A multi-stage cleaning method for inhibiting pin fin black spot defects of an IGBT heat dissipation substrate

By employing a multi-stage cleaning method, combined with ultrasonic and microfluidic jet technology, the problem of black spot defects on the pin fins of IGBT heat sink substrates was solved, achieving a highly efficient cleaning effect and improving product quality and production efficiency.

CN122032941BActive Publication Date: 2026-06-26HUANGSHAN GUANGJIE SURFACE TREATMENT TECH CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
HUANGSHAN GUANGJIE SURFACE TREATMENT TECH CO LTD
Filing Date
2026-04-17
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

In the existing technology, black spot defects are prone to occur in the needle fin structure of IGBT heat dissipation substrate during the chemical nickel plating process. This is mainly due to incomplete cleaning, especially in the narrow-spaced needle fin area where oil stains are difficult to remove, which affects the product qualification rate and production cost.

Method used

A multi-stage cleaning method is adopted, including ultrasonic coarse cleaning and microfluidic jet fine cleaning. Combined with real-time detection technology, the cleaning parameters and nozzle posture are optimized to ensure the cleanliness of the needle fin surface. Through the multi-stage cleaning method of iterative optimization of ultrasonic cleaning parameters and microfluidic jet fine cleaning, all-round cleaning of the heat dissipation substrate surface is achieved.

Benefits of technology

It significantly reduced the incidence of black spot defects on the heat dissipation substrate pin fins, improved product quality and coating adhesion, met the requirements of downstream customers, and reduced production costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122032941B_ABST
    Figure CN122032941B_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of multi-stage cleaning of IGBT heat dissipation substrates, and solves the technical problem of black spot defects after chemical nickel plating caused by insufficient cleaning of oil stains in the pin fin microstructure of the heat dissipation substrate, especially a multi-stage cleaning method for inhibiting black spot defects of IGBT heat dissipation substrates. In the rough cleaning stage, the key parameters of the ultrasonic generator and the transducer are optimized, the heat dissipation substrate is subjected to rough cleaning, and the surface cleanliness is detected in real time. In the fine cleaning stage, the posture of the hyperspectral camera is accurately controlled for omnidirectional detection of the pin fin, the nozzle posture is adjusted based on the key information of the oil-stained pin fin, the nozzle output flow and pressure are controlled as the optimization target of the micro-flow column jet impact force, and the oil-stained pin fin is subjected to accurate cleaning. The present application realizes multi-angle omnidirectional oil stain cleaning and real-time cleanliness detection of each area on the surface of the heat dissipation substrate, thereby improving product quality to meet the requirements of downstream customers and effectively improving the core competitiveness in the industry.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of multi-stage cleaning technology for IGBT heat sink substrates, and in particular to a multi-stage cleaning method for suppressing black spot defects on the pin fins of IGBT heat sink substrates. Background Technology

[0002] IGBTs (Insulated Gate Bipolar Transistors) are core semiconductor power devices in power electronic systems, widely used in new energy vehicles, industrial drives, and consumer electronics. With the increasing trend towards higher frequencies and higher power densities, IGBTs generate significant heat during switching, leading to elevated junction temperatures and significantly impacting device performance and reliability. Traditional heat dissipation solutions (such as natural convection or simple air cooling) are insufficient for demanding operating conditions, especially in scenarios like electric vehicle motor controllers, where thermal management failures can cause system shutdowns or even safety accidents. Therefore, high-efficiency heat dissipation substrates have become a key technological support for overcoming the thermal bottleneck of IGBTs. To improve heat dissipation efficiency, substrate design has evolved from planar to three-dimensional. Currently, the mainstream heat dissipation substrate is a copper plate with a pin-fin arrangement. This type of substrate increases the surface area through the pin-fin structure, achieving direct liquid cooling and avoiding the accumulation of thermal resistance caused by the thermal grease layer.

[0003] IGBT heat sinks often require nickel plating to improve thermal conductivity, corrosion resistance, and solderability, making it the mainstream choice for IGBT module heat sinks. IGBT heat sinks often have a finned structure, and to improve plating uniformity and adhesion, a chemical nickel plating process is used. The chemical nickel plating process for IGBT copper-based finned heat sinks includes degreasing, surface roughening, activation and desmearing, chemical nickel plating, post-treatment, and quality inspection. The degreasing process uses an alkaline degreasing agent to thoroughly remove residual cutting oil and contaminants from machining, ensuring a clean interface for subsequent processing. Chemical nickel plating requires a very clean substrate surface for nickel alloy deposition; therefore, cleaning is a crucial pre-treatment step in the chemical nickel plating process.

[0004] The problem of black spots appearing on the pin fins during the electroless nickel plating process of copper plates in heat dissipation components is a significant concern. This defect is difficult to remove, severely impacting product yield and failing to meet downstream customer requirements, thus increasing production costs. Analysis of surface defects and the characteristics of the production process reveals that the black spots on the pin fins are primarily caused by incomplete removal of oil stains during the pre-plating cleaning process, especially noticeable in the narrow-spaced pin fin areas. Currently, production involves a single ultrasonic cleaning cycle, without real-time monitoring of the substrate surface cleanliness during this process. Defects are only visually inspected after the electroless nickel plating process, resulting in a lack of quantitative quality control in the intermediate production stages. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a multi-stage cleaning method for suppressing black spot defects in the pin fins of IGBT heat dissipation substrates. This method solves the technical problem of insufficient oil removal in the microstructure of the pin fins of heat dissipation substrates, which is difficult to address in existing technologies, leading to black spot defects after chemical nickel plating. This improves product quality, meets the requirements of downstream customers, and effectively enhances the core competitiveness in the industry.

[0006] To address the aforementioned technical problems, this invention proposes a multi-stage cleaning method for suppressing black spot defects on the pin fins of IGBT heat sink substrates. This method includes the following steps:

[0007] Acquire multi-source parameters, including equipment parameters, geometric parameters of the heat dissipation substrate, and process parameters, during the cleaning process of the heat dissipation substrate;

[0008] The heat dissipation substrate is subjected to ultrasonic rough cleaning based on multi-source parameters, and the surface cleanliness of the heat dissipation substrate is detected in real time during the rough cleaning process to determine whether it meets the first preset condition.

