Measuring the Capability of Current Manufacturing Processes

Measuring the Capability of Current Manufacturing Processes

Publish Date:2017-09-19 19:41:13 Clicks: 162

Up-to-date capability data for the manufacturing processes to build new products have to be available to the new product design team. The data can be used to calculate the design and manufacturing quality of the new product. The data should contain the process average 3nd standard deviation, as well as design guidelines for design for the manufacture (DFM) and early supplier involvement. In addition, the data has to be updated regularly, typically every quarter, so that the design team is working on the latest capability of the manufacturing processes. These processes can be divided into two categories:


1.Processes that are used to build current products similar to the new one, with adequate process capability. These processes should have long-established capability of meeting six sigma (or specific Cpk) requirements. They should also include guidelines for DFM and ESI.

2. Processes building current products that are not capable for all operations. In this case, the manufacturing process engineers should collect a list of alternative manufacturing processes available that can make products with varying quality depending on the specified parameters.

An example of quality data collected for PCB assembly is shown in Table 8.4. This example is for a mix of surface mount technology (SMT) and through-hole (TH) design. Several options are available to the design team for specifying certain manufacturing process parameters. For example, specifying laser stencil or paralene conformal coatings will result in greater quality than etch stencils or acrylic spray coating. The design team has to select process and material parameters based on the quality and cost goals of the new product.

Once these process quality parameters have been identified a measure of typical defect rates for PCBs can be generated. Any new PCBs to be designed can be analyzed for quality, given the component count. The defect rate is normalized by the number of opportunities based on terminations of leads or solder joints per component, as well as the DPMO method, discussed in 

Chapter 4. A quality analysis for a new PCB is shown in Table 8.5, with typical quality levels for the various PCB assembly operations. The PCB is two-sided, with many components of various technologies, including automatic insertion of through-hole (TH) and placement of surface mount technology (SMT).


In addition, the PCB has 40 components leads to be hand soldered and 20 mechanical parts to be assembled. The PCB will also have to soldered and washed. The component counts have been translated to defect opportunities depending on assembly operations such as the number of component leads, solder joints or mechanical assembly. The resultant quality level is 1.51 defects/PCB or 22% FTY. This is very poor quality and will necessitate extensive testing. An exercise such as this example might prove to be very positive for increasing the design team focus on the quality drivers for PCBs discussed next.

For the cases where the quality of current operations are not adequate, a list of drivers should be generated to alert the design team to the critical attributes of the design that will influence the quality of manufacturing. The design team can thus focus on modifying the design to allow manufacturing to build the new product to the specified level of quality (six sigma or Cpk target). An example of such a list for PCBs is shown in Table 8.6. In many of the items in that table, the geometric properties of the components of PCB layout or the PCB warp specifications are shown to be important in increasing the quality of the PCB assembly. Unfortunately,it is difficult to generate a precise amount of quality improvements associated with items on this list.

label: PCB

Copyright 2009-2020 All Rights Reserved by NOD Electronics
Building E, Qixing Industrial Area, Xintang Town, Zengcheng District, Guangzhou 511340, China
Powered by MetInfo 7.2.0 ©2008-2024