AVS 63rd International Symposium & Exhibition
    Vacuum Technology Tuesday Sessions
       Session VT-TuP

Paper VT-TuP1
Smart Measurement and Diagnostics Module for Dry Vacuum Pumps

Tuesday, November 8, 2016, 6:30 pm, Room Hall D

Session: VT Poster Session (and Student Poster Competition)
Presenter: Wan-Sup Cheung, KRISS, Republic of Korea
Authors: W.S. Cheung, KRISS, Republic of Korea
K. Baik, KRISS, Republic of Korea
J.Y. Lim, KRISS, Republic of Korea
Correspondent: Click to Email
This paper addresses recent industrial demands for more reliable predictive maintenance and diagnostics for the failure protection of dry vacuum pumps operated in the semiconductor and flat display chemical processes. Korean leading companies are very expecting to improve the predictive maintenance and self-diagnostics capability sufficient to meet such higher demands for dry vacuum pumps. This project has started to satisfy Korean industrial demands. On the onset of this work, the first technical issue was to examine what kinds of state variables are measured from dry vacuum pumps. Most of them were found to be the static properties such as body and exhaust temperatures, N2-flow rate, motor supply currents of booster and dry pumps, exhaust pressure, etc. Most vacuum pump manufactures have reported that most of vacuum pump failures come from their rotating machinery parts such as rotors, bearings and/or gears. It became apparent that reliable self-diagnostics of vacuum pumps cannot be realized without vibration measurement of those rotating parts. A vacuum pump vibration measurement (VPVM) module has been developed to enable the measurement of three vibration harmonics of rotors, bearings and gears. Tested results of the VPVM module are shown to provide harmonic vibration measurements of rotors, bearings and gears. In addition to the vibration measurement capability of the VPVM module, another technical challenge was to collect the measurements of the state variables available from each vacuum pump via the digital communication interface. This work has attempted to integrate both the traditional state variables and the three harmonic vibration levels into an extended set of state variables required to realize more reliable predictive maintenance and diagnostics of dry vacuum pumps. To obtain the extended state variables, the VPVM module was developed to support three kinds of serial communication ports (RS232C/RS485, CAN and/or SPI) to an individual vacuum pump controller. Furthermore, the VPVM module was designed to provide a 128 MB backup flash ROM to record a time series of the extended state variables in real-time. These records of the extended state variables are used to implement the self-diagnostics and predictive maintenance algorithms of dry vacuum pumps developed and patented by KRISS.