Regression orthogonal test method for adjusting vacuum vapor phase reflow soldering temperature curve

Abstract

The invention relates to a regression orthogonal test method for adjusting a vacuum vapor phase reflow soldering temperature curve, which combines and unifies the advantages of an orthogonal test design method and regression analysis, and selects proper test points within the test range of various factors influencing the vacuum reflow soldering temperature curve. According to the method, a regression model with high precision and good fitting with an actual vacuum reflow soldering curve can be quickly and accurately constructed by using a small number of tests, and the obtained regression coefficient is not influenced by units and values of factors, so that the regression calculation amount is simplified, the limitation of original orthogonal table selection is broken through, and the method is more flexible and convenient to use.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic assembly processes for aerospace products, and particularly to a method for adjusting the temperature curve of vacuum vapor phase reflow soldering. Background Art

[0002] At present, there are three ways to achieve reflow soldering. First, hot air reflow soldering. The printed circuit board assembly is transported through a track, and the temperature values of each temperature zone and the track speed are set to achieve hot air reflow soldering. In this method, during the soldering process, the solder joints are easily disturbed due to mechanical transmission, the temperature compensation efficiency is low in a poor airtight environment, the temperature gradient in each temperature zone is large due to heat transfer by hot air convection, the temperature control accuracy is poor, and it is not easy to adjust the ideal temperature curve. Second, vapor phase reflow soldering. The printed circuit board is placed on a bracket, and the upward movement speed of the bracket from the bottom is controlled. The solder paste coated on the pads of the printed circuit board is reflow soldered and wetted by the vaporization and deposition of the vapor phase liquid in the heating furnace to form solder joints. In this method, the solder joints are easily disturbed due to the up and down movement of the bracket during the soldering process. Compared with hot air reflow soldering, vapor phase reflow soldering is beneficial for the overall heat balance of the printed circuit board through the deposition of the vapor phase liquid, forming solder joints with good wetting. However, with the increase in the size, thickness, and large area copper clad area of the printed circuit board, in order to meet the heat balance of the smallest 0201 package to the largest heat capacity components on the whole board and the soldering quality of lead-free CBGA and BGA, the reflow soldering requires longer time and higher soldering temperature, resulting in oxidized and non-smooth surfaces of the formed solder joints, increasing the BGA soldering void ratio and affecting the soldering quality and reliability. Third, vacuum vapor phase reflow soldering. The printed circuit board is placed flat on the bracket and transported into the furnace without moving. By setting the vacuum degree and adjusting the vapor phase liquid injection amount parameters in different temperature intervals, the vapor phase reflow soldering process of the printed circuit board assembly in a vacuum environment is realized. In this method, the printed circuit board does not move during the soldering process, which is conducive to the wetting and forming of solder joints, effectively controlling and reducing the surface tension and oxidation degree of the solder joints by increasing the vacuum degree, reducing welding quality defects such as voids, bubbles, and bridging. The heat balance during reflow soldering in a closed environment is better and the temperature control accuracy is higher. However, due to the complexity of the interaction between factors such as vacuum degree, vapor phase liquid injection amount, and temperature zone interval, the uncertainty of this open-loop temperature curve function relationship will greatly increase the difficulty of adjusting the temperature curve. There are problems such as poor indicators, bad shapes, and lack of mastery of adjustment methods in a large number of frequent temperature curve adjustments, increasing the process preparation time, low production efficiency resulting in a large consumption cost of vapor phase liquid, unstable process parameters, low process capability, and not meeting the needs of flexible production in the mode of rapid switching of multi-variety and small-batch production. Therefore, based on the above third method, after consulting relevant industry standards, there are only temperature curves applicable to hot air reflow soldering, vapor phase reflow soldering, and the Sn63Pb37 solder paste recommended by manufacturers, which have weak engineering practice guidance for the temperature curve technology of vacuum vapor phase reflow soldering.

[0003] The existing reflow soldering process, such as the Chinese invention with the application publication number CN107124835B and the authorization date of August 30, 2019, discloses a reflow soldering and chip mounting process, which includes the following steps: printing solder paste, chip mounting, intermediate inspection, reflow soldering, and post-furnace inspection. By performing a pressure reduction treatment during reflow soldering, it is possible to avoid the formation of large bubbles and voids during the soldering process, or the explosive exhaust caused by the escape of large bubbles, which may carry out a lot of fine solder balls. By creating a negative pressure in the soldering area through pressure reduction, small bubbles during the soldering process can easily escape, and it is not easy to form large bubbles or cause solder ball splashing. Among them, in a reflow soldering and chip mounting process, the time in the soldering area is specified as 30s to 90s. When the temperature is greater than 150°C, the pressure reduction starts, and the pressure reduction time is 30s to 50s, and the lowest air pressure drops to 40mbar to 2mbar. When adjusting the temperature according to this process parameter using a vacuum vapor reflow soldering equipment, it is found that, first of all, the liquid phase time in the soldering area of 30s - 90s belongs to the liquid phase time range of general reflow soldering. Currently, key devices with BGA and CBGA as the core are increasingly used in products. BGA solder balls generally use lead-free materials and are prone to oxidation. It is necessary to reduce the soldering void defect and oxidation degree by evacuating the air. The lower the vacuum degree, the longer the liquid phase time in the soldering area. At the same time, the soldering temperature for soldering lead-free devices with lead-containing solder paste needs to be controlled within the range of 230°C to 235°C, and the boiling point of the vapor phase liquid is 240°C. The closer the temperature is to the boiling point, the longer the heating-up time. However, the liquid phase time in the soldering area of 30s to 90s can no longer meet the soldering requirements for BGA and CBGA lead-free devices in vacuum vapor reflow soldering; secondly, 120°C to 150°C is the heat preservation area of reflow soldering during the solvent volatilization stage of the solder paste. At this time, the air pressure change in the evacuated furnace causes the furnace temperature to first overshoot downward and then rise for 20s to 30s. The too long heat preservation time causes the solvent in the solder paste to volatilize excessively, reducing the wettability, capillary action, and spreading ability of the molten solder during the soldering stage; finally, the vacuum degree is as low as 40mbar to 2mbar, and multiple gradient extractions are required. If the extraction is too fast at one time, it will cause a sudden change in the vacuum pressure in the furnace, and due to the stress change of the evacuation, components may undergo displacement, tilt, solder bridging and other soldering defects. Summary of the Invention

[0004] The technical problem solved by this application is: overcoming the deficiencies of the prior art, providing a regression orthogonal test method for adjusting the temperature curve of vacuum vapor reflow soldering, realizing the refined requirements for adjusting the temperature curve of vacuum vapor reflow soldering, being particularly suitable for multi-variety and batch production, meeting the needs of flexible production of printed circuit board assemblies with different product input characteristics, reducing rework, and improving the soldering quality and reliability of products.

[0005] Give full play to the role of vacuum degree in reflow soldering in reducing oxidation, lowering surface tension, and improving soldering wetting quality. By means of analysis of variance to fit and optimize the regression equation model, it can more scientifically and effectively guide technicians in the process or engineering technology to discover the objective laws of the mapping of the changes in the levels of various factors in the equipment to the test result data. In the case of complex printed circuit boards and component characteristics, it is possible to quickly and well master the method of adjusting the vacuum vapor phase reflow soldering temperature curve. It solves the technical problems of incorrect application timing of vacuum degree, non-quantification of vacuum degree, low temperature control accuracy of the vapor phase liquid injection volume in the open-loop temperature system, and high complexity of the interaction of various influencing factors, long adjustment time and difficult to master. Design a characteristic selection table for the vacuum vapor phase reflow soldering temperature curve, so that process personnel can be liberated from the complex process of adjusting the temperature curve time and time again. According to the input characteristics of components and printed circuit boards, quickly match and select the best temperature curve from the applied product types and reference pictures, thereby greatly reducing the number of process tests and production preparation time, improving the process capability, ensuring the stability of process parameters and the consistency of products, and being especially suitable for multi-variety and batch production.

