Study of Gradients in the Leader Channel

Student K. S. Nazarova

Supervisor: Professor, Doctor I. S. Stekolnikov

(Presented by the scientific seminar on dielectrics)

The study of atmospheric electricity is of enormous importance from the standpoint of creating reliable lightning protection, which is one of the factors ensuring an uninterrupted supply of electric power to consumers.

This question is especially timely at present in connection with the construction of a large number of transmission lines. The study of the physics of lightning under laboratory conditions is based on the analogy between the nature of lightning and that of the long spark.

As is known, there are two main stages in the development of a discharge, both in lightning and in the long spark \[1\]: the leader stage and the stage of the main discharge. Although the leader stage is the more thoroughly studied, some parameters of the leader still have not received proper treatment in the literature. Thus, the gradients of the leader channel have been studied theoretically and, especially, experimentally to an insufficient degree.

The experimental study of gradients in the leader channel was carried out by the author in the high-voltage gas-discharge laboratory of ENIN, Academy of Sciences of the USSR, under the direction of the head of the HVGR laboratory of ENIN, Doctor of Technical Sciences, Professor I. S. Stekolnikov, using the latest advances in measuring technology, such as a photoelectronic multiplier—PEM, a Kerr cell, and a high-speed electron oscillograph.

To study gradients in the leader channel, two circuits were used.

The block diagram of the first setup, using a PEM and a Kerr cell, is shown in Fig. 1.

In the discharge gap, positive rod—plane, additional electrodes serving as probes were placed. The change in the potential difference between them during the discharge process is fed to the electrodes of the Kerr cell. A lamp was placed in front of the polarizer of the cell. Depending on the change in potential difference between the probes, the Kerr cell transmits a varying light flux from the lamp. This time variation of the light falls on the photocathode of the PEM, which converts the light signal into an electrical one; the latter is then amplified by an amplifier and fed to the deflection plates of an electron oscillograph.

Thus, the EO records the magnitude of the potential difference between the probes at all stages of the discharge, including the leader stage.

To check the operation of the measuring circuit and to calibrate it by voltage, the block diagram (Fig. 2) was used, consisting of a high-voltage rectifier unit, a rectangular-pulse generator, and a delay circuit. The experiments were carried out with a terminating resistance \(R_t = 1350\ \Omega\).

Fig. 1. Block diagram for studying gradients of the leader channel by the first method.

Fig. 2. Circuit for determining the sensitivity of the measuring installations.

During the investigation it became necessary to increase the sensitivity of the Kerr cell by reducing the distance between its electrodes from 5 to 0.8 mm. Oscillograms of the voltage between the probes and of the current in the gap, which is needed for processing the results, are presented in Fig. 3. The distance between the electrodes of the cell is 1.3 mm.

As can be seen from the oscillograms, during the corona stage there is saturation of the FEU, while in the leader stage there are deflections on the oscillograph due to the FEU current.

Thus, because of the large difference in voltages in the corona stage and at the moment of leader passage, it was impossible to measure accurately...

gradients in the leader channel, since when the sensitivity of the Kerr cell was increased, breakdown occurred in it at the corona stage.

However, taking into account the sensitivity of the circuit, it was possible to establish that the gradients in the leader channel are less than 1 kV/cm with a damping resistance \(R_{\mathrm{t}} = 1350\ \Omega\).

Another circuit was used, shown in Fig. 4, representing a modification of the first. A positive

Fig. 3. Parameters of the impulse-voltage generator.

Fig. 3a. Oscillogram of the gradient in a long spark, obtained by measurement by the first method; frequency of the calibration oscillations 500 kc/s.

Fig. 3b. Oscillogram of the current in a long spark. Frequency of the calibration oscillations 500 kc/s.

Fig. 4. Circuit for measuring gradients in a long spark by the second method.

rod—a sphere—was placed in the discharge gap as a probe. The distance between the probe and the rod was varied. During the discharge the leader, developing from the rod, closes the gap between the rod and the probe.

The voltage between the rod and the probe was applied to a capacitive divider consisting of series-connected capacitors of ca—

with capacitances of 3 pF and 55 pF. The voltage of one of the capacitors was applied to the plates of the event oscilloscope (EO No. 1).

The sensitivity of the circuit was determined according to the circuit in Fig. 2.

Oscillography was carried out with damping resistances \(R_T = 1350\ \Omega\) and \(R_T = 30\ \text{k}\Omega\). Oscillograms of the voltage between the probe and the rod are shown in Figs. 5a and b.

Fig. 5. Oscillogram of the leader gradient, obtained in measurement by the second method:
a) \(R_T = 1350\ \Omega,\quad S = 5\ \text{cm};\)
b) \(R_T = 30\ \text{k}\Omega,\quad S = 10\ \text{cm}.\)

In Fig. 5a an oscillogram is presented that was recorded with \(R_T = 1350\ \Omega\) and a distance between the rod and the probe \(S = 5\ \text{cm}\). In Fig. 5b—with \(R_T = 30\ \text{k}\Omega\) and \(S = 10\ \text{cm}\).

The results of the experiments are presented below.

With \(R_T = 30\ \text{k}\Omega\) and \(S = 5\) and \(10\ \text{cm}\), the gradient \(g = 250\ \text{V/cm}\)

    “  \(R_T = 1350\ \Omega\) and \(S = 5\ \text{cm}\)    “  \(g = 250\ \text{V/cm}\)

    “  \(R_T = 1350\ \Omega\) and \(S = 10\ \text{cm}\)   “  \(g = 500\ \text{V/cm}\)

Owing to the comparatively small number of experimental data, the obtained values of the gradients are approximate in character, and investigations in this direction are continuing.

LITERATURE

1. Stekolnikov I. S. Study of the nature of long sparks, I. Izv. AN SSSR, OTN, No. 11, 1952.

Errata

Bull. TPI, vol. 95.