Investigation of Various Factors Affecting Noise Emission from Roller Element Bearings
Open Access
- Author:
- Elston, Jacob
- Graduate Program:
- Mechanical Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- January 04, 2021
- Committee Members:
- Issam Abu-Mahfouz, Thesis Advisor/Co-Advisor
Amit Banerjee, Committee Member
Abdallah Ramini, Committee Member
Issam Abu-Mahfouz, Committee Member
Rick Ciocci, Program Head/Chair - Keywords:
- Bearing
Vibration
Frequency
Time Domain
Frequency Domain
Wavelet - Abstract:
- ABSTRACT A vast amount of research has gone into interpreting vibrational signals emitted from roller element bearings of various designs. Most of this research is dedicated to detecting bearing faults in damaged bearing or predicting the end of useful life for a worn bearing. There is little research dedicated to the exploration of the vibration signals emitted from new bearings operating under various conditions. This paper aims to document the result of scientific experiments, using the scientific method to the greatest extent possible, where bearing parameters are varied and compared. Although many types of roller bearing designs exist, all bearings tested in this paper were spherical roller element type bearings. All tested bearing were composed of only the inner race, outer race, balls, and cage; no bearings were tested with integral mounting block, retaining devices, etc. All bearings had any seals and manufacturer provided lubricant removed. Standard ball bearings, deep groove ball bearings, and ‘low quality’ ball bearings were tested at varied loads, with varied speeds, and different viscosity lubricants. Statistical data (kurtosis, skewness, crest factor, and standard deviation) was collected and compared, along with frequency domain data (FFT, and Burg), and wavelets. The same data was also collected and analyzed for a load cell placed between the test bearing and the hydraulic cylinder used to generate the radial load. The raw data was processed by Matlab and exported to Excel where a multitude of figures were generated that allowed for the easy comparison of data. Conclusions were drawn based on the comparisons made. The goal is to provide additional information to engineers and equipment designers, for selecting the most appropriate bearing, when it is desired to minimize vibrations emissions from operating equipment.