建设你自己IoT:新系统和遗留系统设备监测

全文内容 Full Text

An exploration of the PDN bucket brigade concept. READ MORE... The difference between process capability and design intent. READ MORE... Dynamic flex applications require careful consideration. READ MORE... https://pcdandf.com/pcdesign/index.php/editorial/menu-features/19405-the-charge-of-the-bucket-brigade https://pcdandf.com/pcdesign/index.php/editorial/menu-features/19407-why-don-t-uhdi-capability-charts-tell-the-whole-story https://pcdandf.com/pcdesign/index.php/editorial/menu-features/19406-flex-dfm-guidelines-part-1 https://pcdandf.com/pcdesign/index.php/current-issue/241-designer-s-notebook/19401-rf-pcb-layout-transmission-lines-impedance-and-pdn-design https://pcdandf.com/pcdesign/index.php/current-issue/243-flexperts/19404-flexpert-bending-with-attached-structure https://pcdandf.com/pcdesign/index.php/current-issue/241-designer-s-notebook/19202-tolerances-and-dimensions-for-pcb-fabrication PCB fabricators can pair IoT monitoring with AI to track equipment behavior, troubleshoot setups and build useful operating records. Walk through a printed circuit fabrication shop and ask how people know the equipment is running normally. Some machines have useful displays and records. Others depend on an operator checking a gauge, writing down a reading, or hearing something that doesn't sound right. A lot of knowledge lives with the person standing next to the machine. That works until a problem develops between checks or the person who usually catches it isn't there. A drilling machine stops, a press needs attention and now production has to work around it. Later, someone asks when the problem started, and there may not be much of a record to check. If you've priced a monitoring system, you know the costs can include hardware, installation, and ongoing software fees. Older equipment can add another expense when it needs custom work to connect. A separate sensor and a small computer may give you a simpler place to start. AI can help your team identify what works best for your system and can help you work through the setup. This becomes a basic example of the Internet of Things, or IoT, when the monitoring device connects to a network so its readings can be collected or reviewed. For this article, we're talking about watching equipment your shop already owns. A PCB shop may have newer drilling equipment alongside a lamination press that has been in service for years. The press joins the layers of a multilayer board using heat and pressure. It may still do its job well without offering an easy way to collect and review its operating data. Replacing useful equipment just to get better information is a hard expense to defend. And even a newer machine may not make the reading you need easy to find or compare over time. The people operating and maintaining these machines can usually tell you where more information would help. One machine stops more often than the others. Another has started making an unfamiliar noise. Everyone knows about it, but nobody is sure when it changed or how often it happens. Ask them what they wish they knew sooner. They may be more concerned about repeated stops than a measurement you thought was important. Depending on the equipment, a separate sensor can collect a measurement without connecting to the machine's controller. That might be temperature, vibration, or electrical current. Vibration monitoring can require faster measurements and more careful sensor mounting than a simple temperature log. The right choice depends on the question you're trying to answer and what can be measured reliably. Figure 1. Low-cost sensors and small computers can give PCB fabricators greater visibility into equipment operation without replacing legacy machinery. A basic monitor collects a reading, saves it, and alerts someone when it needs attention. The sensor measures, the computer keeps the record, and a program handles the logging and alerts. For a PCB shop, the question might be whether a machine's