Arc Ignition Methods and Combustion Characteristics of Small-Current Arc Faults in High-Voltage Cables

Abstract

High-voltage cables will continue to operate for a period of time in the event of a small current arc fault, which poses a risk of fire. Two simulated ignition methods, moving electrode and melting fuses, are proposed to analyze the ignition characteristics of low-current arcs. The ignition test was carried out, and the combustion effect was compared. The results indicate that the moving electrode ignition method can achieve long-distance arc ignition test when the current is small and is suitable for simulating the arc ignition situation of cable outer protective layer damage. By controlling the movement speed, it can be ensured that the arc will not be interrupted during the electrode movement process. However, the arc is difficult to sustain using the fuse melting method when the current is small and the distance is long. The fuse melting method is suitable for simulating insulation breakdown situations. The results show that the critical arc duration for cable ignition under five different current conditions of 2–10 A is 28 s, 21 s, 14 s, 9 s, and 4 s, respectively. The maximum height of the cable flame under 2–10 A arc current is 9–52 cm and 16–63 cm, respectively, when the arc duration is 50 s and 100 s. The self-ignition time of the cable after the arc extinguishing is 8–95 s and 14–261 s, respectively. The maximum temperature of the cable flame is positively correlated with arc current, and the maximum flame temperature of the cable under 2–10 A arc current is 540–980 °C. Based on the actual current monitoring data in cable tunnels, the research results can provide reference for the risk assessment and protection of cable tunnel fires.

1. Introduction

With the rapid development of cities, the number of cables in urban power grids is rapidly increasing. There is a certain degree of aging and deterioration in the outer protective layer and insulation layer of cables that have been in service for a long time [1,2]. Then, the breakdown fault of the main insulation or outer protective layer of the cable may occur. High-temperature arcing caused by breakdown leads to frequent cable fire accidents [3,4,5]. At present, most researchers mainly focus on fire accidents caused by large current breakdown, but few researchers focus on fire accidents caused by arc ignition of a small current. Small-current short-circuit arc fault is mainly divided into two kinds. One is that the outer sheath of the high-voltage cable is damaged. The suspended voltage generated by the metal sheath breaks down the outer sheath. Due to the low induced voltage on the metal shielding layer, the arc current formed is relatively small, and the current of the metal shielding layer will not trigger the protective action. Therefore, the fault arc continues to act. The other is when a single-phase ground fault occurs in the distribution network cable. The distribution network can run with the fault for two hours [6,7], causing the fault arc to persist. In recent years, channel fire accidents caused by single-phase grounding faults of 10 kV distribution cable often occur [8,9,10]. Although the arc current of these accidents is generally less than 10 A, the arc temperature is high, reaching thousands of degrees Celsius, and these arcs last for a long time. The cable burns under the continuous action of high temperature. It is urgent to carry out low-current arc ignition tests to study the ignition characteristics under different current amplitudes and durations.

In recent years, a certain amount of research has been carried out on the ignition process of cables under the action of electric arcs. He et al. [11] built a physical test platform for arc fault. The notch was pre-fabricated on the cable insulation layer. The cable was energized at this notch to generate an arc. However, due to the short duration of the current and the small current, the cable was not successfully ignited. Liu et al. [12] carried out dynamic simulation of fire in a 10 kV cable single-phase arc grounding fault ignition tunnel, but the current lasted only 0.5 s. The cable was only ablated and did not continue to burn. Tartawski et al. [13] analyzed the influence of arc ignition on materials by a self-made high-voltage and low-current arcing device. The continuous arc was not generated due to the short duration of this method, and the application conditions were limited. Although the modified composite material has high arc resistance, the material will still be seriously damaged after exposure for more than 60 s. Li et al. [14] used the Jacob’s ladder arc, generated by breakdown air as the fire source to construct a cross-linked polyethylene cable experimental platform with arc ignition of 110 kV voltage level. However, the test power supply is a 20 kHz high-frequency external charging power supply, and the arc current amplitude was only 0.13 A. The current gap with the actual fault state is large; thus, it cannot reflect the arc ignition characteristics under the fault state. Xu et al. [15] constructed a test platform for ignition of a 1 kV distribution cable. The influence of arc power on the ignition process of the cable was investigated by using air breakdown arc combined with fire dynamics simulation. The results show that when the arc power decreases from 21.93 W to 7.15 W, a continuous arc was formed for a certain time, but the time is short and a large fire cannot be formed. The current method of arc ignition cable is quite different from the actual fault arc ignition process. When the low-current arc is difficult to sustain, the cable is difficult to ignite. Further research is needed on ignition testing methods and ignition characteristics for low-current arc faults in cables.

