Issue link: https://iconnect007.uberflip.com/i/1161956
SEPTEMBER 2019 I SMT007 MAGAZINE 53 ment manufacturers in the world. My response was, "Wow. This is serious." The group debated a series of formats and spent at least a year talking about a format. When your machines can all talk to each other, the next concern is going to be, "What about security? I don't want my line hacked, or my information shared over the internet." So, we picked a proto - col that's one of the most secure available—the same format that the banking industry used. Still, there was a lot of debate because this would be a series of choices that we would make in support of electronics manufacturing. We chose the most secure one in the end. We wanted to choose methods that could go machine-to-machine and yet weren't com- pletely tied to a single broker so that it was flexible. We had a significant donation from one of the industry software providers who cre- ated a software developer kit (SDK) that low- ered the barrier of effort for companies to get the messages from their machines into CFX. That SDK was probably one of the most sig- nificant efforts in reducing the amount of time needed to get CFX to market. The IPC CFX SDK is available for free on Github. Our intent for this was to look at the sur- face-mount manufacturing process; it's what we had in mind when we were working on CFX. But as other industries have seen our approach, they have reached out, comment- ing, "I could use CFX for metals forming. And you could find a way of creating messages to build automobiles with CFX given the right set of time and messages." What I'm looking for- ward to is seeing the implementation of how the smart factory comes to life on the founda- tion that CFX forms. I can tell you that there are several active implementations happening. Johnson: That's exactly why it seems to be a model. I'm picking up from you that there's similar momentum going on with automotive? Bergman: With the automotive side, there are multiple efforts and different technology seg- ments with three or four committees. CFX is a single committee with subcommittees and task groups. In the automotive efforts, though, we have a PCB group, press-fit pins group, an assembly group, and a wire har- ness group, reaching across all of IPC's man- ufacturing areas. There's some commonality of leadership, but it's not a single committee driving it. Certainly, the energy is there, and seeing the success has led to additional suc- cesses. There is interest in supporting electric vehicles. There's also interest in pushing more data, particularly within the trucking and agri- culture industries. They have to know where the trucks are at all times, so the data require- ments are increasing. Johnson: What's clear is that CFX is an infra- structure project—a conduit for communica- tion to enable facilities to improve their capa- bilities, consistency, throughput, and yields. Meanwhile, the challenge from automotive for our industry is an application project. The need is to improve our yields by orders of mag- nitude to deliver the kind of reliability that automotive requires for electric vehicles and autonomous cars. I'm getting the feeling from you that there is that same sort of interest and momentum—a "let's get this done" attitude— in automotive. Bergman: In some of our existing standards, we've had active participation from the aero- space industry. The automotive industry is look- ing for the same amount of reliability, but with less cost. They want to reduce the testing and still maintain the quality requirements and expecta- tions. And for the assembly side, automotive wants to allow for components that may not be ready for acceptance by the aerospace industry. There are some process validation requirements in the assembly addendum to address that. In some of our existing standards, we've had active participation from the aerospace industry.