Based on ultra narrowband spectrum

Application of traditional spectral analysis does not produce tangible results for multiple mechanisms that rotate with approximately same speed.For instance, there are several induction motors – pumps, motor means, airmovers with line supply of 50 Hz and rotating with speed of 3,000 r/min. Signal power spectrum of vibration detector in one frequency component of 50 Hz with width of 0.12 Hz has vibrations of all wheels and signal of leakage, i.e. these signals are not separated. To separate spectral component of different sources you should apply enlarging lens mode of spectral analysis. Figure 1 shows diagram of signal power spectrum in constant mesh of frequencies with 0.01 Hz resolution. Frequency components are separated during such spectral analysis and this allows performing vibration measurement of each mechanism separately. The peculiarity of this method is specification of observation time up to 100 seconds and stability of frequencies of vibration sources.

Building spectrum sets as spectral recording enables us to get additional information on instability of frequency and amplitude of the signal. Spectral recording is made as consistent estimated and reflected power spectrums. Frequency rate is indicated in horizontal direction, observation time is indicated in vertical direction. The level of spectral components is indicated by the color. Low levels are highlighted with light colors; high levels are highlighted with dark colors. Figure 2 shows spectral recording of the signal of vibration detector, placed on the support foot of the mechanism with rotating speed about 5,500 r/min. Observation time is 300 seconds. Frequency instability at the interval of observation is 0.5%. Instability depends on rotating moment, load and other factors.

Rotating instability parameter also can be applied in case of gear units control.

See also:

• diagnostics of gear units;
• diagnostics of gear units based on spectrum of vibration signal;
• diagnostics of gear based on inter-spectral characteristics;
• diagnostics of gear units based on inter-correlation characteristics;
• diagnostics of gear units based on spectrum of envelop curve;
• diagnostics of gear units based on order normalization.