Therefore, this research is carried out on ignition methods under the action of sustained arc faults in small-current fault systems in the test. Firstly, an arc ignition test platform was built to verify the feasibility of generating a continuous arc by two arc ignition methods: moving electrode ignition and fuse melting ignition. Then, the advantages and disadvantages of the two arc ignition methods were compared and analyzed. Based on the continuity of the arc and the ignition effect of the cable, the arc ignition method of fuse melting was selected. Finally, arc ignition tests under different conditions were carried out. The ignition characteristics, temperature rise characteristics, and fire trends of the cable were obtained, and then the influence of arc current size and duration on the cable fire was analyzed. The test results can provide reference for fire risk assessment and protection of cables.

2. Test Method for Small-Current Arc Ignition

In the atmospheric gap, the potential gradient needs to reach 30 kV/cm to cause electric field breakdown. Once the arc has been formed, a lower voltage can be used to maintain a longer arc without extinguishing it. In a stable atmospheric pressure environment, the arc maintenance voltage per centimeter length is approximately 15 to 30 volts. Based on the above arc characteristics and the findings of contemporary scholars, two cable ignition simulation methods are proposed in this research. The two methods are moving electrode ignition and fuse melting ignition. This study preliminarily determined the required experimental parameters for two methods and analyzed the respective advantages and disadvantages based on the characteristics of the two ignition methods.

2.1. Moving Electrode Ignition Method

Two electrodes are in contact. The one located on the high-voltage terminal side is the high-voltage electrode, and the other located on the ground terminal side is the ground electrode. The external control is used to move the grounding electrode away from the high-voltage electrode. When the two electrodes separate from the contact state, a high field strength will be formed in the extremely small electrode spacing. Subsequently, an electric arc is generated between the two electrodes through electrical breakdown. As the electrodes are displaced, the distance between the electrodes increases, resulting in a continuous stretching of the arc. This process is to form an arc by moving electrodes. If the process is conducted on the surface of the cable and the arc generated between the two electrodes acts on the surface of the cable body, the arc ignition process of the cable can be achieved under suitable test parameters.

2.1.1. Building the Test Platform

The test platform utilized in this test to ignite the outer sheath of the cable with the moving electrode was primarily composed of four components: the power supply apparatus, protective resistor, measuring apparatus, and arc initiation device. The power supply equipment is a single-phase AC-regulated power supply box with a maximum output current of up to 10 A. The protection resistor employs a wire wound resistor with a resistance of 100 Ω and a rated power of 10 kW to constrain the loop current to the maximum current range that the power supply apparatus is capable of withstanding. The measurement apparatus comprises a high-voltage probe and a current coil. The data are collected through the Tie Pie Handyscope HS5 acquisition card with a collection frequency of 2000 Hz. The layout of the arc-starting device is shown in Figure 1. The ground electrode was arranged above the screw slide. The software was used to set parameters such as moving distance and moving speed. The step electrode drives the screw to rotate, thereby achieving free movement control of the ground electrode.

The power supply equipment, protective resistor, measuring equipment, and arc starting device were wired according to the circuit shown in Figure 1. The wire was led out from the power supply equipment and connected to one end of the protective resistor, and the other end of the protective resistor was connected to the high-voltage electrode. A high-voltage probe and current coil were connected between the protective resistor and the high-voltage electrode, and the ground electrode was connected to the ground side. The moving electrode and the fixed electrode must be in contact. The entire test loop has been constructed, and the layout of the test platform is illustrated in Figure 2. The new 10 kV cables were used in this experiment; the insulation layer of the cable sample used in the test is cross-linked polyethylene XLPE, and the outer sheath is polyvinyl chloride PVC. The cable sample that needs to be simulated for arc ignition was placed under the two test electrodes; the cable must be tightly attached to the lower surfaces of the two electrodes. Therefore, the arc can be ignited on the surface of the cable by moving the electrodes to ignite the cable. This test platform was designed to simulate small-current arc ignition cable tests. As long as the simulated fault current value was controlled within the rated current range of the power supply equipment, this test platform can generate an arc on the surface of the cable to conduct ignition tests.

