Issue link: https://iconnect007.uberflip.com/i/1010078
AUGUST 2018 I SMT007 MAGAZINE 37 demonstrate cases where a high CI value is a poor indicator of SIR performance. Conclusions The design of the cleanliness test system assumes that the more corrosive or conductive the extracted residue, the quicker a current leakage event occurs. [4] In addition, the CI index is marketed as a predictor of product performance in that the lower the CI the less likely a tested product site will suffer performance problems due to the presence of a detrimental ionic residue. [5] However, a comparison to SIR testing results indicates that these assumptions do not hold true when testing no-clean liquid wave solder fluxes. Recall that SIR testing requires no evidence of corrosion on comb patterns, no evidence of electrochemical migration that exceeds 20% of the distance between combs, and maintains 100 MΩ minimum resistance under controlled, elevated temperature and humidity conditions to receive a pass result. In addition, all of the fluxes tested (with the exception of Flux 13) are classified as "L" activity per J-STD-004. This classification is derived from the results of corrosion testing on full-strength flux and indicates that the fluxes are non-corrosive to copper. Flux 13 is a special case, where it is rated "M" activity (which indicates it can be mildly corrosive to copper) but still meets all the requirements to be called a "no-clean" flux per J-STD-004B. The localized cleanliness test system appears to be prone to a high rate false negative errors, with 11 fluxes reporting a dirty result where they pass SIR testing. These results show that users of the localized cleanliness test system could be rejecting useful fluxes based on the results of localized cleanliness testing. This is burdensome to users, as it tends to limit the available fluxes to select when developing new processes. This can also lead to situations where end users have localized cleanliness testing performed and require the assembler to respond to undesirable localized cleanliness testing results. This response can include additional product testing or implementation of a new assembly process in order to receive a desirable result of cleanliness testing, in cases where the original process was acceptable in its current condition. Also, there were two fluxes tested that the localized cleanliness test system suffered false positive errors, reporting clean cleanliness results while they fail SIR testing. These results show that users could be accepting fluxes that will not provide the expected reliability performance in the product's service environment. This can be especially dangerous, as field returns due to poor reliability performance can be very damaging to an assembler's business. The only fluxes that did not experience errors are fluxes 14 and 15, which are both very old technology formulations and are infrequently used for new applications and processes. In fact, Flux 15 is a flux designed for solderabil- ity testing and is not intended for assembly soldering. The study of CI values and their relation- ship to final SIR values shows that there are a significant number of cases where the desir- able results (low CI, high SIR) exist. However, there are also cases where low CI values (desirable) are matched with low SIR Figure 3: CI and final SIR values.