Face Rub-Impact Monitoring of a Dry Gas Seal Using Acoustic Emission

Abstract

Face rub-impact of a dry gas seal was investigated using acoustic emission (AE) technology. A force applicator was designed and mounted on the seal test rig to provide controlled misalignment and thus induce rub-impact of the seal faces. Two types of AE sensor were mounted at different positions on the seal. For the PICO AE sensor mounted directly on the seal ring, the root mean square (RMS) of the original AE signal was sensitive to face rub-impact of the seal. When force was applied, the AE RMS of the PICO sensor gave a distinct periodic waveform with a period consistent with the rotational period above an initial noise signal. The magnitude and shape of the waveform changed as the applied force increased. For the R15α AE sensor mounted on the housing of the seal, no obvious changes could be found from the RMS of the original AE signal synchronously obtained during loading. Two kinds of signal processing methods were tried to eliminate noise. After band-pass filtering, the RMS of the AE signals of the R15α sensor indicated face rub-impact when the misalignment was relatively large. The empirical mode decomposition method using masking signals was found to be more effective than band-pass filtering in eliminating the noise but took much more computational time. The results indicate that the AE technology is a potentially effective tool in monitoring and investigating face rub-impact of dry gas seals.