[0009] If so, the cleanliness of the needle fin surface after ultrasonic coarse cleaning is detected in real time, and the coordinate information of the oil-stained needle fins on the heat dissipation substrate and the height information of the oil stains are obtained.

[0010] If not, optimize the ultrasonic cleaning parameters in the rough cleaning stage until the result of the first real-time detection meets the first preset condition.

[0011] Based on the coordinate information of the oil-stained needle wings and the height information of the oil stains, the oil-stained needle wings are subjected to microfluidic jet cleaning, and the cleanliness of the needle wing surface is detected in real time during the cleaning process to determine whether the second preset condition is met.

[0012] If so, then end the cleaning operation;

[0013] If not, then adjust the microfluidic jet impact force during the microfluidic jet washing process until the second preset condition is met.

[0014] Furthermore, the first preset condition is whether the area of ​​the region where the oil stains have not been cleaned, as identified by the first real-time detection, is less than or equal to a first area threshold.

[0015] The second preset condition is that the area of ​​oil stains on the needle wings being cleaned is less than or equal to the second area threshold.

[0016] Further, the ultrasonic rough cleaning of the heat dissipation substrate includes:

[0017] The radiated sound pressure of the transducer Output frequency and the output current of the ultrasonic generator As a constraint, the output power of the ultrasonic generator is... and the air gap of the transducer The ultrasonic cleaning parameters were obtained by iterative optimization.

[0018] The ultrasonic generator and transducer are controlled according to the optimized ultrasonic cleaning parameters to perform ultrasonic cleaning on the heat sink substrate.

[0019] Furthermore, optimizing the ultrasonic cleaning parameters and performing ultrasonic rough cleaning includes the following steps:

[0020] S21. Set the initial values ​​of the ultrasonic generator parameters, including setting the initial value of the output power. Output power adjustment steps ;

[0021] S22. Set optimization counting variables initial value And calculate the optimal increment of the ultrasonic generator output power. The calculation formula is:

[0022] ;

[0023] In the formula, , These are the maximum and minimum power for driving the ultrasonic generator, respectively;

[0024] S23. Based on the optimized increment Calculate the output power of the ultrasonic generator The calculation formula is:

[0025] ;

[0026] In the formula, , is the initial value of the adjustment coefficient for the output power of the ultrasonic generator;

[0027] S24. Based on the rated current of the ultrasonic generator. Executes the output current of the ultrasonic generator Conditional judgment of relevant constraints, namely:

[0028] Judgment Condition 1: Judgment Is it true? If so, proceed to step S25.

[0029] If not, then execute condition 2: judgment Is it valid?

[0030] If so, then set the initial value of the adjustment coefficient. Optimize counting variables Proceed to step S23;

[0031] If not, proceed to step S21;

[0032] S25, Based on output power Calculate the radiated sound pressure of the transducer The calculation formula is:

[0033] ;

[0034] In the formula, The acoustic impedance of the cleaning fluid; The electroacoustic conversion efficiency of the transducer; The effective area of ​​the transducer radiating head;

[0035] Based on the minimum allowable sound pressure level radiated by the transducer Execution and radiated sound pressure The judgment conditions for relevant constraints are as follows:

[0036] judge If the condition is met, proceed to step S26; otherwise, proceed to step S21.

[0037] S26. Set the initial value of the transducer's air gap. and incremental optimization of air gap For the air gap of the transducer Optimize;

[0038] S27. Set the temperature of the cleaning solution. and the initial concentration of cleaning agent injected into the cleaning solution. Set the optimal increment for the cleaning agent content in the cleaning solution. Set initial values ​​for the optimization variables and calculate the number of optimization steps. ,Right now:

[0039] ;

[0040] In the formula, The maximum amount of cleaning agent that can be injected into the cleaning solution;

[0041] S28. Calculate the cleaning agent content in the optimized cleaning solution. The calculation formula is:

[0042] ;

[0043] in, The number of optimization steps is a variable used to optimize the cleaning agent content.

[0044] S29. Start the ultrasonic coarse cleaning and initialize the count variable for the number of cleaning cycles. ;

[0045] S210. Using a water film continuity test, machine vision is used to identify the back and sides of the heat sink substrate with a hyperspectral camera, and the area of ​​the heat sink substrate where oil stains have not been cleaned is calculated. ;

[0046] And perform conditional checks related to the first area threshold, namely:

[0047] Judgment condition 5: Determine the area of ​​the region Is it valid?

[0048] If so, then the condition is determined. Is it true? If so, then let , Proceed to step S27; otherwise, proceed to step S26.

[0049] If not, the cleanliness of the needle fin surface after ultrasonic coarse washing is detected in real time.

[0050] Furthermore, the optimized air gap Includes the following steps:

[0051] S261. Set optimization variables And calculate the number of air gap optimization steps for the transducer. The calculation formula is:

[0052] ;

[0053] In the formula, , These are the maximum and minimum values ​​of the transducer air gap, respectively;

[0054] S262. Based on the allowable deviation value of the ultrasonic generator output current. Executes the output current of the ultrasonic generator Conditional judgment of relevant constraints, namely:

[0055] Judgment condition 3: Judgment Is it true? If so, proceed to step S263.

[0056] If not, then determine Is it true? If so, then let the optimization variable... Initial value of the air gap adjustment coefficient of the transducer Proceed to step S264; otherwise, proceed to step S261.

[0057] S263. Measure the output frequency of the transducer. And based on the resonant frequency of the transducer and the allowable deviation value of the output frequency. Execution and output frequency Conditional judgment of relevant constraints, namely:

[0058] Judgment condition 4: Judgment Is it true? If so, proceed to step S27.

[0059] If not, then determine Is it true? If so, then let Proceed to step S264; otherwise, proceed to step S261.

[0060] S264. Calculate the optimized air gap of the transducer. Then proceed to step S262, the calculation formula is:

[0061] ;

[0062] In the formula, To optimize variables; This is the initial value of the air gap adjustment coefficient for the transducer.

