Vibration-based sensor design to detect lubrication levels contained within differential gear housings

Open Access
Author:
Wells, Stephen Matthew
Graduate Program:
Acoustics
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
March 26, 2013
Committee Members:
  • Karl Martin Reichard, Thesis Advisor
Keywords:
  • Acoustics
  • Vibration-based sensor design
  • Condition Based Maintenance Plus
  • CBM+
  • Differential Gear Housings
  • Lubrication level detection
  • Structural Acoustics
Abstract:
A root cause of failure of many gear and bearing system failures is loss of lubrication. Arguably, the lubrication system is the most critical part in any mechanical system because it serves many functions including friction reduction, wear reduction, cooling, anti-corrosion, cleaning action and sealing. A loss of lubrication leads to high friction and excessive wear in gears and bearings, which leads to component failure. The key is to find the most effective and efficient method for detecting lubrication loss in a closed mechanical lubrication system. Currently, the drive differentials for most vehicles do not possess a sensor to detect the lubrication level within the differential casing. Due to the lack of space and the dynamics of the gear system inside of the differentials, it is impractical to place a sensor inside of the housing. However, installing an external sensor on the differentials is reasonable and can be an after-market addition to most vehicles. Tests were performed on two different military vehicle differentials to determine if the lubrication level could be detected using a vibration sensor on the outside of the main drive differential. These tests consisted of monitoring the vibration levels generated by the differential at different speeds and lubrication levels. It was confirmed that, at different speeds and lubrication levels, the vibration levels in certain frequency ranges fluctuated depending on those two elements previously stated. By monitoring these vibrations, the lubrication level contained within the differential can successfully be determined with a practical amount of certainty. Future studies will focus on creating a sensor that can be installed on any differential, detect the lubrication level and eventually send warning messages if a leak has formed. Imminent complete lubrication loss can then be immediately corrected, preventing a catastrophic failure.