Method for manufacturing laminated chip inductor element from low-temperature co-fired ferrite raw material belt

Abstract

The invention relates to a method for preparing a laminated chip inductor element from a low-temperature co-fired ferrite raw material belt, and belongs to the technical field of laminated chip inductor elements. S1, slurry preparation; s2, casting is carried out; s3, carrying out silk-screen printing; s4, cutting is conducted; s5, discharging glue; s6, sintering test; s7, quality detection; s8, establishing a quality optimization system, predicting the sintering parameters and the parameters in the general sintering process through a random forest model to obtain new sintering parameters, and predicting the same sintering parameters through a support vector machine model to obtain new sintering parameters; s9, determining parameters; s10, determining a sintering temperature change curve; s11, sintering treatment is carried out; and S12, the sintered product is subjected to end sealing, electroplating, testing and inspection, and then the laminated chip inductor element is obtained. The problem that the production efficiency is affected due to the fact that a large amount of time needs to be consumed for determining the sintering temperature curve of existing different types of laminated chip inductors is solved.

Description

Technical Field

[0001] The invention belongs to the technical field of laminated chip inductor components, and in particular relates to a method for manufacturing a laminated chip inductor component from a low temperature co-fired ferrite raw tape. Background Art

[0002] At present, there are generally the following methods for making multilayer chip inductors at home and abroad, and the principles and characteristics are different. Full wet production method: add adhesives to ferrite or ceramic dielectric materials to make a slurry suitable for waterfall flow, and flow a certain thickness of membrane on the carrier under mechanical drive, and then print the conductive wire electrode slurry, and then print the turn-to-turn connection slurry at the line electrode to form a bump. When flowing the dielectric membrane, the membrane forms a connecting through hole at the bump, and then print the conductive wire electrode at the through hole to make the turn-to-turn connection good. After repeated cycles, an inductor structure of a certain thickness is formed, and then dried, cut, debonded, sintered, terminated, electroplated, tested and inspected; in addition, there are alternating printing wet method and full dry method; However, no matter what manufacturing method is used, the process of debinding and sintering is required. During the debinding and sintering process, in order to ensure the production quality of the multilayer chip inductor, the temperature curve of debinding and sintering needs to be determined according to different material ratios, and the organic matter such as adhesives and solvents used in the manufacturing process in the green body needs to be removed to avoid the formation of bubbles or other defects in the finished product. Moreover, each new material ratio and new product specification requires re-adjustment of the sintering temperature curve design, but the time spent in adjusting the sintering temperature is relatively long, and the determination of the temperature curve needs to be obtained through a series of experiments and analyses. Therefore, it is necessary to perform debinding and sintering again and again when adjusting the temperature curve to determine which temperature curve can meet the production standards, which will take a lot of time, thereby reducing production efficiency. Summary of the invention

[0003] In order to solve the above problems existing in the prior art, the present invention provides a method for making a laminated chip inductor component from a low-temperature co-fired ferrite raw tape, which solves the problem that the sintering temperature curve of different types of existing laminated chip inductors needs to be determined, which takes a lot of time and affects the production efficiency.

[0004] The object of the present invention can be achieved by the following technical solution: A method for making a laminated chip inductor component from a low temperature co-fired ferrite raw tape, comprising the following steps: S1: slurry preparation, mixing the materials according to the proportion and grinding and homogenizing by ball milling; S2: Casting, casting the slurry to form a ferrite raw tape; S3: Screen printing, printing the raw tape layer by layer according to the designed circuit, and forming a complete multi-layer substrate embryo through electrostatic lamination; S4: cutting, dividing the multi-layer substrate embryo into specific small pieces; S5: Debinding: Debinding the product by determining the debinding temperature curve according to the material ratio; S6: Sintering test, taking samples and subjecting the debinding products to sintering process, and recording the parameters during the sintering process; S7: Quality inspection: perform quality inspection on the sintered products and record the inspection results and corresponding sintering parameters; S8: Establish a quality optimization system, feed back the sintering parameters and product quality data to the quality optimization system in real time, predict the sintering parameters and parameters in the general sintering process through the random forest model to obtain new sintering parameters, and predict the same sintering parameters through the support vector machine model to obtain new sintering parameters; S9: Parameter determination, weighted averaging the prediction results of the random forest model and the support vector machine model to obtain new sintering parameters; S10: Determine the sintering temperature change curve, and repeat steps S8-S9 until the quality of the product reaches the preset standard; S11: sintering process, batch sintering the products according to the sintering parameters of step S10; S12: The sintered product is terminated, electroplated, tested and inspected to obtain a laminated chip inductor component.

