Issue link: https://iconnect007.uberflip.com/i/1529411
92 PCB007 MAGAZINE I NOVEMBER 2024 characterization of manufacturing capabili- ties. By using the CAT coupons, mean stan- dard deviations and process capability (Cp and Cpk) can be statistically determined. e per- formance provides data for acceptability and requirements specifications. Both need to be related to the design rule nominals and toler- ances proposed in IPC-2226A. is scenario then provides the settings for CAM's manufac- turability review and design rule checking. Tying design rules to fabrication perfor- mance can help us better understand costs. Tight design rules are no good if no one can build them. For those who can, the designer needs to be pre-warned that yields could be quite low and costs correspondingly higher. is might be tolerable for the designer, pro- vided he is willing to pay. Summary e need for the PCQRR program is great. I had the chance to see it in action at the Taiwan High-Tech Forum at IPC APEX EXPO 2024. e next step is to tie our HDI and UHDI man- ufacturing processes to the new IPC HDI design standards and then to IPC UHDI design standards. Contact IPC if you want to partici- pate or if you support such a program. PCB007 References 1. IPC-2315 and IPC-2226A, ipc.org. 2. "Understanding Process Capability, Quality, and Reliability," by Tim Estes, Ronald Rhodes, and David Wolfe, HDI Handbook, Conductor Analysis Technologies, Inc., 2010. Happy Holden has worked in printed circuit technology since 1970 with Hewlett-Pack- ard, NanYa Westwood, Merix, Foxconn, and Gentex. He is currently a contributing tech- nical editor with I-Connect007, and the author of Automation and Advanced Procedures in PCB Fabrication, and 24 Essential Skills for Engineers. To read past col- umns, click here. In nature, flying animals sense coming changes in their surroundings, including the onset of sud- den turbulence, and quickly adjust to stay safe. En- gineers who design aircraft would like to give their vehicles the same ability to predict incoming distur- bances and respond appropriately. Indeed, disas- ters such as the fatal Singapore Airlines flight this past May in which more than 100 passengers were injured after the plane encountered severe turbu- lence, could be avoided if aircraft had such auto- matic sensing and prediction capabilities combined with mechanisms to stabilize the vehicle. Now a team of researchers from Caltech's Cen- ter for Autonomous Systems and Technolo- gies (CAST) and NVIDIA has taken an important step toward such capabilities. In a new paper in the journal npj Robotics, the team describes a control strategy they have developed for unmanned aer- ial vehicles (UAVs) called FALCON (Fourier Adap- tive Learning and CONtrol). The strategy uses rein- forcement learning, a form of artificial intelligence, to adaptively learn how turbulent wind can change over time and then uses that knowledge to control a UAV based on what it is experiencing in real time. (Source: Caltech) AI-Trained Vehicles Can Adjust to Extreme Turbulence on the Fly