Don't overspecify machine-vision cameras
Avoid adding "fudge factors" to your requirements.
Steve Scheiber, Contributing Technical Editor -- Test & Measurement World, 4/22/2008 12:07:00 PM
Engineers learn very early to prepare specifications carefully but to always include a “fudge factor” to ensure that the result will hit its performance targets. Unfortunately, squeezing that additional increment into a design can incur high costs in money, design time, and manufacturability. Meeting the specifications may prove almost impossible.
When you are selecting cameras for electronics inspection, Bruce Butkus, product line engineer for Edmund Optics, advocates concentrating on what you actually need and not trying to guarantee that no situation will ever exceed what the system can do.
“Machine-vision vendors need to encourage customers to stick to realistic specs,” commented Butkus. “A customer might need a camera and lens that can measure to a thousandth of an inch, but they demand a tenth of a thousandth ‘just to be sure’. Some of the specs have passed through numerous levels of approval—often increasing the requirements along the way—despite the fact that there is no logical reason why the application needs that level of performance. “In addition, you cannot push system limits without considering the impact on lens performance. As cost-effective camera resolutions have increased from 0.3 to 5 megapixels over the past seven years, we can no longer treat lenses simply as accessories. Achieving a higher resolution than necessary, for example, can dramatically increase costs and can also limit the selection of available lenses. Specifying a higher resolution will increase system performance, but you cannot achieve the full benefit if the camera and lens are not optimally matched. “Even if we meet one spec, side issues can negate any advantage. A specification may look very impressive on paper, but a metal shop may have difficulty fabricating the necessary component with sufficient precision. “In addition to rapidly increasing costs, when you push the capability of the technology, you stress other parts of the system as well. Higher resolutions and higher speeds can lead to increasing frame rates, decreased exposure time, and the need for better and more advanced lighting. If you can’t illuminate your target properly, you sacrifice accuracy and repeatability. You also have to increase the diameter of the lens aperture. "But manufacturing a lens with a larger aperture can require increasing tolerances, or image quality may suffer. The chips in the camera may capture the image quickly enough at sufficient resolution, but the communications bus has to be fast enough to off-load the data and the computer core has to be powerful enough to analyze it. Ultra-high-speed cameras may have to buffer the data before transfer to the computer. Processing may require several computers that divide up the data and reassemble it later. "All of these techniques compromise the speed of the inspection. Adding processing capability to increase the speed makes the entire system both more complex and more expensive. And no matter how fast you design the system, you will eventually encounter situations that exceed it. “So before you decide to specify the ultimate inspection system, consider whether your application really needs those levels of performance, and whether achieving them is worth the price.”Talkback
Related Content
Sponsored Links
TMW Resource Center
Browse Resources by Type:
- PCI Express Advances Machine Vision
DALSA | White Paper
VIEW NOW
DALSA | White Paper
VIEW NOW
DALSA | White Paper
VIEW NOW
View More