[0063] Furthermore, the cleanliness of the needle fin surface after ultrasonic rough cleaning is detected in real time, and the coordinate information and oil height information of the oil-stained needle fins on the heat dissipation substrate are obtained, including the following steps:

[0064] S31. Establish a coordinate system on the heat sink substrate, and mark the origin of the coordinate system at the position of the first row and first column of the pin fins on the surface of the heat sink substrate. The coordinates of this position are: ;

[0065] S32. Using fluorescence detection method to show the oil stains present on the pin fin array on the heat dissipation substrate;

[0066] S33. Set the attitude parameters of the hyperspectral camera, including the distance between the hyperspectral camera and the surface of the substrate where the needle fin is located. The angle between the axial direction of the hyperspectral camera lens and the axial direction of the needle fin. Horizontal distance between the center of the hyperspectral camera lens and the axis of the needle fin A hyperspectral camera detects the movement path of a single needle wing;

[0067] S34. Adjust the posture of the hyperspectral camera to perform machine vision recognition on the back and sides of the heat sink substrate based on posture parameters;

[0068] S35. Set the scanning parameters for the hyperspectral detection needle-fin array, including the scanning method as a transverse S-shaped sequential scan, and the starting point of the scan as... The horizontal spacing of the scan is and scanning longitudinal spacing ;

[0069] S36. Scan the pin array using a hyperspectral camera, identify the oil stains on each pin, and count the number of pins with oil stains in the pin array on the heat dissipation substrate. Coordinates of the oil-stained needle-wings , The center height of the oil-stained area on the oil-stained needle wings ;

[0070] And execute the number of oil-stained needles. Conditional judgment of relevant constraints, namely:

[0071] Judgment Condition 6: Judgment If the condition is met, the needles with oil stains will be subjected to microfluidic jet cleaning and a second real-time detection will be performed; otherwise, the cleaning operation will be terminated.

[0072] Furthermore, adjusting the attitude of the hyperspectral camera includes:

[0073] S341. Calculate the maximum value of the angle between the axial direction of the hyperspectral camera lens and the axial direction of the needle fin. and minimum value and determine the conditions If the condition is met, proceed to step S342; otherwise, proceed to step S33.

[0074] Maximum value and minimum value The calculation formula is:

[0075] ;

[0076] In the formula, This function represents the minimum value of two numbers. , These are the horizontal and vertical spacing of the heat dissipation substrate pin fins, respectively. , These are the height and radius of the heat dissipation substrate pin fins, respectively;

[0077] S342. Calculate the actual object distance of the hyperspectral camera lens. and determine the conditions If the condition is met, proceed to step S35; otherwise, proceed to step S33.

[0078] Actual object distance The calculation formula is:

[0079] ;

[0080] in, , These represent the maximum and minimum object distances allowed under conditions where the hyperspectral camera lens can capture images clearly.

[0081] Furthermore, the microfluidic jet cleaning of the oil-containing needle fins includes:

[0082] Based on the coordinate information of the oil-stained needle fins, control the jet nozzle to move to the corresponding needle fin position;

[0083] Based on the height information of the oil stains, the attitude and output parameters of the jet nozzle are adjusted to spray and wash the needle fins containing the oil stains.

[0084] Furthermore, adjusting the output parameters of the jet nozzle includes:

[0085] Based on the oil stain height information and the relative positions of the jet nozzle and the needle fins, the impact force of the microfluidic jet acting on the oil stain area is calculated.

[0086] With the goal of achieving a preset threshold range for the impact force, the output flow rate and output pressure of the jet nozzle are adjusted.

[0087] Furthermore, the needles with oil contamination are subjected to microfluidic jet cleaning and a second real-time detection, including the following steps:

[0088] S41. Set the initial value of the variable representing the needle fin with oil contamination. ;

[0089] S42. Set the attitude of the jet nozzle, including the angle between the jet nozzle axis and the needle fin axis. Horizontal distance between the jet nozzle and the needle axis The distance between the jet nozzle and the substrate surface where the needle fins are located The direction and speed of rotation of the jet nozzle around the axis of the needle fin And calculate the actual distance between the jet nozzle and the center of the oil-stained area on the needle fin. The calculation formula is:

[0090] ;

[0091] in, The distance between the hyperspectral camera and the substrate surface where the needle fins are located; The angle between the axial direction of the hyperspectral camera lens and the axial direction of the needle fin; The horizontal distance between the center of the hyperspectral camera lens and the axis of the needle fin; The height of the center of the oil-stained area on the oil-stained needle fin; The radius of the heat dissipation substrate pin fins;

[0092] S43. Set the number of cleaning cycles for the microfluidic jet. The optimized increments for output flow and output pressure are respectively , Initial value of the output flow rate of the jet nozzle Initial pressure value ;

[0093] S44, Based on actual distance Calculate the impact force of the current microfluidic jet. And perform conditional judgments related to the impact force constraints, namely:

[0094] Judgment Condition 7: Judgment If the condition is met, proceed to step S45; otherwise, proceed to step S43.

[0095] Impact The calculation formula is:

[0096] ;

[0097] in, The minimum allowable injection pressure for the jet nozzle; The density of the cleaning solution; The outlet radius of the jet nozzle; The coefficients for calculating the impact force of microfluidic column jet washing are provided. This is the critical distance at which the jet cross-section does not diverge; It is a natural exponential function;

[0098] S45. Based on the coordinates of the needles with oil stains. The jet nozzle is quickly moved to the needle fin and sprayed;

[0099] S46. Fluorescence detection was used to examine the sprayed needle wings, and the area of ​​oil stains on the needle wings was statistically analyzed using a hyperspectral camera. And perform conditional judgments related to the second area threshold constraint, namely:

[0100] Judgment condition 8: Judgment If the condition is met, proceed to step S47; otherwise, end the cleaning operation.

[0101] S47, Increase the output flow rate of the jet nozzle. Output pressure Calculate the impact force of the microfluidic jet at this time. And perform conditional checks on the constraints related to the impact force, namely:

[0102] Judgment Condition 9: Judgment Is it true? If so, then let Proceed to step S45; otherwise, proceed to step S43.