2.1.2. Realisation of Continuous Arc Ignition and Cable Ignition Processes

To realize the ignition simulation of a small-current arc, the duration of the arc needs to be guaranteed during the arc ignition process of the moving electrode. The longer the arc duration, the longer the cable is heated, and the easier it is to ignite. If the arc duration is short, the arc may be extinguished before the cable ignites. The changing trends of arc duration, electrode moving speed, and current size were drawn through experiments, as shown in Figure 3 and Figure 4. A negative correlation exists between arc duration and electrode movement speed. When the electrode moving speed is high, the arc duration is short, and the cable is difficult to ignite. In contrast, the arc duration is positively correlated with the current. If the current is increased, the arc duration will be longer, which will help ignite the cable. Therefore, the process of moving the electrode to ignite the cable is affected by both the moving speed of the electrode and the current conditions. Based on the simulated fault conditions, five different current conditions from 2 A to 10 A are selected. To clarify the electrode moving speed that can achieve cable ignition under these five current working conditions, ignition tests at different speeds were carried out on the surface of the cable body to ensure that the electrode moving speed could achieve cable ignition under each current working condition.

To determine the electrode moving speed that can achieve cable ignition under each current operating condition, 1 mm/s was first used as the initial electrode moving speed. The dichotomy method is used to determine the electrode moving speed that can achieve cable ignition. Then, the speed can be increased or decreased in steps of 0.01 mm/s. Therefore, the electrode moving speed range under this current condition was finally determined. Taking the 10 A current working condition as an example, an arc was generated on the surface of the cable sample at the moment the electrode moves, as shown in Figure 5. Under this working condition, cable ignition can be achieved when the electrode moving speed is 1 mm/s. Then, tests were conducted by increasing or decreasing at 0.01 mm/s to both ends. The test results show that when the electrode moving speed exceeds 1.28 mm/s, the arc is extinguished before the cable ignites; thus, the simulation process of cable ignition cannot be realized. When the moving speed continues to decrease, the electrode moving speed slows down, resulting in an increase in the time required to achieve ignition at the same distance and causing the protection resistor in the test circuit to continue to heat under the action of the current. Under 10 A current working conditions, when the electrode moving speed is less than 0.46 mm/s, the cable carbonization channel will connect the two electrodes, causing the arc to disappear. Therefore, under 10 A current conditions, the appropriate electrode moving speed range is 0.46~1.28 mm/s.

The same method was used to test the current conditions of 2 A, 4 A, 6 A, and 8 A. The appropriate electrode moving speed range that can achieve cable ignition under various working conditions is shown in

Table 1. Range of electrode movement speeds for different current conditions.

Test Current (A)	Electrode Speed (mm/s)
2	0.02–0.11
4	0.08–0.22
6	0.18–0.73
8	0.29–0.95
10	0.46–1.28

.

Under the 10 A current working condition, based on the determined electrode moving speed range, a median of 0.87 mm/s is taken to carry out electrode arc ignition on the cable surface. The cable was ignited and burned, as shown in Figure 6. Under the continuous action of arc high temperature, the cable gradually decomposes upon heating, producing flammable gases. These gases are mixed with oxygen and ignited at the high temperatures of the arc. As the flame spreads to the periphery of the arc action area, the fire gradually intensifies. The above experimental phenomena show that this method can effectively realize the simulation test of arc ignition process of the cable. When the experiment is stopped, the electrode stops moving. The cable carbonization channel connects two electrodes, and the arc is extinguished.

2.2. Fuse Melting Ignition Method

A gap exists between the high-voltage electrodes and ground electrodes. A fuse was initially used to connect the two electrodes to form a complete circuit. The current in the loop flows from the high-voltage side to the ground side through the fuse. The fuse has a certain resistance. The fuse begins to heat up under the action of the current. When the temperature rises to the melting point of the fuse, the fuse melts. A liquid metal bridge or metal vapor is formed between the two electrodes, eventually causing breakdown and arc formation at the gap.