[0006] The technical solution provided by this application is as follows:

[0007] A regression orthogonal test method for adjusting the vacuum vapor phase reflow soldering temperature curve, including:

[0008] S1. According to the product input characteristics, which include printed circuit board characteristics and component characteristics, determine the key parameters of vacuum vapor phase reflow soldering as vacuum degree, peak temperature, and liquid phase time;

[0009] S2. Make a process board according to the printed circuit board characteristics, set process parameters according to the key parameters determined in S1, and the process parameters include multiple temperature segments; use the process board to perform vacuum vapor phase reflow soldering according to the process parameters and collect the temperature in real time to obtain a time-temperature waveform diagram;

[0010] Determine the pass criterion;

[0011] According to the temperature segments and the waveform diagram, determine the running time and heating rate of each temperature segment;

[0012] Compare the running time and heating rate of each temperature segment with the pass criterion. If the pass criterion is met, the temperature curve of this temperature segment does not need to be adjusted and proceed to S6; if the pass criterion is not met, the temperature curve of this temperature segment is adjusted according to S3;

[0013] S3. For a temperature segment that does not meet the pass criterion, determine the main factors of the regression orthogonal test as vacuum degree, temperature range, and vapor phase liquid injection volume; calculate the upper level, zero level, and lower level for each main factor respectively, and calculate the corresponding coded levels according to the upper level, zero level, and lower level of each factor;

[0014] According to the coding level, select multiple test schemes of regression orthogonal design in the regression orthogonal test table; according to the multiple test schemes of regression orthogonal design, conduct experiments successively using a process board to obtain test index results;

[0015] According to the test index results and the regression equation, calculate and obtain the equation coefficients, and the equation coefficients correspond to the main factors;

[0016] S4. According to the regression equation and equation coefficients in S3, calculate the F-distribution value of each main factor, conduct a significance test on the F-distribution value, judge whether the result is that the main factor is a key factor or not a key factor, determine the corresponding coefficients of the key factors according to the key factors and the regression equation in S3, and determine the sorted regression equation according to the positive and negative signs of the corresponding coefficients of the key factors; according to the sorted regression equation, calculate and obtain the updated process parameters; the updated process parameters include the heating response time, temperature range, and vapor-liquid injection volume;

[0017] S5. Compare the heating response time, temperature range, and vapor-liquid injection volume obtained in S4 with the qualified criteria determined in S2. If the heating response time, temperature range, and vapor-liquid injection volume do not meet the qualified criteria, repeat S3 and S4, and iterate the four process parameters of the heating response time, temperature range, vapor-liquid injection volume, and liquid phase time for each test until the heating response time, temperature range, and vapor-liquid injection volume meet the qualified criteria;

[0018] S6. After each temperature segment of the temperature curve meets the qualified criteria, modify and save in the vacuum vapor reflow soldering equipment according to the finally determined process parameters, and exit the temperature curve debugging mode.

[0019] Optionally, in S2, the process parameters further include vacuum degree, vapor-liquid injection volume, exhaust time, and liquid phase time;

[0020] The qualified criteria include the heating rate and the heating time range;

[0021] The heating rate is calculated based on the lowest temperature, highest temperature, and running time within the temperature segment.

[0022] Optionally, set the process parameters according to the reflow soldering key parameters determined in S1, including:

[0023] The product input characteristics include component characteristics and printed circuit board characteristics. The component characteristics are the package forms of the components. According to the component package forms, the component pin solder ends, package materials, and specific heat capacities can be determined; the printed circuit board characteristics are the printed circuit board area, number of printed circuit board layers, printed circuit board thickness, and printed circuit board copper clad area;

[0024] Determine the vacuum degree process parameters according to the component characteristics and the oxidation degree of the component pin solder ends;

[0025] Determine the peak temperature process parameters according to the component characteristics and the lead percentage in the solder paste;

[0026] Determine the liquid phase time process parameters according to the product input characteristics, vacuum degree, and peak temperature;

[0027] According to the reflow soldering temperature curve recommended by the solder paste manufacturer and the component manufacturer, determine the temperature segments and obtain the temperature range of each temperature segment;

[0028] Determine the exhaust time of the vacuum vapor phase reflow soldering according to the component characteristics and the printed circuit board characteristics to determine the cooling rate process parameters.

[0029] Optionally, the main factors for determining the regression orthogonal test are the vacuum degree, temperature range, and vapor phase liquid injection volume, including:

[0030] The level of the vacuum degree is the vacuum degree range of a temperature segment that does not meet the qualified criterion;

[0031] The level of the temperature range is the temperature range of a temperature segment that does not meet the qualified criterion;

[0032] The level of the vapor phase liquid injection volume is the vapor phase liquid injection volume range of a temperature segment that does not meet the qualified criterion.

[0033] Optionally, the upper level of each of the main factors is the maximum value of its corresponding value range, the lower level is the minimum value of its corresponding value range, and the zero level = (upper level + lower level) / 2.

[0034] Optionally, in the S3, the regression equation is:

[0035] t = a ± b 1 V 1 ±b 2 V 2 ±b 3 V 3 ±b 12 V 1 V 2 ±b 13 V 1 V 3 ,

[0036] where V 1 is the vacuum degree; V 2 is the temperature range; V 3 is the vapor phase liquid injection volume; t is the heating response time;

[0037] n = m 0 +m c, where n is the number of trials, and m 0 is the number of zero-level trials, and m c is the number of trials specified by the orthogonal array;

[0038] where a is the constant coefficient of the regression equation, and t i is the heating response time for each trial, and n is the number of trials;

[0039] where b 1 is the partial regression coefficient of the vacuum degree in the regression equation, is the vacuum degree for each trial, and m c is the number of trials specified by the orthogonal array;

[0040] where b 2 is the partial regression coefficient of the temperature range in the regression equation, is the temperature range for each trial, and m c is the number of trials specified by the orthogonal array;

[0041] where b 3 is the partial regression coefficient of the vapor-liquid injection amount in the regression equation, is the vapor-liquid injection amount for each trial, and m c is the number of trials specified by the orthogonal array;

[0042] where b 12 is the partial regression coefficient of the interaction between the vacuum degree and the temperature range in the regression equation, (V 1 V 2 ) i is the interaction between the vacuum degree and the temperature range for each trial, and m c is the number of trials specified by the orthogonal array;

[0043] where b 13 is the partial regression coefficient of the interaction between the vacuum degree and the vapor-liquid injection amount in the regression equation, (V 1 V 3 ) i is the interaction between the vacuum degree and the vapor-liquid injection amount for each trial, and m c is the number of trials specified by the orthogonal array.

[0044] Optionally, in S4, according to the regression equation and equation coefficients in S3, calculate the F-distribution value of each main factor, perform a significance test on the F-distribution value, and determine whether the result shows that the main factor is a key factor or a non-key factor. Based on the key factors and the regression equation in S3, determine the corresponding coefficients of the key factors. According to the positive or negative sign of the corresponding coefficients of the key factors, determine the sorted regression equation. According to the sorted regression equation, calculate and obtain the heating response time, temperature range, and vapor-liquid injection volume, including:

[0045] According to the regression equation and equation coefficients in S3, calculate the partial regression sum of squares of each main factor; calculate the total sum of squares; calculate the total partial regression sum of squares based on the partial regression sum of squares of each main factor; calculate the residual sum of squares based on the total sum of squares and the total partial regression sum of squares; calculate the degrees of freedom and mean square value based on the partial regression sum of squares of each main factor; calculate the F-distribution value of each main factor based on the mean square value and the residual sum of squares;

[0046] Under the condition of a given significance level α, perform a significance test on the F-distribution value of each main factor, and determine whether the result shows that the main factor is a key factor or a non-key factor;

[0047] Substitute the key factors into the regression equation in S3 to obtain the simplified regression equation; find the positive or negative sign of the corresponding coefficients of the key factors in the simplified regression equation. When the coefficient is positive, it indicates that taking the upper level of the key factor has a good significance on the test index result; when the coefficient is negative, it indicates that taking the lower level of the key factor has a good significance on the test index result;

[0048] Calculate the corresponding coding according to the positive or negative sign of the corresponding coefficients of the key factors, and substitute it back into the simplified regression equation according to the corresponding coding to establish the sorted regression equation;

[0049] According to the sorted regression equation, calculate and obtain the heating response time, temperature range, and vapor-liquid injection volume.