[0103] Microfluidic jet impact force The calculation formula is:

[0104] ;

[0105] in, The maximum allowable injection pressure for the jet nozzle;

[0106] S48. Judgment Conditions Is it true? If so, then let Proceed to step S42; otherwise, end the cleaning operation.

[0107] By employing the above technical solution, the present invention provides a multi-stage cleaning method for suppressing black spot defects on the pin fins of IGBT heat sink substrates, which has at least the following beneficial effects:

[0108] In the ultrasonic coarse cleaning stage, the key parameters related to the ultrasonic generator and transducer are optimized with the radiated sound pressure, output current and output frequency as optimization targets, and the heat dissipation substrate is cleaned multiple times. The cleanliness of the back and sides of the heat dissipation substrate after coarse cleaning is detected in real time using a hyperspectral camera based on machine vision recognition technology.

[0109] In the microfluidic jet cleaning stage, the output flow rate and output pressure of the jet nozzle are controlled with the impact force of the microfluidic jet as the optimization target, and the oil-stained needle fins are precisely cleaned according to the detection results of the hyperspectral camera, thereby realizing multi-angle and all-round oil stain cleaning and real-time cleanliness detection of various areas on the surface of the heat dissipation substrate.

[0110] In summary, this invention proposes a decontamination method that combines ultrasonic multi-stage cleaning with microfluidic jet cleaning, and monitors the surface cleanliness of the heat dissipation substrate in real time at different stages of cleaning, thereby improving product quality to meet the requirements of downstream customers and effectively enhancing core competitiveness in the industry.

[0111] This invention effectively reduces the probability of black spot defects in the pin fin area and improves the coating quality of the heat dissipation substrate by real-time and precise control of the cleanliness of the substrate surface during the cleaning process. Attached Figure Description

[0112] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0113] Figure 1 This is a flowchart of the multi-stage cleaning method in this invention;

[0114] Figure 2This is a schematic diagram illustrating the cleaning detection scanning method used in the microfluidic column jet cleaning stage of the present invention;

[0115] Figure 3 This is a schematic diagram illustrating the determination of the coordinate system during the microfluidic column jet washing stage of the present invention;

[0116] Figure 4 This is a schematic diagram illustrating the determination of the hyperspectral camera attitude during the microfluidic column jet washing stage of the present invention.

[0117] Figure 5 This is a schematic diagram illustrating the determination of nozzle orientation during the microfluidic jet cleaning stage of the present invention.

[0118] Figure 6 This is a comparison chart showing the results of cleaning and electroless nickel plating on the 110300500100 heat dissipation substrate in this invention.

[0119] Figure 7 This is a comparison chart showing the results of cleaning and electroless nickel plating on the 20020100012 heat dissipation substrate of this invention.

[0120] Figure 8 This is a comparison of the results after cleaning and electroless nickel plating of the C7100487 heat dissipation substrate in this invention. Detailed Implementation

[0121] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. This will allow for a full understanding of how the present application uses technical means to solve technical problems and achieve technical effects, and to facilitate its implementation.

[0122] This embodiment proposes a multi-stage cleaning method to suppress black spot defects on the pin fins of IGBT heat sink substrates. It employs a combination of ultrasonic cleaning and microfluidic jet cleaning, with ultrasonic cleaning serving as a coarse wash and microfluidic jet cleaning as a fine wash. The cleanliness of the heat sink substrate surface is monitored in real time at different cleaning stages. In the microfluidic jet fine wash stage, the attitude of a hyperspectral camera is precisely controlled to achieve omnidirectional detection of the pin fins, acquiring information such as the coordinates of the oil-stained pin fins, the area of ​​the oil stains, and the height of the oil stains on the pin fin surface. Based on this key information, the nozzle attitude and nozzle output parameters are determined, thereby achieving an efficient cleaning process for the pin fin surface. Figure 1 As shown, the method includes the following steps:

[0123] S1. Obtain equipment parameters, geometric parameters of the heat sink substrate, and process parameters related to the cleaning process, including the maximum power driving the ultrasonic generator. Minimum power and rated current Maximum value of transducer air gap and minimum value The resonant frequency of the transducer The maximum allowable concentration of cleaning agent to be injected. Electroacoustic conversion efficiency of the transducer Effective area of ​​the transducer radiating head Acoustic impedance of the cleaning fluid The maximum allowable injection pressure of the jet nozzle. and minimum injection pressure Minimum permissible sound pressure level radiated by the transducer ; Lateral spacing of the heat sink base plate pins Longitudinal spacing ,high and radius Initial value of the output power adjustment coefficient of the ultrasonic generator Initial value of the air gap adjustment coefficient of the transducer The allowable deviation value of the output current of the ultrasonic generator. The allowable deviation value of the transducer output frequency. Density of the cleaning fluid The outlet radius of the jet nozzle ; Calculation coefficients of impact force in microfluidic column jet washing The maximum object distance allowed under conditions where the hyperspectral camera lens captures images clearly. and minimum object distance .

[0124] S2. Parameter optimization for the ultrasonic coarse cleaning stage, including the following steps:

[0125] S21. Set the initial values ​​of the ultrasonic generator parameters, including setting the initial value of the output power. Output power adjustment steps ;

[0126] S22. Set optimization counting variables initial value And calculate the optimal increment of the ultrasonic generator output power. The calculation formula is:

[0127] ;

[0128] In the formula, , These are the maximum and minimum power for driving the ultrasonic generator, respectively;

[0129] S23. Based on the optimized increment Calculate the output power of the ultrasonic generator The calculation formula is:

[0130] ;

[0131] In the formula, , is the initial value of the adjustment coefficient for the output power of the ultrasonic generator;

[0132] S24. Based on the rated current of the ultrasonic generator. Executes the output current of the ultrasonic generator Conditional judgment of relevant constraints, namely:

[0133] Judgment Condition 1: Judgment Is it true? If so, proceed to step S25.

[0134] If not, then execute condition 2: judgment Is it valid?