[0005] As a preferred technical solution of the present invention, the sintering parameters include sintering temperature, sintering time and atmosphere content.

[0006] As a preferred technical solution of the present invention, in step S8, the magnet crystal image is first analyzed by scanning electron microscopy to determine the type of general process to be adopted.

[0007] As a preferred technical solution of the present invention, in step S9, the weight ratio of the two models is cross-validated by the mean square error when performing weighted averaging.

[0008] As a preferred technical solution of the present invention, in step S7, the quality inspection of the product includes magnet cracking, magnet white spots and magnet cooked or not cooked, and specific evaluation criteria are formulated.

[0009] As a preferred technical solution of the present invention, in step S8, the prediction results of the random forest model focus on magnet cracking, and the prediction results of the support vector machine model focus on magnet white spots and whether the magnet is cooked or not.

[0010] As a preferred technical solution of the present invention, in step S1, the milled slurry is filtered before the next step is performed.

[0011] As a preferred technical solution of the present invention, in step S1, the particle distribution of the filtered slurry is detected and the detection result is input into the random forest model as a variable.

[0012] The beneficial effects of the present invention are as follows: by establishing a quality optimization system, sintering parameters and product quality data are predicted and optimized through a random forest model and a support vector machine model, and the prediction results of the two models are further optimized and determined to improve the accuracy of the prediction results, and by recording the sintered products and the corresponding sintering parameters, a data feedback mechanism is formed, thereby shortening the R&D and production cycles, reducing the number of trial and error times, thereby improving production efficiency and reducing production costs, and solving the problem that different types of existing multilayer chip inductors need to spend a lot of time to determine the sintering temperature curve, which affects production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0014] Figure 1 The present invention is a flow chart of the method for making. DETAILED DESCRIPTION

[0015] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0016] In the manufacturing process of multilayer chip inductor components, because the adhesive system is the same, the debinding process is defined by referring to the debinding curve of the adhesive. Therefore, the debinding process is the same for all materials. Therefore, the debinding temperature curve is obtained according to the material ratio during the debinding process. Therefore, the current process in the debinding process is basically consistent, so no additional design is required. However, during the sintering process, the sintering process is to make the ceramic material undergo physical and chemical changes at high temperature to form a dense ceramic body and improve the performance and reliability of the inductor. Principle: At high temperatures, the diffusion and rearrangement between ceramic particles lead to the densification of the material and the formation of a continuous grain structure.

[0017] The sintering process generally refers to the whole process of sintering (densification) and crystallization to finally generate ferrite under the condition of being lower than the melting temperature of the material. Ferrite undergoes significant volume shrinkage during the sintering process, and also undergoes great changes in bulk density and porosity. According to the porosity, sintering shrinkage can be divided into three stages: initial, intermediate and late stages. In the initial stage of sintering, the solid particles are in contact with each other, the voids are dispersed and interpenetrated, and as the binder, solvent and other impurities evaporate, the pores escape through the surface of the particles, the porosity decreases, the shrinkage rate increases, and the overall density increases slightly; in the middle stage of sintering, as the temperature increases, the interfaces of each solid particle gradually merge, and the interpenetrating pores are gradually closed and relatively concentrated. The porosity decreases rapidly, the shrinkage rate increases significantly, and the bulk density decreases; in the late stage of sintering, as the sintering temperature further increases, the closed voids shrink, the bulk density increases significantly, and the densification tends to be perfect.

[0018] Therefore, when testing new materials in the sintering process, some abnormalities may occur and the process needs to be adjusted. In fact, the common problems are product magnet cracking, magnet white spots and magnet under-sintering. Therefore, it takes a lot of time to achieve a good and stable production process when designing the sintering process, and in order to shorten this process to improve production efficiency and ensure the stability of production quality.

[0019] See also Figure 1 This embodiment provides a method for manufacturing a laminated chip inductor component from a low temperature co-fired ferrite raw tape, comprising the following steps: S1: slurry preparation, mixing the materials according to the proportion and grinding and homogenizing by ball milling; S2: Casting, casting the slurry to form a ferrite raw tape; S3: Screen printing, printing the raw tape layer by layer according to the designed circuit, and forming a complete multi-layer substrate embryo through electrostatic lamination; S4: cutting, dividing the multi-layer substrate embryo into specific small pieces; S5: Debinding: Debinding the product by determining the debinding temperature curve according to the material ratio; S6: Sintering test, taking samples and subjecting the debinding products to sintering process, and recording the parameters during the sintering process; S7: Quality inspection: perform quality inspection on the sintered products and record the inspection results and corresponding sintering parameters; S8: Establish a quality optimization system, feed back the sintering parameters and product quality data to the quality optimization system in real time, predict the sintering parameters and parameters in the general sintering process through the random forest model to obtain new sintering parameters, and predict the same sintering parameters through the support vector machine model to obtain new sintering parameters; S9: Parameter determination, weighted averaging the prediction results of the random forest model and the support vector machine model to obtain new sintering parameters; S10: Determine the sintering temperature change curve, and repeat steps S8-S9 until the quality of the product reaches the preset standard; S11: sintering process, batch sintering the products according to the sintering parameters of step S10; S12: The sintered product is terminated, electroplated, tested and inspected to obtain a laminated chip inductor component.

