBA at a preset pressure and speed. During the pressing process, the feedback signal of the force sensor ensures uniform distribution of pressure, avoiding local stress concentration. When the pressure is detected to be stable and the positional deviation is within the allowable range, the system determines that the mounting is completed, the mounting head is lifted back to the standby position, and a precise mounting cycle is completed. During the entire mounting process, the system continuously monitors the changes in the ambient temperature and the changes in the reference pad position, and makes real-time compensation adjustments as needed to ensure mounting accuracy and reliability.

[0106] In summary, the flexible PCBA processing positioning optimization method based on the embodiments of the present application is illustrated. By dividing the temperature field region based on temperature data, establishing the mapping relationship between temperature gradient and pad displacement, and combining the fitted stress-strain function to dynamically compensate the thermal displacement of the reference pad, while optimizing the dynamic alignment control of the mounting head, the problem of pad positional deviation caused by thermal expansion and deformation during the processing of the flexible PCBA can be effectively solved. To a certain extent, the precise positioning problem of the IC chip and the pad misalignment caused by the thermal expansion and local deformation of the flexible PCBA during the COF mounting process can be solved, thereby significantly improving the quality and reliability of the flexible PCBA processing.

[0107] Here, those skilled in the art can understand that the specific operations of each step in the above flexible PCBA processing positioning optimization method have been described in detail above with reference to the description of the flexible PCBA processing positioning optimization method of Figure 1 , and therefore, the repeated description will be omitted.

Claims

1. A method for optimizing the positioning during flexible PCBA processing, characterized in that, include: Collect reference pad temperature and position data for the area to be mounted on the flexible PCBA; Based on the temperature data, the area to be mounted is divided into two temperature field regions: the central region and the edge region. The temperature gradient values ​​of the central region and the edge region are calculated, and a mapping relationship between the pad coordinates and the temperature gradient is established. Based on the temperature gradient value, the stress-strain curves of the central region and the edge region are fitted into piecewise continuous functions. The thermal displacement of the reference pads in each region is calculated using the stress-strain function, and the position data of the reference pads are continuously and dynamically compensated. Based on the compensated reference pad position data, the placement head of the COF placement equipment is controlled to adjust the XY direction position and placement angle of the placement head so that the pins of the IC chip are precisely aligned with the reference pad.

2. The flexible PCBA processing positioning optimization method according to claim 1, characterized in that, The temperature distribution data includes the real-time temperature values ​​of each reference pad in the area to be mounted; The temperature distribution data is acquired through a first detection device positioned above the work platform. The location data is acquired through a second detection device positioned above the work platform.

3. The flexible PCBA processing positioning optimization method according to claim 1, characterized in that, Establishing the mapping relationship between the pad coordinates and the temperature gradient includes the following steps: Temperature field partitioning is performed on the area to be mounted to obtain temperature similarity S(P). When the temperature similarity S(P) is greater than a preset threshold, any point P is assigned to the center area; otherwise, it is assigned to the edge area. Local temperature field descriptions for the central and peripheral regions are established separately; Based on the local temperature field expressions of the central and edge regions, the temperature gradient field is calculated to obtain the temperature gradient vector. Based on the temperature gradient vector, a mapping function for the pad coordinates is constructed.

4. The flexible PCBA processing positioning optimization method according to claim 3, characterized in that, The mapping function for the pad coordinates is shown in the following formula: in, Where M(x,y) are the mapped target coordinates, and δ is the displacement coefficient. The vector is a unit direction vector, determined by the direction of the temperature gradient, and κ is the curl influence coefficient. η(x,y) represents the curl of the temperature gradient field, and η(x,y) is the thermal stress influence factor.

5. The flexible PCBA processing positioning optimization method according to claim 1, characterized in that, The piecewise continuous function is divided according to the region. When it is in the central region, based on the material mechanics test data and temperature gradient influence analysis, the stress-strain curve is divided into three characteristic intervals, including the linear elastic interval, the elastoplastic transition interval, and the plastic deformation interval. When in the edge region, the stress-strain curve of the central region is corrected using the boundary effect correction term.

6. The flexible PCBA processing positioning optimization method according to claim 1, characterized in that, Calculating the temperature gradient values ​​of the central region and the edge region includes: Based on the temperature gradient value at the location of each reference pad, and combined with the stress-strain curve at that location, the strain value of the pad under the action of the temperature gradient is calculated. Based on the strain value, the displacement vector of the pad is calculated using the thermoelastic equation.

7. The flexible PCBA processing positioning optimization method according to claim 1, characterized in that, Continuous dynamic compensation of the reference pad position data includes: Establish the correlation equation between pad displacement and time; The compensation amount is calculated at each moment based on the correlation equation, and the compensation amount is superimposed on the original position data of the pad.

8. The flexible PCBA processing positioning optimization method according to claim 7, characterized in that, The compensation amount is calculated as shown in the following formula: P compensated (x,t,t)=P original (x,y)+(Δx(P,t)Δy(P,t))·R(θ) Among them, P compensated (x,y,t) represents the compensated coordinates of the reference pad position, P original (x,y) represents the original design position coordinates of the reference pad, Δx(P,t) represents the thermal displacement component along the x-direction at point P at time t, ΔP(t) is the time response term, Δy(P,t) represents the thermal displacement component along the y-direction at point P at time t, and R(θ) is the local rotation matrix caused by thermal deformation. The rotation angle θ is determined by the curl of the temperature gradient field: Among them, κ θ This is the rotational sensitivity coefficient. This represents the curl of the temperature gradient field.

9. The flexible PCBA processing positioning optimization method according to claim 1, characterized in that, Based on the compensated reference pad position data, the center coordinates and orientation reference line of the target pad array are calculated. Then, based on the relative deviation between the IC chip pin position and the target pad position, the motion trajectory planning of the mounting head is generated.

10. The flexible PCBA processing positioning optimization method according to claim 9, characterized in that, include: The center coordinates and orientation reference lines of the target pad array are calculated by dividing quadrants, selecting weighted reference points, eliminating points with excessive deviations, and fitting orthogonal reference axes. This allows for the accurate calculation of the geometric center and orientation reference lines of the pad array, ultimately achieving a center position and orientation reference determination that is highly resistant to interference and has accurate positioning.

Images