[0050] Optionally, in S4, the total sum of squares: Among them, SS T is the total sum of squares, m c is the number of tests specified by the orthogonal table

[0051] Partial regression sum of squares:

[0052] Among them, SS 1 is the partial regression sum of squares of the vacuum degree, m c is the number of tests specified by the orthogonal table, b 1 is the partial regression coefficient of the vacuum degree in the regression equation;

[0053] Among them, SS 2 is the partial regression sum of squares for the temperature interval, m c is the number of experiments specified by the orthogonal array, b 2 is the partial regression coefficient of the temperature interval in the regression equation;

[0054] Among them, SS 3 is the partial regression sum of squares for the vapor-liquid injection volume, m c is the number of experiments specified by the orthogonal array, b 3 is the partial regression coefficient of the vapor-liquid injection volume in the regression equation;

[0055] Among them, SS 12 is the partial regression sum of squares for the interaction between the vacuum degree and the temperature interval, m c is the number of experiments specified by the orthogonal array, b 12 is the partial regression coefficient of the interaction between the vacuum degree and the temperature interval in the regression equation;

[0056] Among them, SS 13 is the partial regression sum of squares for the interaction between the vacuum degree and the vapor-liquid injection volume, m c is the number of experiments specified by the orthogonal array, b 13 is the partial regression coefficient of the interaction between the vacuum degree and the vapor-liquid injection volume in the regression equation, SS 12 is the partial regression sum of squares for the interaction between the vacuum degree and the temperature interval,;

[0057] Total partial regression sum of squares:

[0058] SS R = SS 1 + SS 2 + SS 3 + SS 12 + SS 13 where SS R is the total partial regression sum of squares, SS 1 is the partial regression sum of squares for the vacuum degree, SS 2 is the partial regression sum of squares for the temperature interval, SS 3 is the partial regression sum of squares for the vapor-liquid injection volume, SS 13 is the partial regression sum of squares for the interaction between the vacuum degree and the vapor-liquid injection volume;

[0059] Residual sum of squares:

[0060] SS e = SS T - SS R , where SS e is the residual sum of squares, SS Tis the total sum of squares, SS R is the total partial regression sum of squares;

[0061] Degree of freedom:

[0062] df R = ∑df 一次项 + ∑df 交互项 = m, where ∑df 一次项 is the partial regression sum of squares of each factor, and ∑df 交互项 is the partial regression sum of squares of the interaction factor;

[0063] df e = ∑df T - df R = n - 1 - m, where df e is the residual degree of freedom, and ∑df T = n - 1, where n is the number of trials;

[0064] Mean square value:

[0065] where MS i is the mean of each factor or interaction factor, SS i is the partial regression sum of squares of each factor and interaction factor, and df i is the degree of freedom of each factor and interaction factor;

[0066] where MS e is the residual mean, SS e is the partial regression sum of squares of the residual; and df e is the residual degree of freedom;

[0067] F - distribution value:

[0068] where F i is the F - distribution value, MS i is the mean of each factor or interaction factor, and MS e is the residual mean.

[0069] Optionally, in S4, at least one F - distribution value significance test is performed on each main factor, and the judgment result of each main factor is obtained as a key factor or a non - key factor, including:

[0070] Query the regression orthogonal experiment table to obtain the F - critical values at the significance levels α = 0.05 and α = 0.01, compare the F - distribution value of the parameter with the F - critical value, and obtain the judgment result of each main factor as a significant factor or a non - significant factor;

[0071] Incorporate the degrees of freedom of the partial regression sum of squares of insignificant factors into the residual sum of squares parameter, recalculate the F-distribution values of each parameter, and conduct another F-distribution significance test to obtain the judgment results of each significant factor as key factors or non-key factors.

[0072] The orthogonal regression experimental design method solves the problem that the traditional orthogonal experiment can only be limited to the established process levels, rather than the optimal solution of process parameters within a certain range of process parameters. With reasonable experimental design and fewer experimental times, it discovers the laws between the levels of influencing factors, establishes a mathematical model of the fitting regression equation, accurately predicts the results of experimental indicators, determines the significance and change trends of influencing factors through F-distribution variance analysis, and simplifies the regression equation.

[0073] The present invention solves the problems of the interaction complexity of process parameters such as vacuum degree, vapor-liquid injection amount, temperature range, and exhaust time, the lack of mastery of the method of frequently adjusting the temperature curve in large quantities, and poor indicators. It iteratively optimizes process parameters to improve the temperature control accuracy of the vapor-liquid injection amount, realizes the refined requirements for adjusting the vacuum vapor-phase reflow soldering temperature curve, thereby greatly reducing the number of experiments and production preparation time, improving the process capability, ensuring the stability of process parameters and the consistency of products, being particularly suitable for multi-variety and batch production, meeting the needs of flexible production of printed circuit board assemblies with different product input characteristics, and improving the welding quality and reliability of products.

[0074] The regression orthogonal experimental method for adjusting the vacuum vapor-phase reflow soldering temperature curve provided by the present invention has the following beneficial effects:

[0075] (1) The regression orthogonal experimental method for adjusting the vacuum vapor-phase reflow soldering temperature curve provided by the present invention combines the temperature curve parameters recommended by industry standards and mainstream solder paste manufacturers, designs a vacuum vapor-phase reflow soldering temperature curve diagram, defines the vacuum reflow soldering process: heating (A) → heat preservation (B1) → transition (B2) → reflow soldering (C) → cooling (D), gives the qualified criterion method and the core temperature point of vacuum vapor-phase reflow soldering, and through the explanation of the welding quality of each technical index parameter of the temperature curve, it helps the understanding and mastery of process and engineering technicians, can quickly and effectively analyze and accurately locate welding defect problems, meet the technical index parameters of the vacuum vapor-phase reflow soldering temperature curve, and has good and fast adjustment methods.

[0076] (2) The regression orthogonal test method for adjusting the vacuum vapor phase reflow soldering temperature curve provided by the present invention gives the timing and setting method of applying vacuum through experimental experience summary, as well as the setting method of the exhaust cooling time from the start of the peak temperature of reflow soldering to before the circuit board is taken out of the furnace, reducing the number of experimental runs and the difficulty of adjusting the curve significantly due to inappropriate process parameters, and improving the welding quality and reliability of the core key devices of BGA, CBGA, and CCGA packages.

[0077] (3) The regression orthogonal test method for adjusting the vacuum vapor phase reflow soldering temperature curve provided by the present invention compares the experimental data of each temperature zone with the criterion of index requirements through observation, discovers the critical and out-of-tolerance data, establishes a regression equation with good goodness of fit using the regression orthogonal test design method, finds the mapping relationship between each factor level and the index, sets the injection amount of the vapor liquid according to the index, and gives the injection amount range of the vapor liquid in each temperature zone, improving the control accuracy of the vapor liquid, reducing waste, saving costs, and environmental pollution.

[0078] (4) The regression orthogonal test method for adjusting the vacuum vapor phase reflow soldering temperature curve provided by the present invention creates a characteristic selection table and a process parameter table for the vacuum vapor phase reflow soldering temperature curve, liberating the process or engineering and technical personnel from the difficult and complex work of adjusting the temperature curve. On the one hand, it reduces the time for adjusting the temperature curve, avoids being close to the equipment and continuously adjusting the curve for more than 2 hours, inhaling too much flux residue and volatile gas of the vapor liquid, resulting in HSE occupational health problems such as respiratory diseases; on the other hand, it reduces the number of process experiments and production preparation time, improves work efficiency, makes the process capability sufficient, and ensures the quality and reliability of the product. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 is a schematic diagram of the vacuum vapor phase reflow soldering temperature curve of the present invention;

[0080] Figure 2 is a schematic diagram of the vacuum pumping interface in the vacuum vapor phase reflow soldering temperature curve;

[0081] Figure 3 is the vacuum pumping process in the vacuum vapor phase reflow soldering temperature curve;

[0082] Figure 4 is a schematic diagram of the adjusted curve of the vacuum vapor phase reflow soldering temperature curve;

[0083] Figure 5 is a schematic diagram showing that the temperature reduction of the vacuum vapor phase reflow soldering is divided into 2 stages;

[0084] Figure 6 is a schematic diagram of the air-cooling parameter setting of the vacuum vapor phase reflow soldering;

[0085] Figure 7 It is a schematic diagram of the exhaust parameter setting for vacuum vapor phase reflow soldering;

[0086] Figure 8 It is a schematic diagram of the temperature curve parameter debugging mode for vacuum vapor phase reflow soldering;

[0087] Figure 9 It is a schematic diagram of the vapor phase liquid and temperature zone parameter setting for vacuum vapor phase reflow soldering;

[0088] Figure 10 It is a flowchart of the method for adjusting the temperature curve of vacuum vapor phase reflow soldering used for regression orthogonal experimental design. Specific implementation manners

[0089] To make the objectives, technical solutions and advantages of the present invention clearer, the following will further describe in detail the disclosed implementation manners of the present invention with reference to the accompanying drawings.

[0090] Figure 1 It is a schematic diagram of the temperature curve of a vacuum vapor phase reflow soldering of the present invention. The temperature curve is divided into four parts, namely: heating zone (A), preheating and insulation B (B1, B2), reflow soldering (C), cooling (D), as shown in Table 1.

