Christopher Scott Larsen

Applications of multi-spectral imaging: failsafe industrial flame detector

Industrial and petrochemical facilities present unique challenges for fire protection and safety. Typical scenarios include detection of an unintended fire in a scene, wherein the scene also includes a flare stack in the background. Maintaining a high level of process and plant safety is a critical concern. In this paper, we present a failsafe industrial flame detector which has significant performance benefits compared to current flame detectors. The design involves use of microbolometer in the MWIR and LWIR spectrum and a dual band filter. This novel flame detector can help industrial facilities to meet their plant safety and critical infrastructure protection requirements while ensuring operational and business readiness at project start-up.

Characterization of vibration transfer paths in nose gearboxes of an AH-64 Apache

Health monitoring of rotorcraft components, which is currently being performed by Health and Usage Monitoring Systems (HUMS) through analyzing vibration signatures of dynamic mechanical components, is very important for their safe and economic operation. Vibration diagnostic algorithms in HUMS analyze vibration signatures associated with faults and quantify them as condition indicators (CI) to predict component behavior. Vibration transfer paths (VTP) play important roles in CI response and are characterized by frequency response functions (FRF) derived from vibration signatures of dynamic mechanical components of a helicopter. With an objective to investigate the difference in VTP of a component in a helicopter and test stand, and to relate that to the CI response, VTP measurements were recorded from 0–50 kHz under similar conditions in the left and right nose gearboxes (NGBs) of an AH-64 Apache and an isolated left NGB in a test stand at NASA Glenn Research Center. The test fixture enabled the application of measured torques – common during an actual operation. Commercial and lab piezo shakers, and an impact hammer were used in both systems to collect the vibration response using two types of commercially available accelerometers under various test conditions. The FRFs of both systems were found to be consistent, and certain real-world installation and maintenance issues, such as sensor alignments, locations and installation torques, had minimal effect on the VTP. However, gear vibration transfer path dynamics appeared to be somewhat dependent on presence of oil, and the lightly-damped ring gear produced sharp and closer transfer path resonances.