Relating Quality Data to Manufacturing Six Sigma or Cpk Levels

There are many steps in the manufacture of printed circuit boards (PCBs). They include the preparation of the components and the fabricated PCB, the placement of the components or their insertion into predetermined locations on the PCBs, and the attachment of the components to the PCBs through the application of solder joints.

In order to control the quality of manufacturing the PCBs, some of these steps have their own recommended specifications from the equipment and material suppliers. However, a direct relationship of these specifications and the defects occurring during the PCB manufacturing steps is not readily discernible. This has sometimes led to a manufacturing process having a high-quality Cpk for the process meeting its individual specifications, yet having a very poor effective PCB assembly yield. This could result in a loss of credibility in the Cpk values in manufacturing.

An example of such a problem is in the SMT assembly operation in PCBs. The assembly consists of applying solder paste onto PCB component pads through a thin metal stencil in a screening machine, then placing the components onto the pads using an automatic placement machine. The components remain on the PCBs because of the tackiness of the solder paste. The final operation consists of passing the PCB through a conveyer oven to reflow the solder. The solder paste suppliers recommend a particular paste volume and height of the solder deposited on the pads and a particular temperature profile for the reflow oven. A Cpk of the solder paste and reflow operations can easily be obtained from control chart or process capability data.

High Cpk levels in solder deposition, oven profiles, and other indirect measurements of quality do not necessarily lead to high yields in PCB assembly. This has resulted in the need to develop composite Cpk analysis based on direct defect analysis for each step of the PCB assembly operations. These will be discussed in Chapter 8.