[0091] Table 1 Explanation of the parameters of the vacuum vapor phase reflow soldering interval

[0092]

[0093] A good temperature curve in terms of technical indicators should have the following characteristics:

[0094] 1) Whether the temperature curve is adjusted well depends crucially on the BGA soldering quality;

[0095] 2) The maximum heat capacity and the minimum heat capacity on the printed circuit board reach a temperature intersection point at the end of preheating, that is, the temperature of the whole board reaches thermal equilibrium;

[0096] 3) The highest peak temperature on the whole board meets the heat resistance requirements of the components, and the lowest peak temperature meets the requirements for solder joint formation;

[0097] 4) The temperature difference between the highest temperature and the lowest temperature on the BGA package is less than 5°C, and generally it is not allowed to exceed 7°C

[0098] 5) The difference between the set temperature and the soldering peak temperature determines the strain and soldering stress of the components on the printed circuit board, which is usually more important than the heating rate or the cooling rate;

[0099] 6) The shorter the LTD, the longer the True TAL.

[0100] The longer the LTD (Liquidus Time Delay), the greater the temperature difference and deformation in the BGA package, and the more likely it is to occur ball socket defects (HoP) and shrinkage fractures. The shorter the True TAL, the shorter the coexistence time of the solder balls at the center and the periphery in the liquid state, and the more likely it is to occur ball socket defects (HoP). Therefore, it is desired that the LTD be shorter and the True TAL be longer.

[0101] 7) According to the reflow soldering temperature curve and the core temperature points recommended by the solder paste manufacturer and the component manufacturer, combined with the actual characteristics of the printed circuit board and components, select the temperature section and recommend the set core temperature parameters, as shown in Table 2.

[0102] Based on the core temperature points, determine the temperature segments and obtain the temperature ranges for each temperature segment;

[0103] Table 2 Core Temperature Parameters

[0104]

[0105]

[0106] The starting temperature is the ambient temperature (23 ± 5) °C, and the end temperature of the heating-up is the peak temperature of up to 230 °C. The peak temperature of lead-free BGA is 235 °C, and the actual core temperature point is set at 230 °C. The temperature is rushed from 230 °C to 235 °C by evacuating the last temperature zone.

[0107] Figure 2 It is a schematic diagram of the evacuation interface in a vacuum vapor reflow soldering temperature curve; it is evacuated in 3 stages, ① Pre-Vac (pre-vacuum), ② Process Vac (evacuated during each temperature process), ③ Final Vac (evacuated after the injection of the last vapor phase liquid and the completion of soldering);

[0108] ① Pre-Vac (pre-vacuum) applies vacuum to the entire process of the printed circuit board from entering the furnace to exiting the furnace, and the vacuum degree does not affect the singularity of the temperature curve during the intermediate process (heating-up, heat preservation, reflow soldering);

[0109] ② Evacuate during each temperature zone, selectively apply vacuum degree in each temperature section to increase the temperature gradient and change the temperature curvature. Especially when the temperature is in an unstable process near the eutectic point, if the vacuum degree design is not appropriate, it will instead have a certain impact on the soldering quality;

[0110] ③ Final Vac: By using the principle that the boiling point of the vapor-phase liquid decreases during vacuum pumping, resulting in a temperature overshoot, the requirement for the peak temperature can be met. For example, for lead-free BGA, the peak temperature is required to be 235°C. By setting the core temperature of the last section to 230°C and increasing the final Vac vacuum degree, the temperature can rise from 230°C to 235°C - 238°C.

[0111] The vacuum degree will accelerate the change of the reflow soldering temperature curvature. A lower vacuum degree will increase the liquid phase time, causing the flux in the solder paste to be over-inactivated, resulting in appearance defects such as rough and wrinkled solder joints. If the vacuum pumping speed is too fast or the valve opening is too large, the surface-mounted components will be displaced under the action of external forces during vacuum pumping, and the solder paste will splash out to form defects such as excess solder balls. The recommended parameters are as follows:

[0112] 1) Vacuum degree gradient setting

[0113] When the vacuum degree is less than 200 mbar, at least three-stage gradient vacuum pumping is required;

[0114] When the vacuum degree is 200 mbar - 500 mbar, at least two-stage gradient vacuum pumping is required;

[0115] When the vacuum degree is above 500 mbar, at least one-stage gradient vacuum pumping is required.

[0116] 2) Vacuum degree holding time: 15 s - 25 s;

[0117] 3) Valve opening ratio: 20% - 50%.

[0118] Figure 3 This is the vacuum pumping process in the vacuum vapor-phase reflow soldering temperature curve. The vacuum pumping time is controlled within 15 s - 25 s. As the vacuum degree decreases, the holding time increases, and the change trend of the soldering temperature decreasing and then rising again becomes more obvious;

[0119] Figure 4 This is a schematic diagram of the vacuum vapor-phase reflow soldering temperature curve adjustment curve. While controlling the injection of the overall vapor-phase liquid in several temperature zones, the injection volume of the last temperature zone also needs to be controlled. The experimental data analysis shows that the higher injection volume in the last section does not necessarily result in a higher temperature and a faster temperature rise. This is because it takes a certain transition time for the cold vapor-phase liquid to be injected and heated to the vapor temperature in the cavity. The length of this time is related to the amount of injected vapor-phase liquid and the state equation of the vapor temperature of the equipment system parameters. This transition time prolongs the liquid phase time. Since the volume of the vacuum equipment cavity is fixed, when the air pressure is saturated and stable, adding an excessive amount of vapor-phase liquid makes the temperature rise change slowly and further prolongs the liquid phase time. An overly long liquid phase time leads to the formation of a thicker IMC layer at the solder joint, reducing the strength of the solder joint against mechanical stress and making it prone to brittle fracture. Therefore, the most effective method is vacuum pumping, using the temperature overshoot characteristics of vacuum pumping to compensate for the peak temperature value;

[0120] Figure 5 It is a schematic diagram of the two-stage cooling process of vacuum vapor phase reflow soldering. Cooling stage 1: The temperature drops from the peak temperature to a range 5°C - 10°C higher than the solder paste eutectic temperature point; Cooling stage 2: From the time the printed circuit board assembly exits the furnace until the temperature drops to 75°C, this stage is cooled by an external blower.

[0121] Figure 6 It is a schematic diagram of the air-cooling parameter settings for vacuum vapor phase reflow soldering. The Double-Stage time in the cold machine parameters is the waiting time for the carrier to transfer the printed circuit board from inside the furnace to under the blower. It is generally advisable to set it to 25s ± 5s. When the printed circuit board arrives, the blower starts working according to the set parameters and stops working when the set parameter value is reached. The transfer table moves out of the working area and the printed circuit board is taken out, ending the entire reflow soldering process. It is advisable to set the air-cooling time of the blower to 150s - 180s, and the temperature reduction is generally advisable to be 25°C - 35°C.

[0122] Figure 7 It is a schematic diagram of the exhaust parameter settings for vacuum vapor phase reflow soldering;

[0123] Combined with the characteristics of the actual printed circuit board and components, the temperature range for setting the exhaust parameters is from the peak temperature dropping to a range 5°C - 10°C higher than the solder paste eutectic temperature point. The exhaust setting is generally in the last stage. Exhausting in the middle stage will affect the shape of the temperature curve and prolong the heating time in the later stage. When the cooling rate of the first stage (stage1) of exhaust exceeds 15s, the temperature will drop by 10°C, and it is extremely easy to cause shrinkage and fracture of BGA solder joints; When the cooling rate of the second stage (stage2) of exhaust is slow, it can be combined with the first stage to achieve a controlled cooling rate of (0.5 - 3) °C / s, and control the temperature when the furnace door is opened to drop to 5°C - 10°C higher than the eutectic temperature point, reducing the formation of disturbed solder joints and component displacement during the transfer process when the furnace door is opened;

[0124] Figure 8 It is a schematic diagram of the temperature curve parameter debugging mode for vacuum vapor phase reflow soldering. The device enters the test mode (test), and by setting the injection volume (injection1) ml of the vapor liquid in each temperature section, switch the mode (Modus) to the temperature mode (tempture), and set the core temperature points according to Table 2. Record the running time h of the temperature rise in each temperature section during the debugging process or obtain it from the exported temperature acquisition data. After the debugging is completed, switch the mode from temperature (tempture) to holding time (holding), record the running time t into the holding time (holding) to save the debugging program, and thus complete the temperature curve adjustment;

[0125] Figure 9It is a schematic diagram of the setting of the vapor liquid and temperature zone parameters for vacuum vapor reflow soldering. Call the debugged temperature curve and run the program once without loading before formal soldering. After confirming that the equipment parameters are running normally, start the normal printed circuit board reflow soldering production.