2.2.1. Building a Test Platform

Based on the above-mentioned moving electrode ignition test platform, the arc starting position was adjusted. To achieve arc ignition through fuse melting, the cable body sample was used to replace the high-voltage electrode, and the internal metal conductor of the cable was connected to the high-voltage end. According to the simulated fault conditions, hole defects were created on the cable insulation layer or outer sheath, and the grounding electrode was arranged on the outer surface of the insulation layer or outer sheath accordingly. The fuse was short-circuited to the metal conductor inside the cable and the ground electrode on the surface of the cable through the hole. The arrangement is shown in Figure 7.

2.2.2. Realization of Continuous Arc Ignition and Cable Ignition Processes

Differences exist in the heating conditions of fuses of the same specification under different current operating conditions. Therefore, the fusing time of the fuse and whether it can be blown are closely related to the current operating conditions and fuse specifications. Five different current conditions (2 A, 4 A, 6 A, 8 A, and 10 A) were selected, and four different specifications of fuses were used. The rated current and diameter were 1 A/0.2 mm, 2 A/0.4 mm, 3 A/0.6 mm, and 4 A/0.8 mm, respectively. The cable and the ground electrode are short-circuited through fuses, and five different working condition currents are passed through the four specifications of fuses. The melting conditions of fuses of different specifications under each current condition are shown in

Table 2. Fusing time of different fuse sizes under various current conditions.

Test Current (A)	Rated Current 1 A Fuse	Rated Current 2 A Fuse	Rated Current 3 A Fuse	Rated Current 4 A Fuse
2	56 s	>300 s	>300 s	>300 s
4	3 s	34 s	34 s	>300 s
6	/	7 s	56 s	>300 s
8	/	/	28 s	75 s
10	/	/	8 s	37 s

. For fuses of the same specification, as the current increases, the degree of heating increases, and the melting time shortens. When the gap between the rated current of the fuse and the test current is large, the fuse will melt before the test current reaches the preset value. Under the same test current, the melting time of the fuse increases as the fuse specification increases, and the closer the rated current of the fuse is to the test current, the more difficult it is to melt the fuse.

Based on the corresponding relationship between the melting time of different test currents and different fuses determined above, it can be ensured that the fuse melts within a controllable time after the current flows through the fuse. However, in the test of the fuse melting condition under different current conditions, it was found that there is instability after the fuse is blown. This is mainly because as the arc length changes, the maintenance voltage and current required for the arc burning process will also change accordingly. With a longer gap, higher voltage and current are required to maintain stable combustion of the arc, therefore ensuring that the arc can be stably generated to ignite the cable after the fuse is blown. According to the corresponding relationship obtained in Table 2, select the corresponding specification fuse based on the principle of the shortest melting time. Taking 1 mm as the initial gap distance, gradually increase the gap distance in steps of 0.5 mm. Test after the fuse blows under five different current conditions. The critical gap distance that stabilizes the arc is obtained.

Taking the 10 A current working condition as an example, select a fuse with a rated current of 3 A to connect the high-voltage terminal and the ground terminal. When the gap gradually increases from 1 mm to 16.5 mm, arcing can be stable at the moment the fuse blows. The arcing moment is shown in Figure 8a. The arc generated between the gaps emits strong white arc light and then continues to arc stably in the gaps. When the gap distance increases to 17 mm, short-term combustion occurs when the fuse blows. A yellow light similar to a flame can be seen between the gaps, as shown in Figure 8b. There was no obvious strong arc as shown in Figure 8a, and then the fire went out and disappeared. Figure 9 shows the gap voltage waveform variation curves when the spacing is 16.5 mm and 17 mm, respectively. Monitoring starts from the moment the current is applied. At about 8 s, the fuse blows, and the gap voltage begins to rise. The voltage waveforms under both spacings showed a saddle shape with arc characteristics and had an arc extinction spike and arc spike. At a spacing of 16.5 mm, stable combustion occurs without arc extinction after arc formation. However, at a 17 mm pitch, the arc can only last about 8.35 s. Therefore, under the 10 A current condition, according to the above test results, the critical distance that can achieve stable arcing after the fuse is blown is 16.5 mm. The same test method was used to determine the critical value of the gap that can form a stable arc after the fuse is blown under four current conditions of 2 A, 4 A, 6 A, and 8 A. The test results are shown in

Table 3. Pitch threshold for stable arc ignition after fuse melting.

