Issue link: https://iconnect007.uberflip.com/i/745476
84 SMT Magazine • November 2016 Goldman: Yes. This interview should be an excel- lent way to perhaps encourage some other EMS providers to get involved in this. Russeau: Mark, Tim, and I would definitely like to see other people get involved—especially to help propose additional perspectives, provide parallel testing for correlation, and to continue to move the bar for IPC test methods capabili- ties higher. Goldman: So you can expand it and get to some point where things either correlate, or come to some real conclusions here. Russeau: Yes, like I said earlier, with industry electronics companies de- signing tighter conductor spacing, smaller board real estate, and high- er component densities there is a renewed interest in understanding what the cleanliness impacts are to those products and how to accu- rately predict those impacts before products go into service. Whether or not we find a correlation is anyone's guess at this point. There's not been enough data gen- erated yet to remove uncertainty. Who knows, this ongoing testing may lead to an entirely new set of techniques to replace the current ones to support the electronics industry ever increasing demands. Goldman: And also higher reliability is required, which is another side to that. Russeau: Absolutely. One of the other challeng- es that we have in reliability is differences in- volving environmentally friendly materials, both PCBs (e.g., halogen free) as well as fluxes (e.g., no-clean). Both are vastly different today than they were many years ago. The companies that manufacture these are trying to find better materials, with no negative impacts on reliabil- ity. Trying to match the methodologies we use for testing to keep up with those changes is be- coming extraordinarily difficult. Goldman: After this paper, what do you see as the next step? Russeau: Excellent question. We're still in the beginning stages of discussing that. What we're probably going to do is to see if we can home in on what the cleanliness requirements should be. As I mentioned earlier, the ion chromatog- raphy cleanliness levels that we used were IEC Electronics. They have a long history with us- ing them. They have some level of confidence in those limits for harsh environments, but we have to decide whether or not they are appli- cable in how we're trying to use them for com- mercial, medical, industrial, and military envi- ronments in this testing. Remember, presently there are no ionic limits defined for either PBA or PB CAF testing. Add to this, another chal- lenge since CAF is an inner layer board phenomena, so will IC test- ing predict it, or do we need to look at other types of test methods to be able to get in and see if there are other chemicals within the inner layers that are causing these prob- lems? Like I said, this is just the be- ginning step to see if there is any correlation. There's much more work to be done. Goldman: It seems like the importance of this isn't going to go away. It's only going to increase. Russeau: Correct, the importance of it is that we have to find ways of advancing the testing methodology to keep pace with the ever in- creasing electronics density technology. The ultimate goal is to improve our ability to risk mitigate and if we're using prehistoric methods from the '70s, '80s and '90s, then it is imper- ative that we update our testing methods and support them with strong data validation. Goldman: Yes, that's a long time ago. Russeau: It is. So that's what Mark, Tim and I are trying to look at while asking for support from others going forward. Goldman: Excellent. Joe, thank you so much for your time. Russeau: You're welcome. SMT INCREASING RELIABILITY THROUGH PREDICTIVE ANALYSIS