Diagnostics of a condition of electric motors
on the basis of the current spectrum

Petoukhov V.S., Ph.D., IEEE member

It is developed software - a program complex for diagnostics of a condition and operating conditions of an electric and mechanical part of electric motors, and the mechanical devices connected with them on the basis of the spectral analysis of a currents.

Now alternating current motors are the largest consumers of electric energy. According to last researches, they consume over 80% of the produced electric power. However while in service, there can be damages of elements of the motor, that in turn leads to its premature exit out of operation.

Following statistical data /1-2/, obtained by numerous researches, the character of damages of alternating current motors:

• Damages of a stator - 38%
• Damages of a rotor - 10%
• Damages of elements of bearings - 40%
• Other damages - 12%

Recently methods of diagnostics of a condition of the electric machines, based on performance of monitoring of a currents with the subsequent performance of the special spectral analysis of the received signal /3-8/, that allows to determine a condition of various elements of the motor with a high degree of reliability. The important advantage of such approach is that carrying out of monitoring of a current of the electric motor can be executed as directly on it, and in an power supply panel.

Substantive conditions of the developed method of diagnostics of electric motors

The hardware-software complex is developed for a condition and operating conditions of an electric and mechanical part of electric motors and the mechanical devices connected with them on the basis of the spectral analysis of a currents of the electric motor.The block - the scheme of a complex is presented on fig. 1.

Fig. 1 Block - scheme of a diagnostic complex

The structure of a complex include demountable current sensors with the linear frequency characteristic, the conditioner of a signal (the filter of low frequencies interfering occurrence of false frequencies of signals (aliasing) at their digitization/9/, the analog-digital converter (ADC), a personal computer with the necessary software for gathering and processing of the information.

Record of signals of a currents is carried out during time necessary for performance of the spectral analysis with the sanction on frequency not less 0.01-0.2 Hz.

Digitized ADC data are transmitted into the personal computer where processing of the received data is carried out, The rotation frequency of the motor and number of rotor bars of its is determined, then the special spectral analysis of a currents is carried out.

The physical principle put in a basis of work of a diagnostic complex, consists that any perturbations in work of an electric and/or mechanical part of the electric motor and the device connected with it lead to changes of a magnetic field in a electric machine and, hence, to weak modulation of a currents consumed by the electric motor.

Thus, presence in a spectrum of a currents of the motor characteristic frequencies of the certain amplitude means presence of damages of an electric and/or mechanical part of the electric motor and the mechanical device connected with it /1/.

As an example below results of the spectral analysis of currents of two same fans installations are resulted: 1 week being operation (fig. 2) and the worked 5 years (fig. 3).

Fig. 2. Spectral of a current of the new electric motor

Fig. 3. Spectral of a current of the electric motor which is being operation within 5 years

In the above-stated figures it is clearly visible as the increase in amount of the frequency corresponding to various kinds of damages, and growth of their amplitudesfor the electric motor, is long being in work.

Opportunities of the developed method of diagnostics from the point of view of detection of various kinds of damages

Damages of a rotor of the motor.

This kind of malfunction is found out on presence of 2 lines of peaks symmetric to line frequency in a spectrum of a current (see fig. 4).

Fig 4. Frequencies, typical for damages of a rotor

Misalignment of shaft of the motor and mechanical load. This kind of malfunction is determined on frequencies, multiple to frequency of rotation of a rotor (see fig. 5)

Fig 5. Typical frequencies at presence of misalignment of shaft of the motor and mechanical load

Defects of a belts of the fan. This kind of malfunction is determined on frequencies, multiple to frequency betating of a belt, the determined length of the last and diameters of pulleys (see fig. 6)

Fig 6. Typical frequencies at presence of defects of a belts

Presence of such defects, are determined also:
• Short circuits of stator windings;
• Rotor rub;
• Damages of bearings (data about bearings of the electric motor and the mechanical device are necessary);
• Increased eccentricity of a rotor (static and/or dynamic);
• Defects of a mechanical part of the devices connected with the electric motor.

Definition of a degree of damages is made on gradation "Damages are absent" - "Damages are found" - "Critical damages" by comparison the amplitude of a signal on characteristic frequencies of damage with amplitude of a signal on line frequency.

At carrying out of repeated measurements on the given equipment the database is formed, allowing to trace changes of development of damages to time that enables to plan removing the equipment in repair.

Besides the above described measurements, carrying out of monitoring of the supply voltage allows to determine its asymmetry, presence of the harmonics and impulses of overvoltages (that is probable at work with frequency regulators of speed rotation) - i.e. those factors which directly influence for the term of service and profitability of the work of the motor.

Well-known, that first two of above-mensioned factors result as in an overheat of stator windings, and to damage of bearings due to occurrence of the high-frequency rotating moments of negative sequence.

Measurements can be executed as directly on electric motor (without any infringement of a mode of its work), and in an power supply panel.
Performance of this work will allow to lead full-scale natural diagnostics and the analysis of operating conditions of park of electric motors and the mechanical devices connected with them, it is essential to reduce as the expenses connected with "unexpected" refusals of the equipment, and to lower unproductive expenses of theelectric power.

The literature

1. W. T. Thomson: " A Review of On-Line Condition Monitoring Techniques for Three-Phase Squirrel-Cage Induction Motors-Past Present and Future " Keynote address at IEEE Symposium on Diagnostics for Electrical Machines, Power Electronics and Drives, Gijon, Spain, Sept. 1999 pp 3-18.
2. EPRI: " Improved Motors for Utility Applications and Improved Motors for Utility Applications, Industry Assessment Study ", Vol 1, EPRI EL-2678, Vol 1 1763-1, final report and EPRI EL-2678, Vol 2,1763-1 final report October 1982
3. V Thorsen and M Dalva: " Condition Monitoring Methods, Failure Identification and Analysis for High Voltage Motors in Petrochemical Industry ", Proc 8a 1EE Int Conf, EMD ' 97, University of Cambridge,No 444, pp 109-113
4. W T Thomson and D Rankin; " Case Histories of Rotor Winding Fault Diagnosis in Induction Motors ", 21 " 1 Int Conf Proc on Condition Monitoring, University College Swansea, March 1987
5. G B Kliman and J Stein: " Induction Motor Fault Detection Via Passive Current Monitoring ", Proc Int Conf (ICEM ' 90), MIT,Boston, USA, 1990, pp 13-17
6. W T Thomson, S J Chalmers and D Rankin: " On-line Current Monitoring and Fault Diagnosis in High Voltage Induction Motors - Case Histories and Cost Savings in Offshore Installations ", Offshore Europe ' 87, Conf Proc SPE September 1987, Aberdeen, SPE 16577/1 - SPE 16577/10
7. Randy R. Schoen, Thomas G. Habetler, Farrukh Kamran, Robert G. Barthel п-╟ Motor Bearing Damage Detection Using Stator Current Monitoring п-? IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL.31, NO.6, November/December 1995
8. William T.Thomson, Mark Fenger п-╟ Current Signature Analysis to Detect Induction Motor Faults п-? IEEE Industry Application Magazine July/August 2001.
9. Sergienko A.V. Digital processing of signals. - SPb., Peter,2002.-608l. Silt.