[0135] If so, then set the initial value of the adjustment coefficient. Optimize the counting variable Proceed to step S23;

[0136] If not, proceed to step S21;

[0137] S25, Based on output power Calculate the radiated sound pressure of the transducer The calculation formula is:

[0138] ;

[0139] In the formula, The acoustic impedance of the cleaning fluid; The electroacoustic conversion efficiency of the transducer; The effective area of ​​the transducer radiating head;

[0140] Based on the minimum allowable sound pressure level radiated by the transducer Execution and radiated sound pressure The judgment conditions for relevant constraints are as follows:

[0141] judge If the condition is met, proceed to step S26; otherwise, proceed to step S21.

[0142] S26. Set the initial value of the transducer's air gap. and incremental optimization of air gap For the air gap of the transducer Optimize;

[0143] As a further implementation of this embodiment, the air gap is optimized. Includes the following steps:

[0144] S261. Set optimization variables And calculate the number of air gap optimization steps for the transducer. The calculation formula is:

[0145] ;

[0146] In the formula, , These are the maximum and minimum values ​​of the transducer air gap, respectively;

[0147] S262. Based on the allowable deviation value of the ultrasonic generator output current. Executes the output current of the ultrasonic generator Conditional judgment of relevant constraints, namely:

[0148] Judgment condition 3: Judgment Is it true? If so, proceed to step S263.

[0149] If not, then determine Is it true? If so, then let the optimization variable... Initial value of the air gap adjustment coefficient of the transducer Proceed to step S264; otherwise, proceed to step S261.

[0150] S263. Measure the output frequency of the transducer. And based on the resonant frequency of the transducer and the allowable deviation value of the output frequency. Execution and output frequency Conditional judgment of relevant constraints, namely:

[0151] Judgment condition 4: Judgment If the condition is true, proceed to step S27; otherwise, determine the condition. Is it true? If so, then let Proceed to step S264; otherwise, proceed to step S261.

[0152] S264. Calculate the optimized air gap of the transducer and proceed to step S262. The calculation formula is:

[0153] ;

[0154] In the formula, To optimize variables; This is the initial value of the air gap adjustment coefficient for the transducer.

[0155] S27. Set the temperature of the cleaning solution. and the initial concentration of cleaning agent injected into the cleaning solution. Set the optimal increment for the cleaning agent content in the cleaning solution. Set initial values ​​for the optimization variables and calculate the number of optimization steps. ,Right now:

[0156] ;

[0157] In the formula, The maximum amount of cleaning agent that can be injected into the cleaning solution;

[0158] S28. Calculate the cleaning agent content in the optimized cleaning solution. The calculation formula is:

[0159] ;

[0160] in, The number of optimization steps is a variable used to optimize the cleaning agent content.

[0161] S29. Start the ultrasonic coarse cleaning and initialize the count variable for the number of cleaning cycles. ;

[0162] S210. Using a water film continuity test, machine vision is used to identify the back and sides of the heat sink substrate with a hyperspectral camera, and the area of ​​the heat sink substrate where oil stains have not been cleaned is calculated. ;

[0163] And perform conditional checks related to the first area threshold, namely:

[0164] Judgment condition 5: Determine the area of ​​the region Is it valid?

[0165] If so, then the condition is determined. Is it true? If so, then let , Proceed to step S27; otherwise, proceed to step S26.

[0166] If not, proceed to step S3.

[0167] S3. Real-time detection of the surface cleanliness of the needle wings after ultrasonic coarse washing, including the following steps:

[0168] S31. Establish a coordinate system on the heat sink substrate, and mark the origin of the coordinate system at the position of the first row and first column of the pin fins on the surface of the heat sink substrate. The coordinates of this position are: ,like Figure 3 As shown.

[0169] S32. Using fluorescence detection method to show the oil stains present on the pin fin array on the heat dissipation substrate;

[0170] S33. Set the attitude parameters of the hyperspectral camera, including the distance between the hyperspectral camera and the surface of the substrate where the needle fin is located. The angle between the axial direction of the hyperspectral camera lens and the axial direction of the needle fin. Horizontal distance between the center of the hyperspectral camera lens and the axis of the needle fin A hyperspectral camera detects the movement path of a single needle wing;

[0171] S34. Adjust the posture of the hyperspectral camera to perform machine vision recognition on the back and sides of the heat sink substrate based on posture parameters;

[0172] As a further implementation of this embodiment, such as Figure 4 As shown, adjusting the attitude of the hyperspectral camera includes:

[0173] S341. Calculate the maximum value of the angle between the axial direction of the hyperspectral camera lens and the axial direction of the needle fin. and minimum value and determine the conditions If the condition is met, proceed to step S342; otherwise, proceed to step S33.

[0174] Maximum value and minimum value The calculation formula is:

[0175] ;

[0176] In the formula, This function represents the minimum value of two numbers. , These are the horizontal and vertical spacing of the heat dissipation substrate pin fins, respectively. , These are the height and radius of the heat dissipation substrate pin fins, respectively;

[0177] S342. Calculate the actual object distance of the hyperspectral camera lens. and determine the conditions If the condition is met, proceed to step S35; otherwise, proceed to step S33.

[0178] Actual object distance The calculation formula is:

[0179] ;

[0180] in, , These are the maximum and minimum object distances allowed under the condition that the hyperspectral camera lens can capture images clearly;

[0181] S35. Set the scanning parameters for the hyperspectral detection needle-fin array, including the scanning method as a transverse S-shaped sequential scan, and the starting point of the scan as... The horizontal spacing of the scan is and scanning longitudinal spacing ,like Figure 2 As shown.

[0182] S36. Scan the pin array using a hyperspectral camera, identify the oil stains on each pin, and count the number of pins with oil stains in the pin array on the heat dissipation substrate. Coordinates of the oil-stained needle-wings , The center height of the oil-stained area on the oil-stained needle wings ;

[0183] And execute the number of oil-stained needles. Conditional judgment of relevant constraints, namely:

[0184] Judgment Condition 6: Judgment If the condition is true, proceed to step S4; otherwise, proceed to step S5.