Test Current (A)	Spacing Size (mm)
2	3
4	5.5
6	7
8	10.5
10	16.5

.

After the above 16.5 mm fuse was blown under the 10 A current condition, a stable and continuous burning arc was generated between the external ground electrode and the internal metal conductor gap of the cable. Under the high temperature of the arc, the cable material on the inner wall of the hole defect is gradually heated and ignited. During the burning process of the cable, the original hole gradually expands, the burning material increases, and the fire intensity increases. In the continuous accumulation of carbides after ablation, a permanent fault is eventually formed between the cable metal conductor and the external ground electrode, and the arc is extinguished. The cable then undergoes self-sustaining combustion. As shown in Figure 10. In the hole defect created, after the fuse connecting the metal conductor of the cable and the ground electrode melted and started arcing, the cable was ignited and burned. This shows the feasibility of this method for carrying out simulated small-current arc ignition cable tests.

2.3. Analysis of Ignition Effects under Two Methods

The above test results show that both methods can simulate the ignition process of cables under the action of a small-current arc. Both methods have their advantages and disadvantages. For the moving electrode ignition method proposed in Section 2.1, since the initial arc is formed at a very small distance at the moment when the electrode moves, even a small current can maintain stable combustion of the arc. Therefore, the current range of fault conditions that this method can simulate is wide. However, during the ignition process of the moving electrode, if the electrode moving speed is too fast or too slow, the test results will be unsatisfactory, and the applicable electrode moving speed may change depending on the cable material and other conditions. This speed range has many prerequisites and generally poor generalizability. For the fuse melting ignition method proposed in Section 2.2, after the fuse is heated and melted, the arcing process generated between the gaps is stable and can continue to act on surrounding cable materials. However, a critical arcing distance exists after the fuse is blown under different currents, and the critical distance is positively correlated with the current value. Therefore, this method makes it difficult to simulate the arc ignition process under smaller current conditions at a larger spacing, and the range of fault current simulations is limited.

Based on the above analysis results, the fuse melting ignition method is improved. Simulation of arc ignition with small amplitude current needs to be implemented at large intervals. Firstly, a higher amplitude current was applied to the loop so that an arc could be formed in the gap after the fuse was blown. After the arc was generated, the current size was adjusted to the small amplitude current condition that needed to be simulated. Then, the fuse melting ignition method was used to carry out the cable ignition simulation test under five different current conditions and different arcing times, and the ignition characteristics and fire risk of the cable under different working conditions were explored and analyzed.

3. Analysis of Arc Waveform Characteristics and Ignition Characteristics

Based on the above research, the ignition test of cables was carried out using the methods of fuse melting and arc initiation. Different currents and arc durations were selected to obtain the ignition situation and combustion degree of cables under different working conditions, and the ignition characteristics of cables under different working conditions were analyzed.

3.1. Arc Ignition Conditions

Cable ignition tests were conducted under different arc current sizes and arc durations to investigate the ignition effect of a low-current arc on cable samples. The experimental steps are as follows:

(1)

Multiple sets of cable test samples with a length of 15 cm and a diameter of 4.5 cm were prepared. The main materials for the cable sample from the outside to the inside were insulation layer and copper conductor.

(2)

A hole defect of about 5 mm was manufactured on the cable insulation layer. According to the data in Table 3, the corresponding specifications of lead wire were selected based on the current situation, and the lead wire was used to short-circuit the metal conductor inside the hole with the external grounding electrode. The ambient temperature was 33 °C, and the humidity was 43%. A thermocouple 7.5 cm was placed directly above the cable. The environment was open and windless.

(3)

Five current conditions (2 A, 4 A, 6 A, 8 A, and 10 A) were selected to conduct ignition tests on cables. An arc is generated between the metal conductor of the cable and the grounding electrode after the fuse is blown, and the arc continues to act on the cable hole position. The ignition arc times of cables under several current conditions were determined through experimental phenomena.

(4)

Under these five current conditions, two different arcing durations were selected to carry out cable ignition tests under different currents and arcing durations. The flame height, flame temperature, arc voltage, and current waveform characteristics during the combustion process after cable ignition and the spontaneous combustion of the cable after arc extinction were obtained. The ignition and combustion characteristics under different currents and arc durations were analyzed.

