Figure 4.5 is an example of the test methodologies available for PCB assemblies. These methodologies can be summarized as follows:
1. Visual test and inspection. These tests use trained operators to inspect PCBs for defects using the naked eye or visual magnification such as microscopes or enlarging lenses. They concentrate on geo- metrical defects that are easily observed by the human eye, but may be difficult for machines to detect, such as solder shapes and shorts.
2. In-circuit test (ICT). This type of test is used to eliminate defects that result from individual components not meeting their specifications. The defects either due to the components being defective as supplied, or becoming defective through the PCB assembly operations. They could either be missing, wrong, placed or inserted into the PCB incorrectly, or become defective because of PCB assembly operations exceeding manufacturing specifications.
The ICT test consists of a machine with electronic means of comparing the components to a preprogrammed value. The components, already soldered in place on the PCBs are reached through a bed of nails fixture that provides contact of the component pads on the PCB to pins in the fixture. Many sources of electronic noise may be present, such as stray capacitance and resistance in the fixture and its wires. In addition, some components in the circuit are used in parallel with other components, so that it is difficult for the tester to isolate the individual component to be tested.
The ICT is not always capable of detecting all component defects because of the tester connections to the circuit. This inability to detect all of the component defects is called defect or test coverage. A low test coverage, under 90% yield of good PCB*s into the next test cycle, will result in the need for a functional test of the PCB.
3.Functional test (FT). This type of test is used to eliminate design- based tis well as assembly defects. The latter occur when the set of
components being assembled meet their individual specifications and are assembled correctly, but the assembled PCB does not meet its systems specifications. Functional testers use the PCBs in a similar fashion to their intended use in the product. In its simplest form, FT is called a box test, which consists of testing the PCBs in a fully functional product.
4.The customer of PCB test is the next level of production. It usually is the product assembly process, where the PCBs are combined with mechanical parts to form the product. The product might undergo additional testing such as burn-in, product test, and system test.
Product test occurs after the product has been assembled with PCBs and other mechanical and input/output modules. Burn-in occurs when the product is subjected to environmental stress conditions, and then tested to see if there were any “infant mortality” failures. System test occurs when the product is combined with other products with maximum-length cabling to form a system configuration similar to those in customer sites.
The different types of PCB tests have different expectation of defects removal. Test engineers usually communicate quality through using the PCB yields from each of the test methods mentioned above, whereas the assembly community communicates through DPU or DPMO. The management sets enterprise goals at certain Cpk levels or six sigma. Using the examples in this chapter, it was shown that quality communications could be just as effective using any of these common methods outlined above.
The PCB test strategy is formulated based on the lowest-cost alternative for removing defects, given the current quality level of the design and manufacturing process. To achieve a successful strategy, the costs of each type of test as well as visual inspection should be known, including nonrecurring costs such as test development and programming, fixture design, and troubleshooting, A test strategy could be developed to provide the proper balance between investing in improving the PCB assembly process capability versus performing test and troubleshooting to remove defects generated in design and assembly. This can be accomplished given the availability of alternative test method costs, and having the defects and their sources quantified using methods outlined in this chapter.