Effect of Thermal Aging on Defect Formation in the Insulation of Enameled Wires

L. E. KUKOLEVA, Yu. P. POKHOLKOV, V. M. SHAMSHUR
Submitted 1974 | SovietRxiv: ru-197401.13610 | Translated from Russian

Abstract Generated abstract

This paper examines whether thermal aging increases through defects in the enamel insulation of low-voltage winding wires, contributing to failures beyond the run-in period. The authors define wire defectiveness from breakdown-voltage distributions and test PEV-2 and PETV-F-35 wire specimens aged at elevated temperatures, using industrial-frequency AC breakdown measurements with shot electrodes. The results indicate that defectiveness remains nearly unchanged for an initial period and then rises sharply at times dependent on aging temperature, while the breakdown strength of undamaged insulation remains relatively high. The study attributes the increase to through cracks formed by structural changes, internal stresses, and bond rupture in the enamel insulation, supporting the use of defectiveness as a parameter for estimating winding reliability during physical aging.

Full Text

EFFECT OF THERMAL AGING ON THE FORMATION OF DEFECTS IN THE INSULATION OF ENAMELED WIRES

L. E. KUKOLEVA, Yu. P. POKHOLKOV, V. M. SHAMSHUR

(Presented by the scientific seminar of the Department of EIKT)

Failures of turn insulation of windings of low-voltage electrotechnical structures during the run-in period are caused mainly by the manifestation of through-damages (defects) [1]. The breakdown voltage of sound insulation of enameled wires, as a rule, greatly exceeds the level of the operating voltage and the level of commutation overvoltages [2].

At present, at TPI, RTM have been developed for calculating the reliability of turn insulation of windings of asynchronous motors during the run-in period, where the principal initial parameters are taken to be the defectiveness of enameled wires \((q)\) and the level of commutation overvoltages. In this case, the defectiveness \(q\) is understood to mean the probability of the occurrence, in a batch, of such specimens whose breakdown voltage is lower than the maximum breakdown voltage of air at a thickness equal to the one-sided thickness of the wire insulation (allowing for overlap along the surface)

\[ q=\frac{n_{\bar u_{\max\ \text{def}}}}{\sum n_i}, \]

where

\(n_{\bar u_{\max\ \text{def}}}\) is the number of specimens broken down by the voltage \(U_{\max\ \text{def}}\),

\(\sum n_i\) is the sum of specimens in the batch.

\[ U_{\max\ \text{def}}=\overline{U}_1+3\sigma_1, \]

where

\(\overline{U}_1\) and \(\sigma_1\) are, respectively, the mean value and the root-mean-square deviation of the breakdown voltage of the wire at the site of through-damage.

Through-damages of enamel insulation of wires, of the type of tears, cuts, and punctures, may appear both during the manufacture of the wires and during the manufacture of windings. Most of these defects during the run-in period of structures, as already stated, are revealed; however, failures of structures continue to be observed, and their frequency increases as the insulation undergoes physical wear under the action of opera-

tional factors (heat, vibration, etc.). It may be assumed that the increase in the failure rate of the turn insulation of asynchronous motors is caused by the appearance, in the insulation during aging, of defects of the through-crack type. The assumption that through cracks arise on the surface of enamel-insulated wire was checked by a number of experiments. Thus, a wire laid in the stator slots was removed from the slots and tested with a breakdown voltage. From the observed values a histogram of the distributions of \(U_{\mathrm{br}}\) was constructed. Then an analogous histogram was constructed for a wire whose insulation had been damaged through and through (ring cuts were made on the insulation). The left-hand parts of the histograms were identical. A similar experiment was also carried out during aging of the enamel-insulated wire. Confirmation of this assumption made it possible to use the estimate of the dependence \(q=f(\tau)\) for estimating the probability of failure-free operation of windings not only during the running-in period, but also over any time interval. For this purpose we investigated the defectiveness of the insulation of PEV-2 and PETV-\(F\)-35 wires of diameter 1.25 mm in the process of thermal aging. PEV-2 wire was chosen as the one most thoroughly studied in all respects, and PETV-\(F\)-35 as the most widely used. The duration of thermal aging is given below.

