Issue link: https://iconnect007.uberflip.com/i/881969
October 2017 • SMT Magazine 23 bottom termination components, people don't think three-dimensionally about them. It's easy to miss these kinds of failures because they're not as apparent—you can't see them with your naked eye. I think there's a big gap between the way the designer is thinking about the design because they're very concerned about the way the connections are being made inside of the board with all of that complexity. I don't think they're always thinking ahead to design- ing for manufacturability." "We talk to a lot of process en- gineers as we're developing our tools and one of the things that's really become interesting for us is that the electronics assembly pro- cess, with some exceptions, is not really a discipline when it comes to engineering," says Roush. "What you tend to talk to are ei- ther electricals or mechanicals that have been learning bits of the other person's profession, and there's a lot of interaction going on. So, the mechanicals may understand how to physically assemble stuff, but there's some electrical considerations that they're not maybe aware of or that they've for- gotten. There's really been a blending of these process engineers where it's predominately mostly during the on-the-job training (OJT) where they've got to figure this out as they go along; if they're lucky, they've had somebody who's dragged them along in their learning process. What's really been interesting is we've talked to people over the last couple years and heard what's going on and how they're solv- ing the problems, and their approach to solving that problem." Burns notes that they have seen the same is- sues across a variety of customers when it comes to BGAs and bottom termination components. "As a CM, a lot of times we'll catch those things because we're looking for it as we know the issue. So, if we catch it during our review upfront, then we can talk to the designers and try to get them to change those designs. Some- times we're successful and sometimes we're not, but part of that issue goes back to the basic ed- ucation of the engineers. How many engineers, either electrical or mechanical, study solder and the flow properties of solder when they go through school? You know, getting their bach- elor's degree even at that level. There's not a whole lot of understanding necessarily by the engineers that 'Hey, this solder is going to flow somewhere and it's going to go down this hole.' They may not even think about it until they get that OJT and that education in the field. It would go a long way toward getting the de- sign engineers educated upfront in the properties of the materials that are going to be involved in this system, because they don't have any clue or they don't even think about," Burns explains. "Some of these new cir- cuit components are purposely warped before you actually ap- ply heat, and then they relax into a flat state once they've had heat. Some of these new packag- es are extremely challenging and getting more so," says Roush. "I know that the designers' in- centive is to pack more density in there and get more processing out of smaller spaces, but where we come in to play in that realm is re- work. You know, they've spent all this time in development and then they say 'Well, we can't launch it until you can prove that you can fix it,' and there's quite a few hoops you end up jumping through at the end of the day to get it to work where you're manipulating custom stencils, or sometimes, we've had to go out and design entire machinery just to be able to sup- port some of the applications out there because we know that when our customers come back to us and say, 'We're trying to rework this com- ponent and no one else can do it' or 'There's just not a solution in the factory.' It takes a lot longer to get the part's acceptance in the indus- try, and that's really the challenge. We've done it over the past couple years where we've done evaluations, and I'm just now seeing them show up on boards. It's taken that long to get the ac- ceptance and I think part of it is that you can work out all that you know. How do you run through the oven and the materials and every- thing else, but when it goes wrong on that rare ACHIEVING THE PERFECT SOLDER JOINT: THE MANY PERSPECTIVES ON SOLDERING Andrew Nunenkamp