Industrial machinery wears down or needs maintenance on a regular but unpredictable schedule. While the traditional approach of responding to failures after-the-fact leads to lost productivity, the advent of wireless sensor networks for portable devices offers a superior approach: Real-time monitoring of the acoustic or vibration characteristics of a machine offers an early warning of trouble or predicts trouble that is about to take place.
The challenges of implementing a real-time vibration monitoring system, however, are many. Designers must decide how to handle the large quantities of data generated by multiple acoustic and vibration sensors distributed throughout a structure, and they must find a solution that fits in a small, low-power footprint.
More Intelligence at the Sensor Edge
Aspinity’s RAMP chip uses its analyze-first edge architecture to let system designers decrease the volume of sensor data that is digitized, analyzed and transmitted, reducing power and data requirements and improving the utility of the collected data.
As an example, instead of a predictive maintenance system that continuously digitizes and generates 1000s of points of data to monitor trends in the magnitudes of certain spectral peaks, Aspinity’s new analog processor-platform can extract the same information from raw analog sensor data, and reduce the data down to just the frequency/energy pairs – up to 100x reduction in data at the sensor.
Reducing the amount of data that is handled through the system is the key. It lowers the power and data requirements for the sensor system and reduces the burden on downstream processors. This efficiency speeds development of more easily deployable, battery-operated, smaller, lower-cost wireless sensor system for preventative and predictive maintenance systems.
For applications where it is not necessary to track and follow all sensor data, the RAMP chip can be programmed to detect and transmit data only for specific acoustic or vibration frequencies that indicate trouble. In this case, only the sensor and the RAMP chip are always on to monitor the environment, while the downstream higher-power chips in the system remain off until the chip detects an event that triggers an alert. At that point, the RAMP chip signals to the other chips that it is time to activate, the compressed sensor data is transmitted, and the alert is received so that further actions can be taken. This ultra-low-power continuous-monitoring solution that transmits only relevant data for analysis is ideal for remote, battery-operated systems where small sensor modules and long lifetimes are required.