In Circuit Pin Testing- An excellent potential source of value creation
Posted 6/11/2013 by Mitch Holtzer
Unlike printing paste, placing components and reflowing solder, ICT is considered a non-value added process. In fact, ICT increases the cycle time of the assembly process. If the ICT is a fail, but the circuit being tested is actually good (known as a false negative), even more time is wasted trying to determine whether or not a good assembly will function properly.
There are two basic types of ICT. They are commonly referred to as clamshell and flying probe. (See figures 1 and 2)
Figure 1 Clamshell ICT Device
Figure 2 Flying Probe ICT Device
The clamshell ICT simultaneously tests dozens of points on a single board. Alpha’s clamshell ICT device, created by Karen Telefsen, uses three types of pins and four different forces. Crown, spear, and blade pins are shown in figure 4. A test vehicle is placed in the clamshell fixture. The clamshell is closed, and electrical resistivity is measured at each of the test pins.
Although clamshell devices are commonly used, they are somewhat impractical for testing solder paste and flux’s ability to be probed. There is a 15 minute cycle time per test. Pin residue build up generally occurs after hundreds of tests, so flux buildup would be a very time consuming process.
Flying probe testing is also commonly used in industry. It is more versatile in that special tooling is usually not required, so high mix applications are a good fit for flying probe testing.
In the Alpha flying probe procedure, one pin type and one force are used, but the test runs for a total of 4,000 strikes. Flux build up on the pins is an important part of the test. In addition, the Alpha test vehicle contains 4 different types of pads. Pad A is a 40 mil (1mm) square pad without vias, pad B is a 40 mil (1mm) square pad with 13 mil (0.33mm) vias, pad C is a 28 mil (0.7mm) round pad without vias, and pad D is a 28 mil (o.77mm) round pad with 13 mil (0.33mm) vias. (see figure 5). There are 1,000 opportunities for each pad type.
Pads with vias tend to collect more flux residue, and generally have lower yield for a given paste or flux. Often these pads are the source of discrimination between high performing paste and flux.
Value is created when the first pass yield of the ICT approaches 100%. This means that the pin probe is able to penetrate flux residue and make positive contact with the test point in the circuit. Value is lost when the probe touches a fully functioning test point, but the test shows an open circuit. This is a false negative. Even though the assembly is perfectly functional, re-testing and extra handling are required, increasing the assembly cycle time.
Based on actual test data, a value in use model has been created to estimate the value of increased first pass ICT yields. In Alpha’s testing, a leading competitive solder paste has a relatively low first pass yield, compared to CVP-390 or OM-340. In mobile phones where functional testing is preferred over ICT, the value is difficult to document.
A value in use model, based on the first pass yield of OM-340 versus the leading competitive solder paste was reviewed with the customer. The customer agreed that the savings in first pass yield could save the customer $128 for every kilogram of paste used! The customer has approved OM-340.
In conclusion, knowing if and how a customer tests assemblies is a potentially powerful way to create value. Alpha has data on every modern paste and flux, and this data can be used to create value in use models. Try it, it works!