Many manufacturers use manual test equipment that consist of standalone COTS (commercial-off-the-shelf) test instruments, such as digital multimeters, oscilloscopes, and hipot testers. COTS test instruments often require configuration and programming using their front-panels. The required sequences of button presses to program these instruments through multiple selection menus can be cumbersome, prone to errors, and subject the test process to test inconsistency.
Quality control processes are driving manufacturers to automate test equipment to ensure consistent configuration control and data collection.??PC-based ATE (automatic test equipment) systems often provide control of the testing process, manage instrument configurations, increase test throughput, and address the issues associated with manual test. Although PC-based ATE is widely accepted, many manufacturers still test manually. PC-based ATE systems deploy a PC with each test system, which requires a minimum level of operator skill level and IT maintenance.
For bench-top systems, the keyboard, monitor, and mouse occupy precious space on the production floor. Operator skill level and language barriers may necessitate significant training for some manufacturers. In particular, setup and configuration during product changeover can be challenging. Also, operators using PC-based ATE systems can be distracted by the accessibility of web browsing and other PC software. Responsibility for IT administration of PCs on the production floor is a gray area at some organizations. These issues can lead to operator error, inefficiency, and inconsistency in the testing process.
Engineers at Bloomy Controls developed a headless ATE system that is simple to operate and doesn't require PCs on the production floor. The system is headless because it doesn't need a keyboard, monitor, touch screen, or mouse. Instead, it contains a physical interface comprised of a few mechanical buttons, LEDs, and optional barcode scanner.
We have been through a few iterations for PWA ATE. Eg big external multiplexors, wire wrap, pcb. The method we use now is a clamshell case, with custom PCB that has the test pins, multiplexing, power and any specialist interface on the main one PCB, then plug in ports for data aquisition modules or RF connectors for SigGens/DSO etc. We build in a USB hub so everything connects to the PC via a single USB cable. We code it in c# (and have learnt to stay away from LabView) and connect it to our manufacturing database for results. even still, ATE is almost as hard as the product development itself.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.