[0126] The present invention provides a regression orthogonal test method for adjusting the temperature curve of vacuum vapor reflow soldering, as Figure 10 shown, including the following steps:

[0127] Step 1: Determine the ranges of 3 key parameters, namely vacuum degree, peak temperature, and liquid phase time, based on the characteristics of the printed circuit board and components as input features;

[0128] In this step, determine the process parameters for setting 3 key parameters of reflow soldering, including:

[0129] Determine that the process parameter of the vacuum degree is that the level of the vacuum degree is 1 mbar to 999 mbar. When the component manufacturer requests not to use soldering in a vacuum environment, the vacuum degree is set to 999 mbar; when soldering surface mount resistors and capacitors, the control range of the vacuum degree is 700 mbar to 999 mbar; when soldering surface mount resistors and capacitors, including components with QFP, SOP, QFN, LCCC, and LGA packages, the control range of the vacuum degree is 500 mbar to 999 mbar; when soldering components with BGA, CBGA, and CCGA packages, the control range of the vacuum degree is 100 mbar to 500 mbar; when there are special usage occasions, the vacuum degree is less than 100 mbar;

[0130] The process parameter of the peak temperature is that the peak temperature range of general components is 215°C to 225°C; when the component pins, solder pads, and solder balls are lead-free materials (such as lead-free solder balls for BGA), the encapsulation temperature range is 230°C to 235°C; when there are special usage environments or requirements, follow the requirements of the component manufacturer or technical documents;

[0131] The process parameter of the liquid phase time is at a level of 60 s to 120 s. For leaded solder paste, the liquid phase time is within the temperature change cycle of 183°C - peak temperature - 183°C, and the liquid phase time is 60 s to 90 s; for lead-free solder paste, the temperature change trend is 217°C - peak temperature - 217°C, and the liquid phase time of the lead-free solder paste is within the temperature change cycle of 217°C - peak temperature - 217°C or when using leaded solder paste to solder lead-free BGA packages, the liquid phase time TAL is 90 s to 120 s.

[0132] In this step, according to the product input features: the component feature is that the component side is a lead-free BGA, and the soldering side is mainly surface mount resistors, capacitors, and components with QFP, SOP, QFN, and SMD packages; the printed circuit board feature is that the number of layers of the printed circuit board is 10 layers and the area of the printed circuit board is 160 cm 2, the printed circuit board has a thickness of 2.5 mm, and the copper-clad area of the printed circuit board accounts for 80% of the total board area;

[0133] According to the product input characteristics, the process parameters of the three key parameters are determined as follows: the vacuum degree range is 100 mbar to 500 mbar, the peak temperature range is 230 °C to 235 °C, and the liquid phase time range is 90 s to 120 s;

[0134] Step 2: Make a process board according to the characteristics of the printed circuit board, set the vacuum degree, temperature segments, temperature range, vapor-liquid injection volume, and exhaust time to obtain the determined process parameters; enter the debugging mode and record the test data of temperature acquisition;

[0135] In this step, set the process parameters according to the key parameters of reflow soldering determined in S1, including:

[0136] The process parameters of temperature segments are from 1 segment to 6 segments. When soldering surface mount resistors, capacitors, QFP, and SOP packaged devices, the thickness of the printed circuit board does not exceed 2 mm, the number of layers of the printed circuit board is not more than 4 layers, and the area of the printed circuit board is not more than 150 cm 2 , when the copper-clad area accounts for no more than 40% of the printed circuit board area, select a 3-segment temperature curve; when soldering devices including BGA, CBGA, CCGA, QFP, SOP, QFN, LCCC, LGA, SMD packaged devices and large heat containers with metal casings, the thickness of the printed circuit board does not exceed 3 mm, the number of layers of the printed circuit board is 6 to 12 layers, and the area range of the printed circuit board is between 150 cm 2 ~900 cm 2 , when the copper-clad area accounts for the printed circuit board area in the range of 40% to 80%, select 4 to 5 segments; when soldering devices including BGA, CBGA, CCGA, LCCC packages and large heat containers with metal casings, the thickness of the printed circuit board is greater than 3 mm, the number of layers of the printed circuit board is greater than 12 layers, and the area of the printed circuit board is greater than 900 cm 2 board, when the copper-clad area accounts for more than 80% of the printed circuit board area, select 6 segments;

[0137] The process parameters of the temperature range are determined according to the reflow soldering temperature curve recommended by the solder paste manufacturer and the component manufacturer, determine the temperature segments, and obtain the temperature range of each temperature segment as follows:

[0138] Temperature segment 1: 25 °C to 100 °C, or 25 °C to 120 °C;

[0139] Temperature segment 2: 100 °C to 150 °C, or 120 °C to 160 °C;

[0140] Temperature segment 3: 150 °C to 180 °C, or 160 °C to 180 °C;

[0141] Temperature segment 4: 180 °C to 215 °C, or 180 °C to 220 °C;

[0142] Temperature segment 5: 215°C to 225°C, or 220°C to 230°C;

[0143] Temperature segment 6: 225°C to 230°C, or 230°C to 235°C;

[0144] The process parameter of the vapor-phase liquid injection volume is that the range of each temperature interval is 0 ml to 1000 ml, and the vapor-phase liquid injection volume of each segment is determined according to the temperature segment; when the temperature segment is 1 to 3 and the temperature is set below 180°C, the control range of the vapor-phase liquid injection volume of each segment is 200 ml to 500 ml; when the temperature segment is 4 and the temperature is set in the range of 180°C to 215°C or 180°C to 220°C, the control range of the vapor-phase liquid injection volume of each segment is 500 ml to 900 ml; when the temperature segment is 5 to 6 and the temperature is set in the range of 215°C to 235°C or 220°C to 235°C, the control range of the vapor-phase liquid injection volume of each segment is 400 ml to 800 ml; the total control range of the vapor-phase liquid injection volume of segments 1 to 6 is 1500 ml to 2500 ml;

[0145] The process parameter of the exhaust time is in the range of 30 s to 120 s. The exhaust time is set according to different material components to control the cooling rate. Plastic encapsulation: <= 2°C / s; Ceramic: (2 - 4)°C / s; Hybrid: (2 - 2.4)°C / s; The exhaust time is divided into two stages. The frequency of stage 1 is 50 Hz and the time is controlled within 5 s to 30 s; the frequency of stage 2 is 60 Hz and the time is controlled within 30 s to 90 s;

[0146] The steps for determining the process parameters are as follows:

[0147] Determine the vacuum degree. According to S1, the range is 100 mbar to 500 mbar. Use a 40x magnifying glass to check the surface of the lead-free BGA solder balls for oxidation. Select a higher vacuum degree. The determined vacuum degree parameter is 200 mbar, and a three-stage gradient vacuum pumping is adopted;

[0148] Determine the temperature segment. Select a range of 4 segments or 5 segments. Considering that the copper-clad area is 80%, the determined temperature segment is 5;

[0149] Determine the temperature range. The injection volume control range of the vapor phase liquid for each of the first 3 segments is 200 ml to 500 ml. Considering 80% of the copper-clad area, when the solder paste temperature reaches close to the eutectic point of 183°C, the components with the maximum and minimum heat capacities on the entire board reach thermal equilibrium. Set a lower injection volume of the vapor phase liquid for the temperature range of 160°C to 180°C to reduce the heating rate, and determine the injection volume of the vapor phase liquid for 160°C to 180°C to be 220 ml. Taking into account that the peak temperature of the lead-free BGA reaches 235°C and the liquid phase time is controlled within 120 s, increase the injection volume of the vapor phase liquid in the 180°C to 220°C range of the fourth temperature zone, control the heating rate at 1.5°C / s ± 0.1°C / s, and set the injection volume control range at 700 ml ± 100 ml according to engineering experience. Relatively reduce the injection volume of the vapor phase liquid in the 220°C to 230°C range of the fifth temperature zone, and use the physical process of vacuum temperature boost to reach the required peak temperature of 235°C, and set the injection volume control range at 450 ml ± 100 ml according to engineering experience;

[0150] Exhaust time. The body of the lead-free BGA is made of ceramic material, and the cooling rate control range is 2°C / s to 4°C / s. According to engineering experience, set the first stage at 50 Hz, and the exhaust time range is 5 s to 30 s, and determine the exhaust time to be 5 s. Set the second stage at 60 Hz, and the exhaust time range is 30 s to 90 s, and determine the exhaust time to be 60 s;

[0151] Specifically, in this step, first take points on the process board (principles: minimum, large heat capacity, BGA center, corners), and use high-temperature solder (melting point above 260°C) to weld the thermocouple at the points taken; secondly, determine the process parameters and enter them into the equipment, preheat by running the program once empty, and finally, put the process board into the equipment, insert the thermocouple on the process board into the equipment sensor interface, and record the corresponding relationship between each sensor number and the temperature acquisition point on the process board; enter the equipment temperature curve debugging mode and open the waveform diagram of real-time sampling, and perform vacuum vapor phase reflow soldering according to the determined process parameters.

