A diagnostic signal-analysis technique, called the Phase Synchronized Enhancement Method (PSEM), has been developed. This method allows NASA engineers to retrieve any unique defect signatures and trends associated with different failure modes and unusual phenomena experienced during Space Shuttle main engine (SSME) testing.

Previous diagnostic studies using the Generalized Hyper-Coherence method have shown that the frequency of the shaft rotational component (sync) of a high-speed SSME turbopump is fluctuating around a center frequency during steady-state operation. These studies further showed the Power Spectral Density (PSD) to be exhibiting a discrete peak, indicative of strong stationarity.

Machinery failure detection has always been a significant technical challenge for NASA's propulsion technology engineers. A reliable engine-health-monitoring system can prevent catastrophic failures and lower costly downtime due to false alarms. PSEM can provide valuable signal-enhancement capability during engine-health monitoring and diagnostics and improve the safety and reliability of NASA's advanced propulsion systems.

This illustration shows the Instantaneous Frequency of sync (N), inner ball pass (IBP), and the 8th harmonic of sync (8N). PSEM algorithm forces the frequency of sync from this periodic variation into a constant frequency, generating a highly desirable effect on the entire signal.

This method of analysis uses the microfrequency variation phenomenon to improve all the sync-related components in a signal. PSEM forces the narrow-band spectral component of sync into a pure-tone discrete component with a constant frequency by transforming its instantaneous phase signal into an equivalent "realignment" time. When the realignment time is corrected, the original sync component will become discrete, generating a highly desirable effect on the entire signal where all the other sync-related components (sync harmonics, cage, ball spin, outer ball pass, and inner ball pass) will automatically become discrete. The resulting discrete signal provides better PSD resolution, which improves engine diagnostic evaluation.

A vibration signal in a rotor system is modeled as an FM signal with multiple spectral components at different carrier frequencies. The instantaneous frequency and instantaneous phase information of each component is recovered using several digital frequency-demodulation methods, such as complex demodulation techniques and the phase-lock- loop method. During steady-state operation of most machinery systems, the instantaneous frequency of sync tends to fluctuate about a center frequency and is not a constant frequency. Since the amplitude of such frequency variation could be much smaller than the bandwidth with PSD analysis, the sync spectral component still appears as a very discrete peak in its PSD.

This technique has potential for commercial application outside NASA's propulsion area. For example, PSEM will greatly increase the PSD resolution during quality-control checks on spindle motors during production and reduce the man-hours needed for system monitoring and diagnostics.

This work was done by Jen-Yi Jong of AI Signal Research, Inc., for the Marshall Space Flight Center. For further information, access the Technical Support Package (TSP) free on-line at www.techbriefs.com under the Electronic Components and Circuits category, or circle no. 109on the TSP Order card in this issue to receive a copy by mail ($5 charge).MFS-26435