This research is focused on the condition monitoring of electrical machines and its long term purpose is to monitor electrical and mechanical faults at the same time, in non-stationary conditions (variable load and speed), with a single piece of hardware. The Instantaneous Angular Speed (IAS) measurement of an electrical machine is proposed and analysed in order to detect the fault development inside it.
Chapter 1 introduces some basic principles about the maintenance of an electrical machine. Machine unscheduled downtimes are frequently caused by bearing faults, and rotor/stator faults. Monitoring systems are needed when the machine is very important for the plant (cost, safety). In this chapter, the electrical machine’s behaviour is also examined. Induction electrical machines have been chosen for this research. A review of the excitation frequencies is reported in the chapter. In the last section, characteristic fault frequencies (from mechanical and electrical sources) are collected.
Chapter 2 presents the IAS measurement and its signal processing. The IAS is the measurement of the shaft rotating speed in order to visualize what’s happening during a single or in multiple turns. There are many measurement methods which are based either analogical to digital conversion or which use counters. Analogical to digital methods use a standard data acquisition board. Counter methods have to use specific hardware that is more expensive, but with less data to store. In this research, the counter method is used, combined with the Elapsed Time (ET) counting technique.
Chapter 3 describes the encoder system. Its output signal is acquired with an oscilloscope and with the counter board. The signal’s differences are highlighted. In this chapter, the measurement’s source of errors are listed: the encoder’s geometrical error, the counter’s quantization error, the clock stability and the general electrical noise.
Chapter 4 collects all the experimental tests done during the PhD research. Three experimental test rigs are shown and two measurements at Nidec ASI S.p.A. are reported. Note that the experimental test rigs were designed and built at the Università degli Studi di Trieste during the three years of the PhD.
Experimental Test Rig 1 (ETR1) is used to understand the electrical motor’s behaviour with varying speed, the difference between the IAS and the speed acquired with the Torsional Laser Vibrometer, the difference between the IAS and the acceleration signal measured with an accelerometer located on the motor’s stator, the effect of the unbalance in the IAS measurement.
Experimental Test Rig 2 (ETR2) allows to examine the load effect on the IAS measurement, the magneto-motive force harmonics, the slip and the rotor effects.
Experimental Test Rig 3 (ETR3) is designed in order to detect the Inner Race Bearing Fault (Ball Pass Frequency Inner - BPFI) with varying load. The acceleration, the voltage and the current are compared with the Instantaneous Angular Speed. The motor is also tested with an unbalanced power supply.
The two measurements at Nidec ASI S.p.A show how the IAS measurement could be implemented in an industrial machine larger than the one tested in the laboratory.
This research presents the pros and cons of the IAS measurement, highlighting the capability of detecting BPFI bearing fault, feeling the load variations owing to the brake system (a synchronous generator), measuring the Fundamental Train Frequency of an healthy bearing, detecting unbalance in the rotor and other special features.
The author would like to thank the Fondo Sociale Europeo, the Regione Friuli Venezia Giulia and Nidec ASI S.p.A (an electrical motor company) for the sponsorship and the collaboration during the three PhD years covered by the SHARM project ”Manutenzione Preventiva Integrata”.