Aging time

Table

Wire grade 140° 160° 180° 200° 220°
PEV-2 14 days 4 days 1 day
PETV-\(F\)-35 — — 14 days 4 days 1 day

The specimens were pieces of wire 120 mm long. Breakdown tests were carried out at an alternating voltage of industrial frequency. Shot with a diameter close to the diameter of the wire was used as the electrodes.

Thermal aging was carried out as follows: the prepared wire specimens were divided into samples (sample size 50 specimens), which were placed in a thermostat. Individual samples were removed

Fig. 1. PEV-2 wire, \(t=180^\circ\)C. Dependence \(q=f(\tau)\)

Fig. 1. PEV-2 wire, \(t=180^\circ\)C. Dependence \(q=f(\tau)\)

removal from the thermostat was carried out after a time specified for each temperature, as indicated in the table.

Based on the observation results for the values of \(U_{\mathrm{pr}}\), histograms were constructed and the magnitude of the wire defectiveness was determined. The form of a typical distribution curve \(f(U_{\mathrm{pr}})\) is given in [3].

Fig. 1 gives the dependence \(q=f(\tau)\) for PEV-2 wire for a temperature of \(180^\circ\). Figs. 2 and 3 show changes in the mean value and the root-mean-square deviation of the breakdown voltage of the entire insulation (including defective spots), \(\overline{U}_1\), \(\sigma_1\), and, separately, the same values characterizing only sound insulation, \(\overline{U}_2\) and \(\sigma_2\). The separation of defective and sound insulation was carried out according to the value of the breakdown voltage \(U_{\max\,\mathrm{def}}=1.6\ \mathrm{kV}\).

Fig. 2. PEV-2 wire, \(t=180^\circ\mathrm{C}\). Dependence \(U=t(\tau)\)

Fig. 2. PEV-2 wire, \(t=180^\circ\mathrm{C}\). Dependence \(U=t(\tau)\)

As can be seen from the figures, the defectiveness of the wires does not change over a certain period of time, and then increases sharply. The time at which the value \(q\) begins to increase is determined by the aging temperature. A comparison of these times with the total duration of the cycles recommended by GOST 10519-63 shows that

Fig. 3. PEV-2 wire, \(t=180^\circ\mathrm{C}\). Dependence \(\sigma=f(\tau)\)

Fig. 3. PEV-2 wire, \(t=180^\circ\mathrm{C}\). Dependence \(\sigma=f(\tau)\)

the growth of defects begins after 9–11 cycles, i.e., after approximately the same number of cycles that determines the average service life of the windings. The latter makes it possible to assume that failures of windings during the period of physical aging are also caused by the presence of through defects appearing as a result of aging of the enamel insulation of the wires. This is also confirmed by the fact that the value \(\overline{U}_2\) (the mean breakdown voltage of sound insulation) remains fairly high even for long aging times. Taking into account the small value of \(\sigma_2\), it may be said that breakdown of such (undamaged) insulation under the action of switching overvoltages is unlikely.

Defects formed during thermal aging of enamel wires are through cracks formed as a result of structural changes in the enamel insulation, which lead to the appearance of internal mechanical stresses and to bond rupture.

Conclusions

Observation of the change in defectiveness of the insulation of enamel wires of grades PEV-2 and PETV-\(F\)-35 in the process of thermal aging shows that, in the range of times approximately corresponding to the average service life, a sharp increase in defectiveness is observed—the appearance of through cracks.

References

  1. RTM OAA 682 022-71. Three-phase asynchronous motors for general use with power from 0.12 to 100 kW. Tomsk, 1971.

  2. O. D. Gol'dberg, N. S. Suvorov. “The influence of switching overvoltages on the reliability of low-voltage asynchronous motors.” Elektrotekhnika, 1968, No. 5.

  3. Yu. P. Pokholkov, V. I. Derevyanko. “On a method for determining the defectiveness of winding enamel wires used in low-voltage electrotechnical equipment.” In the present collection.

Submission history

Effect of Thermal Aging on Defect Formation in the Insulation of Enameled Wires