[0152] Record the heating response time t of each temperature zone. According to the temperature curve waveform diagram, record and calculate the temperature range, heating response time t, liquid phase time, and peak temperature process parameters. The process record of the actual vacuum vapor phase reflow soldering temperature curve of the lead-free BGA is shown in Table 3.

[0153] Specifically include:

[0154] According to the aerospace industry standard QJ3173 "Requirements for Reflow Soldering Technology of Aerospace Electronic and Electrical Products" and the key parameters in S1, determine the acceptance criteria. The acceptance criteria include the heating rate and heating time range, and the heating rate and heating time range are obtained according to the standard.

[0155] Determine the heating response time t for each temperature range based on the temperature segmentation and waveform diagram;

[0156] Calculate the heating rate based on the lowest temperature, highest temperature, and heating response time t within the temperature segmentation;

[0157] Compare the heating response time t and the heating rate with the acceptance criteria. If the acceptance criteria are met, it indicates that the temperature curve for this temperature segmentation does not need to be adjusted and proceed to Step 5; if the acceptance criteria are not met, the temperature curve for the corresponding temperature segmentation needs to be adjusted according to Step 3. That is, adjust the temperature curve for each temperature segmentation that does not meet the acceptance criteria according to Step 3.

[0158] Table 3 Record Table of 5 - segment Temperature Curves for Orthogonal Experiment

[0159]

[0160] Step 3: Based on the regression orthogonal experiment table (the regression orthogonal experiment table is from "Experimental Design and Data Processing (Second Edition)"), establish a regression equation to determine the influence degree of each factor. If the test data is qualified, proceed to Step 5;

[0161] In this step, for the 5th temperature zone of 220°C - 230°C, the heating response time t of 16 s and 0.33°C / s is judged not to meet the standard range (8 s - 10 s, 0.50°C / s - 0.625°C / s). Conduct a regression orthogonal experiment on the 5th temperature zone to determine the functional relationship between the heating response time t and the three main factors of vacuum degree, temperature range, and vapor - liquid injection volume.

[0162] Specifically, the regression orthogonal experiment includes:

[0163] (1) Conduct factor - level coding

[0164] Based on process experience, determine the three main factors as vacuum degree, temperature range, and vapor - liquid injection volume. The level of the temperature range is that the temperature range of the 5th temperature zone is 225°C - 230°C, the level of the vacuum degree is 200 mbar - 400 mbar, and the level of the vapor - liquid injection volume is 350 ml - 550 ml;

[0165] Calculate the upper level, zero level, and lower level for each of the three main factors respectively. Based on the upper level, zero level, and lower level of each factor, calculate the corresponding factor - level coding;

[0166] The factor levels are: vacuum degree x1 = 200 mbar - 400 mbar, temperature range x2 = 225°C - 230°C, vapor - liquid injection volume x3 = 350 ml - 550 ml.

[0167] Among them, xj1 is the lower level of factor x j x j2 is the upper level of factor x j x j0 is the zero level of factor x j ;

[0168] where, △j is the change interval of factor x j ;

[0169] Taking the vacuum degree as an example, the upper level x12 = 400 mbar and the lower level x11 = 200 mbar are determined, and the zero level is calculated

[0170] The coding of factor levels is: perform a linear transformation on the factor levels, that is: where, V j is the coding of factor X j which is also called the canonical variable, X j is the natural variable of the factor, x j0 is the zero level of factor x j and △j is the change interval of factor x j ;

[0171] In this example, the codings of x j2 , x j0 , x j1 are calculated to be V j2 = 1, V j0 = 0, V j1 = -1 respectively, and the calculation method is as follows:

[0172] Calculate the upper level coding

[0173] Calculate the zero level coding

[0174] Calculate the lower level coding

[0175] △1 = x12 - x10 = 400 - 300 = 100 mbar, or △1 = x10 - x11 = 300 - 200 = 100 mbar

[0176] Code the other factor levels, and the coding results are shown in Table 4

[0177] Table 4 Factor Level Coding Table

[0178]

[0179] (2) Selection of Orthogonal Array and Determination of Test Scheme

[0180] According to the factor level coding table, considering the interaction between factors, the orthogonal table is determined to have 5 factors and 2 levels. Select the orthogonal table L from the regression orthogonal test table according to the coding obtained in the previous step 8 (2 7 ). After coding transformation, Table 5 is obtained. Zero-level experiments are not conducted, and the total number of experiments n = m 0 +m c = 0 + 8 = 8. Among them, m 0 is the number of zero-level (0) experiments, and m c is the number of experiments specified by the orthogonal table.

[0181] Table 5 Regression orthogonal design experimental scheme and experimental results

[0182]

[0183]

[0184] According to the regression orthogonal design experimental scheme in Table 5 (including the standardized variables of factor levels and the input parameters of the vacuum reflow soldering furnace), use the process board to conduct 8 experiments in sequence to obtain the experimental index results;

[0185] Based on the experimental index results (heating-up response time and peak temperature), calculate the equation coefficients according to the regression equation, and list the intermediate process results of the relevant calculations, as shown in Table 6.

[0186] Table 6 Regression orthogonal design calculation table

[0187]

[0188] The regression equation is from the book "Experimental Design and Data Processing (Second Edition)". According to the regression equation:

[0189] t = a ± b 1 V 1 ±b 2 V 2 ±b 3 V 3 ±b 12 V 1 V 2 ±b 13 V 1 V 3 , calculate the equation coefficients as follows:

[0190]

[0191] Among them,

[0192] The regression equation obtained is:

[0193] t = 12.125 + 0.125V1 +0.875 V 2 -2.875 V 3 +0.375 V 1 V 2 -0.375 V 1 V 3

[0194] According to the absolute values of the partial regression coefficients in the regression equation, the primary and secondary order of each factor and interaction is: V 3 > V 2 > V 1 V 2 = V 1 V 3 > V 1 , and the priority order of the main factors affecting the heating-up time t is the vapor-phase liquid injection amount, the temperature range, and the vacuum degree in sequence.

[0195] Step 4: Through the analysis of variance table (Table 7), obtain the well-fitted regression equation, determine the vapor-phase liquid injection amount calculated from the heating-up response time t and the temperature range, modify the equipment process parameters, re-conduct the vacuum vapor-phase reflow soldering using the process board, record the test index results (heating-up response time and peak temperature) and the liquid phase time for judgment. If it is unqualified, repeat Step 4; if it is qualified, enter Step 5;

[0196] In this step, the analysis of variance of the regression equation is as follows:

[0197] Total sum of squares:

[0198] Partial sum of squares of regression:

[0199]

[0200] Total partial sum of squares of regression:

[0201] SS R = SS 1 + SS 2 + SS 3 + SS 12 + SS 13 = 0.125 + 6.125 + 66.125 + 1.125 + 1.125 = 74.625 Residual sum of squares: SS e = SS T - SS R = 74.875 - 74.625 = 0.25

[0202] According to the above parameters, including the total sum of squares, partial regression sum of squares, total partial regression sum of squares, partial regression sum of squares, residual sum of squares, degrees of freedom, mean square value, and F-distribution value, the calculation results of the first analysis of variance table are obtained, as shown in Table 7.

[0203] Table 7 The First Analysis of Variance Table

[0204]

[0205] Significance test of the F-distribution value:

[0206] Query the regression orthogonal experiment table (the regression orthogonal experiment table is from the book "Experimental Design and Data Processing (Second Edition)"), obtain the F critical values at the significant levels α = 0.05 and α = 0.01, and compare the F-distribution value of the parameter with the F critical value obtained from the table. The significance test judgment criteria are as follows:

[0207] When the F-distribution value ≤ F critical value, it indicates that the influence of the factor on the test index result is not significant;

[0208] When the F-distribution value > F critical value, it indicates that the factor has a significant influence on the test index result;

[0209] When the F-distribution value >> F critical value, it indicates that the factor has a very significant influence on the test index result;

[0210] Finally, the judgment result of each main factor is obtained as a significant factor or a non-significant factor;

[0211] Given the significance level α = 0.01 or 0.05, perform a significance test on the F-distribution values of the parameters V 1 、V 2 、V 3 、V 1 V 2 、V 1 V 3 ;

[0212] Check the F-distribution table to obtain the critical value, compare the F-distribution value with the F critical value. The greater the gap between the F value and the critical value, the more significant and important the influence of the factor or interaction on the test result;

[0213] In this example, query the F-distribution table, F 0.05 (1, 2) = 18.51, F 0.01 (1, 2) = 98.49, F 0.05 (5, 2) = 19.30, F 0.01 (5, 2) = 99.30;

[0214] When the significant level α = 0.05, the factors V 2 (temperature range) and V3 (Vapor liquid injection volume) has a significant effect on the test index result (heating response time t), and other factors are not significant. Incorporate the partial regression sum of squares and degrees of freedom of V in the parameters 1 , V 1 V 2 , V 1 V 3 into the residual SS e , and then perform the second - order analysis of variance, as shown in Table 8.

