Develop a robust process for the hottest lead-free

  • Detail

Lead free welding: develop a robust process

this paper will study the method to determine what parameters have the greatest and minimum impact on lead-free welding. The purpose is to establish a lead-free process with controlled quality and repeatability

develop a set of robust methods

whether a welding process is robust depends on its ability to maintain a stable output (qualification rate) for various inputs. Input changes are caused by "noise" factors. Even before the printed circuit board (PCB) enters the reflow furnace, some factors will change in a surface mount assembly

first, there are changes in the materials used in the process. These changes exist in the characteristics of solder paste, such as composition, lubricant, powder and oxide; The material of the board, considering different suppliers and different storage characteristics; And components. Secondly, changes may occur in the first part of the surface mount process: solder paste printing and collapse and component mounting. Third, the noise factor can come from the indoor conditions of the manufacturing area - temperature and humidity. These input variables require the best heating curve, which must be least sensitive to all variables, and a method to quantify process capability

reflow curve

in terms of reflow soldering, the use of lead-free alloy directly affects the process temperature, thus affecting the heating curve. Increasing the melting temperature reduces the process window because the time above the liquidus and the maximum allowable temperature of 250 ° C (to prevent component damage and plate delamination) have not changed

triangle (heating up to peak value) curve

we can distinguish between those critical processes that are close to the realistic limit of reflow soldering and those less critical processes. For the process that PCB is relatively easy to heat and the temperature of components and board materials are close to each other, a triangular temperature curve can be used (Figure 1). Triangular temperature curve is recommended for products such as computer motherboard, which has small temperature difference in assembly (small Δ T)。

Figure 1. Triangle reflux temperature curve figure 2. Temperature rise insulation peak temperature curve

triangle temperature curve has some advantages. For example, if the solder paste is properly formulated for the lead-free triangle temperature curve, brighter solder joints and improved solderability will be obtained. However, the activation time and temperature of the flux must conform to the higher temperature of the lead-free temperature curve. The heating rate of triangle curve is controlled by the whole process, and it remains more or less the same in this process. As a result, the stress in the PCB material is small during welding. Compared with the traditional temperature rise insulation peak curve, the energy cost is also lower

temperature rise insulation peak temperature curve

safe and reliable

smaller components rise faster than larger components and heat sinks. Therefore, in order to meet the liquidus of all components, pay attention to: if the difference between the calibrated display and the dynamometer display is relatively large, it is better to use the temperature rise insulation peak temperature curve for these processes (Fig. 2). The purpose of insulation is to reduce Δ T。

in several areas of the temperature rise insulation peak temperature curve, if not properly controlled, it may cause too much stress in the material. First, the preheating rate should be limited to 4 ° c/second or less, depending on the specification. The flux element in the solder paste should be targeted at this curve formula, because too high insulation temperature can damage the performance of the solder paste; Sufficient activator must be retained in the peak area where oxidation is particularly serious. The second temperature rise slope occurs at the inlet of the peak zone, with a typical limit of 3 ° c/second

The third part of the temperature curve is the cooling zone, and special attention should be paid to reducing stress. For example, the maximum cooling rate of a ceramic chip capacitor is -2~-4 ° C/s. Therefore, a controlled cooling process is required, because the reliability of special materials and the structure of welding joints are also affected

for any process, the best temperature curve can be determined by a Taguchi test. The use of noise factors in the test will help determine which curve is least sensitive to variables and more stable

evaluation process

statistical process control (SPC) is used to stabilize and maintain the process under control. In welding, SPC is used to reduce variability and provide process capability. Typically, x-bar-range chart and performance analysis are used for this purpose. X-Y coordinate diagram is a graphical representation of statistical calculation of measured variables, where the average value and amplitude (maximum minimum) of each group are used to monitor the change of average value or range; This magnitude is used as a measure of the variable. Statistically significant changes may indicate process drift, trends, cycle patterns, or runaway conditions caused by special reasons

when the most influential parameters of the welding process (as defined by Taguchi test) are subject to statistical process control (SPC), the stability and performance improvement of the process can be easily achieved. For example, in a welding equipment, hardware and software are designed to keep important parameters within the specified range of set points. However, even when a parameter is within the starting deviation limit (no alarm occurs), it may have been statistically out of control, or show an unexpected state due to historical data

buying only hardware and software may not necessarily lead to a successful SPC. A key consideration is the reduction of variability. There is a difference between special cause variables and general cause variables. Control charts are used to eliminate special cause variables, that is, any variables that may be associated with attributable causes. Performance charts are used to reduce common cause variables, which are inherent in any process and can only be reduced by process changes

in a reflow soldering process, typical parameters of SPC include conveyor speed, gas or heater temperature, time above liquidus and maximum peak temperature. In a wave soldering machine, typical parameters include conveyor speed, contact time, preheating temperature (PCB or heater) and the amount of flux acting on the PCB