[0185] S4. Parameter control during the microfluidic column jet washing stage includes the following steps:

[0186] S41. Set the initial value of the variable representing the needle fin with oil contamination. ;

[0187] S42. Set the attitude of the jet nozzle, such as... Figure 5 As shown. This includes the angle between the jet nozzle axis and the needle fin axis. Horizontal distance between the jet nozzle and the needle fin axis The distance between the jet nozzle and the substrate surface where the needle fins are located The direction and speed of rotation of the jet nozzle around the axis of the needle fin And calculate the actual distance between the jet nozzle and the center of the oil-stained area on the needle fin. The calculation formula is:

[0188] ;

[0189] in, The distance between the hyperspectral camera and the substrate surface where the needle fins are located; The angle between the axial direction of the hyperspectral camera lens and the axial direction of the needle fin; The horizontal distance between the center of the hyperspectral camera lens and the axis of the needle fin; The height of the center of the oil-stained area on the oil-stained needle fin; The radius of the heat dissipation substrate pin fins;

[0190] S43. Set the number of cleaning cycles for the microfluidic jet. The optimized increments for output flow and output pressure are respectively , Initial value of the output flow rate of the jet nozzle Initial pressure value ;

[0191] S44, Based on actual distance Calculate the impact force of the current microfluidic jet. And perform conditional judgments related to the impact force constraints, namely:

[0192] Judgment Condition 7: Judgment If the condition is met, proceed to step S45; otherwise, proceed to step S43.

[0193] Impact The calculation formula is:

[0194] ;

[0195] in, The minimum allowable injection pressure for the jet nozzle; The density of the cleaning solution; The outlet radius of the jet nozzle; The coefficients for calculating the impact force of microfluidic column jet washing are provided. This is the critical distance at which the jet cross-section does not diverge; It is a natural exponential function;

[0196] S45. Based on the coordinates of the needles with oil stains. The jet nozzle is quickly moved to the needle fin and sprayed;

[0197] S46. Fluorescence detection was used to examine the sprayed needle wings, and the area of ​​oil stains on the needle wings was statistically analyzed using a hyperspectral camera. And perform conditional judgments related to the second area threshold constraint, namely:

[0198] Judgment condition 8: Judgment If the condition is met, proceed to step S47; otherwise, end the cleaning operation.

[0199] S47, Increase the output flow rate of the jet nozzle. Output pressure Calculate the impact force of the microfluidic jet at this time. And perform conditional checks on the constraints related to the impact force, namely:

[0200] Judgment Condition 9: Judgment Is it true? If so, then let Proceed to step S45; otherwise, proceed to step S43.

[0201] Microfluidic jet impact force The calculation formula is:

[0202] ;

[0203] in, The maximum allowable injection pressure for the jet nozzle;

[0204] S48. Judgment Conditions Is it true? If so, then let Proceed to step S42; otherwise, proceed to step S5.

[0205] S5. The cleaning process is complete.

[0206] In this embodiment, seven types of heat dissipation substrates with specifications of 110300500100, 20020100012, and C7100487 were selected as experimental objects. A control group was set up using a single ultrasonic cleaning method, while the experimental group used the multi-stage cleaning method proposed in this invention and underwent chemical plating treatment after cleaning. The black spot defects of the needle fins were observed.

[0207] This embodiment uses the defect incidence rate (the percentage of substrates with black spot defects out of the total number of observations) as a quantitative indicator, and the comparison results are as follows:

[0208] Control group: The black spot defects of the heat sink substrate pin fins were obtained using the existing cleaning method (single ultrasonic cleaning), as shown in Table 1.

[0209] Table 1. Statistics on black spot defects on heat sink substrate pin fins under existing cleaning methods

[0210]

[0211] Experimental group: Heat dissipation substrates of the same specification and batch were selected and the multi-stage cleaning method of the present invention was used. The black spot defect of the pin fins of the heat dissipation substrates was significantly improved, as shown in Table 2.

[0212] Table 2. Statistics of black spot defects on heat dissipation substrate pin fins under multi-stage cleaning method

[0213]

[0214] In summary, the defect rate of single ultrasonic cleaning is 5.1%-12.7%, while the method proposed in this embodiment reduces it to 0.5%-3.4%, with an average reduction of over 70%.

[0215] In this embodiment, three typical specifications of heat dissipation substrates were selected and cleaned using a single ultrasonic cleaning method and the multi-stage cleaning method of this invention. The comparison results of black spot defects on the needle fins after chemical plating of the substrates are as follows: Figures 6-8 As shown. Wherein:

[0216] Specification 110300500100: After a single ultrasonic cleaning, such as Figure 6 As shown in (a), after electroless plating, dense black spots are visible on the surface of the needle fins of the 110300500100 substrate. After multi-stage cleaning, the defect rate decreased from 12.7% to 3.2%. Figure 6 The results of electroless plating shown in (b) show that the black spots on the surface of the needle wings are significantly reduced.

[0217] Specification 20020100012: After a single ultrasonic cleaning, such as Figure 7 As shown in (a), after electroless plating, dense black spots are visible on the surface of the pin fins of the 20020100012 substrate. After multi-stage cleaning, the defect rate decreased from 9.1% to 3.4%, as... Figure 7 The results of electroless plating shown in Figure (b) show that the black spots on the surface of the needle wings are significantly reduced and the surface quality is significantly improved.

[0218] Specification C7100487: After a single ultrasonic cleaning, such as Figure 8 As shown in (a), the surface of the C7100487 substrate pin fins after electroless plating exhibits dense black spots, resembling "ink stains" and affecting thermal conductivity. However, after multi-stage cleaning, as... Figure 8 The results of the chemical plating shown in (b) show that the surface of the substrates in the same batch is as smooth as a mirror, with only a trace of light gray traces remaining, and the black spots visible on the surface of the needle wings are effectively suppressed.

[0219] Comparative experiments show that, through comparative experiments on 210 heat dissipation substrates of 7 specifications, the multi-stage cleaning method proposed in this invention reduces the defect rate by an average of 82.3%, with specification C7100487 achieving a breakthrough improvement from 5.1% to 0.5%. Analysis of variance shows that the standard deviation of the defect rate in the multi-stage cleaning group is only one-third of that in the control group, demonstrating a significant improvement in process stability. X-ray fluorescence spectroscopy confirms that the impurity element content on the substrate surface is reduced by more than 90% after cleaning, directly improving the adhesion and corrosion resistance of subsequent chemical plating layers, effectively suppressing black spot defects on the IGBT heat dissipation substrate pin fins, and significantly improving product surface quality.

