• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.17 no.10
• /
• pp.967-975
• /
• 2007

A rotordynamic analysis was performed with a dry vacuum pump, which is a major equipment in modern semiconductor and LCD manufacturing processes. The system is composed of screw rotors, lobes picking air, helical gears, driving motor, and support rolling element bearings of rotors and motor. The driving motor-screw rotor system has a rated speed of 6,300 rpm, and was modeled utilizing a rotordynamic FE method for analysis, which was verified through 3-D FE analysis and experimental modal analysis. As loadings on the bearings due to the gear action were significant in the system considered, each resultant bearing load was calculated by considering the generalized forces of the gear action as well as the rotor itself. Each resultant bearing loading was used in calculating each stiffness of rolling element bearings. Design goals are to achieve wide separation margins of lateral and torsional critical speeds, and favorable unbalance responses of the rotor in the operating range. Then, a complex rotordynamic analysis of the system was carried out to evaluate its forward synchronous critical speeds, whirl natural frequencies and mode shapes, unbalance responses under various unbalance locations, and torsional interference diagram. Results show that the entire system is well designed in the operating range. In addition, the procedure of rotordynamic analysis for dry vacuum pump rotor-bearing system was proposed and established.

• Tribology and Lubricants
• /
• v.38 no.2
• /
• pp.33-40
• /
• 2022

In this study, we present rotor dynamic analysis and operation test of a turbo expander for a hydrogen liquefaction plant. The turbo expander consists of a turbine and compressor wheel connected to a shaft supported by two hydrostatic radial and thrust bearings. In rotor dynamic analysis, the shaft is modeled as a rigid body, and the equations of motion for the shaft are solved using the unsteady Reynolds equation. Additionally, the operating test of the turbo expander has been performed in the test rig. Pressurized helium is supplied to the bearings at 8.5 bar. Furthermore, we monitor the shaft vibration and flow rate of the helium supplied to the bearings. The rotor dynamic analysis result shows that there are two critical speeds related with the rigid body mode under 40,000 rpm. At the first critical speed of 36,000 rpm, the vibration at the compressor side is maximum, whereas that of the turbine is maximum at the second critical speed of 40,000 rpm. The predicted maximum shaft vibration is 3 ㎛, whereas sub-synchronous vibration is not presented. The operation test results show that there are two critical speeds under the rated speed, and the measured vibration value agrees well with predicted value. The measured flow rate of the helium supplied to the bearing is 2.0 g/s, which also agrees well with the predicted data.

• Journal of Domestic Journal Test
• /
• v.11 no.2
• /
• pp.221-227
• /
• 2023

− A rotordynamic system consists of components that undergo rotational motion. These components include shafts, impellers, thrust collars, and components that support rotation, such as bearings and seals. The motion of this type of rotating system can be modeled as two-dimensional motion and, accordingly, the equa- tion of motion for the rotordynamic system can be represented using complex coordinates. The directional fre- quency response function (dFRF) can be derived from this complex coordinate system and used as an effective analytical tool for rotating machinery. However, the dFRF is not widely used in the field because most pre- vious studies and commercial software are based on real coordinate systems. The objective of the current study is to introduce the dFRF and show that it can be an effective tool in rotordynamic analysis. In this study, the normal frequency response function (nFRF) and dFRF are compared under rotordynamic analysis for isotropic and unisotropic rotors. Results show that in the nFRF, the magnitude of the response is the same for both pos- itive and negative frequencies, and the response is similar under all modes. Consequently, the severity of the mode cannot be identified. However, in the dFRF, the forward and backward modes are clearly distinguishable in the frequency domain of the isotropic rotor, and the severity of the mode can be identified for the uniso- tropic rotor.