Figure 3. X-bar-range diagram describing preheating temperature Figure 4. Sample process capability (CP) diagram

an example of X-Y coordinate diagram shows the hot air temperature in the preheating zone in a wave soldering process (Figure 3). During the whole day, the sampling readings are taken every 10 seconds and divided into groups with five samples in each group. The average value and amplitude are shown in Figure 3. The average temperature is 120.0 ° C and the set point is also 120 ° C. This data comes from thermocouples installed in the preheating module. All machine data from processes, settings, and measurements are recorded. The management information file can be imported into SPC Software, which will produce X-Y coordinate diagram and performance analysis diagram like Figure 3

we accept the preheating temperature (hot air) of 120 ° C ± 2 ° C, because we know that as long as the measured temperature is within this limit, the temperature of the plate will not fluctuate and remain within the flux specification. This data, together with the upper control limit (ucl=122 ° C) and the lower control limit (lcl=118 ° C), will return a process capability value: here, CP = process capability, s = standard deviation

the CP diagram in Figure 4 shows that the process is capable of preheating temperature. We found that cp=3.55; A stable process requires a CP value greater than 1.66


once we run the first batch of products with lead-free solder, we need to quantify the stability of the process. These response factors can be measured on the product, such as counting defects or collecting data from machine settings. For example, the temperature of a board can be measured with a thermocouple installed on the PCB, or the temperature of hot air can be measured in the machine, which is related to the temperature on the PCB

another method of measuring stability is to use a special calibration tool, which rides the instrument on the conveyor belt through the furnace. The advantage of using these tools is that they are very stable. An electronic processing is completed on a 2D plane, and several different parameters can be measured. In most production lines, operators have their own test boards on which thermocouples have been installed. Running the plate in the furnace (or wave soldering machine) will soon damage the test plate because of the high temperature of lead-free welding. The plate will begin to delaminate and really reflect this kind of warping, and the thermocouple may fall off the surface

the parameters that affect the quality most are known from our understanding of tin lead process and the results of Taguchi test. We start counting and collecting data for these parameters. After SPC has proved that a parameter is controlled for a long time (CP> 1.66), the measurement interval can be reduced. Using SPC, we only focus on some of the most important parameters. Pareto diagrams will also help define these parameters to be measured to maintain process stability

x-y coordinate diagram shows the drift, trend, cycle mode of the process or out of control conditions due to special reasons. In some cases, preventive measures can be taken before runaway conditions actually occur

exhaust and temperature conditions

the whole process has changed with the introduction of lead-free alloy. The temperature has risen in all modules of the machine. For reflow soldering, higher temperature range and peak temperature are obtained. For the cooling zone, a more effective cooling method than normal is required, because the peak temperature is higher. The furnace should be designed to meet these higher temperatures, but the machine temperature should be verified at this stage of implementation

lead free solder paste has different chemical composition from traditional tin lead formula. Therefore, we have to deal with other residues evaporated at different and higher temperatures. Thermogravimetric analysis can help define where and at what temperature the material can evaporate. An adequate flux management system is needed to control the removal of all residues. In addition, the exhaust and exhaust settings should be verified before the process is implemented in the production environment

evaluate reliability

reliability tests should be carried out to predict the life cycle of products and compare data with the standard of tin lead process. More conclusions about the strength of lead-free solder joints are obtained from shear, tensile and temperature cycling tests. The cross section will show the intermetallic layer and the increased thickness, which is also related to reliability

process release is used to implement

now we have reached the next milestone in the implementation stage. Once all the conditions are met, we can get the following results:

the process is stable and repeatable

the machine condition is controlled

the quality and reliability of the welding point are within the specification

the cost is acceptable

therefore, the process can be issued for implementation. So far, tests have been carried out on independent machines, or machines in laboratories or demonstration rooms, or production line machines during downtime. The next step is to transfer the technology to the production line. However, a lot of work needs to be done before production starts. These include: project schedule, quality problems, out of control action plan (ocap), and operator training

project schedule

create a schedule for all implementation actions. This schedule will consider purchasing materials and necessary machine accessories, organizing personnel and materials for adjustment, writing procedures and ocap, and training operators and engineers

quality problems

(wave soldering) the solder in the tin pot will be polluted after long-term production. Try to establish the specification of the maximum allowable contamination of the alloy. Customer specifications or guidelines from research institutions can help determine

Copyright © 2011 JIN SHI