[0020] By establishing a quality optimization system, the sintering parameters and product quality data are predicted and optimized through the random forest model and the support vector machine model, and the prediction results of the two models are further optimized and determined to improve the accuracy of the prediction results. By recording the sintered products and the corresponding sintering parameters, a data feedback mechanism is formed, thereby shortening the R&D and production cycle, reducing the number of trial and error, thereby improving production efficiency and reducing production costs, and solving the problem that different types of existing multilayer chip inductors require a lot of time to determine the sintering temperature curve, which affects production efficiency.

[0021] In order to better improve product quality problems, reduce the occurrence of defects, and improve the overall reliability and performance stability of inductor components, in this embodiment, the sintering parameters include sintering temperature, sintering time and atmosphere content, and the sintering process requires strict control of temperature, time and atmosphere to ensure uniform sintering and stable performance of the material. In addition, magnet cracking is generally improved by starting with the slurry formula, while magnet white spots and under-firing are improved by sintering technology. Among them, white spots are generally improved by cooling and refining the potting process, and under-firing is improved by increasing the furnace temperature and extending the insulation time. Therefore, by detecting specific parameters and improving the prediction weight, the quality and performance of the magnet during the sintering process can be improved more quickly.

[0022] The more data the quality optimization system records when determining the sintering process, the more accurate the prediction result will be. Therefore, in order to reduce the number of predictions while improving the prediction accuracy, in one embodiment, in step S8, the magnet crystal image is first analyzed by a scanning electron microscope to determine the type of general process to be used. Because the sintering parameters of the general process are needed as reference data when making predictions, the magnet crystal image is first analyzed by a scanning electron microscope to determine the type of general process to be used, and then the sintering parameters of the process are input into the quality optimization system to improve the accuracy of the sintering parameter prediction. At the same time, a large amount of data is used to shorten the number of predictions, thereby improving production efficiency.

[0023] Since the methods used by the random forest model and the support vector machine (SVM) model for prediction are different, in order to improve the prediction accuracy of the relationship between the temperature curve and the quality of the finished product during the production process, thereby optimizing the production process, reducing time waste, and improving product quality, in one embodiment, in step S9, the weight ratio of the two models is cross-validated by the mean square error when performing weighted averaging. Since the random forest is an integrated learning method, by constructing multiple decision trees and voting on their results, it can reduce overfitting and improve the robustness of the prediction, and can process large-scale data, which is suitable for capturing complex nonlinear relationships between features. SVM can perform classification and regression by finding the optimal hyperplane, is suitable for processing high-dimensional data, can effectively deal with small sample problems, and can process nonlinear relationships through different kernel functions. Therefore, it is necessary to process the prediction results of the quantity model to obtain more accurate prediction results. In step S9, the weight ratio of the two models is cross-validated by weighted averaging and mean square error, and reverse calculation is performed through the cross-validation performance of the model to ensure that the model with better performance occupies a larger weight. Through weighted averaging, the advantages of different models can be fully utilized, the deviation that may be caused by a single model can be reduced, and the stability and accuracy of the prediction can be improved.

[0024] In order to improve the prediction accuracy of the quality optimization system, in one embodiment, in step S7, the quality inspection of the product includes magnet cracking, magnet white spots and magnet undercooking, and specific evaluation criteria are formulated. Common magnet cracking, magnet white spots and magnet undercooking are used to formulate specific evaluation criteria, such as the length of cracks, the number and distribution of white spots, and the density after sintering, which will be used for quantitative and qualitative analysis to ensure the consistency of the inspection. Thereby significantly improving the prediction accuracy of the quality optimization system.

[0025] Since different models predict in different ways and process different data types, in order to improve the accuracy of the quality optimization system prediction, in one embodiment, in step S8, the prediction results of the random forest model focus on magnet cracking, and the prediction results of the support vector machine model focus on magnet white spots and magnet undercooking, because the main characteristics of magnet cracking are determined by the formula and uniformity of the slurry, and the random forest can identify the most important features, which helps to optimize the process parameters and reduce the number of experiments. Therefore, the random forest model is used to accurately predict the situation of magnet cracking to reduce the occurrence of magnet cracking. Since the change of sintering temperature, sintering time control and atmosphere content during sintering are all uncertain parameters, these parameters will cause the product to have magnet white spots and magnet undercooking, and SVM is suitable for complex nonlinear problems and can effectively handle different results in the sintering process. Therefore, SVM is used to effectively analyze and predict different problems in the sintering process, providing a scientific basis for improving processes and raw materials. At the same time, it can flexibly respond to changes in production, thereby improving the overall quality of the product.

