As populations of cities increase, and rail traffic increases accordingly, noise pollution from trains will become more of an issue for those residing near the track and those travelling on it. Squeal is just a part of the noise pollution that trains emit, however the level of knowledge of its causes needs to be increased if it is to be better understood. Investigation into what causes squeal and how it can be mitigated or prevented is important for the development of rail infrastructure, especially in urban environments.
The rail test rig (referred to as R2DT, meaning the Rolling Contact Two Disk Test Rig) at The University of Queensland (UQ) has been used for squeal investigation and will be an important tool in developing knowledge of the phenomenon. As for any investigation, accurate data is paramount for good results. This thesis details the development, testing and use of an upgraded data acquisition system, including software (written with LabVIEW) and hardware upgrades.
The software (called R2DT Measurement Centre) was developed specifically for the test rig but is versatile in that other analogue sensors can be used with it. The software facilitates easy acquisition, monitoring and storage (in raw and/or processed forms) of analogue sensor data. It also was designed to acquire shaft speed and wheel slip data, and this worked on a test bench, but due to electrical noise in the outputs of the rotary encoders (connected to the test rig shafts to acquire speed data) it did not work on the test rig. For the same reason, another section of the program, that was written to acquire the profile of the lower wheel relative to its angle, did not work. A speed control program, written by a previous UQ student was successfully incorporated into the program.
Benchmarking tests performed on the processed data of the test rig confirmed that the program will give accurate values for the required parameters being measured, provided a suitable calibration is performed. A brief case study showed how the program can be used for squeal investigation. A statistical error analysis has been performed.