[0220] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0221] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. Since the above embodiments are substantially similar to the method embodiments, their descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0222] The above embodiments provide a detailed description of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A multi-stage cleaning method for suppressing black spot defects on the pin fins of IGBT heat sink substrates, characterized in that, The method includes the following steps: Acquire multi-source parameters, including equipment parameters, geometric parameters of the heat dissipation substrate, and process parameters, during the cleaning process of the heat dissipation substrate; The heat dissipation substrate is subjected to ultrasonic rough cleaning based on multi-source parameters, and the surface cleanliness of the heat dissipation substrate is detected in real time during the rough cleaning process to determine whether it meets the first preset condition. If so, the cleanliness of the needle fin surface after ultrasonic coarse cleaning is detected in real time, and the coordinate information of the oil-stained needle fins on the heat dissipation substrate and the height information of the oil stains are obtained. If not, optimize the ultrasonic cleaning parameters in the rough cleaning stage until the result of the first real-time detection meets the first preset condition. The ultrasonic rough cleaning of the heat dissipation substrate includes: The radiated sound pressure of the transducer Output frequency and the output current of the ultrasonic generator As a constraint, the output power of the ultrasonic generator is... and the air gap of the transducer The ultrasonic cleaning parameters were obtained by iterative optimization. The ultrasonic generator and transducer are controlled to perform ultrasonic cleaning on the heat sink substrate according to the optimized ultrasonic cleaning parameters. Optimizing the ultrasonic cleaning parameters and performing ultrasonic rough cleaning includes the following steps: S21. Set the initial values ​​of the ultrasonic generator parameters, including setting the initial value of the output power. Output power adjustment steps ; S22. Set optimization counting variables initial value And calculate the optimal increment of the ultrasonic generator output power. The calculation formula is: ; In the formula, , These are the maximum and minimum power for driving the ultrasonic generator, respectively; S23. Based on the optimized increment Calculate the output power of the ultrasonic generator The calculation formula is: ; In the formula, , is the initial value of the adjustment coefficient for the output power of the ultrasonic generator; S24. Based on the rated current of the ultrasonic generator. Executes the output current of the ultrasonic generator Conditional judgment of relevant constraints, namely: Judgment Condition 1: Judgment Is it true? If so, proceed to step S25. If not, then execute condition 2: judgment Is it valid? If so, then set the initial value of the adjustment coefficient. Optimize the counting variable Proceed to step S23; If not, proceed to step S21; S25, Based on output power Calculate the radiated sound pressure of the transducer The calculation formula is: ; In the formula, The acoustic impedance of the cleaning fluid; The electroacoustic conversion efficiency of the transducer; The effective area of ​​the transducer radiating head; Based on the minimum allowable sound pressure level radiated by the transducer Execution and radiated sound pressure The judgment conditions for relevant constraints are as follows: judge If the condition is met, proceed to step S26; otherwise, proceed to step S21. S26, air gap of the transducer Optimize; S27. Set the temperature of the cleaning solution. and the initial concentration of cleaning agent injected into the cleaning solution. Set the optimal increment for the cleaning agent content in the cleaning solution. Set initial values ​​for the optimization variables and calculate the number of optimization steps. ,Right now: ; In the formula, The maximum amount of cleaning agent that can be injected into the cleaning solution; S28. Calculate the cleaning agent content in the optimized cleaning solution. The calculation formula is: ; in, The number of optimization steps is a variable used to optimize the cleaning agent content. S29. Start the ultrasonic coarse cleaning and initialize the count variable for the number of cleaning cycles. ; S210. Using a water film continuity test, machine vision is used to identify the back and sides of the heat sink substrate with a hyperspectral camera, and the area of ​​the heat sink substrate where oil stains have not been cleaned is calculated. ; And perform conditional checks related to the first area threshold, namely: Judgment condition 5: Determine the area of ​​the region Is it valid? If so, then the condition is determined. Is it true? If so, then let , Proceed to step S28; otherwise, proceed to step S27. If not, the cleanliness of the needle fin surface after ultrasonic coarse washing will be detected in real time. Based on the coordinate information of the oil-stained needle wings and the height information of the oil stains, the oil-stained needle wings are subjected to microfluidic jet cleaning, and the cleanliness of the needle wing surface is detected in real time during the cleaning process to determine whether the second preset condition is met. If so, then end the cleaning operation; If not, then adjust the microfluidic jet impact force during the microfluidic jet washing process until the second preset condition is met.

2. The multi-stage cleaning method according to claim 1, characterized in that, The first preset condition is whether the area of ​​the untreated oil stains identified by the first real-time detection is equal to the first area threshold. The second preset condition is that the area of ​​oil stains on the needle wings being cleaned is equal to the second area threshold.

3. The multi-stage cleaning method according to claim 1, characterized in that, The optimized air gap Includes the following steps: S261. Set the initial value of the transducer's air gap. and incremental optimization of air gap Set optimization variables Initial value of the air gap adjustment coefficient of the transducer And calculate the number of air gap optimization steps for the transducer. The calculation formula is: ; In the formula, , These are the maximum and minimum values ​​of the transducer air gap, respectively; S262. Based on the allowable deviation value of the ultrasonic generator output current. Executes the output current of the ultrasonic generator Conditional judgment of relevant constraints, namely: Judgment condition 3: Judgment Is it true? If so, proceed to step S263. If not, then determine Does this hold true? If so, then let the optimization variable... Initial value of the air gap adjustment coefficient of the transducer Proceed to step S264; otherwise, proceed to step S261. S263. Measure the output frequency of the transducer. And based on the resonant frequency of the transducer and the allowable deviation value of the output frequency. Execution and output frequency Conditional judgment of relevant constraints, namely: Judgment condition 4: Judgment Is it true? If so, proceed to step S27. If not, then determine Is it true? If so, then let Proceed to step S264; otherwise, proceed to step S261. S264. Calculate the optimized air gap of the transducer. Then proceed to step S262, the calculation formula is: ; In the formula, To optimize variables; This is the initial value of the air gap adjustment coefficient for the transducer.

