Issue link: https://iconnect007.uberflip.com/i/1424540
22 SMT007 MAGAZINE I NOVEMBER 2021 But in all that time, and with all the chang- es our industry has gone through, the accep- tance criteria had not changed before 2020. Of course, one could argue that, even still, the criteria wasn't so much changed as it was removed completely. Beyond that, we also need to remember what the most important part of this whole discussion might be: the ROSE test was nev- er intended to be used for determining what a qualified manufacturing process is. e idea was to only use it as a process monitoring tool during production to detect gross changes. It didn't take long for most assembly houses to start using the ROSE test for process qualifi- cation because, quite honestly, in most cases it wasn't hard to pass. at passing result gave you a warm fuzzy feeling about your process and it met a published industry standard. Ev- erything was great, right? Well, maybe, may- be not. at is one of the biggest issues with using the old acceptance criteria on far more tech- nologically advanced assemblies. You passed the test but what did it mean without objective evidence specific to your process and material set? For decades, a lot of manufacturers never thought about that part of the equation. Con- tract manufacturers just kept building prod- uct, and in some cases, they were not clean enough to be used in the intended end use en- vironment. is oen manifested in electrical leakage-related failures as shown in Figure 3 of an assembly that passed the ROSE test. e assembly in Figure 3 used a manual sol- der operation for the plated through-hole (PTH) connector pins. is was the only part of the process that was leaving high levels of ionics, but when that was averaged out across the full surface area, it was not detected as an issue. is would be a great time to bring up the topic of full board extractions vs. local- ized cleanliness analysis in areas of concern or sensitive components—but I digress. e point is if this company had done more test- ing up front to determine what a golden board ROSE test result should have been, instead of blindly adopting 1.56 µg/cm 2 , this might have been detected before it became a much larger issue. To end my soliloquy on the past, I'll just say we can do better and that is where the new J-STD-001H comes in. Creating objective evidence to meet the new standard isn't as difficult to achieve as some seem to think. e underlying data may already exist in some form to satisfy the new require- ment. I think this part of the process is being overlooked by many, but it is right there in black and white within the standard. J-STD-001H Section 8.1 lays out three ways to create objective evidence. e second note references historical evi- dence, looking at returns, war- ranty service records, failure analysis results, etc., that show the issues for those failures aren't related to cleanliness Figure 3: Passed ROSE test.