Analysis of the Reliability Level of Series "P" DC Machines and Electromechanical Converters Based on Bench Test Results
Abstract Generated abstract
This paper analyzes bench durability tests of series “P” direct-current machines and electromechanical converters conducted at the Elektromashina plant beginning in 1963. It describes test procedures, operating-hour distributions, observed failures, and the performance of key assemblies such as commutator-brush systems, insulation, windings, bearings, and automatic-control components. The results indicate elevated failure rates during an initial run-in period, about 2000 to 6000 hours for DC machines and 2000 to 4000 hours for converters, with failures linked both to hidden manufacturing defects and to component-element breakdowns. Statistical processing of the bench-test data suggests that time to failure-free operation for the tested machines follows an exponential distribution, supporting the use of accelerated testing to improve early-life reliability.
Full Text
ANALYSIS OF THE RELIABILITY LEVEL OF SERIES “P” DC MACHINES AND ELECTROMECHANICAL CONVERTERS BASED ON BENCH TEST RESULTS
A. N. Nimirovskii
(Presented by the scientific seminar of the Departments of Electrical Machines and General Electrical Engineering)
In connection with the fact that consumers set before electrical-machine-building plants the task of significantly increasing the service life and operating life of electrical machines, the Elektromashina plant (Prokopyevsk) began to carry it out.
Beginning in 1963, 28 DC electrical machines of frames I to VI were tested for durability on the plant’s benches. The test conditions were created in accordance with the requirements of the technical specifications for these products. In the course of these tests, special attention was paid to
1) the stability of the output electromechanical parameters;
2) the reliability of operation of the component elements in automatic-control circuits;
3) the stability of operation of the commutator-brush apparatus;
4) checking variants of the design of individual modernized units, as well as the influence on reliability of improvements in the technology and in inspection and measuring means;
5) checking modifications of insulating and winding materials, impregnating varnishes, and types of lubricant used in bearing assemblies, etc.
The tests were conducted in cycles of 2000 hours, after which a preventive inspection was carried out, with the defects observed entered in the test report.
Analyzing the tested machines and those of series “P” that are still undergoing tests from the standpoint of reliability and durability, the following may be stated. For 28 machines, the operating time in hours was distributed as follows:
up to 5000 hours — 10 machines,
up to 10000 hours — 5 machines,
up to 15000 hours — 7 machines,
up to 20000 hours — 4 machines,
over 20000 hours — 2 machines.
The largest number of failures was recorded in the period from 2000 to 6000 hours of operation. The dependence of the failure frequency during bench tests on operating time is shown in Fig. 1.
Three motors were removed from testing owing to extremely unstable commutation caused by hidden defects in the commutators. Four motors were removed from testing because of short circuits in the armature windings. One case of breakdown of the armature slot insulation was recorded.
Fig. 1. Dependence of the failure frequency during bench testing of series “P” machines on the duration of their operation.
In total, 13 machines failed; the remaining 15 continue to operate.
After completing its service life, each motor was subjected to a comprehensive investigation both as a whole and by individual assemblies and structural elements.
Simultaneously with the direct-current machines, electromechanical converters with capacities from 1 to 50 kW were also tested. The 14 converters were distributed by accumulated operating hours as follows:
up to 4000 hours — 7 units,
up to 8000 hours — 3 units,
over 10000 hours — 4 units.
The greatest number of failures during the bench-test period was recorded in the interval from 2000 to 4000 hours of operation. Of the 14 converters, 4 were finally removed from bench testing for various reasons, while 10 continue to operate.
It should be noted that 6 converter failures were caused by failure of component elements (capacitors, resistors, diodes, etc., in automatic-control systems), and 8 failures occurred because of hidden defects of a manufacturing nature (deviations from the technology, design, etc.).
Fig. 2. Dependence of the failure frequency during bench testing of converters on the duration of their operation.
The dependence of the failure rate during the bench-test period on the duration of operation is shown in Fig. 2. As can be seen from Fig. 1 and Fig. 2, there is a run-in period amounting to 2000–4000 hours for converters and 2000–6000 hours for machines of the “P” series, during which an increased failure rate is observed. In order to shorten the run-in period, accelerated testing of the machines should be carried out to find ways of increasing reliability in the initial period of operation.
Statistical processing of the results of bench tests of direct-current machines and single-frame electromechanical converters showed that the distribution of the time of failure-free operation of these machines obeys the exponential law.