A microprocessor-controlled system for real time signature analysis

A system of hardware and software has been developed which will aid in the monitoring of the operating conditions of large axial flow compressors. The system monitors the machinery health by a technique known as signature analysis.

The application of signature analysis techniques to machinery health monitoring is not new. Most applications, however, are off-line analyses of machinery operating under fixed conditions. The system described in this study monitors the machine on-line, and monitors machines that operate under a wide range of conditions.

The system uses a frequency spectrum analyzer to produce digital spectrum data and a microprocessor to perform the signature analysis. The software is written in Intel 8080 assembly language.

Propulsion Wind Tunnel Compressor, C-1, at the Arnold Engineering Development Center, is used for all experimental work and data collection. The new system provides a significant improvement in monitoring two problem areas for C-1: rotating stall detection and false signal detection. Although the data presented deal only with these two problem areas, the system that is developed is versatile enough to handle a variety of monitoring tasks.