3.2. Cable Ignition and Combustion Characteristics

(1)

Characteristics of arc voltage waveform

During the arc ignition cable test, the gap voltage waveform changes at each stage of the arc ignition cable were measured by a high-voltage probe, as shown in Figure 11.

At the initial stage, the cable metal conductor and the grounding electrode are shorted by the lead wire in the gap, and the gap voltage is almost zero at this time. As the lead wire melts, an arc is generated in the gap, and the voltage waveform presents a saddle-like shape, accompanied by extinguishing and ignition spikes at the zero-crossing point. This is consistent with the traditional arc voltage waveform. The cable ignites and burns under the action of an arc. The arc gradually stabilizes in the flame state. The distortion degree of the extinguishing peak and the ignition peak decreases. Carbonization and erosion gradually occur during the process of cable combustion, ultimately leading to permanent faults. The arc extinguishes, and the gap voltage waveform changes from a rectangular arc wave to a nearly sine wave.

(2)

Cable ignition time

Figure 12 shows the development process of cable ignition under 10 A current condition. The test begins after the circuit is closed. At the 8th second of the circuit current, the fuse is melted, and a bright arc is generated. During the continuous high-temperature action of the arc, the cable material continuously decomposes under heat, producing combustible gases. After 4 s of arc action, the bright silver white arc light between the cable metal conductor and the grounding electrode transformed into a yellow flame, accompanied by sparks splashing around, igniting the cable. The critical arc duration and arc power for cable ignition under five current conditions are shown in

Table 4. Cable ignition time under various current conditions.

Testing Current	Arc Power	Ignition Time
2	75 W	28 s
4	112 W	21 s
6	189 W	14 s
8	507 W	9 s
10	820 W	4 s

.

A cable ignition test was conducted under 10 A current conditions, and it was found that the cable burned more fiercely and the fire rapidly increased after being ignited. After the arc ignition time reaches about 135 s, the surrounding cables of the arc-affected area are completely eroded, and the flame gradually spreads outward. At this time, the arc is far away from the flame, and the high temperature during the arc burning process is difficult to directly promote flame combustion, so the fire is greatly reduced, as shown in Figure 13.

To focus on analyzing the combustion characteristics of cables during intense combustion processes, the time periods from cable ignition to fire reduction under five working conditions are considered, with two types of arc ignition times selected: 50 s and 100 s. Then, cable ignition and combustion tests were conducted under five current conditions ranging from 2–10 A to analyze the characteristics of cable ignition and combustion at different arc times under each current condition.

(3)

Cable burning degree

As shown in Figure 14, the combustion of the cable occurs when the arc time reaches 100 s under five different current conditions. As the current increases, the arc energy gradually increases. The energy transmitted to the surrounding cable materials during the arc combustion process is also higher, resulting in more intense cable combustion. The size of the fire increases with the increase of electric current. The flame heights under these five current conditions can reach 16 cm, 32 cm, 38 cm, 51 cm, and 63 cm, respectively. Figure 15 shows the cable erosion after 100 s of arc action under five current conditions. Based on the above reasons, a larger current causes a stronger fire during the cable combustion process, resulting in a larger and deeper burn area of the cable.

As shown in Figure 16, the combustion process of the cable and the appearance of the cable after erosion were observed when the arc was applied for 50 s and 100 s under a current of 10 A. Under the same current conditions, as the arc duration increases and the cable erosion time increases, the flame spreads around the arc action area, the erosion area expands, and the fire burns more fiercely.

Table 5. Flame height under various arc conditions.

Arc Duration	2 A	4 A	6 A	8 A	10 A
50 s	9 cm	24 cm	29 cm	44 cm	52 cm
100 s	16 cm	32 cm	38 cm	51 cm	63 cm

shows the comparison of flame height during cable combustion under different current conditions when the arc lasts for 50 and 100 s, respectively.

During the combustion process under different arc conditions, temperature changes were obtained by arranging thermocouples above the cable. According to the data obtained from Table 5, the minimum value of the maximum flame height data of the cable during combustion is 9 cm. The height of 9 cm is higher than the height of 7.5 cm, where the thermocouple is arranged. Therefore, the thermocouple is located within the flame envelope, and the temperature measured by the thermocouple is considered to be the flame temperature of the cable burning under different working conditions.