[0026] In order to ensure the uniformity of the slurry and reduce quality problems caused by the unevenness of the slurry, in one embodiment, in step S1, the ground slurry is filtered before executing the next step. After grinding, the slurry will be filtered through a filter mesh with an appropriate pore size to remove large particles, bubbles and impurities in the slurry to avoid defects caused by them in subsequent processes, which will help to achieve better density and higher product stability in subsequent molding and sintering processes.

[0027] Since the focus parameter of the random forest model is magnet cracking, and magnet cracking is mainly caused by slurry uniformity, in order to improve the accuracy of the random forest model prediction, in one embodiment, in step S1, the distribution of slurry particles after filtration is detected and the detection result is input as a variable into the random forest model. By detecting the distribution of slurry particles after filtration in step S1 and inputting the result as a variable into the random forest model, the prediction accuracy of the model for magnet cracking can be effectively improved. This not only helps to understand the impact of slurry uniformity on product quality, but also can provide data support for subsequent production processes, thereby improving the overall quality and production efficiency of the product.

[0028] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A method for manufacturing a laminated chip inductor component from a low temperature co-fired ferrite raw tape, characterized in that: The following steps are involved: S1: slurry preparation, mixing the materials according to the proportion and grinding and homogenizing by ball milling; S2: Casting, casting the slurry to form a ferrite raw tape; S3: Screen printing, printing the raw tape layer by layer according to the designed circuit, and forming a complete multi-layer substrate embryo through electrostatic lamination; S4: cutting, dividing the multi-layer substrate embryo into specific small pieces; S5: Debinding: Debinding the product by determining the debinding temperature curve according to the material ratio; S6: Sintering test, taking samples and subjecting the debinding products to sintering process, and recording the parameters during the sintering process; S7: Quality inspection: perform quality inspection on the sintered products and record the inspection results and corresponding sintering parameters; S8: Establish a quality optimization system, feed back the sintering parameters and product quality data to the quality optimization system in real time, predict the sintering parameters and parameters in the general sintering process through the random forest model to obtain new sintering parameters, and predict the same sintering parameters through the support vector machine model to obtain new sintering parameters; S9: Parameter determination, weighted averaging the prediction results of the random forest model and the support vector machine model to obtain new sintering parameters; S10: Determine the sintering temperature change curve, and repeat steps S8-S9 until the quality of the product reaches the preset standard; S11: sintering process, batch sintering the products according to the sintering parameters of step S10; S12: The sintered product is terminated, electroplated, tested and inspected to obtain a laminated chip inductor component.

2. The method for manufacturing a laminated chip inductor component from a low temperature co-fired ferrite raw tape according to claim 1, characterized in that: The sintering parameters include sintering temperature, sintering time and atmosphere content.

3. The method for manufacturing a laminated chip inductor component from a low temperature co-fired ferrite raw tape according to claim 1, characterized in that: In step S8, the magnet crystal image is first analyzed by scanning electron microscopy to determine the general process type to be adopted.

4. The method for manufacturing a laminated chip inductor component from a low temperature co-fired ferrite raw tape according to claim 1, characterized in that: In step S9, the weight ratio of the two models is cross-validated by the mean square error when performing weighted averaging.

5. The method for manufacturing a laminated chip inductor component from a low temperature co-fired ferrite raw tape according to claim 1, characterized in that: In step S7, the quality inspection of the product includes magnet cracking, magnet white spots and magnet cooked or not cooked, and specific evaluation criteria are formulated.

6. The method for manufacturing a laminated chip inductor component from a low temperature co-fired ferrite raw tape according to claim 5, characterized in that: In step S8, the prediction results of the random forest model focus on magnet cracking, and the prediction results of the support vector machine model focus on magnet white spots and whether the magnet is cooked or not.

7. The method for manufacturing a laminated chip inductor component from a low temperature co-fired ferrite raw tape according to claim 1, characterized in that: In step S1, the milled slurry is filtered before the next step.

8. The method for manufacturing a laminated chip inductor component from a low temperature co-fired ferrite raw tape according to claim 7, characterized in that: In step S1, the particle distribution of the filtered slurry is detected and the detection results are input into the random forest model as variables.

Images