Circuits Assembly - December 2008 - (Page 34) Wave Soldering Analyzing Pb-Free Defects Using Partial DoE Optimal assemblies don’t come about from drop-in materials replacement. n important step in reducing and preventing defects during soldering is defining the correct process settings. It’s a good practice to design proper experiments and analyze their results to help determine these settings. When the number of test boards and components for experimentation are limited, use of Partial Factorial, Taguchi or Response Surface Experiments makes for a good approach. These techniques permit end-users to understand the process by minimizing time, resources and cost. The objective of these experiments is to determine the factors and their levels that minimize the variation of a product around a target response. The champion settings will result in robust products resistant to change in operational and environmental conditions. The experiment begins with a brainstorming session where the problem is stated; the factors are selected; the measurable output is chosen, and the experimental design is selected. Then the experiment is run and output measured. Data are analyzed and the best settings are identified. It is important to schedule confirmation runs based on the champion settings; if these results are good, the settings can be implemented into production. To illustrate the concept, a Taguchi experiment was used to develop a robust process for Pbfree wave soldering. The control factors selected were solder temperature, contact time, preheat temperature and flux amount. Circuits Assembly DECEMBER 2008 A Ursula Marquez de Tino is a process and research engineer at Vitronics Soltec, based in the Unovis SMT Lab (vitronics-soltec.com); umarquez@vsww.com. For practical Table 1. Factors and Levels for the Taguchi Experiment reasons, there Level 1 Level 2 Factors Level 3 were no “noise” Solder 250 (A1) 260 (A2) 275 (A3) Temperature (˚C) elements in 1.8 (B1) 3 (B2) Contact Time (s) 4.2 (B3) this experiPreheat ment. The 90 (C1) 110 (C2) 130 (C3) Temperature (˚C) output charFlux Amount 355 (D1) 474 (D2) 639 (D3) (mg/cm2) acteristic was the number of pins without bridges (maximum 200 pins/ board) and through-hole penetration (2 = 100%, 1 = partially wetted, 0 = Figure 1. Taguchi analysis for bridging. Higher scores are not wetted). better. This experiment used an L9 orthogonal array design, which means nine experimental runs with four factors and three Figure 2.Taguchi analysis for through-hole penetration. levels investiHigher scores are better. gated (Table 1). are shown in Figures 1 and Ma te r i a l s used for 2, respectively. In both analythis ex p er i m en t were ses, higher scores meant betSnAg3.8Cu0.7Sb0.25% alloy ter results. (Table 1 also shows with a VOC-free flux (<2% codes used in the graphics.) solids). The test board was For bridging, contact time double-sided (160 x 100 x 1.6 and preheat temperature were mm) with plated through-holes more important, while for and Cu OSP Entek Plus surface through-hole penetration, prefinish. A Delta wave soldering heat temperature was the main machine equipped with a spray contributor. The optimal setnozzle fluxer, three bottom side tings are not the same for bridgpreheaters, and nitrogen was ing and through-hole penetraused. tion. Therefore, some sacrifices Eighteen boards were run must be made. For our analysis, (nine runs with one repetithrough-hole penetration was tion). The results for bridging more important. and through-hole penetration circuitsassembly.com 34 http://www.vitronics-soltec.com http://www.circuitsassembly.com
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.