As shown in Figure 17, the flame temperature variation curves at 50 and 100 s of arc ignition under five different current conditions are presented. The starting point for curve drawing is the moment of arc initiation. Before the cable ignition time, the temperature measured at the thermocouple is the ambient temperature. At the moment of cable combustion, the temperature curve rapidly rises due to the high flame temperature. After the flame temperature reaches a certain value, due to the action of the arc, the AC arc will drive the flame to oscillate periodically, and the free radicals and ions generated by the arc will also affect the stability of the flame. Therefore, the temperature fluctuates within a certain range after rising to the flame temperature.

At an arc time of 50 s, the average flame temperature and its temperature fluctuation range under 2–10 A current conditions are 421 (394–506) °C, 542 (483–685) °C, 686 (610–754) °C, 757 (709–826) °C, and 934 (885–976) °C, respectively. The data are 410 (335–538) °C, 558 (485–670) °C, 680 (610–733) °C, 778 (691–837) °C, and 903 (781–979) °C at 100 s arcing time, respectively. Under two different arc ignition times, the same flame temperature under corresponding current conditions is almost the same, and the temperature during the combustion process is not affected by the arc ignition time. However, as the current increases, the arc power increases, and the energy transmitted to the surrounding area during the arc burning process is higher. The flame temperature is positively correlated with the current.

(4)

Cable self-sustaining combustion

According to the combustion process of the cable under the action of an arc in the above experiments, the initial combustion of the cable is mainly concentrated at the location where the arc is generated. With the increase of current and arc time, although the flame combustion range has increased, it is mainly concentrated around the arc burning area.

As shown in Figure 18, the cable combustion situation after arc extinction is observed under two combustion conditions: 2 A/50 s and 10 A/100 s. The fire intensity is significantly reduced in both operating conditions compared to the combustion process under the aforementioned arc action. When the current is low and the arc time is short, the flame cannot sustain spontaneous combustion after the arc is extinguished. After the arc is extinguished, the cable burns for only 8 s before the flame is extinguished, and its burning area does not further expand. When the current is large and the burning time is long, although the fire weakens after the arc is extinguished, it can continue to burn and spread to both ends of the cable sample until the cable is completely burned out. After the arc was extinguished, the cable continued to burn for 620 s, causing the damaged area of the cable to further expand. The relationship between the residual flame time and the arc current after the arc was extinguished is shown in Figure 19.

4. Conclusions

In order to explore the ignition characteristics and fire risk of high-voltage cable under the continuous action of low-current fault arc, two test simulation methods, moving electrode ignition and fuse melting ignition, are proposed in this test, and the results under different current conditions and different arc time are analyzed. Based on the arc ignition situation, the following conclusions are drawn:

(1)

The moving electrode method can achieve long-distance arc ignition when the current is small. By controlling the moving speed of the electrode, it can be ensured that the arc will not be interrupted during the movement of the electrode. This method is suitable for simulating the ignition situation of cable outer sheath damage. The test operation of the fuse melting method is relatively simple and does not require moving electrodes. The arc burning process is stable. This method is suitable for simulating the arc ignition situation of single-phase breakdown of the insulation layer.

(2)

After improving the fuse melting ignition method, arc ignition tests under five current conditions were carried out on the cable body sample. Under the continuous action of the arc under five different working conditions of 2 A, 4 A, 6 A, 8 A, and 10 A, the time of cable ignition is 28 s, 21 s, 14 s, 9 s, and 4 s, respectively.

(3)

The flame height and the burning time of the cable under the action of the arc are positively correlated with the current and the arcing time. The flame temperature changes with the current when the cable burns. When the current rises to 10 A and the arcing time reaches 100 s, the maximum height of flame combustion can reach 63 cm, and the maximum temperature can reach 979 °C. After the cable is burned for 261 s, the whole cable sample is completely ablated. The ignition characteristics of the cable under several different current conditions from 2 A to 10 A obtained in this paper, combined with the actual fault record data, can provide a reference for judging whether it is necessary to remove the fault within a certain period of time so as to avoid the cable burning after a long-term operation fault, causing more serious short-circuit or fire accidents.