ON THE EFFECTIVENESS OF FLAW DETECTION IN ROTORS PRIOR TO MACHINING

V. P. Obruchnik, G. G. Okunev

(Recommended by the scientific and technical seminar of the Department of Electric Drive and Automation of Industrial Installations, Research Institute of AEM)

The scientific-technical and production base that has now taken shape at electrical-machine-building plants makes it possible to fully ensure the production of electric motors only of high quality and without in-plant losses. However, in the practical activity of plants, rejects and the low quality of some portion of electric motors have existed and continue to exist. This applies to electric motors that are simple in design, such as asynchronous motors with a squirrel-cage rotor. In these machines the rotor is the simplest unit, and the opinion has taken hold that the rotor problem, as such, does not exist in improving the quality and reliability of squirrel-cage asynchronous electric motors.

Nevertheless, since the appearance of the cast aluminum cage, the question of controlling its parameters—which may deviate substantially from the calculated values owing to defects arising during rotor casting [2]—has not been removed for manufacturers and operating personnel. Serious factors affecting the quality of the squirrel cage are: the pressing pressure of the rotor-iron stack, the temperature of the aluminum, the temperature of the rotor-iron stack, the rate at which the rotor slots are filled with aluminum, the filling pressure, the uniformity of rotor cooling, the presence of impurities in the aluminum, the contact resistance between the bar and the rotor iron, and also the casting method. All this can lead to breaks, misruns, cavities, and other defects in the bars of the squirrel cage. Excessive pressing of the rotor stack and nonuniform cooling often cause cracks in the bars. An unsuitable aluminum temperature may cause constrictions, ruptures, porosity, and cavities in the bars.

Investigations of squirrel-cage asynchronous electric motors of frame sizes 2–11 carried out at several plants in the country have established that rejects due to internal defects of the cast squirrel cage of rotors range from 0.3 to 11% of the total output of these motors. Thus, for example, for the plants of Moscow, Kharkov, Yaroslavl, Kemerovo, and Ulan-Ude, the average percentage of such rejects is 4.1%. It may be assumed that rejects of electric motors due to internal defects of rotors occur at all electrical-machine-building plants in the country, and if they are taken to be 4%, then the losses at manufacturing plants amount to enormous sums.

Let us show this by the example of specific calculations based on data on the output of squirrel-cage electric motors with power up to 100 kW in 1967. Assuming that 5 million of these motors were produced in 1967 [1] and that their power is distributed approximately equally by frame size, one can compile data Tables No. 1 and No. 2. In these tables the wholesale prices

electric motors are taken from [3], and their quantity by frame size was established on the basis that it is inversely proportional to power. For example, if a motor of the ninth frame size has a power of 100 kW, and one of the first frame size has 0.8 kW, then there will be \(100 : 0.8 = 125\) times as many motors of the first frame size as of the ninth.

Table 1

Table 2

According to Tables 1 and 2, the overall average losses from rejects of defective rotors of a single series of electric motors of frame sizes 2–9, produced in 1967, amounted to:

\[ \frac{1}{2}(688 + 619.8) = 654 \text{ thousand rubles per year.} \]

Losses from rejected rotors can be reduced to a minimum if this reject is detected immediately after casting of the squirrel cage, that is, before machining the rotor barrel. Then, first, the need for...

...it is necessary to write off rejected rotors as scrap metal, since it will not be difficult to repair them by recasting, while retaining the scarce metal of the rotor pack; second, when internal defects in rotors are detected during the first operations after casting, their further processing is suspended, which makes it possible to save human labor and not occupy equipment for the production of an unusable motor. The savings from reducing rejects at manufacturing plants are presented by the data in Table 3. They amount to approximately 477.6 thousand rubles per year.

Table 3

Average savings: \( \frac{1}{2}(503.0 + 452.2) = 477.6 \) thousand rubles.

Well-known methods for sorting rotors, for example, the three-ammeter method, the method of measuring slip at the nominal load of electric motors, comparison of the mechanical characteristics of motors, measurement of deviations in short-circuit currents (the PZR method), and others, do not make it possible to detect all defects in cast rotors and may pass defective motors with breaks in one, two, and even three bars of the squirrel cage [6, 7, 8].

Naturally, electric motors with internal defects in the rotor squirrel cage are unreliable in service and fail quickly. Investigations carried out in 1961 [2, 6], and mathematical statistics, established that at consumer plants, for rotor-related reasons, 1.4–2% of electric motors of the total number in operation fail prematurely every year. Approximately the same percentage, and even more, of rotor defects is not detected at manufacturing plants by the control methods used. Obviously this coincidence is not accidental. There are data in the literature [8] that in a rotor with damaged bars, the current distribution is disturbed, leading to asymmetry of the secondary currents, which, in a loaded motor, causes a cascading process of destruction of the squirrel cage.

Additional expenditures have to be made for the capital repair of electric motors that have failed prematurely. They considerably exceed the losses from rejects at manufacturing plants.

If we assume that, of the 2,640 thousand electric motors of frame sizes 2–9 produced in 1967, 1.4% failed prematurely because of rotor defects, then it is not difficult to find the number of motors that consumers had to repair in order to eliminate the manufacturers’ defects:

\[ 2640 \cdot 0.014 = 3696 \text{ units.} \]

The annual costs of carrying out capital repairs per motor, as an average representative of the power range (3.1–5) kW, amounted to [4]:

wages — 15 standard hours · 1.6 repair-complexity units · 0.443 rub/hour = 10.63 rubles,

materials — 110% of the workers’ basic wages = 11.69 rubles,

shop expenses—150% of the basic wage = 15.94 rubles.
Total—38.26 rubles.

