Effect of Gas Inclusion Shape on Electrical Insulation Failures

L. V. Alypova, V. S. Dmitrevsky
Submitted 1974 | SovietRxiv: ru-197401.13602 | Translated from Russian

Abstract Generated abstract

This paper examines whether long-term breakdown of solid electrical insulation containing gas inclusions is governed primarily by erosive wear from partial discharges or by enhanced local electric fields around ionized inclusions. Using three-layer polyethylene film specimens with artificial holes of different diameters but equal total cross-sectional area, the authors compared breakdown paths under applied voltages while accounting for differences in partial-discharge inception voltage and expected discharge intensity. In all tested specimens at 20 and 27 kV, breakdown occurred through the small-diameter inclusions, which corresponded to the greater local field intensity rather than the higher partial-discharge intensity. The results support the view that inclusions elongated in the direction of the electric field are especially dangerous and that partial-discharge inception voltage alone may not determine insulation lifetime.

Full Text

INFLUENCE OF THE SHAPE OF A GAS INCLUSION ON ELECTRICAL INSULATION FAILURES

L. V. Alypova, V. S. Dmitrevsky

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

Partial discharges in gas inclusions of a solid dielectric are accompanied by the formation of a flux of electrons and ions bombarding the walls of the inclusion, by chemically active compounds, and by an increase in temperature. The simultaneous action of the flux of electrons and ions and of chemically active substances causes destruction of the solid dielectric.

Erosive destruction of a dielectric under the action of partial discharges has been observed experimentally by a number of investigators on models with flat pores [1—3]. In practical insulation, gas inclusions usually have an ellipsoidal shape; in them a rapid extinction of partial discharges is observed [4]. Naturally, in such gas inclusions the erosive wear of the insulation will proceed very slowly and is unlikely to cause breakdown of the insulation. Meanwhile, in [5] it is indicated that, with a shorter duration of voltage application, specimens in which the extinction of partial discharges occurs more rapidly break down.

In [6] a hypothesis was put forward concerning the determining influence on insulation breakdown of local field intensities in the solid dielectric. If the inhomogeneity is an ellipsoid of revolution, then the magnitude of the local field intensity \(E_{\mathrm{m}}\), at an average field intensity \(E\), is found from [7]

\[ E_{\mathrm{m}} = \frac{E}{ 1-\left(1-\frac{1}{v}\right)e^{-\frac{0.41}{y^{1.22}}} }, \tag{1} \]

where

\(v\) is the ratio of the total specific conductivities of the inhomogeneity and the dielectric,

\[ y=\frac{a}{b} \]

is the ratio of the axes of the ellipsoid of revolution of the inhomogeneity (\(a\) is the length of the inclusion in the direction of the field, \(b\) is the length of the inclusion in the direction perpendicular to the field).

For an ionized gas inclusion, \(v\) may be taken as close to infinity. Then the greatest local field intensity is determined by the quantity \(y\). It follows from [1] that \(E_{\mathrm{m}}\) will be the greater, the smaller \(b\) is and the larger \(a\) is.

The aim of the present work is to determine the main cause of breakdown of electrical insulation when gas inclusions are present in it and a voltage is applied for a long time; namely, whether the principal cause is erosion wear by partial discharges or high local field strengths leading to rupture of bonds in the material.

The experiments were carried out on polyethylene film with artificial gas inclusions. The specimen consisted of three layers of film 0.19 mm thick. Holes of diameter 1 mm (1 hole) and diameter 0.1 mm (100 holes) were made in the middle layer of the film. The number of holes of small diameter was chosen so that the total cross-sectional area of the small-diameter holes would be equal to the cross-sectional area of the large-diameter hole. Thus, the influence of the dielectric area on insulation breakdown was eliminated. A sketch of the specimens is given in Fig. 1, where the following are denoted: 1—small-diameter holes and 2—large-diameter holes. On analogous specimens, but with one type of gas inclusion, the partial-discharge inception voltage \((U_{\mathrm{н}})\) was measured. The average value of \(U_{\mathrm{н}}\) for specimens with holes of type 1 was 2.6 kV, and for specimens with holes of type 2, 2.0 kV. The intensity of partial discharges is directly proportional to \((U - U_{\mathrm{н}})\), i.e., the higher partial-discharge intensity should occur with inclusions of type 2.

Fig. 1. Specimen construction: 1—small-diameter hole, 2—large-diameter hole.

Fig. 1. Specimen construction: 1—small-diameter hole, 2—large-diameter hole.

Consequently, if insulation breakdown under prolonged application of voltage is caused by erosion of the dielectric by partial discharges, then the breakdown channel should pass through an inclusion of type 2. If, however, the greatest local field strength produced around an ionized gas inclusion is responsible for breakdown under prolonged application of voltage, then the breakdown channel should pass through inclusions of type 1, since in this case \(y\) is 10 times greater than for an inclusion of type 2 (1). Table 1 gives the results of testing specimens at different applied voltages.

Table 1

Applied voltage, kV Number of specimens tested Number of specimens with a breakdown channel through an inclusion, type I Number of specimens with a breakdown channel through an inclusion, type II
20 15 15 0
27 15 15 0

The data of Table 1 show that insulation breakdown always occurred at the point of greatest local field strength.

The experiments performed make it possible to draw the following conclusions:

  1. The greatest danger to insulation is posed by gas inclusions elongated in the direction of the electric field.

  2. The partial-discharge inception voltage in practical structures does not determine their service life.

  3. The high rate of partial-discharge decay, characteristic of inclusions of small diameters, indicates the existence of gas inclusions elongated in the direction of the field.

References

  1. S. N. Koikov, A. N. Tsikin. Change in the breakdown voltage, thickness, and weight of polymer films under ionization aging. In: Breakdown of Dielectrics and Semiconductors. Energiya, 1964, pp. 311–314.

  2. I. H. Mason. Deterioration and breakdown of dielectrics resulting from internal discharges. Proc. of the IEE, 1951, 98, No. 1, 44–58.

  3. N. S. Ilchenko, G. I. Gavrilyuk, A. V. Kovalev. Influence of the magnitude of ionization intensity on the service life of polyethylene α-film. Izvestiya vuzov, “Energetika,” No. 3, 1966, 20–26.

  4. V. S. Dmitrevsky, R. M. Kessenikh, Yu. N. Shumilov. Decay of ionization in gas inclusions of organic dielectrics. Elektrotekhnika, in press.

  5. A. N. Perfiletov. Electrical strength of cable polyethylene and cables with polyethylene insulation. In: Breakdown of Dielectrics and Semiconductors. Energiya, 1964, pp. 117–122.

  6. V. F. Grafov, A. A. Vorob’ev, V. S. Dmitrevsky, V. A. Lkhamazhapov, V. I. Merkulov. Aging of solid dielectrics and the reliability of electrical insulation. Izv. SO AN SSSR, in press.

  7. V. S. Dmitrevsky, V. I. Merkulov. Distribution function of the probabilities of breakdown electric-field strengths of polymer dielectrics. Izvestiya vuzov, “Fizika,” No. 1, 1971, pp. 49–55.

Submission history

Effect of Gas Inclusion Shape on Electrical Insulation Failures