Search

The effect of metal core on flame spread behavior over polyethylene (PE)-insulated electrical wire was experimentally investigated by varying applied AC voltage (V AC) and frequency (f AC). The present experimental results with Cu-core were compared with those conducted previously with NiCr-core. Flame spread rate (FSR) with Cu-core for the baseline case with no electric field was larger than that with NiCr-core. Both FSR and flame size were appreciably influenced by applied AC electric fields and behaved differently for Cu- and NiCr-cores. For Cu-core, FSR behavior could be classified into four regimes as f AC increases: increasing FSR (regime I), decreasing FSR (II), and again increasing FSR (III) and decreasing FSR (IV). While for NiCr-core, it has been categorized into two regimes I and II. FSR behavior was qualitatively similar to that of flame width in regimes I, II, and IV. While such a relationship was not satisfied in regime III, which can be attributed to the formation of molten PE film in the burnt wire side as well as the formation of globular molten PE in front of spreading flame edge, emphasizing the important role of complex molten PE behavior. Molten PE dripping was observed for NiCr-core, while for Cu-core, such dripping was not observed. Electrospray and di-electrophoresis phenomena occurred with Cu-core along with a formation of molten PE film on the burnt wire side via continuous di-electrophoresis phenomenon from molten PE droplet. When the frequency was excessive, flame extinction occurred via two routes: appreciable reduction of flame size (both for Cu- and NiCr-cores) and detachment of molten PE (for Cu-core). These extinction frequencies were correlated well with the voltage. [ABSTRACT FROM AUTHOR]

The effect of DC electric field on the characteristics of flame spread over polyethylene-insulated twin electrical wires was studied by varying wire gap (S) and voltage (V DC). Under an applied electric field, the flame spread rate (FSR), flame width, leaning direction of the interacting twin flames varied substantially with varying the voltage and wire gap. The flame spread rate was initially larger for the wire with negative voltage (spreading flame with negative charge; SF −) than the wire with positive voltage (SF +), but the two eventually became the same in the developed region when a quasi-steady state was reached. The FSR behavior could be classified into two regimes; twin flame spread (regime I) and single flame spread (regime II) after the extinction of SF+. Under regime I, three sub-regimes were identified depending on the wire gap and voltage. For the twin flame spread, the flame spread rate initially decreased with increasing voltage as the flame leaned toward the burnt wire. As the two flames interacted, the flame spread rate increased because of the ionic wind effect, and eventually decreased because of the loss of molten PE mass and the electrospray phenomenon. In regime II after the extinction of SF+, the single flame spread showed a transient behavior since the influences of electric field from burnt and unburned wire sections of SF+ wire varied with flame spread. When the voltage was increased even further, SF– was extinguished by streamer generation and, at excessive voltages, an electrical short occurred. The flame spread rates for twin flame spread were best correlated with the electric field intensity in the form of | V DC |0.91/ S 0.72. [ABSTRACT FROM AUTHOR]

Abstract The effect of AC electric field on flame spread over polyethylene (PE)-insulated electrical wire was experimentally investigated by varying the applied AC voltage (V AC) and frequency (f AC) with special attention to the effect of the thickness of the PE insulation material (T ins). The diameter of the Ni–Cr wire was 0.5 mm and T ins was 0.15, 0.3, and 0.5 mm, yielding outer diameters (D out) of 0.8, 1.1, and 1.5 mm, respectively. For the baseline cases with no electric field, the flame spread rate (FSR) (flame size) decreased (increased) with insulation thickness. Both FSR and flame size were appreciably influenced by applied AC electric fields. The FSR behaviors under applied electric fields could be classified into three sub-regimes as AC frequency increased: regime I exhibited a decreasing FSR as AC frequency increased, regime II exhibited an increasing FSR, and regime III again exhibited a decreasing FSR. Molten PE dripped from the wire (resulting in mass loss); exhibited di-electrophoresis (some molten PE moved from the main molten zone toward the burnt wire, forming globules in the process); and developed electrosprays (ejection of small droplets from the molten PE surface). For D out = 0.8 mm, the FSR behavior was similar to that of the flame width, such that the behavior could be explained by the thermal balance mechanism. When a low voltage and high frequency were applied to wires with D out = 1.1 and 1.5 mm, molten PE droplets detached and moved to the burnt wires continuously (although sometimes intermittently) from the main body of molten PE; the FSR behavior thus deviated from that of regime I. Droplet detachment was attributable to a di-electrophoresis. The distance moved correlated well with the difference in electric field intensities of burnt and unburned wires. Appreciable dripping of molten PE occurred at high voltages and moderate frequencies in regime II. When the frequency was excessive, flame extinction occurred via two routes: appreciable reduction of flame size when D out = 0.8 mm in regime I and appreciable fuel mass loss via dripping of molten PE during flame spread when D out = 0.8 and 1.1 mm in regime III. These extinction frequencies correlated well with V AC / D out. When high voltage and frequency were applied in the D out = 1.5 mm case, droplets detached and moved to the burnt wire via di-electrophoresis; subsequently a series of fine droplets ejected from the surface via electrospraying, while the molten PE region grew and subsequently dripped. In such cases, flame extinction did not occur because di-electrophoresis increased the flame width and thereby the FSR over the experimental ranges of V AC and f AC. [ABSTRACT FROM AUTHOR]