The total amount of forced annual expenditures for carrying out major repairs of electric motors that prematurely fail because of rotor defects will be:

\[ 38.26 \cdot 3696 = 1414 \text{ thousand rubles.} \]

This amount will increase still further if one adds to it the costs of dismantling failed motors and installing new ones, if one takes into account losses from downtime of technological equipment during this period, etc.

Thus, the aggregate losses from rotor defects in three-phase squirrel-cage asynchronous electric motors, attributable to electrical-machine-building enterprises (attributable for the time being involuntarily and explainably), in 1967 for motors of frames 2–9 amounted to:

\[ 1414 \text{ thousand rubles} + 477.6 \text{ thousand rubles} = 1891.6 \text{ thousand rubles.} \]

Calculations have shown that in 1970 these losses may exceed 2.45 million rubles per year. As can be seen, the seemingly nonexistent rotor problem results in appreciable losses for the state.

It should be noted that the figures given for monetary losses from defects in the rotors of asynchronous motors are greatly understated, since the rejection percentages were taken only from test data for motors of frames 2–9 of general industrial design. Higher requirements are imposed on special-purpose machines, mine electric motors, motors used in chemical enterprises, etc., and the rejection percentage due to rotor causes is considerably higher. For example, if in the electric-drive systems of machine tools operating under nonsevere conditions breaks in one or two bars of the rotor cage may be tolerated, then when motors operate in an explosive atmosphere even cavities or thinning of the bars may lead to disaster. No defects of the rotor cage are desirable in motors to which increased vibration requirements are applied.

We have also not cited losses from rotor defects in motors of frames 10–11. It is known that in manufacturing the rotor cage of these machines, scrap is most probable, and its untimely detection leads to appreciable material losses. For example, if a defect in the rotor cage of one frame-11 motor is detected after assembly of the machine, the loss will amount to 230–250 rubles.

At Tomsk Polytechnic Institute an instrument has been developed that detects practically all the main casting defects of the rotor cage of three-phase asynchronous squirrel-cage electric motors produced by domestic industry; moreover, the defects are detected immediately after casting of the cage, that is, before machining of the rotor body. This rotor flaw detector, called RDS-T, has already successfully passed tests at six electrical-machine-building enterprises of the USSR.

During testing of the flaw detectors, rotors of motors of frames 2–11 of almost all type designs were checked. It was established that the installations made it possible to reliably detect defects of the squirrel-cage winding leading to a change in the active resistance of one bar by 15% and more, including breaks, cracks, thinning, cavities, etc. [9]. The RDS-T flaw detectors proved to be simple in design and execution and very reliable in operation. They can be installed in any production area and built into conveyor and automatic lines. The average productivity of the RDS-T is not less than 50 rotors per hour. The cost of one flaw detector, when manufactured individually under the conditions of factory electrical workshops, is 560–700 rubles.

Enterprises that have put the RDS-T rotor flaw detectors into operation have given these installations a high appraisal. This appraisal was also facilitated by the fact that the flaw detectors immediately provided the plants with considerable savings. For example, the Vladimir Ilyich Plant in Moscow

receives from flaw detection of rotors on the RDS-T an annual saving of 48–52 thousand rubles.

Obviously, installations for checking the squirrel-cage parameters of three-phase asynchronous motors before machining the rotor barrel should be of interest to all electrical-machine-building—and possibly also repair—enterprises in our country. Glavelektromash, in turn, should discuss the question of producing a large batch of these flaw detectors in order to satisfy the numerous orders from its plants for the manufacture of the RDS-T that are being received by the Tomsk Polytechnic Institute, which, naturally, cannot fulfill them, being simply an educational higher-education institution.

Thus, even the few examples and approximate calculations reveal substantial losses caused by imperfect methods of detecting rotor defects—losses amounting to millions of rubles. Naturally, such a situation is impermissible when there are, in principle, possibilities for practically eliminating rejects in the rotors of motors being produced, and for doing so at the very first operations of rotor manufacture, with a large technical-and-economic effect. It may be possible to solve this problem quickly by widely introducing RDS-T flaw detectors. But in one way or another, the managers of enterprises producing electric motors, innovating engineers, and researchers must, in the very near future, solve this problem positively.

REFERENCES

1. B. I. Kuznetsov. Production of electric motors with power up to 100 W. The Electrical-Engineering Industry of the USSR. Jubilee collection, 1967.

2. N. A. Tishchenko. The problem of reliability of electric motors. Elektrichestvo, Nos. 11, 12, 1961.

3. Price List of Wholesale Prices for Electric Motors No. 1501. Moscow, 1967.

4. Unified System of Scheduled Preventive Maintenance and Rational Operation of Technological Equipment at Machine-Building Enterprises. “Mashinostroenie,” 1967.

5. Standard Methodology for Determining the Economic Effectiveness of Introducing New Technology, Mechanization, and Automation of Production Processes in Industry. Publishing House of the USSR Academy of Sciences, 1963.

6. A. A. Saprykin. Defects of rotor bars. Mining Electromechanics. Collection of the Donetsk Coal Institute, No. 19, 1961.

7. A. A. Taits. The state of and experience in repairing electrical equipment at industrial enterprises. Collection: Problems of Reliability and Repair of Electrical Machines. No. 2. Moscow, 1961.

8. V. N. Cherevatov. Causes of breakage of rotor bars in asynchronous motors of the USSR Academy of Sciences. Collection of works on problems of electromechanics. Issue 8. 1963.

9. V. P. Obrusnik, A. I. Zaitsev, E. G. Zavatskii. Development and study of an RDS-T-type rotor flaw detector (scientific and technical report of the Scientific Research Institute of Electrical Machines). Tomsk, 1969.