[0215] Table 8 Second - order analysis of variance table

[0216]

[0217] Query the F - distribution table, and we get F 0.05 (1, 5) = 6.61, F 0.01 (1, 5) = 16.26;

[0218] When the significance level α = 0.01 or α = 0.05, the factor V1 (vacuum degree) has no significant effect on the test index result (heating response time t) within the range of 200 mbar - 400 mbar for soldering lead - free BGA;

[0219] When the significance level α = 0.05, the factor V 2 (temperature range) has a significant effect;

[0220] When the significance level α = 0.01, the factor V 3 (vapor liquid injection volume) has a very significant effect on the test index result (heating response time t);

[0221] Simplify the original regression equation, and according to the coding formula, we get:

[0222] For the factor V 2 the coefficient takes the upper level, where the X2 factor is the temperature range;

[0223] For the factor V 3 the coefficient takes the lower level, where the X3 factor is the vapor liquid injection volume;

[0224] Substitute into the simplified regression equation:

[0225]

[0226] We get, after arrangement, the regression equation: t = - 54.5625 + 0.35V 2 - 0.02875V 3

[0227] According to the technical indicators (heating response time t) in the passing criteria, the range is 8s to 10s. When X 2 (temperature range) is at 225°C, 227.5°C, and 230°C respectively, according to the sorted regression equation, the vapor-phase liquid injection volume X 3 is obtained, as shown in Table 9.

[0228] Table 9 Back-calculated vapor-phase liquid injection volume

[0229]

[0230]

[0231] According to Table 9, under the condition that the technical indicator (heating response time t) is controlled within 8s to 10s, the sorted regression equation is calculated to make the vapor-phase liquid injection volume accurate to within 10ml from 450ml ± 100ml, and the control accuracy of the vapor-phase liquid injection volume is increased from 22.2% to 2.2%, effectively reducing the waste of the vapor-phase liquid injection volume and improving the temperature control accuracy and accuracy of the vacuum vapor-phase reflow soldering adjustment method.

[0232] Step Five: After the temperature curve technical indicators meet the passing criteria, exit the temperature curve debugging mode;

[0233] In this step, the optimized process parameters determined in the 5th paragraph are a peak temperature setting of 227.5°C, the actual equipment parameter setting is 228°C, the vapor-phase liquid injection volume is 559ml, and the vacuum degree is 400mbar. After modification, it is saved in the vacuum vapor-phase reflow soldering equipment;

[0234] Re-perform vacuum vapor-phase reflow soldering using the process board. The process parameters obtained from the sampling data of the time-temperature waveform diagram are a peak temperature of 235.7°C (criterion: 235°C ± 1°C), a liquid phase time of 102s (criterion: 90s to 120s), and a heating response time of 8s (criterion: 8s to 10s). After confirmation, it meets the technical indicator requirements of the passing criteria;

[0235] According to the product input characteristics, the key parameters of the temperature curve are determined, and a temperature curve feature selection table for vacuum vapor-phase reflow soldering is designed, and the temperature curves are numbered, as shown in Table 10;

[0236] The key parameters of the vacuum vapor-phase reflow soldering in Step S1 are determined as the vacuum degree, peak temperature, and liquid phase time, and a process parameter table for the temperature curve of vacuum vapor-phase reflow soldering is designed, as shown in Table 11;

[0237] Combined with different product input characteristics, Tables 10 and 11 are continuously accumulated and iterated;

[0238] Liberate the process personnel from the difficult and complex task of adjusting the temperature curve time and time again. Through the product input characteristics in Table 10, the applied product models, and the picture information, the process personnel can more accurately and quickly select the most suitable vacuum vapor phase reflow soldering temperature curve, thereby significantly reducing the number of tests and production preparation time, improving the process capability, ensuring the stability of process parameters and the consistency of products. The stable process capability is especially suitable for multi-variety and batch production, meeting the needs of flexible production of printed circuit board assemblies with different product input characteristics, reducing rework, and improving the welding quality and reliability of products.

[0239] Table 10 Vacuum Vapor Phase Reflow Soldering Temperature Curve Feature Selection Table

[0240]

[0241] Table 11 Vacuum Vapor Phase Reflow Soldering Temperature Curve Process Parameter Table

[0242] Number: TC1

[0243]

[0244]

[0245] The present invention has been described in detail above in combination with specific embodiments and exemplary examples. However, these descriptions should not be construed as limitations on the present invention. Those skilled in the art understand that without departing from the spirit and scope of the present invention, various equivalent substitutions, modifications, or improvements can be made to the technical solutions and their implementation manners of the present invention, and these all fall within the scope of the present invention. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, directly or indirectly applied to other related technical fields, is similarly included in the patent protection scope of the present invention. The protection scope of the present invention is subject to the appended claims.

[0246] The content not described in detail in the specification of the present invention belongs to the well-known technology of those skilled in the art.

Claims

1. A regression orthogonal test method for adjusting the temperature curve of vacuum vapor phase reflow soldering, characterized in that: include: S1. According to the product input characteristics, which include printed circuit board characteristics and component characteristics, the key parameters of vacuum vapor phase reflow soldering are determined to be vacuum degree, peak temperature and liquid phase time; S2. Making a process board according to the characteristics of the printed board, setting process parameters according to the key parameters determined in S1, the process parameters including multiple temperature segments; Use a process board to perform vacuum vapor phase reflow soldering according to process parameters and collect temperature in real time to obtain a time-temperature waveform diagram; Determine the qualified criterion; compare the waveform corresponding to the temperature segment with the qualified criterion. If the qualified criterion is met, the temperature curve of the temperature segment does not need to be adjusted, and enter S6; if the qualified criterion is not met, the temperature curve of the temperature segment is adjusted according to S3; S3. For the temperature segments that do not meet the qualification criteria, determine the main factors of the regression orthogonal test as vacuum degree, temperature range and vapor phase liquid injection amount; calculate the coding level for the main factors; According to the coding level, multiple test schemes of regression orthogonal design are selected in the regression orthogonal test table; according to the multiple test schemes of regression orthogonal design, process boards are used in sequence to conduct experiments to obtain test index results; According to the test index results and the regression equation, the equation coefficients are calculated and obtained, and the equation coefficients correspond to the main factors; S4. Calculate the F distribution value of each main factor based on the regression equation and equation coefficients in S3, perform a significance test based on the F distribution value, and determine whether the main factor is a key factor or a non-key factor; determine the corresponding coefficient of the key factor based on the key factor and the regression equation in S3; According to the positive and negative signs of the coefficients corresponding to the key factors, the sorted regression equation is determined; according to the sorted regression equation, the updated process parameters are calculated; S5, comparing the updated process parameters obtained in S4 with the qualified criteria determined in S2. If the updated process parameters do not meet the qualified criteria, repeat S3 and S4, and iterate the four process parameters of temperature rise response time, temperature range, vapor phase liquid injection amount and liquid phase time in each test until the updated process parameters meet the qualified criteria; S6. After each temperature segment of the temperature curve meets the qualification criteria, the final determined process parameters are used as parameters for vacuum vapor phase reflow soldering.

2. The regression orthogonal test method for adjusting the temperature curve of vacuum vapor phase reflow soldering according to claim 1, characterized in that: In said S2, the process parameters also include vacuum degree, vapor phase liquid injection amount, exhaust time and liquid phase time; The qualification criteria include heating rate and heating time range; The heating rate is calculated based on the minimum temperature, maximum temperature and operating time within the temperature segment.

