Detecting foreign objects in process streams.

Looks like ice cream. Smells like ice cream. But does it sound like ice cream? A new ultrasonic technology could tell ice cream manufacturers if a foreign object has fallen into their product before a customer finds it at the bottom of an ice cream cone.

Researchers at the Department of Energy's Pacific Northwest National Laboratory (PNNL) have developed a patented inspection method that uses sound waves to detect foreign objects in processing streams. The method can detect cartilage, bone fragments, metal, and plastic--essentially anything that shouldn't be in the product. Battelle, which operates PNNL, funded the research, and the method is available for licensing.

The PNNL-developed method combines ultrasonic and optical capabilities. According to Aaron Diaz, PNNL staff scientist, this method is "the only one we're aware of that uses both acoustics and optics." Diaz added: "Because the method can be automated, it's inherently safer and more effective than inspecting certain types of process streams manually. Automation eliminates the need for costly and slower inspections using methods such as X-ray, which typically requires added safety precautions and complex operator training."

For example, "first-step" baby foods, which usually consist of pureed fruits and vegetables, could be examined with this method.

"Our process involves multiple sensors configured along the process stream," Diaz said. "We flood the stream with sound pulses of acoustic energy, while at the same time transmitting optical energy. We're looking for reflected echoes that we can correlate with location in the stream relative to the product's flow rate. If the received acoustic response is abnormal, it indicates there's something in the stream that shouldn't be there, and we know, generally, where the foreign object is."