4. The multi-stage cleaning method according to claim 1, characterized in that, The cleanliness of the needle fin surface after ultrasonic rough cleaning is detected in real time, and the coordinate information and oil height information of the oil-stained needle fins on the heat dissipation substrate are obtained. This includes the following steps: S31. Establish a coordinate system on the heat sink substrate, and mark the origin of the coordinate system at the position of the first row and first column of the pin fins on the surface of the heat sink substrate. The coordinates of this position are: ; S32. Using fluorescence detection method to show the oil stains present on the pin fin array on the heat dissipation substrate; S33. Set the attitude parameters of the hyperspectral camera, including the distance between the hyperspectral camera and the surface of the substrate where the needle fin is located. The angle between the axial direction of the hyperspectral camera lens and the axial direction of the needle fin. Horizontal distance between the center of the hyperspectral camera lens and the axis of the needle fin A hyperspectral camera detects the movement path of a single needle wing; S34. Adjust the posture of the hyperspectral camera to perform machine vision recognition on the back and sides of the heat sink substrate based on posture parameters; S35. Set the scanning parameters for the hyperspectral detection needle-fin array, including the scanning method as a transverse S-shaped sequential scan, and the starting point of the scan as... The horizontal spacing of the scan is and scanning longitudinal spacing ; S36. Scan the pin array using a hyperspectral camera, identify the oil stains on each pin, and count the number of pins with oil stains in the pin array on the heat dissipation substrate. Coordinates of the oil-stained needle wings , The center height of the oil-stained area on the oil-stained needle wings ; And execute the number of oil-stained needle wings present. Conditional judgment of relevant constraints, namely: Judgment Condition 6: Judgment If the condition is met, the needles with oil stains will be subjected to microfluidic jet cleaning and a second real-time detection will be performed; otherwise, the cleaning operation will be terminated.

5. The multi-stage cleaning method according to claim 4, characterized in that, Adjusting the attitude of the hyperspectral camera includes: S341. Calculate the maximum value of the angle between the axial direction of the hyperspectral camera lens and the axial direction of the needle fin. and minimum value and determine the conditions If the condition is met, proceed to step S342; otherwise, proceed to step S33. Maximum value and minimum value The calculation formula is: ; In the formula, This function represents the minimum value of two numbers. , These are the horizontal and vertical spacing of the heat dissipation substrate pin fins, respectively. , These are the height and radius of the heat dissipation substrate pin fins, respectively; S342. Calculate the actual object distance of the hyperspectral camera lens. and determine the conditions If the condition is met, proceed to step S35; otherwise, proceed to step S33. Actual object distance The calculation formula is: ; in, , These represent the maximum and minimum object distances allowed under conditions where the hyperspectral camera lens can capture images clearly.

6. The multi-stage cleaning method according to claim 1, characterized in that, Microfluidic jet cleaning of oil-contaminated needles includes: Based on the coordinate information of the oil-stained needle fins, control the jet nozzle to move to the corresponding needle fin position; Based on the height information of the oil stains, the attitude and output parameters of the jet nozzle are adjusted to spray and wash the needle fins containing the oil stains.

7. The multi-stage cleaning method according to claim 6, characterized in that, Adjusting the output parameters of the jet nozzle includes: Based on the oil stain height information and the relative positions of the jet nozzle and the needle fins, the impact force of the microfluidic jet acting on the oil stain area is calculated. With the goal of achieving a preset threshold range for the impact force, the output flow rate and output pressure of the jet nozzle are adjusted.

8. The multi-stage cleaning method according to claim 1, characterized in that, The process of performing microfluidic jet cleaning on oil-contaminated needles and conducting a second real-time inspection includes the following steps: S41. Set the initial value of the variable representing the needle fin with oil contamination. ; S42. Set the attitude of the jet nozzle, including the angle between the jet nozzle axis and the needle fin axis. Horizontal distance between the jet nozzle and the needle fin axis The distance between the jet nozzle and the substrate surface where the needle fins are located The direction and speed of rotation of the jet nozzle around the needle fin axis And calculate the actual distance between the jet nozzle and the center of the oil-stained area on the needle fin. The calculation formula is: ; in, The distance between the hyperspectral camera and the substrate surface where the needle fins are located; The angle between the axial direction of the hyperspectral camera lens and the axial direction of the needle fin; The horizontal distance between the center of the hyperspectral camera lens and the axis of the needle fin; The height of the center of the oil-stained area on the oil-stained needle fin; The radius of the heat dissipation substrate pin fins; S43. Set the number of cleaning cycles for the microfluidic jet. The optimized increments for output flow and output pressure are respectively , Initial value of the output flow rate of the jet nozzle Initial pressure value ; S44, Based on actual distance Calculate the impact force of the current microfluidic jet. And perform conditional judgments related to the impact force constraints, namely: Judgment Condition 7: Judgment If the condition is met, proceed to step S45; otherwise, proceed to step S43. Impact The calculation formula is: ; in, The minimum allowable injection pressure for the jet nozzle; The density of the cleaning solution; The outlet radius of the jet nozzle; The coefficients for calculating the impact force of microfluidic column jet washing are provided. This is the critical distance at which the jet cross-section does not diverge; It is a natural exponential function; S45. Based on the coordinates of the needles with oil stains. The jet nozzle is quickly moved to the needle fin and sprayed; S46. Fluorescence detection was used to examine the sprayed needle wings, and the area of ​​oil stains on the needle wings was statistically analyzed using a hyperspectral camera. And perform conditional judgments related to the second area threshold constraint, namely: Judgment condition 8: Judgment If the condition is met, proceed to step S47; otherwise, end the cleaning operation. S47, Increase the output flow rate of the jet nozzle. Output pressure Calculate the impact force of the microfluidic jet at this time. And perform conditional checks on the constraints related to the impact force, namely: Judgment Condition 9: Judgment Is it true? If so, then let Proceed to step S45; otherwise, proceed to step S43. Microfluidic jet impact force The calculation formula is: ; in, The maximum allowable injection pressure for the jet nozzle; S48. Judgment Conditions Is it true? If so, then let Proceed to step S42; otherwise, end the cleaning operation.