3. The regression orthogonal test method for adjusting the vacuum vapor phase reflow soldering temperature curve according to claim 1, characterized in that: Set the process parameters according to the key parameters of reflow soldering determined in S1, including: Product input features include component features and printed circuit board features. Component features refer to the packaging form of components. According to the packaging form of components, the soldering ends of component pins, packaging materials, and specific heat capacity can be determined. Printed circuit board features include printed circuit board area, number of printed circuit board layers, printed circuit board thickness, and printed circuit board copper coverage area. Determine the vacuum process parameters based on the component characteristics and the degree of oxidation of the component lead soldering ends; Determine the peak temperature process parameter based on component characteristics and the percentage of lead in the solder paste; Determine the liquidus time process parameters based on product input characteristics, vacuum level, and peak temperature; According to the reflow soldering temperature curve recommended by the solder paste manufacturer and the component manufacturer, determine the temperature segmentation and obtain the temperature range of each temperature segment; The exhaust time of vacuum vapor phase reflow soldering is determined according to the characteristics of components and printed circuit boards to determine the cooling rate process parameters.

4. The regression orthogonal test method for adjusting the temperature curve of vacuum vapor phase reflow soldering according to claim 1, characterized in that: The main factors for determining the regression orthogonal test are vacuum degree, temperature range and vapor phase liquid injection amount, including: The vacuum level is a vacuum range of a temperature segment that does not meet the qualification criterion; The level of the temperature interval is a temperature interval of a temperature segment that does not meet the qualification criterion; The level of the vapor phase liquid injection amount is a vapor phase liquid injection amount range for a temperature segment that does not meet the qualification criterion.

5. The regression orthogonal test method for adjusting the temperature curve of vacuum vapor phase reflow soldering according to claim 4, characterized in that: The upper level of each of the main factors is the maximum value of its corresponding value interval, the lower level is the minimum value of its corresponding value interval, and the zero level = (upper level + lower level) / 2.

6. The regression orthogonal test method for adjusting the vacuum vapor phase reflow soldering temperature curve according to claim 1, characterized in that: In S3, the regression equation is: t=a±b1V1±b2V2±b3V3±b 12 V1V2±b 13 V1V3, Wherein, V1 is the vacuum degree; V2 is the temperature range; V3 is the amount of vapor phase liquid injected; t is the temperature rise response time; n=m0+m c , where m0 is the number of zero-level trials, m c is the number of trials specified for the orthogonal array; Among them, a is the constant coefficient of the regression equation, t i is the temperature rise response time of each test, and n is the number of tests; Among them, b1 is the partial regression coefficient of vacuum degree in the regression equation, is the vacuum degree of each test, m c is the number of trials specified for the orthogonal array; Among them, b2 is the partial regression coefficient of the temperature interval in the regression equation, is the temperature range of each test, m c is the number of trials specified for the orthogonal array; Among them, b3 is the partial regression coefficient of the vapor phase liquid injection amount in the regression equation, is the amount of vapor liquid injected in each test, m c is the number of trials specified for the orthogonal array; Among them, b 12 is the partial regression coefficient of the interaction between vacuum degree and temperature range in the regression equation, (V1V2) i is the interaction between vacuum degree and temperature range in each test, m c is the number of trials specified for the orthogonal array; Among them, b 13 is the partial regression coefficient of the interaction between vacuum degree and vapor phase liquid injection amount in the regression equation, (V1V3) i is the interaction between the vacuum degree and the amount of vapor phase liquid injected in each test, m c is the number of trials specified by the orthogonal array.

7. The regression orthogonal test method for adjusting the temperature curve of vacuum vapor phase reflow soldering according to claim 1, characterized in that: In S4, according to the regression equation and equation coefficient of S3, the F distribution value of each main factor is calculated, and the significance test of the F distribution value is performed. The result of judging whether the main factor is a key factor or a non-key factor is determined. According to the key factor and the regression equation in S3, the corresponding coefficient of the key factor is determined. According to the positive and negative signs of the corresponding coefficient of the key factor, the sorted regression equation is determined; According to the sorted regression equation, the temperature rise response time, temperature range and vapor phase liquid injection amount are calculated, including: According to the regression equation and equation coefficient of S3, calculate the partial regression sum of squares of each main factor; calculate the total sum of squares; calculate the total partial regression sum of squares according to the partial regression sum of squares of each main factor; calculate the residual sum of squares according to the total sum of squares and the total partial regression sum of squares; calculate the degree of freedom and mean square value according to the partial regression sum of squares of each main factor; calculate the F distribution value of each main factor according to the mean square value and the residual sum of squares; Under the condition of a given significance level α, a significance test of the F distribution value is performed on each main factor, and the judgment result is whether the main factor is a key factor or a non-key factor; Substitute the key factors into the regression equation in S3 to obtain the simplified regression equation; find the positive and negative signs of the coefficients corresponding to the key factors in the simplified regression equation. When the coefficients are positive, it means that the key factors have a significant impact on the test index results when they are at the upper level; when the coefficients are negative, it means that the key factors have a significant impact on the test index results when they are at the lower level; Calculate the corresponding code according to the positive and negative signs of the corresponding coefficients of the key factors, back-substitute the simplified regression equation according to the corresponding codes, and establish the sorted regression equation; According to the reorganized regression equation, the temperature rise response time, temperature range and vapor phase liquid injection amount are calculated.

8. The regression orthogonal test method for adjusting the temperature curve of vacuum vapor phase reflow soldering according to claim 7, characterized in that: In S4, Total sum of squares: Among them, SS T is the total sum of squares, m c is the number of trials specified for the orthogonal array; Partial regression sum of squares: Among them, SS1 is the partial regression square sum of vacuum degree, m c is the number of tests specified by the orthogonal table, b1 is the partial regression coefficient of vacuum degree in the regression equation; Among them, SS2 is the partial regression sum of squares for the temperature range, m c is the number of tests specified by the orthogonal table, b2 is the partial regression coefficient of the temperature range in the regression equation; Among them, SS3 is the partial regression sum of squares of vapor phase liquid injection, m c is the number of tests specified by the orthogonal table, b3 is the partial regression coefficient of the vapor phase liquid injection amount in the regression equation; Among them, SS 12 is the partial regression sum of squares of the interaction between vacuum degree and temperature range, m c is the number of trials specified by the orthogonal array, b 12 is the partial regression coefficient of the interaction between vacuum degree and temperature range in the regression equation; Among them, SS 13 is the partial regression square sum of the interaction between vacuum degree and vapor phase liquid injection amount, m c is the number of trials specified by the orthogonal array, b 13 is the partial regression coefficient of the interaction between vacuum degree and vapor phase liquid injection amount in the regression equation, SS 12 is the partial regression sum of squares of the interaction between vacuum degree and temperature range; Total partial regression sum of squares: SS R =SS1+SS2+SS3+SS 12 +SS 13 , where SS R is the total partial regression sum of squares, SS1 is the partial regression sum of squares for vacuum degree, SS2 is the partial regression sum of squares for temperature range, SS3 is the partial regression sum of squares for vapor phase liquid injection, SS 13 is the partial regression sum of squares of the interaction between vacuum degree and vapor phase liquid injection amount; Residual sum of squares: SS e =SS T -SS R , where SS e Residual sum of squares, SS T is the total sum of squares, SS R is the total partial regression sum of squares; Degrees of Freedom: df R =∑df 一次项 +∑df 交互项 =m, where ∑df 一次项 is the partial regression sum of squares of each factor, ∑df 交互项 is the partial regression sum of squares of the interaction factors; df e =∑df T -df R =n-1-m, where df e is the residual degrees of freedom, ∑df T =n-1, n is the number of trials; Mean Square Value: Among them, MS i is the mean of each factor or interaction factor, SS i is the partial regression sum of squares of each factor and interaction factor, df i is the degrees of freedom of each factor and interaction factor; Among them, MS e is the residual mean, SS e is the partial regression sum of squares of the residuals; df e is the residual degrees of freedom; F distribution value: Among them, F i is the F distribution value, MS i is the mean of each factor or interaction factor, MS e is the residual mean.

9. The regression orthogonal test method for adjusting the temperature curve of vacuum vapor phase reflow soldering according to claim 7, characterized in that: In S4, at least one F distribution value significance test is performed on each main factor to obtain a judgment result of each main factor as a key factor or a non-key factor, including: Query the regression orthogonal test table to obtain the F critical value when the significance level is α = 0.05 and α = 0.01, compare the F distribution value of the parameter with the F critical value, and obtain the judgment result of each main factor as a significant factor or an insignificant factor; The partial regression squares and degrees of freedom of the insignificant factors are incorporated into the residual squares parameters, the F distribution values ​​of each parameter are recalculated, and then an F distribution significance test is performed to obtain the judgment result of each significant factor as a key factor or a non-key factor.

Images